System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting user information; means for inputting an image, a hashtag, or a text that a user likes; means for analyzing the user information, the image, the hashtag, and the text; means for selecting and recommending a personalized fashion item based on an analysis result; means for generating a specific recommendation reason for the recommended fashion item; means for proposing another item or a cosmetic that can be easily combined with the recommended fashion item; and means for displaying the recommended fashion item and the proposed another item or cosmetic to the user.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] When shopping online, users often feel anxious about choosing clothes that suit them, have difficulty imagining the right size, and young people who are not confident in their fashion sense have trouble choosing a style. Furthermore, since it is difficult to select the right fashion items, it is necessary to reduce the number of mistakes made when shopping online. [Means for solving the problem]

[0005] The present invention includes a means for collecting user information, a means for allowing users to input their favorite images, hashtags, and text, a means for analyzing this information and selecting and recommending personalized fashion items, a means for generating specific reasons for recommending the recommended fashion items, and a means for suggesting and displaying to the user other items and cosmetics that are easily combined with the fashion items.

[0006] This system allows users to confidently select fashion items that suit them, improving their online shopping experience. Furthermore, by providing recommendations based on the user's preferences, body type, personal color, past purchase history, and the latest trend information, it reduces anxiety about choosing a style and helps prevent mistakes when shopping online.

[0007] The "means for collecting user information" is a system component that has the function of collecting information such as user preferences, body type, personal color, past purchase history, and the latest trends.

[0008] "Means for users to input their favorite images, hashtags, and text" is a component of the system that provides an interface for users to specifically input their fashion preferences and style images.

[0009] The "means of analysis" refers to a component of the system that analyzes and interprets information collected from users, as well as input images, hashtags, and text, and performs processing to recommend fashion items that suit the user.

[0010] The "recommendation means" is a component of the system that has the function of selecting and suggesting fashion items suitable for the user based on the analysis results.

[0011] The "means for generating reasons for recommendation" is a system component that has the function of creating specific statements and reasons for why a recommended fashion item is suitable for the user.

[0012] The "means for suggesting other items and cosmetics" is a component of the system that has the function of suggesting to the user other fashion items and cosmetics that can be used in combination with the recommended fashion item.

[0013] The "display means" is a component of the system that provides an interface that visually displays information about recommended fashion items and other suggested items and cosmetics to the user.

[0014] A "means for logging in" is a system component that allows a user to access the system using their authentication information and initiate an individual session.

[0015] The "database means" is a component of a storage system for centrally storing and managing collected user information, analysis results, recommendation contents, etc.

[0016] The "means for selecting a product" is a system component that provides an interface for the user to select a specific item from the recommended fashion items. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] Natural language description of the program

[0039] The system is programmed using the following process:

[0040] 1. Collection of User Information

[0041] A session begins when a user logs in to an EC site.

[0042] The device displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[0043] The server stores this information in a database.

[0044] 2. Recommendation function

[0045] The device displays an image, hashtag, and text input field to the user, and the user enters their favorite image, hashtag, or text.

[0046] The server analyzes this input data and combines it with the user's profile information to interpret it. The analysis uses natural language processing technology, particularly the ChatGPT API, to make abstract images concrete.

[0047] The server selects fashion items that suit the user based on the analysis results.

[0048] The device displays a list of recommendations to the user, including a selection of items such as a pastel pink blouse, white denim, and a camera bag.

[0049] 3. Boost function

[0050] The user selects the product of interest from the recommendation list.

[0051] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[0052] The server will suggest other items or cosmetics that go well with the product (such as "white denim" or "light blue sneakers").

[0053] The device displays this information to the user, making it visually convincing.

[0054] Specific examples

[0055] Example 1: Recommendation function

[0056] A user enters "casual" and "likes pastel colors" in their profile.

[0057] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[0058] The server analyzes the input data and retrieves the appropriate items via the ChatGPT API.

[0059] The server selects "pastel pink blouse," "white denim," and "camera bag" as recommendations and displays them on the device.

[0060] Example 2: Boost function

[0061] The user selects "Pastel Pink Blouse" from the recommendations presented.

[0062] The server generates reasons for recommending this item, explaining, for example, "This pastel pink suits your personal color and is perfect for casual spring occasions."

[0063] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[0064] The terminal displays these coordination suggestions to the user and supports overall styling.

[0065] As a result, the present invention provides users with personalized fashion suggestions and a sense of security when shopping online.

[0066] The processing flow will be explained below.

[0067] Step 1:

[0068] A user logs in to an e-commerce site.

[0069] Step 2:

[0070] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters preferences, body type information, personal color, past purchase history, the latest trends, etc.

[0071] Step 3:

[0072] The server receives the user's input and stores it in a database.

[0073] Step 4:

[0074] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[0075] Step 5:

[0076] The server analyzes the user's input data, which involves applying natural language processing based on the user's profile information and sending a request to the ChatGPT API to materialize the abstract image.

[0077] Step 6:

[0078] The server analyzes the response from the ChatGPT API and selects the appropriate fashion items for the user.

[0079] Step 7:

[0080] The server generates a list of selected fashion items (e.g., "pastel pink blouse," "white denim," "camera bag," etc.) and stores it in a database.

[0081] Step 8:

[0082] The device displays the recommendation list to the user.

[0083] Step 9:

[0084] The user selects the product of interest from the recommendation list.

[0085] Step 10:

[0086] The server generates specific reasons for recommending the selected products, such as "This pastel pink is a perfect match for your personal color and also goes well with a casual spring vibe."

[0087] Step 11:

[0088] The server will suggest other items (e.g., "white denim" or "light blue sneakers") and cosmetics that go well with the selected product.

[0089] Step 12:

[0090] The terminal displays this information to the user, allowing them to understand it visually.

[0091] By following the above steps, users can confidently select fashion items that suit them, reducing the chances of making mistakes when shopping online.

[0092] Example 1

[0093] 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."

[0094] In conventional online shopping, it is difficult for users to find products that suit them. Especially when it comes to fashion items, recommendations do not sufficiently take into account the user's individual preferences, body type, personal color, etc. Furthermore, since there is no specific reason for the recommended product, users lack a sense of security when making a purchase decision. Furthermore, suggestions for coordinating the recommended product with other items are also insufficient. This leads to problems such as low user satisfaction and a decrease in purchasing motivation.

[0095] 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.

[0096] In this invention, the server includes means for collecting user information, means for users to input their favorite images, hashtags, and text, means for analyzing the user information and the images, hashtags, and text, means for selecting and recommending personalized products based on the analysis results, means for generating specific reasons for recommending the recommended products, means for suggesting other items that are easy to combine with the recommended products, and means for displaying the recommended products and other suggested items to the user. This allows the server to provide users with personalized fashion item recommendations and specific reasons for the recommendations, and can also provide coordination suggestions. This allows users to easily find fashion items that suit them and make purchasing decisions with confidence.

[0097] "User information" refers to information such as the user's individual preferences, body type, personal color, past purchase history, and the latest trends.

[0098] "Image" refers to an image that visually represents a style or design that a user likes.

[0099] A "hashtag" is a symbolic word that users use to identify keywords or themes of interest or concern.

[0100] "Text" refers to a data format in which a user inputs their opinions or wishes as text.

[0101] "Analysis" refers to the process of analyzing user preferences and trends based on user information, images, hashtags, and text.

[0102] "Personalized products" refer to products that are optimized to a user's individual preferences and needs based on analysis results.

[0103] "Recommendation" refers to the act of suggesting to a user that they purchase a particular product.

[0104] "Reasons for recommendation" refers to text that explains the specific appeal of the recommended product and its suitability for the user.

[0105] "Other Items" refers to additional products that can be used in conjunction with the recommended product to provide an enhanced experience.

[0106] "Display" refers to the act of visually providing information to a user via a terminal.

[0107] This invention is implemented by a system including the following components and processes: This system collects user information, recommends personalized products based on the user's preferences and needs, and also suggests specific reasons for recommending the products and related items.

[0108] Hardware and Software

[0109] This system mainly uses the following hardware and software:

[0110] Device: The device used by the user to access the site (e.g., computer, smartphone, tablet)

[0111] Server: A server system for collecting, storing, and analyzing data.

[0112] Database: A database system that stores user information and recommendation information

[0113] Generative AI models: APIs for using natural language processing techniques (e.g., ChatGPT API)

[0114] What the program does

[0115] The process is based on the following steps:

[0116] 1. User Information Collection:

[0117] A session begins when a user logs in to an EC site.

[0118] The device displays a login page and the user enters their existing account information.

[0119] The server authenticates the user's login information and begins the session.

[0120] 2. Displaying survey forms and profile entry pages:

[0121] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[0122] 3. Saving User Information:

[0123] The server stores the information entered by the user in a database in real time.

[0124] 4. Providing recommendation features:

[0125] The device displays an image, hashtags, and a text entry field to the user.

[0126] The user enters an image, hashtags, and text.

[0127] The server receives the input data and analyzes it using a generative AI model.

[0128] Based on the analysis results, the server selects personalized products and generates a recommendation list.

[0129] The device displays the recommendation list to the user.

[0130] 5. Providing boosting features:

[0131] The user selects the product of interest from the recommendation list.

[0132] The server generates specific reasons for recommending the selected product.

[0133] The server will suggest related items (e.g., "white denim" or "light blue sneakers").

[0134] The device will display the recommendation reason and related items to the user.

[0135] Specific examples

[0136] Here are some examples and prompts:

[0137] Example 1: Recommendation function

[0138] A user enters "casual" and "likes pastel colors" in their profile.

[0139] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[0140] The server analyzes the input data and obtains suitable items via a generative AI model.

[0141] The server selects "pastel pink blouse," "white denim," and "camera bag" from the recommendation list and displays them on the device.

[0142] Example 2: Boost function

[0143] The user selects "Pastel Pink Blouse" from the recommendations presented.

[0144] The server generates reasons for recommending this item, explaining, for example, "This pastel pink suits your personal color and is perfect for casual spring occasions."

[0145] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[0146] The terminal displays these coordination suggestions to the user and supports overall styling.

[0147] Example prompt sentence:

[0148] "What casual fashion items would be perfect for a spring picnic? Our users love pastel colors and casual styles."

[0149] As a result, the system of the present invention enables personalized recommendations of fashion items to users, along with specific reasons for the recommendations and coordination suggestions, thereby improving the sense of security and satisfaction of online shopping.

[0150] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0151] Step 1: User logs in to the e-commerce site

[0152] The user accesses the EC site and logs in with an existing account. Authentication is performed using the ID and password entered by the user.

[0153] The server checks the user's credentials against a database and starts a session. The input is the user's credentials and the output is a session ID.

[0154] Step 2: Displaying the survey form and profile entry page

[0155] The terminal displays a questionnaire form or profile entry page to the logged-in user.

[0156] The user enters information into the form, such as their preferences, body type, personal color, past purchase history, the latest trends, etc. The input is the user's personal information, and the output is the user's response data.

[0157] Step 3: Save user information

[0158] The server receives the information entered by the user and stores it in a database.

[0159] This data is used for subsequent analysis and recommendations. The input is the user's personal information data, and the output is a user profile stored in a database.

[0160] Step 4: Displaying an image, hashtags, and text input fields

[0161] The device displays a recommendation page, providing the user with an image, hashtags, and a text input field.

[0162] The user enters their desired image, hashtag, or text in the fields provided. The input is the user's input data (image, hashtag, text), and the output is the input data sent to the server.

[0163] Step 5: Collecting User Input Data

[0164] The server receives the images, hashtags, and text sent by the user and stores them in a database. The input is the user's input data, and the output is the data stored in the database for analysis.

[0165] Step 6: Data analysis and item selection

[0166] The server retrieves the user's profile information and newly entered data from the database and analyzes it using a generative AI model. Specifically, it uses the ChatGPT API to perform natural language processing and abstractly understand the user's preferences.

[0167] As a result of the analysis, the system selects the most suitable product for the user. The input is the profile data and the user's input data, and the output is a list of recommended products.

[0168] Step 7: Generate and display the recommendation list

[0169] Based on the analysis results, the server lists the most suitable products for the user.

[0170] The terminal visually displays this recommendation list to the user. The input is a list of recommended products, and the output is the recommendation list displayed to the user.

[0171] Step 8: Product Selection and Recommendation Generation

[0172] The user selects the product of interest from the recommendation list.

[0173] The server uses a generative AI model to generate specific reasons for recommending the selected product. For example, it generates an explanation such as, "This pastel pink blouse is perfect for your personal color and goes well with a casual spring look." The input is the selected product information, and the output is the reason for the recommendation.

[0174] Step 9: Suggest other items

[0175] The server will suggest other items that go well with the selected item (e.g., "white denim" or "light blue sneakers")

[0176] The server selects related items and sends them to the terminal along with a recommendation list. The input is the selected product information, and the output is a list of suggested items.

[0177] Step 10: View the proposal

[0178] The device displays the recommended product, the reason for the recommendation, and related items to the user.

[0179] The user reviews the overall styling and makes a purchasing decision. The input is data from the server and the output is the information displayed to the user.

[0180] This allows the system to suggest personalized fashion items to users, improving their online shopping experience.

[0181] (Application example 1)

[0182] 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."

[0183] The user experience in online shopping lacks visual information compared to shopping in a physical store, making it difficult to provide personalized product suggestions and encourage purchases. Finding products that match a user's preferences, body type, and personal color is particularly difficult when it comes to fashion items, which can cause users to hesitate before making a purchase. Therefore, there is a demand for systems that provide a more interactive and personalized shopping experience.

[0184] 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.

[0185] In this invention, the server includes means for collecting user information, means for inputting user-preferred images, hashtags, and text, means for analyzing the user information and the images, hashtags, and text, means for selecting and recommending personalized fashion items based on the analysis results, means for generating specific recommendation reasons for the recommended fashion items, means for suggesting other items and cosmetics that are easily combined with the fashion items, means for displaying the recommended fashion items and the suggested other items and cosmetics to the user, and means for visually displaying the recommended fashion items using smart glasses, thereby enabling users to enjoy a more interactive and personalized shopping experience and select and purchase products based on visual information.

[0186] "User information" refers to information such as the user's preferences, body type, personal color, past purchase history, and the latest trends.

[0187] "Images" are images or visual content that a user visually enjoys.

[0188] A "hashtag" is a symbol used to promote information sharing among people with the same hobbies or interests by tagging specific keywords or phrases with "."

[0189] "Text" refers to sentences or character information entered by the user.

[0190] "Analysis Methods" refers to algorithms and technologies used to process user information, images, hashtags, and text to understand and analyze their content.

[0191] "Personalized fashion items" refer to clothing and accessories that are identified and suggested to a user based on their individual information and preferences.

[0192] A "recommendation means" is a system or method that recommends specific fashion items to users based on the analysis results.

[0193] "Specific reasons for recommendation" refers to a detailed explanation of the recommended fashion item and the reasons for its recommendation.

[0194] "Means for suggesting other items or cosmetics" refers to a method or system for presenting other clothing or cosmetics that go well with the recommended fashion item to the user.

[0195] "Visual display means" refers to a method of visually presenting personalized products or offers to a user using a device such as smart glasses.

[0196] "Generative AI model" refers to artificial intelligence technology for generating and analyzing information based on user input and prompts.

[0197] The present invention provides a system for providing personalized fashion item recommendations to users using smart glasses. A specific implementation method thereof will be described below.

[0198] Program processing flow

[0199] 1. Collection of User Information

[0200] When a user logs in to the e-commerce site, the server begins collecting user information. The terminal displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends. This information is sent to the server and stored in a database.

[0201] 2. Implementation of recommendation function

[0202] When a user logs in, the device displays an image, hashtags, and text input field. The user then enters their desired image, hashtag, or text. The server analyzes this input data and interprets it in combination with the user's profile information. The analysis uses natural language processing technology, particularly the ChatGPT API, a generative AI model, to make abstract images concrete.

[0203] 3. Selection of recommended items

[0204] Based on the analysis results, the server selects fashion items that suit the user. Specifically, it also generates personalized reasons for the recommendation. The device then displays a recommendation list to the user. The recommendation list includes the selected items, such as a pastel pink blouse, white denim, and a camera bag.

[0205] 4. Visual display using smart glasses

[0206] The recommended fashion items are displayed visually to the user using the smart glasses, allowing the user to view product details and facilitate purchasing decisions.

[0207] Specific examples

[0208] Example 1: Recommendation function

[0209] A user enters "casual" and "likes pastel colors" in their profile. They then enter "a perfect outfit for a spring picnic" in the text field and add the hashtag "spring casual." The server analyzes this input data and retrieves suitable items via the ChatGPT API of the generative AI model. As a result, "pastel pink blouse," "white denim," and "camera bag" are displayed in the recommendation list.

[0210] Example 2: Boost function

[0211] The user selects a "pastel pink blouse" from the recommendations presented. The server generates a reason for recommending this item, explaining that "this pastel pink suits your personal color and is perfect for spring casual occasions." Furthermore, other items that go well with the blouse, such as "white denim" and "light blue sneakers," are also suggested. The user can visually check these coordination suggestions through the smart glasses, receiving support for overall styling.

[0212] Specific hardware and software used

[0213] Hardware: Smart glasses, user devices (smartphones and PCs)

[0214] Software: E-commerce website platform, generative AI model (ChatGPT API), database

[0215] Examples of prompt statements

[0216] User profile: {'Preferences': 'Casual', 'Favorite color': 'Pastel'}

[0217] User Input: Perfect outfit for a spring picnic Spring Casual

[0218] Recommended items:

[0219] In this way, the system can provide users with a more personalized and visually appealing shopping experience.

[0220] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0221] Step 1:

[0222] Collection of User Information

[0223] Users log in to the e-commerce site using a terminal. The terminal displays a questionnaire form and a profile entry page, where users enter their preferences, body type, personal color, past purchase history, the latest trends, etc. This information is sent to the server and stored in a database.

[0224] Input: User login information, survey responses, profile information

[0225] Output: Saved user information (database)

[0226] Step 2:

[0227] User image, hashtag and text input

[0228] The device displays an image, hashtags, and text input fields to the user, who then enters the image, hashtag, or text of their choice, which is then sent to the server.

[0229] Input: Images, hashtags, and text entered by the user

[0230] Output: Data entered by the user

[0231] Step 3:

[0232] Data analysis

[0233] The server analyzes the collected user information and the input image, hashtag, and text. It uses the ChatGPT API, a generative AI model, to generate prompts. Specifically, it converts abstract images into concrete fashion items based on the user's profile information and text input.

[0234] Input: User information, image, hashtag, text, prompt

[0235] Output: A list of fashion items as the analysis result

[0236] Step 4:

[0237] Personalized fashion item selection and recommendations

[0238] Based on the analysis results, the server selects and recommends fashion items that suit the user. Specific reasons for recommending the items are also provided. The list of selected items is then sent to the device.

[0239] Input: Analysis results, user information, prompt text

[0240] Output: A list of selected fashion items and reasons for their recommendation

[0241] Step 5:

[0242] Preparing for visual display

[0243] The device receives the list of recommended fashion items and the reasons for the recommendations, and prepares them for display on the smart glasses.

[0244] Input: List of selected fashion items and reasons for recommendation

[0245] Output: Data ready for display

[0246] Step 6:

[0247] Visual display with smart glasses

[0248] By using the smart glasses, users can visually see the recommended fashion items and the reasons for their recommendations, allowing them to view product details and make purchasing decisions more easily.

[0249] Input: Data ready to be displayed

[0250] Output: Visual product display via smart glasses

[0251] 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.

[0252] Natural language description of the program

[0253] The system is programmed using the following process:

[0254] 1. Collection of User Information

[0255] A session begins when a user logs in to an EC site.

[0256] The device displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[0257] The server stores this information in a database.

[0258] 2. Recommendation function

[0259] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[0260] The server analyzes these inputs, which involves applying natural language processing based on the user's profile information and sending requests to the ChatGPT API to turn the abstract image into a concrete image.

[0261] Based on the analysis results, the server selects fashion items suitable for the user.

[0262] The device displays a list of recommendations to the user, including a selection of items such as a pastel pink blouse, white denim, and a camera bag.

[0263] 3. Boost function

[0264] The user selects the product of interest from the recommendation list.

[0265] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[0266] The server will suggest other items or cosmetics that go well with the product (such as "white denim" or "light blue sneakers").

[0267] The device displays this information to the user, making it visually convincing.

[0268] 4. Utilizing the Emotion Engine

[0269] The server analyzes the user's input information and selection behavior, and recognizes the user's emotions using an emotion engine.

