System

Abstract

A system is provided.SOLUTION: A system comprising: means for obtaining user preference information; means for obtaining owned clothing data; means for recommending optimal clothes based on the user preference information and the owned clothing data; and means for generating a reward for the recommended clothes.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In today's world where clothing choices are becoming more diverse, users often find it difficult to find clothing that perfectly matches their preferences. This is particularly true when matching existing clothing with existing clothing or finding a coordinated outfit that suits their preferences. Furthermore, users may feel that compliments from store clerks or the results of AI fittings lack credibility, leading to a lack of confidence in the clothes they choose. In these circumstances, a new system is needed to help users feel confident in the clothes they choose. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system that includes a means for acquiring user preference information, a means for acquiring data on owned clothing, a means for recommending optimal clothing based on the user's preference information and the user's clothing data, and a means for generating compliments for the recommended clothing. Specifically, the system converts the user's preference information and clothing description data into TF-IDF vectors and calculates the cosine similarity between the vectors to select optimal clothing. Furthermore, the system generates and provides more human-like compliments to the user based on the color and style of the selected clothing, allowing the user to feel confident in their clothing selection.

[0006] "User preference information" is information that indicates the user's personal preferences, such as the style, color, and design that the user prefers.

[0007] "Owned clothing data" is data that includes information on clothing that the user already owns.

[0008] "Recommending optimal clothing" refers to proposing the most suitable clothing to the user based on the user's preference information and clothing data.

[0009] "Generating compliments" refers to generating words to praise the goodness and attractiveness of the recommended outfit.

[0010] The "system" refers to a set of devices and software including a means for acquiring user preference information, a means for acquiring user clothing data, a means for recommending clothes, and a means for generating compliments.

[0011] "TF-IDF vectorization" is a method of calculating the features of a document using word frequency and inverse document frequency to represent the features as a numerical vector.

[0012] "Cosine similarity" is a method for measuring the similarity of two vectors by calculating the cosine value of the angle between the vectors.

[0013] "Color and style" refers to the specific characteristics of the appearance or design of a garment, including specifically shades and fashion types. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

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

[0024] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0031] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] The present invention relates to a system for recommending optimal clothing based on a user's preference information and clothing data, and generating compliments for the selected clothing. Specific embodiments of the system are described below.

[0036] System configuration

[0037] The system of the present invention comprises a user terminal, a server, and a database. The user terminal provides an interface for users to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and catalog information of fashion items.

[0038] System Operation

[0039] 1. Enter user information

[0040] The user logs in to the system and inputs their preferences (preferred style, color, design, etc.), which are then sent to the server by the terminal.

[0041] 2. Obtaining User Data

[0042] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database.

[0043] 3. Acquiring fashion item data

[0044] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[0045] 4. Running the recommendation algorithm

[0046] The server converts the descriptions of fashion items into TF-IDF vectors, and also converts the user's preference information into vectors in the same way.

[0047] The server calculates the cosine similarity between these vectors to identify the items that most closely match the user's preferences, and recommends the most similar items to the user.

[0048] 5. Generating compliments

[0049] The server generates compliments for each selected item, building feedback that praises its quality and appeal based on color and style.

[0050] 6. Displaying the results

[0051] The server sends the generated list of recommended items and the compliments to the device.

[0052] The user terminal displays these results to the user, who can then check the suggested outfits and their feedback.

[0053] Specific examples

[0054] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[0055] 2. The server retrieves data such as the user's preferences and past purchase history from the database.

[0056] 3. The server retrieves information on all fashion items from the database and converts the item descriptions into TF-IDF vectors.

[0057] 4. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the item that best matches the user's preference.

[0058] 5. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!"

[0059] 6. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[0060] The above is an embodiment of the system of the present invention, which allows users to find the best outfit based on their preferences and feel confident in their choice.

[0061] The processing flow will be explained below.

[0062] Step 1:

[0063] A user logs in to the system using a terminal and inputs their preference information (e.g., preferred style, color, design), and the terminal transmits the input data to the server.

[0064] Step 2:

[0065] The server retrieves basic information about the user from the database, such as preference information and past purchase history corresponding to the user ID.

[0066] Step 3:

[0067] The server retrieves all the fashion item information from the database, including the name, description, color, style, etc. of the outfit.

[0068] Step 4:

[0069] The server converts the descriptions of fashion items into TF-IDF vectors, which is a process for expressing clothing features as numerical vectors.

[0070] Step 5:

[0071] The server also converts the user's preference information into a TF-IDF vector, which allows the user's preferences to be expressed as a numerical vector.

[0072] Step 6:

[0073] The server calculates the cosine similarity between these vectors, and uses the cosine similarity to identify the fashion item that most closely matches the user's preferences.

[0074] Step 7:

[0075] The server selects the top five fashion items from among the items with the highest similarity scores, thereby recommending the most suitable outfit for the user.

[0076] Step 8:

[0077] The server generates compliments for each selected item, generating feedback that praises its quality and appeal based on color and style.

[0078] Step 9:

[0079] The server sends the generated list of recommended items and the compliments to the device.

[0080] Step 10:

[0081] The device will display the recommended outfits and feedback to the user, and the user can check the suggested outfits and their feedback.

[0082] Example 1

[0083] 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."

[0084] In recent years, interest in fashion has increased, and many people want to choose clothes that suit their own tastes. However, choosing the best outfit from the vast number of fashion items available is not easy, and the process takes time and effort. It is also difficult to be confident in one's choices. Furthermore, existing recommendation systems may not be accurate enough to recommend items that precisely match a user's tastes.

[0085] 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.

[0086] In this invention, the server includes means for acquiring user preference information, means for acquiring user clothing data, means for recommending optimal clothing based on the user preference information and the user clothing data, and means for generating compliments for the recommended clothing using a generative AI model. This makes it possible to select optimal clothing based on the user's preferences and generate compliments to give the user confidence.

