System

Abstract

To provide a system capable of efficiently selling unsold stock.SOLUTION: A system includes: means for collecting sales data and inventory data and selecting an apparel item with sales below a certain threshold; means for generating a prompt sentence instructing the creation of a coordinate combining the selected apparel item with other items; means for generating the coordinate combining the apparel item with other items using the generated prompt sentence and a generative AI model; means for generating a prompt sentence instructing the creation of an image of the coordinated apparel item being worn on a model, a catch phrase, and an item description; means for generating the image of the coordinated apparel item being worn on a model, the catch phrase, and the item description using the generated prompt sentence and the generative AI model; and means for creating a sales page using the image of the apparel item being worn on a model, the catch phrase, and the item description.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In the apparel industry, the disposal of unsold inventory is a major issue. The reuse of used clothing from consumers is also a challenge. To solve these issues, a method for efficiently selling unsold inventory and used clothing from consumers is required. [Means for solving the problem]

[0005] This invention uses AI to select apparel items that have been in stock for a long time and coordinate them across brands. It then generates images of the coordinated apparel items being worn by models, a catchy slogan, and product descriptions, and creates a sales page. The AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase via social media as well. The AI ​​also generates images of models wearing the items and product descriptions on the same platform for consumers' used clothing, and sells them. This solves the issues of disposing of unsold inventory and reusing consumers' used clothing. [Brief explanation of the drawings]

[0006] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11]FIG. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0011] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0012] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0017] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0018] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0019] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0020] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0024] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0025] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] One embodiment of the present invention is an apparel sales system using AI. In this system, AI first selects apparel items that have been in stock for a long time. For example, it analyzes sales and inventory data and automatically selects items with slowing sales. Next, it coordinates the selected items across brands. This makes it possible to suggest combinations that work well together, even among products from different brands. Images of the coordinated items worn by models, catchy slogans, and product descriptions are automatically generated by the AI. This enables the efficient creation of sales pages that maximize the appeal of the products. Furthermore, the AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media. Furthermore, AI also generates images of models wearing the items and product descriptions for the consumer's used clothing on the same platform and sells them. This encourages consumers to reuse their used clothing and achieves zero waste.

[0029] "Example 2"

[0030] One embodiment of the present invention is an apparel sales system using AI. In this system, AI first selects apparel items that have been in stock for a long time. For example, it analyzes sales and inventory data and automatically selects items with slowing sales. Next, it coordinates the selected items across brands. This makes it possible to suggest combinations that work well together, even among products from different brands. Images of the coordinated items worn by models, catchy slogans, and product descriptions are automatically generated by the AI. This enables the efficient creation of sales pages that maximize the appeal of the products. Furthermore, the AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media. Furthermore, AI also generates images of models wearing the items and product descriptions for the consumer's used clothing on the same platform and sells them. This encourages consumers to reuse their used clothing and achieves zero waste.

[0031] The processing flow of each embodiment will be described below.

[0032] "Example 1"

[0033] Step 1: The AI ​​selects apparel items that have been in stock for a long time. Specifically, it analyzes sales and inventory data and automatically selects items whose sales are slowing.

[0034] Step 2: Coordinate the selected products across brands, making it possible to propose combinations that go well together even when products from different brands are used.

[0035] Step 3: Images of models wearing the coordinated products, along with catchy slogans and product descriptions, are automatically created by generative AI. This makes it possible to efficiently create sales pages that maximize the appeal of the products.

[0036] Step 4: The AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase the product via social media.

[0037] Step 5: Consumers can also purchase their used clothing on the same platform, with AI generating images of models wearing the clothing and product descriptions. This will encourage consumers to reuse their used clothing and achieve zero waste.

[0038] Example 1

[0039] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] Conventional apparel sales systems lacked efficient methods for selling long-term inventory. Furthermore, there were no effective methods for suggesting outfits that combined products from different brands, or for reusing consumers' old clothes. This resulted in increased inventory backlog and waste, and reduced sales efficiency.

[0041] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0042] In this invention, the server includes means for collecting sales data and inventory data, means for analyzing the collected data to select long-term inventory items, means for proposing outfits by combining the selected items with items from different brands, means for generating images of the coordinated items worn by models, catch phrases, and product descriptions using a generative AI model, means for creating a sales page using the generated images of the coordinated items worn by models, catch phrases, and product descriptions, means for automatically setting sale prices by executing a pricing algorithm based on past sales data, means for automatically posting the created sales page to social media, means for allowing consumers to purchase via social media, and means for generating images of the items worn by models and product descriptions and selling consumers' used clothing on the same platform. This makes it possible to efficiently sell long-term inventory items, propose attractive outfits by combining items from different brands, and encourage consumers to reuse used clothing.

[0043] "Sales data" refers to data that includes transaction information regarding the sale of apparel products.

[0044] "Inventory data" is data that indicates the current inventory status of apparel products.

[0045] "Long-term inventory" refers to apparel products that remain unsold for a certain period of time.

[0046] "AI" is a system that uses artificial intelligence technology to analyze data and make predictions.

[0047] "Coordination" is a fashion combination that combines products from different brands.

[0048] A "generative AI model" is an artificial intelligence model for generating natural language and images.

[0049] "Model wearing image" refers to an image of a model wearing an apparel product.

[0050] A "catchphrase" is a short promotional phrase that draws out the appeal of a product.

[0051] A "product description" is a text that explains the features and benefits of a product.

[0052] A "sales page" is a web page for selling apparel products online.

[0053] "Sale price" refers to the selling price of a product.

[0054] "SNS" refers to social networking services, which are online platforms for users to share information.

[0055] "Consumer" refers to a customer who purchases apparel products.

[0056] "Used clothing" refers to second-hand apparel items that consumers have previously purchased and wish to resell.

[0057] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[0058] The server first collects sales data and inventory data. To do this, MySQL (registered trademark) is used as the database management system (DBMS). The server retrieves sales data for the past six months and current inventory data using SQL queries.

[0059] Next, the server analyzes the collected data and selects items that have become long-term inventory. This analysis is performed using Python (registered trademark) and TENSORFLOW (registered trademark). Items whose sales have fallen below a certain threshold are selected as long-term inventory. For example, items whose sales have been less than 10 units in the past six months are selected.

[0060] For the selected items, the server proposes outfits by combining them with products from different brands. To propose outfits, OpenCV is used to evaluate color compatibility and Scikit-learn is used to evaluate style compatibility. Specifically, a color wheel is used to evaluate color compatibility, and a clustering algorithm is used to evaluate style compatibility.

[0061] Next, the server uses a generative AI model (e.g., GPT-4 (registered trademark)) to generate an image of a model wearing the coordinated products, a catchy slogan, and a product description. The following prompt sentence is input to the generative AI model:

[0062] "This is a long-term inventory item. Please generate a model image, tagline, and product description based on the following information: Product name: XXXX, Brand: YYYY, Color: ZZZZ, Style: AAAA"

[0063] Based on the generated information, the server uses an image generation API (e.g., DALL-E) to generate an image of the model wearing the product.

[0064] The server then runs a pricing algorithm based on past sales data to determine the optimal sale price. To do this, it uses XGBoost to build a pricing model and predict the optimal sale price for each product.

[0065] The server then creates a sales page based on the generated model images, tagline, product description, and sale price. The sales page is built using HTML and CSS. The link to the created sales page and the generated tagline are automatically posted to social media using an API (Application Programming Interface).

[0066] Additionally, users can input photos of used clothing and basic information (brand, color, style, etc.) from their devices. The server uses this information to generate an image of the clothing being worn by a model and a product description using a generative AI model. For example, the following prompt sentence can be input into the generative AI model:

[0067] "This used clothing item is for sale by a user. Please generate a model image, tagline, and product description based on the following information: Product name: XXXX, Brand: YYYY, Color: ZZZZ, Style: AAAA"

[0068] Based on the generated information, the server creates a sales page and automatically posts it to social media, allowing users to easily sell their used clothing.

[0069] In this way, the system of the present invention makes it possible to efficiently sell long-term inventory items, propose attractive outfits that combine products from different brands, and encourage consumers to reuse used clothing.

[0070] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0071] Step 1: The server collects sales and inventory data. Specifically, the server connects to a MySQL database and retrieves sales data for the past six months and current inventory data using an SQL query. The input is the database connection information and query, and the output is sales data and inventory data.

[0072] Step 2: The server analyzes the collected data and selects items that have become long-term inventory. Specifically, the server uses Python and TensorFlow to analyze sales data and selects items whose sales fall below a certain threshold as long-term inventory. The inputs are sales data and inventory data, and the output is a list of long-term inventory items.

[0073] Step 3: The server combines the selected long-term inventory items with products from different brands to suggest outfits. Specifically, the server evaluates color compatibility using OpenCV and style compatibility using Scikit-learn. The input is a list of long-term inventory items and product data from other brands, and the output is a list of outfit suggestions.

[0074] Step 4: The server uses the generative AI model to generate images of the coordinated products worn by a model, a catchphrase, and a product description. Specifically, the server inputs a prompt to the generative AI model and obtains the generated text and images. The input is a list of outfit suggestions and a prompt, and the output is an image of the product worn by a model, a catchphrase, and a product description.

[0075] Step 5: The server creates a sales page using the generated model image, tagline, and product description. Specifically, the server builds the sales page using HTML and CSS. The input is the model image, tagline, and product description, and the output is an HTML file for the sales page.

[0076] Step 6: The server runs a pricing algorithm based on past sales data to automatically set the sale price. Specifically, the server uses XGBoost to build a pricing model and predict the optimal sale price for each product. The input is past sales data, and the output is the sale price.

[0077] Step 7: The server automatically posts the created sales page to social media. Specifically, the server uses an API to post the sales page link and the generated tagline. The input is the sales page link and tagline, and the output is the result of posting to social media.

[0078] Step 8: The user enters a photo of the used clothing and basic information (brand, color, style, etc.) from the device. Specifically, the user enters this information using a web form. The input is a photo of the used clothing and basic information, and the output is the data sent to the server.

[0079] Step 9: The server uses the generative AI model to generate an image of a model wearing the item and a product description based on the information entered by the user. Specifically, the server inputs a prompt into the generative AI model and obtains the generated text and image. The input is a photo of the used clothing, basic information, and the prompt, and the output is an image of a model wearing the item and a product description.

[0080] Step 10: The server creates a sales page based on the generated information and automatically posts it to social media. Specifically, the server uses HTML and CSS to build the sales page and posts it using an API. The input is an image of the model wearing the product, a product description, and a link to the sales page, and the output is the result of posting to social media.

[0081] (Application example 1)

[0082] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0083] Conventional apparel sales systems have faced the problem of making it difficult to efficiently sell long-term inventory items, requiring a great deal of effort for inventory management and sales promotion. Furthermore, they lacked systems for suggesting outfits that combine products from different brands or for reusing consumers' old clothing, making it impossible to provide an attractive purchasing experience for consumers. Furthermore, they lacked the ability to automatically generate product descriptions using generative AI models or automatically post them to information sharing services, making sales page creation and promotional activities inefficient.

[0084] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for selecting long-term inventory apparel items, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for enabling consumers to purchase the items via the information sharing service, means for generating images of the items worn by models and product descriptions for consumers' used clothing on the same platform and selling them, means for analyzing sales data and inventory data, means for proposing coordination by combining products from different brands, means for generating product descriptions using a generative AI model, and means for posting the generated product descriptions to the information sharing service. This enables efficient sales of long-term inventory items, attractive coordination suggestions combining products from different brands, reuse of consumers' used clothing, and efficient creation of sales pages and promotional activities using the automatic generation function utilizing the generative AI model.

[0085] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0086] "Apparel products" refers to fashion-related products such as clothing and fashion accessories.

[0087] "Creating coordination that transcends brand boundaries" refers to proposing styling by combining products from different brands.

[0088] "Model wearing image" refers to an image of a model actually wearing the product.

[0089] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0090] "Product description" refers to text that explains the detailed information and features of a product.

[0091] "Sales page" refers to a web page that lists product information and allows consumers to purchase the product.

[0092] "Sale price" refers to the selling price of a product.

[0093] "Information sharing services" refers to online platforms such as social networking sites.

[0094] "Used consumer clothing" refers to second-hand clothing that is owned by a consumer.

[0095] "Platform" refers to the underlying system for providing specific services or functions.

[0096] "Sales Data" refers to data regarding product sales performance.

[0097] "Inventory Data" refers to data regarding the availability of products.

[0098] A "generative AI model" refers to a model that uses artificial intelligence to generate text and images.

[0099] "Prompt" refers to a text instruction that is input to a generative AI model.

[0100] A system for implementing the present invention includes a server, a user terminal, and a generative AI model. A specific embodiment of the system will be described below.

[0101] Server Processing

[0102] The server is equipped with a database for analyzing sales and inventory data and a data analysis module using artificial intelligence (AI). The server integrates sales and inventory data and automatically selects apparel items that have been in stock for a long time. The selected items are then analyzed using an AI clustering algorithm to suggest outfits that can be combined with products from different brands.

[0103] Next, the server uses the generative AI model to automatically generate images of the coordinated items worn by models, taglines, and product descriptions. The generated content is used to create a sales page. The sale price on the sales page is automatically set by the AI.

[0104] Furthermore, the server has the function of automatically posting the generated sales page to information sharing services (such as SNS), allowing consumers to purchase products through the information sharing service.

[0105] User terminal processing

[0106] The user device provides an interface for consumers to browse and purchase products. Users can check product information and complete the purchase process through the sales page. Users can also upload their own used clothing to the platform, and images of models wearing the clothing and product descriptions can be automatically generated using a generative AI model. This allows users to reuse and sell their used clothing.

[0107] Processing generative AI models

[0108] A generative AI model is an artificial intelligence model for generating text and images. Specifically, it uses natural language processing models such as GPT-2 to generate product descriptions and catchy slogans. Image generation models (e.g., GANs) are used to generate images of models wearing the products.

[0109] Prompt Sentence Examples

[0110] The following text will be used as an example of a prompt sentence to be input into the generative AI model.

[0111] Product Name: Slim Fit Jeans, Features: Stretch Material, Dark Blue, Casual

[0112] By inputting this prompt, the generative AI model generates the following product description:

[0113] These slim-fit jeans are made from a stretchy fabric for ease of movement and comfort, and their dark blue color is perfect for a casual look and pairs well with any top.

[0114] In this way, by combining the server, user terminal, and generative AI model, an efficient and attractive apparel product sales system can be realized.

[0115] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0116] Step 1: The server retrieves sales data and inventory data from the database. This data is integrated to select apparel items that have become long-term inventory. Specifically, products whose sales have been sluggish for a certain period of time are filtered out and listed as long-term inventory items. The input is sales data and inventory data, and the output is a list of long-term inventory items.

[0117] Step 2: The server proposes outfits by combining the selected long-term inventory items with products from different brands. It uses an AI clustering algorithm to analyze product characteristics and group items that go well together. The input is a list of long-term inventory items, and the output is a list of outfit suggestions.

[0118] Step 3: The server uses the generative AI model to automatically generate images of the coordinated items worn by a model, a catchphrase, and a product description. Specifically, the server inputs a prompt into the generative AI model and retrieves the generated text and images. The input is a list of outfit suggestions and the prompt, and the output is an image of the item worn by a model, a catchphrase, and a product description.

[0119] Step 4: The server creates a sales page using the generated model images, tagline, and product description. The layout and design of the sales page are automatically generated and product information is placed on it. The input is the model images, tagline, and product description, and the output is the completed sales page.

[0120] Step 5: The server automatically sets the sale price for the sales page using AI. It analyzes past sales data and market trends to calculate the optimal price. The input is the sales page and sales data, and the output is a sales page with the sale price set.

[0121] Step 6: The server automatically posts the created sales page to an information sharing service (such as SNS). Using an API, it posts the sales page link and product information. The input is the sales page with the set sale price, and the output is a notification that posting to the information sharing service has been completed.

[0122] Step 7: The user browses and purchases the product through the information sharing service. The user clicks on the link to the sales page and checks the product information. The input is the post on the information sharing service, and the output is the user's purchase procedure.

[0123] Step 8: The user uploads their own secondhand clothing to the platform, and the generative AI model automatically generates an image of the clothing being worn by a model and a product description. The user's device inputs the image and characteristics of the secondhand clothing and sends a prompt to the generative AI model. The input is the image and characteristics of the secondhand clothing, and the output is an image of the clothing being worn by a model and a product description.

[0124] Step 9: The server uses the generated model image and product description to create a sales page for the used clothing and posts it to the information sharing service. The input is the model image and product description, and the output is the sales page for the used clothing and a notification that posting to the information sharing service has been completed.

[0125] Example 2

[0126] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0127] Conventional apparel sales systems have difficulty efficiently selling products that have been in stock for a long time, and lack mechanisms for suggesting outfits that combine products from different brands or encouraging consumers to reuse used clothing. This leads to issues such as increased inventory backlog and waste, and reduced sales efficiency.

[0128] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0129] In this invention, the server includes means for collecting and analyzing sales data and inventory data, means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information transmission medium, means for consumers to purchase via the information transmission medium, and means for generating images of models wearing the apparel items and product descriptions and selling consumers' used clothing on the same platform. This enables efficient sales of long-term inventory items, suggestions for coordinating outfits that combine products from different brands, and promotion of consumers' reuse of used clothing.

[0130] "Sales data" is data that includes information regarding sales of apparel products.

[0131] "Inventory data" refers to data that includes information regarding the inventory status of apparel products.

[0132] "Long-term inventory" refers to apparel products that remain unsold for a certain period of time.

[0133] "Artificial intelligence" is a technology that uses machine learning and data analysis to automatically make judgments and predictions.

[0134] "Coordination" refers to proposing styling by combining multiple apparel items.

[0135] A "model wearing image" is an image of a model wearing an apparel product.

[0136] A "catchphrase" is a short phrase that succinctly conveys the appeal of a product.

[0137] A "product description" is a detailed description of the product's features and benefits.

[0138] A "sales page" is a web page for selling apparel products online.

[0139] "Sale price" refers to the selling price of a product.

[0140] An "information dissemination medium" is a platform for widely disseminating information, such as social media or websites.

[0141] "Consumers" refers to general customers who purchase apparel products.

[0142] "Used clothing" refers to apparel products that have been previously purchased and used by consumers.

[0143] The "platform" is a system that serves as the foundation for selling apparel products and disseminating information.

[0144] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[0145] First, the server collects and analyzes sales and inventory data. For this, a database management system (e.g., MySQL) is used. The data is read using the Python pandas library and converted into a data frame. Next, the sales and inventory data are analyzed and time series analysis is performed to identify sales trends.

[0146] Based on the analysis results, the server selects apparel items that have been in long-term inventory. Specifically, it selects items whose sales over the past six months are below a certain threshold. This selection is made using artificial intelligence (AI).

[0147] The server then coordinates the selected items across brands, using a generative AI model (e.g., GPT-4 from OpenAI (registered trademark)). The server inputs a prompt into the generative AI model and receives coordination suggestions.

[0148] The server then uses the generative AI model to automatically generate images of the coordinated products worn by the model, a catchphrase, and a product description. DALL-E is used for image generation, and GPT-4 is used for generating the catchphrase and product description.

[0149] In addition, the server uses machine learning algorithms (e.g., scikit-learn) to automatically set the sale price on the sales page, calculating the optimal price based on past sales data.

[0150] The server creates a sales page based on the generated model images, catchphrases, product descriptions, and sale prices. The created sales page is then automatically posted to information sharing media (e.g., social media), allowing consumers to purchase the product via the information sharing media.

[0151] Users can also upload product information to the platform to sell their used clothing. The server then uses a generative AI model to automatically generate images of models wearing the clothing and product descriptions, creating a sales page. This encourages users to reuse their used clothing and achieve zero waste.

[0152] As a concrete example, the following prompt sentence is input into the generative AI model:

[0153] "Analyze sales data from the past six months and select apparel items that are experiencing slower sales. Coordinate the selected items with products from different brands, generate model images, taglines, and product descriptions, set a sale price for the sales page, and automatically post to social media."

[0154] Using this prompt, the generative AI model can perform the specified task and efficiently create a sales page.

[0155] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0156] Step 1: The server collects and analyzes sales and inventory data. As input, it obtains sales and inventory data from a database management system (e.g., MySQL). Specifically, it reads the data using Python's pandas library and converts it into a data frame. As output, it performs time series analysis to obtain analytical results and identify sales trends.

[0157] Step 2: The server selects apparel items that have become long-term inventory based on the analysis results. The analysis results obtained in Step 1 are used as input. Specifically, it selects items whose sales over the past six months are below a certain threshold. The output is a list of selected long-term inventory items.

[0158] Step 3: The server coordinates the selected products across brands. It uses the list of long-term inventory items obtained in Step 2 as input. Specifically, it inputs a prompt to a generative AI model (e.g., OpenAI's GPT-4) and receives coordination suggestions. The output is a list of coordinated products.

[0159] Step 4: The server uses the generative AI model to automatically generate images of the coordinated items worn by the model, taglines, and product descriptions. The list of coordinated items obtained in Step 3 is used as input. Specifically, DALL-E is used to generate images, and GPT-4 is used to generate taglines and product descriptions. The output is an image of the model wearing the items, taglines, and product descriptions.

[0160] Step 5: The server uses a machine learning algorithm (e.g., scikit-learn) to automatically set the sale price for the sales page. It uses past sales data and the product information obtained in Step 4 as input. Specifically, it trains a machine learning model to calculate the optimal price for the new product. The output is the sale price.

[0161] Step 6: The server creates a sales page based on the generated model image, catchphrase, product description, and sale price. The product information obtained in step 4 and the sale price obtained in step 5 are used as input. Specifically, the server generates a web page using HTML and CSS. The completed sales page is obtained as output.

