System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting body shape information and past purchase history of a user; means for training a machine learning model based on the collected information; and means for recommending an optimal product for the user using the trained machine learning model.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] On conventional fashion e-commerce sites, the clothes purchased by users often do not fit properly or are not the right size, resulting in returns and wasteful consumption, which has a negative impact on the environment. It is also difficult to find products that fit specific tastes or body types, creating a need for improved user experience (UX). Furthermore, when offering custom-made clothing, it is technically difficult to quickly generate appropriate designs that meet users' needs. [Means for solving the problem]

[0005] The present invention provides a system that recommends optimal products to users by collecting information on a user's body type and past purchase history and training a machine learning model based on the collected information. The system also includes a means for analyzing requests from users using an interactive AI engine, a means for searching for relevant product information based on the analysis results, and a means for displaying the search results to the user. The system also includes a means for generating designs using a generation AI based on text input from the user and providing the designs to the user. Additionally, the system includes a means for analyzing photos of the user and images of the product the user plans to purchase, using an AI model to determine their compatibility, and presenting the results to the user, thereby increasing user satisfaction and reducing wasteful consumption and environmental impact.

[0006] "User's body type information" is data indicating the user's physical dimensions and characteristics, such as height, weight, waist size, hip size, and bust size.

[0007] "Past purchase history" refers to information about products that a user has purchased in the past, and specifically includes data such as product type, size, color, purchase date and time, and brand.

[0008] A "machine learning model" is a collection of algorithms that use collected data to learn patterns and relationships and make predictions or classifications based on new data.

[0009] "Recommending" means proposing optimal products and services based on the user's preferences and needs.

[0010] An "interactive AI engine" is an artificial intelligence system that enables natural language dialogue with users, and has the ability to understand users' requests and questions and generate appropriate responses.

[0011] "Generative AI" is artificial intelligence that generates new content (such as designs or images) based on input such as text and images.

[0012] The "AI compatibility model" is an algorithm that analyzes user photos and product images and evaluates how well they match each other.

[0013] "Customization" refers to the process of creating customized products based on a user's specific needs and preferences. [Brief explanation of the drawings]

[0014] [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. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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, a 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), and an APU (Accelerated Processing Unit).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. 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."

[0035] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history, uses AI to recommend optimal fashion items, and also provides custom-made clothing based on the user's requests. This system is configured as follows.

[0036] Recommendation function implementation

[0037] server

[0038] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering.

[0039] Terminal

[0040] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[0041] User

[0042] Users enter their purchase history, body type information, etc., and when they access the site, they can receive product suggestions that suit them.

[0043] Specific examples

[0044] When a user purchases a T-shirt, the app will display the best size T-shirt and related recommended items based on their past purchase history and body type information.

[0045] Implementing conversational AI search functionality

[0046] server

[0047] The server is equipped with a conversational AI engine that analyzes natural language requests from users, searches for relevant product information from a database based on the user's input, and returns the results.

[0048] Terminal

[0049] The terminal has a chat window where users can freely input requests and questions. The terminal passes the requests to the server, which then displays the results of the server's analysis to the user.

[0050] User

[0051] For example, users can enter, "I'm looking for a casual summer dress," and receive suggestions from the conversational AI.

[0052] Specific examples

[0053] When a user types "I'm looking for a sports jacket" into the chat window, the AI ​​engine searches for the appropriate products and displays a list of recommended sports jackets.

[0054] Implementing custom features using generative AI

[0055] server

[0056] The server uses generative AI to generate clothing designs based on textual information entered by the user, such as a specific request like "I want a red dress with a belt."

[0057] Terminal

[0058] The terminal transmits the user's custom-made request to the server and displays the generated design.

[0059] User

[0060] Users can enter information into a custom request form, view the generated design, and purchase it.

[0061] Specific examples

[0062] A user enters "I want a blue chiffon blouse" into the custom-made request form, and the generative AI generates a design based on that text and presents it to the user.

[0063] Implementation of AI compatibility assessment function

[0064] server

[0065] The server uses an AI model to analyze the photos uploaded by the user and images of the products they plan to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[0066] Terminal

[0067] The device sends the user's image to the server and displays the results of the compatibility assessment, which can be used as a reference when making a purchasing decision.

[0068] User

[0069] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server.

[0070] Specific examples

[0071] When a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the image and displays a result such as "This dress fits you 88%."

[0072] By combining these features, users can easily find the products that best suit them, minimizing post-purchase dissatisfaction and waste. Furthermore, by allowing users to easily create and purchase customized, made-to-order clothing, the service not only increases user satisfaction but also contributes to the realization of sustainable fashion.

[0073] The processing flow will be explained below.

[0074] Specific processing of the recommendation function

[0075] Step 1:

[0076] Server: When a user registers on the site, they are asked to enter their body information (height, weight, waist size, etc.) and this information is saved in a database.

[0077] Step 2:

[0078] Server: When a user completes a product purchase, the purchase history (product name, category, size, color, etc.) is saved in a database.

[0079] Step 3:

[0080] Server: Once a certain amount of data has been accumulated, the user data is used to train a machine learning model, using algorithms such as Collaborative Filtering and Content-based Filtering.

[0081] Step 4:

[0082] Terminal: When a user logs in to the site, the currently logged-in user ID is sent to the server.

[0083] Step 5:

[0084] Server: Based on the submitted user ID, the trained model generates the optimal recommended products.

[0085] Step 6:

[0086] Server: Returns the generated recommended product list to the terminal.

[0087] Step 7:

[0088] Terminal: Displays the received recommended product list to the user.

[0089] Specific processing of conversational AI search function

[0090] Step 1:

[0091] User: Launches the site's conversational AI search widget and enters a question or request in natural language (e.g., "I'm looking for a casual summer dress").

[0092] Step 2:

[0093] Terminal: Sends user input to the server.

[0094] Step 3:

[0095] Server: A conversational AI engine analyzes the user's natural language input and understands their requests and questions.

[0096] Step 4:

[0097] Server: Based on the analysis results, search for related product information from the database.

[0098] Step 5:

[0099] Server: Returns search results to the device.

[0100] Step 6:

[0101] Terminal: Displays search results to the user.

[0102] Specific processing of custom features using generative AI

[0103] Step 1:

[0104] User: Enters details into a custom request form (e.g., "I would like a red dress with a belt").

[0105] Step 2:

[0106] Terminal: Sends request information to the server.

[0107] Step 3:

[0108] Server: Uses generative AI to generate designs based on user requests.

[0109] Step 4:

[0110] Server: Sends the generated design image back to the device.

[0111] Step 5:

[0112] Terminal: Displays the generated design to the user and asks for their confirmation.

[0113] Step 6:

[0114] User: Checks the generated design and confirms the order if satisfied.

[0115] Specific processing of the AI ​​compatibility judgment function

[0116] Step 1:

[0117] Users upload a photo of themselves and an image of the clothing they plan to purchase to the site.

[0118] Step 2:

[0119] Terminal: Sends the uploaded image data to the server.

[0120] Step 3:

[0121] Server: Analyzes the image using a compatibility assessment AI model. Specifically, it compares the user's photo with the image of the clothing and scores the degree of fit.

[0122] Step 4:

[0123] Server: Generates the analyzed fitness score and other feedback.

[0124] Step 5:

[0125] Server: Sends the generated fitness score and feedback back to the device.

[0126] Step 6:

[0127] Device: Presents fitness scores and feedback to the user.

[0128] These concrete steps allow users to easily find the perfect product and order custom-made clothing, minimizing post-purchase frustration and waste.

[0129] Example 1

[0130] Next, a description will be given of 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."

[0131] Conventional fashion e-commerce systems have limited functionality to provide users with optimal products, lacking in recommendation accuracy and customization capabilities. Furthermore, they lack comprehensive functionality to improve the user experience, such as conversational AI search, custom-made features, and pre-purchase compatibility assessment. As a result, users have difficulty finding the right products, which reduces their motivation to purchase.

[0132] 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.

[0133] In this invention, the server includes: means for collecting a user's physical information and historical purchase records; means for training a machine learning model based on the collected information; means for recommending appropriate products to the user using the trained machine learning model; means including an interactive AI engine for analyzing interactive requests from the user; means for searching for related product information based on the generated analysis results and displaying it to the user; means for generating designs using artificial intelligence based on the user's text input information; means for providing the generated designs to the user; and means for performing image analysis and compatibility assessment, scoring the most suitable products for the user, and displaying them. This enables a comprehensive system that integrates a variety of functions, such as product recommendations optimized for the user, search using interactive AI, custom-made generation, and compatibility assessment.

[0134] "User's physical information" refers to data related to the user's physical shape, such as height, weight, and waist size.

[0135] "Historical purchase records" refers to data and history of products purchased by a user in the past.

[0136] A "machine learning model" is a set of algorithms that uses collected data to learn user preferences and behavioral patterns and recommend optimal products.

[0137] An "interactive artificial intelligence engine" is an artificial intelligence system that uses natural language processing technology to analyze user requests and generate appropriate responses.

[0138] "Generative AI" is an AI technology that generates specific outputs (e.g., designs) based on user input.

[0139] "Image analysis" is the process of analyzing images of users and products, extracting features, and comparing and judging them.

[0140] "Compatibility assessment" is the process of quantifying how well a user and product match based on the results of image analysis.

[0141] "Scoring" refers to expressing the results of a relevance assessment as a numerical value.

[0142] "Searching for product information" refers to extracting related product data from a database based on a user request.

[0143] "Generating a design" is the process of using generative AI to create a specific design or item based on a user request.

[0144] "Recommending appropriate products to users" refers to selecting and providing products that best match the user's preferences and needs through machine learning models.

[0145] "Searching for related product information and displaying it to the user" refers to extracting related product data based on the analysis results of the interactive AI engine and displaying it on the user's device.

[0146] MODE FOR CARRYING OUT THE INVENTION

[0147] This invention relates to a fashion e-commerce system that utilizes a user's physical information and historical purchase records, uses artificial intelligence to recommend optimal fashion items, and even provides custom-made clothing based on the user's requests.

[0148] Recommendation function implementation

[0149] server

[0150] The server first collects the user's submitted physical information (e.g., height, weight, waist size) and historical purchase records and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. The model uses algorithms such as collaborative filtering and content-based filtering.

[0151] Terminal

[0152] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[0153] User

[0154] Users enter their physical information and purchase history on the e-commerce site, and when they return to the site, they can receive recommendations for products that are best suited to them.

[0155] Specific examples

[0156] When a user purchases a T-shirt, the server recommends a size "L" white T-shirt or blue jeans based on their past purchase history and physical information. The device displays this information on the e-commerce site, allowing the user to consider the purchase.

[0157] Implementing conversational AI search functionality

[0158] server

[0159] The server is equipped with a conversational AI engine that analyzes requests made in natural language by the user, searches for relevant product information from a database based on the analyzed request, and returns the results.

[0160] Terminal

[0161] The terminal has a chat window where users can freely enter questions or requests. The terminal sends these to the server, which then displays the analysis results to the user.

[0162] User

[0163] Users can type "I'm looking for a casual summer dress" into the chat window and receive suggestions from the conversational AI.

[0164] Specific examples

[0165] When a user types "I'm looking for a sports jacket" into a chat window, the server's conversational AI engine searches for the keyword "sports jacket" and displays the results as a list in the chat window.

[0166] Implementing custom features using generative AI

[0167] server

[0168] The server uses generative AI to generate clothing designs based on textual information entered by the user, such as a specific request like "I want a red dress with a belt."

[0169] Terminal

[0170] The terminal transmits the user's custom-made request to the server and displays the generated design.

[0171] User

[0172] Users can enter specific information into a custom request form, view the generated design, and purchase it.

[0173] Specific examples

[0174] If a user enters "I want a blue chiffon blouse" into a form, the server will generate a design using a generative AI model based on that request and display the result to the user.

[0175] Implementation of AI compatibility assessment function

[0176] server

[0177] The server uses an AI model to analyze the photos of the user uploaded by the user and the images of the product they are planning to purchase, and uses image analysis technology to quantify the fit between the user and the product.

[0178] Terminal

[0179] The device sends the user's image to the server and displays the results of the compatibility assessment, which the user can use as a reference when making a purchase decision.

[0180] User

[0181] Users upload a full-body photo of themselves and an image of the product they plan to purchase and check the analysis results.

[0182] Specific examples

[0183] When a user uploads a full-body photo and an image of the dress they want to try on, the server analyzes the image and provides results such as "This dress fits you 88%."

[0184] Example prompts for generative AI models

[0185] "I want a red dress with a belt."

[0186] "I want a blue chiffon blouse."

[0187] "I'm looking for a casual summer dress."

[0188] I'm looking for a sports jacket.

[0189] This invention significantly improves the user experience by allowing the server, terminal, and user to appropriately link these means, allowing users to easily find the best products and easily create and purchase customized, made-to-order products.

[0190] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0191] Recommendation processing steps

[0192] Step 1:

[0193] server

[0194] Collect physical information and historical purchase records from users. This input data is first stored in a database.

[0195] Specific actions

[0196] The user enters their height, weight, waist size, purchase history, etc. into a form. This information is immediately sent to the server and stored in a database.

[0197] Step 2:

[0198] server

[0199] The collected data is used to train machine learning models that learn user preferences using algorithms such as Collaborative Filtering and Content-based Filtering.

[0200] Specific actions

[0201] The server reads the stored data and uses algorithms to train the model, which improves its ability to predict what products will suit users.

[0202] Step 3:

[0203] Terminal

[0204] When a user accesses an e-commerce site, a recommendation request including the user ID is sent from the terminal to the server.

[0205] Specific actions

[0206] When a user logs in to the site, a recommendation request is automatically sent to the server, and the user ID is attached so personalized products are returned.

[0207] Step 4:

[0208] server

[0209] Using a trained machine learning model, the system recommends the best products for the user, generating a recommendation list and sending it to the device.

[0210] Specific actions

[0211] The server retrieves relevant information from a database based on the user ID, uses a machine learning model to generate a recommendation list, and sends it back to the device.

[0212] Step 5:

[0213] Terminal

[0214] The terminal displays the recommended products returned from the server, and the user can select from the displayed products.

[0215] Specific actions

[0216] The recommended products are displayed in the user interface, allowing the user to review them and select their favorite products.

[0217] Conversational AI search function processing steps

[0218] Step 1:

[0219] User

[0220] Enter your natural language request into the chat window, including the specific product category and details you require.

[0221] Specific actions

[0222] The user types "I'm looking for a casual summer dress" into the chat window. This becomes the input data.

[0223] Step 2:

[0224] Terminal

[0225] The user's request is sent to the server, and the input content in the chat window is passed as is as data.

[0226] Specific actions

[0227] The terminal transmits the input natural language data to the server.

[0228] Step 3:

[0229] server

[0230] The conversational AI engine analyzes the user's request and searches the database for relevant product information.

[0231] Specific actions

[0232] The conversational AI engine extracts keywords such as "casual," "summer," and "dress," and searches for corresponding products from the database.

[0233] Step 4:

[0234] server

[0235] The search results are generated as a chat list and sent back to the device.

[0236] Specific actions

[0237] The searched products are returned in list form to the terminal.

[0238] Step 5:

[0239] Terminal

[0240] The terminal displays the returned search results to the user, who can then check the product details in the chat window.

[0241] Specific actions

[0242] The product list will be displayed in the chat window, allowing the user to check the contents.

[0243] Processing steps for custom-made features using generative AI

[0244] Step 1:

[0245] User

[0246] Fill out the custom request form with specific details and submit it. For example, enter a detailed request such as "I want a red dress with a belt."

[0247] Specific actions

[0248] A user enters "I want a blue chiffon blouse" into a custom-made request form and submits it.

[0249] Step 2:

[0250] Terminal

[0251] The device sends a request from the user to the server, and this request data becomes the input for the generation AI.

[0252] Specific actions

[0253] The terminal transmits the requested information to the server as is.

[0254] Step 3:

[0255] server

[0256] The generative AI model generates designs based on requests, automatically generating clothing designs based on input parameters.

[0257] Specific actions

[0258] The server passes the request information to the generative AI model, which then generates a specific design that reflects the requirements, such as "blue color" and "chiffon material."

[0259] Step 4:

[0260] server

[0261] The generated design is formatted for a user interface and sent back to the terminal.

[0262] Specific actions

[0263] The generated design data is converted into a format that can be confirmed by the user and sent back to the terminal.

[0264] Step 5:

[0265] Terminal

[0266] The terminal displays the generated design to the user, who can then review it and proceed with the purchase.

[0267] Specific actions

[0268] The generated design is displayed on the terminal, and the user can confirm the design before proceeding with the purchase.

[0269] Processing steps of the AI ​​compatibility assessment function

[0270] Step 1:

[0271] User

[0272] Upload a full-body photo of yourself and an image of the product you plan to purchase.

[0273] Specific actions

[0274] Users upload their full-body photos and product images to the site, which serve as input data.

[0275] Step 2:

[0276] Terminal

[0277] The device sends the uploaded image to the server, where the image data is analyzed.

[0278] Specific actions

[0279] The terminal sends the photo and product image to the server.

[0280] Step 3:

[0281] server

[0282] It uses image analysis technology to compare the user's photo with the product image to determine compatibility. This analysis uses image recognition technology with machine learning.

[0283] Specific actions

[0284] The server's AI analyzes the photo and product image and scores how well it fits.

[0285] Step 4:

[0286] server

[0287] A score is generated as a result of the relevance determination and sent back to the terminal.

[0288] Specific actions

[0289] The judgment results are generated as a score that is easy for the user to understand and are sent back to the terminal.

[0290] Step 5:

[0291] Terminal

[0292] The terminal displays the results of the compatibility assessment to the user, who can then use these results to make product selections.

[0293] Specific actions

[0294] For example, the result "This dress fits you 88%" may be displayed to the user to help them make a purchase.

[0295] (Application example 1)

[0296] Next, a description will be given of Application 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."

[0297] While existing fashion e-commerce platforms have recommendation systems based on users' body type information and past purchase history, they do not offer visual try-on in virtual stores or product recommendations through real-time voice interaction. Furthermore, the process of generating custom designs using generative AI and providing those designs to users is incomplete, failing to sufficiently improve user satisfaction. Furthermore, the process of users generating specific designs using prompts is complex.

[0298] 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.

[0299] In this invention, the server includes means for collecting a user's body type information and past purchase history, means for training a machine learning model based on the collected information, means for recommending optimal products to the user using the trained machine learning model, means for allowing the user to visualize and try on items in a virtual shop, and means for responding to the user's questions and requests via voice recognition. This allows the user to receive recommendations for optimal products in real time in the virtual shop, visually try on items, and have their questions and requests about the products answered through voice interaction. Furthermore, the process of generating and providing custom designs based on the user's specific requirements using generative AI is simplified, thereby improving user satisfaction.

[0300] "User's body type information" is information about the user's physical characteristics such as height, weight, and waist size.

[0301] "Past purchase history" is a record of products that a user has purchased in the past.

[0302] A "machine learning model" is a program that uses algorithms to find patterns and make predictions or classifications based on collected data.

[0303] "Recommendation methods" are methods that use machine learning models to present optimal products based on data such as the user's body type and purchase history.

[0304] A "virtual shop" is a virtual store that users can access via the Internet to select, try on, and purchase products.

[0305] "Visualization" is the process of showing the user what the product will look like in the virtual shop.

[0306] "Trying on" is the process by which a user simulates how clothes and accessories will look on them in a virtual environment.

[0307] "Speech recognition" is a technology that analyzes a user's voice and converts it into text.

[0308] An "interactive AI engine" is a program that analyzes requests and questions from users in natural language and generates appropriate responses.

[0309] "Generative AI" is an artificial intelligence algorithm that creates new designs and content based on information and requests provided by users.

[0310] A "prompt sentence" is a piece of text given to a generation AI as specific input, and serves as a guide to control the direction of the generated output.

[0311] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history, uses AI to recommend optimal fashion items, and also provides custom-made clothing based on the user's requests. This system is configured as follows.

[0312] Recommendation function implementation

[0313] server

[0314] The server first collects the user's body type information (height, weight, waist size, etc.) and past purchase history and stores them in a database. It then trains a machine learning model based on this data to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering. Specifically, each time a user accesses the e-commerce site, the trained model presents the user with personalized product recommendations.

[0315] Terminal

[0316] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server. The device displays the recommended products returned by the server, helping the user easily find the right product. Using smart glasses or a head-mounted display, the user can visualize and try on products in a virtual store.

[0317] User

[0318] Users enter their purchase history, body type, and other information. When they access the site, they can receive product suggestions that suit them. They can also visually try on and purchase products in a virtual store.

[0319] Implementing conversational AI search functionality

[0320] server

[0321] The server is equipped with a conversational AI engine that analyzes natural language requests from users. Based on the user's input, it searches for relevant product information from a database and returns the results. The conversational AI engine uses speech recognition technology to respond to the user's voice questions and convert them into text.

[0322] Terminal

[0323] The terminal has a chat window where users can freely input requests and questions. It also responds to user voice requests using a voice dialogue function. The terminal passes these requests to the server, and the server's analysis results are displayed to the user.

[0324] User

[0325] For example, users can input "I want a casual summer dress" and receive suggestions from the conversational AI. Voice input is possible, making it highly convenient.

[0326] Implementing custom features using generative AI

[0327] server

[0328] The server uses a generative AI to generate clothing designs based on text input from the user, such as a specific request like "I want a red dress with a belt." The generative AI then uses prompts to suggest the best design.

[0329] Terminal

[0330] The terminal sends the user's custom-made request to the server and displays the generated design, allowing the user to confirm the design and decide to purchase it.

[0331] User

[0332] Users can enter information into a custom-made request form, view the generated design, and purchase it. For example, if you enter "I want a blue chiffon blouse," the AI ​​will generate a design based on that text and present it to the user.

[0333] Implementation of AI compatibility assessment function

[0334] server

[0335] The server uses an AI model to analyze the photos uploaded by the user and images of the product they are planning to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[0336] Terminal

[0337] The device sends the user's image to the server and displays the results of the compatibility assessment, which the user can use to make a purchasing decision.

[0338] User

[0339] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server. Specifically, when a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the images and displays results such as "This dress fits you 88%."