[0270] The server dynamically adjusts the recommendation content based on the user's emotions. For example, if the user is unsure about which product to choose, the server may change the emphasis on the recommended item or suggest additional similar products.

[0271] The server analyzes the user's reactions in real time and adjusts the reason for the recommendation accordingly. For example, if the user is interested but anxious, the reason will be changed to emphasize positive feedback.

[0272] Specific examples

[0273] Example 1: Combining recommendation features with an emotion engine

[0274] A user enters "casual" and "likes pastel colors" in their profile.

[0275] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[0276] The server analyzes the input data and retrieves the appropriate items via the ChatGPT API.

[0277] The server selects "pastel pink blouse," "white denim," and "camera bag" as recommendations and displays them on the device.

[0278] The server analyzes the user's reactions in real time and recognizes that the user is interested in the recommended item but is also feeling anxious.

[0279] The server adjusts the recommendation statement to read, "This pastel pink suits your personal color and is perfect for casual spring occasions. Many users also rate this item highly."

[0280] Example 2: Combining boosting and emotion engines

[0281] The user selects "Pastel Pink Blouse" from the presented recommendation list.

[0282] The server generates reasons for recommending this item, explaining, for example, "This pastel pink is a perfect match for your personal color and also goes well with a casual spring vibe."

[0283] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[0284] The server analyzes user responses and checks to see which ones are more positive.

[0285] The server displays the original reasons for the recommendation and adds suggestions to encourage additional purchases (e.g., "This outfit is eligible for a 30% off campaign!").

[0286] The device displays this information to the user and supports overall styling.

[0287] As a result, the present invention not only provides users with personalized fashion suggestions and a sense of security when shopping online, but also uses an emotion engine to achieve even more accurate recommendations and support.

[0288] The processing flow will be explained below.

[0289] Step 1:

[0290] A user logs in to an e-commerce site and a session begins.

[0291] Step 2:

[0292] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters their preferences, body type information, personal color, past purchase history, the latest trend information, etc.

[0293] Step 3:

[0294] The server receives the user's input and stores it in a database.

[0295] Step 4:

[0296] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[0297] Step 5:

[0298] The server analyzes the user's input data, which involves applying natural language processing based on the user's profile information and sending a request to the ChatGPT API to materialize the abstract image.

[0299] Step 6:

[0300] The server analyzes the response from the ChatGPT API and selects the appropriate fashion items for the user.

[0301] Step 7:

[0302] The server generates a list of selected fashion items and stores it in a database.

[0303] Step 8:

[0304] The device displays the recommendation list to the user.

[0305] Step 9:

[0306] The user selects the product of interest from the recommendation list.

[0307] Step 10:

[0308] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[0309] Step 11:

[0310] The server will suggest other items (e.g., "white denim" or "light blue sneakers") and cosmetics that go well with the selected product.

[0311] Step 12:

[0312] The terminal displays this information to the user, allowing them to understand it visually.

[0313] Step 13:

[0314] The server analyzes the user's input information and selection behavior, and recognizes the user's emotions using an emotion engine.

[0315] Step 14:

[0316] The server dynamically adjusts the recommendation content based on the user's emotions. For example, if the user is unsure about which product to choose, the server may change the emphasis on the recommended item or suggest additional similar products.

[0317] Step 15:

[0318] The server analyzes the user's reactions in real time and adjusts the reason for the recommendation accordingly. For example, if the user is interested but anxious, the reason will be changed to emphasize positive feedback.

[0319] Through the above steps, users can confidently choose fashion items that suit them, and by using the emotion engine, even more accurate recommendations and support can be achieved.

[0320] Example 2

[0321] 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."

[0322] Conventional online fashion recommendation systems do not fully consider users' subjective preferences and emotions when making recommendations, making it difficult to provide products that truly satisfy users. Furthermore, conventional systems cannot respond to changes in users' emotions, and therefore do not sufficiently motivate users to purchase additional products that they are temporarily interested in.

[0323] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting user information; means for inputting a user's favorite images, hashtags, and text; means for analyzing the user information and the images, hashtags, and text; means for selecting and recommending personalized fashion items based on the analysis results; means for generating specific recommendation reasons for the recommended fashion items; means for suggesting other items and cosmetics that are easily combined with the fashion items; means for displaying the recommended fashion items and the suggested other items and cosmetics to the user; means for analyzing the user's input information and behavior and recognizing emotions; means for dynamically adjusting the recommendation content based on the recognized user emotions; and means for analyzing the user's reaction to the recommended items in real time and dynamically adjusting the recommendation reasons in accordance with the analysis results. This enables personalized recommendations of fashion items based on the user's preferences and emotions, and dynamic adjustment of recommendations in accordance with the user's emotions.

[0324] "Means for collecting user information" refers to the interface and data collection process for collecting information such as the user's preferences, body type, personal color, past purchase history, and the latest trends.

[0325] A "means for users to input their favorite images, hashtags, and text" is an interface that provides forms or input fields for users to enter information about their preferences and style in visual or textual form.

[0326] The "means for analyzing the user information and the images, hashtags, and text" refers to a system that analyzes collected user information and input data using natural language processing and image analysis algorithms.

[0327] The "means for selecting and recommending personalized fashion items based on analysis results" is a system that selects fashion items suitable for the user from the analysis results and provides recommendation information on them.

[0328] The "means for generating a specific reason for recommendation for the recommended fashion item" is a system that generates a specific reason for recommendation related to the selected fashion item and provides it to the user.

[0329] The "means for suggesting other items and cosmetics that can be easily combined with the fashion item" is a system that suggests to the user other related items and cosmetics that can be easily combined with the recommended fashion item.

[0330] The "means for displaying the recommended fashion items and other suggested items and cosmetics to the user" is an interface for visually displaying the selected fashion items and other suggested related items and cosmetics to the user.

[0331] The "means for analyzing the user's input information and behavior and recognizing emotions" is an emotion analysis system that analyzes the user's input data and behavior history and estimates the user's emotional state.

[0332] The "means for dynamically adjusting recommendation content based on recognized user emotions" is a system that adjusts recommendation content in real time based on analyzed emotional information.

[0333] "Means for analyzing user reactions to the recommended item in real time and dynamically adjusting the reason for recommendation based on the analysis results" refers to a system that collects and analyzes user reaction data in real time and dynamically changes the reason for recommendation based on the results.

[0334] A "generative AI model" is an artificial intelligence model that generates natural language and recognizes data patterns based on user information and behavioral data.

[0335] A "prompt" is an instruction or question that is input to a generative AI model to output recommendations.

[0336] System Overview

[0337] This invention is a system that recommends personalized fashion items based on a user's preferences and emotional state. The system is realized by combining four main functions: user information collection, recommendation function, boost function, and emotion engine.

[0338] Hardware and software used

[0339] Server: Used to analyze data and select recommended items. The main software includes Python natural language processing libraries (NLTK and spaCy), sentiment analysis libraries (TextBlob and VADER), and generative AI models (ChatGPT API).

[0340] Device: This serves as the interface where users can input information and view recommended items and descriptions. This is typically a web browser or a mobile app.

[0341] Database: Used to store user information. Popular databases include AWS RDS (Relational Database Service) and MySQL.

[0342] Processing Details

[0343] 1. Collection of User Information

[0344] When a user logs in to an e-commerce site, the device displays a questionnaire form and a profile entry page. The user enters information such as their preferred fashion style, body type, personal color, past purchase history, and the latest trends. This information is sent to the server and stored in a database such as AWS RDS or MySQL.

[0345] 2. Recommendation function

[0346] The device displays an image, hashtags, and a text input field to the user. The user uploads an image of their choice and enters text and hashtags. The server analyzes the received data using Python's natural language processing library (NLTK or spaCy) and generates a prompt for the ChatGPT API based on the analysis results. Based on the response from the ChatGPT API, the server selects appropriate fashion items and displays them on the device.

[0347] 3. Boost function

[0348] When a user selects a product from the recommendation list, the server generates specific reasons for recommending that product. For example, a reason might be provided such as, "This pastel pink blouse matches your personal color perfectly." Other related items (such as "white denim" or "light blue sneakers") are also suggested. This information is displayed on the device, allowing the user to visually understand the recommendation.

[0349] 4. Utilizing the Emotion Engine

[0350] The server analyzes the user's input information and behavior using emotion analysis libraries (TextBlob and VADER) to recognize the user's emotions. If the user is unsure about which product to choose, the server dynamically adjusts the emphasis of the recommended items. For example, it adds information about limited-time sales or special offers. The user's reactions are analyzed in real time, and if the user shows anxiety, for example, it changes the reason statement to emphasize positive feedback.

[0351] Specific examples

[0352] Combining recommendation functionality with an emotion engine

[0353] A user enters "casual" and "I like pastel colors" in their profile. They then write "A perfect outfit for a spring picnic" in the text field and add the hashtag "spring casual." The server analyzes the input data and selects "pastel pink blouse," "white denim," and "camera bag" as recommendations, which are displayed on the device. The server analyzes the user's reactions in real time, and if the user is interested in a recommended item but has concerns, it dynamically adjusts the reason for the recommendation to something like, "This pastel pink suits your personal color and is perfect for spring casual occasions."

[0354] Prompt Sentence Examples

[0355] "Suggest a casual outfit perfect for a spring picnic. Your user likes pastel pink and prefers casual styles."

[0356] This invention makes it possible to provide personalized fashion suggestions based on a user's preferences and emotions, improving the online shopping experience.

[0357] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0358] Step 1:

[0359] A user logs in to an e-commerce site.

[0360] Input: User ID and password

[0361] Operation: The terminal displays the login screen. The user enters their ID and password and presses the "Login" button.

[0362] Output: A session is started and the user's profile information is loaded.

[0363] Step 2:

[0364] The device will display a survey form and profile entry page.

[0365] Input: Display requirements for survey forms and profile entry pages

[0366] How it works: The device generates an input form with fields such as "preferred fashion style," "body type information," and "personal color."

[0367] Output: The form that is displayed to the user

[0368] Step 3:

[0369] The user enters information into a survey form or profile entry page and submits it.

[0370] Input: Information such as "preferred fashion style," "body type information," "personal color," "past purchase history," and "latest trends"

[0371] What happens: The user enters information into the fields and presses the "Submit" button.

[0372] Output: The entered user information is sent to the server.

[0373] Step 4:

[0374] The server stores the user information in a database.

[0375] Input: Information entered by the user

[0376] How it works: The server receives the data and stores it in a database such as AWS RDS or MySQL.

[0377] Output: User information stored in the database

[0378] Step 5:

[0379] The device displays an image, hashtags, and a text entry field to the user.

[0380] Input: Display requirements for images, hashtags, and text input fields

[0381] How it works: Your device will display options such as "Upload your favorite image," "Related hashtags," and "Describe your desired style."

[0382] Output: Input fields that are displayed to the user

[0383] Step 6:

[0384] The user enters an image, text, and hashtags.

[0385] Input: User-uploaded images, entered text, hashtags

[0386] How it works: User uploads an image, enters text and hashtags.

[0387] Output: The entered data is sent to the server.

[0388] Step 7:

[0389] The server parses the input data.

[0390] Input: Image, Text, Hashtag

[0391] How it works: The server parses the data using Python natural language processing libraries (NLTK or spaCy) and generates prompts to the ChatGPT API based on the results of the analysis.

[0392] Output: Generated prompt, response data from ChatGPT API

[0393] Step 8:

[0394] The server selects recommended items based on the analysis results.

[0395] Input: Response data from ChatGPT API, user profile information

[0396] Operation: The server uses the response data to create a list of fashion items suitable for the user.

[0397] Output: A list of selected recommended items

[0398] Step 9:

[0399] The device displays a recommendation list to the user.

[0400] Input: Recommended item list

[0401] What it does: Your device will display a list of items such as "Pastel Pink Blouse," "White Denim," and "Camera Bag."

[0402] Output: The recommendation list displayed to the user

[0403] Step 10:

[0404] The user selects the product of interest from the recommendation list.

[0405] Input: Products you are interested in from the recommendation list

[0406] Action: The user clicks on the selected item.

[0407] Output: Selected product information is sent to the server.

[0408] Step 11:

[0409] The server generates specific recommendation reasons.

[0410] Input: Selected product information, user profile information

[0411] How it works: The server generates a recommendation reason based on the user's profile and product attributes, for example, "This pastel pink blouse matches your personal color."

[0412] Output: Generated recommendation reason text

[0413] Step 12:

[0414] The server suggests other related items.

[0415] Input: Selected product information

[0416] What it does: The server lists items that go well with the selected item, such as "white denim" or "light blue sneakers."

[0417] Output: A list of suggested related items

[0418] Step 13:

[0419] The device will display the recommendation reason and related items to the user.

[0420] Input: Generated recommendation reason text, suggested related item list

[0421] What it does: The device displays this information to the user in a visually convincing way.

[0422] Output: Displayed recommendation reason and related items

[0423] Step 14:

[0424] The server analyzes the user's input information and behavior.

[0425] Input: User input data and behavior history

[0426] How it works: The server analyzes the data using a sentiment analysis library (TextBlob or VADER) to recognize the user's sentiment.

[0427] Output: Recognized emotion data

[0428] Step 15:

[0429] The server dynamically adjusts the recommendation content based on emotions.

[0430] Input: Recognized emotion data, recommended item list

[0431] How it works: If the server detects negative sentiment, it changes the emphasis of the recommended items and adds limited-time sales and special offers.

[0432] Output: Dynamically adjusted recommendations

[0433] Step 16:

[0434] The server analyzes user responses in real time and adjusts the wording of the recommendation reasons.

[0435] Input: User response data (viewing time, number of clicks, etc.)

[0436] How it works: The server analyzes the user's response and modifies the generated reason statement based on the results. For example, if the user is feeling anxious, it emphasizes more positive feedback.

[0437] Output: Adjusted recommendation reason

[0438] (Application example 2)

[0439] 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."

[0440] While conventional fashion recommendation systems suggest personalized items to users when shopping online, they do not address the in-store shopping experience. Furthermore, they do not provide recommendations that take into account the user's real-time movements and emotions, which prevents them from fully enhancing user satisfaction. Furthermore, there is a need for a system that effectively utilizes devices such as eye-tracking technology, smart glasses, and head-mounted displays to enable users to obtain the information they want in-store in real time.

[0441] 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.

[0442] In this invention, the server includes an eye tracking unit that identifies a specific product when the user moves around the store and directs their gaze at the product, a unit that dynamically adjusts recommendations for the identified product according to the user's emotional state, and a unit that displays the generated recommendation reasons and recommendation content on smart glasses or a head-mounted display, thereby improving the shopping experience in a physical store and making it possible to provide appropriate information and recommendations in real time for products that interest the user.

[0443] "Means for collecting user information" refers to a function for inputting and recording user preferences, body type, personal color, past purchase history, the latest trends, and the like.

[0444] The "means for users to input their favorite images, hashtags, and text" refers to an interface that allows users to input images, hashtags, text, etc. that specifically indicate their own fashion style and preferences.

[0445] "Means for analyzing user information and the images, hashtags, and text" refers to a function for analyzing collected user information and input images, hashtags, and text to understand user preferences and trends.

[0446] "Means for selecting and recommending personalized fashion items" refers to a system that selects and suggests the most suitable fashion items for users based on the analysis results.

[0447] The "means for generating specific reasons for recommendation" is a function for automatically generating detailed explanations and reasons why recommended fashion items are preferred.

[0448] The "means for suggesting other items and cosmetics" is a function for automatically suggesting other fashion items and cosmetics that can be easily combined with the recommended fashion items.

[0449] The "means for displaying to the user" is an interface for visually presenting information about the recommended fashion items and other suggested items and cosmetics to the user.

[0450] The "eye tracking means" is a technology for identifying and tracking a particular product when the user directs their gaze at the product.

[0451] "Dynamic adjustment means" is a function that allows recommendations to be changed and adjusted in real time according to the user's emotions and behavior.

[0452] "Means for displaying on smart glasses or a head-mounted display" refers to a function for displaying the generated recommendation reasons and recommendation content on a smart device in real time.

[0453] Specific configurations and methods for implementing the present invention are described below.

[0454] 1. Collection of User Information

[0455] The terminal uses smart glasses and mobile devices to collect user information. Users use these devices to input their preferred style, body type, personal color, past purchase history, and the latest trend information. This information is sent to a server and stored in a database.

[0456] 2. Eye tracking and product identification

[0457] The smart glasses use built-in eye-tracking technology (e.g., Tobii Eye Tracker SDK) to identify the products the user is looking at in the store, recognize the products the user is looking at, and send the related information to a server.

[0458] 3. Recommendation Generation and Sentiment Analysis

[0459] The server generates personalized recommendations based on the aforementioned user information and eye-tracking information. This includes suggesting fashion items based on the user's input and profile using a generative AI model (e.g., OpenAI API). It also uses an emotion analysis engine to analyze the user's emotional state from their facial expressions and behavior and dynamically adjust the recommendations.

[0460] 4. Displaying recommendation reasons and dynamic adjustment

[0461] The device displays the recommendation content and specific reasons for the recommendation in real time on smart glasses or a head-mounted display. Based on the results of the sentiment analysis engine, the device adjusts the recommendation reasons and the display of recommended items. For example, if the user is interested but anxious, it will emphasize positive feedback.

[0462] Specific examples

[0463] For example, if a user wears smart glasses and enters a store and looks at a particular product, fashion items related to that product will be automatically recommended. The recommendation will be displayed on the visual device along with the reason for the recommendation generated by the server based on the user's information. This allows users to enjoy shopping while receiving detailed information and personalized recommendations.

[0464] Prompt Sentence Examples

[0465] An example of a prompt generated by the server is:

[0466] markdown

[0467] User Profile:

[0468] Favorite style: Casual, pastel colors

[0469] Body type: Medium build

[0470] Past purchases: Jeans, blouses, casual shoes

[0471] User input:

[0472] Tags: Spring Casual

[0473] Keywords: Spring Picnic

[0474] Recommended Items:

[0475] pastel pink blouse

[0476] White denim

[0477] camera bag

[0478] Recommended reasons include:

[0479] This pastel pink blouse is perfect for your personal color and is perfect for casual spring occasions!

[0480] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0481] Step 1:

[0482] The user logs in using smart glasses or a mobile device. After logging in, the user enters their preferred style, body type, personal color, past purchase history, latest trend information, etc. This information is sent to the server and stored in a database.

[0483] Input: User profile information (style, body type, color, purchase history, trend information)

[0484] Output: User information stored in the database

[0485] Step 2:

[0486] As a user moves around the store, the smart glasses' eye-tracking technology (Tobii Eye Tracker SDK) is used to identify the products they are looking at. The glasses then transmit the user's gaze data to a server.

[0487] Input: User gaze data

[0488] Output: Identified product information

[0489] Step 3:

[0490] The server analyzes the user's gaze data and saved profile information to generate personalized recommendations for the user. A generative AI model (OpenAI API) is used to suggest fashion items based on the user's prompts.

[0491] Input: Identified product information, user profile information

[0492] Output: A list of recommended fashion items

[0493] Step 4:

[0494] The server inputs the user's prompt information into the generative AI model to generate specific recommendation reasons, including specific reasons and explanations for the new recommendation.

[0495] Input: Identified product information, user profile information, prompt information

[0496] Output: Specific reasons for recommendation

[0497] Step 5:

[0498] The server uses an emotion analysis engine to analyze the user's emotional state based on their facial expressions and behavior. Based on the user's emotions, the recommendation content is dynamically adjusted. For example, if the user is interested but anxious, the server emphasizes positive feedback.

[0499] Input: gaze data, facial expression data, behavior data

[0500] Output: Adjusted recommendation content

[0501] Step 6:

[0502] The device (smart glasses or head-mounted display) displays the generated recommendation content and specific reasons for the recommendation in real time. The user can consider products based on this information and obtain additional information as needed.

[0503] Input: Adjusted recommendation content, specific reasons for recommendation

[0504] Output: Present information to the user

[0505] Step 7:

[0506] The user selects recommended fashion items or other suggested items through their device. Detailed information about the selected item and related items are displayed. The device then sends the information back to the server, which generates additional recommendations as needed.

[0507] Input: User selection information

[0508] Output: Show detailed information and related items

[0509] Through these steps, users can enjoy a personalized shopping experience in a physical store.

[0510] 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.

[0511] 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.

[0512] 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.

[0513] [Second embodiment]

[0514] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0515] 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.

[0516] 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).

[0517] 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.

[0518] 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.

[0519] 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).

[0520] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0521] 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.

[0522] 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.

[0523] 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.

[0524] In the smart glasses 214, 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.