[0162] Step 7: The server automatically posts the created sales page to the information dissemination medium. The sales page obtained in step 6 is used as input. Specifically, it posts to the social networking site using an API. The output is the sales page published on the social networking site.

[0163] Step 8: The user uploads product information to the platform to sell their used clothing. The input is an image of the used clothing and basic information provided by the user. Specifically, the server uses the generative AI model to automatically generate an image of a model wearing the clothing and a product description, and creates a sales page. The output is a sales page for the used clothing.

[0164] (Application example 2)

[0165] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0166] With conventional apparel sales systems, it was difficult to efficiently sell long-term inventory items, and it was also difficult to suggest outfits that combined products from different brands. Furthermore, it was risky for consumers to make a purchase decision without actually trying on the product, which discouraged them from purchasing. There was a need to solve these issues and improve the consumer purchasing experience.

[0167] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0168] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an SNS, means for consumers to purchase via the SNS, means for generating images of the consumers' used clothing worn by models and product descriptions on the same platform and selling them, means for applying coordination to a user's avatar and performing a virtual try-on, and means for consumers to purchase items through the virtual try-on. This enables efficient sales of long-term inventory items, suggestions for coordinating outfits that combine items from different brands, and an improved consumer purchasing experience through the virtual try-on.

[0169] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0170] "Apparel products" refers to fashion-related products such as clothing and accessories.

[0171] "Coordinating across brand boundaries" refers to proposing styling by combining products from different brands.

[0172] "Model wearing image" refers to an image of a model actually wearing the product.

[0173] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0174] "Product description" refers to text that explains the detailed information and features of a product.

[0175] "Sales Page" refers to a web page for selling products online.

[0176] "Sale price" refers to the selling price of a product.

[0177] "SNS" is an abbreviation for social networking service, which refers to an online platform for users to share information.

[0178] "Consumer" refers to the general user who purchases the product.

[0179] "Used clothing" refers to clothing that has been previously used.

[0180] "Platform" refers to the underlying system for providing specific services or functions.

[0181] An "avatar" refers to a character or image used to represent a user in a virtual space.

[0182] "Virtual try-on" refers to users trying on products in a virtual space.

[0183] The system for implementing this invention has the following configuration. First, the server analyzes sales data and inventory data to select apparel items that have been in stock for a long time. This makes it possible to automatically select items whose sales are slowing. Next, AI is used to coordinate the selected apparel items across brand boundaries. This makes it possible to suggest combinations that go well together, even when products from different brands are combined.

[0184] The server uses a generative AI model to generate images of models wearing coordinated apparel, as well as catchy slogans and product descriptions. A sales page is then created using the generated images of models wearing the apparel, catchy slogans and product descriptions. The AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media as well.

[0185] In addition, consumers can also sell their used clothing on the same platform, generating images of models wearing the clothing and product descriptions, which will encourage consumers to reuse their used clothing and achieve zero waste.

[0186] Users can use smartphones, smart glasses, or head-mounted displays to virtually try on products. Coordination is applied to the user's avatar, allowing them to check out products through virtual try-on. This allows users to consider purchasing products without actually trying them on.

[0187] Specifically, the server uses software such as TensorFlow, OpenCV, and PIL (Python Imaging Library) to process data and perform calculations, such as loading product data, generating outfits, and running virtual try-ons.

[0188] An example of a prompt is as follows:

[0189] "Please select apparel items that have been in stock for a long time and coordinate them with items from different brands."

[0190] In this way, it will be possible to efficiently sell long-term inventory items, suggest outfits that combine products from different brands, and improve the consumer purchasing experience through virtual try-on.

[0191] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0192] Step 1: The server analyzes sales and inventory data to select apparel items that have become long-term inventory. Sales and inventory data are required as input, and a list of long-term inventory items is obtained as output. Specifically, an AI algorithm is used to automatically extract items that have remained unsold for a certain period of time.

[0193] Step 2: The server coordinates the selected apparel items across brands. It requires a list of long-term inventory items as input, and produces a list of coordinated items as output. Specifically, it uses an AI model to combine items from different brands and generate coordinated outfits that go well together.

[0194] Step 3: The server generates images of the coordinated apparel items worn by a model, a catchy slogan, and a product description. The input is a list of coordinated items, and the output is an image of the item worn by a model, a catchy slogan, and a product description. Specifically, a generative AI model is used to automatically generate images and text that highlight the appeal of the items.

[0195] Step 4: The server creates a sales page using the generated model image, catch phrase, and product description. The inputs are the model image, catch phrase, and product description, and the output is a sales page. Specifically, a web page generation tool is used to automatically create a sales page that includes detailed product information.

[0196] Step 5: The server automatically sets the sale price for the sales page. The sales page is required as input, and the sales page with the sale price set is obtained as output. Specifically, an AI algorithm is used to set the optimal sales price taking into account the market price and inventory status.

[0197] Step 6: The server automatically posts the created sales page to social media. The input is a sales page with a set sales price, and the output is the sales page posted to social media. Specifically, the server uses the social media API to post a link to the sales page along with a tagline and image.

[0198] Step 7: The user performs a virtual try-on using a smartphone, smart glasses, or a head-mounted display. The input is a list of products coordinated with the user's avatar, and the output is the results of the virtual try-on. Specifically, the coordinated items are applied to the user's avatar, and the user checks the appearance of the products through the virtual try-on.

[0199] Step 8: The user purchases the product through the virtual try-on. The input is the results of the virtual try-on, and the output is the product for which the purchase procedure has been completed. Specifically, the user checks the results of the virtual try-on and proceeds with the purchase procedure for the product they like.

[0200] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0201] "Example 1"

[0202] As one embodiment of the present invention, a system is provided that includes an emotion engine that uses AI to recognize consumer emotions. This emotion engine analyzes emotions from consumer social media posts. For example, it analyzes consumer emotions such as joy, sadness, surprise, anger, and fear from text, images, videos, etc. posted by consumers on social media. The analysis is performed using technologies such as text analysis, image recognition, and video analysis. The analyzed emotions are used to adjust product coordination. For example, if a consumer posts many times expressing joy, apparel items in bright colors can be selected and coordinated. On the other hand, if a consumer posts many times expressing surprise, apparel items with unique designs can be selected and coordinated. This makes it possible to suggest products that match the consumer's emotions, thereby increasing the consumer's desire to purchase.

[0203] "Example 2"

[0204] As one embodiment of the present invention, a system is provided that includes an emotion engine that uses AI to recognize consumer emotions. This emotion engine analyzes emotions from consumer social media posts. For example, it analyzes consumer emotions such as joy, sadness, surprise, anger, and fear from text, images, videos, etc. posted by consumers on social media. The analysis is performed using technologies such as text analysis, image recognition, and video analysis. The analyzed emotions are used to adjust product coordination. For example, if a consumer posts many times expressing joy, apparel items in bright colors can be selected and coordinated. On the other hand, if a consumer posts many times expressing surprise, apparel items with unique designs can be selected and coordinated. This makes it possible to suggest products that match the consumer's emotions, thereby increasing the consumer's desire to purchase.

[0205] The processing flow of each embodiment will be described below.

[0206] "Example 1"

[0207] Step 1: AI collects consumer social media posts, including text, images, and video data.

[0208] Step 2: The emotion engine analyzes the collected data to determine consumer emotions, using technologies such as text analysis, image recognition, and video analysis to extract emotions such as joy, sadness, surprise, anger, and fear.

[0209] Step 3: The AI ​​coordinates product outfits based on the analyzed emotions. For example, if a consumer posts a lot of content expressing happiness, it will select and coordinate brightly colored apparel items. On the other hand, if a consumer posts a lot of content expressing surprise, it will select and coordinate apparel items with unique designs.

[0210] Step 4: AI recommends coordinated products to consumers. This makes it possible to recommend products that match the consumer's emotions, increasing their desire to purchase.

[0211] Example 1

[0212] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0213] Conventional apparel sales systems have made it difficult to efficiently sell long-term inventory items or propose products that match consumer sentiment. They also lacked mechanisms for suggesting outfits that combine products from different brands or for encouraging consumers to reuse secondhand items. This has led to problems of backlogged inventory and a decline in consumer purchasing motivation.

[0214] The identification process by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means. In this invention, the server includes means for selecting long-term inventory items, means for coordinating the selected items across brands, means for generating images of the coordinated items worn by a model, a catchy slogan, and a product description, means for creating a sales page using the generated images of the items worn by a model, the catchy slogan, and the product description, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing platform, means for allowing consumers to purchase the items via the information sharing platform, means for generating images of the items worn by a model and a product description for selling the consumer's used items on the same platform, and means for analyzing consumer emotions and suggesting product coordination that matches the consumer's emotions. This enables efficient sales of long-term inventory items, suggestions for products that match consumer emotions, suggestions for coordinating items that combine products from different brands, and promotion of reuse of consumer used items.

[0215] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0216] "Means for selecting" refers to a method or device for selecting an object based on specific conditions.

[0217] "Means for coordinating" refers to a method or device for proposing a combination of multiple products.

[0218] "Model wearing image" refers to an image of a model wearing a product.

[0219] A "catchphrase" is a short sentence that succinctly expresses the features and appeal of a product.

[0220] "Product description" refers to text that explains the detailed information and features of a product.

[0221] "Sales Page" refers to a web page that displays product information and allows consumers to purchase the product.

[0222] "Sale price" refers to the selling price of a product.

[0223] An "information sharing platform" refers to online services, such as social media and websites, that allow for the widespread sharing of information.

[0224] "Consumer" refers to the general user who purchases goods.

[0225] "Used goods" refers to goods that have been previously used.

[0226] "Means for analyzing emotions" refers to methods or devices for reading emotions from text, images, etc.

[0227] A "generative AI model" refers to a model that uses artificial intelligence to generate text and images.

[0228] A "prompt sentence" refers to an input sentence that gives instructions to a generative AI model.

[0229] The present invention is a system for efficiently selling products that have been in stock for a long time and for proposing products that match consumer sentiment. Specific embodiments of this system will be described below.

[0230] Hardware and software used

[0231] This system operates with three main components: a server, a terminal, and a user. The server manages sales and inventory data using a database management system (e.g., MySQL). Programs are developed using Python, with TensorFlow and PyTorch as key libraries for training and inference of AI models. Generative AI models include BERT, GPT-3 (registered trademark), and DALL-E. The terminal provides an interface for users to access the system through a web browser.

[0232] Program processing

[0233] The server first collects sales and inventory data. Specifically, it connects to a MySQL database and queries the sales history for the past six months and the current inventory status. Next, it analyzes the collected data and selects products that have been in stock for a long time. This analysis is performed using the Python pandas library.

[0234] Based on the selected products, the server generates coordinated outfits that go well together, even if the items are from different brands. It uses a generative AI model (e.g., GPT-3) to suggest product combinations. Images of models wearing the outfits, catchy slogans, and product descriptions are automatically created using a generative AI model (e.g., DALL-E or GPT-3).

[0235] Furthermore, the server automatically sets the sales price for the sales page using AI and posts the created sales page to social media, allowing consumers to purchase products via social media. The AI ​​also generates images of models wearing the products and product descriptions on the same platform, and sells consumers' used clothing. This encourages consumers to reuse their used clothing and achieves zero waste.

[0236] The server analyzes emotions from consumers' social media posts and suggests outfits that match those emotions. It uses NLP and image recognition technology to analyze emotions. Specifically, it uses the Hugging Face Transformers library and OpenCV.

[0237] Specific examples

[0238] Below is an example of a prompt sentence to input to the generative AI model.

[0239] 1. Analysis of sales and inventory data

[0240] Prompt: "Based on sales and inventory data from the past six months, please select the product that has been in stock for a long time."

[0241] 2. Coordinate Generation

[0242] Prompt: "Using the selected products, suggest a coordinated look that works well together, even if it's from a different brand."

[0243] 3. Model Wearing Images and Copy Generation

[0244] Prompt: "Generate a model image, tagline, and product description based on this outfit."

[0245] 4. Sentiment analysis and coordination

[0246] Prompt: "Analyze the emotions of consumers in their social media posts and suggest product coordination that matches those emotions."

[0247] Using these prompts, the system can automatically analyze the data and generate optimal outfits and sales pages.

[0248] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0249] Step 1: Collect sales and inventory data

[0250] The server collects sales and inventory data from the database. Specifically, it connects to the MySQL database and queries the sales history for the past six months and the current inventory status. The input requires database connection information and a query statement, and the output is sales and inventory data.

[0251] Specific behavior:

[0252] The server executes the SQL query SELECT FROM sales WHERE date >= DATE_SUB(CURDATE(), INTERVAL 6 MONTH).

[0253] The server executes the SQL query SELECT FROM inventory to obtain the current inventory status.

[0254] Step 2: Select long-term inventory items

[0255] The server analyzes the collected data and selects products that have been in long-term inventory. It uses the Python pandas library to identify products with sluggish sales. Sales data and inventory data are required as input, and a list of long-term inventory items is obtained as output.

[0256] Specific behavior:

[0257] The server uses the pandas groupby function to aggregate sales for each product.

[0258] The server filters out items whose sales are below a certain threshold.

[0259] Step 3: Generate coordinates

[0260] The server generates coordinated outfits based on the selected items, even for items from different brands, and uses a generative AI model (e.g., GPT-3) to suggest combinations of items. The input is a list of long-term inventory items, and the output is a suggested outfit.

[0261] Specific behavior:

[0262] The server inputs a prompt to the generative AI model: "Using the selected products, please suggest a coordinated look that goes well with other products from different brands."

[0263] The server receives the generated coordinates and stores them in a database.

[0264] Step 4: Generate model images and tagline

[0265] The server uses a generative AI model (e.g., DALL-E or GPT-3) to generate images of a model wearing coordinated products, a catchy slogan, and a product description. The input is a coordination suggestion, and the output is an image of a model wearing the product, a catchy slogan, and a product description.

[0266] Specific behavior:

[0267] The server inputs a prompt to the generative AI model: "Based on this outfit, please generate an image of the model wearing the outfit, a catchy slogan, and a product description."

[0268] The server receives the generated images and text and stores them as materials for the sales page.

[0269] Step 5: Create a sales page and set your sale price

[0270] The server creates a sales page based on the generated materials, and the AI ​​automatically sets the sale price. Pricing takes into account past sales data and market trends. The inputs include images of models wearing the products, a catchy slogan, and a product description, and the output is a completed sales page.

[0271] Specific behavior:

[0272] The server generates the sales page using the HTML template.

[0273] The server inputs a prompt statement to the AI ​​model: "Please set the optimal sale price for this product."

[0274] The server reflects the set price on the sales page.

[0275] Step 6: Post your sales page on social media

[0276] The server automatically posts the completed sales page to social media. The posting is done using the social media API. The input required is the sales page URL and a tagline, and the output is the completed posting to social media.

[0277] Specific behavior:

[0278] The server connects to the SNS API using an authentication token.

[0279] The server posts the sales page URL and the generated catchphrase to social media.

[0280] Step 7: Selling your used clothing

[0281] Users upload their own secondhand clothes to the platform. The server automatically generates images and descriptions of the uploaded secondhand clothes using a generative AI model. The input is an image of the secondhand clothes, and the output is an image of a model wearing the clothes and a product description.

[0282] Specific behavior:

[0283] Users upload images of used clothing.

[0284] The server inputs a prompt to the generative AI model: "Please generate an image of a model wearing this vintage clothing item and a product description."

[0285] The server reflects the generated images and text on the sales page.

[0286] Step 8: Sentiment analysis and coordination adjustment

[0287] The server analyzes emotions from consumers' social media posts and suggests outfits that match those emotions. Emotions are analyzed using NLP and image recognition technology. The input is the consumer's social media post, and the output is outfit suggestions that match the emotion.

[0288] Specific behavior:

[0289] The server retrieves the consumer's posts using the SNS API.

[0290] The server uses Hugging Face's Transformers library to analyze the sentiment of the text.

[0291] Based on the analysis results, the server inputs a prompt statement to the generative AI model: "Please suggest a product coordination that matches the consumer's emotions."

[0292] The server proposes the generated coordinates to the consumer.

[0293] (Application example 1)

[0294] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0295] In conventional apparel sales systems, long-term inventory items often remained unsold, making inventory management difficult. It was also difficult to propose coordinated outfits across brand boundaries, which prevented consumers from motivating them to purchase. Furthermore, product proposals did not take into account consumer emotions, making it impossible to provide optimal products to individual consumers. To solve these issues, efficient sales promotion of long-term inventory items and coordinated outfit proposals based on consumer emotions were necessary.

[0296] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0297] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for consumers to purchase via the information sharing service, means for generating images of models wearing the apparel items and product descriptions for consumers' used clothing on the same platform and selling them, means for analyzing consumer sentiment and adjusting coordination based on the sentiment, means for generating product pages using a generative AI model, and means for inputting prompts into the generative AI model to generate images and text. This enables efficient sales promotion of long-term inventory items and optimal coordination suggestions based on consumer sentiment.

[0298] "Long-term inventory" refers to products that have been stored as inventory for a certain period of time or longer and remain unsold in the normal sales cycle.

[0299] "Apparel products" refers to wearable products such as clothing and fashion accessories.

[0300] "Creating outfits that transcend brand boundaries" refers to combining products from different brands to create a unified fashion style.

[0301] "Model wearing images" refers to photographs or images of a model actually wearing the product.

[0302] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0303] "Product description" refers to text that explains the detailed information, features, and usage of a product.

[0304] "Sales Page" refers to a web page for selling products online.

[0305] "Sale price" refers to the selling price of a product.

[0306] "Information sharing services" refer to online platforms for sharing information, such as social media and blogs.

[0307] "Consumer" refers to an individual who purchases goods or services.

[0308] "Secondhand clothing" refers to previously used clothing or fashion items.

[0309] "Platform" refers to the underlying system for providing specific services or functions.

[0310] "Analyzing emotions" refers to reading people's emotions from text, images, videos, etc.

[0311] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to generate images or text.

[0312] A "prompt" refers to an instruction or question that is input into a generative AI model.

[0313] The system for implementing this invention is composed of three main elements: a server, a terminal, and a user. The specific operation of each element and the hardware and software used will be explained below.

[0314] Server Operation

[0315] The server includes programs to perform the following functions:

[0316] 1. Select long-term inventory items:

[0317] The server analyzes the inventory database and selects apparel items that have been in stock for a certain period of time, using an artificial intelligence (AI) model.

[0318] 2. Coordination suggestions that transcend brand boundaries:

[0319] The server then combines selected long-term inventory items across brands to suggest a cohesive fashion style, again using an AI model.

[0320] 3. Generate model images and product descriptions:

[0321] The server uses a generative AI model to generate images of a model wearing the coordinated products, a catchy copy, and a product description. The generative AI model generates images and text by inputting a prompt sentence.

[0322] 4. Sales page creation and automatic posting:

[0323] The server creates a sales page using the generated images of the model wearing the product, a catchy slogan, and a product description, and automatically posts it to an information sharing service (SNS).

[0324] 5. Consumer sentiment analysis:

[0325] The server analyzes the emotions of consumers in their social media posts and adjusts outfits based on those emotions. This process is performed using an emotion analysis engine.

[0326] Device behavior

[0327] The terminal (smartphone) provides an interface for users to view and purchase apparel products.

[0328] 1. Product browsing:

[0329] The user can view the sales page generated by the server through the terminal.

[0330] 2. Purchase procedure:

[0331] Users can purchase products through the terminal, and the purchase procedure is carried out through an electronic payment service.

[0332] User Actions

[0333] The user uses the terminal to browse and purchase apparel items.

[0334] 1. Social Media Posts:

[0335] Users express their feelings by posting on the SNS, and these posts are analyzed by the server and used to adjust coordination.

[0336] Hardware and software used

[0337] Hardware: Servers, smartphones

[0338] Software: Python, transformers library, requests library, sentiment analysis engine, generative AI model

[0339] Examples of specific examples and prompts

[0340] As a concrete example, the following is an example of a prompt sentence to be input to the generative AI model.

[0341] Sentiment analysis prompt: "Analyze the sentiment of the following text: 'I love this new dress!'"

[0342] Image generation prompt: "Generate an image of a Red Dress."

[0343] Text generation prompt: "Generate a catchy description for a Red Dress."

[0344] This allows the server to efficiently promote long-term inventory items and propose optimal outfits based on consumer sentiment.

[0345] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0346] Step 1: The server analyzes the inventory database to select apparel items that have become long-term inventory. It uses the inventory database as input and obtains a list of long-term inventory items as output. This process uses an artificial intelligence (AI) model to select items that have been in stock for more than a certain period of time (e.g., 180 days).

[0347] Step 2: The server coordinates the selected long-term inventory items across brands. It uses a list of long-term inventory items as input and obtains coordinated product combinations as output. This process also uses an AI model to combine products from different brands to suggest a cohesive fashion style.

[0348] Step 3: The server uses the generative AI model to generate an image of a model wearing the coordinated items, a catchy slogan, and a product description. It uses the coordinated product combination as input and obtains an image of a model wearing the items, a catchy slogan, and a product description as output. The generative AI model receives a prompt sentence as input and generates images and text. Examples of specific prompt sentences include "Generate an image of a Red Dress." and "Generate a catchy description for a Red Dress."

[0349] Step 4: The server creates a sales page using the generated model images, taglines, and product descriptions. The generated images and text are used as input, and a sales page is obtained as output. This sales page includes detailed product information, prices, a purchase button, and more.

[0350] Step 5: The server automatically sets the sale price for the sales page. It uses the sales page and market data as input and sets the optimal sale price as output. This process uses a pricing algorithm to determine the price based on competitor prices and demand forecasts.