[0340] By combining these features, users can easily find the products that best suit them, minimizing post-purchase dissatisfaction and waste. Furthermore, by allowing users to easily create and purchase customized, made-to-order clothing, it is possible to increase user satisfaction and contribute to the realization of sustainable fashion.

[0341] Specific hardware and software used

[0342] Hardware: Smart glasses, head-mounted displays

[0343] Software: Python, cloud computing services (image analysis services), natural language processing libraries (conversational AI, generative AI models), image processing libraries

[0344] Example prompt: "A blue chiffon blouse"

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

[0346] Step 1:

[0347] The server receives the user's body type information (height, weight, waist size, etc.) and past purchase history and stores them in a database. At this time, the input data is the body type information and purchase history provided by the user, which the server receives and stores in the database in an appropriate format. This creates a data set that reflects the user's individual characteristics.

[0348] Step 2:

[0349] The server trains a machine learning model based on the collected body type information and purchase history. The input data is the user's saved body type information and purchase history, and the output is a model that predicts the most suitable fashion items for each user. Specifically, the model is trained using algorithms such as Collaborative Filtering and Content-based Filtering.

[0350] Step 3:

[0351] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server. Request data including the user's ID and session information is sent as input, and the server returns a list of recommended products. This allows the user to receive suggestions for the most suitable fashion items.

[0352] Step 4:

[0353] The terminal displays recommended products to the user in a virtual shop, allowing them to visualize and try on the products. The input is a list of recommended products returned from the server, and the output is an environment in which the user can try on and visualize the products using AR or VR technology. Specifically, smart glasses or a head-mounted display are used to display 3D models of the products.

[0354] Step 5:

[0355] The server uses a conversational AI engine to analyze natural language requests from users. The input is the user's voice or text question or request, and the output is the analyzed result, which is related product information or a response. The conversational AI engine uses speech recognition technology to convert the user's voice into text and generate an appropriate response.

[0356] Step 6:

[0357] The device receives the analysis results from the server and displays them to the user on a screen or via voice. The input is the response or product information generated by the conversational AI engine, and the output is the information displayed to the user. This allows the user to quickly and accurately obtain the information they need.

[0358] Step 7:

[0359] The server uses a generative AI to generate clothing designs based on text information entered by the user. The input is the text information entered by the user into the custom-made request form, and the output is the clothing design created by the generative AI. For requests such as "I want a blue chiffon blouse," the server generates a design based on the prompt text.

[0360] Step 8:

[0361] The device receives the generated design from the server and displays it to the user. The input is the design created by the generative AI, and the output is the user's screen where the design can be viewed. This allows the user to view the custom-made design and proceed with further customization or purchase.

[0362] Step 9:

[0363] The server analyzes the user's uploaded photo and the product image of the product they are planning to purchase to determine compatibility. The input is the user's photo and the product image, and the output is a compatibility score. Using image analysis technology, an AI model determines the fit between the user and the product.

[0364] Step 10:

[0365] The device receives the relevance assessment results from the server and presents them to the user. The input is the server's relevance score, and the output is feedback on the user's screen. This provides the user with information to help them make a purchasing decision and make an appropriate choice.

[0366] 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.

[0367] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history to recommend optimal fashion items using AI, and even provides custom-made clothing based on the user's requests. It also combines an emotion engine that recognizes the user's emotions to provide a more personalized service. This system is configured as follows:

[0368] Recommendation function implementation

[0369] server

[0370] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering.

[0371] Terminal

[0372] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[0373] User

[0374] Users enter their purchase history, body type information, etc., and when they access the site, they can receive product suggestions that suit them.

[0375] Specific examples

[0376] When a user purchases a T-shirt, the app will display the best size T-shirt and related recommended items based on their past purchase history and body type information.

[0377] Implementing conversational AI search functionality

[0378] server

[0379] The server is equipped with a conversational AI engine that analyzes natural language requests from users, searches for relevant product information from a database based on the user's input, and returns the results. It also uses an emotion engine to recognize the user's emotional state and tailor its response accordingly.

[0380] Terminal

[0381] The terminal has a chat window where users can freely input requests and questions. The terminal passes the requests to the server, which then displays the results of the server's analysis to the user.

[0382] User

[0383] For example, users can input "I want a casual summer dress" and receive suggestions from the conversational AI. They can also receive more appropriate products and advice depending on the user's emotional state while inputting.

[0384] Specific examples

[0385] When a user types "I'm looking for a sports jacket" in the chat window, the AI ​​engine searches for the appropriate product and displays a list of sports jacket recommendations. If the user is excited by the emotion engine, it will suggest more active designs.

[0386] Implementing custom features using generative AI

[0387] server

[0388] The server uses generative AI to generate clothing designs based on text input from the user, such as a specific request like "I want a red dress with a belt," and an emotional engine to recognize the user's emotional state and adjust the tone and style of the design.

[0389] Terminal

[0390] The terminal transmits the user's custom-made request to the server and displays the generated design.

[0391] User

[0392] Users can enter information into a custom request form, review the generated design, and purchase it. They can also receive design suggestions based on their emotional state.

[0393] Specific examples

[0394] A user enters "I want a blue chiffon blouse" into the custom-made request form, and the generative AI generates a design based on that text and presents it to the user. If the user is relaxed, the emotion engine suggests a softer design.

[0395] Implementation of AI compatibility assessment function

[0396] server

[0397] The server uses an AI model to analyze the photos uploaded by the user and images of the products they plan to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[0398] Terminal

[0399] The device sends the user's image to the server and displays the results of the compatibility assessment, which can be used as a reference when making a purchasing decision.

[0400] User

[0401] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server.

[0402] Specific examples

[0403] When a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the image and displays a result such as "This dress fits you 88%."

[0404] Specific implementation of adjustments using the emotion engine

[0405] server

[0406] The server is equipped with an emotion engine that recognizes the user's emotional state from input information and images. Based on the results, the engine adjusts the accuracy of recommendations and the response of the conversational AI. For example, if the user is excited about a purchase, the emotion engine will adjust to provide additional suggestions.

[0407] Terminal

[0408] The device displays information adjusted by the emotion engine to the user, improving the user experience.

[0409] User

[0410] Users can receive suggestions optimized by the emotion engine and enjoy personalized services according to their individual emotional state.

[0411] Specific examples

[0412] If the user is recognized as feeling comfortable, relaxation items and designs with a relaxing effect will be suggested.

[0413] The combination of these features allows users to easily find the perfect product for them, easily order custom-made clothing, and receive personalized service based on their emotions, minimizing post-purchase dissatisfaction and waste.

[0414] The processing flow will be explained below.

[0415] Specific processing of the recommendation function

[0416] Step 1:

[0417] Server: When a user registers on the site, they are asked to enter their physical information (height, weight, waist size, etc.) and past purchase history, which is then saved in a database.

[0418] Step 2:

[0419] Server: When a user purchases a product, the purchase history (product name, category, size, color, etc.) is saved in a database.

[0420] Step 3:

[0421] Server: Periodically trains machine learning models using stored user data, using algorithms such as Collaborative Filtering and Content-based Filtering.

[0422] Step 4:

[0423] Terminal: The user logs in to the e-commerce site and sends a request for recommended products to the server.

[0424] Step 5:

[0425] Server: Based on the received user ID, the trained model is used to generate the optimal recommended product list.

[0426] Step 6:

[0427] Server: Sends the generated recommended product list to the terminal.

[0428] Step 7:

[0429] Terminal: Displays the sent recommended product list to the user.

[0430] Specific processing of conversational AI search function

[0431] Step 1:

[0432] User: Launches the site's conversational AI search widget and enters a question or request in natural language (e.g., "I'm looking for a casual summer dress").

[0433] Step 2:

[0434] Terminal: Sends user input to the server.

[0435] Step 3:

[0436] Server: Uses a conversational AI engine to analyze the user's natural language input and understand their requests and questions.

[0437] Step 4:

[0438] Server: Based on the analysis results, it uses an emotion engine to recognize the user's emotional state.

[0439] Step 5:

[0440] Server: Based on the recognized emotions and analysis results, it searches for related product information from the database.

[0441] Step 6:

[0442] Server: Sends search results to the device.

[0443] Step 7:

[0444] Terminal: Displays search results to the user.

[0445] Specific processing of custom features using generative AI

[0446] Step 1:

[0447] User: Enters details into a custom request form (e.g., "I would like a red dress with a belt").

[0448] Step 2:

[0449] Terminal: Sends request information to the server.

[0450] Step 3:

[0451] Server: Uses generative AI to generate designs based on user requests.

[0452] Step 4:

[0453] Server: The generated design image is adjusted based on the user's emotional state using an emotion engine.

[0454] Step 5:

[0455] Server: Sends the adjusted design image to the device.

[0456] Step 6:

[0457] Terminal: The adjusted design image is displayed to the user and confirmation is requested.

[0458] Step 7:

[0459] User: Checks the generated design and confirms the order if satisfied.

[0460] Specific processing of the AI ​​compatibility judgment function

[0461] Step 1:

[0462] Users upload a photo of themselves and an image of the clothing they plan to purchase to the site.

[0463] Step 2:

[0464] Terminal: Sends the uploaded image data to the server.

[0465] Step 3:

[0466] Server: Analyzes the image using a fit determination AI model. Specifically, it compares the user's photo with the image of the clothing and scores the degree of fit.

[0467] Step 4:

[0468] Server: Generates analyzed fitness scores and feedback.

[0469] Step 5:

[0470] Server: Sends the generated fitness score and feedback to the device.

[0471] Step 6:

[0472] Device: Presents fitness scores and feedback to the user.

[0473] Specific implementation of adjustments using the emotion engine

[0474] Step 1:

[0475] User: Provides input information and images when accessing the site, searching for products, interacting with the site, etc.

[0476] Step 2:

[0477] Terminal: Sends user input and images to the server.

[0478] Step 3:

[0479] Server: Using the emotion engine, recognizes emotions from user input information and images.

[0480] Step 4:

[0481] Server: Based on the recognized emotions, it adjusts recommendation results, conversational AI responses, and generative AI designs.

[0482] Step 5:

[0483] Server: Sends the adjustment results to the terminal.

[0484] Step 6:

[0485] Terminal: Display tailored information and suggestions to the user.

[0486] These concrete steps allow users to easily find the perfect product, easily order custom-made clothing, and receive personalized service based on their emotions, minimizing post-purchase dissatisfaction and waste.

[0487] Example 2

[0488] Next, a description will be given of 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."

[0489] Conventional fashion e-commerce systems do not fully utilize a user's body type information or past purchase history, making it difficult to recommend optimal products to the user. Furthermore, personalized recommendations that take the user's emotional state into consideration are not made, resulting in a poor user experience. Furthermore, it is difficult to respond quickly and accurately to requests for custom-made items, which is a factor that reduces user satisfaction. The present invention aims to solve these problems.

[0490] 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.

[0491] In this invention, the server includes: means for collecting a user's body type information and past purchase history; means for training a machine learning model based on the collected information; means for recommending optimal products to the user using the trained machine learning model; means for analyzing the user's natural language input; means for searching for related product information based on the analysis results by the dialogue engine; means for recognizing the user's emotional state and adjusting suggestions based on the results; and means for generating custom designs based on the user's input information using generative AI technology. This makes it possible to comprehensively utilize the user's body type information, purchase history, and emotional state to propose optimal products and provide customized services to individual users.

[0492] 1. "Means of collection" refers to the equipment and software used to receive a user's body information and past purchase history and store it in a database.

[0493] 2. "Training means" refers to the process of using collected data to train a machine learning model to predict the best products for a user.

[0494] 3. "Recommendation method" refers to a system or algorithm that uses a trained machine learning model to suggest optimal products to users.

[0495] 4. "Dialogue engine" refers to software or a system that analyzes a user's natural language input and extracts relevant information based on that input.

[0496] 5. "Searching means" refers to the process of searching for appropriate product information from the database based on the results analyzed by the dialogue engine.

[0497] 6. "Emotion engine" refers to an algorithm that recognizes a user's emotional state from input information and images, and adjusts the system's responses and suggestions based on the results.

[0498] 7. "Generative AI technology" refers to artificial intelligence technology that automatically generates custom designs based on text information entered by the user.

[0499] 8. "Custom Design" refers to a unique clothing or product design created according to a user's specific requirements.

[0500] 9. "Means of analysis" refers to the process of analyzing uploaded images and input information to determine the suitability and suitability of products that meet the user's requirements.

[0501] 10. "Means of scoring" refers to the process of quantifying the suitability of a product based on the analysis results and evaluating its fit for the user.

[0502] This invention is a fashion e-commerce system that utilizes a user's body type information and past purchase history to recommend the most suitable fashion items using AI, and even provides custom-made clothing based on the user's requests. It also combines an emotion engine that recognizes the user's emotions to provide a more personalized service. This system consists of three main components: a server, a terminal, and a user.

[0503] server

[0504] The server first collects the user's body type information (height, weight, waist size, etc.) and past purchase history, and stores this data in a database. This collection is done using a web framework such as Django. A trained machine learning model is then applied to recommend the best products for the user based on the collected data. Machine learning tools such as SciKit-Learn and TENSORFLOW (registered trademark) are used for this training and prediction.

[0505] The server then houses a dialogue engine that analyzes the user's natural language input. This dialogue engine utilizes advanced natural language processing models such as GPT-3 (registered trademark) and searches a database for relevant product information based on the analysis results. The server also houses an emotion engine that uses NLP libraries and other tools to recognize the user's emotional state from their input and images.

[0506] It also uses generative AI technology, such as DALL-E, to generate custom designs based on the user's specific requests.

[0507] Terminal

[0508] The device provides a front-end interface for users to access. Through a web browser or mobile app, users can input their body type information and purchase history and use the conversational AI search function to find products. It also has a chat window that displays the analysis results of the dialogue engine and tailored suggestions from the emotion engine. Front-end frameworks such as React.js and Vue.js are used for this display.

[0509] The device then sends the user's uploaded photos and images of the products they plan to purchase to the server, and displays the results of the matching. OpenCV and TensorFlow are used for this analysis and matching.

[0510] User

[0511] Users first input their body type and past purchase history, which allows them to receive personalized product suggestions. Next, when searching for products using the conversational AI search function, they can input their request in natural language. Users can input a prompt statement such as "I'm looking for a casual summer dress." They can also enter a specific request, such as "I'm looking for a blue chiffon blouse," into the custom-made request form, where they can view and purchase the generated custom design.

[0512] For example, if a user wants to buy a T-shirt, the system will display the best size T-shirt and other related recommended items based on their past purchase history and body type information. This allows users to easily find the products that best suit them. They can also easily order custom-made clothing and receive personalized service based on their emotions. This minimizes post-purchase dissatisfaction and waste.

[0513] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0514] Step 1: Enter and collect user information

[0515] The user inputs information about their body type (height, weight, waist size, etc.) and also provides past purchase history data. This input data is sent from the terminal to the server, which then stores this data in a database. Specifically, the user inputs information into a web form, and the form data is sent to the server.

[0516] Input: Body type information, past purchase history data

[0517] Output: Body shape information and purchase history data stored in a database

[0518] Step 2: Training the machine learning model

[0519] The server trains a machine learning model based on the collected user data. During this process, machine learning tools such as SciKit-Learn and TensorFlow are used to process the data and train the model. After training is complete, the model is saved and used for future predictions.

[0520] Input: User data in the database

[0521] Output: A trained machine learning model

[0522] Step 3: Predicting recommended products

[0523] The server uses a trained machine learning model to predict the best product for the user. The prediction result is encoded in JSON format and sent back to the device. Specifically, the server inputs the current user data into the trained model and generates a prediction result.

[0524] Input: current user data, trained model

[0525] Output: Predicted recommended product list (JSON format)

[0526] Step 4: Display recommended products

[0527] The device receives the recommended product list returned from the server and displays it on the user interface. Specifically, the device uses a front-end framework such as React.js or Vue.js to parse the JSON data and display the product list.

[0528] Input: Recommended product list (JSON format)

[0529] Output: A list of products displayed on the user interface

[0530] Step 5: Enter your request in natural language

[0531] Using the conversational AI search feature, users input their requests in natural language, such as "I'm looking for a casual summer dress" into a chat window on a website or mobile app.

[0532] Input: Natural language request text

[0533] Output: Sends the request text to the server

[0534] Step 6: Natural Language Analysis and Emotion Recognition

[0535] The server's dialogue engine receives the user's input text and performs natural language analysis and emotion recognition. This process uses natural language processing models and emotion recognition algorithms such as GPT-3. The analysis results are used in the next step.

[0536] Input: Natural language request text

[0537] Output: Analysis results and emotional state

[0538] Step 7: Find and refine product information

[0539] The server searches the database based on the analysis results to retrieve relevant product information. At the same time, it adjusts the content of the suggestions using an emotion engine. Specifically, it executes an SQL query to retrieve product information from the database and adjusts the response content according to the emotion analysis results.

[0540] Input: Analysis results and emotional state

[0541] Output: Adjusted product information list

[0542] Step 8: Viewing conversational AI responses

[0543] The terminal receives the adjusted product list from the server and displays it as a response in the chat window. Specifically, the terminal receives the response data and dynamically generates content to be displayed in the chat window.

[0544] Input: Adjusted product information list

[0545] Output: Product suggestions displayed in the chat window

[0546] Step 9: Fill out your customization request

[0547] The user enters specific information into the custom order request form, for example, text such as "I would like a blue chiffon blouse."

[0548] Input: Custom request text

[0549] Output: Request text sent to the server

[0550] Step 10: Generate a design using generative AI

[0551] The server uses a generative AI model, such as DALL-E, to generate a custom design based on the user's input. The generated design image is then sent to the device.

[0552] Input: Custom request text

[0553] Output: Generated design image

[0554] Step 11: Display your custom design

[0555] The device displays the generated design image received from the server. Specifically, the design is displayed using the HTML Canvas API.

[0556] Input: Generated design image

[0557] Output: Your custom design displayed on the user interface

[0558] Step 12: Conducting a conformance determination

[0559] Users upload photos of themselves and the products they plan to purchase from their devices to a server, which then uses an AI model to analyze the images and calculate a suitability score.

[0560] Input: User image, product image

[0561] Output: Relevance score

[0562] Step 13: Viewing the relevance results

[0563] The terminal displays the relevance score received from the server. Specifically, the terminal visualizes the relevance score and provides it to the user.

[0564] Input: Suitability score

[0565] Output: Relevance assessment results displayed on the user interface

[0566] Step 14: Optimizing recommendations with an emotion engine

[0567] The server uses an emotion engine to recognize the user's emotional state from input information and images, and adjusts the suggestions based on the results.

[0568] Input: User input information, images

[0569] Output: Adjusted proposal

[0570] Step 15: Viewing optimized suggestions

[0571] The terminal displays the dynamically generated suggestion content to the user, optimized by the emotion engine.

[0572] Input: Adjusted proposal

[0573] Output: Optimization suggestions displayed on the user interface

[0574] Through these steps, users can easily find the products that best suit them, easily order custom-made clothing, and receive personalized service based on their emotions.

[0575] (Application example 2)

[0576] Next, a description will be given of Application 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."

[0577] Conventional fashion e-commerce systems recommend products based only on a user's body type and past purchase history, making it difficult to provide personalized suggestions that take into account a user's individual emotional state and temporary preferences. This often results in users being unable to find products that satisfy them, and also leads to increased dissatisfaction and returns after purchase. Furthermore, the systems lacked interactive features that allow users to input their requirements in natural language and features that generate custom-made designs, resulting in a lack of methods to improve the user experience.

[0578] 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.

[0579] In this invention, the server includes: means for collecting a user's body type information and past purchase history; means for training a machine learning model based on the collected information; means for recommending optimal products to the user using the trained machine learning model; and means for adjusting the recommendations according to the user's emotional state using an emotion recognition engine that identifies the user's emotional state. This enables highly personalized product proposals that take the user's individual emotional state into consideration. The server also includes means for analyzing a user's request using an interactive artificial intelligence engine; means for searching for related product information based on the analysis results; means for displaying the search results to the user; and means including an emotion recognition engine that identifies the user's emotional state and adjusts the analysis results. This enables intuitive operation using natural language and improves user satisfaction. The server also includes means for generating designs using artificial intelligence based on the user's text input information; means for providing the generated designs to the user; and means including an emotion recognition engine that adjusts the tone and style of the designs based on the user's emotional state. This makes it easy to provide custom designs tailored to the user's preferences.

[0580] "User's body type information" refers to physical measurement data such as the user's height, weight, waist size, etc.

[0581] "Past purchase history" refers to information such as the products a user has purchased in the past, the purchase dates, and the quantities.

[0582] A "machine learning model" refers to an artificial intelligence algorithm that learns patterns based on data and makes predictions and classifications.

[0583] An "emotion recognition engine" refers to an artificial intelligence technology that recognizes emotions from a user's facial expressions and text input, and adjusts responses based on the results.

[0584] "Recommendation" refers to the act of suggesting products or services that are suitable for a user.

[0585] An "interactive artificial intelligence engine" refers to an artificial intelligence technology that analyzes requests made in natural language by users and generates appropriate responses.

[0586] "Generative AI" refers to AI technology that generates new designs and content based on user input data.

[0587] "Tone and style of design" refers to elements of design such as color, shape, and atmosphere.

[0588] The present invention relates to a fashion e-commerce system that collects a user's body type information and past purchase history, and recommends personalized products based on the user's emotional state using an emotion recognition engine. This system includes the following configuration and processing steps.

[0589] System configuration

[0590] server

[0591] The server includes the following means:

[0592] 1. Collection of body type information and purchase history: Collect body type information such as height, weight, waist size, and past purchase history from users and store them in a database.

[0593] 2. Training machine learning models: Along with the collected information, algorithms such as Collaborative Filtering and Content-based Filtering are used to train machine learning models.

[0594] 3. Use of Emotion Recognition Engine: The app has a built-in emotion recognition engine to identify the user's emotional state (relaxed, excited, etc.). This engine recognizes emotions from images and input text uploaded by the user and analyzes the data.

[0595] 4. Recommendation adjustment: Adjust the recommendation results obtained from the machine learning model based on the results of the emotion recognition engine.

[0596] Terminal

[0597] The terminal includes the following means:

[0598] 1. Data transmission and reception: When a user accesses an e-commerce site, their body type information, purchase history, and images and text for emotion recognition are sent to the server. The server also displays the recommendations and generated designs.