[0525] 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."

[0526] Natural language description of the program

[0527] The system is programmed using the following process:

[0528] 1. Collection of User Information

[0529] A session begins when a user logs in to an EC site.

[0530] The device displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[0531] The server stores this information in a database.

[0532] 2. Recommendation function

[0533] The device displays an image, hashtag, and text input field to the user, and the user enters their favorite image, hashtag, or text.

[0534] The server analyzes this input data and combines it with the user's profile information to interpret it. The analysis uses natural language processing technology, particularly the ChatGPT API, to make abstract images concrete.

[0535] The server selects fashion items that suit the user based on the analysis results.

[0536] The device displays a list of recommendations to the user, including a selection of items such as a pastel pink blouse, white denim, and a camera bag.

[0537] 3. Boost function

[0538] The user selects the product of interest from the recommendation list.

[0539] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[0540] The server will suggest other items or cosmetics that go well with the product (such as "white denim" or "light blue sneakers").

[0541] The device displays this information to the user, making it visually convincing.

[0542] Specific examples

[0543] Example 1: Recommendation function

[0544] A user enters "casual" and "likes pastel colors" in their profile.

[0545] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[0546] The server analyzes the input data and retrieves the appropriate items via the ChatGPT API.

[0547] The server selects "pastel pink blouse," "white denim," and "camera bag" as recommendations and displays them on the device.

[0548] Example 2: Boost function

[0549] The user selects "Pastel Pink Blouse" from the recommendations presented.

[0550] The server generates reasons for recommending this item, explaining, for example, "This pastel pink suits your personal color and is perfect for casual spring occasions."

[0551] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[0552] The terminal displays these coordination suggestions to the user and supports overall styling.

[0553] As a result, the present invention provides users with personalized fashion suggestions and a sense of security when shopping online.

[0554] The processing flow will be explained below.

[0555] Step 1:

[0556] A user logs in to an e-commerce site.

[0557] Step 2:

[0558] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters preferences, body type information, personal color, past purchase history, the latest trends, etc.

[0559] Step 3:

[0560] The server receives the user's input and stores it in a database.

[0561] Step 4:

[0562] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[0563] Step 5:

[0564] The server analyzes the user's input data, which involves applying natural language processing based on the user's profile information and sending a request to the ChatGPT API to materialize the abstract image.

[0565] Step 6:

[0566] The server analyzes the response from the ChatGPT API and selects the appropriate fashion items for the user.

[0567] Step 7:

[0568] The server generates a list of selected fashion items (e.g., "pastel pink blouse," "white denim," "camera bag," etc.) and stores it in a database.

[0569] Step 8:

[0570] The device displays the recommendation list to the user.

[0571] Step 9:

[0572] The user selects the product of interest from the recommendation list.

[0573] Step 10:

[0574] The server generates specific reasons for recommending the selected products, such as "This pastel pink is a perfect match for your personal color and also goes well with a casual spring vibe."

[0575] Step 11:

[0576] The server will suggest other items (e.g., "white denim" or "light blue sneakers") and cosmetics that go well with the selected product.

[0577] Step 12:

[0578] The terminal displays this information to the user, allowing them to understand it visually.

[0579] By following the above steps, users can confidently select fashion items that suit them, reducing the chances of making mistakes when shopping online.

[0580] Example 1

[0581] 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."

[0582] In conventional online shopping, it is difficult for users to find products that suit them. Especially when it comes to fashion items, recommendations do not sufficiently take into account the user's individual preferences, body type, personal color, etc. Furthermore, since there is no specific reason for the recommended product, users lack a sense of security when making a purchase decision. Furthermore, suggestions for coordinating the recommended product with other items are also insufficient. This leads to problems such as low user satisfaction and a decrease in purchasing motivation.

[0583] 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.

[0584] In this invention, the server includes means for collecting user information, means for users to input their favorite images, hashtags, and text, means for analyzing the user information and the images, hashtags, and text, means for selecting and recommending personalized products based on the analysis results, means for generating specific reasons for recommending the recommended products, means for suggesting other items that are easy to combine with the recommended products, and means for displaying the recommended products and other suggested items to the user. This allows the server to provide users with personalized fashion item recommendations and specific reasons for the recommendations, and can also provide coordination suggestions. This allows users to easily find fashion items that suit them and make purchasing decisions with confidence.

[0585] "User information" refers to information such as the user's individual preferences, body type, personal color, past purchase history, and the latest trends.

[0586] "Image" refers to an image that visually represents a style or design that a user likes.

[0587] A "hashtag" is a symbolic word that users use to identify keywords or themes of interest or concern.

[0588] "Text" refers to a data format in which a user inputs their opinions or wishes as text.

[0589] "Analysis" refers to the process of analyzing user preferences and trends based on user information, images, hashtags, and text.

[0590] "Personalized products" refer to products that are optimized to a user's individual preferences and needs based on analysis results.

[0591] "Recommendation" refers to the act of suggesting to a user that they purchase a particular product.

[0592] "Reasons for recommendation" refers to text that explains the specific appeal of the recommended product and its suitability for the user.

[0593] "Other Items" refers to additional products that can be used in conjunction with the recommended product to provide an enhanced experience.

[0594] "Display" refers to the act of visually providing information to a user via a terminal.

[0595] This invention is implemented by a system including the following components and processes: This system collects user information, recommends personalized products based on the user's preferences and needs, and also suggests specific reasons for recommending the products and related items.

[0596] Hardware and Software

[0597] This system mainly uses the following hardware and software:

[0598] Device: The device used by the user to access the site (e.g., computer, smartphone, tablet)

[0599] Server: A server system for collecting, storing, and analyzing data.

[0600] Database: A database system that stores user information and recommendation information

[0601] Generative AI models: APIs for using natural language processing techniques (e.g., ChatGPT API)

[0602] What the program does

[0603] The process is based on the following steps:

[0604] 1. User Information Collection:

[0605] A session begins when a user logs in to an EC site.

[0606] The device displays a login page and the user enters their existing account information.

[0607] The server authenticates the user's login information and begins the session.

[0608] 2. Displaying survey forms and profile entry pages:

[0609] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[0610] 3. Saving User Information:

[0611] The server stores the information entered by the user in a database in real time.

[0612] 4. Providing recommendation features:

[0613] The device displays an image, hashtags, and a text entry field to the user.

[0614] The user enters an image, hashtags, and text.

[0615] The server receives the input data and analyzes it using a generative AI model.

[0616] Based on the analysis results, the server selects personalized products and generates a recommendation list.

[0617] The device displays the recommendation list to the user.

[0618] 5. Providing boosting features:

[0619] The user selects the product of interest from the recommendation list.

[0620] The server generates specific reasons for recommending the selected product.

[0621] The server will suggest related items (e.g., "white denim" or "light blue sneakers").

[0622] The device will display the recommendation reason and related items to the user.

[0623] Specific examples

[0624] Here are some examples and prompts:

[0625] Example 1: Recommendation function

[0626] A user enters "casual" and "likes pastel colors" in their profile.

[0627] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[0628] The server analyzes the input data and obtains suitable items via a generative AI model.

[0629] The server selects "pastel pink blouse," "white denim," and "camera bag" from the recommendation list and displays them on the device.

[0630] Example 2: Boost function

[0631] The user selects "Pastel Pink Blouse" from the recommendations presented.

[0632] The server generates reasons for recommending this item, explaining, for example, "This pastel pink suits your personal color and is perfect for casual spring occasions."

[0633] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[0634] The terminal displays these coordination suggestions to the user and supports overall styling.

[0635] Example prompt sentence:

[0636] "What casual fashion items would be perfect for a spring picnic? Our users love pastel colors and casual styles."

[0637] As a result, the system of the present invention enables personalized recommendations of fashion items to users, along with specific reasons for the recommendations and coordination suggestions, thereby improving the sense of security and satisfaction of online shopping.

[0638] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0639] Step 1: User logs in to the e-commerce site

[0640] The user accesses the EC site and logs in with an existing account. Authentication is performed using the ID and password entered by the user.

[0641] The server checks the user's credentials against a database and starts a session. The input is the user's credentials and the output is a session ID.

[0642] Step 2: Displaying the survey form and profile entry page

[0643] The terminal displays a questionnaire form or profile entry page to the logged-in user.

[0644] The user enters information into the form, such as their preferences, body type, personal color, past purchase history, the latest trends, etc. The input is the user's personal information, and the output is the user's response data.

[0645] Step 3: Save user information

[0646] The server receives the information entered by the user and stores it in a database.

[0647] This data is used for subsequent analysis and recommendations. The input is the user's personal information data, and the output is a user profile stored in a database.

[0648] Step 4: Displaying an image, hashtags, and text input fields

[0649] The device displays a recommendation page, providing the user with an image, hashtags, and a text input field.

[0650] The user enters their desired image, hashtag, or text in the fields provided. The input is the user's input data (image, hashtag, text), and the output is the input data sent to the server.

[0651] Step 5: Collecting User Input Data

[0652] The server receives the images, hashtags, and text sent by the user and stores them in a database. The input is the user's input data, and the output is the data stored in the database for analysis.

[0653] Step 6: Data analysis and item selection

[0654] The server retrieves the user's profile information and newly entered data from the database and analyzes it using a generative AI model. Specifically, it uses the ChatGPT API to perform natural language processing and abstractly understand the user's preferences.

[0655] As a result of the analysis, the system selects the most suitable product for the user. The input is the profile data and the user's input data, and the output is a list of recommended products.

[0656] Step 7: Generate and display the recommendation list

[0657] Based on the analysis results, the server lists the most suitable products for the user.

[0658] The terminal visually displays this recommendation list to the user. The input is a list of recommended products, and the output is the recommendation list displayed to the user.

[0659] Step 8: Product Selection and Recommendation Generation

[0660] The user selects the product of interest from the recommendation list.

[0661] The server uses a generative AI model to generate specific reasons for recommending the selected product. For example, it generates an explanation such as, "This pastel pink blouse is perfect for your personal color and goes well with a casual spring look." The input is the selected product information, and the output is the reason for the recommendation.

[0662] Step 9: Suggest other items

[0663] The server will suggest other items that go well with the selected item (e.g., "white denim" or "light blue sneakers")

[0664] The server selects related items and sends them to the terminal along with a recommendation list. The input is the selected product information, and the output is a list of suggested items.

[0665] Step 10: View the proposal

[0666] The device displays the recommended product, the reason for the recommendation, and related items to the user.

[0667] The user reviews the overall styling and makes a purchasing decision. The input is data from the server and the output is the information displayed to the user.

[0668] This allows the system to suggest personalized fashion items to users, improving their online shopping experience.

[0669] (Application example 1)

[0670] 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."

[0671] The user experience in online shopping lacks visual information compared to shopping in a physical store, making it difficult to provide personalized product suggestions and encourage purchases. Finding products that match a user's preferences, body type, and personal color is particularly difficult when it comes to fashion items, which can cause users to hesitate before making a purchase. Therefore, there is a demand for systems that provide a more interactive and personalized shopping experience.

[0672] 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.

[0673] In this invention, the server includes means for collecting user information, means for inputting user-preferred images, hashtags, and text, means for analyzing the user information and the images, hashtags, and text, means for selecting and recommending personalized fashion items based on the analysis results, means for generating specific recommendation reasons for the recommended fashion items, means for suggesting other items and cosmetics that are easily combined with the fashion items, means for displaying the recommended fashion items and the suggested other items and cosmetics to the user, and means for visually displaying the recommended fashion items using smart glasses, thereby enabling users to enjoy a more interactive and personalized shopping experience and select and purchase products based on visual information.

[0674] "User information" refers to information such as the user's preferences, body type, personal color, past purchase history, and the latest trends.

[0675] "Images" are images or visual content that a user visually enjoys.

[0676] A "hashtag" is a symbol used to promote information sharing among people with the same hobbies or interests by tagging specific keywords or phrases with "."

[0677] "Text" refers to sentences or character information entered by the user.

[0678] "Analysis Methods" refers to algorithms and technologies used to process user information, images, hashtags, and text to understand and analyze their content.

[0679] "Personalized fashion items" refer to clothing and accessories that are identified and suggested to a user based on their individual information and preferences.

[0680] A "recommendation means" is a system or method that recommends specific fashion items to users based on the analysis results.

[0681] "Specific reasons for recommendation" refers to a detailed explanation of the recommended fashion item and the reasons for its recommendation.

[0682] "Means for suggesting other items or cosmetics" refers to a method or system for presenting other clothing or cosmetics that go well with the recommended fashion item to the user.

[0683] "Visual display means" refers to a method of visually presenting personalized products or offers to a user using a device such as smart glasses.

[0684] "Generative AI model" refers to artificial intelligence technology for generating and analyzing information based on user input and prompts.

[0685] The present invention provides a system for providing personalized fashion item recommendations to users using smart glasses. A specific implementation method thereof will be described below.

[0686] Program processing flow

[0687] 1. Collection of User Information

[0688] When a user logs in to the e-commerce site, the server begins collecting user information. The terminal displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends. This information is sent to the server and stored in a database.

[0689] 2. Implementation of recommendation function

[0690] When a user logs in, the device displays an image, hashtags, and text input field. The user then enters their desired image, hashtag, or text. The server analyzes this input data and interprets it in combination with the user's profile information. The analysis uses natural language processing technology, particularly the ChatGPT API, a generative AI model, to make abstract images concrete.

[0691] 3. Selection of recommended items

[0692] Based on the analysis results, the server selects fashion items that suit the user. Specifically, it also generates personalized reasons for the recommendation. The device then displays a recommendation list to the user. The recommendation list includes the selected items, such as a pastel pink blouse, white denim, and a camera bag.

[0693] 4. Visual display using smart glasses

[0694] The recommended fashion items are displayed visually to the user using the smart glasses, allowing the user to view product details and facilitate purchasing decisions.

[0695] Specific examples

[0696] Example 1: Recommendation function

[0697] A user enters "casual" and "likes pastel colors" in their profile. They then enter "a perfect outfit for a spring picnic" in the text field and add the hashtag "spring casual." The server analyzes this input data and retrieves suitable items via the ChatGPT API of the generative AI model. As a result, "pastel pink blouse," "white denim," and "camera bag" are displayed in the recommendation list.

[0698] Example 2: Boost function

[0699] The user selects a "pastel pink blouse" from the recommendations presented. The server generates a reason for recommending this item, explaining that "this pastel pink suits your personal color and is perfect for spring casual occasions." Furthermore, other items that go well with the blouse, such as "white denim" and "light blue sneakers," are also suggested. The user can visually check these coordination suggestions through the smart glasses, receiving support for overall styling.

[0700] Specific hardware and software used

[0701] Hardware: Smart glasses, user devices (smartphones and PCs)

[0702] Software: E-commerce website platform, generative AI model (ChatGPT API), database

[0703] Examples of prompt statements

[0704] User profile: {'Preferences': 'Casual', 'Favorite color': 'Pastel'}

[0705] User Input: Perfect outfit for a spring picnic Spring Casual

[0706] Recommended items:

[0707] In this way, the system can provide users with a more personalized and visually appealing shopping experience.

[0708] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0709] Step 1:

[0710] Collection of User Information

[0711] Users log in to the e-commerce site using a terminal. The terminal displays a questionnaire form and a profile entry page, where users enter their preferences, body type, personal color, past purchase history, the latest trends, etc. This information is sent to the server and stored in a database.

[0712] Input: User login information, survey responses, profile information

[0713] Output: Saved user information (database)

[0714] Step 2:

[0715] User image, hashtag and text input

[0716] The device displays an image, hashtags, and text input fields to the user, who then enters the image, hashtag, or text of their choice, which is then sent to the server.

[0717] Input: Images, hashtags, and text entered by the user

[0718] Output: Data entered by the user

[0719] Step 3:

[0720] Data analysis

[0721] The server analyzes the collected user information and the input image, hashtag, and text. It uses the ChatGPT API, a generative AI model, to generate prompts. Specifically, it converts abstract images into concrete fashion items based on the user's profile information and text input.

[0722] Input: User information, image, hashtag, text, prompt

[0723] Output: A list of fashion items as the analysis result

[0724] Step 4:

[0725] Personalized fashion item selection and recommendations

[0726] Based on the analysis results, the server selects and recommends fashion items that suit the user. Specific reasons for recommending the items are also provided. The list of selected items is then sent to the device.

[0727] Input: Analysis results, user information, prompt text

[0728] Output: A list of selected fashion items and reasons for their recommendation

[0729] Step 5:

[0730] Preparing for visual display

[0731] The device receives the list of recommended fashion items and the reasons for the recommendations, and prepares them for display on the smart glasses.

[0732] Input: List of selected fashion items and reasons for recommendation

[0733] Output: Data ready for display

[0734] Step 6:

[0735] Visual display with smart glasses

[0736] By using the smart glasses, users can visually see the recommended fashion items and the reasons for their recommendations, allowing them to view product details and make purchasing decisions more easily.

[0737] Input: Data ready to be displayed

[0738] Output: Visual product display via smart glasses

[0739] 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.

[0740] Natural language description of the program

[0741] The system is programmed using the following process:

[0742] 1. Collection of User Information

[0743] A session begins when a user logs in to an EC site.

[0744] The device displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[0745] The server stores this information in a database.

[0746] 2. Recommendation function

[0747] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[0748] The server analyzes these inputs, which involves applying natural language processing based on the user's profile information and sending requests to the ChatGPT API to turn the abstract image into a concrete image.

[0749] Based on the analysis results, the server selects fashion items suitable for the user.

[0750] The device displays a list of recommendations to the user, including a selection of items such as a pastel pink blouse, white denim, and a camera bag.

[0751] 3. Boost function

[0752] The user selects the product of interest from the recommendation list.

[0753] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[0754] The server will suggest other items or cosmetics that go well with the product (such as "white denim" or "light blue sneakers").

[0755] The device displays this information to the user, making it visually convincing.

[0756] 4. Utilizing the Emotion Engine

[0757] The server analyzes the user's input information and selection behavior, and recognizes the user's emotions using an emotion engine.

[0758] The server dynamically adjusts the recommendation content based on the user's emotions. For example, if the user is unsure about which product to choose, the server may change the emphasis on the recommended item or suggest additional similar products.

[0759] The server analyzes the user's reactions in real time and adjusts the reason for the recommendation accordingly. For example, if the user is interested but anxious, the reason will be changed to emphasize positive feedback.

[0760] Specific examples

[0761] Example 1: Combining recommendation features with an emotion engine

[0762] A user enters "casual" and "likes pastel colors" in their profile.

[0763] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[0764] The server analyzes the input data and retrieves the appropriate items via the ChatGPT API.

[0765] The server selects "pastel pink blouse," "white denim," and "camera bag" as recommendations and displays them on the device.

[0766] The server analyzes the user's reactions in real time and recognizes that the user is interested in the recommended item but is also feeling anxious.

[0767] The server adjusts the recommendation statement to read, "This pastel pink suits your personal color and is perfect for casual spring occasions. Many users also rate this item highly."

[0768] Example 2: Combining boosting and emotion engines

[0769] The user selects "Pastel Pink Blouse" from the presented recommendation list.

[0770] The server generates reasons for recommending this item, explaining, for example, "This pastel pink is a perfect match for your personal color and also goes well with a casual spring vibe."

[0771] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[0772] The server analyzes user responses and checks to see which ones are more positive.

[0773] The server displays the original reasons for the recommendation and adds suggestions to encourage additional purchases (e.g., "This outfit is eligible for a 30% off campaign!").

[0774] The device displays this information to the user and supports overall styling.

[0775] As a result, the present invention not only provides users with personalized fashion suggestions and a sense of security when shopping online, but also uses an emotion engine to achieve even more accurate recommendations and support.

[0776] The processing flow will be explained below.

[0777] Step 1:

[0778] A user logs in to an e-commerce site and a session begins.

[0779] Step 2:

[0780] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters their preferences, body type information, personal color, past purchase history, the latest trend information, etc.

[0781] Step 3:

[0782] The server receives the user's input and stores it in a database.

[0783] Step 4:

[0784] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[0785] Step 5:

[0786] The server analyzes the user's input data, which involves applying natural language processing based on the user's profile information and sending a request to the ChatGPT API to materialize the abstract image.

[0787] Step 6:

[0788] The server analyzes the response from the ChatGPT API and selects the appropriate fashion items for the user.

[0789] Step 7:

[0790] The server generates a list of selected fashion items and stores it in a database.

[0791] Step 8:

[0792] The device displays the recommendation list to the user.

[0793] Step 9:

[0794] The user selects the product of interest from the recommendation list.