[0351] Step 6: The server automatically posts the created sales page to an information sharing service (SNS). The sales page is used as input and the SNS post is made as output. This process uses the SNS API, and the posted content includes an image of the model wearing the product, a catchy slogan, and a product description.

[0352] Step 7: The user browses the sales page generated by the server through their terminal and purchases the product. The sales page is used as input and the purchase procedure is performed as output. The purchase procedure is performed through an electronic payment service.

[0353] Step 8: The server analyzes the sentiment of the consumer's social media posts and adjusts the outfit based on that sentiment. It uses the social media post data as input and obtains the sentiment analysis results as output. This process uses a sentiment analysis engine to read emotions from text, images, and videos. An example of a specific prompt is "Analyze the sentiment of the following text: 'I love this new dress!'"

[0354] Step 9: The server also generates images of models wearing the consumer's used clothing and product descriptions for sale on the same platform. Used clothing information is used as input, and images of models wearing the clothing and product descriptions are obtained as output. This process uses a generative AI model, which generates images and text based on prompts.

[0355] Example 2

[0356] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0357] Conventional apparel sales systems were unable to efficiently process long-term inventory, resulting in backlogged inventory and declining sales. Furthermore, it was difficult to propose outfits that transcended brand boundaries, preventing consumers from purchasing. Furthermore, product suggestions were not based on consumer sentiment, making it difficult to provide services that meet the needs of individual consumers. To solve these issues, efficient inventory management utilizing AI technology and outfit suggestions based on consumer sentiment were required.

[0358] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0359] In this invention, the server includes means for selecting long-term inventory items, means for coordinating the selected items across brand boundaries, means for generating images of the coordinated items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the items worn by models, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing platform, means for consumers to purchase via the information sharing platform, means for generating images of the items worn by models and product descriptions and selling consumers' used items on the same platform, and means for analyzing consumer sentiment and adjusting product coordination based on the analysis results. This enables efficient processing of long-term inventory items, attractive coordination suggestions across brand boundaries, and individual responses based on consumer sentiment.

[0360] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0361] "Goods" means goods sold to consumers.

[0362] "Brand" refers to a name or logo that is used to identify the products or services offered by a particular company or manufacturer.

[0363] "Coordination" refers to proposing a combination of multiple products.

[0364] "Model Wearing Image" refers to a photograph of a model wearing a product.

[0365] A "catchphrase" is a short sentence that succinctly expresses the features and appeal of a product.

[0366] "Product description" refers to text that explains the detailed information and features of a product.

[0367] "Sales Page" means a web page for selling products online.

[0368] "Sale Price" refers to the price at which an item is sold.

[0369] An "information sharing platform" refers to online services for sharing information, such as social media and websites.

[0370] "Consumer" refers to an individual who purchases a product.

[0371] "Used goods" refers to goods that have been previously used.

[0372] "Emotion" refers to the psychological state of the consumer.

[0373] "Analysis" refers to analyzing data to extract meaning.

[0374] "Artificial intelligence" refers to the technology that allows computers to act in a way that mimics human intelligence.

[0375] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[0376] The server first collects sales and inventory data from the database, using SQL queries to retrieve past sales history and current inventory status. It then uses Python and TensorFlow to analyze the collected data and identify slowing sales. This allows it to select products that have been in stock for a long time.

[0377] For the selected products, the server uses a generative AI model (for example, OpenAI's GPT-4) to generate a coordinated outfit that transcends brand boundaries. Specifically, the following prompt sentence is input into the generative AI model:

[0378] "This item has been in stock for a long time, so please coordinate it with other brand items. Please generate a model image, tagline, and product description."

[0379] Based on the generated outfit information, the server uses an HTML template to create a sales page, which includes images of the product being worn by a model, a tagline, and a product description. Furthermore, a pricing model is built using Python and Scikit-learn to automatically set the sale price for the sales page.

[0380] The created sales page is automatically posted to an information-sharing platform using SNS APIs, allowing consumers to purchase products via the information-sharing platform.

[0381] Users can also sell their own second-hand goods on the platform. When a user uploads an image of their used clothing, the server uses a generative AI model to generate an image of the model wearing the item and a product description. Specifically, the following prompt sentence is input into the generative AI model:

[0382] "Please generate an image of a model wearing this vintage clothing along with a product description."

[0383] Furthermore, the server uses NLP technology (e.g., the BERT model) and image recognition technology (e.g., OpenCV) to analyze consumer sentiment. Based on the analysis results, it coordinates products to match the consumer's sentiment. For example, if a consumer posts many times expressing happy emotions, it selects and coordinates products in bright colors.

[0384] In this way, it becomes possible to efficiently process long-term inventory items, propose attractive coordination that transcends brand boundaries, and respond to individual customer needs based on consumer sentiment.

[0385] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0386] Step 1: The server collects sales and inventory data from the database. The input is the database connection information and an SQL query. Specifically, the server executes the SQL query to obtain past sales history and current inventory status. The output is sales and inventory data.

[0387] Step 2: The server analyzes the collected sales and inventory data to identify products that have become long-term inventory. The server uses sales and inventory data as input. Specifically, the server uses Python and TensorFlow to predict sales trends and identify products with slowing sales. The output is a list of long-term inventory products.

[0388] Step 3: The server uses the generative AI model to generate a cross-brand coordination based on the long-term inventory product list. The long-term inventory product list and a prompt sentence are used as input. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0389] "This item has been in stock for a long time, so please coordinate it with other brand items. Please generate a model image, tagline, and product description."

[0390] As an output, coordinate information is obtained.

[0391] Step 4: The server creates a sales page based on the generated coordination information. The coordination information and HTML template are used as input. Specifically, the server embeds an image of a model wearing the item, a catchy slogan, and a product description into the HTML template to generate the sales page. The sales page is obtained as output.

[0392] Step 5: The server automatically sets the sale price for the sales page. The sales page and pricing model are used as input. Specifically, the server uses Python and Scikit-learn to build a pricing model and predict the sale price for the sales page. The output is the sales page with the sale price set.

[0393] Step 6: The server automatically posts the created sales page to the information sharing platform. The sales page and the information sharing platform's API are used as input. Specifically, the server posts the sales page using the information sharing platform's API. The output is the sales page posted on the information sharing platform.

[0394] Step 7: The user uploads an image of their used clothing to sell on the platform. The image of the used clothing is used as input. Specifically, the server uses the generative AI model to generate an image of the model wearing the clothing and a product description. The following prompt sentence is input into the generative AI model:

[0395] "Please generate an image of a model wearing this vintage clothing along with a product description."

[0396] The output is an image of a model wearing the product and a product description.

[0397] Step 8: The server analyzes the consumer's sentiment and adjusts product coordination based on the analysis results. The input is the consumer's social media posting data. Specifically, the server analyzes sentiment using NLP technology (e.g., the BERT model) or image recognition technology (e.g., OpenCV). The output is the sentiment analysis result. Based on the analysis results, the server inputs the following prompt sentence into the generative AI model:

[0398] "This consumer often posts expressing joy, so choose apparel items in bright colors and coordinate them together."

[0399] As an output, emotion-based coordination information is obtained.

[0400] (Application example 2)

[0401] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0402] In the conventional apparel sales system, there was a lack of efficient ways to sell long-term inventory, making inventory management difficult. It was also difficult to propose coordination that transcended the boundaries of brands, which prevented consumers from motivating purchases. Furthermore, product proposals were not based on consumer emotions, making it impossible to provide the best products for each individual consumer.

[0403] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0404] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for consumers to purchase via the information sharing service, means for generating images of the consumers wearing the apparel items on models and product descriptions and selling their used clothing on the same platform, means for analyzing consumer sentiment, and means for making product suggestions based on the analyzed sentiment. This enables efficient sales of long-term inventory items, coordination suggestions across brands, and optimal product suggestions based on consumer sentiment.

[0405] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0406] "Apparel products" refers to clothing items such as clothing and fashion accessories.

[0407] "Crossing brand boundaries" refers to combining products from different brands.

[0408] "Coordination" refers to styling multiple apparel items together.

[0409] "Model Wearing Image" refers to a photograph of a model wearing an apparel product.

[0410] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0411] "Product description" refers to text that explains the detailed information and features of a product.

[0412] "Sales page" refers to a web page that lists product information and allows consumers to purchase the product.

[0413] "Sale price" refers to the selling price of a product.

[0414] "Information sharing services" refers to online platforms such as social networking sites.

[0415] "Consumer" refers to an individual who purchases a product.

[0416] "Used clothing" refers to clothing that has been previously purchased and used by a consumer.

[0417] "Platform" refers to a system that serves as the foundation for providing multiple functions and services.

[0418] "Analyzing emotions" refers to reading consumer emotions from data.

[0419] "Product suggestion" refers to recommending a specific product to a consumer.

[0420] "Artificial intelligence" refers to technology that allows machines to act in a way that mimics human intelligence.

[0421] A system for implementing the present invention has the following configuration.

[0422] First, the server selects apparel items that have become long-term inventory. This is achieved by analyzing sales and inventory data. Specifically, the server retrieves sales and inventory data from the database and selects items that have remained unsold for a certain period of time as long-term inventory items.

[0423] Next, the selected apparel items are coordinated across brands. The server uses artificial intelligence to combine products from different brands and propose optimal outfits. For example, it uses a clustering algorithm to group products based on color or design characteristics and find combinations that go well together.

[0424] Images of models wearing coordinated apparel products, catch phrases, and product descriptions are automatically generated using a generative AI model. The server inputs the following prompt sentence into the generative AI model and obtains the generated text.

[0425] Example prompt sentence:

[0426] Product Name: Refreshing Summer Dress

[0427] Description: This dress features a lightweight material and vibrant color. It's designed to be comfortable even on hot summer days, making it perfect for relaxing at the beach or a cafe.

[0428] The server creates a sales page using the generated images of models wearing the product, catchy slogans, and product descriptions. The server automatically sets the sale price on the sales page. This is achieved by calculating the optimal price based on past sales data and market trends.

[0429] The created sales page is automatically posted to the information sharing service by the server, allowing consumers to purchase the product via the information sharing service.

[0430] Consumers can also sell their used clothing on the same platform. Based on the information provided by consumers, the server uses a generative AI model to generate images of models wearing the clothing and product descriptions, and creates a sales page.

[0431] To analyze consumer sentiment, the server collects social media posts from consumers and analyzes them using a sentiment analysis algorithm. Based on the analyzed sentiment, the server makes optimal product recommendations to consumers. For example, if a consumer posts frequently expressing joy, the server will recommend brightly colored apparel products.

[0432] In this way, the server enables efficient sales of long-term inventory items, coordination suggestions that transcend brand boundaries, and optimal product suggestions based on consumer sentiment.

[0433] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0434] Step 1: The server retrieves sales and inventory data from the database. This collects basic data for selecting apparel items that have become long-term inventory. The input is sales and inventory data, and the output is the integration result of these data.

[0435] Step 2: The server analyzes the acquired data and selects items that have remained unsold for a certain period of time as long-term inventory items. Specifically, it filters out items with sales below a certain level. The input is the integrated sales data and inventory data, and the output is a list of long-term inventory items.

[0436] Step 3: The server uses artificial intelligence to coordinate the selected long-term inventory items across brands. It uses a clustering algorithm to group products based on color and design characteristics and find combinations that go well together. The input is a list of long-term inventory items, and the output is a list of coordinated items.

[0437] Step 4: The server uses the generative AI model to generate images of coordinated apparel items worn by a model, a tagline, and a product description. The prompt text is input into the generative AI model and the generated text is obtained. The input is a list of coordinated items and the prompt text, and the output is the generated images of the items worn by a model, a tagline, and a product description.

[0438] Step 5: The server creates a sales page using the generated model images, catchphrases, and product descriptions. The input is the generated content, and the output is the completed sales page.

[0439] Step 6: The server automatically sets the sale price on the sales page. It calculates the optimal price based on past sales data and market trends. The input is the sales page and past sales data, and the output is the set sale price.

[0440] Step 7: The server automatically posts the created sales page to the information sharing service, allowing consumers to purchase the product via the information sharing service. The input is the completed sales page, and the output is the result of posting to the information sharing service.

[0441] Step 8: The server uses the generative AI model to generate images of a model wearing the item and a product description based on the consumer's used clothing information, and creates a sales page. The input is the consumer's used clothing information and prompt text, and the output is the generated image of a model wearing the item and a product description.

[0442] Step 9: The server collects the consumer's SNS posts and analyzes their emotions using a sentiment analysis algorithm. The input is the consumer's SNS posts, and the output is the analyzed sentiment data.

[0443] Step 10: The server makes optimal product recommendations to the consumer based on the analyzed emotions. For example, if a consumer posts a lot of happy emotions, it will suggest brightly colored apparel products. The input is the analyzed emotion data, and the output is a list of suggested products.

[0444] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0445] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0446] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0447] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0448] [Second embodiment]

[0449] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0450] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0451] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0452] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0453] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0454] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0455] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0456] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0457] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0458] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0459] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0460] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0461] "Example 1"

[0462] One embodiment of the present invention is an apparel sales system using AI. In this system, AI first selects apparel items that have been in stock for a long time. For example, it analyzes sales and inventory data and automatically selects items with slowing sales. Next, it coordinates the selected items across brands. This makes it possible to suggest combinations that work well together, even among products from different brands. Images of the coordinated items worn by models, catchy slogans, and product descriptions are automatically generated by the AI. This enables the efficient creation of sales pages that maximize the appeal of the products. Furthermore, the AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media. Furthermore, AI also generates images of models wearing the items and product descriptions for the consumer's used clothing on the same platform and sells them. This encourages consumers to reuse their used clothing and achieves zero waste.

[0463] "Example 2"

[0464] One embodiment of the present invention is an apparel sales system using AI. In this system, AI first selects apparel items that have been in stock for a long time. For example, it analyzes sales and inventory data and automatically selects items with slowing sales. Next, it coordinates the selected items across brands. This makes it possible to suggest combinations that work well together, even among products from different brands. Images of the coordinated items worn by models, catchy slogans, and product descriptions are automatically generated by the AI. This enables the efficient creation of sales pages that maximize the appeal of the products. Furthermore, the AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media. Furthermore, AI also generates images of models wearing the items and product descriptions for the consumer's used clothing on the same platform and sells them. This encourages consumers to reuse their used clothing and achieves zero waste.

[0465] The processing flow of each embodiment will be described below.

[0466] "Example 1"

[0467] Step 1: The AI ​​selects apparel items that have been in stock for a long time. Specifically, it analyzes sales and inventory data and automatically selects items whose sales are slowing.

[0468] Step 2: Coordinate the selected products across brands, making it possible to propose combinations that go well together even when products from different brands are used.

[0469] Step 3: Images of models wearing the coordinated products, along with catchy slogans and product descriptions, are automatically created by generative AI. This makes it possible to efficiently create sales pages that maximize the appeal of the products.

[0470] Step 4: The AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase the product via social media.

[0471] Step 5: Consumers can also purchase their used clothing on the same platform, with AI generating images of models wearing the clothing and product descriptions. This will encourage consumers to reuse their used clothing and achieve zero waste.

[0472] Example 1

[0473] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0474] Conventional apparel sales systems lacked efficient methods for selling long-term inventory. Furthermore, there were no effective methods for suggesting outfits that combined products from different brands, or for reusing consumers' old clothes. This resulted in increased inventory backlog and waste, and reduced sales efficiency.

[0475] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0476] In this invention, the server includes means for collecting sales data and inventory data, means for analyzing the collected data to select long-term inventory items, means for proposing outfits by combining the selected items with items from different brands, means for generating images of the coordinated items worn by models, catch phrases, and product descriptions using a generative AI model, means for creating a sales page using the generated images of the coordinated items worn by models, catch phrases, and product descriptions, means for automatically setting sale prices by executing a pricing algorithm based on past sales data, means for automatically posting the created sales page to social media, means for allowing consumers to purchase via social media, and means for generating images of the items worn by models and product descriptions and selling consumers' used clothing on the same platform. This makes it possible to efficiently sell long-term inventory items, propose attractive outfits by combining items from different brands, and encourage consumers to reuse used clothing.

[0477] "Sales data" refers to data that includes transaction information regarding the sale of apparel products.

[0478] "Inventory data" is data that indicates the current inventory status of apparel products.

[0479] "Long-term inventory" refers to apparel products that remain unsold for a certain period of time.

[0480] "AI" is a system that uses artificial intelligence technology to analyze data and make predictions.

[0481] "Coordination" is a fashion combination that combines products from different brands.

[0482] A "generative AI model" is an artificial intelligence model for generating natural language and images.

[0483] "Model wearing image" refers to an image of a model wearing an apparel product.

[0484] A "catchphrase" is a short promotional phrase that draws out the appeal of a product.

[0485] A "product description" is a text that explains the features and benefits of a product.

[0486] A "sales page" is a web page for selling apparel products online.

[0487] "Sale price" refers to the selling price of a product.

[0488] "SNS" refers to social networking services, which are online platforms for users to share information.

[0489] "Consumer" refers to a customer who purchases apparel products.

[0490] "Used clothing" refers to second-hand apparel items that consumers have previously purchased and wish to resell.

[0491] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[0492] The server first collects sales and inventory data. To do this, MySQL is used as the database management system (DBMS). The server retrieves sales data for the past six months and current inventory data using SQL queries.

[0493] The server then analyzes the collected data and selects items that have become long-term inventory. This analysis is performed using Python and TensorFlow. Items whose sales have fallen below a certain threshold are selected as long-term inventory. For example, items whose sales have been less than 10 units in the past six months are selected.

[0494] For the selected items, the server proposes outfits by combining them with products from different brands. To propose outfits, OpenCV is used to evaluate color compatibility and Scikit-learn is used to evaluate style compatibility. Specifically, a color wheel is used to evaluate color compatibility, and a clustering algorithm is used to evaluate style compatibility.

[0495] Next, the server uses a generative AI model (e.g., GPT-4) to generate an image of a model wearing the coordinated items, along with a catchy slogan and product description. The following prompt is input to the generative AI model:

[0496] "This is a long-term inventory item. Please generate a model image, tagline, and product description based on the following information: Product name: XXXX, Brand: YYYY, Color: ZZZZ, Style: AAAA"

[0497] Based on the generated information, the server uses an image generation API (e.g., DALL-E) to generate an image of the model wearing the product.

[0498] The server then runs a pricing algorithm based on past sales data to determine the optimal sale price. To do this, it uses XGBoost to build a pricing model and predict the optimal sale price for each product.

[0499] Next, the server creates a sales page based on the generated model images, tagline, product description, and sale price. The sales page is built using HTML and CSS. The link to the created sales page and the generated tagline are automatically posted to social media using an API.

[0500] Additionally, users can input photos of used clothing and basic information (brand, color, style, etc.) from their devices. The server uses this information to generate an image of the clothing being worn by a model and a product description using a generative AI model. For example, the following prompt sentence can be input into the generative AI model:

[0501] "This used clothing item is for sale by a user. Please generate a model image, tagline, and product description based on the following information: Product name: XXXX, Brand: YYYY, Color: ZZZZ, Style: AAAA"

[0502] Based on the generated information, the server creates a sales page and automatically posts it to social media, allowing users to easily sell their used clothing.

[0503] In this way, the system of the present invention makes it possible to efficiently sell long-term inventory items, propose attractive outfits that combine products from different brands, and encourage consumers to reuse used clothing.

[0504] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0505] Step 1: The server collects sales and inventory data. Specifically, the server connects to a MySQL database and retrieves sales data for the past six months and current inventory data using an SQL query. The input is the database connection information and query, and the output is sales data and inventory data.

[0506] Step 2: The server analyzes the collected data and selects items that have become long-term inventory. Specifically, the server uses Python and TensorFlow to analyze sales data and selects items whose sales fall below a certain threshold as long-term inventory. The inputs are sales data and inventory data, and the output is a list of long-term inventory items.

[0507] Step 3: The server combines the selected long-term inventory items with products from different brands to suggest outfits. Specifically, the server evaluates color compatibility using OpenCV and style compatibility using Scikit-learn. The input is a list of long-term inventory items and product data from other brands, and the output is a list of outfit suggestions.

[0508] Step 4: The server uses the generative AI model to generate images of the coordinated products worn by a model, a catchphrase, and a product description. Specifically, the server inputs a prompt to the generative AI model and obtains the generated text and images. The input is a list of outfit suggestions and a prompt, and the output is an image of the product worn by a model, a catchphrase, and a product description.

[0509] Step 5: The server creates a sales page using the generated model image, tagline, and product description. Specifically, the server builds the sales page using HTML and CSS. The input is the model image, tagline, and product description, and the output is an HTML file for the sales page.

[0510] Step 6: The server runs a pricing algorithm based on past sales data to automatically set the sale price. Specifically, the server uses XGBoost to build a pricing model and predict the optimal sale price for each product. The input is past sales data, and the output is the sale price.

[0511] Step 7: The server automatically posts the created sales page to social media. Specifically, the server uses an API to post the sales page link and the generated tagline. The input is the sales page link and tagline, and the output is the result of posting to social media.

[0512] Step 8: The user enters a photo of the used clothing and basic information (brand, color, style, etc.) from the device. Specifically, the user enters this information using a web form. The input is a photo of the used clothing and basic information, and the output is the data sent to the server.

[0513] Step 9: The server uses the generative AI model to generate an image of a model wearing the item and a product description based on the information entered by the user. Specifically, the server inputs a prompt into the generative AI model and obtains the generated text and image. The input is a photo of the used clothing, basic information, and the prompt, and the output is an image of a model wearing the item and a product description.

[0514] Step 10: The server creates a sales page based on the generated information and automatically posts it to social media. Specifically, the server uses HTML and CSS to build the sales page and posts it using an API. The input is an image of the model wearing the product, a product description, and a link to the sales page, and the output is the result of posting to social media.