[0087] "User preference information" is information about personal fashion preferences such as the user's preferred style, color, and design.

[0088] "Owned clothing data" is information about clothing currently owned by the user, including the item name, description, color, style, and so on.

[0089] The "means for recommending optimal clothing" is a function for identifying and recommending the most suitable clothing for a user using the user's preference information and clothing data.

[0090] A "generative AI model" refers to an artificial intelligence algorithm that generates a response in natural language in response to a certain input (prompt), such as a generative language model.

[0091] The "means for generating compliments" is a function that uses a generative AI model to create feedback praising the good qualities and attractiveness of the selected outfit.

[0092] MODE FOR CARRYING OUT THE INVENTION

[0093] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Specific embodiments of the system are described below.

[0094] System configuration

[0095] This system consists of a user terminal, a server, and a database. The user terminal provides an interface for users to access the system and input their preference information. The server processes the user's preference information and clothing data and recommends the most suitable outfits. The database stores user information, clothing data, and catalog information of fashion items.

[0096] System Operation

[0097] 1. Enter user information

[0098] The user logs into the system using a terminal and inputs preference information (preferred style, color, design, etc.).

[0099] Example: A user inputs that they like "elegant style" and the color "blue." The device sends this information to the server through the interface.

[0100] 2. Obtaining User Data

[0101] The server retrieves basic information about the user (preferences, past purchase history, etc.) from a database. This information is retrieved using an SQL database (e.g., MySQL).

[0102] Example: The server executes an SQL query to retrieve user preferences such as "elegant" and "blue."

[0103] 3. Acquiring fashion item data

[0104] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[0105] Example: The server executes an SQL query to retrieve item information such as "an elegant blue dress" from a database.

[0106] 4. Running the recommendation algorithm

[0107] The server converts the descriptions of fashion items into TF-IDF vectors and also converts the user's preference information into vectors. This is done using Python and natural language processing libraries (e.g., scikit-learn, NLTK).

[0108] Cosine similarity is calculated to identify items that most closely match the user's preferences.

[0109] Example: The description of each item is vectorized using TF-IDF vectorization, and the cosine similarity is calculated to select dresses that are "elegant" and "blue."

[0110] 5. Generating compliments

[0111] The server generates a compliment for each selected item. To do this, it uses a generative AI model (e.g., GPT-3). Specifically, it generates a compliment for the selected item by inputting a prompt sentence into the generative AI model.

[0112] Example prompt: "Generate a compliment about this elegant blue dress."

[0113] Example: A generative AI model can generate compliments such as, "The dress you chose is very elegant and graceful!"

[0114] 6. Displaying the results

[0115] The server transmits the generated list of recommended items and the compliments to the terminal, which displays them.

[0116] Example: The server sends the generated item list and compliments in JSON format, and the web browser or smartphone app on the user's device displays "The blue dress you chose is very elegant and wonderful!"

[0117] In this way, the present invention can recommend optimal outfits based on a user's preferences, providing confidence in the choice. Furthermore, by using a generative AI model, the generation of compliments can be automated, providing users with rapid positive feedback.

[0118] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0119] Step 1: Enter your information

[0120] Users log in to the system using a terminal and enter their preference information. Specifically, the user enters their authentication information on the login screen and transitions to the customer preference interface. There, the user enters their preferred style, color, design, etc. into the input form.

[0121] Input: Authentication information, preference information (preferred style, color, design, etc.)

[0122] Output: User preference information is sent to the server

[0123] Step 2: Retrieving User Data

[0124] The server executes an SQL query to retrieve basic user information from the database, including past purchase history and registered preference information.

[0125] Input: User ID

[0126] Output: Basic information of the user (preference information, past purchase history, etc.)

[0127] Step 3: Obtaining fashion item data

[0128] The server retrieves all the fashion item information from the database. Specifically, the server executes SQL queries to retrieve the name, description, color, style, etc. of each fashion item.

[0129] Input: None

[0130] Output: Fashion item information (name, description, color, style, etc.)

[0131] Step 4: Run the recommendation algorithm

[0132] The server converts the descriptions of fashion items and user preference information into TF-IDF vectors. Specifically, it uses Python and natural language processing libraries (e.g., scikit-learn, NLTK) to vectorize the text data and calculate the cosine similarity between them to identify the most suitable items.

[0133] Input: Description of fashion item, user preference information

[0134] Output: A list of items that best fit the user's preferences

[0135] Step 5: Generate compliments

[0136] The server generates a compliment for the selected item using a generative AI model (e.g., GPT-3). Specifically, the server inputs a prompt sentence into the generative AI model to generate feedback.

[0137] Input: Recommended item

[0138] Output: Generated compliment

[0139] Step 6: View the results

[0140] The server sends the generated list of recommended items and compliments to the device, which then displays them. Specifically, the device parses the received JSON data and displays it to the user in a web browser or application.

[0141] Input: Recommended item list, compliments

[0142] Output: A list of recommended items and a compliment displayed to the user

[0143] (Application example 1)

[0144] 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."

[0145] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. In particular, conventional systems require users to select a vast number of fashion items themselves, and this has led to issues with insufficient appropriate recommendations and feedback. There is a need for systems that can improve user satisfaction and recommend optimal clothing in a rational and timely manner.

[0146] 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.

[0147] In this invention, the server includes a means for acquiring user preference information, a means for acquiring user clothing data, a means for recommending optimal clothing based on the user's preference information and the user's clothing data, a means for generating compliments for the recommended clothing, and a means for displaying these results. This allows users to easily find optimal clothing based on their preferences and feel confident in their selection. Furthermore, a recommendation algorithm using TF-IDF vectorization and cosine similarity calculations achieves highly accurate recommendations. Furthermore, by using prompt sentences that generate compliments using a generative AI model, more natural and persuasive feedback can be provided.

[0148] definition statement

[0149] "User preference information" is information about fashion such as the style, color, and design that the user prefers.