[0599] 2. Interactive features: Provides chat windows and forms that allow users to enter requests and questions in natural language, allowing users to enter specific requests such as "I want a casual summer dress."

[0600] User

[0601] The user performs the following activities:

[0602] 1. Information input: Enter your body type and past purchase history, as well as requests and questions about fashion items in natural language.

[0603] 2. Image Upload: Upload your own photo and use it to analyze your emotional state with the emotion recognition engine.

[0604] Details of the hardware and software you will use

[0605] Smartphone: An interface that allows users to enter information, upload images, and view recommendation results.

[0606] PIL (Python Imaging Library): Used to load user images.

[0607] EmotionRecognizer: An emotion recognition library used to recognize the emotional state of a user from their image.

[0608] RecommenderSystem: A system that recommends optimal products based on the user's body type information and past purchase history.

[0609] Specific examples

[0610] 1. The user enters their body information (height 170cm, weight 65kg, waist size 75cm) into a smartphone app, along with the T-shirts, jeans, sweatshirts, etc. they have purchased in the past.

[0611] 2. The user uploads a photo and EmotionRecognizer recognizes the "relaxed state" from the photo.

[0612] 3. The Recommender System recommends the most suitable relaxing wear based on the user's data.

[0613] 4. The recommendation results are displayed to the user, suggesting "comfortable relaxing wear."

[0614] Example prompts for generative AI models

[0615] "If the user is relaxing, suggest the best relaxing wear."

[0616] In this way, the present invention realizes providing a highly personalized fashion e-commerce service that meets the individual needs of users and takes into account their emotional state.

[0617] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0618] Step 1:

[0619] User input of information

[0620] A user enters their body information (e.g., height, weight, waist size) and past purchase history into a smartphone app. This input information is sent from the device to a server. Specifically, the user provides information using an input form, and the input data is stored on the server through the device interface.

[0621] input:

[0622] User's body type information and past purchase history

[0623] output:

[0624] User information data stored on the server

[0625] Step 2:

[0626] Image upload and emotion recognition

[0627] A user uploads a photo of themselves. This photo data is sent from the device to a server, where an emotion recognition engine (e.g., EmotionRecognizer) is used to analyze the user's emotional state. Specifically, after the image data arrives at the server, the emotion recognition engine receives it as input and outputs the user's emotional state (e.g., relaxed, excited, etc.).

[0628] input:

[0629] User photo data

[0630] output:

[0631] User emotional state data

[0632] Step 3:

[0633] Training a machine learning model

[0634] The server trains a machine learning model using algorithms such as Collaborative Filtering and Content-based Filtering based on the collected user data and past purchase history. This builds a model for predicting user preferences and trends. Specifically, the machine learning model receives user data as input and learns patterns to improve the accuracy of product recommendations.

[0635] input:

[0636] Collected user data and purchase history

[0637] output:

[0638] Trained machine learning models

[0639] Step 4:

[0640] Recommendation generation and adjustment

[0641] The server uses the trained machine learning model to recommend appropriate products to the user. At this time, it also takes into account the user's emotional state obtained from the emotion recognition engine to adjust the recommendations. Specifically, it receives the user's emotional state as input and generates an optimal product list based on the output of the machine learning model.

[0642] input:

[0643] Trained machine learning model and user emotional state data

[0644] output:

[0645] Recommended product list

[0646] Step 5:

[0647] Displaying recommendations

[0648] The terminal displays the recommended product list sent from the server to the user. The user can check the suggested products through the interface of the smartphone app. Specifically, the product list is sent to the terminal and displayed on the user interface of the terminal.

[0649] input:

[0650] Recommended product list

[0651] output:

[0652] Recommended products displayed on the device

[0653] Step 6:

[0654] Interactive information provision

[0655] The server uses an interactive AI engine based on the user's request to analyze the request in natural language and search for related product information. For example, if a user inputs "I want a casual summer dress," the AI ​​engine will analyze the request and search for related product information.

[0656] input:

[0657] Natural language requests from users

[0658] output:

[0659] Parsed request and associated product information

[0660] Step 7:

[0661] Custom-made designs with generative AI

[0662] The server generates new clothing designs using an artificial intelligence model generated based on the user's text input. During this process, the user's emotional state is also taken into account using an emotion recognition engine. The specific prompt used is, "If the user is relaxed, please suggest the most suitable relaxing wear."

[0663] input:

[0664] User text input information and emotional state

[0665] output:

[0666] Generated custom designs

[0667] In this way, a personalized fashion e-commerce service is provided that comprehensively takes into account the individual needs and emotional state of the user.

[0668] 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.

[0669] 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> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) 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.

[0670] 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.

[0671] [Second embodiment]

[0672] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0673] 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.

[0674] 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).

[0675] 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.

[0676] 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.

[0677] 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).

[0678] 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. 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.

[0679] 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.

[0680] 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.

[0681] 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.

[0682] 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.

[0683] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0684] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history, uses AI to recommend optimal fashion items, and also provides custom-made clothing based on the user's requests. This system is configured as follows.

[0685] Recommendation function implementation

[0686] server

[0687] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering.

[0688] Terminal

[0689] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[0690] User

[0691] Users enter their purchase history, body type information, etc., and when they access the site, they can receive product suggestions that suit them.

[0692] Specific examples

[0693] When a user purchases a T-shirt, the app will display the best size T-shirt and related recommended items based on their past purchase history and body type information.

[0694] Implementing conversational AI search functionality

[0695] server

[0696] The server is equipped with a conversational AI engine that analyzes natural language requests from users, searches for relevant product information from a database based on the user's input, and returns the results.

[0697] Terminal

[0698] The terminal has a chat window where users can freely input requests and questions. The terminal passes the requests to the server, which then displays the results of the server's analysis to the user.

[0699] User

[0700] For example, users can enter, "I'm looking for a casual summer dress," and receive suggestions from the conversational AI.

[0701] Specific examples

[0702] When a user types "I'm looking for a sports jacket" into the chat window, the AI ​​engine searches for the appropriate products and displays a list of recommended sports jackets.

[0703] Implementing custom features using generative AI

[0704] server

[0705] The server uses generative AI to generate clothing designs based on textual information entered by the user, such as a specific request like "I want a red dress with a belt."

[0706] Terminal

[0707] The terminal transmits the user's custom-made request to the server and displays the generated design.

[0708] User

[0709] Users can enter information into a custom request form, view the generated design, and purchase it.

[0710] Specific examples

[0711] A user enters "I want a blue chiffon blouse" into the custom-made request form, and the generative AI generates a design based on that text and presents it to the user.

[0712] Implementation of AI compatibility assessment function

[0713] server

[0714] The server uses an AI model to analyze the photos uploaded by the user and images of the products they plan to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[0715] Terminal

[0716] The device sends the user's image to the server and displays the results of the compatibility assessment, which can be used as a reference when making a purchasing decision.

[0717] User

[0718] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server.

[0719] Specific examples

[0720] When a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the image and displays a result such as "This dress fits you 88%."

[0721] By combining these features, users can easily find the products that best suit them, minimizing post-purchase dissatisfaction and waste. Furthermore, by allowing users to easily create and purchase customized, made-to-order clothing, the service not only increases user satisfaction but also contributes to the realization of sustainable fashion.

[0722] The processing flow will be explained below.

[0723] Specific processing of the recommendation function

[0724] Step 1:

[0725] Server: When a user registers on the site, they are asked to enter their body information (height, weight, waist size, etc.) and this information is saved in a database.

[0726] Step 2:

[0727] Server: When a user completes a product purchase, the purchase history (product name, category, size, color, etc.) is saved in a database.

[0728] Step 3:

[0729] Server: Once a certain amount of data has been accumulated, the user data is used to train a machine learning model, using algorithms such as Collaborative Filtering and Content-based Filtering.

[0730] Step 4:

[0731] Terminal: When a user logs in to the site, the currently logged-in user ID is sent to the server.

[0732] Step 5:

[0733] Server: Based on the submitted user ID, the trained model generates the optimal recommended products.

[0734] Step 6:

[0735] Server: Returns the generated recommended product list to the terminal.

[0736] Step 7:

[0737] Terminal: Displays the received recommended product list to the user.

[0738] Specific processing of conversational AI search function

[0739] Step 1:

[0740] User: Launches the site's conversational AI search widget and enters a question or request in natural language (e.g., "I'm looking for a casual summer dress").

[0741] Step 2:

[0742] Terminal: Sends user input to the server.

[0743] Step 3:

[0744] Server: A conversational AI engine analyzes the user's natural language input and understands their requests and questions.

[0745] Step 4:

[0746] Server: Based on the analysis results, search for related product information from the database.

[0747] Step 5:

[0748] Server: Returns search results to the device.

[0749] Step 6:

[0750] Terminal: Displays search results to the user.

[0751] Specific processing of custom features using generative AI

[0752] Step 1:

[0753] User: Enters details into a custom request form (e.g., "I would like a red dress with a belt").

[0754] Step 2:

[0755] Terminal: Sends request information to the server.

[0756] Step 3:

[0757] Server: Uses generative AI to generate designs based on user requests.

[0758] Step 4:

[0759] Server: Sends the generated design image back to the device.

[0760] Step 5:

[0761] Terminal: Displays the generated design to the user and asks for their confirmation.

[0762] Step 6:

[0763] User: Checks the generated design and confirms the order if satisfied.

[0764] Specific processing of the AI ​​compatibility judgment function

[0765] Step 1:

[0766] Users upload a photo of themselves and an image of the clothing they plan to purchase to the site.

[0767] Step 2:

[0768] Terminal: Sends the uploaded image data to the server.

[0769] Step 3:

[0770] Server: Analyzes the image using a compatibility assessment AI model. Specifically, it compares the user's photo with the image of the clothing and scores the degree of fit.

[0771] Step 4:

[0772] Server: Generates the analyzed fitness score and other feedback.

[0773] Step 5:

[0774] Server: Sends the generated fitness score and feedback back to the device.

[0775] Step 6:

[0776] Device: Presents fitness scores and feedback to the user.

[0777] These concrete steps allow users to easily find the perfect product and order custom-made clothing, minimizing post-purchase frustration and waste.

[0778] Example 1

[0779] Next, a description will be given of 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."

[0780] Conventional fashion e-commerce systems have limited functionality to provide users with optimal products, lacking in recommendation accuracy and customization capabilities. Furthermore, they lack comprehensive functionality to improve the user experience, such as conversational AI search, custom-made features, and pre-purchase compatibility assessment. As a result, users have difficulty finding the right products, which reduces their motivation to purchase.

[0781] 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.

[0782] In this invention, the server includes: means for collecting a user's physical information and historical purchase records; means for training a machine learning model based on the collected information; means for recommending appropriate products to the user using the trained machine learning model; means including an interactive AI engine for analyzing interactive requests from the user; means for searching for related product information based on the generated analysis results and displaying it to the user; means for generating designs using artificial intelligence based on the user's text input information; means for providing the generated designs to the user; and means for performing image analysis and compatibility assessment, scoring the most suitable products for the user, and displaying them. This enables a comprehensive system that integrates a variety of functions, such as product recommendations optimized for the user, search using interactive AI, custom-made generation, and compatibility assessment.

[0783] "User's physical information" refers to data related to the user's physical shape, such as height, weight, and waist size.

[0784] "Historical purchase records" refers to data and history of products purchased by a user in the past.

[0785] A "machine learning model" is a set of algorithms that uses collected data to learn user preferences and behavioral patterns and recommend optimal products.

[0786] An "interactive artificial intelligence engine" is an artificial intelligence system that uses natural language processing technology to analyze user requests and generate appropriate responses.

[0787] "Generative AI" is an AI technology that generates specific outputs (e.g., designs) based on user input.

[0788] "Image analysis" is the process of analyzing images of users and products, extracting features, and comparing and judging them.

[0789] "Compatibility assessment" is the process of quantifying how well a user and product match based on the results of image analysis.

[0790] "Scoring" refers to expressing the results of a relevance assessment as a numerical value.

[0791] "Searching for product information" refers to extracting related product data from a database based on a user request.

[0792] "Generating a design" is the process of using generative AI to create a specific design or item based on a user request.

[0793] "Recommending appropriate products to users" refers to selecting and providing products that best match the user's preferences and needs through machine learning models.

[0794] "Searching for related product information and displaying it to the user" refers to extracting related product data based on the analysis results of the interactive AI engine and displaying it on the user's device.

[0795] MODE FOR CARRYING OUT THE INVENTION

[0796] This invention relates to a fashion e-commerce system that utilizes a user's physical information and historical purchase records, uses artificial intelligence to recommend optimal fashion items, and even provides custom-made clothing based on the user's requests.

[0797] Recommendation function implementation

[0798] server

[0799] The server first collects the user's submitted physical information (e.g., height, weight, waist size) and historical purchase records and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. The model uses algorithms such as collaborative filtering and content-based filtering.

[0800] Terminal

[0801] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[0802] User

[0803] Users enter their physical information and purchase history on the e-commerce site, and when they return to the site, they can receive recommendations for products that are best suited to them.

[0804] Specific examples

[0805] When a user purchases a T-shirt, the server recommends a size "L" white T-shirt or blue jeans based on their past purchase history and physical information. The device displays this information on the e-commerce site, allowing the user to consider the purchase.

[0806] Implementing conversational AI search functionality

[0807] server

[0808] The server is equipped with a conversational AI engine that analyzes requests made in natural language by the user, searches for relevant product information from a database based on the analyzed request, and returns the results.

[0809] Terminal

[0810] The terminal has a chat window where users can freely enter questions or requests. The terminal sends these to the server, which then displays the analysis results to the user.

[0811] User

[0812] Users can type "I'm looking for a casual summer dress" into the chat window and receive suggestions from the conversational AI.

[0813] Specific examples

[0814] When a user types "I'm looking for a sports jacket" into a chat window, the server's conversational AI engine searches for the keyword "sports jacket" and displays the results as a list in the chat window.

[0815] Implementing custom features using generative AI

[0816] server

[0817] The server uses generative AI to generate clothing designs based on textual information entered by the user, such as a specific request like "I want a red dress with a belt."

[0818] Terminal

[0819] The terminal transmits the user's custom-made request to the server and displays the generated design.

[0820] User

[0821] Users can enter specific information into a custom request form, view the generated design, and purchase it.

[0822] Specific examples

[0823] If a user enters "I want a blue chiffon blouse" into a form, the server will generate a design using a generative AI model based on that request and display the result to the user.

[0824] Implementation of AI compatibility assessment function

[0825] server

[0826] The server uses an AI model to analyze the photos of the user uploaded by the user and the images of the product they are planning to purchase, and uses image analysis technology to quantify the fit between the user and the product.

[0827] Terminal

[0828] The device sends the user's image to the server and displays the results of the compatibility assessment, which the user can use as a reference when making a purchase decision.

[0829] User

[0830] Users upload a full-body photo of themselves and an image of the product they plan to purchase and check the analysis results.

[0831] Specific examples

[0832] When a user uploads a full-body photo and an image of the dress they want to try on, the server analyzes the image and provides results such as "This dress fits you 88%."

[0833] Example prompts for generative AI models

[0834] "I want a red dress with a belt."

[0835] "I want a blue chiffon blouse."

[0836] "I'm looking for a casual summer dress."

[0837] I'm looking for a sports jacket.

[0838] This invention significantly improves the user experience by allowing the server, terminal, and user to appropriately link these means, allowing users to easily find the best products and easily create and purchase customized, made-to-order products.

[0839] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0840] Recommendation processing steps

[0841] Step 1:

[0842] server

[0843] Collect physical information and historical purchase records from users. This input data is first stored in a database.

[0844] Specific actions

[0845] The user enters their height, weight, waist size, purchase history, etc. into a form. This information is immediately sent to the server and stored in a database.

[0846] Step 2:

[0847] server

[0848] The collected data is used to train machine learning models that learn user preferences using algorithms such as Collaborative Filtering and Content-based Filtering.

[0849] Specific actions

[0850] The server reads the stored data and uses algorithms to train the model, which improves its ability to predict what products will suit users.

[0851] Step 3:

[0852] Terminal

[0853] When a user accesses an e-commerce site, a recommendation request including the user ID is sent from the terminal to the server.

[0854] Specific actions

[0855] When a user logs in to the site, a recommendation request is automatically sent to the server, and the user ID is attached so personalized products are returned.

[0856] Step 4:

[0857] server

[0858] Using a trained machine learning model, the system recommends the best products for the user, generating a recommendation list and sending it to the device.

[0859] Specific actions

[0860] The server retrieves relevant information from a database based on the user ID, uses a machine learning model to generate a recommendation list, and sends it back to the device.

[0861] Step 5:

[0862] Terminal

[0863] The terminal displays the recommended products returned from the server, and the user can select from the displayed products.

[0864] Specific actions

[0865] The recommended products are displayed in the user interface, allowing the user to review them and select their favorite products.

[0866] Conversational AI search function processing steps

[0867] Step 1:

[0868] User

[0869] Enter your natural language request into the chat window, including the specific product category and details you require.

[0870] Specific actions

[0871] The user types "I'm looking for a casual summer dress" into the chat window. This becomes the input data.

[0872] Step 2:

[0873] Terminal

[0874] The user's request is sent to the server, and the input content in the chat window is passed as is as data.

[0875] Specific actions

[0876] The terminal transmits the input natural language data to the server.

[0877] Step 3:

[0878] server

[0879] The conversational AI engine analyzes the user's request and searches the database for relevant product information.

[0880] Specific actions

[0881] The conversational AI engine extracts keywords such as "casual," "summer," and "dress," and searches for corresponding products from the database.

[0882] Step 4:

[0883] server

[0884] The search results are generated as a chat list and sent back to the device.

[0885] Specific actions

[0886] The searched products are returned in list form to the terminal.

[0887] Step 5:

[0888] Terminal

[0889] The terminal displays the returned search results to the user, who can then check the product details in the chat window.

[0890] Specific actions

[0891] The product list will be displayed in the chat window, allowing the user to check the contents.

[0892] Processing steps for custom-made features using generative AI

[0893] Step 1:

[0894] User

[0895] Fill out the custom request form with specific details and submit it. For example, enter a detailed request such as "I want a red dress with a belt."

[0896] Specific actions

[0897] A user enters "I want a blue chiffon blouse" into a custom-made request form and submits it.

[0898] Step 2:

[0899] Terminal

[0900] The device sends a request from the user to the server, and this request data becomes the input for the generation AI.

[0901] Specific actions

[0902] The terminal transmits the requested information to the server as is.

[0903] Step 3:

[0904] server

[0905] The generative AI model generates designs based on requests, automatically generating clothing designs based on input parameters.

[0906] Specific actions

[0907] The server passes the request information to the generative AI model, which then generates a specific design that reflects the requirements, such as "blue color" and "chiffon material."

[0908] Step 4:

[0909] server

[0910] The generated design is formatted for a user interface and sent back to the terminal.

[0911] Specific actions

[0912] The generated design data is converted into a format that can be confirmed by the user and sent back to the terminal.

[0913] Step 5:

[0914] Terminal

[0915] The terminal displays the generated design to the user, who can then review it and proceed with the purchase.

[0916] Specific actions

[0917] The generated design is displayed on the terminal, and the user can confirm the design before proceeding with the purchase.

[0918] Processing steps of the AI ​​compatibility assessment function

[0919] Step 1:

[0920] User

[0921] Upload a full-body photo of yourself and an image of the product you plan to purchase.

[0922] Specific actions

[0923] Users upload their full-body photos and product images to the site, which serve as input data.

[0924] Step 2:

[0925] Terminal

[0926] The device sends the uploaded image to the server, where the image data is analyzed.

[0927] Specific actions

[0928] The terminal sends the photo and product image to the server.

[0929] Step 3:

[0930] server

[0931] It uses image analysis technology to compare the user's photo with the product image to determine compatibility. This analysis uses image recognition technology with machine learning.

[0932] Specific actions

[0933] The server's AI analyzes the photo and product image and scores how well it fits.

[0934] Step 4:

[0935] server

[0936] A score is generated as a result of the relevance determination and sent back to the terminal.

[0937] Specific actions

[0938] The judgment results are generated as a score that is easy for the user to understand and are sent back to the terminal.

[0939] Step 5:

[0940] Terminal

[0941] The terminal displays the results of the compatibility assessment to the user, who can then use these results to make product selections.

[0942] Specific actions

[0943] For example, the result "This dress fits you 88%" may be displayed to the user to help them make a purchase.

[0944] (Application example 1)

[0945] Next, a description will be given of Application 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."

[0946] While existing fashion e-commerce platforms have recommendation systems based on users' body type information and past purchase history, they do not offer visual try-on in virtual stores or product recommendations through real-time voice interaction. Furthermore, the process of generating custom designs using generative AI and providing those designs to users is incomplete, failing to sufficiently improve user satisfaction. Furthermore, the process of users generating specific designs using prompts is complex.

[0947] 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.

[0948] In this invention, the server includes means for collecting a user's body type information and past purchase history, means for training a machine learning model based on the collected information, means for recommending optimal products to the user using the trained machine learning model, means for allowing the user to visualize and try on items in a virtual shop, and means for responding to the user's questions and requests via voice recognition. This allows the user to receive recommendations for optimal products in real time in the virtual shop, visually try on items, and have their questions and requests about the products answered through voice interaction. Furthermore, the process of generating and providing custom designs based on the user's specific requirements using generative AI is simplified, thereby improving user satisfaction.

[0949] "User's body type information" is information about the user's physical characteristics such as height, weight, and waist size.

[0950] "Past purchase history" is a record of products that a user has purchased in the past.

[0951] A "machine learning model" is a program that uses algorithms to find patterns and make predictions or classifications based on collected data.

[0952] "Recommendation methods" are methods that use machine learning models to present optimal products based on data such as the user's body type and purchase history.

[0953] A "virtual shop" is a virtual store that users can access via the Internet to select, try on, and purchase products.

[0954] "Visualization" is the process of showing the user what the product will look like in the virtual shop.

[0955] "Trying on" is the process by which a user simulates how clothes and accessories will look on them in a virtual environment.

[0956] "Speech recognition" is a technology that analyzes a user's voice and converts it into text.

[0957] An "interactive AI engine" is a program that analyzes requests and questions from users in natural language and generates appropriate responses.

[0958] "Generative AI" is an artificial intelligence algorithm that creates new designs and content based on information and requests provided by users.