[0515] (Application example 1)

[0516] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0517] Conventional apparel sales systems have faced the problem of making it difficult to efficiently sell long-term inventory items, requiring a great deal of effort for inventory management and sales promotion. Furthermore, they lacked systems for suggesting outfits that combine products from different brands or for reusing consumers' old clothing, making it impossible to provide an attractive purchasing experience for consumers. Furthermore, they lacked the ability to automatically generate product descriptions using generative AI models or automatically post them to information sharing services, making sales page creation and promotional activities inefficient.

[0518] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for selecting long-term inventory apparel items, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for enabling consumers to purchase the items via the information sharing service, means for generating images of the items worn by models and product descriptions for consumers' used clothing on the same platform and selling them, means for analyzing sales data and inventory data, means for proposing coordination by combining products from different brands, means for generating product descriptions using a generative AI model, and means for posting the generated product descriptions to the information sharing service. This enables efficient sales of long-term inventory items, attractive coordination suggestions combining products from different brands, reuse of consumers' used clothing, and efficient creation of sales pages and promotional activities using the automatic generation function utilizing the generative AI model.

[0519] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0520] "Apparel products" refers to fashion-related products such as clothing and fashion accessories.

[0521] "Creating coordination that transcends brand boundaries" refers to proposing styling by combining products from different brands.

[0522] "Model wearing image" refers to an image of a model actually wearing the product.

[0523] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0524] "Product description" refers to text that explains the detailed information and features of a product.

[0525] "Sales page" refers to a web page that lists product information and allows consumers to purchase the product.

[0526] "Sale price" refers to the selling price of a product.

[0527] "Information sharing services" refers to online platforms such as social networking sites.

[0528] "Used consumer clothing" refers to second-hand clothing that is owned by a consumer.

[0529] "Platform" refers to the underlying system for providing specific services or functions.

[0530] "Sales Data" refers to data regarding product sales performance.

[0531] "Inventory Data" refers to data regarding the availability of products.

[0532] A "generative AI model" refers to a model that uses artificial intelligence to generate text and images.

[0533] "Prompt" refers to a text instruction that is input to a generative AI model.

[0534] A system for implementing the present invention includes a server, a user terminal, and a generative AI model. A specific embodiment of the system will be described below.

[0535] Server Processing

[0536] The server is equipped with a database for analyzing sales and inventory data and a data analysis module using artificial intelligence (AI). The server integrates sales and inventory data and automatically selects apparel items that have been in stock for a long time. The selected items are then analyzed using an AI clustering algorithm to suggest outfits that can be combined with products from different brands.

[0537] Next, the server uses the generative AI model to automatically generate images of the coordinated items worn by models, taglines, and product descriptions. The generated content is used to create a sales page. The sale price on the sales page is automatically set by the AI.

[0538] Furthermore, the server has the function of automatically posting the generated sales page to information sharing services (such as SNS), allowing consumers to purchase products through the information sharing service.

[0539] User terminal processing

[0540] The user device provides an interface for consumers to browse and purchase products. Users can check product information and complete the purchase process through the sales page. Users can also upload their own used clothing to the platform, and images of models wearing the clothing and product descriptions can be automatically generated using a generative AI model. This allows users to reuse and sell their used clothing.

[0541] Processing generative AI models

[0542] A generative AI model is an artificial intelligence model for generating text and images. Specifically, it uses natural language processing models such as GPT-2 to generate product descriptions and catchy slogans. Image generation models (e.g., GANs) are used to generate images of models wearing the products.

[0543] Prompt Sentence Examples

[0544] The following text will be used as an example of a prompt sentence to be input into the generative AI model.

[0545] Product Name: Slim Fit Jeans, Features: Stretch Material, Dark Blue, Casual

[0546] By inputting this prompt, the generative AI model generates the following product description:

[0547] These slim-fit jeans are made from a stretchy fabric for ease of movement and comfort, and their dark blue color is perfect for a casual look and pairs well with any top.

[0548] In this way, by combining the server, user terminal, and generative AI model, an efficient and attractive apparel product sales system can be realized.

[0549] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0550] Step 1: The server retrieves sales data and inventory data from the database. This data is integrated to select apparel items that have become long-term inventory. Specifically, products whose sales have been sluggish for a certain period of time are filtered out and listed as long-term inventory items. The input is sales data and inventory data, and the output is a list of long-term inventory items.

[0551] Step 2: The server proposes outfits by combining the selected long-term inventory items with products from different brands. It uses an AI clustering algorithm to analyze product characteristics and group items that go well together. The input is a list of long-term inventory items, and the output is a list of outfit suggestions.

[0552] Step 3: The server uses the generative AI model to automatically generate images of the coordinated items worn by a model, a catchphrase, and a product description. Specifically, the server inputs a prompt into the generative AI model and retrieves the generated text and images. The input is a list of outfit suggestions and the prompt, and the output is an image of the item worn by a model, a catchphrase, and a product description.

[0553] Step 4: The server creates a sales page using the generated model images, tagline, and product description. The layout and design of the sales page are automatically generated and product information is placed on it. The input is the model images, tagline, and product description, and the output is the completed sales page.

[0554] Step 5: The server automatically sets the sale price for the sales page using AI. It analyzes past sales data and market trends to calculate the optimal price. The input is the sales page and sales data, and the output is a sales page with the sale price set.

[0555] Step 6: The server automatically posts the created sales page to an information sharing service (such as SNS). Using an API, it posts the sales page link and product information. The input is the sales page with the set sale price, and the output is a notification that posting to the information sharing service has been completed.

[0556] Step 7: The user browses and purchases the product through the information sharing service. The user clicks on the link to the sales page and checks the product information. The input is the post on the information sharing service, and the output is the user's purchase procedure.

[0557] Step 8: The user uploads their own secondhand clothing to the platform, and the generative AI model automatically generates an image of the clothing being worn by a model and a product description. The user's device inputs the image and characteristics of the secondhand clothing and sends a prompt to the generative AI model. The input is the image and characteristics of the secondhand clothing, and the output is an image of the clothing being worn by a model and a product description.

[0558] Step 9: The server uses the generated model image and product description to create a sales page for the used clothing and posts it to the information sharing service. The input is the model image and product description, and the output is the sales page for the used clothing and a notification that posting to the information sharing service has been completed.

[0559] Example 2

[0560] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0561] Conventional apparel sales systems have difficulty efficiently selling products that have been in stock for a long time, and lack mechanisms for suggesting outfits that combine products from different brands or encouraging consumers to reuse used clothing. This leads to issues such as increased inventory backlog and waste, and reduced sales efficiency.

[0562] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0563] In this invention, the server includes means for collecting and analyzing sales data and inventory data, means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information transmission medium, means for consumers to purchase via the information transmission medium, and means for generating images of models wearing the apparel items and product descriptions and selling consumers' used clothing on the same platform. This enables efficient sales of long-term inventory items, suggestions for coordinating outfits that combine products from different brands, and promotion of consumers' reuse of used clothing.

[0564] "Sales data" is data that includes information regarding sales of apparel products.

[0565] "Inventory data" refers to data that includes information regarding the inventory status of apparel products.

[0566] "Long-term inventory" refers to apparel products that remain unsold for a certain period of time.

[0567] "Artificial intelligence" is a technology that uses machine learning and data analysis to automatically make judgments and predictions.

[0568] "Coordination" refers to proposing styling by combining multiple apparel items.

[0569] A "model wearing image" is an image of a model wearing an apparel product.

[0570] A "catchphrase" is a short phrase that succinctly conveys the appeal of a product.

[0571] A "product description" is a detailed description of the product's features and benefits.

[0572] A "sales page" is a web page for selling apparel products online.

[0573] "Sale price" refers to the selling price of a product.

[0574] An "information dissemination medium" is a platform for widely disseminating information, such as social media or websites.

[0575] "Consumers" refers to general customers who purchase apparel products.

[0576] "Used clothing" refers to apparel products that have been previously purchased and used by consumers.

[0577] The "platform" is a system that serves as the foundation for selling apparel products and disseminating information.

[0578] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[0579] First, the server collects and analyzes sales and inventory data. For this, a database management system (e.g., MySQL) is used. The data is read using the Python pandas library and converted into a data frame. Next, the sales and inventory data are analyzed and time series analysis is performed to identify sales trends.

[0580] Based on the analysis results, the server selects apparel items that have been in long-term inventory. Specifically, it selects items whose sales over the past six months are below a certain threshold. This selection is made using artificial intelligence (AI).

[0581] The server then coordinates the selected items across brands, using a generative AI model (e.g., OpenAI's GPT-4) to provide a prompt and receive coordination suggestions.

[0582] The server then uses the generative AI model to automatically generate images of the coordinated products worn by the model, a catchphrase, and a product description. DALL-E is used for image generation, and GPT-4 is used for generating the catchphrase and product description.

[0583] In addition, the server uses machine learning algorithms (e.g., scikit-learn) to automatically set the sale price on the sales page, calculating the optimal price based on past sales data.

[0584] The server creates a sales page based on the generated model images, catchphrases, product descriptions, and sale prices. The created sales page is then automatically posted to information sharing media (e.g., social media), allowing consumers to purchase the product via the information sharing media.

[0585] Users can also upload product information to the platform to sell their used clothing. The server then uses a generative AI model to automatically generate images of models wearing the clothing and product descriptions, creating a sales page. This encourages users to reuse their used clothing and achieve zero waste.

[0586] As a concrete example, the following prompt sentence is input into the generative AI model:

[0587] "Analyze sales data from the past six months and select apparel items that are experiencing slower sales. Coordinate the selected items with products from different brands, generate model images, taglines, and product descriptions, set a sale price for the sales page, and automatically post to social media."

[0588] Using this prompt, the generative AI model can perform the specified task and efficiently create a sales page.

[0589] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0590] Step 1: The server collects and analyzes sales and inventory data. As input, it obtains sales and inventory data from a database management system (e.g., MySQL). Specifically, it reads the data using Python's pandas library and converts it into a data frame. As output, it performs time series analysis to obtain analytical results and identify sales trends.

[0591] Step 2: The server selects apparel items that have become long-term inventory based on the analysis results. The analysis results obtained in Step 1 are used as input. Specifically, it selects items whose sales over the past six months are below a certain threshold. The output is a list of selected long-term inventory items.

[0592] Step 3: The server coordinates the selected products across brands. It uses the list of long-term inventory items obtained in Step 2 as input. Specifically, it inputs a prompt to a generative AI model (e.g., OpenAI's GPT-4) and receives coordination suggestions. The output is a list of coordinated products.

[0593] Step 4: The server uses the generative AI model to automatically generate images of the coordinated items worn by the model, taglines, and product descriptions. The list of coordinated items obtained in Step 3 is used as input. Specifically, DALL-E is used to generate images, and GPT-4 is used to generate taglines and product descriptions. The output is an image of the model wearing the items, taglines, and product descriptions.

[0594] Step 5: The server uses a machine learning algorithm (e.g., scikit-learn) to automatically set the sale price for the sales page. It uses past sales data and the product information obtained in Step 4 as input. Specifically, it trains a machine learning model to calculate the optimal price for the new product. The output is the sale price.

[0595] Step 6: The server creates a sales page based on the generated model image, catchphrase, product description, and sale price. The product information obtained in step 4 and the sale price obtained in step 5 are used as input. Specifically, the server generates a web page using HTML and CSS. The completed sales page is obtained as output.

[0596] Step 7: The server automatically posts the created sales page to the information dissemination medium. The sales page obtained in step 6 is used as input. Specifically, it posts to the social networking site using an API. The output is the sales page published on the social networking site.

[0597] Step 8: The user uploads product information to the platform to sell their used clothing. The input is an image of the used clothing and basic information provided by the user. Specifically, the server uses the generative AI model to automatically generate an image of a model wearing the clothing and a product description, and creates a sales page. The output is a sales page for the used clothing.

[0598] (Application example 2)

[0599] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0600] With conventional apparel sales systems, it was difficult to efficiently sell long-term inventory items, and it was also difficult to suggest outfits that combined products from different brands. Furthermore, it was risky for consumers to make a purchase decision without actually trying on the product, which discouraged them from purchasing. There was a need to solve these issues and improve the consumer purchasing experience.

[0601] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0602] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an SNS, means for consumers to purchase via the SNS, means for generating images of the consumers' used clothing worn by models and product descriptions on the same platform and selling them, means for applying coordination to a user's avatar and performing a virtual try-on, and means for consumers to purchase items through the virtual try-on. This enables efficient sales of long-term inventory items, suggestions for coordinating outfits that combine items from different brands, and an improved consumer purchasing experience through the virtual try-on.

[0603] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0604] "Apparel products" refers to fashion-related products such as clothing and accessories.

[0605] "Coordinating across brand boundaries" refers to proposing styling by combining products from different brands.

[0606] "Model wearing image" refers to an image of a model actually wearing the product.

[0607] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0608] "Product description" refers to text that explains the detailed information and features of a product.

[0609] "Sales Page" refers to a web page for selling products online.

[0610] "Sale price" refers to the selling price of a product.

[0611] "SNS" is an abbreviation for social networking service, which refers to an online platform for users to share information.

[0612] "Consumer" refers to the general user who purchases the product.

[0613] "Used clothing" refers to clothing that has been previously used.

[0614] "Platform" refers to the underlying system for providing specific services or functions.

[0615] An "avatar" refers to a character or image used to represent a user in a virtual space.

[0616] "Virtual try-on" refers to users trying on products in a virtual space.

[0617] The system for implementing this invention has the following configuration. First, the server analyzes sales data and inventory data to select apparel items that have been in stock for a long time. This makes it possible to automatically select items whose sales are slowing. Next, AI is used to coordinate the selected apparel items across brand boundaries. This makes it possible to suggest combinations that go well together, even when products from different brands are combined.

[0618] The server uses a generative AI model to generate images of models wearing coordinated apparel, as well as catchy slogans and product descriptions. A sales page is then created using the generated images of models wearing the apparel, catchy slogans and product descriptions. The AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media as well.

[0619] In addition, consumers can also sell their used clothing on the same platform, generating images of models wearing the clothing and product descriptions, which will encourage consumers to reuse their used clothing and achieve zero waste.

[0620] Users can use smartphones, smart glasses, or head-mounted displays to virtually try on products. Coordination is applied to the user's avatar, allowing them to check out products through virtual try-on. This allows users to consider purchasing products without actually trying them on.

[0621] Specifically, the server uses software such as TensorFlow, OpenCV, and PIL (Python Imaging Library) to process data and perform calculations, such as loading product data, generating outfits, and running virtual try-ons.

[0622] An example of a prompt is as follows:

[0623] "Please select apparel items that have been in stock for a long time and coordinate them with items from different brands."

[0624] In this way, it will be possible to efficiently sell long-term inventory items, suggest outfits that combine products from different brands, and improve the consumer purchasing experience through virtual try-on.

[0625] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0626] Step 1: The server analyzes sales and inventory data to select apparel items that have become long-term inventory. Sales and inventory data are required as input, and a list of long-term inventory items is obtained as output. Specifically, an AI algorithm is used to automatically extract items that have remained unsold for a certain period of time.

[0627] Step 2: The server coordinates the selected apparel items across brands. It requires a list of long-term inventory items as input, and produces a list of coordinated items as output. Specifically, it uses an AI model to combine items from different brands and generate coordinated outfits that go well together.

[0628] Step 3: The server generates images of the coordinated apparel items worn by a model, a catchy slogan, and a product description. The input is a list of coordinated items, and the output is an image of the item worn by a model, a catchy slogan, and a product description. Specifically, a generative AI model is used to automatically generate images and text that highlight the appeal of the items.

[0629] Step 4: The server creates a sales page using the generated model image, catch phrase, and product description. The inputs are the model image, catch phrase, and product description, and the output is a sales page. Specifically, a web page generation tool is used to automatically create a sales page that includes detailed product information.

[0630] Step 5: The server automatically sets the sale price for the sales page. The sales page is required as input, and the sales page with the sale price set is obtained as output. Specifically, an AI algorithm is used to set the optimal sales price taking into account the market price and inventory status.

[0631] Step 6: The server automatically posts the created sales page to social media. The input is a sales page with a set sales price, and the output is the sales page posted to social media. Specifically, the server uses the social media API to post a link to the sales page along with a tagline and image.

[0632] Step 7: The user performs a virtual try-on using a smartphone, smart glasses, or a head-mounted display. The input is a list of products coordinated with the user's avatar, and the output is the results of the virtual try-on. Specifically, the coordinated items are applied to the user's avatar, and the user checks the appearance of the products through the virtual try-on.

[0633] Step 8: The user purchases the product through the virtual try-on. The input is the results of the virtual try-on, and the output is the product for which the purchase procedure has been completed. Specifically, the user checks the results of the virtual try-on and proceeds with the purchase procedure for the product they like.

[0634] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0635] "Example 1"

[0636] As one embodiment of the present invention, a system is provided that includes an emotion engine that uses AI to recognize consumer emotions. This emotion engine analyzes emotions from consumer social media posts. For example, it analyzes consumer emotions such as joy, sadness, surprise, anger, and fear from text, images, videos, etc. posted by consumers on social media. The analysis is performed using technologies such as text analysis, image recognition, and video analysis. The analyzed emotions are used to adjust product coordination. For example, if a consumer posts many times expressing joy, apparel items in bright colors can be selected and coordinated. On the other hand, if a consumer posts many times expressing surprise, apparel items with unique designs can be selected and coordinated. This makes it possible to suggest products that match the consumer's emotions, thereby increasing the consumer's desire to purchase.

[0637] "Example 2"

[0638] As one embodiment of the present invention, a system is provided that includes an emotion engine that uses AI to recognize consumer emotions. This emotion engine analyzes emotions from consumer social media posts. For example, it analyzes consumer emotions such as joy, sadness, surprise, anger, and fear from text, images, videos, etc. posted by consumers on social media. The analysis is performed using technologies such as text analysis, image recognition, and video analysis. The analyzed emotions are used to adjust product coordination. For example, if a consumer posts many times expressing joy, apparel items in bright colors can be selected and coordinated. On the other hand, if a consumer posts many times expressing surprise, apparel items with unique designs can be selected and coordinated. This makes it possible to suggest products that match the consumer's emotions, thereby increasing the consumer's desire to purchase.

[0639] The processing flow of each embodiment will be described below.

[0640] "Example 1"

[0641] Step 1: AI collects consumer social media posts, including text, images, and video data.

[0642] Step 2: The emotion engine analyzes the collected data to determine consumer emotions, using technologies such as text analysis, image recognition, and video analysis to extract emotions such as joy, sadness, surprise, anger, and fear.

[0643] Step 3: The AI ​​coordinates product outfits based on the analyzed emotions. For example, if a consumer posts a lot of content expressing happiness, it will select and coordinate brightly colored apparel items. On the other hand, if a consumer posts a lot of content expressing surprise, it will select and coordinate apparel items with unique designs.

[0644] Step 4: AI recommends coordinated products to consumers. This makes it possible to recommend products that match the consumer's emotions, increasing their desire to purchase.

[0645] Example 1

[0646] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0647] Conventional apparel sales systems have made it difficult to efficiently sell long-term inventory items or propose products that match consumer sentiment. They also lacked mechanisms for suggesting outfits that combine products from different brands or for encouraging consumers to reuse secondhand items. This has led to problems of backlogged inventory and a decline in consumer purchasing motivation.

[0648] The identification process by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means. In this invention, the server includes means for selecting long-term inventory items, means for coordinating the selected items across brands, means for generating images of the coordinated items worn by a model, a catchy slogan, and a product description, means for creating a sales page using the generated images of the items worn by a model, the catchy slogan, and the product description, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing platform, means for allowing consumers to purchase the items via the information sharing platform, means for generating images of the items worn by a model and a product description for selling the consumer's used items on the same platform, and means for analyzing consumer emotions and suggesting product coordination that matches the consumer's emotions. This enables efficient sales of long-term inventory items, suggestions for products that match consumer emotions, suggestions for coordinating items that combine products from different brands, and promotion of reuse of consumer used items.

[0649] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0650] "Means for selecting" refers to a method or device for selecting an object based on specific conditions.

[0651] "Means for coordinating" refers to a method or device for proposing a combination of multiple products.

[0652] "Model wearing image" refers to an image of a model wearing a product.

[0653] A "catchphrase" is a short sentence that succinctly expresses the features and appeal of a product.

[0654] "Product description" refers to text that explains the detailed information and features of a product.

[0655] "Sales Page" refers to a web page that displays product information and allows consumers to purchase the product.

[0656] "Sale price" refers to the selling price of a product.

[0657] An "information sharing platform" refers to online services, such as social media and websites, that allow for the widespread sharing of information.

[0658] "Consumer" refers to the general user who purchases goods.

[0659] "Used goods" refers to goods that have been previously used.

[0660] "Means for analyzing emotions" refers to methods or devices for reading emotions from text, images, etc.

[0661] A "generative AI model" refers to a model that uses artificial intelligence to generate text and images.

[0662] A "prompt sentence" refers to an input sentence that gives instructions to a generative AI model.

[0663] The present invention is a system for efficiently selling products that have been in stock for a long time and for proposing products that match consumer sentiment. Specific embodiments of this system will be described below.

[0664] Hardware and software used

[0665] This system operates with three main components: a server, a terminal, and a user. The server manages sales and inventory data using a database management system (e.g., MySQL). Programs are developed using Python, with TensorFlow and PyTorch as the main libraries for training and inference of AI models. Generative AI models include BERT, GPT-3, and DALL-E. The terminal provides an interface for users to access the system through a web browser.

[0666] Program processing

[0667] The server first collects sales and inventory data. Specifically, it connects to a MySQL database and queries the sales history for the past six months and the current inventory status. Next, it analyzes the collected data and selects products that have been in stock for a long time. This analysis is performed using the Python pandas library.