[0795] Step 10:

[0796] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[0797] Step 11:

[0798] The server will suggest other items (e.g., "white denim" or "light blue sneakers") and cosmetics that go well with the selected product.

[0799] Step 12:

[0800] The terminal displays this information to the user, allowing them to understand it visually.

[0801] Step 13:

[0802] The server analyzes the user's input information and selection behavior, and recognizes the user's emotions using an emotion engine.

[0803] Step 14:

[0804] The server dynamically adjusts the recommendation content based on the user's emotions. For example, if the user is unsure about which product to choose, the server may change the emphasis on the recommended item or suggest additional similar products.

[0805] Step 15:

[0806] The server analyzes the user's reactions in real time and adjusts the reason for the recommendation accordingly. For example, if the user is interested but anxious, the reason will be changed to emphasize positive feedback.

[0807] Through the above steps, users can confidently choose fashion items that suit them, and by using the emotion engine, even more accurate recommendations and support can be achieved.

[0808] Example 2

[0809] 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."

[0810] Conventional online fashion recommendation systems do not fully consider users' subjective preferences and emotions when making recommendations, making it difficult to provide products that truly satisfy users. Furthermore, conventional systems cannot respond to changes in users' emotions, and therefore do not sufficiently motivate users to purchase additional products that they are temporarily interested in.

[0811] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting user information; means for inputting a user's favorite images, hashtags, and text; means for analyzing the user information and the images, hashtags, and text; means for selecting and recommending personalized fashion items based on the analysis results; means for generating specific recommendation reasons for the recommended fashion items; means for suggesting other items and cosmetics that are easily combined with the fashion items; means for displaying the recommended fashion items and the suggested other items and cosmetics to the user; means for analyzing the user's input information and behavior and recognizing emotions; means for dynamically adjusting the recommendation content based on the recognized user emotions; and means for analyzing the user's reaction to the recommended items in real time and dynamically adjusting the recommendation reasons in accordance with the analysis results. This enables personalized recommendations of fashion items based on the user's preferences and emotions, and dynamic adjustment of recommendations in accordance with the user's emotions.

[0812] "Means for collecting user information" refers to the interface and data collection process for collecting information such as the user's preferences, body type, personal color, past purchase history, and the latest trends.

[0813] A "means for users to input their favorite images, hashtags, and text" is an interface that provides forms or input fields for users to enter information about their preferences and style in visual or textual form.

[0814] The "means for analyzing the user information and the images, hashtags, and text" refers to a system that analyzes collected user information and input data using natural language processing and image analysis algorithms.

[0815] The "means for selecting and recommending personalized fashion items based on analysis results" is a system that selects fashion items suitable for the user from the analysis results and provides recommendation information on them.

[0816] The "means for generating a specific reason for recommendation for the recommended fashion item" is a system that generates a specific reason for recommendation related to the selected fashion item and provides it to the user.

[0817] The "means for suggesting other items and cosmetics that can be easily combined with the fashion item" is a system that suggests to the user other related items and cosmetics that can be easily combined with the recommended fashion item.

[0818] The "means for displaying the recommended fashion items and other suggested items and cosmetics to the user" is an interface for visually displaying the selected fashion items and other suggested related items and cosmetics to the user.

[0819] The "means for analyzing the user's input information and behavior and recognizing emotions" is an emotion analysis system that analyzes the user's input data and behavior history and estimates the user's emotional state.

[0820] The "means for dynamically adjusting recommendation content based on recognized user emotions" is a system that adjusts recommendation content in real time based on analyzed emotional information.

[0821] "Means for analyzing user reactions to the recommended item in real time and dynamically adjusting the reason for recommendation based on the analysis results" refers to a system that collects and analyzes user reaction data in real time and dynamically changes the reason for recommendation based on the results.

[0822] A "generative AI model" is an artificial intelligence model that generates natural language and recognizes data patterns based on user information and behavioral data.

[0823] A "prompt" is an instruction or question that is input to a generative AI model to output recommendations.

[0824] System Overview

[0825] This invention is a system that recommends personalized fashion items based on a user's preferences and emotional state. The system is realized by combining four main functions: user information collection, recommendation function, boost function, and emotion engine.

[0826] Hardware and software used

[0827] Server: Used to analyze data and select recommended items. The main software includes Python natural language processing libraries (NLTK and spaCy), sentiment analysis libraries (TextBlob and VADER), and generative AI models (ChatGPT API).

[0828] Device: This serves as the interface where users can input information and view recommended items and descriptions. This is typically a web browser or a mobile app.

[0829] Database: Used to store user information. Popular databases include AWS RDS (Relational Database Service) and MySQL.

[0830] Processing Details

[0831] 1. Collection of User Information

[0832] When a user logs in to an e-commerce site, the device displays a questionnaire form and a profile entry page. The user enters information such as their preferred fashion style, body type, personal color, past purchase history, and the latest trends. This information is sent to the server and stored in a database such as AWS RDS or MySQL.

[0833] 2. Recommendation function

[0834] The device displays an image, hashtags, and a text input field to the user. The user uploads an image of their choice and enters text and hashtags. The server analyzes the received data using Python's natural language processing library (NLTK or spaCy) and generates a prompt for the ChatGPT API based on the analysis results. Based on the response from the ChatGPT API, the server selects appropriate fashion items and displays them on the device.

[0835] 3. Boost function

[0836] When a user selects a product from the recommendation list, the server generates specific reasons for recommending that product. For example, a reason might be provided such as, "This pastel pink blouse matches your personal color perfectly." Other related items (such as "white denim" or "light blue sneakers") are also suggested. This information is displayed on the device, allowing the user to visually understand the recommendation.

[0837] 4. Utilizing the Emotion Engine

[0838] The server analyzes the user's input information and behavior using emotion analysis libraries (TextBlob and VADER) to recognize the user's emotions. If the user is unsure about which product to choose, the server dynamically adjusts the emphasis of the recommended items. For example, it adds information about limited-time sales or special offers. The user's reactions are analyzed in real time, and if the user shows anxiety, for example, it changes the reason statement to emphasize positive feedback.

[0839] Specific examples

[0840] Combining recommendation functionality with an emotion engine

[0841] A user enters "casual" and "I like pastel colors" in their profile. They then write "A perfect outfit for a spring picnic" in the text field and add the hashtag "spring casual." The server analyzes the input data and selects "pastel pink blouse," "white denim," and "camera bag" as recommendations, which are displayed on the device. The server analyzes the user's reactions in real time, and if the user is interested in a recommended item but has concerns, it dynamically adjusts the reason for the recommendation to something like, "This pastel pink suits your personal color and is perfect for spring casual occasions."

[0842] Prompt Sentence Examples

[0843] "Suggest a casual outfit perfect for a spring picnic. Your user likes pastel pink and prefers casual styles."

[0844] This invention makes it possible to provide personalized fashion suggestions based on a user's preferences and emotions, improving the online shopping experience.

[0845] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0846] Step 1:

[0847] A user logs in to an e-commerce site.

[0848] Input: User ID and password

[0849] Operation: The terminal displays the login screen. The user enters their ID and password and presses the "Login" button.

[0850] Output: A session is started and the user's profile information is loaded.

[0851] Step 2:

[0852] The device will display a survey form and profile entry page.

[0853] Input: Display requirements for survey forms and profile entry pages

[0854] How it works: The device generates an input form with fields such as "preferred fashion style," "body type information," and "personal color."

[0855] Output: The form that is displayed to the user

[0856] Step 3:

[0857] The user enters information into a survey form or profile entry page and submits it.

[0858] Input: Information such as "preferred fashion style," "body type information," "personal color," "past purchase history," and "latest trends"

[0859] What happens: The user enters information into the fields and presses the "Submit" button.

[0860] Output: The entered user information is sent to the server.

[0861] Step 4:

[0862] The server stores the user information in a database.

[0863] Input: Information entered by the user

[0864] How it works: The server receives the data and stores it in a database such as AWS RDS or MySQL.

[0865] Output: User information stored in the database

[0866] Step 5:

[0867] The device displays an image, hashtags, and a text entry field to the user.

[0868] Input: Display requirements for images, hashtags, and text input fields

[0869] How it works: Your device will display options such as "Upload your favorite image," "Related hashtags," and "Describe your desired style."

[0870] Output: Input fields that are displayed to the user

[0871] Step 6:

[0872] The user enters an image, text, and hashtags.

[0873] Input: User-uploaded images, entered text, hashtags

[0874] How it works: User uploads an image, enters text and hashtags.

[0875] Output: The entered data is sent to the server.

[0876] Step 7:

[0877] The server parses the input data.

[0878] Input: Image, Text, Hashtag

[0879] How it works: The server parses the data using Python natural language processing libraries (NLTK or spaCy) and generates prompts to the ChatGPT API based on the results of the analysis.

[0880] Output: Generated prompt, response data from ChatGPT API

[0881] Step 8:

[0882] The server selects recommended items based on the analysis results.

[0883] Input: Response data from ChatGPT API, user profile information

[0884] Operation: The server uses the response data to create a list of fashion items suitable for the user.

[0885] Output: A list of selected recommended items

[0886] Step 9:

[0887] The device displays a recommendation list to the user.

[0888] Input: Recommended item list

[0889] What it does: Your device will display a list of items such as "Pastel Pink Blouse," "White Denim," and "Camera Bag."

[0890] Output: The recommendation list displayed to the user

[0891] Step 10:

[0892] The user selects the product of interest from the recommendation list.

[0893] Input: Products you are interested in from the recommendation list

[0894] Action: The user clicks on the selected item.

[0895] Output: Selected product information is sent to the server.

[0896] Step 11:

[0897] The server generates specific recommendation reasons.

[0898] Input: Selected product information, user profile information

[0899] How it works: The server generates a recommendation reason based on the user's profile and product attributes, for example, "This pastel pink blouse matches your personal color."

[0900] Output: Generated recommendation reason text

[0901] Step 12:

[0902] The server suggests other related items.

[0903] Input: Selected product information

[0904] What it does: The server lists items that go well with the selected item, such as "white denim" or "light blue sneakers."

[0905] Output: A list of suggested related items

[0906] Step 13:

[0907] The device will display the recommendation reason and related items to the user.

[0908] Input: Generated recommendation reason text, suggested related item list

[0909] What it does: The device displays this information to the user in a visually convincing way.

[0910] Output: Displayed recommendation reason and related items

[0911] Step 14:

[0912] The server analyzes the user's input information and behavior.

[0913] Input: User input data and behavior history

[0914] How it works: The server analyzes the data using a sentiment analysis library (TextBlob or VADER) to recognize the user's sentiment.

[0915] Output: Recognized emotion data

[0916] Step 15:

[0917] The server dynamically adjusts the recommendation content based on emotions.

[0918] Input: Recognized emotion data, recommended item list

[0919] How it works: If the server detects negative sentiment, it changes the emphasis of the recommended items and adds limited-time sales and special offers.

[0920] Output: Dynamically adjusted recommendations

[0921] Step 16:

[0922] The server analyzes user responses in real time and adjusts the wording of the recommendation reasons.

[0923] Input: User response data (viewing time, number of clicks, etc.)

[0924] How it works: The server analyzes the user's response and modifies the generated reason statement based on the results. For example, if the user is feeling anxious, it emphasizes more positive feedback.

[0925] Output: Adjusted recommendation reason

[0926] (Application example 2)

[0927] 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."

[0928] While conventional fashion recommendation systems suggest personalized items to users when shopping online, they do not address the in-store shopping experience. Furthermore, they do not provide recommendations that take into account the user's real-time movements and emotions, which prevents them from fully enhancing user satisfaction. Furthermore, there is a need for a system that effectively utilizes devices such as eye-tracking technology, smart glasses, and head-mounted displays to enable users to obtain the information they want in-store in real time.

[0929] 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.

[0930] In this invention, the server includes an eye tracking unit that identifies a specific product when the user moves around the store and directs their gaze at the product, a unit that dynamically adjusts recommendations for the identified product according to the user's emotional state, and a unit that displays the generated recommendation reasons and recommendation content on smart glasses or a head-mounted display, thereby improving the shopping experience in a physical store and making it possible to provide appropriate information and recommendations in real time for products that interest the user.

[0931] "Means for collecting user information" refers to a function for inputting and recording user preferences, body type, personal color, past purchase history, the latest trends, and the like.

[0932] The "means for users to input their favorite images, hashtags, and text" refers to an interface that allows users to input images, hashtags, text, etc. that specifically indicate their own fashion style and preferences.

[0933] "Means for analyzing user information and the images, hashtags, and text" refers to a function for analyzing collected user information and input images, hashtags, and text to understand user preferences and trends.

[0934] "Means for selecting and recommending personalized fashion items" refers to a system that selects and suggests the most suitable fashion items for users based on the analysis results.

[0935] The "means for generating specific reasons for recommendation" is a function for automatically generating detailed explanations and reasons why recommended fashion items are preferred.

[0936] The "means for suggesting other items and cosmetics" is a function for automatically suggesting other fashion items and cosmetics that can be easily combined with the recommended fashion items.

[0937] The "means for displaying to the user" is an interface for visually presenting information about the recommended fashion items and other suggested items and cosmetics to the user.

[0938] The "eye tracking means" is a technology for identifying and tracking a particular product when the user directs their gaze at the product.

[0939] "Dynamic adjustment means" is a function that allows recommendations to be changed and adjusted in real time according to the user's emotions and behavior.

[0940] "Means for displaying on smart glasses or a head-mounted display" refers to a function for displaying the generated recommendation reasons and recommendation content on a smart device in real time.

[0941] Specific configurations and methods for implementing the present invention are described below.

[0942] 1. Collection of User Information

[0943] The terminal uses smart glasses and mobile devices to collect user information. Users use these devices to input their preferred style, body type, personal color, past purchase history, and the latest trend information. This information is sent to a server and stored in a database.

[0944] 2. Eye tracking and product identification

[0945] The smart glasses use built-in eye-tracking technology (e.g., Tobii Eye Tracker SDK) to identify the products the user is looking at in the store, recognize the products the user is looking at, and send the related information to a server.

[0946] 3. Recommendation Generation and Sentiment Analysis

[0947] The server generates personalized recommendations based on the aforementioned user information and eye-tracking information. This includes suggesting fashion items based on the user's input and profile using a generative AI model (e.g., OpenAI API). It also uses an emotion analysis engine to analyze the user's emotional state from their facial expressions and behavior and dynamically adjust the recommendations.

[0948] 4. Displaying recommendation reasons and dynamic adjustment

[0949] The device displays the recommendation content and specific reasons for the recommendation in real time on smart glasses or a head-mounted display. Based on the results of the sentiment analysis engine, the device adjusts the recommendation reasons and the display of recommended items. For example, if the user is interested but anxious, it will emphasize positive feedback.

[0950] Specific examples

[0951] For example, if a user wears smart glasses and enters a store and looks at a particular product, fashion items related to that product will be automatically recommended. The recommendation will be displayed on the visual device along with the reason for the recommendation generated by the server based on the user's information. This allows users to enjoy shopping while receiving detailed information and personalized recommendations.

[0952] Prompt Sentence Examples

[0953] An example of a prompt generated by the server is:

[0954] markdown

[0955] User Profile:

[0956] Favorite style: Casual, pastel colors

[0957] Body type: Medium build

[0958] Past purchases: Jeans, blouses, casual shoes

[0959] User input:

[0960] Tags: Spring Casual

[0961] Keywords: Spring Picnic

[0962] Recommended Items:

[0963] pastel pink blouse

[0964] White denim

[0965] camera bag

[0966] Recommended reasons include:

[0967] This pastel pink blouse is perfect for your personal color and is perfect for casual spring occasions!

[0968] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0969] Step 1:

[0970] The user logs in using smart glasses or a mobile device. After logging in, the user enters their preferred style, body type, personal color, past purchase history, latest trend information, etc. This information is sent to the server and stored in a database.

[0971] Input: User profile information (style, body type, color, purchase history, trend information)

[0972] Output: User information stored in the database

[0973] Step 2:

[0974] As a user moves around the store, the smart glasses' eye-tracking technology (Tobii Eye Tracker SDK) is used to identify the products they are looking at. The glasses then transmit the user's gaze data to a server.

[0975] Input: User gaze data

[0976] Output: Identified product information

[0977] Step 3:

[0978] The server analyzes the user's gaze data and saved profile information to generate personalized recommendations for the user. A generative AI model (OpenAI API) is used to suggest fashion items based on the user's prompts.

[0979] Input: Identified product information, user profile information

[0980] Output: A list of recommended fashion items

[0981] Step 4:

[0982] The server inputs the user's prompt information into the generative AI model to generate specific recommendation reasons, including specific reasons and explanations for the new recommendation.

[0983] Input: Identified product information, user profile information, prompt information

[0984] Output: Specific reasons for recommendation

[0985] Step 5:

[0986] The server uses an emotion analysis engine to analyze the user's emotional state based on their facial expressions and behavior. Based on the user's emotions, the recommendation content is dynamically adjusted. For example, if the user is interested but anxious, the server emphasizes positive feedback.

[0987] Input: gaze data, facial expression data, behavior data

[0988] Output: Adjusted recommendation content

[0989] Step 6:

[0990] The device (smart glasses or head-mounted display) displays the generated recommendation content and specific reasons for the recommendation in real time. The user can consider products based on this information and obtain additional information as needed.

[0991] Input: Adjusted recommendation content, specific reasons for recommendation

[0992] Output: Present information to the user

[0993] Step 7:

[0994] The user selects recommended fashion items or other suggested items through their device. Detailed information about the selected item and related items are displayed. The device then sends the information back to the server, which generates additional recommendations as needed.

[0995] Input: User selection information

[0996] Output: Show detailed information and related items

[0997] Through these steps, users can enjoy a personalized shopping experience in a physical store.

[0998] 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.

[0999] 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.

[1000] 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.

[1001] [Third embodiment]

[1002] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1003] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1004] 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).

[1005] 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.

[1006] 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.

[1007] 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).

[1008] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1009] 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.

[1010] 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.

[1011] 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.

[1012] 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.

[1013] 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."

[1014] Natural language description of the program

[1015] The system is programmed using the following process:

[1016] 1. Collection of User Information

[1017] A session begins when a user logs in to an EC site.

[1018] The device displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[1019] The server stores this information in a database.

[1020] 2. Recommendation function

[1021] The device displays an image, hashtag, and text input field to the user, and the user enters their favorite image, hashtag, or text.

[1022] The server analyzes this input data and combines it with the user's profile information to interpret it. The analysis uses natural language processing technology, particularly the ChatGPT API, to make abstract images concrete.

[1023] The server selects fashion items that suit the user based on the analysis results.

[1024] The device displays a list of recommendations to the user, including a selection of items such as a pastel pink blouse, white denim, and a camera bag.

[1025] 3. Boost function

[1026] The user selects the product of interest from the recommendation list.

[1027] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[1028] The server will suggest other items or cosmetics that go well with the product (such as "white denim" or "light blue sneakers").

[1029] The device displays this information to the user, making it visually convincing.

[1030] Specific examples

[1031] Example 1: Recommendation function

[1032] A user enters "casual" and "likes pastel colors" in their profile.

[1033] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[1034] The server analyzes the input data and retrieves the appropriate items via the ChatGPT API.

[1035] The server selects "pastel pink blouse," "white denim," and "camera bag" as recommendations and displays them on the device.

[1036] Example 2: Boost function

[1037] The user selects "Pastel Pink Blouse" from the recommendations presented.

[1038] The server generates reasons for recommending this item, explaining, for example, "This pastel pink suits your personal color and is perfect for casual spring occasions."

[1039] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[1040] The terminal displays these coordination suggestions to the user and supports overall styling.

[1041] As a result, the present invention provides users with personalized fashion suggestions and a sense of security when shopping online.

[1042] The processing flow will be explained below.

[1043] Step 1:

[1044] A user logs in to an e-commerce site.

[1045] Step 2:

[1046] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters preferences, body type information, personal color, past purchase history, the latest trends, etc.

[1047] Step 3:

[1048] The server receives the user's input and stores it in a database.

[1049] Step 4:

[1050] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[1051] Step 5:

[1052] The server analyzes the user's input data, which involves applying natural language processing based on the user's profile information and sending a request to the ChatGPT API to materialize the abstract image.

[1053] Step 6:

[1054] The server analyzes the response from the ChatGPT API and selects the appropriate fashion items for the user.

[1055] Step 7:

[1056] The server generates a list of selected fashion items (e.g., "pastel pink blouse," "white denim," "camera bag," etc.) and stores it in a database.

[1057] Step 8:

[1058] The device displays the recommendation list to the user.