[0959] A "prompt sentence" is a piece of text given to a generation AI as specific input, and serves as a guide to control the direction of the generated output.

[0960] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history, uses AI to recommend optimal fashion items, and also provides custom-made clothing based on the user's requests. This system is configured as follows.

[0961] Recommendation function implementation

[0962] server

[0963] The server first collects the user's body type information (height, weight, waist size, etc.) and past purchase history and stores them in a database. It then trains a machine learning model based on this data to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering. Specifically, each time a user accesses the e-commerce site, the trained model presents the user with personalized product recommendations.

[0964] Terminal

[0965] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server. The device displays the recommended products returned by the server, helping the user easily find the right product. Using smart glasses or a head-mounted display, the user can visualize and try on products in a virtual store.

[0966] User

[0967] Users enter their purchase history, body type, and other information. When they access the site, they can receive product suggestions that suit them. They can also visually try on and purchase products in a virtual store.

[0968] Implementing conversational AI search functionality

[0969] server

[0970] The server is equipped with a conversational AI engine that analyzes natural language requests from users. Based on the user's input, it searches for relevant product information from a database and returns the results. The conversational AI engine uses speech recognition technology to respond to the user's voice questions and convert them into text.

[0971] Terminal

[0972] The terminal has a chat window where users can freely input requests and questions. It also responds to user voice requests using a voice dialogue function. The terminal passes these requests to the server, and the server's analysis results are displayed to the user.

[0973] User

[0974] For example, users can input "I want a casual summer dress" and receive suggestions from the conversational AI. Voice input is possible, making it highly convenient.

[0975] Implementing custom features using generative AI

[0976] server

[0977] The server uses a generative AI to generate clothing designs based on text input from the user, such as a specific request like "I want a red dress with a belt." The generative AI then uses prompts to suggest the best design.

[0978] Terminal

[0979] The terminal sends the user's custom-made request to the server and displays the generated design, allowing the user to confirm the design and decide to purchase it.

[0980] User

[0981] Users can enter information into a custom-made request form, view the generated design, and purchase it. For example, if you enter "I want a blue chiffon blouse," the AI ​​will generate a design based on that text and present it to the user.

[0982] Implementation of AI compatibility assessment function

[0983] server

[0984] The server uses an AI model to analyze the photos uploaded by the user and images of the product they are planning to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[0985] Terminal

[0986] The device sends the user's image to the server and displays the results of the compatibility assessment, which the user can use to make a purchasing decision.

[0987] User

[0988] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server. Specifically, when a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the images and displays results such as "This dress fits you 88%."

[0989] By combining these features, users can easily find the products that best suit them, minimizing post-purchase dissatisfaction and waste. Furthermore, by allowing users to easily create and purchase customized, made-to-order clothing, it is possible to increase user satisfaction and contribute to the realization of sustainable fashion.

[0990] Specific hardware and software used

[0991] Hardware: Smart glasses, head-mounted displays

[0992] Software: Python, cloud computing services (image analysis services), natural language processing libraries (conversational AI, generative AI models), image processing libraries

[0993] Example prompt: "A blue chiffon blouse"

[0994] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0995] Step 1:

[0996] The server receives the user's body type information (height, weight, waist size, etc.) and past purchase history and stores them in a database. At this time, the input data is the body type information and purchase history provided by the user, which the server receives and stores in the database in an appropriate format. This creates a data set that reflects the user's individual characteristics.

[0997] Step 2:

[0998] The server trains a machine learning model based on the collected body type information and purchase history. The input data is the user's saved body type information and purchase history, and the output is a model that predicts the most suitable fashion items for each user. Specifically, the model is trained using algorithms such as Collaborative Filtering and Content-based Filtering.

[0999] Step 3:

[1000] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server. Request data including the user's ID and session information is sent as input, and the server returns a list of recommended products. This allows the user to receive suggestions for the most suitable fashion items.

[1001] Step 4:

[1002] The terminal displays recommended products to the user in a virtual shop, allowing them to visualize and try on the products. The input is a list of recommended products returned from the server, and the output is an environment in which the user can try on and visualize the products using AR or VR technology. Specifically, smart glasses or a head-mounted display are used to display 3D models of the products.

[1003] Step 5:

[1004] The server uses a conversational AI engine to analyze natural language requests from users. The input is the user's voice or text question or request, and the output is the analyzed result, which is related product information or a response. The conversational AI engine uses speech recognition technology to convert the user's voice into text and generate an appropriate response.

[1005] Step 6:

[1006] The device receives the analysis results from the server and displays them to the user on a screen or via voice. The input is the response or product information generated by the conversational AI engine, and the output is the information displayed to the user. This allows the user to quickly and accurately obtain the information they need.

[1007] Step 7:

[1008] The server uses a generative AI to generate clothing designs based on text information entered by the user. The input is the text information entered by the user into the custom-made request form, and the output is the clothing design created by the generative AI. For requests such as "I want a blue chiffon blouse," the server generates a design based on the prompt text.

[1009] Step 8:

[1010] The device receives the generated design from the server and displays it to the user. The input is the design created by the generative AI, and the output is the user's screen where the design can be viewed. This allows the user to view the custom-made design and proceed with further customization or purchase.

[1011] Step 9:

[1012] The server analyzes the user's uploaded photo and the product image of the product they are planning to purchase to determine compatibility. The input is the user's photo and the product image, and the output is a compatibility score. Using image analysis technology, an AI model determines the fit between the user and the product.

[1013] Step 10:

[1014] The device receives the relevance assessment results from the server and presents them to the user. The input is the server's relevance score, and the output is feedback on the user's screen. This provides the user with information to help them make a purchasing decision and make an appropriate choice.

[1015] 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.

[1016] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history to recommend optimal fashion items using AI, and even provides custom-made clothing based on the user's requests. It also combines an emotion engine that recognizes the user's emotions to provide a more personalized service. This system is configured as follows:

[1017] Recommendation function implementation

[1018] server

[1019] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering.

[1020] Terminal

[1021] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[1022] User

[1023] Users enter their purchase history, body type information, etc., and when they access the site, they can receive product suggestions that suit them.

[1024] Specific examples

[1025] When a user purchases a T-shirt, the app will display the best size T-shirt and related recommended items based on their past purchase history and body type information.

[1026] Implementing conversational AI search functionality

[1027] server

[1028] The server is equipped with a conversational AI engine that analyzes natural language requests from users, searches for relevant product information from a database based on the user's input, and returns the results. It also uses an emotion engine to recognize the user's emotional state and tailor its response accordingly.

[1029] Terminal

[1030] The terminal has a chat window where users can freely input requests and questions. The terminal passes the requests to the server, which then displays the results of the server's analysis to the user.

[1031] User

[1032] For example, users can input "I want a casual summer dress" and receive suggestions from the conversational AI. They can also receive more appropriate products and advice depending on the user's emotional state while inputting.

[1033] Specific examples

[1034] When a user types "I'm looking for a sports jacket" in the chat window, the AI ​​engine searches for the appropriate product and displays a list of sports jacket recommendations. If the user is excited by the emotion engine, it will suggest more active designs.

[1035] Implementing custom features using generative AI

[1036] server

[1037] The server uses generative AI to generate clothing designs based on text input from the user, such as a specific request like "I want a red dress with a belt," and an emotional engine to recognize the user's emotional state and adjust the tone and style of the design.

[1038] Terminal

[1039] The terminal transmits the user's custom-made request to the server and displays the generated design.

[1040] User

[1041] Users can enter information into a custom request form, review the generated design, and purchase it. They can also receive design suggestions based on their emotional state.

[1042] Specific examples

[1043] A user enters "I want a blue chiffon blouse" into the custom-made request form, and the generative AI generates a design based on that text and presents it to the user. If the user is relaxed, the emotion engine suggests a softer design.

[1044] Implementation of AI compatibility assessment function

[1045] server

[1046] The server uses an AI model to analyze the photos uploaded by the user and images of the products they plan to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[1047] Terminal

[1048] The device sends the user's image to the server and displays the results of the compatibility assessment, which can be used as a reference when making a purchasing decision.

[1049] User

[1050] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server.

[1051] Specific examples

[1052] When a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the image and displays a result such as "This dress fits you 88%."

[1053] Specific implementation of adjustments using the emotion engine

[1054] server

[1055] The server is equipped with an emotion engine that recognizes the user's emotional state from input information and images. Based on the results, the engine adjusts the accuracy of recommendations and the response of the conversational AI. For example, if the user is excited about a purchase, the emotion engine will adjust to provide additional suggestions.

[1056] Terminal

[1057] The device displays information adjusted by the emotion engine to the user, improving the user experience.

[1058] User

[1059] Users can receive suggestions optimized by the emotion engine and enjoy personalized services according to their individual emotional state.

[1060] Specific examples

[1061] If the user is recognized as feeling comfortable, relaxation items and designs with a relaxing effect will be suggested.

[1062] The combination of these features allows users to easily find the perfect product for them, easily order custom-made clothing, and receive personalized service based on their emotions, minimizing post-purchase dissatisfaction and waste.

[1063] The processing flow will be explained below.

[1064] Specific processing of the recommendation function

[1065] Step 1:

[1066] Server: When a user registers on the site, they are asked to enter their physical information (height, weight, waist size, etc.) and past purchase history, which is then saved in a database.

[1067] Step 2:

[1068] Server: When a user purchases a product, the purchase history (product name, category, size, color, etc.) is saved in a database.

[1069] Step 3:

[1070] Server: Periodically trains machine learning models using stored user data, using algorithms such as Collaborative Filtering and Content-based Filtering.

[1071] Step 4:

[1072] Terminal: The user logs in to the e-commerce site and sends a request for recommended products to the server.

[1073] Step 5:

[1074] Server: Based on the received user ID, the trained model is used to generate the optimal recommended product list.

[1075] Step 6:

[1076] Server: Sends the generated recommended product list to the terminal.

[1077] Step 7:

[1078] Terminal: Displays the sent recommended product list to the user.

[1079] Specific processing of conversational AI search function

[1080] Step 1:

[1081] User: Launches the site's conversational AI search widget and enters a question or request in natural language (e.g., "I'm looking for a casual summer dress").

[1082] Step 2:

[1083] Terminal: Sends user input to the server.

[1084] Step 3:

[1085] Server: Uses a conversational AI engine to analyze the user's natural language input and understand their requests and questions.

[1086] Step 4:

[1087] Server: Based on the analysis results, it uses an emotion engine to recognize the user's emotional state.

[1088] Step 5:

[1089] Server: Based on the recognized emotions and analysis results, it searches for related product information from the database.

[1090] Step 6:

[1091] Server: Sends search results to the device.

[1092] Step 7:

[1093] Terminal: Displays search results to the user.

[1094] Specific processing of custom features using generative AI

[1095] Step 1:

[1096] User: Enters details into a custom request form (e.g., "I would like a red dress with a belt").

[1097] Step 2:

[1098] Terminal: Sends request information to the server.

[1099] Step 3:

[1100] Server: Uses generative AI to generate designs based on user requests.

[1101] Step 4:

[1102] Server: The generated design image is adjusted based on the user's emotional state using an emotion engine.

[1103] Step 5:

[1104] Server: Sends the adjusted design image to the device.

[1105] Step 6:

[1106] Terminal: The adjusted design image is displayed to the user and confirmation is requested.

[1107] Step 7:

[1108] User: Checks the generated design and confirms the order if satisfied.

[1109] Specific processing of the AI ​​compatibility judgment function

[1110] Step 1:

[1111] Users upload a photo of themselves and an image of the clothing they plan to purchase to the site.

[1112] Step 2:

[1113] Terminal: Sends the uploaded image data to the server.

[1114] Step 3:

[1115] Server: Analyzes the image using a fit determination AI model. Specifically, it compares the user's photo with the image of the clothing and scores the degree of fit.

[1116] Step 4:

[1117] Server: Generates analyzed fitness scores and feedback.

[1118] Step 5:

[1119] Server: Sends the generated fitness score and feedback to the device.

[1120] Step 6:

[1121] Device: Presents fitness scores and feedback to the user.

[1122] Specific implementation of adjustments using the emotion engine

[1123] Step 1:

[1124] User: Provides input information and images when accessing the site, searching for products, interacting with the site, etc.

[1125] Step 2:

[1126] Terminal: Sends user input and images to the server.

[1127] Step 3:

[1128] Server: Using the emotion engine, recognizes emotions from user input information and images.

[1129] Step 4:

[1130] Server: Based on the recognized emotions, it adjusts recommendation results, conversational AI responses, and generative AI designs.

[1131] Step 5:

[1132] Server: Sends the adjustment results to the terminal.

[1133] Step 6:

[1134] Terminal: Display tailored information and suggestions to the user.

[1135] These concrete steps allow users to easily find the perfect product, easily order custom-made clothing, and receive personalized service based on their emotions, minimizing post-purchase dissatisfaction and waste.

[1136] Example 2

[1137] Next, a description will be given of 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."

[1138] Conventional fashion e-commerce systems do not fully utilize a user's body type information or past purchase history, making it difficult to recommend optimal products to the user. Furthermore, personalized recommendations that take the user's emotional state into consideration are not made, resulting in a poor user experience. Furthermore, it is difficult to respond quickly and accurately to requests for custom-made items, which is a factor that reduces user satisfaction. The present invention aims to solve these problems.

[1139] 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.

[1140] In this invention, the server includes: means for collecting a user's body type information and past purchase history; means for training a machine learning model based on the collected information; means for recommending optimal products to the user using the trained machine learning model; means for analyzing the user's natural language input; means for searching for related product information based on the analysis results by the dialogue engine; means for recognizing the user's emotional state and adjusting suggestions based on the results; and means for generating custom designs based on the user's input information using generative AI technology. This makes it possible to comprehensively utilize the user's body type information, purchase history, and emotional state to propose optimal products and provide customized services to individual users.

[1141] 1. "Means of collection" refers to the equipment and software used to receive a user's body information and past purchase history and store it in a database.

[1142] 2. "Training means" refers to the process of using collected data to train a machine learning model to predict the best products for a user.

[1143] 3. "Recommendation method" refers to a system or algorithm that uses a trained machine learning model to suggest optimal products to users.

[1144] 4. "Dialogue engine" refers to software or a system that analyzes a user's natural language input and extracts relevant information based on that input.

[1145] 5. "Searching means" refers to the process of searching for appropriate product information from the database based on the results analyzed by the dialogue engine.

[1146] 6. "Emotion engine" refers to an algorithm that recognizes a user's emotional state from input information and images, and adjusts the system's responses and suggestions based on the results.

[1147] 7. "Generative AI technology" refers to artificial intelligence technology that automatically generates custom designs based on text information entered by the user.

[1148] 8. "Custom Design" refers to a unique clothing or product design created according to a user's specific requirements.

[1149] 9. "Means of analysis" refers to the process of analyzing uploaded images and input information to determine the suitability and suitability of products that meet the user's requirements.

[1150] 10. "Means of scoring" refers to the process of quantifying the suitability of a product based on the analysis results and evaluating its fit for the user.

[1151] This invention is a fashion e-commerce system that utilizes a user's body type information and past purchase history to recommend the most suitable fashion items using AI, and even provides custom-made clothing based on the user's requests. It also combines an emotion engine that recognizes the user's emotions to provide a more personalized service. This system consists of three main components: a server, a terminal, and a user.

[1152] server

[1153] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history, and stores this data in a database. This collection is done using a web framework such as Django. It then applies a trained machine learning model to recommend the best products for the user based on the collected data. Machine learning tools such as SciKit-Learn and TensorFlow are used for this training and prediction.

[1154] The server then houses a dialogue engine that analyzes the user's natural language input. This dialogue engine uses advanced natural language processing models such as GPT-3 and searches a database for relevant product information based on the analysis results. The server also houses an emotion engine that uses NLP libraries to recognize the user's emotional state from their input and images.

[1155] It also uses generative AI technology, such as DALL-E, to generate custom designs based on the user's specific requests.

[1156] Terminal

[1157] The device provides a front-end interface for users to access. Through a web browser or mobile app, users can input their body type information and purchase history and use the conversational AI search function to find products. It also has a chat window that displays the analysis results of the dialogue engine and tailored suggestions from the emotion engine. Front-end frameworks such as React.js and Vue.js are used for this display.

[1158] The device then sends the user's uploaded photos and images of the products they plan to purchase to the server, and displays the results of the matching. OpenCV and TensorFlow are used for this analysis and matching.

[1159] User

[1160] Users first input their body type and past purchase history, which allows them to receive personalized product suggestions. Next, when searching for products using the conversational AI search function, they can input their request in natural language. Users can input a prompt statement such as "I'm looking for a casual summer dress." They can also enter a specific request, such as "I'm looking for a blue chiffon blouse," into the custom-made request form, where they can view and purchase the generated custom design.

[1161] For example, if a user wants to buy a T-shirt, the system will display the best size T-shirt and other related recommended items based on their past purchase history and body type information. This allows users to easily find the products that best suit them. They can also easily order custom-made clothing and receive personalized service based on their emotions. This minimizes post-purchase dissatisfaction and waste.

[1162] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1163] Step 1: Enter and collect user information

[1164] The user inputs information about their body type (height, weight, waist size, etc.) and also provides past purchase history data. This input data is sent from the terminal to the server, which then stores this data in a database. Specifically, the user inputs information into a web form, and the form data is sent to the server.

[1165] Input: Body type information, past purchase history data

[1166] Output: Body shape information and purchase history data stored in a database

[1167] Step 2: Training the machine learning model

[1168] The server trains a machine learning model based on the collected user data. During this process, machine learning tools such as SciKit-Learn and TensorFlow are used to process the data and train the model. After training is complete, the model is saved and used for future predictions.

[1169] Input: User data in the database

[1170] Output: A trained machine learning model

[1171] Step 3: Predicting recommended products

[1172] The server uses a trained machine learning model to predict the best product for the user. The prediction result is encoded in JSON format and sent back to the device. Specifically, the server inputs the current user data into the trained model and generates a prediction result.

[1173] Input: current user data, trained model

[1174] Output: Predicted recommended product list (JSON format)

[1175] Step 4: Display recommended products

[1176] The device receives the recommended product list returned from the server and displays it on the user interface. Specifically, the device uses a front-end framework such as React.js or Vue.js to parse the JSON data and display the product list.

[1177] Input: Recommended product list (JSON format)

[1178] Output: A list of products displayed on the user interface

[1179] Step 5: Enter your request in natural language

[1180] Using the conversational AI search feature, users input their requests in natural language, such as "I'm looking for a casual summer dress" into a chat window on a website or mobile app.

[1181] Input: Natural language request text

[1182] Output: Sends the request text to the server

[1183] Step 6: Natural Language Analysis and Emotion Recognition

[1184] The server's dialogue engine receives the user's input text and performs natural language analysis and emotion recognition. This process uses natural language processing models and emotion recognition algorithms such as GPT-3. The analysis results are used in the next step.

[1185] Input: Natural language request text

[1186] Output: Analysis results and emotional state

[1187] Step 7: Find and refine product information

[1188] The server searches the database based on the analysis results to retrieve relevant product information. At the same time, it adjusts the content of the suggestions using an emotion engine. Specifically, it executes an SQL query to retrieve product information from the database and adjusts the response content according to the emotion analysis results.

[1189] Input: Analysis results and emotional state

[1190] Output: Adjusted product information list

[1191] Step 8: Viewing conversational AI responses

[1192] The terminal receives the adjusted product list from the server and displays it as a response in the chat window. Specifically, the terminal receives the response data and dynamically generates content to be displayed in the chat window.

[1193] Input: Adjusted product information list

[1194] Output: Product suggestions displayed in the chat window

[1195] Step 9: Fill out your customization request

[1196] The user enters specific information into the custom order request form, for example, text such as "I would like a blue chiffon blouse."

[1197] Input: Custom request text

[1198] Output: Request text sent to the server

[1199] Step 10: Generate a design using generative AI

[1200] The server uses a generative AI model, such as DALL-E, to generate a custom design based on the user's input. The generated design image is then sent to the device.

[1201] Input: Custom request text

[1202] Output: Generated design image

[1203] Step 11: Display your custom design

[1204] The device displays the generated design image received from the server. Specifically, the design is displayed using the HTML Canvas API.

[1205] Input: Generated design image

[1206] Output: Your custom design displayed on the user interface

[1207] Step 12: Conducting a conformance determination

[1208] Users upload photos of themselves and the products they plan to purchase from their devices to a server, which then uses an AI model to analyze the images and calculate a suitability score.

[1209] Input: User image, product image

[1210] Output: Relevance score

[1211] Step 13: Viewing the relevance results

[1212] The terminal displays the relevance score received from the server. Specifically, the terminal visualizes the relevance score and provides it to the user.

[1213] Input: Suitability score

[1214] Output: Relevance assessment results displayed on the user interface

[1215] Step 14: Optimizing recommendations with an emotion engine

[1216] The server uses an emotion engine to recognize the user's emotional state from input information and images, and adjusts the suggestions based on the results.

[1217] Input: User input information, images

[1218] Output: Adjusted proposal

[1219] Step 15: Viewing optimized suggestions

[1220] The terminal displays the dynamically generated suggestion content to the user, optimized by the emotion engine.

[1221] Input: Adjusted proposal

[1222] Output: Optimization suggestions displayed on the user interface

[1223] Through these steps, users can easily find the products that best suit them, easily order custom-made clothing, and receive personalized service based on their emotions.

[1224] (Application example 2)

[1225] Next, a description will be given of Application 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."

[1226] Conventional fashion e-commerce systems recommend products based only on a user's body type and past purchase history, making it difficult to provide personalized suggestions that take into account a user's individual emotional state and temporary preferences. This often results in users being unable to find products that satisfy them, and also leads to increased dissatisfaction and returns after purchase. Furthermore, the systems lacked interactive features that allow users to input their requirements in natural language and features that generate custom-made designs, resulting in a lack of methods to improve the user experience.

[1227] 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.