[0668] Based on the selected products, the server generates coordinated outfits that go well together, even if the items are from different brands. It uses a generative AI model (e.g., GPT-3) to suggest product combinations. Images of models wearing the outfits, catchy slogans, and product descriptions are automatically created using a generative AI model (e.g., DALL-E or GPT-3).

[0669] Furthermore, the server automatically sets the sales price for the sales page using AI and posts the created sales page to social media, allowing consumers to purchase products via social media. The AI ​​also generates images of models wearing the products and product descriptions on the same platform, and sells consumers' used clothing. This encourages consumers to reuse their used clothing and achieves zero waste.

[0670] The server analyzes emotions from consumers' social media posts and suggests outfits that match those emotions. It uses NLP and image recognition technology to analyze emotions. Specifically, it uses the Hugging Face Transformers library and OpenCV.

[0671] Specific examples

[0672] Below is an example of a prompt sentence to input to the generative AI model.

[0673] 1. Analysis of sales and inventory data

[0674] Prompt: "Based on sales and inventory data from the past six months, please select the product that has been in stock for a long time."

[0675] 2. Coordinate Generation

[0676] Prompt: "Using the selected products, suggest a coordinated look that works well together, even if it's from a different brand."

[0677] 3. Model Wearing Images and Copy Generation

[0678] Prompt: "Generate a model image, tagline, and product description based on this outfit."

[0679] 4. Sentiment analysis and coordination

[0680] Prompt: "Analyze the emotions of consumers in their social media posts and suggest product coordination that matches those emotions."

[0681] Using these prompts, the system can automatically analyze the data and generate optimal outfits and sales pages.

[0682] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0683] Step 1: Collect sales and inventory data

[0684] The server collects sales and inventory data from the database. Specifically, it connects to the MySQL database and queries the sales history for the past six months and the current inventory status. The input requires database connection information and a query statement, and the output is sales and inventory data.

[0685] Specific behavior:

[0686] The server executes the SQL query SELECT FROM sales WHERE date >= DATE_SUB(CURDATE(), INTERVAL 6 MONTH).

[0687] The server executes the SQL query SELECT FROM inventory to obtain the current inventory status.

[0688] Step 2: Select long-term inventory items

[0689] The server analyzes the collected data and selects products that have been in long-term inventory. It uses the Python pandas library to identify products with sluggish sales. Sales data and inventory data are required as input, and a list of long-term inventory items is obtained as output.

[0690] Specific behavior:

[0691] The server uses the pandas groupby function to aggregate sales for each product.

[0692] The server filters out items whose sales are below a certain threshold.

[0693] Step 3: Generate coordinates

[0694] The server generates coordinated outfits based on the selected items, even for items from different brands, and uses a generative AI model (e.g., GPT-3) to suggest combinations of items. The input is a list of long-term inventory items, and the output is a suggested outfit.

[0695] Specific behavior:

[0696] The server inputs a prompt to the generative AI model: "Using the selected products, please suggest a coordinated look that goes well with other products from different brands."

[0697] The server receives the generated coordinates and stores them in a database.

[0698] Step 4: Generate model images and tagline

[0699] The server uses a generative AI model (e.g., DALL-E or GPT-3) to generate images of a model wearing coordinated products, a catchy slogan, and a product description. The input is a coordination suggestion, and the output is an image of a model wearing the product, a catchy slogan, and a product description.

[0700] Specific behavior:

[0701] The server inputs a prompt to the generative AI model: "Based on this outfit, please generate an image of the model wearing the outfit, a catchy slogan, and a product description."

[0702] The server receives the generated images and text and stores them as materials for the sales page.

[0703] Step 5: Create a sales page and set your sale price

[0704] The server creates a sales page based on the generated materials, and the AI ​​automatically sets the sale price. Pricing takes into account past sales data and market trends. The inputs include images of models wearing the products, a catchy slogan, and a product description, and the output is a completed sales page.

[0705] Specific behavior:

[0706] The server generates the sales page using the HTML template.

[0707] The server inputs a prompt statement to the AI ​​model: "Please set the optimal sale price for this product."

[0708] The server reflects the set price on the sales page.

[0709] Step 6: Post your sales page on social media

[0710] The server automatically posts the completed sales page to social media. The posting is done using the social media API. The input required is the sales page URL and a tagline, and the output is the completed posting to social media.

[0711] Specific behavior:

[0712] The server connects to the SNS API using an authentication token.

[0713] The server posts the sales page URL and the generated catchphrase to social media.

[0714] Step 7: Selling your used clothing

[0715] Users upload their own secondhand clothes to the platform. The server automatically generates images and descriptions of the uploaded secondhand clothes using a generative AI model. The input is an image of the secondhand clothes, and the output is an image of a model wearing the clothes and a product description.

[0716] Specific behavior:

[0717] Users upload images of used clothing.

[0718] The server inputs a prompt to the generative AI model: "Please generate an image of a model wearing this vintage clothing item and a product description."

[0719] The server reflects the generated images and text on the sales page.

[0720] Step 8: Sentiment analysis and coordination adjustment

[0721] The server analyzes emotions from consumers' social media posts and suggests outfits that match those emotions. Emotions are analyzed using NLP and image recognition technology. The input is the consumer's social media post, and the output is outfit suggestions that match the emotion.

[0722] Specific behavior:

[0723] The server retrieves the consumer's posts using the SNS API.

[0724] The server uses Hugging Face's Transformers library to analyze the sentiment of the text.

[0725] Based on the analysis results, the server inputs a prompt statement to the generative AI model: "Please suggest a product coordination that matches the consumer's emotions."

[0726] The server proposes the generated coordinates to the consumer.

[0727] (Application example 1)

[0728] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0729] In conventional apparel sales systems, long-term inventory items often remained unsold, making inventory management difficult. It was also difficult to propose coordinated outfits across brand boundaries, which prevented consumers from motivating them to purchase. Furthermore, product proposals did not take into account consumer emotions, making it impossible to provide optimal products to individual consumers. To solve these issues, efficient sales promotion of long-term inventory items and coordinated outfit proposals based on consumer emotions were necessary.

[0730] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0731] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for consumers to purchase via the information sharing service, means for generating images of models wearing the apparel items and product descriptions for consumers' used clothing on the same platform and selling them, means for analyzing consumer sentiment and adjusting coordination based on the sentiment, means for generating product pages using a generative AI model, and means for inputting prompts into the generative AI model to generate images and text. This enables efficient sales promotion of long-term inventory items and optimal coordination suggestions based on consumer sentiment.

[0732] "Long-term inventory" refers to products that have been stored as inventory for a certain period of time or longer and remain unsold in the normal sales cycle.

[0733] "Apparel products" refers to wearable products such as clothing and fashion accessories.

[0734] "Creating outfits that transcend brand boundaries" refers to combining products from different brands to create a unified fashion style.

[0735] "Model wearing images" refers to photographs or images of a model actually wearing the product.

[0736] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0737] "Product description" refers to text that explains the detailed information, features, and usage of a product.

[0738] "Sales Page" refers to a web page for selling products online.

[0739] "Sale price" refers to the selling price of a product.

[0740] "Information sharing services" refer to online platforms for sharing information, such as social media and blogs.

[0741] "Consumer" refers to an individual who purchases goods or services.

[0742] "Secondhand clothing" refers to previously used clothing or fashion items.

[0743] "Platform" refers to the underlying system for providing specific services or functions.

[0744] "Analyzing emotions" refers to reading people's emotions from text, images, videos, etc.

[0745] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to generate images or text.

[0746] A "prompt" refers to an instruction or question that is input into a generative AI model.

[0747] The system for implementing this invention is composed of three main elements: a server, a terminal, and a user. The specific operation of each element and the hardware and software used will be explained below.

[0748] Server Operation

[0749] The server includes programs to perform the following functions:

[0750] 1. Select long-term inventory items:

[0751] The server analyzes the inventory database and selects apparel items that have been in stock for a certain period of time, using an artificial intelligence (AI) model.

[0752] 2. Coordination suggestions that transcend brand boundaries:

[0753] The server then combines selected long-term inventory items across brands to suggest a cohesive fashion style, again using an AI model.

[0754] 3. Generate model images and product descriptions:

[0755] The server uses a generative AI model to generate images of a model wearing the coordinated products, a catchy copy, and a product description. The generative AI model generates images and text by inputting a prompt sentence.

[0756] 4. Sales page creation and automatic posting:

[0757] The server creates a sales page using the generated images of the model wearing the product, a catchy slogan, and a product description, and automatically posts it to an information sharing service (SNS).

[0758] 5. Consumer sentiment analysis:

[0759] The server analyzes the emotions of consumers in their social media posts and adjusts outfits based on those emotions. This process is performed using an emotion analysis engine.

[0760] Device behavior

[0761] The terminal (smartphone) provides an interface for users to view and purchase apparel products.

[0762] 1. Product browsing:

[0763] The user can view the sales page generated by the server through the terminal.

[0764] 2. Purchase procedure:

[0765] Users can purchase products through the terminal, and the purchase procedure is carried out through an electronic payment service.

[0766] User Actions

[0767] The user uses the terminal to browse and purchase apparel items.

[0768] 1. Social Media Posts:

[0769] Users express their feelings by posting on the SNS, and these posts are analyzed by the server and used to adjust coordination.

[0770] Hardware and software used

[0771] Hardware: Servers, smartphones

[0772] Software: Python, transformers library, requests library, sentiment analysis engine, generative AI model

[0773] Examples of specific examples and prompts

[0774] As a concrete example, the following is an example of a prompt sentence to be input to the generative AI model.

[0775] Sentiment analysis prompt: "Analyze the sentiment of the following text: 'I love this new dress!'"

[0776] Image generation prompt: "Generate an image of a Red Dress."

[0777] Text generation prompt: "Generate a catchy description for a Red Dress."

[0778] This allows the server to efficiently promote long-term inventory items and propose optimal outfits based on consumer sentiment.

[0779] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0780] Step 1: The server analyzes the inventory database to select apparel items that have become long-term inventory. It uses the inventory database as input and obtains a list of long-term inventory items as output. This process uses an artificial intelligence (AI) model to select items that have been in stock for more than a certain period of time (e.g., 180 days).

[0781] Step 2: The server coordinates the selected long-term inventory items across brands. It uses a list of long-term inventory items as input and obtains coordinated product combinations as output. This process also uses an AI model to combine products from different brands to suggest a cohesive fashion style.

[0782] Step 3: The server uses the generative AI model to generate an image of a model wearing the coordinated items, a catchy slogan, and a product description. It uses the coordinated product combination as input and obtains an image of a model wearing the items, a catchy slogan, and a product description as output. The generative AI model receives a prompt sentence as input and generates images and text. Examples of specific prompt sentences include "Generate an image of a Red Dress." and "Generate a catchy description for a Red Dress."

[0783] Step 4: The server creates a sales page using the generated model images, taglines, and product descriptions. The generated images and text are used as input, and a sales page is obtained as output. This sales page includes detailed product information, prices, a purchase button, and more.

[0784] Step 5: The server automatically sets the sale price for the sales page. It uses the sales page and market data as input and sets the optimal sale price as output. This process uses a pricing algorithm to determine the price based on competitor prices and demand forecasts.

[0785] Step 6: The server automatically posts the created sales page to an information sharing service (SNS). The sales page is used as input and the SNS post is made as output. This process uses the SNS API, and the posted content includes an image of the model wearing the product, a catchy slogan, and a product description.

[0786] Step 7: The user browses the sales page generated by the server through their terminal and purchases the product. The sales page is used as input and the purchase procedure is performed as output. The purchase procedure is performed through an electronic payment service.

[0787] Step 8: The server analyzes the sentiment of the consumer's social media posts and adjusts the outfit based on that sentiment. It uses the social media post data as input and obtains the sentiment analysis results as output. This process uses a sentiment analysis engine to read emotions from text, images, and videos. An example of a specific prompt is "Analyze the sentiment of the following text: 'I love this new dress!'"

[0788] Step 9: The server also generates images of models wearing the consumer's used clothing and product descriptions for sale on the same platform. Used clothing information is used as input, and images of models wearing the clothing and product descriptions are obtained as output. This process uses a generative AI model, which generates images and text based on prompts.

[0789] Example 2

[0790] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0791] Conventional apparel sales systems were unable to efficiently process long-term inventory, resulting in backlogged inventory and declining sales. Furthermore, it was difficult to propose outfits that transcended brand boundaries, preventing consumers from purchasing. Furthermore, product suggestions were not based on consumer sentiment, making it difficult to provide services that meet the needs of individual consumers. To solve these issues, efficient inventory management utilizing AI technology and outfit suggestions based on consumer sentiment were required.

[0792] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0793] In this invention, the server includes means for selecting long-term inventory items, means for coordinating the selected items across brand boundaries, means for generating images of the coordinated items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the items worn by models, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing platform, means for consumers to purchase via the information sharing platform, means for generating images of the items worn by models and product descriptions and selling consumers' used items on the same platform, and means for analyzing consumer sentiment and adjusting product coordination based on the analysis results. This enables efficient processing of long-term inventory items, attractive coordination suggestions across brand boundaries, and individual responses based on consumer sentiment.

[0794] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0795] "Goods" means goods sold to consumers.

[0796] "Brand" refers to a name or logo that is used to identify the products or services offered by a particular company or manufacturer.

[0797] "Coordination" refers to proposing a combination of multiple products.

[0798] "Model Wearing Image" refers to a photograph of a model wearing a product.

[0799] A "catchphrase" is a short sentence that succinctly expresses the features and appeal of a product.

[0800] "Product description" refers to text that explains the detailed information and features of a product.

[0801] "Sales Page" means a web page for selling products online.

[0802] "Sale Price" refers to the price at which an item is sold.

[0803] An "information sharing platform" refers to online services for sharing information, such as social media and websites.

[0804] "Consumer" refers to an individual who purchases a product.

[0805] "Used goods" refers to goods that have been previously used.

[0806] "Emotion" refers to the psychological state of the consumer.

[0807] "Analysis" refers to analyzing data to extract meaning.

[0808] "Artificial intelligence" refers to the technology that allows computers to act in a way that mimics human intelligence.

[0809] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[0810] The server first collects sales and inventory data from the database, using SQL queries to retrieve past sales history and current inventory status. It then uses Python and TensorFlow to analyze the collected data and identify slowing sales. This allows it to select products that have been in stock for a long time.

[0811] For the selected products, the server uses a generative AI model (for example, OpenAI's GPT-4) to generate a coordinated outfit that transcends brand boundaries. Specifically, the following prompt sentence is input into the generative AI model:

[0812] "This item has been in stock for a long time, so please coordinate it with other brand items. Please generate a model image, tagline, and product description."

[0813] Based on the generated outfit information, the server uses an HTML template to create a sales page, which includes images of the product being worn by a model, a tagline, and a product description. Furthermore, a pricing model is built using Python and Scikit-learn to automatically set the sale price for the sales page.

[0814] The created sales page is automatically posted to an information-sharing platform using SNS APIs, allowing consumers to purchase products via the information-sharing platform.

[0815] Users can also sell their own second-hand goods on the platform. When a user uploads an image of their used clothing, the server uses a generative AI model to generate an image of the model wearing the item and a product description. Specifically, the following prompt sentence is input into the generative AI model:

[0816] "Please generate an image of a model wearing this vintage clothing along with a product description."

[0817] Furthermore, the server uses NLP technology (e.g., the BERT model) and image recognition technology (e.g., OpenCV) to analyze consumer sentiment. Based on the analysis results, it coordinates products to match the consumer's sentiment. For example, if a consumer posts many times expressing happy emotions, it selects and coordinates products in bright colors.

[0818] In this way, it becomes possible to efficiently process long-term inventory items, propose attractive coordination that transcends brand boundaries, and respond to individual customer needs based on consumer sentiment.

[0819] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0820] Step 1: The server collects sales and inventory data from the database. The input is the database connection information and an SQL query. Specifically, the server executes the SQL query to obtain past sales history and current inventory status. The output is sales and inventory data.

[0821] Step 2: The server analyzes the collected sales and inventory data to identify products that have become long-term inventory. The server uses sales and inventory data as input. Specifically, the server uses Python and TensorFlow to predict sales trends and identify products with slowing sales. The output is a list of long-term inventory products.

[0822] Step 3: The server uses the generative AI model to generate a cross-brand coordination based on the long-term inventory product list. The long-term inventory product list and a prompt sentence are used as input. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0823] "This item has been in stock for a long time, so please coordinate it with other brand items. Please generate a model image, tagline, and product description."

[0824] As an output, coordinate information is obtained.

[0825] Step 4: The server creates a sales page based on the generated coordination information. The coordination information and HTML template are used as input. Specifically, the server embeds an image of a model wearing the item, a catchy slogan, and a product description into the HTML template to generate the sales page. The sales page is obtained as output.

[0826] Step 5: The server automatically sets the sale price for the sales page. The sales page and pricing model are used as input. Specifically, the server uses Python and Scikit-learn to build a pricing model and predict the sale price for the sales page. The output is the sales page with the sale price set.

[0827] Step 6: The server automatically posts the created sales page to the information sharing platform. The sales page and the information sharing platform's API are used as input. Specifically, the server posts the sales page using the information sharing platform's API. The output is the sales page posted on the information sharing platform.

[0828] Step 7: The user uploads an image of their used clothing to sell on the platform. The image of the used clothing is used as input. Specifically, the server uses the generative AI model to generate an image of the model wearing the clothing and a product description. The following prompt sentence is input into the generative AI model:

[0829] "Please generate an image of a model wearing this vintage clothing along with a product description."

[0830] The output is an image of a model wearing the product and a product description.

[0831] Step 8: The server analyzes the consumer's sentiment and adjusts product coordination based on the analysis results. The input is the consumer's social media posting data. Specifically, the server analyzes sentiment using NLP technology (e.g., the BERT model) or image recognition technology (e.g., OpenCV). The output is the sentiment analysis result. Based on the analysis results, the server inputs the following prompt sentence into the generative AI model:

[0832] "This consumer often posts expressing joy, so choose apparel items in bright colors and coordinate them together."

[0833] As an output, emotion-based coordination information is obtained.

[0834] (Application example 2)

[0835] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0836] In the conventional apparel sales system, there was a lack of efficient ways to sell long-term inventory, making inventory management difficult. It was also difficult to propose coordination that transcended the boundaries of brands, which prevented consumers from motivating purchases. Furthermore, product proposals were not based on consumer emotions, making it impossible to provide the best products for each individual consumer.

[0837] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0838] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for consumers to purchase via the information sharing service, means for generating images of the consumers wearing the apparel items on models and product descriptions and selling their used clothing on the same platform, means for analyzing consumer sentiment, and means for making product suggestions based on the analyzed sentiment. This enables efficient sales of long-term inventory items, coordination suggestions across brands, and optimal product suggestions based on consumer sentiment.

[0839] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0840] "Apparel products" refers to clothing items such as clothing and fashion accessories.

[0841] "Crossing brand boundaries" refers to combining products from different brands.

[0842] "Coordination" refers to styling multiple apparel items together.

[0843] "Model Wearing Image" refers to a photograph of a model wearing an apparel product.

[0844] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0845] "Product description" refers to text that explains the detailed information and features of a product.

[0846] "Sales page" refers to a web page that lists product information and allows consumers to purchase the product.

[0847] "Sale price" refers to the selling price of a product.

[0848] "Information sharing services" refers to online platforms such as social networking sites.

[0849] "Consumer" refers to an individual who purchases a product.

[0850] "Used clothing" refers to clothing that has been previously purchased and used by a consumer.

[0851] "Platform" refers to a system that serves as the foundation for providing multiple functions and services.

[0852] "Analyzing emotions" refers to reading consumer emotions from data.

[0853] "Product suggestion" refers to recommending a specific product to a consumer.

[0854] "Artificial intelligence" refers to technology that allows machines to act in a way that mimics human intelligence.

[0855] A system for implementing the present invention has the following configuration.

[0856] First, the server selects apparel items that have become long-term inventory. This is achieved by analyzing sales and inventory data. Specifically, the server retrieves sales and inventory data from the database and selects items that have remained unsold for a certain period of time as long-term inventory items.

[0857] Next, the selected apparel items are coordinated across brands. The server uses artificial intelligence to combine products from different brands and propose optimal outfits. For example, it uses a clustering algorithm to group products based on color or design characteristics and find combinations that go well together.

[0858] Images of models wearing coordinated apparel products, catch phrases, and product descriptions are automatically generated using a generative AI model. The server inputs the following prompt sentence into the generative AI model and obtains the generated text.

[0859] Example prompt sentence:

[0860] Product Name: Refreshing Summer Dress

[0861] Description: This dress features a lightweight material and vibrant color. It's designed to be comfortable even on hot summer days, making it perfect for relaxing at the beach or a cafe.

[0862] The server creates a sales page using the generated images of models wearing the product, catchy slogans, and product descriptions. The server automatically sets the sale price on the sales page. This is achieved by calculating the optimal price based on past sales data and market trends.

[0863] The created sales page is automatically posted to the information sharing service by the server, allowing consumers to purchase the product via the information sharing service.

[0864] Consumers can also sell their used clothing on the same platform. Based on the information provided by consumers, the server uses a generative AI model to generate images of models wearing the clothing and product descriptions, and creates a sales page.

[0865] To analyze consumer sentiment, the server collects social media posts from consumers and analyzes them using a sentiment analysis algorithm. Based on the analyzed sentiment, the server makes optimal product recommendations to consumers. For example, if a consumer posts frequently expressing joy, the server will recommend brightly colored apparel products.

[0866] In this way, the server enables efficient sales of long-term inventory items, coordination suggestions that transcend brand boundaries, and optimal product suggestions based on consumer sentiment.