[1059] Step 9:

[1060] The user selects the product of interest from the recommendation list.

[1061] Step 10:

[1062] The server generates specific reasons for recommending the selected products, such as "This pastel pink is a perfect match for your personal color and also goes well with a casual spring vibe."

[1063] Step 11:

[1064] The server will suggest other items (e.g., "white denim" or "light blue sneakers") and cosmetics that go well with the selected product.

[1065] Step 12:

[1066] The terminal displays this information to the user, allowing them to understand it visually.

[1067] By following the above steps, users can confidently select fashion items that suit them, reducing the chances of making mistakes when shopping online.

[1068] Example 1

[1069] 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."

[1070] In conventional online shopping, it is difficult for users to find products that suit them. Especially when it comes to fashion items, recommendations do not sufficiently take into account the user's individual preferences, body type, personal color, etc. Furthermore, since there is no specific reason for the recommended product, users lack a sense of security when making a purchase decision. Furthermore, suggestions for coordinating the recommended product with other items are also insufficient. This leads to problems such as low user satisfaction and a decrease in purchasing motivation.

[1071] 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.

[1072] In this invention, the server includes means for collecting user information, means for users to input their favorite images, hashtags, and text, means for analyzing the user information and the images, hashtags, and text, means for selecting and recommending personalized products based on the analysis results, means for generating specific reasons for recommending the recommended products, means for suggesting other items that are easy to combine with the recommended products, and means for displaying the recommended products and other suggested items to the user. This allows the server to provide users with personalized fashion item recommendations and specific reasons for the recommendations, and can also provide coordination suggestions. This allows users to easily find fashion items that suit them and make purchasing decisions with confidence.

[1073] "User information" refers to information such as the user's individual preferences, body type, personal color, past purchase history, and the latest trends.

[1074] "Image" refers to an image that visually represents a style or design that a user likes.

[1075] A "hashtag" is a symbolic word that users use to identify keywords or themes of interest or concern.

[1076] "Text" refers to a data format in which a user inputs their opinions or wishes as text.

[1077] "Analysis" refers to the process of analyzing user preferences and trends based on user information, images, hashtags, and text.

[1078] "Personalized products" refer to products that are optimized to a user's individual preferences and needs based on analysis results.

[1079] "Recommendation" refers to the act of suggesting to a user that they purchase a particular product.

[1080] "Reasons for recommendation" refers to text that explains the specific appeal of the recommended product and its suitability for the user.

[1081] "Other Items" refers to additional products that can be used in conjunction with the recommended product to provide an enhanced experience.

[1082] "Display" refers to the act of visually providing information to a user via a terminal.

[1083] This invention is implemented by a system including the following components and processes: This system collects user information, recommends personalized products based on the user's preferences and needs, and also suggests specific reasons for recommending the products and related items.

[1084] Hardware and Software

[1085] This system mainly uses the following hardware and software:

[1086] Device: The device used by the user to access the site (e.g., computer, smartphone, tablet)

[1087] Server: A server system for collecting, storing, and analyzing data.

[1088] Database: A database system that stores user information and recommendation information

[1089] Generative AI models: APIs for using natural language processing techniques (e.g., ChatGPT API)

[1090] What the program does

[1091] The process is based on the following steps:

[1092] 1. User Information Collection:

[1093] A session begins when a user logs in to an EC site.

[1094] The device displays a login page and the user enters their existing account information.

[1095] The server authenticates the user's login information and begins the session.

[1096] 2. Displaying survey forms and profile entry pages:

[1097] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[1098] 3. Saving User Information:

[1099] The server stores the information entered by the user in a database in real time.

[1100] 4. Providing recommendation features:

[1101] The device displays an image, hashtags, and a text entry field to the user.

[1102] The user enters an image, hashtags, and text.

[1103] The server receives the input data and analyzes it using a generative AI model.

[1104] Based on the analysis results, the server selects personalized products and generates a recommendation list.

[1105] The device displays the recommendation list to the user.

[1106] 5. Providing boosting features:

[1107] The user selects the product of interest from the recommendation list.

[1108] The server generates specific reasons for recommending the selected product.

[1109] The server will suggest related items (e.g., "white denim" or "light blue sneakers").

[1110] The device will display the recommendation reason and related items to the user.

[1111] Specific examples

[1112] Here are some examples and prompts:

[1113] Example 1: Recommendation function

[1114] A user enters "casual" and "likes pastel colors" in their profile.

[1115] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[1116] The server analyzes the input data and obtains suitable items via a generative AI model.

[1117] The server selects "pastel pink blouse," "white denim," and "camera bag" from the recommendation list and displays them on the device.

[1118] Example 2: Boost function

[1119] The user selects "Pastel Pink Blouse" from the recommendations presented.

[1120] The server generates reasons for recommending this item, explaining, for example, "This pastel pink suits your personal color and is perfect for casual spring occasions."

[1121] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[1122] The terminal displays these coordination suggestions to the user and supports overall styling.

[1123] Example prompt sentence:

[1124] "What casual fashion items would be perfect for a spring picnic? Our users love pastel colors and casual styles."

[1125] As a result, the system of the present invention enables personalized recommendations of fashion items to users, along with specific reasons for the recommendations and coordination suggestions, thereby improving the sense of security and satisfaction of online shopping.

[1126] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1127] Step 1: User logs in to the e-commerce site

[1128] The user accesses the EC site and logs in with an existing account. Authentication is performed using the ID and password entered by the user.

[1129] The server checks the user's credentials against a database and starts a session. The input is the user's credentials and the output is a session ID.

[1130] Step 2: Displaying the survey form and profile entry page

[1131] The terminal displays a questionnaire form or profile entry page to the logged-in user.

[1132] The user enters information into the form, such as their preferences, body type, personal color, past purchase history, the latest trends, etc. The input is the user's personal information, and the output is the user's response data.

[1133] Step 3: Save user information

[1134] The server receives the information entered by the user and stores it in a database.

[1135] This data is used for subsequent analysis and recommendations. The input is the user's personal information data, and the output is a user profile stored in a database.

[1136] Step 4: Displaying an image, hashtags, and text input fields

[1137] The device displays a recommendation page, providing the user with an image, hashtags, and a text input field.

[1138] The user enters their desired image, hashtag, or text in the fields provided. The input is the user's input data (image, hashtag, text), and the output is the input data sent to the server.

[1139] Step 5: Collecting User Input Data

[1140] The server receives the images, hashtags, and text sent by the user and stores them in a database. The input is the user's input data, and the output is the data stored in the database for analysis.

[1141] Step 6: Data analysis and item selection

[1142] The server retrieves the user's profile information and newly entered data from the database and analyzes it using a generative AI model. Specifically, it uses the ChatGPT API to perform natural language processing and abstractly understand the user's preferences.

[1143] As a result of the analysis, the system selects the most suitable product for the user. The input is the profile data and the user's input data, and the output is a list of recommended products.

[1144] Step 7: Generate and display the recommendation list

[1145] Based on the analysis results, the server lists the most suitable products for the user.

[1146] The terminal visually displays this recommendation list to the user. The input is a list of recommended products, and the output is the recommendation list displayed to the user.

[1147] Step 8: Product Selection and Recommendation Generation

[1148] The user selects the product of interest from the recommendation list.

[1149] The server uses a generative AI model to generate specific reasons for recommending the selected product. For example, it generates an explanation such as, "This pastel pink blouse is perfect for your personal color and goes well with a casual spring look." The input is the selected product information, and the output is the reason for the recommendation.

[1150] Step 9: Suggest other items

[1151] The server will suggest other items that go well with the selected item (e.g., "white denim" or "light blue sneakers")

[1152] The server selects related items and sends them to the terminal along with a recommendation list. The input is the selected product information, and the output is a list of suggested items.

[1153] Step 10: View the proposal

[1154] The device displays the recommended product, the reason for the recommendation, and related items to the user.

[1155] The user reviews the overall styling and makes a purchasing decision. The input is data from the server and the output is the information displayed to the user.

[1156] This allows the system to suggest personalized fashion items to users, improving their online shopping experience.

[1157] (Application example 1)

[1158] 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."

[1159] The user experience in online shopping lacks visual information compared to shopping in a physical store, making it difficult to provide personalized product suggestions and encourage purchases. Finding products that match a user's preferences, body type, and personal color is particularly difficult when it comes to fashion items, which can cause users to hesitate before making a purchase. Therefore, there is a demand for systems that provide a more interactive and personalized shopping experience.

[1160] 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.

[1161] In this invention, the server includes means for collecting user information, means for inputting user-preferred images, hashtags, and text, means for analyzing the user information and the images, hashtags, and text, means for selecting and recommending personalized fashion items based on the analysis results, means for generating specific recommendation reasons for the recommended fashion items, means for suggesting other items and cosmetics that are easily combined with the fashion items, means for displaying the recommended fashion items and the suggested other items and cosmetics to the user, and means for visually displaying the recommended fashion items using smart glasses, thereby enabling users to enjoy a more interactive and personalized shopping experience and select and purchase products based on visual information.

[1162] "User information" refers to information such as the user's preferences, body type, personal color, past purchase history, and the latest trends.

[1163] "Images" are images or visual content that a user visually enjoys.

[1164] A "hashtag" is a symbol used to promote information sharing among people with the same hobbies or interests by tagging specific keywords or phrases with "."

[1165] "Text" refers to sentences or character information entered by the user.

[1166] "Analysis Methods" refers to algorithms and technologies used to process user information, images, hashtags, and text to understand and analyze their content.

[1167] "Personalized fashion items" refer to clothing and accessories that are identified and suggested to a user based on their individual information and preferences.

[1168] A "recommendation means" is a system or method that recommends specific fashion items to users based on the analysis results.

[1169] "Specific reasons for recommendation" refers to a detailed explanation of the recommended fashion item and the reasons for its recommendation.

[1170] "Means for suggesting other items or cosmetics" refers to a method or system for presenting other clothing or cosmetics that go well with the recommended fashion item to the user.

[1171] "Visual display means" refers to a method of visually presenting personalized products or offers to a user using a device such as smart glasses.

[1172] "Generative AI model" refers to artificial intelligence technology for generating and analyzing information based on user input and prompts.

[1173] The present invention provides a system for providing personalized fashion item recommendations to users using smart glasses. A specific implementation method thereof will be described below.

[1174] Program processing flow

[1175] 1. Collection of User Information

[1176] When a user logs in to the e-commerce site, the server begins collecting user information. The terminal displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends. This information is sent to the server and stored in a database.

[1177] 2. Implementation of recommendation function

[1178] When a user logs in, the device displays an image, hashtags, and text input field. The user then enters their desired image, hashtag, or text. The server analyzes this input data and interprets it in combination with the user's profile information. The analysis uses natural language processing technology, particularly the ChatGPT API, a generative AI model, to make abstract images concrete.

[1179] 3. Selection of recommended items

[1180] Based on the analysis results, the server selects fashion items that suit the user. Specifically, it also generates personalized reasons for the recommendation. The device then displays a recommendation list to the user. The recommendation list includes the selected items, such as a pastel pink blouse, white denim, and a camera bag.

[1181] 4. Visual display using smart glasses

[1182] The recommended fashion items are displayed visually to the user using the smart glasses, allowing the user to view product details and facilitate purchasing decisions.

[1183] Specific examples

[1184] Example 1: Recommendation function

[1185] A user enters "casual" and "likes pastel colors" in their profile. They then enter "a perfect outfit for a spring picnic" in the text field and add the hashtag "spring casual." The server analyzes this input data and retrieves suitable items via the ChatGPT API of the generative AI model. As a result, "pastel pink blouse," "white denim," and "camera bag" are displayed in the recommendation list.

[1186] Example 2: Boost function

[1187] The user selects a "pastel pink blouse" from the recommendations presented. The server generates a reason for recommending this item, explaining that "this pastel pink suits your personal color and is perfect for spring casual occasions." Furthermore, other items that go well with the blouse, such as "white denim" and "light blue sneakers," are also suggested. The user can visually check these coordination suggestions through the smart glasses, receiving support for overall styling.

[1188] Specific hardware and software used

[1189] Hardware: Smart glasses, user devices (smartphones and PCs)

[1190] Software: E-commerce website platform, generative AI model (ChatGPT API), database

[1191] Examples of prompt statements

[1192] User profile: {'Preferences': 'Casual', 'Favorite color': 'Pastel'}

[1193] User Input: Perfect outfit for a spring picnic Spring Casual

[1194] Recommended items:

[1195] In this way, the system can provide users with a more personalized and visually appealing shopping experience.

[1196] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1197] Step 1:

[1198] Collection of User Information

[1199] Users log in to the e-commerce site using a terminal. The terminal displays a questionnaire form and a profile entry page, where users enter their preferences, body type, personal color, past purchase history, the latest trends, etc. This information is sent to the server and stored in a database.

[1200] Input: User login information, survey responses, profile information

[1201] Output: Saved user information (database)

[1202] Step 2:

[1203] User image, hashtag and text input

[1204] The device displays an image, hashtags, and text input fields to the user, who then enters the image, hashtag, or text of their choice, which is then sent to the server.

[1205] Input: Images, hashtags, and text entered by the user

[1206] Output: Data entered by the user

[1207] Step 3:

[1208] Data analysis

[1209] The server analyzes the collected user information and the input image, hashtag, and text. It uses the ChatGPT API, a generative AI model, to generate prompts. Specifically, it converts abstract images into concrete fashion items based on the user's profile information and text input.

[1210] Input: User information, image, hashtag, text, prompt

[1211] Output: A list of fashion items as the analysis result

[1212] Step 4:

[1213] Personalized fashion item selection and recommendations

[1214] Based on the analysis results, the server selects and recommends fashion items that suit the user. Specific reasons for recommending the items are also provided. The list of selected items is then sent to the device.

[1215] Input: Analysis results, user information, prompt text

[1216] Output: A list of selected fashion items and reasons for their recommendation

[1217] Step 5:

[1218] Preparing for visual display

[1219] The device receives the list of recommended fashion items and the reasons for the recommendations, and prepares them for display on the smart glasses.

[1220] Input: List of selected fashion items and reasons for recommendation

[1221] Output: Data ready for display

[1222] Step 6:

[1223] Visual display with smart glasses

[1224] By using the smart glasses, users can visually see the recommended fashion items and the reasons for their recommendations, allowing them to view product details and make purchasing decisions more easily.

[1225] Input: Data ready to be displayed

[1226] Output: Visual product display via smart glasses

[1227] 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.

[1228] Natural language description of the program

[1229] The system is programmed using the following process:

[1230] 1. Collection of User Information

[1231] A session begins when a user logs in to an EC site.

[1232] The device displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[1233] The server stores this information in a database.

[1234] 2. Recommendation function

[1235] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[1236] The server analyzes these inputs, which involves applying natural language processing based on the user's profile information and sending requests to the ChatGPT API to turn the abstract image into a concrete image.

[1237] Based on the analysis results, the server selects fashion items suitable for the user.

[1238] The device displays a list of recommendations to the user, including a selection of items such as a pastel pink blouse, white denim, and a camera bag.

[1239] 3. Boost function

[1240] The user selects the product of interest from the recommendation list.

[1241] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[1242] The server will suggest other items or cosmetics that go well with the product (such as "white denim" or "light blue sneakers").

[1243] The device displays this information to the user, making it visually convincing.

[1244] 4. Utilizing the Emotion Engine

[1245] The server analyzes the user's input information and selection behavior, and recognizes the user's emotions using an emotion engine.

[1246] The server dynamically adjusts the recommendation content based on the user's emotions. For example, if the user is unsure about which product to choose, the server may change the emphasis on the recommended item or suggest additional similar products.

[1247] The server analyzes the user's reactions in real time and adjusts the reason for the recommendation accordingly. For example, if the user is interested but anxious, the reason will be changed to emphasize positive feedback.

[1248] Specific examples

[1249] Example 1: Combining recommendation features with an emotion engine

[1250] A user enters "casual" and "likes pastel colors" in their profile.

[1251] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[1252] The server analyzes the input data and retrieves the appropriate items via the ChatGPT API.

[1253] The server selects "pastel pink blouse," "white denim," and "camera bag" as recommendations and displays them on the device.

[1254] The server analyzes the user's reactions in real time and recognizes that the user is interested in the recommended item but is also feeling anxious.

[1255] The server adjusts the recommendation statement to read, "This pastel pink suits your personal color and is perfect for casual spring occasions. Many users also rate this item highly."

[1256] Example 2: Combining boosting and emotion engines

[1257] The user selects "Pastel Pink Blouse" from the presented recommendation list.

[1258] The server generates reasons for recommending this item, explaining, for example, "This pastel pink is a perfect match for your personal color and also goes well with a casual spring vibe."

[1259] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[1260] The server analyzes user responses and checks to see which ones are more positive.

[1261] The server displays the original reasons for the recommendation and adds suggestions to encourage additional purchases (e.g., "This outfit is eligible for a 30% off campaign!").

[1262] The device displays this information to the user and supports overall styling.

[1263] As a result, the present invention not only provides users with personalized fashion suggestions and a sense of security when shopping online, but also uses an emotion engine to achieve even more accurate recommendations and support.

[1264] The processing flow will be explained below.

[1265] Step 1:

[1266] A user logs in to an e-commerce site and a session begins.

[1267] Step 2:

[1268] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters their preferences, body type information, personal color, past purchase history, the latest trend information, etc.

[1269] Step 3:

[1270] The server receives the user's input and stores it in a database.

[1271] Step 4:

[1272] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[1273] Step 5:

[1274] The server analyzes the user's input data, which involves applying natural language processing based on the user's profile information and sending a request to the ChatGPT API to materialize the abstract image.

[1275] Step 6:

[1276] The server analyzes the response from the ChatGPT API and selects the appropriate fashion items for the user.

[1277] Step 7:

[1278] The server generates a list of selected fashion items and stores it in a database.

[1279] Step 8:

[1280] The device displays the recommendation list to the user.

[1281] Step 9:

[1282] The user selects the product of interest from the recommendation list.

[1283] Step 10:

[1284] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[1285] Step 11:

[1286] The server will suggest other items (e.g., "white denim" or "light blue sneakers") and cosmetics that go well with the selected product.

[1287] Step 12:

[1288] The terminal displays this information to the user, allowing them to understand it visually.

[1289] Step 13:

[1290] The server analyzes the user's input information and selection behavior, and recognizes the user's emotions using an emotion engine.

[1291] Step 14:

[1292] The server dynamically adjusts the recommendation content based on the user's emotions. For example, if the user is unsure about which product to choose, the server may change the emphasis on the recommended item or suggest additional similar products.

[1293] Step 15:

[1294] The server analyzes the user's reactions in real time and adjusts the reason for the recommendation accordingly. For example, if the user is interested but anxious, the reason will be changed to emphasize positive feedback.

[1295] Through the above steps, users can confidently choose fashion items that suit them, and by using the emotion engine, even more accurate recommendations and support can be achieved.

[1296] Example 2

[1297] 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."

[1298] Conventional online fashion recommendation systems do not fully consider users' subjective preferences and emotions when making recommendations, making it difficult to provide products that truly satisfy users. Furthermore, conventional systems cannot respond to changes in users' emotions, and therefore do not sufficiently motivate users to purchase additional products that they are temporarily interested in.

[1299] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting user information; means for inputting a user's favorite images, hashtags, and text; means for analyzing the user information and the images, hashtags, and text; means for selecting and recommending personalized fashion items based on the analysis results; means for generating specific recommendation reasons for the recommended fashion items; means for suggesting other items and cosmetics that are easily combined with the fashion items; means for displaying the recommended fashion items and the suggested other items and cosmetics to the user; means for analyzing the user's input information and behavior and recognizing emotions; means for dynamically adjusting the recommendation content based on the recognized user emotions; and means for analyzing the user's reaction to the recommended items in real time and dynamically adjusting the recommendation reasons in accordance with the analysis results. This enables personalized recommendations of fashion items based on the user's preferences and emotions, and dynamic adjustment of recommendations in accordance with the user's emotions.

[1300] "Means for collecting user information" refers to the interface and data collection process for collecting information such as the user's preferences, body type, personal color, past purchase history, and the latest trends.

[1301] A "means for users to input their favorite images, hashtags, and text" is an interface that provides forms or input fields for users to enter information about their preferences and style in visual or textual form.

[1302] The "means for analyzing the user information and the images, hashtags, and text" refers to a system that analyzes collected user information and input data using natural language processing and image analysis algorithms.

[1303] The "means for selecting and recommending personalized fashion items based on analysis results" is a system that selects fashion items suitable for the user from the analysis results and provides recommendation information on them.