[1228] In this invention, the server includes: means for collecting a user's body type information and past purchase history; means for training a machine learning model based on the collected information; means for recommending optimal products to the user using the trained machine learning model; and means for adjusting the recommendations according to the user's emotional state using an emotion recognition engine that identifies the user's emotional state. This enables highly personalized product proposals that take the user's individual emotional state into consideration. The server also includes means for analyzing a user's request using an interactive artificial intelligence engine; means for searching for related product information based on the analysis results; means for displaying the search results to the user; and means including an emotion recognition engine that identifies the user's emotional state and adjusts the analysis results. This enables intuitive operation using natural language and improves user satisfaction. The server also includes means for generating designs using artificial intelligence based on the user's text input information; means for providing the generated designs to the user; and means including an emotion recognition engine that adjusts the tone and style of the designs based on the user's emotional state. This makes it easy to provide custom designs tailored to the user's preferences.

[1229] "User's body type information" refers to physical measurement data such as the user's height, weight, waist size, etc.

[1230] "Past purchase history" refers to information such as the products a user has purchased in the past, the purchase dates, and the quantities.

[1231] A "machine learning model" refers to an artificial intelligence algorithm that learns patterns based on data and makes predictions and classifications.

[1232] An "emotion recognition engine" refers to an artificial intelligence technology that recognizes emotions from a user's facial expressions and text input, and adjusts responses based on the results.

[1233] "Recommendation" refers to the act of suggesting products or services that are suitable for a user.

[1234] An "interactive artificial intelligence engine" refers to an artificial intelligence technology that analyzes requests made in natural language by users and generates appropriate responses.

[1235] "Generative AI" refers to AI technology that generates new designs and content based on user input data.

[1236] "Tone and style of design" refers to elements of design such as color, shape, and atmosphere.

[1237] The present invention relates to a fashion e-commerce system that collects a user's body type information and past purchase history, and recommends personalized products based on the user's emotional state using an emotion recognition engine. This system includes the following configuration and processing steps.

[1238] System configuration

[1239] server

[1240] The server includes the following means:

[1241] 1. Collection of body type information and purchase history: Collect body type information such as height, weight, waist size, and past purchase history from users and store them in a database.

[1242] 2. Training machine learning models: Along with the collected information, algorithms such as Collaborative Filtering and Content-based Filtering are used to train machine learning models.

[1243] 3. Use of Emotion Recognition Engine: The app has a built-in emotion recognition engine to identify the user's emotional state (relaxed, excited, etc.). This engine recognizes emotions from images and input text uploaded by the user and analyzes the data.

[1244] 4. Recommendation adjustment: Adjust the recommendation results obtained from the machine learning model based on the results of the emotion recognition engine.

[1245] Terminal

[1246] The terminal includes the following means:

[1247] 1. Data transmission and reception: When a user accesses an e-commerce site, their body type information, purchase history, and images and text for emotion recognition are sent to the server. The server also displays the recommendations and generated designs.

[1248] 2. Interactive features: Provides chat windows and forms that allow users to enter requests and questions in natural language, allowing users to enter specific requests such as "I want a casual summer dress."

[1249] User

[1250] The user performs the following activities:

[1251] 1. Information input: Enter your body type and past purchase history, as well as requests and questions about fashion items in natural language.

[1252] 2. Image Upload: Upload your own photo and use it to analyze your emotional state with the emotion recognition engine.

[1253] Details of the hardware and software you will use

[1254] Smartphone: An interface that allows users to enter information, upload images, and view recommendation results.

[1255] PIL (Python Imaging Library): Used to load user images.

[1256] EmotionRecognizer: An emotion recognition library used to recognize the emotional state of a user from their image.

[1257] RecommenderSystem: A system that recommends optimal products based on the user's body type information and past purchase history.

[1258] Specific examples

[1259] 1. The user enters their body information (height 170cm, weight 65kg, waist size 75cm) into a smartphone app, along with the T-shirts, jeans, sweatshirts, etc. they have purchased in the past.

[1260] 2. The user uploads a photo and EmotionRecognizer recognizes the "relaxed state" from the photo.

[1261] 3. The Recommender System recommends the most suitable relaxing wear based on the user's data.

[1262] 4. The recommendation results are displayed to the user, suggesting "comfortable relaxing wear."

[1263] Example prompts for generative AI models

[1264] "If the user is relaxing, suggest the best relaxing wear."

[1265] In this way, the present invention realizes providing a highly personalized fashion e-commerce service that meets the individual needs of users and takes into account their emotional state.

[1266] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1267] Step 1:

[1268] User input of information

[1269] A user enters their body information (e.g., height, weight, waist size) and past purchase history into a smartphone app. This input information is sent from the device to a server. Specifically, the user provides information using an input form, and the input data is stored on the server through the device interface.

[1270] input:

[1271] User's body type information and past purchase history

[1272] output:

[1273] User information data stored on the server

[1274] Step 2:

[1275] Image upload and emotion recognition

[1276] A user uploads a photo of themselves. This photo data is sent from the device to a server, where an emotion recognition engine (e.g., EmotionRecognizer) is used to analyze the user's emotional state. Specifically, after the image data arrives at the server, the emotion recognition engine receives it as input and outputs the user's emotional state (e.g., relaxed, excited, etc.).

[1277] input:

[1278] User photo data

[1279] output:

[1280] User emotional state data

[1281] Step 3:

[1282] Training a machine learning model

[1283] The server trains a machine learning model using algorithms such as Collaborative Filtering and Content-based Filtering based on the collected user data and past purchase history. This builds a model for predicting user preferences and trends. Specifically, the machine learning model receives user data as input and learns patterns to improve the accuracy of product recommendations.

[1284] input:

[1285] Collected user data and purchase history

[1286] output:

[1287] Trained machine learning models

[1288] Step 4:

[1289] Recommendation generation and adjustment

[1290] The server uses the trained machine learning model to recommend appropriate products to the user. At this time, it also takes into account the user's emotional state obtained from the emotion recognition engine to adjust the recommendations. Specifically, it receives the user's emotional state as input and generates an optimal product list based on the output of the machine learning model.

[1291] input:

[1292] Trained machine learning model and user emotional state data

[1293] output:

[1294] Recommended product list

[1295] Step 5:

[1296] Displaying recommendations

[1297] The terminal displays the recommended product list sent from the server to the user. The user can check the suggested products through the interface of the smartphone app. Specifically, the product list is sent to the terminal and displayed on the user interface of the terminal.

[1298] input:

[1299] Recommended product list

[1300] output:

[1301] Recommended products displayed on the device

[1302] Step 6:

[1303] Interactive information provision

[1304] The server uses an interactive AI engine based on the user's request to analyze the request in natural language and search for related product information. For example, if a user inputs "I want a casual summer dress," the AI ​​engine will analyze the request and search for related product information.

[1305] input:

[1306] Natural language requests from users

[1307] output:

[1308] Parsed request and associated product information

[1309] Step 7:

[1310] Custom-made designs with generative AI

[1311] The server generates new clothing designs using an artificial intelligence model generated based on the user's text input. During this process, the user's emotional state is also taken into account using an emotion recognition engine. The specific prompt used is, "If the user is relaxed, please suggest the most suitable relaxing wear."

[1312] input:

[1313] User text input information and emotional state

[1314] output:

[1315] Generated custom designs

[1316] In this way, a personalized fashion e-commerce service is provided that comprehensively takes into account the individual needs and emotional state of the user.

[1317] 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.

[1318] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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.

[1319] 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.

[1320] [Third embodiment]

[1321] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1322] 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.

[1323] 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).

[1324] 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.

[1325] 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.

[1326] 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).

[1327] 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. 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.

[1328] 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.

[1329] 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.

[1330] 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.

[1331] 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.

[1332] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1333] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history, uses AI to recommend optimal fashion items, and also provides custom-made clothing based on the user's requests. This system is configured as follows.

[1334] Recommendation function implementation

[1335] server

[1336] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering.

[1337] Terminal

[1338] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[1339] User

[1340] Users enter their purchase history, body type information, etc., and when they access the site, they can receive product suggestions that suit them.

[1341] Specific examples

[1342] When a user purchases a T-shirt, the app will display the best size T-shirt and related recommended items based on their past purchase history and body type information.

[1343] Implementing conversational AI search functionality

[1344] server

[1345] The server is equipped with a conversational AI engine that analyzes natural language requests from users, searches for relevant product information from a database based on the user's input, and returns the results.

[1346] Terminal

[1347] The terminal has a chat window where users can freely input requests and questions. The terminal passes the requests to the server, which then displays the results of the server's analysis to the user.

[1348] User

[1349] For example, users can enter, "I'm looking for a casual summer dress," and receive suggestions from the conversational AI.

[1350] Specific examples

[1351] When a user types "I'm looking for a sports jacket" into the chat window, the AI ​​engine searches for the appropriate products and displays a list of recommended sports jackets.

[1352] Implementing custom features using generative AI

[1353] server

[1354] The server uses generative AI to generate clothing designs based on textual information entered by the user, such as a specific request like "I want a red dress with a belt."

[1355] Terminal

[1356] The terminal transmits the user's custom-made request to the server and displays the generated design.

[1357] User

[1358] Users can enter information into a custom request form, view the generated design, and purchase it.

[1359] Specific examples

[1360] A user enters "I want a blue chiffon blouse" into the custom-made request form, and the generative AI generates a design based on that text and presents it to the user.

[1361] Implementation of AI compatibility assessment function

[1362] server

[1363] The server uses an AI model to analyze the photos uploaded by the user and images of the products they plan to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[1364] Terminal

[1365] The device sends the user's image to the server and displays the results of the compatibility assessment, which can be used as a reference when making a purchasing decision.

[1366] User

[1367] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server.

[1368] Specific examples

[1369] When a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the image and displays a result such as "This dress fits you 88%."

[1370] By combining these features, users can easily find the products that best suit them, minimizing post-purchase dissatisfaction and waste. Furthermore, by allowing users to easily create and purchase customized, made-to-order clothing, the service not only increases user satisfaction but also contributes to the realization of sustainable fashion.

[1371] The processing flow will be explained below.

[1372] Specific processing of the recommendation function

[1373] Step 1:

[1374] Server: When a user registers on the site, they are asked to enter their body information (height, weight, waist size, etc.) and this information is saved in a database.

[1375] Step 2:

[1376] Server: When a user completes a product purchase, the purchase history (product name, category, size, color, etc.) is saved in a database.

[1377] Step 3:

[1378] Server: Once a certain amount of data has been accumulated, the user data is used to train a machine learning model, using algorithms such as Collaborative Filtering and Content-based Filtering.

[1379] Step 4:

[1380] Terminal: When a user logs in to the site, the currently logged-in user ID is sent to the server.

[1381] Step 5:

[1382] Server: Based on the submitted user ID, the trained model generates the optimal recommended products.

[1383] Step 6:

[1384] Server: Returns the generated recommended product list to the terminal.

[1385] Step 7:

[1386] Terminal: Displays the received recommended product list to the user.

[1387] Specific processing of conversational AI search function

[1388] Step 1:

[1389] User: Launches the site's conversational AI search widget and enters a question or request in natural language (e.g., "I'm looking for a casual summer dress").

[1390] Step 2:

[1391] Terminal: Sends user input to the server.

[1392] Step 3:

[1393] Server: A conversational AI engine analyzes the user's natural language input and understands their requests and questions.

[1394] Step 4:

[1395] Server: Based on the analysis results, search for related product information from the database.

[1396] Step 5:

[1397] Server: Returns search results to the device.

[1398] Step 6:

[1399] Terminal: Displays search results to the user.

[1400] Specific processing of custom features using generative AI

[1401] Step 1:

[1402] User: Enters details into a custom request form (e.g., "I would like a red dress with a belt").

[1403] Step 2:

[1404] Terminal: Sends request information to the server.

[1405] Step 3:

[1406] Server: Uses generative AI to generate designs based on user requests.

[1407] Step 4:

[1408] Server: Sends the generated design image back to the device.

[1409] Step 5:

[1410] Terminal: Displays the generated design to the user and asks for their confirmation.

[1411] Step 6:

[1412] User: Checks the generated design and confirms the order if satisfied.

[1413] Specific processing of the AI ​​compatibility judgment function

[1414] Step 1:

[1415] Users upload a photo of themselves and an image of the clothing they plan to purchase to the site.

[1416] Step 2:

[1417] Terminal: Sends the uploaded image data to the server.

[1418] Step 3:

[1419] Server: Analyzes the image using a compatibility assessment AI model. Specifically, it compares the user's photo with the image of the clothing and scores the degree of fit.

[1420] Step 4:

[1421] Server: Generates the analyzed fitness score and other feedback.

[1422] Step 5:

[1423] Server: Sends the generated fitness score and feedback back to the device.

[1424] Step 6:

[1425] Device: Presents fitness scores and feedback to the user.

[1426] These concrete steps allow users to easily find the perfect product and order custom-made clothing, minimizing post-purchase frustration and waste.

[1427] Example 1

[1428] Next, a description will be given of 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."

[1429] Conventional fashion e-commerce systems have limited functionality to provide users with optimal products, lacking in recommendation accuracy and customization capabilities. Furthermore, they lack comprehensive functionality to improve the user experience, such as conversational AI search, custom-made features, and pre-purchase compatibility assessment. As a result, users have difficulty finding the right products, which reduces their motivation to purchase.

[1430] 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.

[1431] In this invention, the server includes: means for collecting a user's physical information and historical purchase records; means for training a machine learning model based on the collected information; means for recommending appropriate products to the user using the trained machine learning model; means including an interactive AI engine for analyzing interactive requests from the user; means for searching for related product information based on the generated analysis results and displaying it to the user; means for generating designs using artificial intelligence based on the user's text input information; means for providing the generated designs to the user; and means for performing image analysis and compatibility assessment, scoring the most suitable products for the user, and displaying them. This enables a comprehensive system that integrates a variety of functions, such as product recommendations optimized for the user, search using interactive AI, custom-made generation, and compatibility assessment.

[1432] "User's physical information" refers to data related to the user's physical shape, such as height, weight, and waist size.

[1433] "Historical purchase records" refers to data and history of products purchased by a user in the past.

[1434] A "machine learning model" is a set of algorithms that uses collected data to learn user preferences and behavioral patterns and recommend optimal products.

[1435] An "interactive artificial intelligence engine" is an artificial intelligence system that uses natural language processing technology to analyze user requests and generate appropriate responses.

[1436] "Generative AI" is an AI technology that generates specific outputs (e.g., designs) based on user input.

[1437] "Image analysis" is the process of analyzing images of users and products, extracting features, and comparing and judging them.

[1438] "Compatibility assessment" is the process of quantifying how well a user and product match based on the results of image analysis.

[1439] "Scoring" refers to expressing the results of a relevance assessment as a numerical value.

[1440] "Searching for product information" refers to extracting related product data from a database based on a user request.

[1441] "Generating a design" is the process of using generative AI to create a specific design or item based on a user request.

[1442] "Recommending appropriate products to users" refers to selecting and providing products that best match the user's preferences and needs through machine learning models.

[1443] "Searching for related product information and displaying it to the user" refers to extracting related product data based on the analysis results of the interactive AI engine and displaying it on the user's device.

[1444] MODE FOR CARRYING OUT THE INVENTION

[1445] This invention relates to a fashion e-commerce system that utilizes a user's physical information and historical purchase records, uses artificial intelligence to recommend optimal fashion items, and even provides custom-made clothing based on the user's requests.

[1446] Recommendation function implementation

[1447] server

[1448] The server first collects the user's submitted physical information (e.g., height, weight, waist size) and historical purchase records and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. The model uses algorithms such as collaborative filtering and content-based filtering.

[1449] Terminal

[1450] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[1451] User

[1452] Users enter their physical information and purchase history on the e-commerce site, and when they return to the site, they can receive recommendations for products that are best suited to them.

[1453] Specific examples

[1454] When a user purchases a T-shirt, the server recommends a size "L" white T-shirt or blue jeans based on their past purchase history and physical information. The device displays this information on the e-commerce site, allowing the user to consider the purchase.

[1455] Implementing conversational AI search functionality

[1456] server

[1457] The server is equipped with a conversational AI engine that analyzes requests made in natural language by the user, searches for relevant product information from a database based on the analyzed request, and returns the results.

[1458] Terminal

[1459] The terminal has a chat window where users can freely enter questions or requests. The terminal sends these to the server, which then displays the analysis results to the user.

[1460] User

[1461] Users can type "I'm looking for a casual summer dress" into the chat window and receive suggestions from the conversational AI.

[1462] Specific examples

[1463] When a user types "I'm looking for a sports jacket" into a chat window, the server's conversational AI engine searches for the keyword "sports jacket" and displays the results as a list in the chat window.

[1464] Implementing custom features using generative AI

[1465] server

[1466] The server uses generative AI to generate clothing designs based on textual information entered by the user, such as a specific request like "I want a red dress with a belt."

[1467] Terminal

[1468] The terminal transmits the user's custom-made request to the server and displays the generated design.

[1469] User

[1470] Users can enter specific information into a custom request form, view the generated design, and purchase it.

[1471] Specific examples

[1472] If a user enters "I want a blue chiffon blouse" into a form, the server will generate a design using a generative AI model based on that request and display the result to the user.

[1473] Implementation of AI compatibility assessment function

[1474] server

[1475] The server uses an AI model to analyze the photos of the user uploaded by the user and the images of the product they are planning to purchase, and uses image analysis technology to quantify the fit between the user and the product.

[1476] Terminal

[1477] The device sends the user's image to the server and displays the results of the compatibility assessment, which the user can use as a reference when making a purchase decision.

[1478] User

[1479] Users upload a full-body photo of themselves and an image of the product they plan to purchase and check the analysis results.

[1480] Specific examples

[1481] When a user uploads a full-body photo and an image of the dress they want to try on, the server analyzes the image and provides results such as "This dress fits you 88%."

[1482] Example prompts for generative AI models

[1483] "I want a red dress with a belt."

[1484] "I want a blue chiffon blouse."

[1485] "I'm looking for a casual summer dress."

[1486] I'm looking for a sports jacket.

[1487] This invention significantly improves the user experience by allowing the server, terminal, and user to appropriately link these means, allowing users to easily find the best products and easily create and purchase customized, made-to-order products.

[1488] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1489] Recommendation processing steps

[1490] Step 1:

[1491] server

[1492] Collect physical information and historical purchase records from users. This input data is first stored in a database.

[1493] Specific actions

[1494] The user enters their height, weight, waist size, purchase history, etc. into a form. This information is immediately sent to the server and stored in a database.

[1495] Step 2:

[1496] server

[1497] The collected data is used to train machine learning models that learn user preferences using algorithms such as Collaborative Filtering and Content-based Filtering.

[1498] Specific actions

[1499] The server reads the stored data and uses algorithms to train the model, which improves its ability to predict what products will suit users.

[1500] Step 3:

[1501] Terminal

[1502] When a user accesses an e-commerce site, a recommendation request including the user ID is sent from the terminal to the server.

[1503] Specific actions

[1504] When a user logs in to the site, a recommendation request is automatically sent to the server, and the user ID is attached so personalized products are returned.

[1505] Step 4:

[1506] server

[1507] Using a trained machine learning model, the system recommends the best products for the user, generating a recommendation list and sending it to the device.

[1508] Specific actions

[1509] The server retrieves relevant information from a database based on the user ID, uses a machine learning model to generate a recommendation list, and sends it back to the device.

[1510] Step 5:

[1511] Terminal

[1512] The terminal displays the recommended products returned from the server, and the user can select from the displayed products.

[1513] Specific actions

[1514] The recommended products are displayed in the user interface, allowing the user to review them and select their favorite products.

[1515] Conversational AI search function processing steps

[1516] Step 1:

[1517] User

[1518] Enter your natural language request into the chat window, including the specific product category and details you require.

[1519] Specific actions

[1520] The user types "I'm looking for a casual summer dress" into the chat window. This becomes the input data.

[1521] Step 2:

[1522] Terminal

[1523] The user's request is sent to the server, and the input content in the chat window is passed as is as data.

[1524] Specific actions

[1525] The terminal transmits the input natural language data to the server.

[1526] Step 3:

[1527] server

[1528] The conversational AI engine analyzes the user's request and searches the database for relevant product information.

[1529] Specific actions

[1530] The conversational AI engine extracts keywords such as "casual," "summer," and "dress," and searches for corresponding products from the database.

[1531] Step 4:

[1532] server

[1533] The search results are generated as a chat list and sent back to the device.

[1534] Specific actions

[1535] The searched products are returned in list form to the terminal.

[1536] Step 5:

[1537] Terminal

[1538] The terminal displays the returned search results to the user, who can then check the product details in the chat window.

[1539] Specific actions

[1540] The product list will be displayed in the chat window, allowing the user to check the contents.

[1541] Processing steps for custom-made features using generative AI

[1542] Step 1:

[1543] User

[1544] Fill out the custom request form with specific details and submit it. For example, enter a detailed request such as "I want a red dress with a belt."

[1545] Specific actions

[1546] A user enters "I want a blue chiffon blouse" into a custom-made request form and submits it.

[1547] Step 2:

[1548] Terminal

[1549] The device sends a request from the user to the server, and this request data becomes the input for the generation AI.

[1550] Specific actions

[1551] The terminal transmits the requested information to the server as is.

[1552] Step 3:

[1553] server

[1554] The generative AI model generates designs based on requests, automatically generating clothing designs based on input parameters.

[1555] Specific actions

[1556] The server passes the request information to the generative AI model, which then generates a specific design that reflects the requirements, such as "blue color" and "chiffon material."

[1557] Step 4:

[1558] server

[1559] The generated design is formatted for a user interface and sent back to the terminal.

[1560] Specific actions

[1561] The generated design data is converted into a format that can be confirmed by the user and sent back to the terminal.

[1562] Step 5:

[1563] Terminal

[1564] The terminal displays the generated design to the user, who can then review it and proceed with the purchase.

[1565] Specific actions

[1566] The generated design is displayed on the terminal, and the user can confirm the design before proceeding with the purchase.

[1567] Processing steps of the AI ​​compatibility assessment function

[1568] Step 1:

[1569] User

[1570] Upload a full-body photo of yourself and an image of the product you plan to purchase.

[1571] Specific actions

[1572] Users upload their full-body photos and product images to the site, which serve as input data.

[1573] Step 2:

[1574] Terminal

[1575] The device sends the uploaded image to the server, where the image data is analyzed.

[1576] Specific actions

[1577] The terminal sends the photo and product image to the server.

[1578] Step 3:

[1579] server

[1580] It uses image analysis technology to compare the user's photo with the product image to determine compatibility. This analysis uses image recognition technology with machine learning.

[1581] Specific actions

[1582] The server's AI analyzes the photo and product image and scores how well it fits.