[0150] "Owned clothing data" is detailed information about clothing items owned by the user, including the name, description, color, style, etc. of each item.

[0151] "Means for recommending optimal clothing" refers to a method or device for selecting the most suitable clothing for a user using an algorithm based on the user's preference information and clothing data.

[0152] "Compliment generator" refers to an algorithm or mechanism for generating positive feedback on a recommended outfit.

[0153] "Means for displaying these results" refers to a method or device for displaying the recommended outfits and compliments on the outfits on the user terminal.

[0154] "TF-IDF vectorization" is a technique for quantifying text data as vectors, and is a technology that converts it into a vector representation by taking into account the importance of each word.

[0155] "Cosine similarity" is an index for measuring the similarity between two vectors, and is calculated based on the angle of the vectors.

[0156] "Generative AI model" refers to an algorithm or system that uses artificial intelligence to automatically create text or other artifacts.

[0157] A "prompt sentence" is an input sentence used to elicit a specific response from a generative AI model.

[0158] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Specific embodiments of this system are described below.

[0159] System configuration

[0160] The system of the present invention consists of the following main parts:

[0161] User terminal

[0162] server

[0163] Database

[0164] These parts work together to provide optimal outfit recommendations and compliment generation for the user.

[0165] User device functions

[0166] The user terminal provides an interface that receives input from the user and displays the results. The user logs in using the terminal and inputs preference information (preferred style, color, design, etc.). The terminal sends this information to the server.

[0167] Server Features

[0168] The server has the following functions:

[0169] 1. Data Acquisition:

[0170] The server obtains the user's preference information and clothing data from the database.

[0171] 2. Data Processing:

[0172] The server converts preference information and clothing description data into TF-IDF vectors and calculates the cosine similarity to select the most suitable outfit.

[0173] 3. Compliment Generation:

[0174] The server generates natural and complimenting words for the selected outfit, using a generative AI model to input prompts and create feedback accordingly.

[0175] Database

[0176] The database stores user information, clothing data, and fashion item catalog information, allowing the server to quickly and efficiently retrieve the required information.

[0177] Example of operation

[0178] 1. Log in and enter your preferences:

[0179] The user logs in and inputs his / her preference information of "elegant style" and "blue."

[0180] 2. Retrieving data from the database:

[0181] The server retrieves the user's preference information, past purchase history, and clothing data from the database.

[0182] 3. Clothing recommendations:

[0183] The server uses TF-IDF vectorization and cosine similarity calculation to select the item that best matches the user's preferences, such as an "elegant blue dress."

[0184] 4. Generating compliments:

[0185] The server uses a generative AI model to generate compliments such as, "The dress you chose is very elegant and graceful!"

[0186] Hardware and software used

[0187] Server: The server on which the Python program runs (using a server framework such as Flask).

[0188] User device: Smartphone (including iOS and Android).

[0189] Examples of prompt statements

[0190] Here is an example of a prompt that can be fed into a generative AI model:

[0191] text

[0192] "Please tell me the algorithm for a system that recommends the most suitable outfit based on the user's preference information and clothing data, and generates compliments for the selected outfit. The user's preference information is "elegant style" and "blue." Please also take into account past purchase data and clothing data."

[0193] The above is a specific embodiment for carrying out the invention. This system allows a user to find the best outfit based on their own preferences and wear the outfit they have chosen with confidence.

[0194] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0195] Processing Step Description

[0196] Step 1:

[0197] User login and preference input

[0198] The user logs in to the device and inputs their preferences (style, color, design, etc.), which are then sent from the device to the server.

[0199] Input: User login information, preferences

[0200] Output: Preference information sent to the server

[0201] Step 2:

[0202] Retrieving user data from the database

[0203] The server accesses the database to obtain the user's preferences, past purchase history, and clothing data, thereby gathering all data related to the user.

[0204] Input: Database query, user identity

[0205] Output: Acquired user data, owned clothing data

[0206] Step 3:

[0207] Obtaining information on fashion items

[0208] The server retrieves information about all fashion items (item name, description, color, style, etc.) from the database and creates a list.

[0209] Input: Database query

[0210] Output: A list of retrieved fashion items

[0211] Step 4:

[0212] TF-IDF vectorization of data

[0213] The server converts the acquired user preference information and fashion item description data into TF-IDF vectors, which converts each piece of data into a numerical value and puts it in a format suitable for calculations.

[0214] Input: User preference information, fashion item description data

[0215] Output: Vectorized user preference information and item description data

[0216] Step 5:

[0217] Calculating cosine similarity

[0218] The server calculates the cosine similarity using the TF-IDF vectorized data, and selects the item that most closely matches the user's preferences.

[0219] Input: Vectorized user preference information and item description data

[0220] Output: A list of items with high similarity

[0221] Step 6:

[0222] Selection of recommended items

[0223] Based on the results of the cosine similarity calculation, the server selects several items with the highest similarity and generates a list of recommended items.

[0224] Input: List of highly similar items

[0225] Output: Recommended item list

[0226] Step 7:

[0227] Generating compliments

[0228] The server generates compliments for the selected recommended items using a generative AI model. The server inputs prompts to the model to obtain natural and positive compliments.

[0229] Input: Recommended item list, prompt text

[0230] Output: Compliment generation

[0231] Step 8:

[0232] Displaying the results

[0233] The server sends a list of recommended items and compliments to the user's device, which then displays them. The user can check the suggested outfit and the feedback.

[0234] Input: Recommended item list, compliments

[0235] Output: The result displayed on the user's terminal

[0236] Through the above processing steps, the user can easily find the best outfit based on their preferences and feel confident in their choice.

[0237] 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.

[0238] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide feedback based on the user's emotions. Specific embodiments of the system are described below.