[0867] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0868] Step 1: The server retrieves sales and inventory data from the database. This collects basic data for selecting apparel items that have become long-term inventory. The input is sales and inventory data, and the output is the integration result of these data.

[0869] Step 2: The server analyzes the acquired data and selects items that have remained unsold for a certain period of time as long-term inventory items. Specifically, it filters out items with sales below a certain level. The input is the integrated sales data and inventory data, and the output is a list of long-term inventory items.

[0870] Step 3: The server uses artificial intelligence to coordinate the selected long-term inventory items across brands. It uses a clustering algorithm to group products based on color and design characteristics and find combinations that go well together. The input is a list of long-term inventory items, and the output is a list of coordinated items.

[0871] Step 4: The server uses the generative AI model to generate images of coordinated apparel items worn by a model, a tagline, and a product description. The prompt text is input into the generative AI model and the generated text is obtained. The input is a list of coordinated items and the prompt text, and the output is the generated images of the items worn by a model, a tagline, and a product description.

[0872] Step 5: The server creates a sales page using the generated model images, catchphrases, and product descriptions. The input is the generated content, and the output is the completed sales page.

[0873] Step 6: The server automatically sets the sale price on the sales page. It calculates the optimal price based on past sales data and market trends. The input is the sales page and past sales data, and the output is the set sale price.

[0874] Step 7: The server automatically posts the created sales page to the information sharing service, allowing consumers to purchase the product via the information sharing service. The input is the completed sales page, and the output is the result of posting to the information sharing service.

[0875] Step 8: The server uses the generative AI model to generate images of a model wearing the item and a product description based on the consumer's used clothing information, and creates a sales page. The input is the consumer's used clothing information and prompt text, and the output is the generated image of a model wearing the item and a product description.

[0876] Step 9: The server collects the consumer's SNS posts and analyzes their emotions using a sentiment analysis algorithm. The input is the consumer's SNS posts, and the output is the analyzed sentiment data.

[0877] Step 10: The server makes optimal product recommendations to the consumer based on the analyzed emotions. For example, if a consumer posts a lot of happy emotions, it will suggest brightly colored apparel products. The input is the analyzed emotion data, and the output is a list of suggested products.

[0878] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0879] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0880] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0881] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0882] [Third embodiment]

[0883] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0884] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0885] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0886] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0887] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0888] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0889] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0890] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0891] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0892] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0893] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0894] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0895] "Example 1"

[0896] One embodiment of the present invention is an apparel sales system using AI. In this system, AI first selects apparel items that have been in stock for a long time. For example, it analyzes sales and inventory data and automatically selects items with slowing sales. Next, it coordinates the selected items across brands. This makes it possible to suggest combinations that work well together, even among products from different brands. Images of the coordinated items worn by models, catchy slogans, and product descriptions are automatically generated by the AI. This enables the efficient creation of sales pages that maximize the appeal of the products. Furthermore, the AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media. Furthermore, AI also generates images of models wearing the items and product descriptions for the consumer's used clothing on the same platform and sells them. This encourages consumers to reuse their used clothing and achieves zero waste.

[0897] "Example 2"

[0898] One embodiment of the present invention is an apparel sales system using AI. In this system, AI first selects apparel items that have been in stock for a long time. For example, it analyzes sales and inventory data and automatically selects items with slowing sales. Next, it coordinates the selected items across brands. This makes it possible to suggest combinations that work well together, even among products from different brands. Images of the coordinated items worn by models, catchy slogans, and product descriptions are automatically generated by the AI. This enables the efficient creation of sales pages that maximize the appeal of the products. Furthermore, the AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media. Furthermore, AI also generates images of models wearing the items and product descriptions for the consumer's used clothing on the same platform and sells them. This encourages consumers to reuse their used clothing and achieves zero waste.

[0899] The processing flow of each embodiment will be described below.

[0900] "Example 1"

[0901] Step 1: The AI ​​selects apparel items that have been in stock for a long time. Specifically, it analyzes sales and inventory data and automatically selects items whose sales are slowing.

[0902] Step 2: Coordinate the selected products across brands, making it possible to propose combinations that go well together even when products from different brands are used.

[0903] Step 3: Images of models wearing the coordinated products, along with catchy slogans and product descriptions, are automatically created by generative AI. This makes it possible to efficiently create sales pages that maximize the appeal of the products.

[0904] Step 4: The AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase the product via social media.

[0905] Step 5: Consumers can also purchase their used clothing on the same platform, with AI generating images of models wearing the clothing and product descriptions. This will encourage consumers to reuse their used clothing and achieve zero waste.

[0906] Example 1

[0907] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0908] Conventional apparel sales systems lacked efficient methods for selling long-term inventory. Furthermore, there were no effective methods for suggesting outfits that combined products from different brands, or for reusing consumers' old clothes. This resulted in increased inventory backlog and waste, and reduced sales efficiency.

[0909] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0910] In this invention, the server includes means for collecting sales data and inventory data, means for analyzing the collected data to select long-term inventory items, means for proposing outfits by combining the selected items with items from different brands, means for generating images of the coordinated items worn by models, catch phrases, and product descriptions using a generative AI model, means for creating a sales page using the generated images of the coordinated items worn by models, catch phrases, and product descriptions, means for automatically setting sale prices by executing a pricing algorithm based on past sales data, means for automatically posting the created sales page to social media, means for allowing consumers to purchase via social media, and means for generating images of the items worn by models and product descriptions and selling consumers' used clothing on the same platform. This makes it possible to efficiently sell long-term inventory items, propose attractive outfits by combining items from different brands, and encourage consumers to reuse used clothing.

[0911] "Sales data" refers to data that includes transaction information regarding the sale of apparel products.

[0912] "Inventory data" is data that indicates the current inventory status of apparel products.

[0913] "Long-term inventory" refers to apparel products that remain unsold for a certain period of time.

[0914] "AI" is a system that uses artificial intelligence technology to analyze data and make predictions.

[0915] "Coordination" is a fashion combination that combines products from different brands.

[0916] A "generative AI model" is an artificial intelligence model for generating natural language and images.

[0917] "Model wearing image" refers to an image of a model wearing an apparel product.

[0918] A "catchphrase" is a short promotional phrase that draws out the appeal of a product.

[0919] A "product description" is a text that explains the features and benefits of a product.

[0920] A "sales page" is a web page for selling apparel products online.

[0921] "Sale price" refers to the selling price of a product.

[0922] "SNS" refers to social networking services, which are online platforms for users to share information.

[0923] "Consumer" refers to a customer who purchases apparel products.

[0924] "Used clothing" refers to second-hand apparel items that consumers have previously purchased and wish to resell.

[0925] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[0926] The server first collects sales and inventory data. To do this, MySQL is used as the database management system (DBMS). The server retrieves sales data for the past six months and current inventory data using SQL queries.

[0927] The server then analyzes the collected data and selects items that have become long-term inventory. This analysis is performed using Python and TensorFlow. Items whose sales have fallen below a certain threshold are selected as long-term inventory. For example, items whose sales have been less than 10 units in the past six months are selected.

[0928] For the selected items, the server proposes outfits by combining them with products from different brands. To propose outfits, OpenCV is used to evaluate color compatibility and Scikit-learn is used to evaluate style compatibility. Specifically, a color wheel is used to evaluate color compatibility, and a clustering algorithm is used to evaluate style compatibility.

[0929] Next, the server uses a generative AI model (e.g., GPT-4) to generate an image of a model wearing the coordinated items, along with a catchy slogan and product description. The following prompt is input to the generative AI model:

[0930] "This is a long-term inventory item. Please generate a model image, tagline, and product description based on the following information: Product name: XXXX, Brand: YYYY, Color: ZZZZ, Style: AAAA"

[0931] Based on the generated information, the server uses an image generation API (e.g., DALL-E) to generate an image of the model wearing the product.

[0932] The server then runs a pricing algorithm based on past sales data to determine the optimal sale price. To do this, it uses XGBoost to build a pricing model and predict the optimal sale price for each product.

[0933] Next, the server creates a sales page based on the generated model images, tagline, product description, and sale price. The sales page is built using HTML and CSS. The link to the created sales page and the generated tagline are automatically posted to social media using an API.

[0934] Additionally, users can input photos of used clothing and basic information (brand, color, style, etc.) from their devices. The server uses this information to generate an image of the clothing being worn by a model and a product description using a generative AI model. For example, the following prompt sentence can be input into the generative AI model:

[0935] "This used clothing item is for sale by a user. Please generate a model image, tagline, and product description based on the following information: Product name: XXXX, Brand: YYYY, Color: ZZZZ, Style: AAAA"

[0936] Based on the generated information, the server creates a sales page and automatically posts it to social media, allowing users to easily sell their used clothing.

[0937] In this way, the system of the present invention makes it possible to efficiently sell long-term inventory items, propose attractive outfits that combine products from different brands, and encourage consumers to reuse used clothing.

[0938] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0939] Step 1: The server collects sales and inventory data. Specifically, the server connects to a MySQL database and retrieves sales data for the past six months and current inventory data using an SQL query. The input is the database connection information and query, and the output is sales data and inventory data.

[0940] Step 2: The server analyzes the collected data and selects items that have become long-term inventory. Specifically, the server uses Python and TensorFlow to analyze sales data and selects items whose sales fall below a certain threshold as long-term inventory. The inputs are sales data and inventory data, and the output is a list of long-term inventory items.

[0941] Step 3: The server combines the selected long-term inventory items with products from different brands to suggest outfits. Specifically, the server evaluates color compatibility using OpenCV and style compatibility using Scikit-learn. The input is a list of long-term inventory items and product data from other brands, and the output is a list of outfit suggestions.

[0942] Step 4: The server uses the generative AI model to generate images of the coordinated products worn by a model, a catchphrase, and a product description. Specifically, the server inputs a prompt to the generative AI model and obtains the generated text and images. The input is a list of outfit suggestions and a prompt, and the output is an image of the product worn by a model, a catchphrase, and a product description.

[0943] Step 5: The server creates a sales page using the generated model image, tagline, and product description. Specifically, the server builds the sales page using HTML and CSS. The input is the model image, tagline, and product description, and the output is an HTML file for the sales page.

[0944] Step 6: The server runs a pricing algorithm based on past sales data to automatically set the sale price. Specifically, the server uses XGBoost to build a pricing model and predict the optimal sale price for each product. The input is past sales data, and the output is the sale price.

[0945] Step 7: The server automatically posts the created sales page to social media. Specifically, the server uses an API to post the sales page link and the generated tagline. The input is the sales page link and tagline, and the output is the result of posting to social media.

[0946] Step 8: The user enters a photo of the used clothing and basic information (brand, color, style, etc.) from the device. Specifically, the user enters this information using a web form. The input is a photo of the used clothing and basic information, and the output is the data sent to the server.

[0947] Step 9: The server uses the generative AI model to generate an image of a model wearing the item and a product description based on the information entered by the user. Specifically, the server inputs a prompt into the generative AI model and obtains the generated text and image. The input is a photo of the used clothing, basic information, and the prompt, and the output is an image of a model wearing the item and a product description.

[0948] Step 10: The server creates a sales page based on the generated information and automatically posts it to social media. Specifically, the server uses HTML and CSS to build the sales page and posts it using an API. The input is an image of the model wearing the product, a product description, and a link to the sales page, and the output is the result of posting to social media.

[0949] (Application example 1)

[0950] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0951] Conventional apparel sales systems have faced the problem of making it difficult to efficiently sell long-term inventory items, requiring a great deal of effort for inventory management and sales promotion. Furthermore, they lacked systems for suggesting outfits that combine products from different brands or for reusing consumers' old clothing, making it impossible to provide an attractive purchasing experience for consumers. Furthermore, they lacked the ability to automatically generate product descriptions using generative AI models or automatically post them to information sharing services, making sales page creation and promotional activities inefficient.

[0952] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for selecting long-term inventory apparel items, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for enabling consumers to purchase the items via the information sharing service, means for generating images of the items worn by models and product descriptions for consumers' used clothing on the same platform and selling them, means for analyzing sales data and inventory data, means for proposing coordination by combining products from different brands, means for generating product descriptions using a generative AI model, and means for posting the generated product descriptions to the information sharing service. This enables efficient sales of long-term inventory items, attractive coordination suggestions combining products from different brands, reuse of consumers' used clothing, and efficient creation of sales pages and promotional activities using the automatic generation function utilizing the generative AI model.

[0953] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[0954] "Apparel products" refers to fashion-related products such as clothing and fashion accessories.

[0955] "Creating coordination that transcends brand boundaries" refers to proposing styling by combining products from different brands.

[0956] "Model wearing image" refers to an image of a model actually wearing the product.

[0957] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[0958] "Product description" refers to text that explains the detailed information and features of a product.

[0959] "Sales page" refers to a web page that lists product information and allows consumers to purchase the product.

[0960] "Sale price" refers to the selling price of a product.

[0961] "Information sharing services" refers to online platforms such as social networking sites.

[0962] "Used consumer clothing" refers to second-hand clothing that is owned by a consumer.

[0963] "Platform" refers to the underlying system for providing specific services or functions.

[0964] "Sales Data" refers to data regarding product sales performance.

[0965] "Inventory Data" refers to data regarding the availability of products.

[0966] A "generative AI model" refers to a model that uses artificial intelligence to generate text and images.

[0967] "Prompt" refers to a text instruction that is input to a generative AI model.

[0968] A system for implementing the present invention includes a server, a user terminal, and a generative AI model. A specific embodiment of the system will be described below.

[0969] Server Processing

[0970] The server is equipped with a database for analyzing sales and inventory data and a data analysis module using artificial intelligence (AI). The server integrates sales and inventory data and automatically selects apparel items that have been in stock for a long time. The selected items are then analyzed using an AI clustering algorithm to suggest outfits that can be combined with products from different brands.

[0971] Next, the server uses the generative AI model to automatically generate images of the coordinated items worn by models, taglines, and product descriptions. The generated content is used to create a sales page. The sale price on the sales page is automatically set by the AI.

[0972] Furthermore, the server has the function of automatically posting the generated sales page to information sharing services (such as SNS), allowing consumers to purchase products through the information sharing service.

[0973] User terminal processing

[0974] The user device provides an interface for consumers to browse and purchase products. Users can check product information and complete the purchase process through the sales page. Users can also upload their own used clothing to the platform, and images of models wearing the clothing and product descriptions can be automatically generated using a generative AI model. This allows users to reuse and sell their used clothing.

[0975] Processing generative AI models

[0976] A generative AI model is an artificial intelligence model for generating text and images. Specifically, it uses natural language processing models such as GPT-2 to generate product descriptions and catchy slogans. Image generation models (e.g., GANs) are used to generate images of models wearing the products.

[0977] Prompt Sentence Examples

[0978] The following text will be used as an example of a prompt sentence to be input into the generative AI model.

[0979] Product Name: Slim Fit Jeans, Features: Stretch Material, Dark Blue, Casual

[0980] By inputting this prompt, the generative AI model generates the following product description:

[0981] These slim-fit jeans are made from a stretchy fabric for ease of movement and comfort, and their dark blue color is perfect for a casual look and pairs well with any top.

[0982] In this way, by combining the server, user terminal, and generative AI model, an efficient and attractive apparel product sales system can be realized.

[0983] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0984] Step 1: The server retrieves sales data and inventory data from the database. This data is integrated to select apparel items that have become long-term inventory. Specifically, products whose sales have been sluggish for a certain period of time are filtered out and listed as long-term inventory items. The input is sales data and inventory data, and the output is a list of long-term inventory items.

[0985] Step 2: The server proposes outfits by combining the selected long-term inventory items with products from different brands. It uses an AI clustering algorithm to analyze product characteristics and group items that go well together. The input is a list of long-term inventory items, and the output is a list of outfit suggestions.

[0986] Step 3: The server uses the generative AI model to automatically generate images of the coordinated items worn by a model, a catchphrase, and a product description. Specifically, the server inputs a prompt into the generative AI model and retrieves the generated text and images. The input is a list of outfit suggestions and the prompt, and the output is an image of the item worn by a model, a catchphrase, and a product description.

[0987] Step 4: The server creates a sales page using the generated model images, tagline, and product description. The layout and design of the sales page are automatically generated and product information is placed on it. The input is the model images, tagline, and product description, and the output is the completed sales page.

[0988] Step 5: The server automatically sets the sale price for the sales page using AI. It analyzes past sales data and market trends to calculate the optimal price. The input is the sales page and sales data, and the output is a sales page with the sale price set.

[0989] Step 6: The server automatically posts the created sales page to an information sharing service (such as SNS). Using an API, it posts the sales page link and product information. The input is the sales page with the set sale price, and the output is a notification that posting to the information sharing service has been completed.

[0990] Step 7: The user browses and purchases the product through the information sharing service. The user clicks on the link to the sales page and checks the product information. The input is the post on the information sharing service, and the output is the user's purchase procedure.

[0991] Step 8: The user uploads their own secondhand clothing to the platform, and the generative AI model automatically generates an image of the clothing being worn by a model and a product description. The user's device inputs the image and characteristics of the secondhand clothing and sends a prompt to the generative AI model. The input is the image and characteristics of the secondhand clothing, and the output is an image of the clothing being worn by a model and a product description.

[0992] Step 9: The server uses the generated model image and product description to create a sales page for the used clothing and posts it to the information sharing service. The input is the model image and product description, and the output is the sales page for the used clothing and a notification that posting to the information sharing service has been completed.

[0993] Example 2

[0994] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0995] Conventional apparel sales systems have difficulty efficiently selling products that have been in stock for a long time, and lack mechanisms for suggesting outfits that combine products from different brands or encouraging consumers to reuse used clothing. This leads to issues such as increased inventory backlog and waste, and reduced sales efficiency.

[0996] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0997] In this invention, the server includes means for collecting and analyzing sales data and inventory data, means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information transmission medium, means for consumers to purchase via the information transmission medium, and means for generating images of models wearing the apparel items and product descriptions and selling consumers' used clothing on the same platform. This enables efficient sales of long-term inventory items, suggestions for coordinating outfits that combine products from different brands, and promotion of consumers' reuse of used clothing.

[0998] "Sales data" is data that includes information regarding sales of apparel products.

[0999] "Inventory data" refers to data that includes information regarding the inventory status of apparel products.

[1000] "Long-term inventory" refers to apparel products that remain unsold for a certain period of time.

[1001] "Artificial intelligence" is a technology that uses machine learning and data analysis to automatically make judgments and predictions.

[1002] "Coordination" refers to proposing styling by combining multiple apparel items.

[1003] A "model wearing image" is an image of a model wearing an apparel product.

[1004] A "catchphrase" is a short phrase that succinctly conveys the appeal of a product.

[1005] A "product description" is a detailed description of the product's features and benefits.

[1006] A "sales page" is a web page for selling apparel products online.

[1007] "Sale price" refers to the selling price of a product.

[1008] An "information dissemination medium" is a platform for widely disseminating information, such as social media or websites.

[1009] "Consumers" refers to general customers who purchase apparel products.

[1010] "Used clothing" refers to apparel products that have been previously purchased and used by consumers.

[1011] The "platform" is a system that serves as the foundation for selling apparel products and disseminating information.

[1012] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[1013] First, the server collects and analyzes sales and inventory data. For this, a database management system (e.g., MySQL) is used. The data is read using the Python pandas library and converted into a data frame. Next, the sales and inventory data are analyzed and time series analysis is performed to identify sales trends.

[1014] Based on the analysis results, the server selects apparel items that have been in long-term inventory. Specifically, it selects items whose sales over the past six months are below a certain threshold. This selection is made using artificial intelligence (AI).

[1015] The server then coordinates the selected items across brands, using a generative AI model (e.g., OpenAI's GPT-4) to provide a prompt and receive coordination suggestions.

[1016] The server then uses the generative AI model to automatically generate images of the coordinated products worn by the model, a catchphrase, and a product description. DALL-E is used for image generation, and GPT-4 is used for generating the catchphrase and product description.

[1017] In addition, the server uses machine learning algorithms (e.g., scikit-learn) to automatically set the sale price on the sales page, calculating the optimal price based on past sales data.

[1018] The server creates a sales page based on the generated model images, catchphrases, product descriptions, and sale prices. The created sales page is then automatically posted to information sharing media (e.g., social media), allowing consumers to purchase the product via the information sharing media.

[1019] Users can also upload product information to the platform to sell their used clothing. The server then uses a generative AI model to automatically generate images of models wearing the clothing and product descriptions, creating a sales page. This encourages users to reuse their used clothing and achieve zero waste.

[1020] As a concrete example, the following prompt sentence is input into the generative AI model:

[1021] "Analyze sales data from the past six months and select apparel items that are experiencing slower sales. Coordinate the selected items with products from different brands, generate model images, taglines, and product descriptions, set a sale price for the sales page, and automatically post to social media."

[1022] Using this prompt, the generative AI model can perform the specified task and efficiently create a sales page.

[1023] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1024] Step 1: The server collects and analyzes sales and inventory data. As input, it obtains sales and inventory data from a database management system (e.g., MySQL). Specifically, it reads the data using Python's pandas library and converts it into a data frame. As output, it performs time series analysis to obtain analytical results and identify sales trends.

[1025] Step 2: The server selects apparel items that have become long-term inventory based on the analysis results. The analysis results obtained in Step 1 are used as input. Specifically, it selects items whose sales over the past six months are below a certain threshold. The output is a list of selected long-term inventory items.

[1026] Step 3: The server coordinates the selected products across brands. It uses the list of long-term inventory items obtained in Step 2 as input. Specifically, it inputs a prompt to a generative AI model (e.g., OpenAI's GPT-4) and receives coordination suggestions. The output is a list of coordinated products.