[1583] Step 4:

[1584] server

[1585] A score is generated as a result of the relevance determination and sent back to the terminal.

[1586] Specific actions

[1587] The judgment results are generated as a score that is easy for the user to understand and are sent back to the terminal.

[1588] Step 5:

[1589] Terminal

[1590] The terminal displays the results of the compatibility assessment to the user, who can then use these results to make product selections.

[1591] Specific actions

[1592] For example, the result "This dress fits you 88%" may be displayed to the user to help them make a purchase.

[1593] (Application example 1)

[1594] Next, a description will be given of Application 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."

[1595] While existing fashion e-commerce platforms have recommendation systems based on users' body type information and past purchase history, they do not offer visual try-on in virtual stores or product recommendations through real-time voice interaction. Furthermore, the process of generating custom designs using generative AI and providing those designs to users is incomplete, failing to sufficiently improve user satisfaction. Furthermore, the process of users generating specific designs using prompts is complex.

[1596] 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.

[1597] In this invention, the server includes means for collecting a user's body type information and past purchase history, means for training a machine learning model based on the collected information, means for recommending optimal products to the user using the trained machine learning model, means for allowing the user to visualize and try on items in a virtual shop, and means for responding to the user's questions and requests via voice recognition. This allows the user to receive recommendations for optimal products in real time in the virtual shop, visually try on items, and have their questions and requests about the products answered through voice interaction. Furthermore, the process of generating and providing custom designs based on the user's specific requirements using generative AI is simplified, thereby improving user satisfaction.

[1598] "User's body type information" is information about the user's physical characteristics such as height, weight, and waist size.

[1599] "Past purchase history" is a record of products that a user has purchased in the past.

[1600] A "machine learning model" is a program that uses algorithms to find patterns and make predictions or classifications based on collected data.

[1601] "Recommendation methods" are methods that use machine learning models to present optimal products based on data such as the user's body type and purchase history.

[1602] A "virtual shop" is a virtual store that users can access via the Internet to select, try on, and purchase products.

[1603] "Visualization" is the process of showing the user what the product will look like in the virtual shop.

[1604] "Trying on" is the process by which a user simulates how clothes and accessories will look on them in a virtual environment.

[1605] "Speech recognition" is a technology that analyzes a user's voice and converts it into text.

[1606] An "interactive AI engine" is a program that analyzes requests and questions from users in natural language and generates appropriate responses.

[1607] "Generative AI" is an artificial intelligence algorithm that creates new designs and content based on information and requests provided by users.

[1608] A "prompt sentence" is a piece of text given to a generation AI as specific input, and serves as a guide to control the direction of the generated output.

[1609] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history, uses AI to recommend optimal fashion items, and also provides custom-made clothing based on the user's requests. This system is configured as follows.

[1610] Recommendation function implementation

[1611] server

[1612] The server first collects the user's body type information (height, weight, waist size, etc.) and past purchase history and stores them in a database. It then trains a machine learning model based on this data to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering. Specifically, each time a user accesses the e-commerce site, the trained model presents the user with personalized product recommendations.

[1613] Terminal

[1614] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server. The device displays the recommended products returned by the server, helping the user easily find the right product. Using smart glasses or a head-mounted display, the user can visualize and try on products in a virtual store.

[1615] User

[1616] Users enter their purchase history, body type, and other information. When they access the site, they can receive product suggestions that suit them. They can also visually try on and purchase products in a virtual store.

[1617] Implementing conversational AI search functionality

[1618] server

[1619] The server is equipped with a conversational AI engine that analyzes natural language requests from users. Based on the user's input, it searches for relevant product information from a database and returns the results. The conversational AI engine uses speech recognition technology to respond to the user's voice questions and convert them into text.

[1620] Terminal

[1621] The terminal has a chat window where users can freely input requests and questions. It also responds to user voice requests using a voice dialogue function. The terminal passes these requests to the server, and the server's analysis results are displayed to the user.

[1622] User

[1623] For example, users can input "I want a casual summer dress" and receive suggestions from the conversational AI. Voice input is possible, making it highly convenient.

[1624] Implementing custom features using generative AI

[1625] server

[1626] The server uses a generative AI to generate clothing designs based on text input from the user, such as a specific request like "I want a red dress with a belt." The generative AI then uses prompts to suggest the best design.

[1627] Terminal

[1628] The terminal sends the user's custom-made request to the server and displays the generated design, allowing the user to confirm the design and decide to purchase it.

[1629] User

[1630] Users can enter information into a custom-made request form, view the generated design, and purchase it. For example, if you enter "I want a blue chiffon blouse," the AI ​​will generate a design based on that text and present it to the user.

[1631] Implementation of AI compatibility assessment function

[1632] server

[1633] The server uses an AI model to analyze the photos uploaded by the user and images of the product they are planning to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[1634] Terminal

[1635] The device sends the user's image to the server and displays the results of the compatibility assessment, which the user can use to make a purchasing decision.

[1636] User

[1637] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server. Specifically, when a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the images and displays results such as "This dress fits you 88%."

[1638] By combining these features, users can easily find the products that best suit them, minimizing post-purchase dissatisfaction and waste. Furthermore, by allowing users to easily create and purchase customized, made-to-order clothing, it is possible to increase user satisfaction and contribute to the realization of sustainable fashion.

[1639] Specific hardware and software used

[1640] Hardware: Smart glasses, head-mounted displays

[1641] Software: Python, cloud computing services (image analysis services), natural language processing libraries (conversational AI, generative AI models), image processing libraries

[1642] Example prompt: "A blue chiffon blouse"

[1643] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1644] Step 1:

[1645] The server receives the user's body type information (height, weight, waist size, etc.) and past purchase history and stores them in a database. At this time, the input data is the body type information and purchase history provided by the user, which the server receives and stores in the database in an appropriate format. This creates a data set that reflects the user's individual characteristics.

[1646] Step 2:

[1647] The server trains a machine learning model based on the collected body type information and purchase history. The input data is the user's saved body type information and purchase history, and the output is a model that predicts the most suitable fashion items for each user. Specifically, the model is trained using algorithms such as Collaborative Filtering and Content-based Filtering.

[1648] Step 3:

[1649] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server. Request data including the user's ID and session information is sent as input, and the server returns a list of recommended products. This allows the user to receive suggestions for the most suitable fashion items.

[1650] Step 4:

[1651] The terminal displays recommended products to the user in a virtual shop, allowing them to visualize and try on the products. The input is a list of recommended products returned from the server, and the output is an environment in which the user can try on and visualize the products using AR or VR technology. Specifically, smart glasses or a head-mounted display are used to display 3D models of the products.

[1652] Step 5:

[1653] The server uses a conversational AI engine to analyze natural language requests from users. The input is the user's voice or text question or request, and the output is the analyzed result, which is related product information or a response. The conversational AI engine uses speech recognition technology to convert the user's voice into text and generate an appropriate response.

[1654] Step 6:

[1655] The device receives the analysis results from the server and displays them to the user on a screen or via voice. The input is the response or product information generated by the conversational AI engine, and the output is the information displayed to the user. This allows the user to quickly and accurately obtain the information they need.

[1656] Step 7:

[1657] The server uses a generative AI to generate clothing designs based on text information entered by the user. The input is the text information entered by the user into the custom-made request form, and the output is the clothing design created by the generative AI. For requests such as "I want a blue chiffon blouse," the server generates a design based on the prompt text.

[1658] Step 8:

[1659] The device receives the generated design from the server and displays it to the user. The input is the design created by the generative AI, and the output is the user's screen where the design can be viewed. This allows the user to view the custom-made design and proceed with further customization or purchase.

[1660] Step 9:

[1661] The server analyzes the user's uploaded photo and the product image of the product they are planning to purchase to determine compatibility. The input is the user's photo and the product image, and the output is a compatibility score. Using image analysis technology, an AI model determines the fit between the user and the product.

[1662] Step 10:

[1663] The device receives the relevance assessment results from the server and presents them to the user. The input is the server's relevance score, and the output is feedback on the user's screen. This provides the user with information to help them make a purchasing decision and make an appropriate choice.

[1664] 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.

[1665] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history to recommend optimal fashion items using AI, and even provides custom-made clothing based on the user's requests. It also combines an emotion engine that recognizes the user's emotions to provide a more personalized service. This system is configured as follows:

[1666] Recommendation function implementation

[1667] server

[1668] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering.

[1669] Terminal

[1670] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[1671] User

[1672] Users enter their purchase history, body type information, etc., and when they access the site, they can receive product suggestions that suit them.

[1673] Specific examples

[1674] When a user purchases a T-shirt, the app will display the best size T-shirt and related recommended items based on their past purchase history and body type information.

[1675] Implementing conversational AI search functionality

[1676] server

[1677] The server is equipped with a conversational AI engine that analyzes natural language requests from users, searches for relevant product information from a database based on the user's input, and returns the results. It also uses an emotion engine to recognize the user's emotional state and tailor its response accordingly.

[1678] Terminal

[1679] The terminal has a chat window where users can freely input requests and questions. The terminal passes the requests to the server, which then displays the results of the server's analysis to the user.

[1680] User

[1681] For example, users can input "I want a casual summer dress" and receive suggestions from the conversational AI. They can also receive more appropriate products and advice depending on the user's emotional state while inputting.

[1682] Specific examples

[1683] When a user types "I'm looking for a sports jacket" in the chat window, the AI ​​engine searches for the appropriate product and displays a list of sports jacket recommendations. If the user is excited by the emotion engine, it will suggest more active designs.

[1684] Implementing custom features using generative AI

[1685] server

[1686] The server uses generative AI to generate clothing designs based on text input from the user, such as a specific request like "I want a red dress with a belt," and an emotional engine to recognize the user's emotional state and adjust the tone and style of the design.

[1687] Terminal

[1688] The terminal transmits the user's custom-made request to the server and displays the generated design.

[1689] User

[1690] Users can enter information into a custom request form, review the generated design, and purchase it. They can also receive design suggestions based on their emotional state.

[1691] Specific examples

[1692] A user enters "I want a blue chiffon blouse" into the custom-made request form, and the generative AI generates a design based on that text and presents it to the user. If the user is relaxed, the emotion engine suggests a softer design.

[1693] Implementation of AI compatibility assessment function

[1694] server

[1695] The server uses an AI model to analyze the photos uploaded by the user and images of the products they plan to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[1696] Terminal

[1697] The device sends the user's image to the server and displays the results of the compatibility assessment, which can be used as a reference when making a purchasing decision.

[1698] User

[1699] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server.

[1700] Specific examples

[1701] When a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the image and displays a result such as "This dress fits you 88%."

[1702] Specific implementation of adjustments using the emotion engine

[1703] server

[1704] The server is equipped with an emotion engine that recognizes the user's emotional state from input information and images. Based on the results, the engine adjusts the accuracy of recommendations and the response of the conversational AI. For example, if the user is excited about a purchase, the emotion engine will adjust to provide additional suggestions.

[1705] Terminal

[1706] The device displays information adjusted by the emotion engine to the user, improving the user experience.

[1707] User

[1708] Users can receive suggestions optimized by the emotion engine and enjoy personalized services according to their individual emotional state.

[1709] Specific examples

[1710] If the user is recognized as feeling comfortable, relaxation items and designs with a relaxing effect will be suggested.

[1711] The combination of these features allows users to easily find the perfect product for them, easily order custom-made clothing, and receive personalized service based on their emotions, minimizing post-purchase dissatisfaction and waste.

[1712] The processing flow will be explained below.

[1713] Specific processing of the recommendation function

[1714] Step 1:

[1715] Server: When a user registers on the site, they are asked to enter their physical information (height, weight, waist size, etc.) and past purchase history, which is then saved in a database.

[1716] Step 2:

[1717] Server: When a user purchases a product, the purchase history (product name, category, size, color, etc.) is saved in a database.

[1718] Step 3:

[1719] Server: Periodically trains machine learning models using stored user data, using algorithms such as Collaborative Filtering and Content-based Filtering.

[1720] Step 4:

[1721] Terminal: The user logs in to the e-commerce site and sends a request for recommended products to the server.

[1722] Step 5:

[1723] Server: Based on the received user ID, the trained model is used to generate the optimal recommended product list.

[1724] Step 6:

[1725] Server: Sends the generated recommended product list to the terminal.

[1726] Step 7:

[1727] Terminal: Displays the sent recommended product list to the user.

[1728] Specific processing of conversational AI search function

[1729] Step 1:

[1730] User: Launches the site's conversational AI search widget and enters a question or request in natural language (e.g., "I'm looking for a casual summer dress").

[1731] Step 2:

[1732] Terminal: Sends user input to the server.

[1733] Step 3:

[1734] Server: Uses a conversational AI engine to analyze the user's natural language input and understand their requests and questions.

[1735] Step 4:

[1736] Server: Based on the analysis results, it uses an emotion engine to recognize the user's emotional state.

[1737] Step 5:

[1738] Server: Based on the recognized emotions and analysis results, it searches for related product information from the database.

[1739] Step 6:

[1740] Server: Sends search results to the device.

[1741] Step 7:

[1742] Terminal: Displays search results to the user.

[1743] Specific processing of custom features using generative AI

[1744] Step 1:

[1745] User: Enters details into a custom request form (e.g., "I would like a red dress with a belt").

[1746] Step 2:

[1747] Terminal: Sends request information to the server.

[1748] Step 3:

[1749] Server: Uses generative AI to generate designs based on user requests.

[1750] Step 4:

[1751] Server: The generated design image is adjusted based on the user's emotional state using an emotion engine.

[1752] Step 5:

[1753] Server: Sends the adjusted design image to the device.

[1754] Step 6:

[1755] Terminal: The adjusted design image is displayed to the user and confirmation is requested.

[1756] Step 7:

[1757] User: Checks the generated design and confirms the order if satisfied.

[1758] Specific processing of the AI ​​compatibility judgment function

[1759] Step 1:

[1760] Users upload a photo of themselves and an image of the clothing they plan to purchase to the site.

[1761] Step 2:

[1762] Terminal: Sends the uploaded image data to the server.

[1763] Step 3:

[1764] Server: Analyzes the image using a fit determination AI model. Specifically, it compares the user's photo with the image of the clothing and scores the degree of fit.

[1765] Step 4:

[1766] Server: Generates analyzed fitness scores and feedback.

[1767] Step 5:

[1768] Server: Sends the generated fitness score and feedback to the device.

[1769] Step 6:

[1770] Device: Presents fitness scores and feedback to the user.

[1771] Specific implementation of adjustments using the emotion engine

[1772] Step 1:

[1773] User: Provides input information and images when accessing the site, searching for products, interacting with the site, etc.

[1774] Step 2:

[1775] Terminal: Sends user input and images to the server.

[1776] Step 3:

[1777] Server: Using the emotion engine, recognizes emotions from user input information and images.

[1778] Step 4:

[1779] Server: Based on the recognized emotions, it adjusts recommendation results, conversational AI responses, and generative AI designs.

[1780] Step 5:

[1781] Server: Sends the adjustment results to the terminal.

[1782] Step 6:

[1783] Terminal: Display tailored information and suggestions to the user.

[1784] These concrete steps allow users to easily find the perfect product, easily order custom-made clothing, and receive personalized service based on their emotions, minimizing post-purchase dissatisfaction and waste.

[1785] Example 2

[1786] Next, a description will be given of 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."

[1787] Conventional fashion e-commerce systems do not fully utilize a user's body type information or past purchase history, making it difficult to recommend optimal products to the user. Furthermore, personalized recommendations that take the user's emotional state into consideration are not made, resulting in a poor user experience. Furthermore, it is difficult to respond quickly and accurately to requests for custom-made items, which is a factor that reduces user satisfaction. The present invention aims to solve these problems.

[1788] 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.

[1789] In this invention, the server includes: means for collecting a user's body type information and past purchase history; means for training a machine learning model based on the collected information; means for recommending optimal products to the user using the trained machine learning model; means for analyzing the user's natural language input; means for searching for related product information based on the analysis results by the dialogue engine; means for recognizing the user's emotional state and adjusting suggestions based on the results; and means for generating custom designs based on the user's input information using generative AI technology. This makes it possible to comprehensively utilize the user's body type information, purchase history, and emotional state to propose optimal products and provide customized services to individual users.

[1790] 1. "Means of collection" refers to the equipment and software used to receive a user's body information and past purchase history and store it in a database.

[1791] 2. "Training means" refers to the process of using collected data to train a machine learning model to predict the best products for a user.

[1792] 3. "Recommendation method" refers to a system or algorithm that uses a trained machine learning model to suggest optimal products to users.

[1793] 4. "Dialogue engine" refers to software or a system that analyzes a user's natural language input and extracts relevant information based on that input.

[1794] 5. "Searching means" refers to the process of searching for appropriate product information from the database based on the results analyzed by the dialogue engine.

[1795] 6. "Emotion engine" refers to an algorithm that recognizes a user's emotional state from input information and images, and adjusts the system's responses and suggestions based on the results.

[1796] 7. "Generative AI technology" refers to artificial intelligence technology that automatically generates custom designs based on text information entered by the user.

[1797] 8. "Custom Design" refers to a unique clothing or product design created according to a user's specific requirements.

[1798] 9. "Means of analysis" refers to the process of analyzing uploaded images and input information to determine the suitability and suitability of products that meet the user's requirements.

[1799] 10. "Means of scoring" refers to the process of quantifying the suitability of a product based on the analysis results and evaluating its fit for the user.

[1800] This invention is a fashion e-commerce system that utilizes a user's body type information and past purchase history to recommend the most suitable fashion items using AI, and even provides custom-made clothing based on the user's requests. It also combines an emotion engine that recognizes the user's emotions to provide a more personalized service. This system consists of three main components: a server, a terminal, and a user.

[1801] server

[1802] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history, and stores this data in a database. This collection is done using a web framework such as Django. It then applies a trained machine learning model to recommend the best products for the user based on the collected data. Machine learning tools such as SciKit-Learn and TensorFlow are used for this training and prediction.

[1803] The server then houses a dialogue engine that analyzes the user's natural language input. This dialogue engine uses advanced natural language processing models such as GPT-3 and searches a database for relevant product information based on the analysis results. The server also houses an emotion engine that uses NLP libraries to recognize the user's emotional state from their input and images.

[1804] It also uses generative AI technology, such as DALL-E, to generate custom designs based on the user's specific requests.

[1805] Terminal

[1806] The device provides a front-end interface for users to access. Through a web browser or mobile app, users can input their body type information and purchase history and use the conversational AI search function to find products. It also has a chat window that displays the analysis results of the dialogue engine and tailored suggestions from the emotion engine. Front-end frameworks such as React.js and Vue.js are used for this display.

[1807] The device then sends the user's uploaded photos and images of the products they plan to purchase to the server, and displays the results of the matching. OpenCV and TensorFlow are used for this analysis and matching.

[1808] User

[1809] Users first input their body type and past purchase history, which allows them to receive personalized product suggestions. Next, when searching for products using the conversational AI search function, they can input their request in natural language. Users can input a prompt statement such as "I'm looking for a casual summer dress." They can also enter a specific request, such as "I'm looking for a blue chiffon blouse," into the custom-made request form, where they can view and purchase the generated custom design.

[1810] For example, if a user wants to buy a T-shirt, the system will display the best size T-shirt and other related recommended items based on their past purchase history and body type information. This allows users to easily find the products that best suit them. They can also easily order custom-made clothing and receive personalized service based on their emotions. This minimizes post-purchase dissatisfaction and waste.

[1811] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1812] Step 1: Enter and collect user information

[1813] The user inputs information about their body type (height, weight, waist size, etc.) and also provides past purchase history data. This input data is sent from the terminal to the server, which then stores this data in a database. Specifically, the user inputs information into a web form, and the form data is sent to the server.

[1814] Input: Body type information, past purchase history data

[1815] Output: Body shape information and purchase history data stored in a database

[1816] Step 2: Training the machine learning model

[1817] The server trains a machine learning model based on the collected user data. During this process, machine learning tools such as SciKit-Learn and TensorFlow are used to process the data and train the model. After training is complete, the model is saved and used for future predictions.

[1818] Input: User data in the database

[1819] Output: A trained machine learning model

[1820] Step 3: Predicting recommended products

[1821] The server uses a trained machine learning model to predict the best product for the user. The prediction result is encoded in JSON format and sent back to the device. Specifically, the server inputs the current user data into the trained model and generates a prediction result.

[1822] Input: current user data, trained model

[1823] Output: Predicted recommended product list (JSON format)

[1824] Step 4: Display recommended products

[1825] The device receives the recommended product list returned from the server and displays it on the user interface. Specifically, the device uses a front-end framework such as React.js or Vue.js to parse the JSON data and display the product list.

[1826] Input: Recommended product list (JSON format)

[1827] Output: A list of products displayed on the user interface

[1828] Step 5: Enter your request in natural language

[1829] Using the conversational AI search feature, users input their requests in natural language, such as "I'm looking for a casual summer dress" into a chat window on a website or mobile app.

[1830] Input: Natural language request text

[1831] Output: Sends the request text to the server

[1832] Step 6: Natural Language Analysis and Emotion Recognition

[1833] The server's dialogue engine receives the user's input text and performs natural language analysis and emotion recognition. This process uses natural language processing models and emotion recognition algorithms such as GPT-3. The analysis results are used in the next step.

[1834] Input: Natural language request text

[1835] Output: Analysis results and emotional state

[1836] Step 7: Find and refine product information

[1837] The server searches the database based on the analysis results to retrieve relevant product information. At the same time, it adjusts the content of the suggestions using an emotion engine. Specifically, it executes an SQL query to retrieve product information from the database and adjusts the response content according to the emotion analysis results.

[1838] Input: Analysis results and emotional state

[1839] Output: Adjusted product information list

[1840] Step 8: Viewing conversational AI responses

[1841] The terminal receives the adjusted product list from the server and displays it as a response in the chat window. Specifically, the terminal receives the response data and dynamically generates content to be displayed in the chat window.

[1842] Input: Adjusted product information list

[1843] Output: Product suggestions displayed in the chat window

[1844] Step 9: Fill out your customization request

[1845] The user enters specific information into the custom order request form, for example, text such as "I would like a blue chiffon blouse."

[1846] Input: Custom request text

[1847] Output: Request text sent to the server

[1848] Step 10: Generate a design using generative AI

[1849] The server uses a generative AI model, such as DALL-E, to generate a custom design based on the user's input. The generated design image is then sent to the device.