[1304] The "means for generating a specific reason for recommendation for the recommended fashion item" is a system that generates a specific reason for recommendation related to the selected fashion item and provides it to the user.

[1305] The "means for suggesting other items and cosmetics that can be easily combined with the fashion item" is a system that suggests to the user other related items and cosmetics that can be easily combined with the recommended fashion item.

[1306] The "means for displaying the recommended fashion items and other suggested items and cosmetics to the user" is an interface for visually displaying the selected fashion items and other suggested related items and cosmetics to the user.

[1307] The "means for analyzing the user's input information and behavior and recognizing emotions" is an emotion analysis system that analyzes the user's input data and behavior history and estimates the user's emotional state.

[1308] The "means for dynamically adjusting recommendation content based on recognized user emotions" is a system that adjusts recommendation content in real time based on analyzed emotional information.

[1309] "Means for analyzing user reactions to the recommended item in real time and dynamically adjusting the reason for recommendation based on the analysis results" refers to a system that collects and analyzes user reaction data in real time and dynamically changes the reason for recommendation based on the results.

[1310] A "generative AI model" is an artificial intelligence model that generates natural language and recognizes data patterns based on user information and behavioral data.

[1311] A "prompt" is an instruction or question that is input to a generative AI model to output recommendations.

[1312] System Overview

[1313] This invention is a system that recommends personalized fashion items based on a user's preferences and emotional state. The system is realized by combining four main functions: user information collection, recommendation function, boost function, and emotion engine.

[1314] Hardware and software used

[1315] Server: Used to analyze data and select recommended items. The main software includes Python natural language processing libraries (NLTK and spaCy), sentiment analysis libraries (TextBlob and VADER), and generative AI models (ChatGPT API).

[1316] Device: This serves as the interface where users can input information and view recommended items and descriptions. This is typically a web browser or a mobile app.

[1317] Database: Used to store user information. Popular databases include AWS RDS (Relational Database Service) and MySQL.

[1318] Processing Details

[1319] 1. Collection of User Information

[1320] When a user logs in to an e-commerce site, the device displays a questionnaire form and a profile entry page. The user enters information such as their preferred fashion style, body type, personal color, past purchase history, and the latest trends. This information is sent to the server and stored in a database such as AWS RDS or MySQL.

[1321] 2. Recommendation function

[1322] The device displays an image, hashtags, and a text input field to the user. The user uploads an image of their choice and enters text and hashtags. The server analyzes the received data using Python's natural language processing library (NLTK or spaCy) and generates a prompt for the ChatGPT API based on the analysis results. Based on the response from the ChatGPT API, the server selects appropriate fashion items and displays them on the device.

[1323] 3. Boost function

[1324] When a user selects a product from the recommendation list, the server generates specific reasons for recommending that product. For example, a reason might be provided such as, "This pastel pink blouse matches your personal color perfectly." Other related items (such as "white denim" or "light blue sneakers") are also suggested. This information is displayed on the device, allowing the user to visually understand the recommendation.

[1325] 4. Utilizing the Emotion Engine

[1326] The server analyzes the user's input information and behavior using emotion analysis libraries (TextBlob and VADER) to recognize the user's emotions. If the user is unsure about which product to choose, the server dynamically adjusts the emphasis of the recommended items. For example, it adds information about limited-time sales or special offers. The user's reactions are analyzed in real time, and if the user shows anxiety, for example, it changes the reason statement to emphasize positive feedback.

[1327] Specific examples

[1328] Combining recommendation functionality with an emotion engine

[1329] A user enters "casual" and "I like pastel colors" in their profile. They then write "A perfect outfit for a spring picnic" in the text field and add the hashtag "spring casual." The server analyzes the input data and selects "pastel pink blouse," "white denim," and "camera bag" as recommendations, which are displayed on the device. The server analyzes the user's reactions in real time, and if the user is interested in a recommended item but has concerns, it dynamically adjusts the reason for the recommendation to something like, "This pastel pink suits your personal color and is perfect for spring casual occasions."

[1330] Prompt Sentence Examples

[1331] "Suggest a casual outfit perfect for a spring picnic. Your user likes pastel pink and prefers casual styles."

[1332] This invention makes it possible to provide personalized fashion suggestions based on a user's preferences and emotions, improving the online shopping experience.

[1333] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1334] Step 1:

[1335] A user logs in to an e-commerce site.

[1336] Input: User ID and password

[1337] Operation: The terminal displays the login screen. The user enters their ID and password and presses the "Login" button.

[1338] Output: A session is started and the user's profile information is loaded.

[1339] Step 2:

[1340] The device will display a survey form and profile entry page.

[1341] Input: Display requirements for survey forms and profile entry pages

[1342] How it works: The device generates an input form with fields such as "preferred fashion style," "body type information," and "personal color."

[1343] Output: The form that is displayed to the user

[1344] Step 3:

[1345] The user enters information into a survey form or profile entry page and submits it.

[1346] Input: Information such as "preferred fashion style," "body type information," "personal color," "past purchase history," and "latest trends"

[1347] What happens: The user enters information into the fields and presses the "Submit" button.

[1348] Output: The entered user information is sent to the server.

[1349] Step 4:

[1350] The server stores the user information in a database.

[1351] Input: Information entered by the user

[1352] How it works: The server receives the data and stores it in a database such as AWS RDS or MySQL.

[1353] Output: User information stored in the database

[1354] Step 5:

[1355] The device displays an image, hashtags, and a text entry field to the user.

[1356] Input: Display requirements for images, hashtags, and text input fields

[1357] How it works: Your device will display options such as "Upload your favorite image," "Related hashtags," and "Describe your desired style."

[1358] Output: Input fields that are displayed to the user

[1359] Step 6:

[1360] The user enters an image, text, and hashtags.

[1361] Input: User-uploaded images, entered text, hashtags

[1362] How it works: User uploads an image, enters text and hashtags.

[1363] Output: The entered data is sent to the server.

[1364] Step 7:

[1365] The server parses the input data.

[1366] Input: Image, Text, Hashtag

[1367] How it works: The server parses the data using Python natural language processing libraries (NLTK or spaCy) and generates prompts to the ChatGPT API based on the results of the analysis.

[1368] Output: Generated prompt, response data from ChatGPT API

[1369] Step 8:

[1370] The server selects recommended items based on the analysis results.

[1371] Input: Response data from ChatGPT API, user profile information

[1372] Operation: The server uses the response data to create a list of fashion items suitable for the user.

[1373] Output: A list of selected recommended items

[1374] Step 9:

[1375] The device displays a recommendation list to the user.

[1376] Input: Recommended item list

[1377] What it does: Your device will display a list of items such as "Pastel Pink Blouse," "White Denim," and "Camera Bag."

[1378] Output: The recommendation list displayed to the user

[1379] Step 10:

[1380] The user selects the product of interest from the recommendation list.

[1381] Input: Products you are interested in from the recommendation list

[1382] Action: The user clicks on the selected item.

[1383] Output: Selected product information is sent to the server.

[1384] Step 11:

[1385] The server generates specific recommendation reasons.

[1386] Input: Selected product information, user profile information

[1387] How it works: The server generates a recommendation reason based on the user's profile and product attributes, for example, "This pastel pink blouse matches your personal color."

[1388] Output: Generated recommendation reason text

[1389] Step 12:

[1390] The server suggests other related items.

[1391] Input: Selected product information

[1392] What it does: The server lists items that go well with the selected item, such as "white denim" or "light blue sneakers."

[1393] Output: A list of suggested related items

[1394] Step 13:

[1395] The device will display the recommendation reason and related items to the user.

[1396] Input: Generated recommendation reason text, suggested related item list

[1397] What it does: The device displays this information to the user in a visually convincing way.

[1398] Output: Displayed recommendation reason and related items

[1399] Step 14:

[1400] The server analyzes the user's input information and behavior.

[1401] Input: User input data and behavior history

[1402] How it works: The server analyzes the data using a sentiment analysis library (TextBlob or VADER) to recognize the user's sentiment.

[1403] Output: Recognized emotion data

[1404] Step 15:

[1405] The server dynamically adjusts the recommendation content based on emotions.

[1406] Input: Recognized emotion data, recommended item list

[1407] How it works: If the server detects negative sentiment, it changes the emphasis of the recommended items and adds limited-time sales and special offers.

[1408] Output: Dynamically adjusted recommendations

[1409] Step 16:

[1410] The server analyzes user responses in real time and adjusts the wording of the recommendation reasons.

[1411] Input: User response data (viewing time, number of clicks, etc.)

[1412] How it works: The server analyzes the user's response and modifies the generated reason statement based on the results. For example, if the user is feeling anxious, it emphasizes more positive feedback.

[1413] Output: Adjusted recommendation reason

[1414] (Application example 2)

[1415] 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."

[1416] While conventional fashion recommendation systems suggest personalized items to users when shopping online, they do not address the in-store shopping experience. Furthermore, they do not provide recommendations that take into account the user's real-time movements and emotions, which prevents them from fully enhancing user satisfaction. Furthermore, there is a need for a system that effectively utilizes devices such as eye-tracking technology, smart glasses, and head-mounted displays to enable users to obtain the information they want in-store in real time.

[1417] 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.

[1418] In this invention, the server includes an eye tracking unit that identifies a specific product when the user moves around the store and directs their gaze at the product, a unit that dynamically adjusts recommendations for the identified product according to the user's emotional state, and a unit that displays the generated recommendation reasons and recommendation content on smart glasses or a head-mounted display, thereby improving the shopping experience in a physical store and making it possible to provide appropriate information and recommendations in real time for products that interest the user.

[1419] "Means for collecting user information" refers to a function for inputting and recording user preferences, body type, personal color, past purchase history, the latest trends, and the like.

[1420] The "means for users to input their favorite images, hashtags, and text" refers to an interface that allows users to input images, hashtags, text, etc. that specifically indicate their own fashion style and preferences.

[1421] "Means for analyzing user information and the images, hashtags, and text" refers to a function for analyzing collected user information and input images, hashtags, and text to understand user preferences and trends.

[1422] "Means for selecting and recommending personalized fashion items" refers to a system that selects and suggests the most suitable fashion items for users based on the analysis results.

[1423] The "means for generating specific reasons for recommendation" is a function for automatically generating detailed explanations and reasons why recommended fashion items are preferred.

[1424] The "means for suggesting other items and cosmetics" is a function for automatically suggesting other fashion items and cosmetics that can be easily combined with the recommended fashion items.

[1425] The "means for displaying to the user" is an interface for visually presenting information about the recommended fashion items and other suggested items and cosmetics to the user.

[1426] The "eye tracking means" is a technology for identifying and tracking a particular product when the user directs their gaze at the product.

[1427] "Dynamic adjustment means" is a function that allows recommendations to be changed and adjusted in real time according to the user's emotions and behavior.

[1428] "Means for displaying on smart glasses or a head-mounted display" refers to a function for displaying the generated recommendation reasons and recommendation content on a smart device in real time.

[1429] Specific configurations and methods for implementing the present invention are described below.

[1430] 1. Collection of User Information

[1431] The terminal uses smart glasses and mobile devices to collect user information. Users use these devices to input their preferred style, body type, personal color, past purchase history, and the latest trend information. This information is sent to a server and stored in a database.

[1432] 2. Eye tracking and product identification

[1433] The smart glasses use built-in eye-tracking technology (e.g., Tobii Eye Tracker SDK) to identify the products the user is looking at in the store, recognize the products the user is looking at, and send the related information to a server.

[1434] 3. Recommendation Generation and Sentiment Analysis

[1435] The server generates personalized recommendations based on the aforementioned user information and eye-tracking information. This includes suggesting fashion items based on the user's input and profile using a generative AI model (e.g., OpenAI API). It also uses an emotion analysis engine to analyze the user's emotional state from their facial expressions and behavior and dynamically adjust the recommendations.

[1436] 4. Displaying recommendation reasons and dynamic adjustment

[1437] The device displays the recommendation content and specific reasons for the recommendation in real time on smart glasses or a head-mounted display. Based on the results of the sentiment analysis engine, the device adjusts the recommendation reasons and the display of recommended items. For example, if the user is interested but anxious, it will emphasize positive feedback.

[1438] Specific examples

[1439] For example, if a user wears smart glasses and enters a store and looks at a particular product, fashion items related to that product will be automatically recommended. The recommendation will be displayed on the visual device along with the reason for the recommendation generated by the server based on the user's information. This allows users to enjoy shopping while receiving detailed information and personalized recommendations.

[1440] Prompt Sentence Examples

[1441] An example of a prompt generated by the server is:

[1442] markdown

[1443] User Profile:

[1444] Favorite style: Casual, pastel colors

[1445] Body type: Medium build

[1446] Past purchases: Jeans, blouses, casual shoes

[1447] User input:

[1448] Tags: Spring Casual

[1449] Keywords: Spring Picnic

[1450] Recommended Items:

[1451] pastel pink blouse

[1452] White denim

[1453] camera bag

[1454] Recommended reasons include:

[1455] This pastel pink blouse is perfect for your personal color and is perfect for casual spring occasions!

[1456] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1457] Step 1:

[1458] The user logs in using smart glasses or a mobile device. After logging in, the user enters their preferred style, body type, personal color, past purchase history, latest trend information, etc. This information is sent to the server and stored in a database.

[1459] Input: User profile information (style, body type, color, purchase history, trend information)

[1460] Output: User information stored in the database

[1461] Step 2:

[1462] As a user moves around the store, the smart glasses' eye-tracking technology (Tobii Eye Tracker SDK) is used to identify the products they are looking at. The glasses then transmit the user's gaze data to a server.

[1463] Input: User gaze data

[1464] Output: Identified product information

[1465] Step 3:

[1466] The server analyzes the user's gaze data and saved profile information to generate personalized recommendations for the user. A generative AI model (OpenAI API) is used to suggest fashion items based on the user's prompts.

[1467] Input: Identified product information, user profile information

[1468] Output: A list of recommended fashion items

[1469] Step 4:

[1470] The server inputs the user's prompt information into the generative AI model to generate specific recommendation reasons, including specific reasons and explanations for the new recommendation.

[1471] Input: Identified product information, user profile information, prompt information

[1472] Output: Specific reasons for recommendation

[1473] Step 5:

[1474] The server uses an emotion analysis engine to analyze the user's emotional state based on their facial expressions and behavior. Based on the user's emotions, the recommendation content is dynamically adjusted. For example, if the user is interested but anxious, the server emphasizes positive feedback.

[1475] Input: gaze data, facial expression data, behavior data

[1476] Output: Adjusted recommendation content

[1477] Step 6:

[1478] The device (smart glasses or head-mounted display) displays the generated recommendation content and specific reasons for the recommendation in real time. The user can consider products based on this information and obtain additional information as needed.

[1479] Input: Adjusted recommendation content, specific reasons for recommendation

[1480] Output: Present information to the user

[1481] Step 7:

[1482] The user selects recommended fashion items or other suggested items through their device. Detailed information about the selected item and related items are displayed. The device then sends the information back to the server, which generates additional recommendations as needed.

[1483] Input: User selection information

[1484] Output: Show detailed information and related items

[1485] Through these steps, users can enjoy a personalized shopping experience in a physical store.

[1486] 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.

[1487] 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.

[1488] 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.

[1489] [Fourth embodiment]

[1490] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1491] 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.

[1492] 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).

[1493] 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.

[1494] 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.

[1495] 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).

[1496] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1497] 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.

[1498] 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.

[1499] 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.

[1500] 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.

[1501] 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.

[1502] 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."

[1503] Natural language description of the program

[1504] The system is programmed using the following process:

[1505] 1. Collection of User Information

[1506] A session begins when a user logs in to an EC site.

[1507] The device displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[1508] The server stores this information in a database.

[1509] 2. Recommendation function

[1510] The device displays an image, hashtag, and text input field to the user, and the user enters their favorite image, hashtag, or text.

[1511] The server analyzes this input data and combines it with the user's profile information to interpret it. The analysis uses natural language processing technology, particularly the ChatGPT API, to make abstract images concrete.

[1512] The server selects fashion items that suit the user based on the analysis results.

[1513] The device displays a list of recommendations to the user, including a selection of items such as a pastel pink blouse, white denim, and a camera bag.

[1514] 3. Boost function

[1515] The user selects the product of interest from the recommendation list.

[1516] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[1517] The server will suggest other items or cosmetics that go well with the product (such as "white denim" or "light blue sneakers").

[1518] The device displays this information to the user, making it visually convincing.

[1519] Specific examples

[1520] Example 1: Recommendation function

[1521] A user enters "casual" and "likes pastel colors" in their profile.

[1522] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[1523] The server analyzes the input data and retrieves the appropriate items via the ChatGPT API.

[1524] The server selects "pastel pink blouse," "white denim," and "camera bag" as recommendations and displays them on the device.

[1525] Example 2: Boost function

[1526] The user selects "Pastel Pink Blouse" from the recommendations presented.

[1527] The server generates reasons for recommending this item, explaining, for example, "This pastel pink suits your personal color and is perfect for casual spring occasions."

[1528] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[1529] The terminal displays these coordination suggestions to the user and supports overall styling.

[1530] As a result, the present invention provides users with personalized fashion suggestions and a sense of security when shopping online.

[1531] The processing flow will be explained below.

[1532] Step 1:

[1533] A user logs in to an e-commerce site.

[1534] Step 2:

[1535] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters preferences, body type information, personal color, past purchase history, the latest trends, etc.

[1536] Step 3:

[1537] The server receives the user's input and stores it in a database.

[1538] Step 4:

[1539] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[1540] Step 5:

[1541] The server analyzes the user's input data, which involves applying natural language processing based on the user's profile information and sending a request to the ChatGPT API to materialize the abstract image.

[1542] Step 6:

[1543] The server analyzes the response from the ChatGPT API and selects the appropriate fashion items for the user.

[1544] Step 7:

[1545] The server generates a list of selected fashion items (e.g., "pastel pink blouse," "white denim," "camera bag," etc.) and stores it in a database.

[1546] Step 8:

[1547] The device displays the recommendation list to the user.

[1548] Step 9:

[1549] The user selects the product of interest from the recommendation list.

[1550] Step 10:

[1551] The server generates specific reasons for recommending the selected products, such as "This pastel pink is a perfect match for your personal color and also goes well with a casual spring vibe."

[1552] Step 11:

[1553] The server will suggest other items (e.g., "white denim" or "light blue sneakers") and cosmetics that go well with the selected product.

[1554] Step 12:

[1555] The terminal displays this information to the user, allowing them to understand it visually.

[1556] By following the above steps, users can confidently select fashion items that suit them, reducing the chances of making mistakes when shopping online.

[1557] Example 1

[1558] 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."

[1559] In conventional online shopping, it is difficult for users to find products that suit them. Especially when it comes to fashion items, recommendations do not sufficiently take into account the user's individual preferences, body type, personal color, etc. Furthermore, since there is no specific reason for the recommended product, users lack a sense of security when making a purchase decision. Furthermore, suggestions for coordinating the recommended product with other items are also insufficient. This leads to problems such as low user satisfaction and a decrease in purchasing motivation.

[1560] 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.

[1561] In this invention, the server includes means for collecting user information, means for users to input their favorite images, hashtags, and text, means for analyzing the user information and the images, hashtags, and text, means for selecting and recommending personalized products based on the analysis results, means for generating specific reasons for recommending the recommended products, means for suggesting other items that are easy to combine with the recommended products, and means for displaying the recommended products and other suggested items to the user. This allows the server to provide users with personalized fashion item recommendations and specific reasons for the recommendations, and can also provide coordination suggestions. This allows users to easily find fashion items that suit them and make purchasing decisions with confidence.

[1562] "User information" refers to information such as the user's individual preferences, body type, personal color, past purchase history, and the latest trends.

[1563] "Image" refers to an image that visually represents a style or design that a user likes.

[1564] A "hashtag" is a symbolic word that users use to identify keywords or themes of interest or concern.

[1565] "Text" refers to a data format in which a user inputs their opinions or wishes as text.

[1566] "Analysis" refers to the process of analyzing user preferences and trends based on user information, images, hashtags, and text.

[1567] "Personalized products" refer to products that are optimized to a user's individual preferences and needs based on analysis results.

[1568] "Recommendation" refers to the act of suggesting to a user that they purchase a particular product.

[1569] "Reasons for recommendation" refers to text that explains the specific appeal of the recommended product and its suitability for the user.

[1570] "Other Items" refers to additional products that can be used in conjunction with the recommended product to provide an enhanced experience.

[1571] "Display" refers to the act of visually providing information to a user via a terminal.