[0239] System configuration

[0240] The system of the present invention comprises a user terminal, a server, a database, and an emotion engine. The user terminal provides an interface for the user to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and fashion item catalog information. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, input text, etc., and sends the results to the server.

[0241] System Operation

[0242] 1. Enter user information

[0243] A user logs in to the system and inputs their preference information (e.g., preferred style, color, design, etc.). The terminal sends this information to the server.

[0244] 2. Acquiring emotional information

[0245] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. This emotion information is sent to the server.

[0246] 3. Obtaining User Data

[0247] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database.

[0248] 4. Acquiring fashion item data

[0249] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[0250] 5. Running the recommendation algorithm

[0251] The server converts the descriptions of fashion items into TF-IDF vectors, and also converts the user's preference information into vectors in the same way.

[0252] The server calculates the cosine similarity between these vectors to identify the items that most closely match the user's preferences, and recommends the most similar items to the user.

[0253] 6. Generating compliments

[0254] The server generates compliments for each selected item, building feedback that praises its quality and appeal based on color and style.

[0255] The server can adjust the content and tone of the compliment based on the user's emotional information obtained from the emotion engine.

[0256] 7. Displaying the results

[0257] The server sends the generated list of recommended items and the compliments to the device.

[0258] The user terminal displays these results to the user, who can then check the suggested outfits and their feedback.

[0259] Specific examples

[0260] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[0261] 2. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The emotion information is sent to the server.

[0262] 3. The server retrieves data such as the user's preferences and past purchase history from the database.

[0263] 4. The server retrieves all fashion item information from the database and converts the item descriptions into TF-IDF vectors.

[0264] 5. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the item that best matches the user's preference.

[0265] 6. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!". Based on the emotional information that the user is pleased, the server adjusts the tone of the compliment to be more positive.

[0266] 7. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[0267] The above is an embodiment of the system of the present invention. This system allows users to find the best outfit based on their preferences and feel confident in their choice. Furthermore, by incorporating an emotion engine, it is possible to provide feedback according to the user's emotions.

[0268] The processing flow will be explained below.

[0269] Step 1:

[0270] A user logs in to the system using a terminal and inputs his / her preference information (e.g., preferred style, color, design, etc.). The terminal transmits the input data to the server.

[0271] Step 2:

[0272] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. This emotion information is then sent to the server.

[0273] Step 3:

[0274] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database. Specifically, it issues a DB query to retrieve related data.

[0275] Step 4:

[0276] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database. The retrieved data is then used in the recommendation algorithm.

[0277] Step 5:

[0278] The server converts the descriptions of fashion items into TF-IDF (Term Frequency-Inverse Document Frequency) vectors, which is a process of representing each item description as a numerical vector.

[0279] Step 6:

[0280] The server also converts the user's preference information into a TF-IDF vector, which allows the user's preferences to be expressed as a numerical vector.

[0281] Step 7:

[0282] The server calculates the cosine similarity between these vectors, and uses the cosine similarity to identify the fashion item that most closely matches the user's preferences.

[0283] Step 8:

[0284] The server selects the top five fashion items from the items with the highest similarity scores, and the selected items are recommended to the user.

[0285] Step 9:

[0286] The server generates compliments for each selected item, generates feedback praising its goodness and attractiveness based on its color and style, and adjusts the content and tone of the compliments based on the user's emotional information obtained from the emotion engine.

[0287] Step 10:

[0288] The server sends the generated list of recommended items and the compliments to the device.

[0289] Step 11:

[0290] The device will display the recommended outfits and feedback to the user, and the user can check the suggested outfits and their feedback.

[0291] Specific examples

[0292] If a user logs in and inputs that they like "elegant style" and the color "blue," the system sends this preference information to the server. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The emotion information is sent to the server. The server retrieves the user's preference information and data such as past purchase history from the database. The server retrieves information on all fashion items from the database and converts the item descriptions into TF-IDF vectors. The server also converts the user's preference information into vectors and calculates cosine similarity to identify the item that best matches the user's preferences. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!" Based on the emotional information that the user is happy, the system adjusts the tone of the compliment to be more positive. The server sends the recommended item list and the compliment to the terminal, which displays them. The user can review the results and wear the outfit they chose with confidence.

[0293] Example 2

[0294] 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."

[0295] Conventional fashion recommendation systems can only suggest outfits based on the user's preference information and clothing data, but they have the problem of not being able to provide feedback that takes into account the user's emotions. This means that the system is unable to fully improve the user's feelings and satisfaction with the suggested outfits, resulting in limited effectiveness.

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

[0297] In this invention, the server includes means for acquiring user preference information, means for acquiring owned clothing data, and means for recommending optimal clothing based on the user preference information and owned clothing data. This makes it possible to suggest optimal clothing based on the user preference information and owned clothing data, thereby improving the user's feelings and satisfaction with the suggested clothing.

[0298] In addition, in this invention, the server includes means for recognizing emotions from the user's facial expressions, tone of voice, input text, etc., means for adjusting the content and tone of the compliments based on the emotional information, and means for generating compliments for the recommended outfit. This makes it possible to provide feedback that takes into consideration the user's emotions, further increasing user satisfaction.

[0299] "User preference information" refers to information about personal preferences such as the user's preferred style, color, and design.

[0300] "Owned clothing data" is information about clothing owned by the user, and includes detailed data such as the type, color, style, and brand of the item.

[0301] "Recommendation means" is a system function that selects optimal clothing based on the user's preference information and clothing data and presents it to the user.

[0302] The "means for generating compliments" is a function of the system that generates feedback praising the goodness and attractiveness of the recommended outfit.

[0303] "Means for recognizing emotions" refers to the system's function of analyzing and determining the user's emotions from the user's facial expressions, tone of voice, input text, etc.

[0304] The "means for adjusting the content and tone of the compliment based on emotional information" is a system function for adjusting the content and tone of the generated compliment to the user's current emotional state based on information obtained from the means for recognizing emotions.