[1850] Input: Custom request text

[1851] Output: Generated design image

[1852] Step 11: Display your custom design

[1853] The device displays the generated design image received from the server. Specifically, the design is displayed using the HTML Canvas API.

[1854] Input: Generated design image

[1855] Output: Your custom design displayed on the user interface

[1856] Step 12: Conducting a conformance determination

[1857] Users upload photos of themselves and the products they plan to purchase from their devices to a server, which then uses an AI model to analyze the images and calculate a suitability score.

[1858] Input: User image, product image

[1859] Output: Relevance score

[1860] Step 13: Viewing the relevance results

[1861] The terminal displays the relevance score received from the server. Specifically, the terminal visualizes the relevance score and provides it to the user.

[1862] Input: Suitability score

[1863] Output: Relevance assessment results displayed on the user interface

[1864] Step 14: Optimizing recommendations with an emotion engine

[1865] The server uses an emotion engine to recognize the user's emotional state from input information and images, and adjusts the suggestions based on the results.

[1866] Input: User input information, images

[1867] Output: Adjusted proposal

[1868] Step 15: Viewing optimized suggestions

[1869] The terminal displays the dynamically generated suggestion content to the user, optimized by the emotion engine.

[1870] Input: Adjusted proposal

[1871] Output: Optimization suggestions displayed on the user interface

[1872] Through these steps, users can easily find the products that best suit them, easily order custom-made clothing, and receive personalized service based on their emotions.

[1873] (Application example 2)

[1874] Next, a description will be given of Application 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."

[1875] Conventional fashion e-commerce systems recommend products based only on a user's body type and past purchase history, making it difficult to provide personalized suggestions that take into account a user's individual emotional state and temporary preferences. This often results in users being unable to find products that satisfy them, and also leads to increased dissatisfaction and returns after purchase. Furthermore, the systems lacked interactive features that allow users to input their requirements in natural language and features that generate custom-made designs, resulting in a lack of methods to improve the user experience.

[1876] 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.

[1877] In this invention, the server includes: means for collecting a user's body type information and past purchase history; means for training a machine learning model based on the collected information; means for recommending optimal products to the user using the trained machine learning model; and means for adjusting the recommendations according to the user's emotional state using an emotion recognition engine that identifies the user's emotional state. This enables highly personalized product proposals that take the user's individual emotional state into consideration. The server also includes means for analyzing a user's request using an interactive artificial intelligence engine; means for searching for related product information based on the analysis results; means for displaying the search results to the user; and means including an emotion recognition engine that identifies the user's emotional state and adjusts the analysis results. This enables intuitive operation using natural language and improves user satisfaction. The server also includes means for generating designs using artificial intelligence based on the user's text input information; means for providing the generated designs to the user; and means including an emotion recognition engine that adjusts the tone and style of the designs based on the user's emotional state. This makes it easy to provide custom designs tailored to the user's preferences.

[1878] "User's body type information" refers to physical measurement data such as the user's height, weight, waist size, etc.

[1879] "Past purchase history" refers to information such as the products a user has purchased in the past, the purchase dates, and the quantities.

[1880] A "machine learning model" refers to an artificial intelligence algorithm that learns patterns based on data and makes predictions and classifications.

[1881] An "emotion recognition engine" refers to an artificial intelligence technology that recognizes emotions from a user's facial expressions and text input, and adjusts responses based on the results.

[1882] "Recommendation" refers to the act of suggesting products or services that are suitable for a user.

[1883] An "interactive artificial intelligence engine" refers to an artificial intelligence technology that analyzes requests made in natural language by users and generates appropriate responses.

[1884] "Generative AI" refers to AI technology that generates new designs and content based on user input data.

[1885] "Tone and style of design" refers to elements of design such as color, shape, and atmosphere.

[1886] The present invention relates to a fashion e-commerce system that collects a user's body type information and past purchase history, and recommends personalized products based on the user's emotional state using an emotion recognition engine. This system includes the following configuration and processing steps.

[1887] System configuration

[1888] server

[1889] The server includes the following means:

[1890] 1. Collection of body type information and purchase history: Collect body type information such as height, weight, waist size, and past purchase history from users and store them in a database.

[1891] 2. Training machine learning models: Along with the collected information, algorithms such as Collaborative Filtering and Content-based Filtering are used to train machine learning models.

[1892] 3. Use of Emotion Recognition Engine: The app has a built-in emotion recognition engine to identify the user's emotional state (relaxed, excited, etc.). This engine recognizes emotions from images and input text uploaded by the user and analyzes the data.

[1893] 4. Recommendation adjustment: Adjust the recommendation results obtained from the machine learning model based on the results of the emotion recognition engine.

[1894] Terminal

[1895] The terminal includes the following means:

[1896] 1. Data transmission and reception: When a user accesses an e-commerce site, their body type information, purchase history, and images and text for emotion recognition are sent to the server. The server also displays the recommendations and generated designs.

[1897] 2. Interactive features: Provides chat windows and forms that allow users to enter requests and questions in natural language, allowing users to enter specific requests such as "I want a casual summer dress."

[1898] User

[1899] The user performs the following activities:

[1900] 1. Information input: Enter your body type and past purchase history, as well as requests and questions about fashion items in natural language.

[1901] 2. Image Upload: Upload your own photo and use it to analyze your emotional state with the emotion recognition engine.

[1902] Details of the hardware and software you will use

[1903] Smartphone: An interface that allows users to enter information, upload images, and view recommendation results.

[1904] PIL (Python Imaging Library): Used to load user images.

[1905] EmotionRecognizer: An emotion recognition library used to recognize the emotional state of a user from their image.

[1906] RecommenderSystem: A system that recommends optimal products based on the user's body type information and past purchase history.

[1907] Specific examples

[1908] 1. The user enters their body information (height 170cm, weight 65kg, waist size 75cm) into a smartphone app, along with the T-shirts, jeans, sweatshirts, etc. they have purchased in the past.

[1909] 2. The user uploads a photo and EmotionRecognizer recognizes the "relaxed state" from the photo.

[1910] 3. The Recommender System recommends the most suitable relaxing wear based on the user's data.

[1911] 4. The recommendation results are displayed to the user, suggesting "comfortable relaxing wear."

[1912] Example prompts for generative AI models

[1913] "If the user is relaxing, suggest the best relaxing wear."

[1914] In this way, the present invention realizes providing a highly personalized fashion e-commerce service that meets the individual needs of users and takes into account their emotional state.

[1915] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1916] Step 1:

[1917] User input of information

[1918] A user enters their body information (e.g., height, weight, waist size) and past purchase history into a smartphone app. This input information is sent from the device to a server. Specifically, the user provides information using an input form, and the input data is stored on the server through the device interface.

[1919] input:

[1920] User's body type information and past purchase history

[1921] output:

[1922] User information data stored on the server

[1923] Step 2:

[1924] Image upload and emotion recognition

[1925] A user uploads a photo of themselves. This photo data is sent from the device to a server, where an emotion recognition engine (e.g., EmotionRecognizer) is used to analyze the user's emotional state. Specifically, after the image data arrives at the server, the emotion recognition engine receives it as input and outputs the user's emotional state (e.g., relaxed, excited, etc.).

[1926] input:

[1927] User photo data

[1928] output:

[1929] User emotional state data

[1930] Step 3:

[1931] Training a machine learning model

[1932] The server trains a machine learning model using algorithms such as Collaborative Filtering and Content-based Filtering based on the collected user data and past purchase history. This builds a model for predicting user preferences and trends. Specifically, the machine learning model receives user data as input and learns patterns to improve the accuracy of product recommendations.

[1933] input:

[1934] Collected user data and purchase history

[1935] output:

[1936] Trained machine learning models

[1937] Step 4:

[1938] Recommendation generation and adjustment

[1939] The server uses the trained machine learning model to recommend appropriate products to the user. At this time, it also takes into account the user's emotional state obtained from the emotion recognition engine to adjust the recommendations. Specifically, it receives the user's emotional state as input and generates an optimal product list based on the output of the machine learning model.

[1940] input:

[1941] Trained machine learning model and user emotional state data

[1942] output:

[1943] Recommended product list

[1944] Step 5:

[1945] Displaying recommendations

[1946] The terminal displays the recommended product list sent from the server to the user. The user can check the suggested products through the interface of the smartphone app. Specifically, the product list is sent to the terminal and displayed on the user interface of the terminal.

[1947] input:

[1948] Recommended product list

[1949] output:

[1950] Recommended products displayed on the device

[1951] Step 6:

[1952] Interactive information provision

[1953] The server uses an interactive AI engine based on the user's request to analyze the request in natural language and search for related product information. For example, if a user inputs "I want a casual summer dress," the AI ​​engine will analyze the request and search for related product information.

[1954] input:

[1955] Natural language requests from users

[1956] output:

[1957] Parsed request and associated product information

[1958] Step 7:

[1959] Custom-made designs with generative AI

[1960] The server generates new clothing designs using an artificial intelligence model generated based on the user's text input. During this process, the user's emotional state is also taken into account using an emotion recognition engine. The specific prompt used is, "If the user is relaxed, please suggest the most suitable relaxing wear."

[1961] input:

[1962] User text input information and emotional state

[1963] output:

[1964] Generated custom designs

[1965] In this way, a personalized fashion e-commerce service is provided that comprehensively takes into account the individual needs and emotional state of the user.

[1966] 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.

[1967] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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.

[1968] 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.

[1969] [Fourth embodiment]

[1970] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1971] 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.

[1972] 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).

[1973] 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.

[1974] 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.

[1975] 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).

[1976] 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. 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.

[1977] 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.

[1978] 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.

[1979] 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.

[1980] 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.

[1981] 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.

[1982] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. 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."

[1983] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history, uses AI to recommend optimal fashion items, and also provides custom-made clothing based on the user's requests. This system is configured as follows.

[1984] Recommendation function implementation

[1985] server

[1986] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering.

[1987] Terminal

[1988] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[1989] User

[1990] Users enter their purchase history, body type information, etc., and when they access the site, they can receive product suggestions that suit them.

[1991] Specific examples

[1992] When a user purchases a T-shirt, the app will display the best size T-shirt and related recommended items based on their past purchase history and body type information.

[1993] Implementing conversational AI search functionality

[1994] server

[1995] The server is equipped with a conversational AI engine that analyzes natural language requests from users, searches for relevant product information from a database based on the user's input, and returns the results.

[1996] Terminal

[1997] The terminal has a chat window where users can freely input requests and questions. The terminal passes the requests to the server, which then displays the results of the server's analysis to the user.

[1998] User

[1999] For example, users can enter, "I'm looking for a casual summer dress," and receive suggestions from the conversational AI.

[2000] Specific examples

[2001] When a user types "I'm looking for a sports jacket" into the chat window, the AI ​​engine searches for the appropriate products and displays a list of recommended sports jackets.

[2002] Implementing custom features using generative AI

[2003] server

[2004] The server uses generative AI to generate clothing designs based on textual information entered by the user, such as a specific request like "I want a red dress with a belt."

[2005] Terminal

[2006] The terminal transmits the user's custom-made request to the server and displays the generated design.

[2007] User

[2008] Users can enter information into a custom request form, view the generated design, and purchase it.

[2009] Specific examples

[2010] A user enters "I want a blue chiffon blouse" into the custom-made request form, and the generative AI generates a design based on that text and presents it to the user.

[2011] Implementation of AI compatibility assessment function

[2012] server

[2013] The server uses an AI model to analyze the photos uploaded by the user and images of the products they plan to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[2014] Terminal

[2015] The device sends the user's image to the server and displays the results of the compatibility assessment, which can be used as a reference when making a purchasing decision.

[2016] User

[2017] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server.

[2018] Specific examples

[2019] When a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the image and displays a result such as "This dress fits you 88%."

[2020] By combining these features, users can easily find the products that best suit them, minimizing post-purchase dissatisfaction and waste. Furthermore, by allowing users to easily create and purchase customized, made-to-order clothing, the service not only increases user satisfaction but also contributes to the realization of sustainable fashion.

[2021] The processing flow will be explained below.

[2022] Specific processing of the recommendation function

[2023] Step 1:

[2024] Server: When a user registers on the site, they are asked to enter their body information (height, weight, waist size, etc.) and this information is saved in a database.

[2025] Step 2:

[2026] Server: When a user completes a product purchase, the purchase history (product name, category, size, color, etc.) is saved in a database.

[2027] Step 3:

[2028] Server: Once a certain amount of data has been accumulated, the user data is used to train a machine learning model, using algorithms such as Collaborative Filtering and Content-based Filtering.

[2029] Step 4:

[2030] Terminal: When a user logs in to the site, the currently logged-in user ID is sent to the server.

[2031] Step 5:

[2032] Server: Based on the submitted user ID, the trained model generates the optimal recommended products.

[2033] Step 6:

[2034] Server: Returns the generated recommended product list to the terminal.

[2035] Step 7:

[2036] Terminal: Displays the received recommended product list to the user.

[2037] Specific processing of conversational AI search function

[2038] Step 1:

[2039] User: Launches the site's conversational AI search widget and enters a question or request in natural language (e.g., "I'm looking for a casual summer dress").

[2040] Step 2:

[2041] Terminal: Sends user input to the server.

[2042] Step 3:

[2043] Server: A conversational AI engine analyzes the user's natural language input and understands their requests and questions.

[2044] Step 4:

[2045] Server: Based on the analysis results, search for related product information from the database.

[2046] Step 5:

[2047] Server: Returns search results to the device.

[2048] Step 6:

[2049] Terminal: Displays search results to the user.

[2050] Specific processing of custom features using generative AI

[2051] Step 1:

[2052] User: Enters details into a custom request form (e.g., "I would like a red dress with a belt").

[2053] Step 2:

[2054] Terminal: Sends request information to the server.

[2055] Step 3:

[2056] Server: Uses generative AI to generate designs based on user requests.

[2057] Step 4:

[2058] Server: Sends the generated design image back to the device.

[2059] Step 5:

[2060] Terminal: Displays the generated design to the user and asks for their confirmation.

[2061] Step 6:

[2062] User: Checks the generated design and confirms the order if satisfied.

[2063] Specific processing of the AI ​​compatibility judgment function

[2064] Step 1:

[2065] Users upload a photo of themselves and an image of the clothing they plan to purchase to the site.

[2066] Step 2:

[2067] Terminal: Sends the uploaded image data to the server.

[2068] Step 3:

[2069] Server: Analyzes the image using a compatibility assessment AI model. Specifically, it compares the user's photo with the image of the clothing and scores the degree of fit.

[2070] Step 4:

[2071] Server: Generates the analyzed fitness score and other feedback.

[2072] Step 5:

[2073] Server: Sends the generated fitness score and feedback back to the device.

[2074] Step 6:

[2075] Device: Presents fitness scores and feedback to the user.

[2076] These concrete steps allow users to easily find the perfect product and order custom-made clothing, minimizing post-purchase frustration and waste.

[2077] Example 1

[2078] Next, a description will be given of 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."

[2079] Conventional fashion e-commerce systems have limited functionality to provide users with optimal products, lacking in recommendation accuracy and customization capabilities. Furthermore, they lack comprehensive functionality to improve the user experience, such as conversational AI search, custom-made features, and pre-purchase compatibility assessment. As a result, users have difficulty finding the right products, which reduces their motivation to purchase.

[2080] 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.

[2081] In this invention, the server includes: means for collecting a user's physical information and historical purchase records; means for training a machine learning model based on the collected information; means for recommending appropriate products to the user using the trained machine learning model; means including an interactive AI engine for analyzing interactive requests from the user; means for searching for related product information based on the generated analysis results and displaying it to the user; means for generating designs using artificial intelligence based on the user's text input information; means for providing the generated designs to the user; and means for performing image analysis and compatibility assessment, scoring the most suitable products for the user, and displaying them. This enables a comprehensive system that integrates a variety of functions, such as product recommendations optimized for the user, search using interactive AI, custom-made generation, and compatibility assessment.

[2082] "User's physical information" refers to data related to the user's physical shape, such as height, weight, and waist size.

[2083] "Historical purchase records" refers to data and history of products purchased by a user in the past.

[2084] A "machine learning model" is a set of algorithms that uses collected data to learn user preferences and behavioral patterns and recommend optimal products.

[2085] An "interactive artificial intelligence engine" is an artificial intelligence system that uses natural language processing technology to analyze user requests and generate appropriate responses.

[2086] "Generative AI" is an AI technology that generates specific outputs (e.g., designs) based on user input.

[2087] "Image analysis" is the process of analyzing images of users and products, extracting features, and comparing and judging them.

[2088] "Compatibility assessment" is the process of quantifying how well a user and product match based on the results of image analysis.

[2089] "Scoring" refers to expressing the results of a relevance assessment as a numerical value.

[2090] "Searching for product information" refers to extracting related product data from a database based on a user request.

[2091] "Generating a design" is the process of using generative AI to create a specific design or item based on a user request.

[2092] "Recommending appropriate products to users" refers to selecting and providing products that best match the user's preferences and needs through machine learning models.

[2093] "Searching for related product information and displaying it to the user" refers to extracting related product data based on the analysis results of the interactive AI engine and displaying it on the user's device.

[2094] MODE FOR CARRYING OUT THE INVENTION

[2095] This invention relates to a fashion e-commerce system that utilizes a user's physical information and historical purchase records, uses artificial intelligence to recommend optimal fashion items, and even provides custom-made clothing based on the user's requests.

[2096] Recommendation function implementation

[2097] server

[2098] The server first collects the user's submitted physical information (e.g., height, weight, waist size) and historical purchase records and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. The model uses algorithms such as collaborative filtering and content-based filtering.

[2099] Terminal

[2100] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[2101] User

[2102] Users enter their physical information and purchase history on the e-commerce site, and when they return to the site, they can receive recommendations for products that are best suited to them.

[2103] Specific examples

[2104] When a user purchases a T-shirt, the server recommends a size "L" white T-shirt or blue jeans based on their past purchase history and physical information. The device displays this information on the e-commerce site, allowing the user to consider the purchase.

[2105] Implementing conversational AI search functionality

[2106] server

[2107] The server is equipped with a conversational AI engine that analyzes requests made in natural language by the user, searches for relevant product information from a database based on the analyzed request, and returns the results.

[2108] Terminal

[2109] The terminal has a chat window where users can freely enter questions or requests. The terminal sends these to the server, which then displays the analysis results to the user.

[2110] User

[2111] Users can type "I'm looking for a casual summer dress" into the chat window and receive suggestions from the conversational AI.

[2112] Specific examples

[2113] When a user types "I'm looking for a sports jacket" into a chat window, the server's conversational AI engine searches for the keyword "sports jacket" and displays the results as a list in the chat window.

[2114] Implementing custom features using generative AI

[2115] server

[2116] The server uses generative AI to generate clothing designs based on textual information entered by the user, such as a specific request like "I want a red dress with a belt."

[2117] Terminal

[2118] The terminal transmits the user's custom-made request to the server and displays the generated design.

[2119] User

[2120] Users can enter specific information into a custom request form, view the generated design, and purchase it.

[2121] Specific examples

[2122] If a user enters "I want a blue chiffon blouse" into a form, the server will generate a design using a generative AI model based on that request and display the result to the user.

[2123] Implementation of AI compatibility assessment function

[2124] server

[2125] The server uses an AI model to analyze the photos of the user uploaded by the user and the images of the product they are planning to purchase, and uses image analysis technology to quantify the fit between the user and the product.

[2126] Terminal

[2127] The device sends the user's image to the server and displays the results of the compatibility assessment, which the user can use as a reference when making a purchase decision.

[2128] User

[2129] Users upload a full-body photo of themselves and an image of the product they plan to purchase and check the analysis results.

[2130] Specific examples

[2131] When a user uploads a full-body photo and an image of the dress they want to try on, the server analyzes the image and provides a result such as "This dress fits you 88%."

[2132] Example prompts for generative AI models

[2133] "I want a red dress with a belt."

[2134] "I want a blue chiffon blouse."

[2135] "I'm looking for a casual summer dress."

[2136] I'm looking for a sports jacket.

[2137] This invention significantly improves the user experience by allowing the server, terminal, and user to appropriately link these means, allowing users to easily find the best products and easily create and purchase customized, made-to-order products.

[2138] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2139] Recommendation processing steps

[2140] Step 1:

[2141] server

[2142] Collect physical information and historical purchase records from users. This input data is first stored in a database.

[2143] Specific actions

[2144] The user enters their height, weight, waist size, purchase history, etc. into a form. This information is immediately sent to the server and stored in a database.

[2145] Step 2:

[2146] server

[2147] The collected data is used to train machine learning models that learn user preferences using algorithms such as Collaborative Filtering and Content-based Filtering.

[2148] Specific actions

[2149] The server reads the stored data and uses algorithms to train the model, which improves its ability to predict what products will suit users.

[2150] Step 3:

[2151] Terminal

[2152] When a user accesses an e-commerce site, a recommendation request including the user ID is sent from the terminal to the server.

[2153] Specific actions

[2154] When a user logs in to the site, a recommendation request is automatically sent to the server, and the user ID is attached so personalized products are returned.

[2155] Step 4:

[2156] server

[2157] Using a trained machine learning model, the system recommends the best products for the user, generating a recommendation list and sending it to the device.

[2158] Specific actions

[2159] The server retrieves relevant information from a database based on the user ID, uses a machine learning model to generate a recommendation list, and sends it back to the device.

[2160] Step 5:

[2161] Terminal

[2162] The terminal displays the recommended products returned from the server, and the user can select from the displayed products.

[2163] Specific actions

[2164] The recommended products are displayed in the user interface, allowing the user to review them and select their favorite products.

[2165] Conversational AI search function processing steps

[2166] Step 1:

[2167] User

[2168] Enter your natural language request into the chat window, including the specific product category and details you require.

[2169] Specific actions

[2170] The user types "I'm looking for a casual summer dress" into the chat window. This becomes the input data.

[2171] Step 2:

[2172] Terminal

[2173] The user's request is sent to the server, and the input content in the chat window is passed as is as data.

[2174] Specific actions

[2175] The terminal transmits the input natural language data to the server.

[2176] Step 3:

[2177] server

[2178] The conversational AI engine analyzes the user's request and searches the database for relevant product information.