[1027] Step 4: The server uses the generative AI model to automatically generate images of the coordinated items worn by the model, taglines, and product descriptions. The list of coordinated items obtained in Step 3 is used as input. Specifically, DALL-E is used to generate images, and GPT-4 is used to generate taglines and product descriptions. The output is an image of the model wearing the items, taglines, and product descriptions.

[1028] Step 5: The server uses a machine learning algorithm (e.g., scikit-learn) to automatically set the sale price for the sales page. It uses past sales data and the product information obtained in Step 4 as input. Specifically, it trains a machine learning model to calculate the optimal price for the new product. The output is the sale price.

[1029] Step 6: The server creates a sales page based on the generated model image, catchphrase, product description, and sale price. The product information obtained in step 4 and the sale price obtained in step 5 are used as input. Specifically, the server generates a web page using HTML and CSS. The completed sales page is obtained as output.

[1030] Step 7: The server automatically posts the created sales page to the information dissemination medium. The sales page obtained in step 6 is used as input. Specifically, it posts to the social networking site using an API. The output is the sales page published on the social networking site.

[1031] Step 8: The user uploads product information to the platform to sell their used clothing. The input is an image of the used clothing and basic information provided by the user. Specifically, the server uses the generative AI model to automatically generate an image of a model wearing the clothing and a product description, and creates a sales page. The output is a sales page for the used clothing.

[1032] (Application example 2)

[1033] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1034] With conventional apparel sales systems, it was difficult to efficiently sell long-term inventory items, and it was also difficult to suggest outfits that combined products from different brands. Furthermore, it was risky for consumers to make a purchase decision without actually trying on the product, which discouraged them from purchasing. There was a need to solve these issues and improve the consumer purchasing experience.

[1035] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1036] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an SNS, means for consumers to purchase via the SNS, means for generating images of the consumers' used clothing worn by models and product descriptions on the same platform and selling them, means for applying coordination to a user's avatar and performing a virtual try-on, and means for consumers to purchase items through the virtual try-on. This enables efficient sales of long-term inventory items, suggestions for coordinating outfits that combine items from different brands, and an improved consumer purchasing experience through the virtual try-on.

[1037] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[1038] "Apparel products" refers to fashion-related products such as clothing and accessories.

[1039] "Coordinating across brand boundaries" refers to proposing styling by combining products from different brands.

[1040] "Model wearing image" refers to an image of a model actually wearing the product.

[1041] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[1042] "Product description" refers to text that explains the detailed information and features of a product.

[1043] "Sales Page" refers to a web page for selling products online.

[1044] "Sale price" refers to the selling price of a product.

[1045] "SNS" is an abbreviation for social networking service, which refers to an online platform for users to share information.

[1046] "Consumer" refers to the general user who purchases the product.

[1047] "Used clothing" refers to clothing that has been previously used.

[1048] "Platform" refers to the underlying system for providing specific services or functions.

[1049] An "avatar" refers to a character or image used to represent a user in a virtual space.

[1050] "Virtual try-on" refers to users trying on products in a virtual space.

[1051] The system for implementing this invention has the following configuration. First, the server analyzes sales data and inventory data to select apparel items that have been in stock for a long time. This makes it possible to automatically select items whose sales are slowing. Next, AI is used to coordinate the selected apparel items across brand boundaries. This makes it possible to suggest combinations that go well together, even when products from different brands are combined.

[1052] The server uses a generative AI model to generate images of models wearing coordinated apparel, as well as catchy slogans and product descriptions. A sales page is then created using the generated images of models wearing the apparel, catchy slogans and product descriptions. The AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media as well.

[1053] In addition, consumers can also sell their used clothing on the same platform, generating images of models wearing the clothing and product descriptions, which will encourage consumers to reuse their used clothing and achieve zero waste.

[1054] Users can use smartphones, smart glasses, or head-mounted displays to virtually try on products. Coordination is applied to the user's avatar, allowing them to check out products through virtual try-on. This allows users to consider purchasing products without actually trying them on.

[1055] Specifically, the server uses software such as TensorFlow, OpenCV, and PIL (Python Imaging Library) to process data and perform calculations, such as loading product data, generating outfits, and running virtual try-ons.

[1056] An example of a prompt is as follows:

[1057] "Please select apparel items that have been in stock for a long time and coordinate them with items from different brands."

[1058] In this way, it will be possible to efficiently sell long-term inventory items, suggest outfits that combine products from different brands, and improve the consumer purchasing experience through virtual try-on.

[1059] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1060] Step 1: The server analyzes sales and inventory data to select apparel items that have become long-term inventory. Sales and inventory data are required as input, and a list of long-term inventory items is obtained as output. Specifically, an AI algorithm is used to automatically extract items that have remained unsold for a certain period of time.

[1061] Step 2: The server coordinates the selected apparel items across brands. It requires a list of long-term inventory items as input, and produces a list of coordinated items as output. Specifically, it uses an AI model to combine items from different brands and generate coordinated outfits that go well together.

[1062] Step 3: The server generates images of the coordinated apparel items worn by a model, a catchy slogan, and a product description. The input is a list of coordinated items, and the output is an image of the item worn by a model, a catchy slogan, and a product description. Specifically, a generative AI model is used to automatically generate images and text that highlight the appeal of the items.

[1063] Step 4: The server creates a sales page using the generated model image, catch phrase, and product description. The inputs are the model image, catch phrase, and product description, and the output is a sales page. Specifically, a web page generation tool is used to automatically create a sales page that includes detailed product information.

[1064] Step 5: The server automatically sets the sale price for the sales page. The sales page is required as input, and the sales page with the sale price set is obtained as output. Specifically, an AI algorithm is used to set the optimal sales price taking into account the market price and inventory status.

[1065] Step 6: The server automatically posts the created sales page to social media. The input is a sales page with a set sales price, and the output is the sales page posted to social media. Specifically, the server uses the social media API to post a link to the sales page along with a tagline and image.

[1066] Step 7: The user performs a virtual try-on using a smartphone, smart glasses, or a head-mounted display. The input is a list of products coordinated with the user's avatar, and the output is the results of the virtual try-on. Specifically, the coordinated items are applied to the user's avatar, and the user checks the appearance of the products through the virtual try-on.

[1067] Step 8: The user purchases the product through the virtual try-on. The input is the results of the virtual try-on, and the output is the product for which the purchase procedure has been completed. Specifically, the user checks the results of the virtual try-on and proceeds with the purchase procedure for the product they like.

[1068] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1069] "Example 1"

[1070] As one embodiment of the present invention, a system is provided that includes an emotion engine that uses AI to recognize consumer emotions. This emotion engine analyzes emotions from consumer social media posts. For example, it analyzes consumer emotions such as joy, sadness, surprise, anger, and fear from text, images, videos, etc. posted by consumers on social media. The analysis is performed using technologies such as text analysis, image recognition, and video analysis. The analyzed emotions are used to adjust product coordination. For example, if a consumer posts many times expressing joy, apparel items in bright colors can be selected and coordinated. On the other hand, if a consumer posts many times expressing surprise, apparel items with unique designs can be selected and coordinated. This makes it possible to suggest products that match the consumer's emotions, thereby increasing the consumer's desire to purchase.

[1071] "Example 2"

[1072] As one embodiment of the present invention, a system is provided that includes an emotion engine that uses AI to recognize consumer emotions. This emotion engine analyzes emotions from consumer social media posts. For example, it analyzes consumer emotions such as joy, sadness, surprise, anger, and fear from text, images, videos, etc. posted by consumers on social media. The analysis is performed using technologies such as text analysis, image recognition, and video analysis. The analyzed emotions are used to adjust product coordination. For example, if a consumer posts many times expressing joy, apparel items in bright colors can be selected and coordinated. On the other hand, if a consumer posts many times expressing surprise, apparel items with unique designs can be selected and coordinated. This makes it possible to suggest products that match the consumer's emotions, thereby increasing the consumer's desire to purchase.

[1073] The processing flow of each embodiment will be described below.

[1074] "Example 1"

[1075] Step 1: AI collects consumer social media posts, including text, images, and video data.

[1076] Step 2: The emotion engine analyzes the collected data to determine consumer emotions, using technologies such as text analysis, image recognition, and video analysis to extract emotions such as joy, sadness, surprise, anger, and fear.

[1077] Step 3: The AI ​​coordinates product outfits based on the analyzed emotions. For example, if a consumer posts a lot of content expressing happiness, it will select and coordinate brightly colored apparel items. On the other hand, if a consumer posts a lot of content expressing surprise, it will select and coordinate apparel items with unique designs.

[1078] Step 4: AI recommends coordinated products to consumers. This makes it possible to recommend products that match the consumer's emotions, increasing their desire to purchase.

[1079] Example 1

[1080] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1081] Conventional apparel sales systems have made it difficult to efficiently sell long-term inventory items or propose products that match consumer sentiment. They also lacked mechanisms for suggesting outfits that combine products from different brands or for encouraging consumers to reuse secondhand items. This has led to problems of backlogged inventory and a decline in consumer purchasing motivation.

[1082] The identification process by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means. In this invention, the server includes means for selecting long-term inventory items, means for coordinating the selected items across brands, means for generating images of the coordinated items worn by a model, a catchy slogan, and a product description, means for creating a sales page using the generated images of the items worn by a model, the catchy slogan, and the product description, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing platform, means for allowing consumers to purchase the items via the information sharing platform, means for generating images of the items worn by a model and a product description for selling the consumer's used items on the same platform, and means for analyzing consumer emotions and suggesting product coordination that matches the consumer's emotions. This enables efficient sales of long-term inventory items, suggestions for products that match consumer emotions, suggestions for coordinating items that combine products from different brands, and promotion of reuse of consumer used items.

[1083] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[1084] "Means for selecting" refers to a method or device for selecting an object based on specific conditions.

[1085] "Means for coordinating" refers to a method or device for proposing a combination of multiple products.

[1086] "Model wearing image" refers to an image of a model wearing a product.

[1087] A "catchphrase" is a short sentence that succinctly expresses the features and appeal of a product.

[1088] "Product description" refers to text that explains the detailed information and features of a product.

[1089] "Sales Page" refers to a web page that displays product information and allows consumers to purchase the product.

[1090] "Sale price" refers to the selling price of a product.

[1091] An "information sharing platform" refers to online services, such as social media and websites, that allow for the widespread sharing of information.

[1092] "Consumer" refers to the general user who purchases goods.

[1093] "Used goods" refers to goods that have been previously used.

[1094] "Means for analyzing emotions" refers to methods or devices for reading emotions from text, images, etc.

[1095] A "generative AI model" refers to a model that uses artificial intelligence to generate text and images.

[1096] A "prompt sentence" refers to an input sentence that gives instructions to a generative AI model.

[1097] The present invention is a system for efficiently selling products that have been in stock for a long time and for proposing products that match consumer sentiment. Specific embodiments of this system will be described below.

[1098] Hardware and software used

[1099] This system operates with three main components: a server, a terminal, and a user. The server manages sales and inventory data using a database management system (e.g., MySQL). Programs are developed using Python, with TensorFlow and PyTorch as the main libraries for training and inference of AI models. Generative AI models include BERT, GPT-3, and DALL-E. The terminal provides an interface for users to access the system through a web browser.

[1100] Program processing

[1101] The server first collects sales and inventory data. Specifically, it connects to a MySQL database and queries the sales history for the past six months and the current inventory status. Next, it analyzes the collected data and selects products that have been in stock for a long time. This analysis is performed using the Python pandas library.

[1102] Based on the selected products, the server generates coordinated outfits that go well together, even if the items are from different brands. It uses a generative AI model (e.g., GPT-3) to suggest product combinations. Images of models wearing the outfits, catchy slogans, and product descriptions are automatically created using a generative AI model (e.g., DALL-E or GPT-3).

[1103] Furthermore, the server automatically sets the sales price for the sales page using AI and posts the created sales page to social media, allowing consumers to purchase products via social media. The AI ​​also generates images of models wearing the products and product descriptions on the same platform, and sells consumers' used clothing. This encourages consumers to reuse their used clothing and achieves zero waste.

[1104] The server analyzes emotions from consumers' social media posts and suggests outfits that match those emotions. It uses NLP and image recognition technology to analyze emotions. Specifically, it uses the Hugging Face Transformers library and OpenCV.

[1105] Specific examples

[1106] Below is an example of a prompt sentence to input to the generative AI model.

[1107] 1. Analysis of sales and inventory data

[1108] Prompt: "Based on sales and inventory data from the past six months, please select the product that has been in stock for a long time."

[1109] 2. Coordinate Generation

[1110] Prompt: "Using the selected products, suggest a coordinated look that works well together, even if it's from a different brand."

[1111] 3. Model Wearing Images and Copy Generation

[1112] Prompt: "Generate a model image, tagline, and product description based on this outfit."

[1113] 4. Sentiment analysis and coordination

[1114] Prompt: "Analyze the emotions of consumers in their social media posts and suggest product coordination that matches those emotions."

[1115] Using these prompts, the system can automatically analyze the data and generate optimal outfits and sales pages.

[1116] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1117] Step 1: Collect sales and inventory data

[1118] The server collects sales and inventory data from the database. Specifically, it connects to the MySQL database and queries the sales history for the past six months and the current inventory status. The input requires database connection information and a query statement, and the output is sales and inventory data.

[1119] Specific behavior:

[1120] The server executes the SQL query SELECT FROM sales WHERE date >= DATE_SUB(CURDATE(), INTERVAL 6 MONTH).

[1121] The server executes the SQL query SELECT FROM inventory to obtain the current inventory status.

[1122] Step 2: Select long-term inventory items

[1123] The server analyzes the collected data and selects products that have been in long-term inventory. It uses the Python pandas library to identify products with sluggish sales. Sales data and inventory data are required as input, and a list of long-term inventory items is obtained as output.

[1124] Specific behavior:

[1125] The server uses the pandas groupby function to aggregate sales for each product.

[1126] The server filters out items whose sales are below a certain threshold.

[1127] Step 3: Generate coordinates

[1128] The server generates coordinated outfits based on the selected items, even for items from different brands, and uses a generative AI model (e.g., GPT-3) to suggest combinations of items. The input is a list of long-term inventory items, and the output is a suggested outfit.

[1129] Specific behavior:

[1130] The server inputs a prompt to the generative AI model: "Using the selected products, please suggest a coordinated look that goes well with other products from different brands."

[1131] The server receives the generated coordinates and stores them in a database.

[1132] Step 4: Generate model images and tagline

[1133] The server uses a generative AI model (e.g., DALL-E or GPT-3) to generate images of a model wearing coordinated products, a catchy slogan, and a product description. The input is a coordination suggestion, and the output is an image of a model wearing the product, a catchy slogan, and a product description.

[1134] Specific behavior:

[1135] The server inputs a prompt to the generative AI model: "Based on this outfit, please generate an image of the model wearing the outfit, a catchy slogan, and a product description."

[1136] The server receives the generated images and text and stores them as materials for the sales page.

[1137] Step 5: Create a sales page and set your sale price

[1138] The server creates a sales page based on the generated materials, and the AI ​​automatically sets the sale price. Pricing takes into account past sales data and market trends. The inputs include images of models wearing the products, a catchy slogan, and a product description, and the output is a completed sales page.

[1139] Specific behavior:

[1140] The server generates the sales page using the HTML template.

[1141] The server inputs a prompt statement to the AI ​​model: "Please set the optimal sale price for this product."

[1142] The server reflects the set price on the sales page.

[1143] Step 6: Post your sales page on social media

[1144] The server automatically posts the completed sales page to social media. The posting is done using the social media API. The input required is the sales page URL and a tagline, and the output is the completed posting to social media.

[1145] Specific behavior:

[1146] The server connects to the SNS API using an authentication token.

[1147] The server posts the sales page URL and the generated catchphrase to social media.

[1148] Step 7: Selling your used clothing

[1149] Users upload their own secondhand clothes to the platform. The server automatically generates images and descriptions of the uploaded secondhand clothes using a generative AI model. The input is an image of the secondhand clothes, and the output is an image of a model wearing the clothes and a product description.

[1150] Specific behavior:

[1151] Users upload images of used clothing.

[1152] The server inputs a prompt to the generative AI model: "Please generate an image of a model wearing this vintage clothing item and a product description."

[1153] The server reflects the generated images and text on the sales page.

[1154] Step 8: Sentiment analysis and coordination adjustment

[1155] The server analyzes emotions from consumers' social media posts and suggests outfits that match those emotions. Emotions are analyzed using NLP and image recognition technology. The input is the consumer's social media post, and the output is outfit suggestions that match the emotion.

[1156] Specific behavior:

[1157] The server retrieves the consumer's posts using the SNS API.

[1158] The server uses Hugging Face's Transformers library to analyze the sentiment of the text.

[1159] Based on the analysis results, the server inputs a prompt statement to the generative AI model: "Please suggest a product coordination that matches the consumer's emotions."

[1160] The server proposes the generated coordinates to the consumer.

[1161] (Application example 1)

[1162] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1163] In conventional apparel sales systems, long-term inventory items often remained unsold, making inventory management difficult. It was also difficult to propose coordinated outfits across brand boundaries, which prevented consumers from motivating them to purchase. Furthermore, product proposals did not take into account consumer emotions, making it impossible to provide optimal products to individual consumers. To solve these issues, efficient sales promotion of long-term inventory items and coordinated outfit proposals based on consumer emotions were necessary.

[1164] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1165] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for consumers to purchase via the information sharing service, means for generating images of models wearing the apparel items and product descriptions for consumers' used clothing on the same platform and selling them, means for analyzing consumer sentiment and adjusting coordination based on the sentiment, means for generating product pages using a generative AI model, and means for inputting prompts into the generative AI model to generate images and text. This enables efficient sales promotion of long-term inventory items and optimal coordination suggestions based on consumer sentiment.

[1166] "Long-term inventory" refers to products that have been stored as inventory for a certain period of time or longer and remain unsold in the normal sales cycle.

[1167] "Apparel products" refers to wearable products such as clothing and fashion accessories.

[1168] "Creating outfits that transcend brand boundaries" refers to combining products from different brands to create a unified fashion style.

[1169] "Model wearing images" refers to photographs or images of a model actually wearing the product.

[1170] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[1171] "Product description" refers to text that explains the detailed information, features, and usage of a product.

[1172] "Sales Page" refers to a web page for selling products online.

[1173] "Sale price" refers to the selling price of a product.

[1174] "Information sharing services" refer to online platforms for sharing information, such as social media and blogs.

[1175] "Consumer" refers to an individual who purchases goods or services.

[1176] "Secondhand clothing" refers to previously used clothing or fashion items.

[1177] "Platform" refers to the underlying system for providing specific services or functions.

[1178] "Analyzing emotions" refers to reading people's emotions from text, images, videos, etc.

[1179] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to generate images or text.

[1180] A "prompt" refers to an instruction or question that is input into a generative AI model.

[1181] The system for implementing this invention is composed of three main elements: a server, a terminal, and a user. The specific operation of each element and the hardware and software used will be explained below.

[1182] Server Operation

[1183] The server includes programs to perform the following functions:

[1184] 1. Select long-term inventory items:

[1185] The server analyzes the inventory database and selects apparel items that have been in stock for a certain period of time, using an artificial intelligence (AI) model.

[1186] 2. Coordination suggestions that transcend brand boundaries:

[1187] The server then combines selected long-term inventory items across brands to suggest a cohesive fashion style, again using an AI model.

[1188] 3. Generate model images and product descriptions:

[1189] The server uses a generative AI model to generate images of a model wearing the coordinated products, a catchy copy, and a product description. The generative AI model generates images and text by inputting a prompt sentence.

[1190] 4. Sales page creation and automatic posting:

[1191] The server creates a sales page using the generated images of the model wearing the product, a catchy slogan, and a product description, and automatically posts it to an information sharing service (SNS).

[1192] 5. Consumer sentiment analysis:

[1193] The server analyzes the emotions of consumers in their social media posts and adjusts outfits based on those emotions. This process is performed using an emotion analysis engine.

[1194] Device behavior

[1195] The terminal (smartphone) provides an interface for users to view and purchase apparel products.

[1196] 1. Product browsing:

[1197] The user can view the sales page generated by the server through the terminal.

[1198] 2. Purchase procedure:

[1199] Users can purchase products through the terminal, and the purchase procedure is carried out through an electronic payment service.

[1200] User Actions

[1201] The user uses the terminal to browse and purchase apparel items.

[1202] 1. Social Media Posts:

[1203] Users express their feelings by posting on the SNS, and these posts are analyzed by the server and used to adjust coordination.

[1204] Hardware and software used

[1205] Hardware: Servers, smartphones

[1206] Software: Python, transformers library, requests library, sentiment analysis engine, generative AI model

[1207] Examples of specific examples and prompts

[1208] As a concrete example, the following is an example of a prompt sentence to be input to the generative AI model.

[1209] Sentiment analysis prompt: "Analyze the sentiment of the following text: 'I love this new dress!'"

[1210] Image generation prompt: "Generate an image of a Red Dress."

[1211] Text generation prompt: "Generate a catchy description for a Red Dress."

[1212] This allows the server to efficiently promote long-term inventory items and propose optimal outfits based on consumer sentiment.

[1213] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1214] Step 1: The server analyzes the inventory database to select apparel items that have become long-term inventory. It uses the inventory database as input and obtains a list of long-term inventory items as output. This process uses an artificial intelligence (AI) model to select items that have been in stock for more than a certain period of time (e.g., 180 days).

[1215] Step 2: The server coordinates the selected long-term inventory items across brands. It uses a list of long-term inventory items as input and obtains coordinated product combinations as output. This process also uses an AI model to combine products from different brands to suggest a cohesive fashion style.