[0305] "TF-IDF vectorization means" is a system function that calculates the importance of words using text data and converts it into a vector representation.

[0306] The "means for calculating cosine similarity" is a system function that measures the similarity between vectorized data using cosine angles and calculates the similarity.

[0307] MODE FOR CARRYING OUT THE INVENTION

[0308] This invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide feedback based on the user's emotions. Specific embodiments of this system are described below.

[0309] System configuration

[0310] The system of the present invention comprises a user terminal, a server, a database, and an emotion engine. The user terminal provides an interface for the user to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and fashion item catalog information. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, input text, etc., and sends the results to the server.

[0311] Hardware and Software Use

[0312] User terminals include devices such as PCs, smartphones, and tablets. The server is a server computer with high-performance computing capabilities, and the database can use a commercial relational database system (e.g., MySQL or PostgreSQL).

[0313] The emotion engine is realized using a model that uses deep learning technology (e.g., an emotion analysis model). For example, a camera and image recognition technology are required to analyze a user's facial expressions, and a microphone and voice analysis technology are required to analyze the tone of voice. A deep learning framework (e.g., TensorFlow, PyTorch) is used to implement the emotion engine.

[0314] Natural language processing techniques are used for the recommendation algorithm and compliment generation. Specifically, "TF-IDF vectorization" and "cosine similarity calculation" are used to process text data. These processes can be performed using the Python "scikit-learn" library. In addition, generative AI models (e.g., GPT models) are used for natural language generation.

[0315] Specific examples

[0316] A specific example of the operation of the system is shown below.

[0317] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[0318] 2. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. This emotion information is sent to the server.

[0319] 3. The server retrieves data such as the user's preferences and past purchase history from the database.

[0320] 4. The server retrieves all fashion item information from the database and converts the item descriptions into TF-IDF vectors.

[0321] 5. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the most suitable item.

[0322] 6. For example, if an "elegant blue dress" is selected, the server generates a compliment for this item, such as "The dress you chose is very elegant and graceful!". Based on the emotional information that the user is pleased, the tone of the compliment is adjusted to be more positive.

[0323] 7. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[0324] Prompt Sentence Examples

[0325] "Recommend the most suitable fashion items based on the user's preferences and past purchase history. Also, generate compliments for the selected items and adjust the tone according to the user's emotions."

[0326] As a result, this system utilizes the user's preference information and emotional information to recommend the most suitable outfits and provide the user with a high level of satisfaction.

[0327] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0328] Step 1:

[0329] A user logs in to the system. The user enters login information, which the terminal sends to the server. The server then refers to a database to authenticate the login information and authenticates the user. If authentication is successful, a screen for entering the user's preferences is displayed.

[0330] Step 2:

[0331] The user inputs their preference information. For example, they select "elegant style" and "blue color." The terminal sends this preference information to the server. The server stores the input preference information in a database. This preference information becomes the input for subsequent processing.

[0332] Step 3:

[0333] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. For example, facial expression data is acquired from a camera and analyzed using a deep learning model to recognize "joy." This emotion information is sent to the server, which temporarily stores the received emotion information.

[0334] Step 4:

[0335] The server retrieves basic information about the user (preferences, past purchase history, etc.) from a database. The server issues SQL queries to extract this information from the database and stores it in memory. This basic information becomes the input for the recommendation algorithm.

[0336] Step 5:

[0337] The server retrieves all fashion item information from the database. Specifically, detailed data such as item name, description, color, and style are extracted using SQL queries. The server receives this information and prepares to convert each item description into a TF-IDF vector.

[0338] Step 6:

[0339] The server converts the fashion item descriptions into TF-IDF vectors. Using the Python "scikit-learn" library, the item descriptions are converted into numerical vectors. User preference information is also vectorized in the same way. These vectors are used as inputs for the cosine similarity calculation.

[0340] Step 7:

[0341] The server calculates the cosine similarity and identifies the item that most closely matches the user's preferences. It outputs a similarity score by calculating the cosine similarity between vectors. It selects the item with the highest score and creates a list of recommended items.

[0342] Step 8:

[0343] The server generates compliments for the selected item. Using a generative AI model (e.g., a GPT model), it generates text that praises the good points and attractiveness of the selected item. It includes specific feedback using color and style information. It also adjusts the tone of the compliments based on emotional information from an emotion engine.

[0344] Step 9:

[0345] The server sends the generated list of recommended items and compliments to the device. The generated text data and item information are sent to the device in JSON format. The device displays these results to the user. The user can check the suggested outfits and their feedback and wear the outfits they have chosen with confidence.

[0346] (Application example 2)

[0347] 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."

[0348] Conventional fashion recommendation systems can recommend clothing based on a user's preferences, but they are unable to provide emotional feedback on the selection. This leaves users with little confidence in the recommended clothing, resulting in low satisfaction with their selection. Furthermore, because previous systems do not take the user's emotional state into account, they are unable to provide appropriate feedback that reflects the user's psychological state at any given time. Therefore, the objective of this invention is to provide a system that takes into account the user's preference information and emotional information, allowing the user to select fashion items with greater satisfaction and confidence.

[0349] 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.

[0350] In this invention, the server includes means for acquiring user preference information, means for acquiring owned clothing data, means for recommending optimal clothing based on the user preference information and the owned clothing data, means for generating compliments for the recommended clothing, means for acquiring user emotion information and adjusting the tone of the compliments based on the emotion information, and means for displaying the optimal clothing after it has been recommended. This makes it possible not only to recommend optimal clothing based on the user preference information, but also to generate compliments for the selected clothing and provide feedback according to the user's emotions.

[0351] "User preference information" is data relating to the user's preferences such as style, color, and design.

[0352] "Owned clothing data" is data about clothing owned by the user, and includes information such as item name, color, style, and description.

[0353] The "means for recommending optimal clothing" is a means for identifying and recommending the most suitable clothing for a user based on the user's preference information and clothing data owned by the user.