[1572] This invention is implemented by a system including the following components and processes: This system collects user information, recommends personalized products based on the user's preferences and needs, and also suggests specific reasons for recommending the products and related items.

[1573] Hardware and Software

[1574] This system mainly uses the following hardware and software:

[1575] Device: The device used by the user to access the site (e.g., computer, smartphone, tablet)

[1576] Server: A server system for collecting, storing, and analyzing data.

[1577] Database: A database system that stores user information and recommendation information

[1578] Generative AI models: APIs for using natural language processing techniques (e.g., ChatGPT API)

[1579] What the program does

[1580] The process is based on the following steps:

[1581] 1. User Information Collection:

[1582] A session begins when a user logs in to an EC site.

[1583] The device displays a login page and the user enters their existing account information.

[1584] The server authenticates the user's login information and begins the session.

[1585] 2. Displaying survey forms and profile entry pages:

[1586] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[1587] 3. Saving User Information:

[1588] The server stores the information entered by the user in a database in real time.

[1589] 4. Providing recommendation features:

[1590] The device displays an image, hashtags, and a text entry field to the user.

[1591] The user enters an image, hashtags, and text.

[1592] The server receives the input data and analyzes it using a generative AI model.

[1593] Based on the analysis results, the server selects personalized products and generates a recommendation list.

[1594] The device displays the recommendation list to the user.

[1595] 5. Providing boosting features:

[1596] The user selects the product of interest from the recommendation list.

[1597] The server generates specific reasons for recommending the selected product.

[1598] The server will suggest related items (e.g., "white denim" or "light blue sneakers").

[1599] The device will display the recommendation reason and related items to the user.

[1600] Specific examples

[1601] Here are some examples and prompts:

[1602] Example 1: Recommendation function

[1603] A user enters "casual" and "likes pastel colors" in their profile.

[1604] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[1605] The server analyzes the input data and obtains suitable items via a generative AI model.

[1606] The server selects "pastel pink blouse," "white denim," and "camera bag" from the recommendation list and displays them on the device.

[1607] Example 2: Boost function

[1608] The user selects "Pastel Pink Blouse" from the recommendations presented.

[1609] The server generates reasons for recommending this item, explaining, for example, "This pastel pink suits your personal color and is perfect for casual spring occasions."

[1610] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[1611] The terminal displays these coordination suggestions to the user and supports overall styling.

[1612] Example prompt sentence:

[1613] "What casual fashion items would be perfect for a spring picnic? Our users love pastel colors and casual styles."

[1614] As a result, the system of the present invention enables personalized recommendations of fashion items to users, along with specific reasons for the recommendations and coordination suggestions, thereby improving the sense of security and satisfaction of online shopping.

[1615] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1616] Step 1: User logs in to the e-commerce site

[1617] The user accesses the EC site and logs in with an existing account. Authentication is performed using the ID and password entered by the user.

[1618] The server checks the user's credentials against a database and starts a session. The input is the user's credentials and the output is a session ID.

[1619] Step 2: Displaying the survey form and profile entry page

[1620] The terminal displays a questionnaire form or profile entry page to the logged-in user.

[1621] The user enters information into the form, such as their preferences, body type, personal color, past purchase history, the latest trends, etc. The input is the user's personal information, and the output is the user's response data.

[1622] Step 3: Save user information

[1623] The server receives the information entered by the user and stores it in a database.

[1624] This data is used for subsequent analysis and recommendations. The input is the user's personal information data, and the output is a user profile stored in a database.

[1625] Step 4: Displaying an image, hashtags, and text input fields

[1626] The device displays a recommendation page, providing the user with an image, hashtags, and a text input field.

[1627] The user enters their desired image, hashtag, or text in the fields provided. The input is the user's input data (image, hashtag, text), and the output is the input data sent to the server.

[1628] Step 5: Collecting User Input Data

[1629] The server receives the images, hashtags, and text sent by the user and stores them in a database. The input is the user's input data, and the output is the data stored in the database for analysis.

[1630] Step 6: Data analysis and item selection

[1631] The server retrieves the user's profile information and newly entered data from the database and analyzes it using a generative AI model. Specifically, it uses the ChatGPT API to perform natural language processing and abstractly understand the user's preferences.

[1632] As a result of the analysis, the system selects the most suitable product for the user. The input is the profile data and the user's input data, and the output is a list of recommended products.

[1633] Step 7: Generate and display the recommendation list

[1634] Based on the analysis results, the server lists the most suitable products for the user.

[1635] The terminal visually displays this recommendation list to the user. The input is a list of recommended products, and the output is the recommendation list displayed to the user.

[1636] Step 8: Product Selection and Recommendation Generation

[1637] The user selects the product of interest from the recommendation list.

[1638] The server uses a generative AI model to generate specific reasons for recommending the selected product. For example, it generates an explanation such as, "This pastel pink blouse is perfect for your personal color and goes well with a casual spring look." The input is the selected product information, and the output is the reason for the recommendation.

[1639] Step 9: Suggest other items

[1640] The server will suggest other items that go well with the selected item (e.g., "white denim" or "light blue sneakers")

[1641] The server selects related items and sends them to the terminal along with a recommendation list. The input is the selected product information, and the output is a list of suggested items.

[1642] Step 10: View the proposal

[1643] The device displays the recommended product, the reason for the recommendation, and related items to the user.

[1644] The user reviews the overall styling and makes a purchasing decision. The input is data from the server and the output is the information displayed to the user.

[1645] This allows the system to suggest personalized fashion items to users, improving their online shopping experience.

[1646] (Application example 1)

[1647] 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."

[1648] The user experience in online shopping lacks visual information compared to shopping in a physical store, making it difficult to provide personalized product suggestions and encourage purchases. Finding products that match a user's preferences, body type, and personal color is particularly difficult when it comes to fashion items, which can cause users to hesitate before making a purchase. Therefore, there is a demand for systems that provide a more interactive and personalized shopping experience.

[1649] 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.

[1650] In this invention, the server includes means for collecting user information, means for inputting user-preferred images, hashtags, and text, means for analyzing the user information and the images, hashtags, and text, means for selecting and recommending personalized fashion items based on the analysis results, means for generating specific recommendation reasons for the recommended fashion items, means for suggesting other items and cosmetics that are easily combined with the fashion items, means for displaying the recommended fashion items and the suggested other items and cosmetics to the user, and means for visually displaying the recommended fashion items using smart glasses, thereby enabling users to enjoy a more interactive and personalized shopping experience and select and purchase products based on visual information.

[1651] "User information" refers to information such as the user's preferences, body type, personal color, past purchase history, and the latest trends.

[1652] "Images" are images or visual content that a user visually enjoys.

[1653] A "hashtag" is a symbol used to promote information sharing among people with the same hobbies or interests by tagging specific keywords or phrases with "."

[1654] "Text" refers to sentences or character information entered by the user.

[1655] "Analysis Methods" refers to algorithms and technologies used to process user information, images, hashtags, and text to understand and analyze their content.

[1656] "Personalized fashion items" refer to clothing and accessories that are identified and suggested to a user based on their individual information and preferences.

[1657] A "recommendation means" is a system or method that recommends specific fashion items to users based on the analysis results.

[1658] "Specific reasons for recommendation" refers to a detailed explanation of the recommended fashion item and the reasons for its recommendation.

[1659] "Means for suggesting other items or cosmetics" refers to a method or system for presenting other clothing or cosmetics that go well with the recommended fashion item to the user.

[1660] "Visual display means" refers to a method of visually presenting personalized products or offers to a user using a device such as smart glasses.

[1661] "Generative AI model" refers to artificial intelligence technology for generating and analyzing information based on user input and prompts.

[1662] The present invention provides a system for providing personalized fashion item recommendations to users using smart glasses. A specific implementation method thereof will be described below.

[1663] Program processing flow

[1664] 1. Collection of User Information

[1665] When a user logs in to the e-commerce site, the server begins collecting user information. The terminal displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends. This information is sent to the server and stored in a database.

[1666] 2. Implementation of recommendation function

[1667] When a user logs in, the device displays an image, hashtags, and text input field. The user then enters their desired image, hashtag, or text. The server analyzes this input data and interprets it in combination with the user's profile information. The analysis uses natural language processing technology, particularly the ChatGPT API, a generative AI model, to make abstract images concrete.

[1668] 3. Selection of recommended items

[1669] Based on the analysis results, the server selects fashion items that suit the user. Specifically, it also generates personalized reasons for the recommendation. The device then displays a recommendation list to the user. The recommendation list includes the selected items, such as a pastel pink blouse, white denim, and a camera bag.

[1670] 4. Visual display using smart glasses

[1671] The recommended fashion items are displayed visually to the user using the smart glasses, allowing the user to view product details and facilitate purchasing decisions.

[1672] Specific examples

[1673] Example 1: Recommendation function

[1674] A user enters "casual" and "likes pastel colors" in their profile. They then enter "a perfect outfit for a spring picnic" in the text field and add the hashtag "spring casual." The server analyzes this input data and retrieves suitable items via the ChatGPT API of the generative AI model. As a result, "pastel pink blouse," "white denim," and "camera bag" are displayed in the recommendation list.

[1675] Example 2: Boost function

[1676] The user selects a "pastel pink blouse" from the recommendations presented. The server generates a reason for recommending this item, explaining that "this pastel pink suits your personal color and is perfect for spring casual occasions." Furthermore, other items that go well with the blouse, such as "white denim" and "light blue sneakers," are also suggested. The user can visually check these coordination suggestions through the smart glasses, receiving support for overall styling.

[1677] Specific hardware and software used

[1678] Hardware: Smart glasses, user devices (smartphones and PCs)

[1679] Software: E-commerce website platform, generative AI model (ChatGPT API), database

[1680] Examples of prompt statements

[1681] User profile: {'Preferences': 'Casual', 'Favorite color': 'Pastel'}

[1682] User Input: Perfect outfit for a spring picnic Spring Casual

[1683] Recommended items:

[1684] In this way, the system can provide users with a more personalized and visually appealing shopping experience.

[1685] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1686] Step 1:

[1687] Collection of User Information

[1688] Users log in to the e-commerce site using a terminal. The terminal displays a questionnaire form and a profile entry page, where users enter their preferences, body type, personal color, past purchase history, the latest trends, etc. This information is sent to the server and stored in a database.

[1689] Input: User login information, survey responses, profile information

[1690] Output: Saved user information (database)

[1691] Step 2:

[1692] User image, hashtag and text input

[1693] The device displays an image, hashtags, and text input fields to the user, who then enters the image, hashtag, or text of their choice, which is then sent to the server.

[1694] Input: Images, hashtags, and text entered by the user

[1695] Output: Data entered by the user

[1696] Step 3:

[1697] Data analysis

[1698] The server analyzes the collected user information and the input image, hashtag, and text. It uses the ChatGPT API, a generative AI model, to generate prompts. Specifically, it converts abstract images into concrete fashion items based on the user's profile information and text input.

[1699] Input: User information, image, hashtag, text, prompt

[1700] Output: A list of fashion items as the analysis result

[1701] Step 4:

[1702] Personalized fashion item selection and recommendations

[1703] Based on the analysis results, the server selects and recommends fashion items that suit the user. Specific reasons for recommending the items are also provided. The list of selected items is then sent to the device.

[1704] Input: Analysis results, user information, prompt text

[1705] Output: A list of selected fashion items and reasons for their recommendation

[1706] Step 5:

[1707] Preparing for visual display

[1708] The device receives the list of recommended fashion items and the reasons for the recommendations, and prepares them for display on the smart glasses.

[1709] Input: List of selected fashion items and reasons for recommendation

[1710] Output: Data ready for display

[1711] Step 6:

[1712] Visual display with smart glasses

[1713] By using the smart glasses, users can visually see the recommended fashion items and the reasons for their recommendations, allowing them to view product details and make purchasing decisions more easily.

[1714] Input: Data ready to be displayed

[1715] Output: Visual product display via smart glasses

[1716] 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.

[1717] Natural language description of the program

[1718] The system is programmed using the following process:

[1719] 1. Collection of User Information

[1720] A session begins when a user logs in to an EC site.

[1721] The device displays a questionnaire form and a profile entry page to the user, and the user enters information such as their preferences, body type, personal color, past purchase history, and the latest trends.

[1722] The server stores this information in a database.

[1723] 2. Recommendation function

[1724] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[1725] The server analyzes these inputs, which involves applying natural language processing based on the user's profile information and sending requests to the ChatGPT API to turn the abstract image into a concrete image.

[1726] Based on the analysis results, the server selects fashion items suitable for the user.

[1727] The device displays a list of recommendations to the user, including a selection of items such as a pastel pink blouse, white denim, and a camera bag.

[1728] 3. Boost function

[1729] The user selects the product of interest from the recommendation list.

[1730] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[1731] The server will suggest other items or cosmetics that go well with the product (such as "white denim" or "light blue sneakers").

[1732] The device displays this information to the user, making it visually convincing.

[1733] 4. Utilizing the Emotion Engine

[1734] The server analyzes the user's input information and selection behavior, and recognizes the user's emotions using an emotion engine.

[1735] The server dynamically adjusts the recommendation content based on the user's emotions. For example, if the user is unsure about which product to choose, the server may change the emphasis on the recommended item or suggest additional similar products.

[1736] The server analyzes the user's reactions in real time and adjusts the reason for the recommendation accordingly. For example, if the user is interested but anxious, the reason will be changed to emphasize positive feedback.

[1737] Specific examples

[1738] Example 1: Combining recommendation features with an emotion engine

[1739] A user enters "casual" and "likes pastel colors" in their profile.

[1740] A user writes "Perfect outfit for a spring picnic" in the text field and adds the hashtag "spring casual."

[1741] The server analyzes the input data and retrieves the appropriate items via the ChatGPT API.

[1742] The server selects "pastel pink blouse," "white denim," and "camera bag" as recommendations and displays them on the device.

[1743] The server analyzes the user's reactions in real time and recognizes that the user is interested in the recommended item but is also feeling anxious.

[1744] The server adjusts the recommendation statement to read, "This pastel pink suits your personal color and is perfect for casual spring occasions. Many users also rate this item highly."

[1745] Example 2: Combining boosting and emotion engines

[1746] The user selects "Pastel Pink Blouse" from the presented recommendation list.

[1747] The server generates reasons for recommending this item, explaining, for example, "This pastel pink is a perfect match for your personal color and also goes well with a casual spring vibe."

[1748] The server will also suggest items that go well with the blouse, such as white denim and light blue sneakers.

[1749] The server analyzes user responses and checks to see which ones are more positive.

[1750] The server displays the original reasons for the recommendation and adds suggestions to encourage additional purchases (e.g., "This outfit is eligible for a 30% off campaign!").

[1751] The device displays this information to the user and supports overall styling.

[1752] As a result, the present invention not only provides users with personalized fashion suggestions and a sense of security when shopping online, but also uses an emotion engine to achieve even more accurate recommendations and support.

[1753] The processing flow will be explained below.

[1754] Step 1:

[1755] A user logs in to an e-commerce site and a session begins.

[1756] Step 2:

[1757] The terminal displays a questionnaire form and a profile entry page to the user, and the user enters their preferences, body type information, personal color, past purchase history, the latest trend information, etc.

[1758] Step 3:

[1759] The server receives the user's input and stores it in a database.

[1760] Step 4:

[1761] The device displays an image, hashtag, and text input field to the user, and the user can enter their favorite image, hashtag, or text.

[1762] Step 5:

[1763] The server analyzes the user's input data, which involves applying natural language processing based on the user's profile information and sending a request to the ChatGPT API to materialize the abstract image.

[1764] Step 6:

[1765] The server analyzes the response from the ChatGPT API and selects the appropriate fashion items for the user.

[1766] Step 7:

[1767] The server generates a list of selected fashion items and stores it in a database.

[1768] Step 8:

[1769] The device displays the recommendation list to the user.

[1770] Step 9:

[1771] The user selects the product of interest from the recommendation list.

[1772] Step 10:

[1773] The server generates specific reasons for recommending selected products, such as "This pastel pink blouse matches your personal color perfectly and goes well with a casual spring look."

[1774] Step 11:

[1775] The server will suggest other items (e.g., "white denim" or "light blue sneakers") and cosmetics that go well with the selected product.

[1776] Step 12:

[1777] The terminal displays this information to the user, allowing them to understand it visually.

[1778] Step 13:

[1779] The server analyzes the user's input information and selection behavior, and recognizes the user's emotions using an emotion engine.

[1780] Step 14:

[1781] The server dynamically adjusts the recommendation content based on the user's emotions. For example, if the user is unsure about which product to choose, the server may change the emphasis on the recommended item or suggest additional similar products.

[1782] Step 15:

[1783] The server analyzes the user's reactions in real time and adjusts the reason for the recommendation accordingly. For example, if the user is interested but anxious, the reason will be changed to emphasize positive feedback.

[1784] Through the above steps, users can confidently choose fashion items that suit them, and by using the emotion engine, even more accurate recommendations and support can be achieved.

[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 robot 414 will be referred to as a "terminal."

[1787] Conventional online fashion recommendation systems do not fully consider users' subjective preferences and emotions when making recommendations, making it difficult to provide products that truly satisfy users. Furthermore, conventional systems cannot respond to changes in users' emotions, and therefore do not sufficiently motivate users to purchase additional products that they are temporarily interested in.

[1788] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting user information; means for inputting a user's favorite images, hashtags, and text; means for analyzing the user information and the images, hashtags, and text; means for selecting and recommending personalized fashion items based on the analysis results; means for generating specific recommendation reasons for the recommended fashion items; means for suggesting other items and cosmetics that are easily combined with the fashion items; means for displaying the recommended fashion items and the suggested other items and cosmetics to the user; means for analyzing the user's input information and behavior and recognizing emotions; means for dynamically adjusting the recommendation content based on the recognized user emotions; and means for analyzing the user's reaction to the recommended items in real time and dynamically adjusting the recommendation reasons in accordance with the analysis results. This enables personalized recommendations of fashion items based on the user's preferences and emotions, and dynamic adjustment of recommendations in accordance with the user's emotions.

[1789] "Means for collecting user information" refers to the interface and data collection process for collecting information such as the user's preferences, body type, personal color, past purchase history, and the latest trends.

[1790] A "means for users to input their favorite images, hashtags, and text" is an interface that provides forms or input fields for users to enter information about their preferences and style in visual or textual form.

[1791] The "means for analyzing the user information and the images, hashtags, and text" refers to a system that analyzes collected user information and input data using natural language processing and image analysis algorithms.

[1792] The "means for selecting and recommending personalized fashion items based on analysis results" is a system that selects fashion items suitable for the user from the analysis results and provides recommendation information on them.

[1793] The "means for generating a specific reason for recommendation for the recommended fashion item" is a system that generates a specific reason for recommendation related to the selected fashion item and provides it to the user.

[1794] The "means for suggesting other items and cosmetics that can be easily combined with the fashion item" is a system that suggests to the user other related items and cosmetics that can be easily combined with the recommended fashion item.

[1795] The "means for displaying the recommended fashion items and other suggested items and cosmetics to the user" is an interface for visually displaying the selected fashion items and other suggested related items and cosmetics to the user.

[1796] The "means for analyzing the user's input information and behavior and recognizing emotions" is an emotion analysis system that analyzes the user's input data and behavior history and estimates the user's emotional state.

[1797] The "means for dynamically adjusting recommendation content based on recognized user emotions" is a system that adjusts recommendation content in real time based on analyzed emotional information.

[1798] "Means for analyzing user reactions to the recommended item in real time and dynamically adjusting the reason for recommendation based on the analysis results" refers to a system that collects and analyzes user reaction data in real time and dynamically changes the reason for recommendation based on the results.

[1799] A "generative AI model" is an artificial intelligence model that generates natural language and recognizes data patterns based on user information and behavioral data.

[1800] A "prompt" is an instruction or question that is input to a generative AI model to output recommendations.

[1801] System Overview

[1802] This invention is a system that recommends personalized fashion items based on a user's preferences and emotional state. The system is realized by combining four main functions: user information collection, recommendation function, boost function, and emotion engine.

[1803] Hardware and software used

[1804] Server: Used to analyze data and select recommended items. The main software includes Python natural language processing libraries (NLTK and spaCy), sentiment analysis libraries (TextBlob and VADER), and generative AI models (ChatGPT API).

[1805] Device: This serves as the interface where users can input information and view recommended items and descriptions. This is typically a web browser or a mobile app.

[1806] Database: Used to store user information. Popular databases include AWS RDS (Relational Database Service) and MySQL.