[1216] Step 3: The server uses the generative AI model to generate an image of a model wearing the coordinated items, a catchy slogan, and a product description. It uses the coordinated product combination as input and obtains an image of a model wearing the items, a catchy slogan, and a product description as output. The generative AI model receives a prompt sentence as input and generates images and text. Examples of specific prompt sentences include "Generate an image of a Red Dress." and "Generate a catchy description for a Red Dress."

[1217] Step 4: The server creates a sales page using the generated model images, taglines, and product descriptions. The generated images and text are used as input, and a sales page is obtained as output. This sales page includes detailed product information, prices, a purchase button, and more.

[1218] Step 5: The server automatically sets the sale price for the sales page. It uses the sales page and market data as input and sets the optimal sale price as output. This process uses a pricing algorithm to determine the price based on competitor prices and demand forecasts.

[1219] Step 6: The server automatically posts the created sales page to an information sharing service (SNS). The sales page is used as input and the SNS post is made as output. This process uses the SNS API, and the posted content includes an image of the model wearing the product, a catchy slogan, and a product description.

[1220] Step 7: The user browses the sales page generated by the server through their terminal and purchases the product. The sales page is used as input and the purchase procedure is performed as output. The purchase procedure is performed through an electronic payment service.

[1221] Step 8: The server analyzes the sentiment of the consumer's social media posts and adjusts the outfit based on that sentiment. It uses the social media post data as input and obtains the sentiment analysis results as output. This process uses a sentiment analysis engine to read emotions from text, images, and videos. An example of a specific prompt is "Analyze the sentiment of the following text: 'I love this new dress!'"

[1222] Step 9: The server also generates images of models wearing the consumer's used clothing and product descriptions for sale on the same platform. Used clothing information is used as input, and images of models wearing the clothing and product descriptions are obtained as output. This process uses a generative AI model, which generates images and text based on prompts.

[1223] Example 2

[1224] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1225] Conventional apparel sales systems were unable to efficiently process long-term inventory, resulting in backlogged inventory and declining sales. Furthermore, it was difficult to propose outfits that transcended brand boundaries, preventing consumers from purchasing. Furthermore, product suggestions were not based on consumer sentiment, making it difficult to provide services that meet the needs of individual consumers. To solve these issues, efficient inventory management utilizing AI technology and outfit suggestions based on consumer sentiment were required.

[1226] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1227] In this invention, the server includes means for selecting long-term inventory items, means for coordinating the selected items across brand boundaries, means for generating images of the coordinated items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the items worn by models, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing platform, means for consumers to purchase via the information sharing platform, means for generating images of the items worn by models and product descriptions and selling consumers' used items on the same platform, and means for analyzing consumer sentiment and adjusting product coordination based on the analysis results. This enables efficient processing of long-term inventory items, attractive coordination suggestions across brand boundaries, and individual responses based on consumer sentiment.

[1228] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[1229] "Goods" means goods sold to consumers.

[1230] "Brand" refers to a name or logo that is used to identify the products or services offered by a particular company or manufacturer.

[1231] "Coordination" refers to proposing a combination of multiple products.

[1232] "Model Wearing Image" refers to a photograph of a model wearing a product.

[1233] A "catchphrase" is a short sentence that succinctly expresses the features and appeal of a product.

[1234] "Product description" refers to text that explains the detailed information and features of a product.

[1235] "Sales Page" means a web page for selling products online.

[1236] "Sale Price" refers to the price at which an item is sold.

[1237] An "information sharing platform" refers to online services for sharing information, such as social media and websites.

[1238] "Consumer" refers to an individual who purchases a product.

[1239] "Used goods" refers to goods that have been previously used.

[1240] "Emotion" refers to the psychological state of the consumer.

[1241] "Analysis" refers to analyzing data to extract meaning.

[1242] "Artificial intelligence" refers to the technology that allows computers to act in a way that mimics human intelligence.

[1243] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[1244] The server first collects sales and inventory data from the database, using SQL queries to retrieve past sales history and current inventory status. It then uses Python and TensorFlow to analyze the collected data and identify slowing sales. This allows it to select products that have been in stock for a long time.

[1245] For the selected products, the server uses a generative AI model (for example, OpenAI's GPT-4) to generate a coordinated outfit that transcends brand boundaries. Specifically, the following prompt sentence is input into the generative AI model:

[1246] "This item has been in stock for a long time, so please coordinate it with other brand items. Please generate a model image, tagline, and product description."

[1247] Based on the generated outfit information, the server uses an HTML template to create a sales page, which includes images of the product being worn by a model, a tagline, and a product description. Furthermore, a pricing model is built using Python and Scikit-learn to automatically set the sale price for the sales page.

[1248] The created sales page is automatically posted to an information-sharing platform using SNS APIs, allowing consumers to purchase products via the information-sharing platform.

[1249] Users can also sell their own second-hand goods on the platform. When a user uploads an image of their used clothing, the server uses a generative AI model to generate an image of the model wearing the item and a product description. Specifically, the following prompt sentence is input into the generative AI model:

[1250] "Please generate an image of a model wearing this vintage clothing along with a product description."

[1251] Furthermore, the server uses NLP technology (e.g., the BERT model) and image recognition technology (e.g., OpenCV) to analyze consumer sentiment. Based on the analysis results, it coordinates products to match the consumer's sentiment. For example, if a consumer posts many times expressing happy emotions, it selects and coordinates products in bright colors.

[1252] In this way, it becomes possible to efficiently process long-term inventory items, propose attractive coordination that transcends brand boundaries, and respond to individual customer needs based on consumer sentiment.

[1253] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1254] Step 1: The server collects sales and inventory data from the database. The input is the database connection information and an SQL query. Specifically, the server executes the SQL query to obtain past sales history and current inventory status. The output is sales and inventory data.

[1255] Step 2: The server analyzes the collected sales and inventory data to identify products that have become long-term inventory. The server uses sales and inventory data as input. Specifically, the server uses Python and TensorFlow to predict sales trends and identify products with slowing sales. The output is a list of long-term inventory products.

[1256] Step 3: The server uses the generative AI model to generate a cross-brand coordination based on the long-term inventory product list. The long-term inventory product list and a prompt sentence are used as input. Specifically, the server inputs the following prompt sentence into the generative AI model:

[1257] "This item has been in stock for a long time, so please coordinate it with other brand items. Please generate a model image, tagline, and product description."

[1258] As an output, coordinate information is obtained.

[1259] Step 4: The server creates a sales page based on the generated coordination information. The coordination information and HTML template are used as input. Specifically, the server embeds an image of a model wearing the item, a catchy slogan, and a product description into the HTML template to generate the sales page. The sales page is obtained as output.

[1260] Step 5: The server automatically sets the sale price for the sales page. The sales page and pricing model are used as input. Specifically, the server uses Python and Scikit-learn to build a pricing model and predict the sale price for the sales page. The output is the sales page with the sale price set.

[1261] Step 6: The server automatically posts the created sales page to the information sharing platform. The sales page and the information sharing platform's API are used as input. Specifically, the server posts the sales page using the information sharing platform's API. The output is the sales page posted on the information sharing platform.

[1262] Step 7: The user uploads an image of their used clothing to sell on the platform. The image of the used clothing is used as input. Specifically, the server uses the generative AI model to generate an image of the model wearing the clothing and a product description. The following prompt sentence is input into the generative AI model:

[1263] "Please generate an image of a model wearing this vintage clothing along with a product description."

[1264] The output is an image of a model wearing the product and a product description.

[1265] Step 8: The server analyzes the consumer's sentiment and adjusts product coordination based on the analysis results. The input is the consumer's social media posting data. Specifically, the server analyzes sentiment using NLP technology (e.g., the BERT model) or image recognition technology (e.g., OpenCV). The output is the sentiment analysis result. Based on the analysis results, the server inputs the following prompt sentence into the generative AI model:

[1266] "This consumer often posts expressing joy, so choose apparel items in bright colors and coordinate them together."

[1267] As an output, emotion-based coordination information is obtained.

[1268] (Application example 2)

[1269] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1270] In the conventional apparel sales system, there was a lack of efficient ways to sell long-term inventory, making inventory management difficult. It was also difficult to propose coordination that transcended the boundaries of brands, which prevented consumers from motivating purchases. Furthermore, product proposals were not based on consumer emotions, making it impossible to provide the best products for each individual consumer.

[1271] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1272] In this invention, the server includes means for selecting apparel items that have been in long-term inventory, means for coordinating the selected apparel items across brands, means for generating images of the coordinated apparel items worn by models, catch phrases, and product descriptions, means for creating a sales page using the generated images of the models wearing the apparel items, catch phrases, and product descriptions, means for automatically setting a sale price for the sales page, means for automatically posting the created sales page to an information sharing service, means for consumers to purchase via the information sharing service, means for generating images of the consumers wearing the apparel items on models and product descriptions and selling their used clothing on the same platform, means for analyzing consumer sentiment, and means for making product suggestions based on the analyzed sentiment. This enables efficient sales of long-term inventory items, coordination suggestions across brands, and optimal product suggestions based on consumer sentiment.

[1273] "Long-term inventory" refers to products that remain unsold for a certain period of time.

[1274] "Apparel products" refers to clothing items such as clothing and fashion accessories.

[1275] "Crossing brand boundaries" refers to combining products from different brands.

[1276] "Coordination" refers to styling multiple apparel items together.

[1277] "Model Wearing Image" refers to a photograph of a model wearing an apparel product.

[1278] A "catchphrase" is a short phrase that succinctly expresses the features and appeal of a product.

[1279] "Product description" refers to text that explains the detailed information and features of a product.

[1280] "Sales page" refers to a web page that lists product information and allows consumers to purchase the product.

[1281] "Sale price" refers to the selling price of a product.

[1282] "Information sharing services" refers to online platforms such as social networking sites.

[1283] "Consumer" refers to an individual who purchases a product.

[1284] "Used clothing" refers to clothing that has been previously purchased and used by a consumer.

[1285] "Platform" refers to a system that serves as the foundation for providing multiple functions and services.

[1286] "Analyzing emotions" refers to reading consumer emotions from data.

[1287] "Product suggestion" refers to recommending a specific product to a consumer.

[1288] "Artificial intelligence" refers to technology that allows machines to act in a way that mimics human intelligence.

[1289] A system for implementing the present invention has the following configuration.

[1290] First, the server selects apparel items that have become long-term inventory. This is achieved by analyzing sales and inventory data. Specifically, the server retrieves sales and inventory data from the database and selects items that have remained unsold for a certain period of time as long-term inventory items.

[1291] Next, the selected apparel items are coordinated across brands. The server uses artificial intelligence to combine products from different brands and propose optimal outfits. For example, it uses a clustering algorithm to group products based on color or design characteristics and find combinations that go well together.

[1292] Images of models wearing coordinated apparel products, catch phrases, and product descriptions are automatically generated using a generative AI model. The server inputs the following prompt sentence into the generative AI model and obtains the generated text.

[1293] Example prompt sentence:

[1294] Product Name: Refreshing Summer Dress

[1295] Description: This dress features a lightweight material and vibrant color. It's designed to be comfortable even on hot summer days, making it perfect for relaxing at the beach or a cafe.

[1296] The server creates a sales page using the generated images of models wearing the product, catchy slogans, and product descriptions. The server automatically sets the sale price on the sales page. This is achieved by calculating the optimal price based on past sales data and market trends.

[1297] The created sales page is automatically posted to the information sharing service by the server, allowing consumers to purchase the product via the information sharing service.

[1298] Consumers can also sell their used clothing on the same platform. Based on the information provided by consumers, the server uses a generative AI model to generate images of models wearing the clothing and product descriptions, and creates a sales page.

[1299] To analyze consumer sentiment, the server collects social media posts from consumers and analyzes them using a sentiment analysis algorithm. Based on the analyzed sentiment, the server makes optimal product recommendations to consumers. For example, if a consumer posts frequently expressing joy, the server will recommend brightly colored apparel products.

[1300] In this way, the server enables efficient sales of long-term inventory items, coordination suggestions that transcend brand boundaries, and optimal product suggestions based on consumer sentiment.

[1301] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1302] Step 1: The server retrieves sales and inventory data from the database. This collects basic data for selecting apparel items that have become long-term inventory. The input is sales and inventory data, and the output is the integration result of these data.

[1303] Step 2: The server analyzes the acquired data and selects items that have remained unsold for a certain period of time as long-term inventory items. Specifically, it filters out items with sales below a certain level. The input is the integrated sales data and inventory data, and the output is a list of long-term inventory items.

[1304] Step 3: The server uses artificial intelligence to coordinate the selected long-term inventory items across brands. It uses a clustering algorithm to group products based on color and design characteristics and find combinations that go well together. The input is a list of long-term inventory items, and the output is a list of coordinated items.

[1305] Step 4: The server uses the generative AI model to generate images of coordinated apparel items worn by a model, a tagline, and a product description. The prompt text is input into the generative AI model and the generated text is obtained. The input is a list of coordinated items and the prompt text, and the output is the generated images of the items worn by a model, a tagline, and a product description.

[1306] Step 5: The server creates a sales page using the generated model images, catchphrases, and product descriptions. The input is the generated content, and the output is the completed sales page.

[1307] Step 6: The server automatically sets the sale price on the sales page. It calculates the optimal price based on past sales data and market trends. The input is the sales page and past sales data, and the output is the set sale price.

[1308] Step 7: The server automatically posts the created sales page to the information sharing service, allowing consumers to purchase the product via the information sharing service. The input is the completed sales page, and the output is the result of posting to the information sharing service.

[1309] Step 8: The server uses the generative AI model to generate images of a model wearing the item and a product description based on the consumer's used clothing information, and creates a sales page. The input is the consumer's used clothing information and prompt text, and the output is the generated image of a model wearing the item and a product description.

[1310] Step 9: The server collects the consumer's SNS posts and analyzes their emotions using a sentiment analysis algorithm. The input is the consumer's SNS posts, and the output is the analyzed sentiment data.

[1311] Step 10: The server makes optimal product recommendations to the consumer based on the analyzed emotions. For example, if a consumer posts a lot of happy emotions, it will suggest brightly colored apparel products. The input is the analyzed emotion data, and the output is a list of suggested products.

[1312] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1313] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1314] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1315] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1316] [Fourth embodiment]

[1317] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1318] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1319] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1320] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1321] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1322] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1323] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1324] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1325] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1326] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1327] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1328] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1329] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1330] "Example 1"

[1331] One embodiment of the present invention is an apparel sales system using AI. In this system, AI first selects apparel items that have been in stock for a long time. For example, it analyzes sales and inventory data and automatically selects items with slowing sales. Next, it coordinates the selected items across brands. This makes it possible to suggest combinations that work well together, even among products from different brands. Images of the coordinated items worn by models, catchy slogans, and product descriptions are automatically generated by the AI. This enables the efficient creation of sales pages that maximize the appeal of the products. Furthermore, the AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media. Furthermore, AI also generates images of models wearing the items and product descriptions for the consumer's used clothing on the same platform and sells them. This encourages consumers to reuse their used clothing and achieves zero waste.

[1332] "Example 2"

[1333] One embodiment of the present invention is an apparel sales system using AI. In this system, AI first selects apparel items that have been in stock for a long time. For example, it analyzes sales and inventory data and automatically selects items with slowing sales. Next, it coordinates the selected items across brands. This makes it possible to suggest combinations that work well together, even among products from different brands. Images of the coordinated items worn by models, catchy slogans, and product descriptions are automatically generated by the AI. This enables the efficient creation of sales pages that maximize the appeal of the products. Furthermore, the AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase products via social media. Furthermore, AI also generates images of models wearing the items and product descriptions for the consumer's used clothing on the same platform and sells them. This encourages consumers to reuse their used clothing and achieves zero waste.

[1334] The processing flow of each embodiment will be described below.

[1335] "Example 1"

[1336] Step 1: The AI ​​selects apparel items that have been in stock for a long time. Specifically, it analyzes sales and inventory data and automatically selects items whose sales are slowing.

[1337] Step 2: Coordinate the selected products across brands, making it possible to propose combinations that go well together even when products from different brands are used.

[1338] Step 3: Images of models wearing the coordinated products, along with catchy slogans and product descriptions, are automatically created by generative AI. This makes it possible to efficiently create sales pages that maximize the appeal of the products.

[1339] Step 4: The AI ​​automatically sets the sales price for the sales page and automatically posts the created sales page to social media, allowing consumers to purchase the product via social media.

[1340] Step 5: Consumers can also purchase their used clothing on the same platform, with AI generating images of models wearing the clothing and product descriptions. This will encourage consumers to reuse their used clothing and achieve zero waste.

[1341] Example 1

[1342] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1343] Conventional apparel sales systems lacked efficient methods for selling long-term inventory. Furthermore, there were no effective methods for suggesting outfits that combined products from different brands, or for reusing consumers' old clothes. This resulted in increased inventory backlog and waste, and reduced sales efficiency.

[1344] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1345] In this invention, the server includes means for collecting sales data and inventory data, means for analyzing the collected data to select long-term inventory items, means for proposing outfits by combining the selected items with items from different brands, means for generating images of the coordinated items worn by models, catch phrases, and product descriptions using a generative AI model, means for creating a sales page using the generated images of the coordinated items worn by models, catch phrases, and product descriptions, means for automatically setting sale prices by executing a pricing algorithm based on past sales data, means for automatically posting the created sales page to social media, means for allowing consumers to purchase via social media, and means for generating images of the items worn by models and product descriptions and selling consumers' used clothing on the same platform. This makes it possible to efficiently sell long-term inventory items, propose attractive outfits by combining items from different brands, and encourage consumers to reuse used clothing.

[1346] "Sales data" refers to data that includes transaction information regarding the sale of apparel products.

[1347] "Inventory data" is data that indicates the current inventory status of apparel products.

[1348] "Long-term inventory" refers to apparel products that remain unsold for a certain period of time.

[1349] "AI" is a system that uses artificial intelligence technology to analyze data and make predictions.

[1350] "Coordination" is a fashion combination that combines products from different brands.

[1351] A "generative AI model" is an artificial intelligence model for generating natural language and images.

[1352] "Model wearing image" refers to an image of a model wearing an apparel product.

[1353] A "catchphrase" is a short promotional phrase that draws out the appeal of a product.

[1354] A "product description" is a text that explains the features and benefits of a product.

[1355] A "sales page" is a web page for selling apparel products online.

[1356] "Sale price" refers to the selling price of a product.

[1357] "SNS" refers to social networking services, which are online platforms for users to share information.

[1358] "Consumer" refers to a customer who purchases apparel products.

[1359] "Used clothing" refers to second-hand apparel items that consumers have previously purchased and wish to resell.

[1360] The present invention relates to an apparel product sales system using AI. A specific embodiment of this system will be described below.

[1361] The server first collects sales and inventory data. To do this, MySQL is used as the database management system (DBMS). The server retrieves sales data for the past six months and current inventory data using SQL queries.

[1362] The server then analyzes the collected data and selects items that have become long-term inventory. This analysis is performed using Python and TensorFlow. Items whose sales have fallen below a certain threshold are selected as long-term inventory. For example, items whose sales have been less than 10 units in the past six months are selected.

[1363] For the selected items, the server proposes outfits by combining them with products from different brands. To propose outfits, OpenCV is used to evaluate color compatibility and Scikit-learn is used to evaluate style compatibility. Specifically, a color wheel is used to evaluate color compatibility, and a clustering algorithm is used to evaluate style compatibility.

[1364] Next, the server uses a generative AI model (e.g., GPT-4) to generate an image of a model wearing the coordinated items, along with a catchy slogan and product description. The following prompt is input to the generative AI model:

[1365] "This is a long-term inventory item. Please generate a model image, tagline, and product description based on the following information: Product name: XXXX, Brand: YYYY, Color: ZZZZ, Style: AAAA"

[1366] Based on the generated information, the server uses an image generation API (e.g., DALL-E) to generate an image of the model wearing the product.

[1367] The server then runs a pricing algorithm based on past sales data to determine the optimal sale price. To do this, it uses XGBoost to build a pricing model and predict the optimal sale price for each product.

[1368] Next, the server creates a sales page based on the generated model images, tagline, product description, and sale price. The sales page is built using HTML and CSS. The link to the created sales page and the generated tagline are automatically posted to social media using an API.

[1369] Additionally, users can input photos of used clothing and basic information (brand, color, style, etc.) from their devices. The server uses this information to generate an image of the clothing being worn by a model and a product description using a generative AI model. For example, the following prompt sentence can be input into the generative AI model:

[1370] "This used clothing item is for sale by a user. Please generate a model image, tagline, and product description based on the following information: Product name: XXXX, Brand: YYYY, Color: ZZZZ, Style: AAAA"

[1371] Based on the generated information, the server creates a sales page and automatically posts it to social media, allowing users to easily sell their used clothing.

[1372] In this way, the system of the pr...

Claims

1. a means for collecting sales data and inventory data to select apparel items with sales below a certain threshold; means for generating a prompt sentence instructing the user to generate a coordination by combining the selected apparel item with other items; A means for generating a coordination by combining the apparel product with other products using the generated prompt sentence and a generation AI model; A means for generating an image of a model wearing the coordinated apparel product, a catch copy, and a prompt for generating a product description; A means for generating an image of a model wearing the coordinated apparel product, a catch copy, and a product description using the generated prompt sentence and a generative AI model; A means for creating a sales page using an image of a model wearing the generated and coordinated apparel product, a catchy copy, and a product description; A system including:

2. The system of claim 1 , wherein the other products include products of a different brand than the apparel products.

3. means for generating a prompt sentence for instructing the setting of a sale price of the apparel item; a means for setting a sale price of the apparel item on the sales page based on at least one of past sales data of the apparel item and market trends using the generated prompt sentence and a generation AI model; The system of claim 1 further comprising:

Images