[0354] The "means for generating compliments" is a means for generating feedback praising the goodness and attractiveness of the recommended outfit.

[0355] "User emotion information" is data related to emotions analyzed from the user's facial expression, tone of voice, input text, and the like.

[0356] The "means for adjusting the tone of the compliment" is a means for adjusting the content and tone of the generated compliment based on the acquired emotional information of the user.

[0357] The "means for displaying the optimal outfit after it has been recommended" is a means for visually presenting information about the selected outfit and feedback thereon to the user.

[0358] This system recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it provides feedback based on the user's emotions. This system can be implemented as an application that runs on smart glasses, head-mounted displays, and smartphones.

[0359] Hardware and software used

[0360] Hardware:

[0361] Smart Glasses

[0362] head-mounted display

[0363] Smartphone

[0364] software:

[0365] Python environment

[0366] TF-IDF Vectorizer (scikit-learn)

[0367] Cosine similarity calculation module (scikit-learn)

[0368] Sentiment analysis engine (AWS Rekognition, Google Cloud Vision, or TextBlob)

[0369] Data processing and calculation

[0370] 1. Data Vectorization:

[0371] The server converts the user's preference information and the description of the clothing data they own into a TF-IDF vector, which allows the data features to be expressed as a numerical vector and made comparable.

[0372] 2. Similarity calculation:

[0373] The server calculates the cosine similarity between the vectorized user preference information and the clothing data, and identifies the items that best match the preferences.

[0374] 3. Compliment Generation:

[0375] The server generates feedback for the selected item, praising its merits and attractiveness, and adjusts the content and tone of the praise based on the user's emotional information obtained from the emotion engine.

[0376] 4. Displaying the results:

[0377] The user device visually presents the recommended item list and compliments received from the server to the user, allowing the user to check the suggested outfit and receive feedback on it.

[0378] Specific examples

[0379] For example, if a user inputs that they like "elegant style" and "blue," the system processes this preference information. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The server vectorizes the user's preference information and clothing data, calculates the cosine similarity, and identifies the item that best matches the user's preference.

[0380] For a selected item (e.g., "an elegant blue dress"), the server generates a compliment such as "The dress you chose is very elegant and graceful!" Furthermore, based on the emotional information that the user is pleased, the tone of the compliment is adjusted to be more positive.

[0381] Prompt Sentence Examples

[0382] Here are some example input prompts for a generative AI model:

[0383] "A system that recommends the most suitable outfit based on the user's preference information and clothing data, and generates compliments for the selected outfit. Furthermore, please combine it with an emotion engine that recognizes the user's emotions and provides emotion-based feedback."

[0384] The above is a specific embodiment of the present invention. This system allows users to find the perfect outfit based on their preferences and feel confident in their choice. Furthermore, by incorporating an emotion engine, it is possible to provide feedback based on the user's emotions.

[0385] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0386] Step 1:

[0387] The user logs in to the device and inputs their preferences (e.g., preferred style, color, design, etc.). The user's input information is acquired by the device and sent to the server. The input data includes the user's preferred style and color, as well as past purchase history. The server stores this information in a database.

[0388] Step 2:

[0389] The server activates an emotion engine and analyzes the user's facial expressions, tone of voice, input text, etc. to obtain emotional information. Specifically, the user's facial expressions and voice data obtained using a camera or microphone are sent to the emotion analysis engine, which extracts emotional information as positive, negative, or neutral. This emotional information is then sent to the server and stored in a database.

[0390] Step 3:

[0391] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database. At the same time, it also retrieves data about all fashion items from the database. This data includes item names, descriptions, colors, styles, etc.

[0392] Step 4:

[0393] The server converts the descriptions of fashion items into TF-IDF vectors. Then, it converts the user's preference information into TF-IDF vectors in the same way. This converts the string data into numerical vectors, making it possible to perform calculations and comparisons.

[0394] Step 5:

[0395] The server uses the vectorized data to calculate the cosine similarity, which identifies items that are closest to the user's preferences. Items with a high cosine similarity are deemed to be suitable for the user's preferences.

[0396] Step 6:

[0397] The server recommends the most similar items to the user by listing them in descending order of cosine similarity, and selecting the top few items. Detailed information about these items (such as names, descriptions, and images) is then retrieved.

[0398] Step 7:

[0399] The server generates compliments for each selected item. The generated compliments highlight the item's quality and attractiveness based on its color and style. Furthermore, the tone of the compliments is adjusted based on the user's emotional information obtained from the emotion engine. For example, if the user is in a positive state, the tone of the compliments will be brighter and more positive.

[0400] Step 8:

[0401] The server sends the generated list of recommended items and compliments to the device. Specifically, it includes the list of recommended items, compliments for each item, and detailed information about each item (image, description, price, etc.). The device visually presents this information to the user so that the user can confirm it.

[0402] The above is the processing flow based on a specific embodiment of the present invention. This system allows users to find the best outfit based on their personal preferences and feel confident about the outfit they choose. It is also possible to provide feedback based on the user's emotions through an emotion engine.

[0403] 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.

[0404] 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.

[0405] 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.

[0406] [Second embodiment]

[0407] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0408] 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.

[0409] 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).

[0410] 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.

[0411] 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.

[0412] 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).

[0413] 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.

[0414] 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.

[0415] 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.

[0416] 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.

[0417] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0418] 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."

[0419] The present invention relates to a system for recommending optimal clothing based on a user's preference information and clothing data, and generating compliments for the selected clothing. Specific embodiments of the system are described below.

[0420] System configuration

[0421] The system of the present invention comprises a user terminal, a server, and a database. The user terminal provides an interface for users to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and catalog information of fashion items.

[0422] System Operation

[0423] 1. Enter user information

[0424] The user logs in to the system and inputs their preferences (preferred style, color, design, etc.), which are then sent to the server by the terminal.