[1807] Processing Details

[1808] 1. Collection of User Information

[1809] When a user logs in to an e-commerce site, the device displays a questionnaire form and a profile entry page. The user enters information such as their preferred fashion style, body type, personal color, past purchase history, and the latest trends. This information is sent to the server and stored in a database such as AWS RDS or MySQL.

[1810] 2. Recommendation function

[1811] The device displays an image, hashtags, and a text input field to the user. The user uploads an image of their choice and enters text and hashtags. The server analyzes the received data using Python's natural language processing library (NLTK or spaCy) and generates a prompt for the ChatGPT API based on the analysis results. Based on the response from the ChatGPT API, the server selects appropriate fashion items and displays them on the device.

[1812] 3. Boost function

[1813] When a user selects a product from the recommendation list, the server generates specific reasons for recommending that product. For example, a reason might be provided such as, "This pastel pink blouse matches your personal color perfectly." Other related items (such as "white denim" or "light blue sneakers") are also suggested. This information is displayed on the device, allowing the user to visually understand the recommendation.

[1814] 4. Utilizing the Emotion Engine

[1815] The server analyzes the user's input information and behavior using emotion analysis libraries (TextBlob and VADER) to recognize the user's emotions. If the user is unsure about which product to choose, the server dynamically adjusts the emphasis of the recommended items. For example, it adds information about limited-time sales or special offers. The user's reactions are analyzed in real time, and if the user shows anxiety, for example, it changes the reason statement to emphasize positive feedback.

[1816] Specific examples

[1817] Combining recommendation functionality with an emotion engine

[1818] A user enters "casual" and "I like pastel colors" in their profile. They then write "A perfect outfit for a spring picnic" in the text field and add the hashtag "spring casual." The server analyzes the input data and selects "pastel pink blouse," "white denim," and "camera bag" as recommendations, which are displayed on the device. The server analyzes the user's reactions in real time, and if the user is interested in a recommended item but has concerns, it dynamically adjusts the reason for the recommendation to something like, "This pastel pink suits your personal color and is perfect for spring casual occasions."

[1819] Prompt Sentence Examples

[1820] "Suggest a casual outfit perfect for a spring picnic. Your user likes pastel pink and prefers casual styles."

[1821] This invention makes it possible to provide personalized fashion suggestions based on a user's preferences and emotions, improving the online shopping experience.

[1822] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1823] Step 1:

[1824] A user logs in to an e-commerce site.

[1825] Input: User ID and password

[1826] Operation: The terminal displays the login screen. The user enters their ID and password and presses the "Login" button.

[1827] Output: A session is started and the user's profile information is loaded.

[1828] Step 2:

[1829] The device will display a survey form and profile entry page.

[1830] Input: Display requirements for survey forms and profile entry pages

[1831] How it works: The device generates an input form with fields such as "preferred fashion style," "body type information," and "personal color."

[1832] Output: The form that is displayed to the user

[1833] Step 3:

[1834] The user enters information into a survey form or profile entry page and submits it.

[1835] Input: Information such as "preferred fashion style," "body type information," "personal color," "past purchase history," and "latest trends"

[1836] What happens: The user enters information into the fields and presses the "Submit" button.

[1837] Output: The entered user information is sent to the server.

[1838] Step 4:

[1839] The server stores the user information in a database.

[1840] Input: Information entered by the user

[1841] How it works: The server receives the data and stores it in a database such as AWS RDS or MySQL.

[1842] Output: User information stored in the database

[1843] Step 5:

[1844] The device displays an image, hashtags, and a text entry field to the user.

[1845] Input: Display requirements for images, hashtags, and text input fields

[1846] How it works: Your device will display options such as "Upload your favorite image," "Related hashtags," and "Describe your desired style."

[1847] Output: Input fields that are displayed to the user

[1848] Step 6:

[1849] The user enters an image, text, and hashtags.

[1850] Input: User-uploaded images, entered text, hashtags

[1851] How it works: User uploads an image, enters text and hashtags.

[1852] Output: The entered data is sent to the server.

[1853] Step 7:

[1854] The server parses the input data.

[1855] Input: Image, Text, Hashtag

[1856] How it works: The server parses the data using Python natural language processing libraries (NLTK or spaCy) and generates prompts to the ChatGPT API based on the results of the analysis.

[1857] Output: Generated prompt, response data from ChatGPT API

[1858] Step 8:

[1859] The server selects recommended items based on the analysis results.

[1860] Input: Response data from ChatGPT API, user profile information

[1861] Operation: The server uses the response data to create a list of fashion items suitable for the user.

[1862] Output: A list of selected recommended items

[1863] Step 9:

[1864] The device displays a recommendation list to the user.

[1865] Input: Recommended item list

[1866] What it does: Your device will display a list of items such as "Pastel Pink Blouse," "White Denim," and "Camera Bag."

[1867] Output: The recommendation list displayed to the user

[1868] Step 10:

[1869] The user selects the product of interest from the recommendation list.

[1870] Input: Products you are interested in from the recommendation list

[1871] Action: The user clicks on the selected item.

[1872] Output: Selected product information is sent to the server.

[1873] Step 11:

[1874] The server generates specific recommendation reasons.

[1875] Input: Selected product information, user profile information

[1876] How it works: The server generates a recommendation reason based on the user's profile and product attributes, for example, "This pastel pink blouse matches your personal color."

[1877] Output: Generated recommendation reason text

[1878] Step 12:

[1879] The server suggests other related items.

[1880] Input: Selected product information

[1881] What it does: The server lists items that go well with the selected item, such as "white denim" or "light blue sneakers."

[1882] Output: A list of suggested related items

[1883] Step 13:

[1884] The device will display the recommendation reason and related items to the user.

[1885] Input: Generated recommendation reason text, suggested related item list

[1886] What it does: The device displays this information to the user in a visually convincing way.

[1887] Output: Displayed recommendation reason and related items

[1888] Step 14:

[1889] The server analyzes the user's input information and behavior.

[1890] Input: User input data and behavior history

[1891] How it works: The server analyzes the data using a sentiment analysis library (TextBlob or VADER) to recognize the user's sentiment.

[1892] Output: Recognized emotion data

[1893] Step 15:

[1894] The server dynamically adjusts the recommendation content based on emotions.

[1895] Input: Recognized emotion data, recommended item list

[1896] How it works: If the server detects negative sentiment, it changes the emphasis of the recommended items and adds limited-time sales and special offers.

[1897] Output: Dynamically adjusted recommendations

[1898] Step 16:

[1899] The server analyzes user responses in real time and adjusts the wording of the recommendation reasons.

[1900] Input: User response data (viewing time, number of clicks, etc.)

[1901] How it works: The server analyzes the user's response and modifies the generated reason statement based on the results. For example, if the user is feeling anxious, it emphasizes more positive feedback.

[1902] Output: Adjusted recommendation reason

[1903] (Application example 2)

[1904] 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 robot 414 will be referred to as a "terminal."

[1905] While conventional fashion recommendation systems suggest personalized items to users when shopping online, they do not address the in-store shopping experience. Furthermore, they do not provide recommendations that take into account the user's real-time movements and emotions, which prevents them from fully enhancing user satisfaction. Furthermore, there is a need for a system that effectively utilizes devices such as eye-tracking technology, smart glasses, and head-mounted displays to enable users to obtain the information they want in-store in real time.

[1906] 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.

[1907] In this invention, the server includes an eye tracking unit that identifies a specific product when the user moves around the store and directs their gaze at the product, a unit that dynamically adjusts recommendations for the identified product according to the user's emotional state, and a unit that displays the generated recommendation reasons and recommendation content on smart glasses or a head-mounted display, thereby improving the shopping experience in a physical store and making it possible to provide appropriate information and recommendations in real time for products that interest the user.

[1908] "Means for collecting user information" refers to a function for inputting and recording user preferences, body type, personal color, past purchase history, the latest trends, and the like.

[1909] The "means for users to input their favorite images, hashtags, and text" refers to an interface that allows users to input images, hashtags, text, etc. that specifically indicate their own fashion style and preferences.

[1910] "Means for analyzing user information and the images, hashtags, and text" refers to a function for analyzing collected user information and input images, hashtags, and text to understand user preferences and trends.

[1911] "Means for selecting and recommending personalized fashion items" refers to a system that selects and suggests the most suitable fashion items for users based on the analysis results.

[1912] The "means for generating specific reasons for recommendation" is a function for automatically generating detailed explanations and reasons why recommended fashion items are preferred.

[1913] The "means for suggesting other items and cosmetics" is a function for automatically suggesting other fashion items and cosmetics that can be easily combined with the recommended fashion items.

[1914] The "means for displaying to the user" is an interface for visually presenting information about the recommended fashion items and other suggested items and cosmetics to the user.

[1915] The "eye tracking means" is a technology for identifying and tracking a particular product when the user directs their gaze at the product.

[1916] "Dynamic adjustment means" is a function that allows recommendations to be changed and adjusted in real time according to the user's emotions and behavior.

[1917] "Means for displaying on smart glasses or a head-mounted display" refers to a function for displaying the generated recommendation reasons and recommendation content on a smart device in real time.

[1918] Specific configurations and methods for implementing the present invention are described below.

[1919] 1. Collection of User Information

[1920] The terminal uses smart glasses and mobile devices to collect user information. Users use these devices to input their preferred style, body type, personal color, past purchase history, and the latest trend information. This information is sent to a server and stored in a database.

[1921] 2. Eye tracking and product identification

[1922] The smart glasses use built-in eye-tracking technology (e.g., Tobii Eye Tracker SDK) to identify the products the user is looking at in the store, recognize the products the user is looking at, and send the related information to a server.

[1923] 3. Recommendation Generation and Sentiment Analysis

[1924] The server generates personalized recommendations based on the aforementioned user information and eye-tracking information. This includes suggesting fashion items based on the user's input and profile using a generative AI model (e.g., OpenAI API). It also uses an emotion analysis engine to analyze the user's emotional state from their facial expressions and behavior and dynamically adjust the recommendations.

[1925] 4. Displaying recommendation reasons and dynamic adjustment

[1926] The device displays the recommendation content and specific reasons for the recommendation in real time on smart glasses or a head-mounted display. Based on the results of the sentiment analysis engine, the device adjusts the recommendation reasons and the display of recommended items. For example, if the user is interested but anxious, it will emphasize positive feedback.

[1927] Specific examples

[1928] For example, if a user wears smart glasses and enters a store and looks at a particular product, fashion items related to that product will be automatically recommended. The recommendation will be displayed on the visual device along with the reason for the recommendation generated by the server based on the user's information. This allows users to enjoy shopping while receiving detailed information and personalized recommendations.

[1929] Prompt Sentence Examples

[1930] An example of a prompt generated by the server is:

[1931] markdown

[1932] User Profile:

[1933] Favorite style: Casual, pastel colors

[1934] Body type: Medium build

[1935] Past purchases: Jeans, blouses, casual shoes

[1936] User input:

[1937] Tags: Spring Casual

[1938] Keywords: Spring Picnic

[1939] Recommended Items:

[1940] pastel pink blouse

[1941] White denim

[1942] camera bag

[1943] Recommended reasons include:

[1944] This pastel pink blouse is perfect for your personal color and is perfect for casual spring occasions!

[1945] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1946] Step 1:

[1947] The user logs in using smart glasses or a mobile device. After logging in, the user enters their preferred style, body type, personal color, past purchase history, latest trend information, etc. This information is sent to the server and stored in a database.

[1948] Input: User profile information (style, body type, color, purchase history, trend information)

[1949] Output: User information stored in the database

[1950] Step 2:

[1951] As a user moves around the store, the smart glasses' eye-tracking technology (Tobii Eye Tracker SDK) is used to identify the products they are looking at. The glasses then transmit the user's gaze data to a server.

[1952] Input: User gaze data

[1953] Output: Identified product information

[1954] Step 3:

[1955] The server analyzes the user's gaze data and saved profile information to generate personalized recommendations for the user. A generative AI model (OpenAI API) is used to suggest fashion items based on the user's prompts.

[1956] Input: Identified product information, user profile information

[1957] Output: A list of recommended fashion items

[1958] Step 4:

[1959] The server inputs the user's prompt information into the generative AI model to generate specific recommendation reasons, including specific reasons and explanations for the new recommendation.

[1960] Input: Identified product information, user profile information, prompt information

[1961] Output: Specific reasons for recommendation

[1962] Step 5:

[1963] The server uses an emotion analysis engine to analyze the user's emotional state based on their facial expressions and behavior. Based on the user's emotions, the recommendation content is dynamically adjusted. For example, if the user is interested but anxious, the server emphasizes positive feedback.

[1964] Input: gaze data, facial expression data, behavior data

[1965] Output: Adjusted recommendation content

[1966] Step 6:

[1967] The device (smart glasses or head-mounted display) displays the generated recommendation content and specific reasons for the recommendation in real time. The user can consider products based on this information and obtain additional information as needed.

[1968] Input: Adjusted recommendation content, specific reasons for recommendation

[1969] Output: Present information to the user

[1970] Step 7:

[1971] The user selects recommended fashion items or other suggested items through their device. Detailed information about the selected item and related items are displayed. The device then sends the information back to the server, which generates additional recommendations as needed.

[1972] Input: User selection information

[1973] Output: Show detailed information and related items

[1974] Through these steps, users can enjoy a personalized shopping experience in a physical store.

[1975] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice 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 voice data.

[1976] 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.

[1977] 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 robot 414.

[1978] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1979] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1980] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1981] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1982] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1983] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1984] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1985] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1986] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1987] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1988] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1989] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1990] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1991] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1992] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1993] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1994] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1995] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1996] The following is further disclosed regarding the above embodiment.

[1997] (Claim 1)

[1998] a means for collecting user information;

[1999] A way for users to input their favorite images, hashtags, and text;

[2000] means for analyzing the user information and the images, hashtags, and text;

[2001] A means for selecting and recommending personalized fashion items based on the analysis results;

[2002] A means for generating a specific reason for recommending the recommended fashion item;

[2003] A means for suggesting other items and cosmetics that are easily combined with the fashion item;

[2004] means for displaying the recommended fashion items and other suggested items and cosmetics to a user;

[2005] A system including:

[2006] (Claim 2)

[2007] a means for users to log in;

[2008] A means to input information such as user preferences, body type, personal color, past purchase history, and the latest trends;

[2009] a database means for storing collected user information;

[2010] 10. The system of claim 1, further comprising:

[2011] (Claim 3)

[2012] A means for users to select fashion items that interest them;

[2013] A means for generating specific reasons for recommending selected fashion items;

[2014] A means for suggesting and presenting other items and cosmetics to the user;

[2015] 10. The system of claim 1, comprising:

[2016] "Example 1"

[2017] (Claim 1)

[2018] a means for collecting user information;

[2019] A way for users to input their favorite images, hashtags, and text;

[2020] means for analyzing the user information and the images, hashtags, and text;

[2021] A means for selecting and recommending personalized products based on the analysis results;

[2022] A means for generating a specific recommendation reason for the recommended product;

[2023] A means for suggesting other items that are easily combined with the product;

[2024] means for displaying the recommended products and other suggested items to a user;

[2025] A system including:

[2026] (Claim 2)

[2027] a means for users to log in;

[2028] A means to input information such as user preferences, body type, personal color, past purchase history, and the latest trends;

[2029] a database means for storing collected user information;

[2030] A means of using generative AI models as natural language processing techniques for analysis;

[2031] 10. The system of claim 1, further comprising:

[2032] (Claim 3)

[2033] A means for users to select products that interest them;

[2034] A means for generating specific recommendation reasons for the selected products;

[2035] means for suggesting and presenting other items to the user;

[2036] 10. The system of claim 1, comprising:

[2037] "Application Example 1"

[2038] (Claim 1)

[2039] a means for collecting user information;

[2040] A way for users to input their favorite images, hashtags, and text;

[2041] means for analyzing the user information and the images, hashtags, and text;

[2042] A means for selecting and recommending personalized fashion items based on the analysis results;

[2043] A means for generating a specific reason for recommending the recommended fashion item;

[2044] A means for suggesting other items and cosmetics that are easily combined with the fashion item;

[2045] means for displaying the recommended fashion items and other suggested items and cosmetics to a user;

[2046] a means for visually displaying the recommended fashion items using smart glasses;

[2047] A system including:

[2048] (Claim 2)

[2049] a means for users to log in;

[2050] A means to input information such as user preferences, body type, personal color, past purchase history, and the latest trends;

[2051] a database means for storing collected user information;

[2052] A means to generate prompt sentences and analyze and recommend fashion items using a generative AI model;

[2053] 10. The system of claim 1, further comprising:

[2054] (Claim 3)

[2055] A means for users to select fashion items that interest them;

[2056] A means for generating specific reasons for recommending selected fashion items;

[2057] A means for suggesting and presenting other items and cosmetics to the user;

[2058] A means for encouraging users to make a purchase through visual display;

[2059] 10. The system of claim 1, comprising:

[2060] "Example 2: Combining Emotion Engines"

[2061] (Claim 1)

[2062] a means for collecting user information;

[2063] A way for users to input their favorite images, hashtags, and text;

[2064] means for analyzing the user information and the images, hashtags, and text;

[2065] A means for selecting and recommending personalized fashion items based on the analysis results;

[2066] A means for generating a specific reason for recommending the recommended fashion item;

[2067] A means for suggesting other items and cosmetics that are easily combined with the fashion item;

[2068] means for displaying the recommended fashion items and other suggested items and cosmetics to a user;

[2069] A means for analyzing user input information and behavior and recognizing emotions;

[2070] means for dynamically adjusting the recommendation content based on the recognized user sentiment;

[2071] means for analyzing user reactions to the recommended items in real time and dynamically adjusting the recommendation reasons in accordance with the analysis results;

[2072] A system including:

[2073] (Claim 2)

[2074] a means for users to log in;

[2075] A means to input information such as user preferences, body type, personal color, past purchase history, and the latest trends;

[2076] a database means for storing collected user information;

[2077] means for generating a prompt sentence using a generative AI model from the data input by the user;

[2078] 10. The system of claim 1, further comprising:

[2079] (Claim 3)

[2080] A means for users to select fashion items that interest them;

[2081] A means for generating specific reasons for recommending selected fashion items;

[2082] A means for suggesting and presenting other items and cosmetics to the user;

[2083] A means for dynamically adjusting the recommendations using a generative AI model when suggesting other items to complement the user's selected item; and

[2084] 10. The system of claim 1, comprising:

[2085] "Application example 2 when combining emotion engines"

[2086] (Claim 1)

[2087] a means for collecting user information;

[2088] A way for users to input their favorite images, hashtags, and text;

[2089] means for analyzing the user information and the images, hashtags, and text;

[2090] A means for selecting and recommending personalized fashion items based on the analysis results;

[2091] A means for generating a specific reason for recommending the recommended fashion item;

[2092] A means for suggesting other items and cosmetics that are easily combined with the fashion item;

[2093] means for displaying the recommended fashion items and other suggested items and cosmetics to a user;

[2094] an eye tracking means for identifying a particular product when the user moves around the store and directs their gaze at the product;

[2095] means for dynamically adjusting recommendations for the identified products according to the emotional state of the user;

[2096] A means for displaying the generated recommendation reasons and recommendation contents on smart glasses or a head-mounted display;

[2097] A system including:

[2098] (Claim 2)

[2099] a means for users to log in;

[2100] A means to input information such as user preferences, body type, personal color, past purchase history, and the latest trends;

[2101] a database means for storing collected user information;

[2102] The system of claim 1 further comprising:

[2103] (Claim 3)

[2104] A means for users to select fashion items that interest them;

[2105] A means for generating specific reasons for recommending selected fashion items;

[2106] A means for suggesting and presenting other items and cosmetics to the user;

[2107] A means for displaying the recommendation content in real time through smart glasses or a head-mounted display;

[2108] 10. The system of claim 1, comprising: [Explanation of symbols]

[2109] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for collecting user information; A way for users to input their favorite images, hashtags, and text; means for analyzing the user information and the images, hashtags, and text; A means for selecting and recommending personalized fashion items based on the analysis results; A means for generating a specific reason for recommending the recommended fashion item; A means for suggesting other items and cosmetics that are easily combined with the fashion item; means for displaying the recommended fashion items and other suggested items and cosmetics to a user; A system including:

2. a means for users to log in; A means to input information such as user preferences, body type, personal color, past purchase history, and the latest trends; a database means for storing collected user information; The system of claim 1 further comprising:

3. A means for users to select fashion items that interest them; A means for generating specific reasons for recommending selected fashion items; A means for suggesting and presenting other items and cosmetics to the user; The system of claim 1 , comprising:

Images