[2179] Specific actions

[2180] The conversational AI engine extracts keywords such as "casual," "summer," and "dress," and searches for corresponding products from the database.

[2181] Step 4:

[2182] server

[2183] The search results are generated as a chat list and sent back to the device.

[2184] Specific actions

[2185] The searched products are returned in list form to the terminal.

[2186] Step 5:

[2187] Terminal

[2188] The terminal displays the returned search results to the user, who can then check the product details in the chat window.

[2189] Specific actions

[2190] The product list will be displayed in the chat window, allowing the user to check the contents.

[2191] Processing steps for custom-made features using generative AI

[2192] Step 1:

[2193] User

[2194] Fill out the custom request form with specific details and submit it. For example, enter a detailed request such as "I want a red dress with a belt."

[2195] Specific actions

[2196] A user enters "I want a blue chiffon blouse" into a custom-made request form and submits it.

[2197] Step 2:

[2198] Terminal

[2199] The device sends a request from the user to the server, and this request data becomes the input for the generation AI.

[2200] Specific actions

[2201] The terminal transmits the requested information to the server as is.

[2202] Step 3:

[2203] server

[2204] The generative AI model generates designs based on requests, automatically generating clothing designs based on input parameters.

[2205] Specific actions

[2206] The server passes the request information to the generative AI model, which then generates a specific design that reflects the requirements, such as "blue color" and "chiffon material."

[2207] Step 4:

[2208] server

[2209] The generated design is formatted for a user interface and sent back to the terminal.

[2210] Specific actions

[2211] The generated design data is converted into a format that can be confirmed by the user and sent back to the terminal.

[2212] Step 5:

[2213] Terminal

[2214] The terminal displays the generated design to the user, who can then review it and proceed with the purchase.

[2215] Specific actions

[2216] The generated design is displayed on the terminal, and the user can confirm the design before proceeding with the purchase.

[2217] Processing steps of the AI ​​compatibility assessment function

[2218] Step 1:

[2219] User

[2220] Upload a full-body photo of yourself and an image of the product you plan to purchase.

[2221] Specific actions

[2222] Users upload their full-body photos and product images to the site, which serve as input data.

[2223] Step 2:

[2224] Terminal

[2225] The device sends the uploaded image to the server, where the image data is analyzed.

[2226] Specific actions

[2227] The terminal sends the photo and product image to the server.

[2228] Step 3:

[2229] server

[2230] It uses image analysis technology to compare the user's photo with the product image to determine compatibility. This analysis uses image recognition technology with machine learning.

[2231] Specific actions

[2232] The server's AI analyzes the photo and product image and scores how well it fits.

[2233] Step 4:

[2234] server

[2235] A score is generated as a result of the relevance determination and sent back to the terminal.

[2236] Specific actions

[2237] The judgment results are generated as a score that is easy for the user to understand and are sent back to the terminal.

[2238] Step 5:

[2239] Terminal

[2240] The terminal displays the results of the compatibility assessment to the user, who can then use these results to make product selections.

[2241] Specific actions

[2242] For example, the result "This dress fits you 88%" may be displayed to the user to help them make a purchase.

[2243] (Application example 1)

[2244] Next, a description will be given of Application 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."

[2245] While existing fashion e-commerce platforms have recommendation systems based on users' body type information and past purchase history, they do not offer visual try-on in virtual stores or product recommendations through real-time voice interaction. Furthermore, the process of generating custom designs using generative AI and providing those designs to users is incomplete, failing to sufficiently improve user satisfaction. Furthermore, the process of users generating specific designs using prompts is complex.

[2246] 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.

[2247] In this invention, the server includes means for collecting a user's body type information and past purchase history, means for training a machine learning model based on the collected information, means for recommending optimal products to the user using the trained machine learning model, means for allowing the user to visualize and try on items in a virtual shop, and means for responding to the user's questions and requests via voice recognition. This allows the user to receive recommendations for optimal products in real time in the virtual shop, visually try on items, and have their questions and requests about the products answered through voice interaction. Furthermore, the process of generating and providing custom designs based on the user's specific requirements using generative AI is simplified, thereby improving user satisfaction.

[2248] "User's body type information" is information about the user's physical characteristics such as height, weight, and waist size.

[2249] "Past purchase history" is a record of products that a user has purchased in the past.

[2250] A "machine learning model" is a program that uses algorithms to find patterns and make predictions or classifications based on collected data.

[2251] "Recommendation methods" are methods that use machine learning models to present optimal products based on data such as the user's body type and purchase history.

[2252] A "virtual shop" is a virtual store that users can access via the Internet to select, try on, and purchase products.

[2253] "Visualization" is the process of showing the user what the product will look like in the virtual shop.

[2254] "Trying on" is the process by which a user simulates how clothes and accessories will look on them in a virtual environment.

[2255] "Speech recognition" is a technology that analyzes a user's voice and converts it into text.

[2256] An "interactive AI engine" is a program that analyzes requests and questions from users in natural language and generates appropriate responses.

[2257] "Generative AI" is an artificial intelligence algorithm that creates new designs and content based on information and requests provided by users.

[2258] A "prompt sentence" is a piece of text given to a generation AI as specific input, and serves as a guide to control the direction of the generated output.

[2259] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history, uses AI to recommend optimal fashion items, and also provides custom-made clothing based on the user's requests. This system is configured as follows.

[2260] Recommendation function implementation

[2261] server

[2262] The server first collects the user's body type information (height, weight, waist size, etc.) and past purchase history and stores them in a database. It then trains a machine learning model based on this data to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering. Specifically, each time a user accesses the e-commerce site, the trained model presents the user with personalized product recommendations.

[2263] Terminal

[2264] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server. The device displays the recommended products returned by the server, helping the user easily find the right product. Using smart glasses or a head-mounted display, the user can visualize and try on products in a virtual store.

[2265] User

[2266] Users enter their purchase history, body type, and other information. When they access the site, they can receive product suggestions that suit them. They can also visually try on and purchase products in a virtual store.

[2267] Implementing conversational AI search functionality

[2268] server

[2269] The server is equipped with a conversational AI engine that analyzes natural language requests from users. Based on the user's input, it searches for relevant product information from a database and returns the results. The conversational AI engine uses speech recognition technology to respond to the user's voice questions and convert them into text.

[2270] Terminal

[2271] The terminal has a chat window where users can freely input requests and questions. It also responds to user voice requests using a voice dialogue function. The terminal passes these requests to the server, and the server's analysis results are displayed to the user.

[2272] User

[2273] For example, users can input "I want a casual summer dress" and receive suggestions from the conversational AI. Voice input is possible, making it highly convenient.

[2274] Implementing custom features using generative AI

[2275] server

[2276] The server uses a generative AI to generate clothing designs based on text input from the user, such as a specific request like "I want a red dress with a belt." The generative AI then uses prompts to suggest the best design.

[2277] Terminal

[2278] The terminal sends the user's custom-made request to the server and displays the generated design, allowing the user to confirm the design and decide to purchase it.

[2279] User

[2280] Users can enter information into a custom-made request form, view the generated design, and purchase it. For example, if you enter "I want a blue chiffon blouse," the AI ​​will generate a design based on that text and present it to the user.

[2281] Implementation of AI compatibility assessment function

[2282] server

[2283] The server uses an AI model to analyze the photos uploaded by the user and images of the product they are planning to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[2284] Terminal

[2285] The device sends the user's image to the server and displays the results of the compatibility assessment, which the user can use to make a purchasing decision.

[2286] User

[2287] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server. Specifically, when a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the images and displays results such as "This dress fits you 88%."

[2288] By combining these features, users can easily find the products that best suit them, minimizing post-purchase dissatisfaction and waste. Furthermore, by allowing users to easily create and purchase customized, made-to-order clothing, it is possible to increase user satisfaction and contribute to the realization of sustainable fashion.

[2289] Specific hardware and software used

[2290] Hardware: Smart glasses, head-mounted displays

[2291] Software: Python, cloud computing services (image analysis services), natural language processing libraries (conversational AI, generative AI models), image processing libraries

[2292] Example prompt: "A blue chiffon blouse"

[2293] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2294] Step 1:

[2295] The server receives the user's body type information (height, weight, waist size, etc.) and past purchase history and stores them in a database. At this time, the input data is the body type information and purchase history provided by the user, which the server receives and stores in the database in an appropriate format. This creates a data set that reflects the user's individual characteristics.

[2296] Step 2:

[2297] The server trains a machine learning model based on the collected body type information and purchase history. The input data is the user's saved body type information and purchase history, and the output is a model that predicts the most suitable fashion items for each user. Specifically, the model is trained using algorithms such as Collaborative Filtering and Content-based Filtering.

[2298] Step 3:

[2299] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server. Request data including the user's ID and session information is sent as input, and the server returns a list of recommended products. This allows the user to receive suggestions for the most suitable fashion items.

[2300] Step 4:

[2301] The terminal displays recommended products to the user in a virtual shop, allowing them to visualize and try on the products. The input is a list of recommended products returned from the server, and the output is an environment in which the user can try on and visualize the products using AR or VR technology. Specifically, smart glasses or a head-mounted display are used to display 3D models of the products.

[2302] Step 5:

[2303] The server uses a conversational AI engine to analyze natural language requests from users. The input is the user's voice or text question or request, and the output is the analyzed result, which is related product information or a response. The conversational AI engine uses speech recognition technology to convert the user's voice into text and generate an appropriate response.

[2304] Step 6:

[2305] The device receives the analysis results from the server and displays them to the user on a screen or via voice. The input is the response or product information generated by the conversational AI engine, and the output is the information displayed to the user. This allows the user to quickly and accurately obtain the information they need.

[2306] Step 7:

[2307] The server uses a generative AI to generate clothing designs based on text information entered by the user. The input is the text information entered by the user into the custom-made request form, and the output is the clothing design created by the generative AI. For requests such as "I want a blue chiffon blouse," the server generates a design based on the prompt text.

[2308] Step 8:

[2309] The device receives the generated design from the server and displays it to the user. The input is the design created by the generative AI, and the output is the user's screen where the design can be viewed. This allows the user to view the custom-made design and proceed with further customization or purchase.

[2310] Step 9:

[2311] The server analyzes the user's uploaded photo and the product image of the product they are planning to purchase to determine compatibility. The input is the user's photo and the product image, and the output is a compatibility score. Using image analysis technology, an AI model determines the fit between the user and the product.

[2312] Step 10:

[2313] The device receives the relevance assessment results from the server and presents them to the user. The input is the server's relevance score, and the output is feedback on the user's screen. This provides the user with information to help them make a purchasing decision and make an appropriate choice.

[2314] 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.

[2315] This invention relates to a fashion e-commerce system that utilizes a user's body type information and past purchase history to recommend optimal fashion items using AI, and even provides custom-made clothing based on the user's requests. It also combines an emotion engine that recognizes the user's emotions to provide a more personalized service. This system is configured as follows:

[2316] Recommendation function implementation

[2317] server

[2318] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history and stores them in a database. This data is then used to train a machine learning model to predict the best products for the user. This model uses algorithms such as collaborative filtering and content-based filtering.

[2319] Terminal

[2320] When a user accesses an e-commerce site, the device sends a personalized recommendation request to the server, and the device displays the recommended products returned by the server, helping the user easily find the right product.

[2321] User

[2322] Users enter their purchase history, body type information, etc., and when they access the site, they can receive product suggestions that suit them.

[2323] Specific examples

[2324] When a user purchases a T-shirt, the app will display the best size T-shirt and related recommended items based on their past purchase history and body type information.

[2325] Implementing conversational AI search functionality

[2326] server

[2327] The server is equipped with a conversational AI engine that analyzes natural language requests from users, searches for relevant product information from a database based on the user's input, and returns the results. It also uses an emotion engine to recognize the user's emotional state and tailor its response accordingly.

[2328] Terminal

[2329] The terminal has a chat window where users can freely input requests and questions. The terminal passes the requests to the server, which then displays the results of the server's analysis to the user.

[2330] User

[2331] For example, users can input "I want a casual summer dress" and receive suggestions from the conversational AI. They can also receive more appropriate products and advice depending on the user's emotional state while inputting.

[2332] Specific examples

[2333] When a user types "I'm looking for a sports jacket" in the chat window, the AI ​​engine searches for the appropriate product and displays a list of sports jacket recommendations. If the user is excited by the emotion engine, it will suggest more active designs.

[2334] Implementing custom features using generative AI

[2335] server

[2336] The server uses generative AI to generate clothing designs based on text input from the user, such as a specific request like "I want a red dress with a belt," and an emotional engine to recognize the user's emotional state and adjust the tone and style of the design.

[2337] Terminal

[2338] The terminal transmits the user's custom-made request to the server and displays the generated design.

[2339] User

[2340] Users can enter information into a custom request form, review the generated design, and purchase it. They can also receive design suggestions based on their emotional state.

[2341] Specific examples

[2342] A user enters "I want a blue chiffon blouse" into the custom-made request form, and the generative AI generates a design based on that text and presents it to the user. If the user is relaxed, the emotion engine suggests a softer design.

[2343] Implementation of AI compatibility assessment function

[2344] server

[2345] The server uses an AI model to analyze the photos uploaded by the user and images of the products they plan to purchase. The model analyzes the images and determines their suitability, generating a score for how well they fit the user.

[2346] Terminal

[2347] The device sends the user's image to the server and displays the results of the compatibility assessment, which can be used as a reference when making a purchasing decision.

[2348] User

[2349] Users upload their own photos and images of the products they plan to purchase, and check the analysis results from the server.

[2350] Specific examples

[2351] When a user uploads a full-body photo of themselves and an image of the dress they want to try on, the server analyzes the image and displays a result such as "This dress fits you 88%."

[2352] Specific implementation of adjustments using the emotion engine

[2353] server

[2354] The server is equipped with an emotion engine that recognizes the user's emotional state from input information and images. Based on the results, the engine adjusts the accuracy of recommendations and the response of the conversational AI. For example, if the user is excited about a purchase, the emotion engine will adjust to provide additional suggestions.

[2355] Terminal

[2356] The device displays information adjusted by the emotion engine to the user, improving the user experience.

[2357] User

[2358] Users can receive suggestions optimized by the emotion engine and enjoy personalized services according to their individual emotional state.

[2359] Specific examples

[2360] If the user is recognized as feeling comfortable, relaxation items and designs with a relaxing effect will be suggested.

[2361] The combination of these features allows users to easily find the perfect product for them, easily order custom-made clothing, and receive personalized service based on their emotions, minimizing post-purchase dissatisfaction and waste.

[2362] The processing flow will be explained below.

[2363] Specific processing of the recommendation function

[2364] Step 1:

[2365] Server: When a user registers on the site, they are asked to enter their physical information (height, weight, waist size, etc.) and past purchase history, which is then saved in a database.

[2366] Step 2:

[2367] Server: When a user purchases a product, the purchase history (product name, category, size, color, etc.) is saved in a database.

[2368] Step 3:

[2369] Server: Periodically trains machine learning models using stored user data, using algorithms such as Collaborative Filtering and Content-based Filtering.

[2370] Step 4:

[2371] Terminal: The user logs in to the e-commerce site and sends a request for recommended products to the server.

[2372] Step 5:

[2373] Server: Based on the received user ID, the trained model is used to generate the optimal recommended product list.

[2374] Step 6:

[2375] Server: Sends the generated recommended product list to the terminal.

[2376] Step 7:

[2377] Terminal: Displays the sent recommended product list to the user.

[2378] Specific processing of conversational AI search function

[2379] Step 1:

[2380] User: Launches the site's conversational AI search widget and enters a question or request in natural language (e.g., "I'm looking for a casual summer dress").

[2381] Step 2:

[2382] Terminal: Sends user input to the server.

[2383] Step 3:

[2384] Server: Uses a conversational AI engine to analyze the user's natural language input and understand their requests and questions.

[2385] Step 4:

[2386] Server: Based on the analysis results, it uses an emotion engine to recognize the user's emotional state.

[2387] Step 5:

[2388] Server: Based on the recognized emotions and analysis results, it searches for related product information from the database.

[2389] Step 6:

[2390] Server: Sends search results to the device.

[2391] Step 7:

[2392] Terminal: Displays search results to the user.

[2393] Specific processing of custom features using generative AI

[2394] Step 1:

[2395] User: Enters details into a custom request form (e.g., "I would like a red dress with a belt").

[2396] Step 2:

[2397] Terminal: Sends request information to the server.

[2398] Step 3:

[2399] Server: Uses generative AI to generate designs based on user requests.

[2400] Step 4:

[2401] Server: The generated design image is adjusted based on the user's emotional state using an emotion engine.

[2402] Step 5:

[2403] Server: Sends the adjusted design image to the device.

[2404] Step 6:

[2405] Terminal: The adjusted design image is displayed to the user and confirmation is requested.

[2406] Step 7:

[2407] User: Checks the generated design and confirms the order if satisfied.

[2408] Specific processing of the AI ​​compatibility judgment function

[2409] Step 1:

[2410] Users upload a photo of themselves and an image of the clothing they plan to purchase to the site.

[2411] Step 2:

[2412] Terminal: Sends the uploaded image data to the server.

[2413] Step 3:

[2414] Server: Analyzes the image using a fit determination AI model. Specifically, it compares the user's photo with the image of the clothing and scores the degree of fit.

[2415] Step 4:

[2416] Server: Generates analyzed fitness scores and feedback.

[2417] Step 5:

[2418] Server: Sends the generated fitness score and feedback to the device.

[2419] Step 6:

[2420] Device: Presents fitness scores and feedback to the user.

[2421] Specific implementation of adjustments using the emotion engine

[2422] Step 1:

[2423] User: Provides input information and images when accessing the site, searching for products, interacting with the site, etc.

[2424] Step 2:

[2425] Terminal: Sends user input and images to the server.

[2426] Step 3:

[2427] Server: Using the emotion engine, recognizes emotions from user input information and images.

[2428] Step 4:

[2429] Server: Based on the recognized emotions, it adjusts recommendation results, conversational AI responses, and generative AI designs.

[2430] Step 5:

[2431] Server: Sends the adjustment results to the terminal.

[2432] Step 6:

[2433] Terminal: Display tailored information and suggestions to the user.

[2434] These concrete steps allow users to easily find the perfect product, easily order custom-made clothing, and receive personalized service based on their emotions, minimizing post-purchase dissatisfaction and waste.

[2435] Example 2

[2436] Next, a description will be given of Example 2. 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."

[2437] Conventional fashion e-commerce systems do not fully utilize a user's body type information or past purchase history, making it difficult to recommend optimal products to the user. Furthermore, personalized recommendations that take the user's emotional state into consideration are not made, resulting in a poor user experience. Furthermore, it is difficult to respond quickly and accurately to requests for custom-made items, which is a factor that reduces user satisfaction. The present invention aims to solve these problems.

[2438] 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.

[2439] In this invention, the server includes: means for collecting a user's body type information and past purchase history; means for training a machine learning model based on the collected information; means for recommending optimal products to the user using the trained machine learning model; means for analyzing the user's natural language input; means for searching for related product information based on the analysis results by the dialogue engine; means for recognizing the user's emotional state and adjusting suggestions based on the results; and means for generating custom designs based on the user's input information using generative AI technology. This makes it possible to comprehensively utilize the user's body type information, purchase history, and emotional state to propose optimal products and provide customized services to individual users.

[2440] 1. "Means of collection" refers to the equipment and software used to receive a user's body information and past purchase history and store it in a database.

[2441] 2. "Training means" refers to the process of using collected data to train a machine learning model to predict the best products for a user.

[2442] 3. "Recommendation method" refers to a system or algorithm that uses a trained machine learning model to suggest optimal products to users.

[2443] 4. "Dialogue engine" refers to software or a system that analyzes a user's natural language input and extracts relevant information based on that input.

[2444] 5. "Searching means" refers to the process of searching for appropriate product information from the database based on the results analyzed by the dialogue engine.

[2445] 6. "Emotion engine" refers to an algorithm that recognizes a user's emotional state from input information and images, and adjusts the system's responses and suggestions based on the results.

[2446] 7. "Generative AI technology" refers to artificial intelligence technology that automatically generates custom designs based on text information entered by the user.

[2447] 8. "Custom Design" refers to a unique clothing or product design created according to a user's specific requirements.

[2448] 9. "Means of analysis" refers to the process of analyzing uploaded images and input information to determine the suitability and suitability of products that meet the user's requirements.

[2449] 10. "Means of scoring" refers to the process of quantifying the suitability of a product based on the analysis results and evaluating its fit for the user.

[2450] This invention is a fashion e-commerce system that utilizes a user's body type information and past purchase history to recommend the most suitable fashion items using AI, and even provides custom-made clothing based on the user's requests. It also combines an emotion engine that recognizes the user's emotions to provide a more personalized service. This system consists of three main components: a server, a terminal, and a user.

[2451] server

[2452] The server first collects the user's body information (height, weight, waist size, etc.) and past purchase history, and stores this data in a database. This collection is done using a web framework such as Django. It then applies a trained machine learning model to recommend the best products for the user based on the collected data. Machine learning tools such as SciKit-Learn and TensorFlow are used for this training and prediction.

[2453] The server then houses a dialogue engine that analyzes the user's natural language input. This dialogue engine uses advanced natural language processing models such as GPT-3 and searches a database for relevant product information based on the analysis results. The server also houses an emotion engine that uses NLP libraries to recognize the user's emotional state from their input and images.

[2454] It also uses generative AI technology, such as DALL-E, to generate custom designs based on the user's specific requests.

[2455] Terminal

[2456] The device provides a front-end interface for users to access. Through a web browser or mobile app, users can input their body type information and purchase history and use the conversational AI search function to find products. It also has a chat window that displays the analysis results of the dialogue engine and tailored suggestions...

Claims

1. A means for collecting user's body type information and past purchase history; a means of training a machine learning model based on the collected information; and A means of using a trained machine learning model to recommend the best products to users; and A system including:

2. A means for analyzing requirements using an interactive AI engine based on a request from a user; A means for searching for related product information based on the analysis results; a means for displaying search results to a user; The system of claim 1 , comprising:

3. A means for generating a design using a generative AI based on user text input information; means for providing the generated design to a user; The system of claim 1 , comprising:

4. Using an AI model to analyze user photos and images of products to be purchased and determine their compatibility; means for presenting the determination result to a user; The system of claim 1 , comprising:

Images