[0425] 2. Obtaining User Data

[0426] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database.

[0427] 3. Acquiring fashion item data

[0428] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[0429] 4. Running the recommendation algorithm

[0430] The server converts the descriptions of fashion items into TF-IDF vectors, and also converts the user's preference information into vectors in the same way.

[0431] The server calculates the cosine similarity between these vectors to identify the items that most closely match the user's preferences, and recommends the most similar items to the user.

[0432] 5. Generating compliments

[0433] The server generates compliments for each selected item, building feedback that praises its quality and appeal based on color and style.

[0434] 6. Displaying the results

[0435] The server sends the generated list of recommended items and the compliments to the device.

[0436] The user terminal displays these results to the user, who can then check the suggested outfits and their feedback.

[0437] Specific examples

[0438] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[0439] 2. The server retrieves data such as the user's preferences and past purchase history from the database.

[0440] 3. The server retrieves information on all fashion items from the database and converts the item descriptions into TF-IDF vectors.

[0441] 4. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the item that best matches the user's preference.

[0442] 5. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!"

[0443] 6. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[0444] The above is an embodiment of the system of the present invention, which allows users to find the best outfit based on their preferences and feel confident in their choice.

[0445] The processing flow will be explained below.

[0446] Step 1:

[0447] A user logs in to the system using a terminal and inputs their preference information (e.g., preferred style, color, design), and the terminal transmits the input data to the server.

[0448] Step 2:

[0449] The server retrieves basic information about the user from the database, such as preference information and past purchase history corresponding to the user ID.

[0450] Step 3:

[0451] The server retrieves all the fashion item information from the database, including the name, description, color, style, etc. of the outfit.

[0452] Step 4:

[0453] The server converts the descriptions of fashion items into TF-IDF vectors, which is a process for expressing clothing features as numerical vectors.

[0454] Step 5:

[0455] The server also converts the user's preference information into a TF-IDF vector, which allows the user's preferences to be expressed as a numerical vector.

[0456] Step 6:

[0457] The server calculates the cosine similarity between these vectors, and uses the cosine similarity to identify the fashion item that most closely matches the user's preferences.

[0458] Step 7:

[0459] The server selects the top five fashion items from among the items with the highest similarity scores, thereby recommending the most suitable outfit for the user.

[0460] Step 8:

[0461] The server generates compliments for each selected item, generating feedback that praises its quality and appeal based on color and style.

[0462] Step 9:

[0463] The server sends the generated list of recommended items and the compliments to the device.

[0464] Step 10:

[0465] The device will display the recommended outfits and feedback to the user, and the user can check the suggested outfits and their feedback.

[0466] Example 1

[0467] 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."

[0468] In recent years, interest in fashion has increased, and many people want to choose clothes that suit their own tastes. However, choosing the best outfit from the vast number of fashion items available is not easy, and the process takes time and effort. It is also difficult to be confident in one's choices. Furthermore, existing recommendation systems may not be accurate enough to recommend items that precisely match a user's tastes.

[0469] 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.

[0470] In this invention, the server includes means for acquiring user preference information, means for acquiring user clothing data, means for recommending optimal clothing based on the user preference information and the user clothing data, and means for generating compliments for the recommended clothing using a generative AI model. This makes it possible to select optimal clothing based on the user's preferences and generate compliments to give the user confidence.

[0471] "User preference information" is information about personal fashion preferences such as the user's preferred style, color, and design.

[0472] "Owned clothing data" is information about clothing currently owned by the user, including the item name, description, color, style, and so on.

[0473] The "means for recommending optimal clothing" is a function for identifying and recommending the most suitable clothing for a user using the user's preference information and clothing data.

[0474] A "generative AI model" refers to an artificial intelligence algorithm that generates a response in natural language in response to a certain input (prompt), such as a generative language model.

[0475] The "means for generating compliments" is a function that uses a generative AI model to create feedback praising the good qualities and attractiveness of the selected outfit.

[0476] MODE FOR CARRYING OUT THE INVENTION

[0477] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Specific embodiments of the system are described below.

[0478] System configuration

[0479] This system consists of a user terminal, a server, and a database. The user terminal provides an interface for users to access the system and input their preference information. The server processes the user's preference information and clothing data and recommends the most suitable outfits. The database stores user information, clothing data, and catalog information of fashion items.

[0480] System Operation

[0481] 1. Enter user information

[0482] The user logs into the system using a terminal and inputs preference information (preferred style, color, design, etc.).

[0483] Example: A user inputs that they like "elegant style" and the color "blue." The device sends this information to the server through the interface.

[0484] 2. Obtaining User Data

[0485] The server retrieves basic information about the user (preferences, past purchase history, etc.) from a database. This information is retrieved using an SQL database (e.g., MySQL).

[0486] Example: The server executes an SQL query to retrieve user preferences such as "elegant" and "blue."

[0487] 3. Acquiring fashion item data

[0488] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[0489] Example: The server executes an SQL query to retrieve item information such as "an elegant blue dress" from a database.

[0490] 4. Running the recommendation algorithm

[0491] The server converts the descriptions of fashion items into TF-IDF vectors and also converts the user's preference information into vectors. This is done using Python and natural language processing libraries (e.g., scikit-learn, NLTK).

[0492] Cosine similarity is calculated to identify items that most closely match the user's preferences.

[0493] Example: The description of each item is vectorized using TF-IDF vectorization, and the cosine similarity is calculated to select dresses that are "elegant" and "blue."

[0494] 5. Generating compliments

[0495] The server generates a compliment for each selected item. To do this, it uses a generative AI model (e.g., GPT-3). Specifically, it generates a compliment for the selected item by inputting a prompt sentence into the generative AI model.

[0496] Example prompt: "Generate a compliment about this elegant blue dress."

[0497] Example: A generative AI model can generate compliments such as, "The dress you chose is very elegant and graceful!"

[0498] 6. Displaying the results

[0499] The server transmits the generated list of recommended items and the compliments to the terminal, which displays them.

[0500] Example: The server sends the generated item list and compliments in JSON format, and the web browser or smartphone app on the user's device displays "The blue dress you chose is very elegant and wonderful!"

[0501] In this way, the present invention can recommend optimal outfits based on a user's preferences, providing confidence in the choice. Furthermore, by using a generative AI model, the generation of compliments can be automated, providing users with rapid positive feedback.

[0502] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0503] Step 1: Enter your information

[0504] Users log in to the system using a terminal and enter their preference information. Specifically, the user enters their authentication information on the login screen and transitions to the customer preference interface. There, the user enters their preferred style, color, design, etc. into the input form.

[0505] Input: Authentication information, preference information (preferred style, color, design, etc.)

[0506] Output: User preference information is sent to the server

[0507] Step 2: Retrieving User Data

[0508] The server executes an SQL query to retrieve basic user information from the database, including past purchase history and registered preference information.

[0509] Input: User ID

[0510] Output: Basic information of the user (preference information, past purchase history, etc.)

[0511] Step 3: Obtaining fashion item data

[0512] The server retrieves all the fashion item information from the database. Specifically, the server executes SQL queries to retrieve the name, description, color, style, etc. of each fashion item.

[0513] Input: None

[0514] Output: Fashion item information (name, description, color, style, etc.)

[0515] Step 4: Run the recommendation algorithm

[0516] The server converts the descriptions of fashion items and user preference information into TF-IDF vectors. Specifically, it uses Python and natural language processing libraries (e.g., scikit-learn, NLTK) to vectorize the text data and calculate the cosine similarity between them to identify the most suitable items.

[0517] Input: Description of fashion item, user preference information

[0518] Output: A list of items that best fit the user's preferences

[0519] Step 5: Generate compliments

[0520] The server generates a compliment for the selected item using a generative AI model (e.g., GPT-3). Specifically, the server inputs a prompt sentence into the generative AI model to generate feedback.

[0521] Input: Recommended item

[0522] Output: Generated compliment

[0523] Step 6: View the results

[0524] The server sends the generated list of recommended items and compliments to the device, which then displays them. Specifically, the device parses the received JSON data and displays it to the user in a web browser or application.

[0525] Input: Recommended item list, compliments

[0526] Output: A list of recommended items and a compliment displayed to the user

[0527] (Application example 1)

[0528] 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."

[0529] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. In particular, conventional systems require users to select a vast number of fashion items themselves, and this has led to issues with insufficient appropriate recommendations and feedback. There is a need for systems that can improve user satisfaction and recommend optimal clothing in a rational and timely manner.

[0530] 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.

[0531] In this invention, the server includes a means for acquiring user preference information, a means for acquiring user clothing data, a means for recommending optimal clothing based on the user's preference information and the user's clothing data, a means for generating compliments for the recommended clothing, and a means for displaying these results. This allows users to easily find optimal clothing based on their preferences and feel confident in their selection. Furthermore, a recommendation algorithm using TF-IDF vectorization and cosine similarity calculations achieves highly accurate recommendations. Furthermore, by using prompt sentences that generate compliments using a generative AI model, more natural and persuasive feedback can be provided.

[0532] definition statement

[0533] "User preference information" is information about fashion such as the style, color, and design that the user prefers.

[0534] "Owned clothing data" is detailed information about clothing items owned by the user, including the name, description, color, style, etc. of each item.

[0535] "Means for recommending optimal clothing" refers to a method or device for selecting the most suitable clothing for a user using an algorithm based on the user's preference information and clothing data.

[0536] "Compliment generator" refers to an algorithm or mechanism for generating positive feedback on a recommended outfit.

[0537] "Means for displaying these results" refers to a method or device for displaying the recommended outfits and compliments on the outfits on the user terminal.

[0538] "TF-IDF vectorization" is a technique for quantifying text data as vectors, and is a technology that converts it into a vector representation by taking into account the importance of each word.

[0539] "Cosine similarity" is an index for measuring the similarity between two vectors, and is calculated based on the angle of the vectors.

[0540] "Generative AI model" refers to an algorithm or system that uses artificial intelligence to automatically create text or other artifacts.

[0541] A "prompt sentence" is an input sentence used to elicit a specific response from a generative AI model.

[0542] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Specific embodiments of this system are described below.

[0543] System configuration

[0544] The system of the present invention consists of the following main parts:

[0545] User terminal

[0546] server

[0547] Database

[0548] These parts work together to provide optimal outfit recommendations and compliment generation for the user.

[0549] User device functions

[0550] The user terminal provides an interface that receives input from the user and displays the results. The user logs in using the terminal and inputs preference information (preferred style, color, design, etc.). The terminal sends this information to the server.

[0551] Server Features

[0552] The server has the following functions:

[0553] 1. Data Acquisition:

[0554] The server obtains the user's preference information and clothing data from the database.

[0555] 2. Data Processing:

[0556] The server converts preference information and clothing description data into TF-IDF vectors and calculates the cosine similarity to select the most suitable outfit.

[0557] 3. Compliment Generation:

[0558] The server generates natural and complimenting words for the selected outfit, using a generative AI model to input prompts and create feedback accordingly.

[0559] Database

[0560] The database stores user information, clothing data, and fashion item catalog information, allowing the server to quickly and efficiently retrieve the required information.

[0561] Example of operation

[0562] 1. Log in and enter your preferences:

[0563] The user logs in and inputs his / her preference information of "elegant style" and "blue."

[0564] 2. Retrieving data from the database:

[0565] The server retrieves the user's preference information, past purchase history, and clothing data from the database.

[0566] 3. Clothing recommendations:

[0567] The server uses TF-IDF vectorization and cosine similarity calculation to select the item that best matches the user's preferences, such as an "elegant blue dress."

[0568] 4. Generating compliments:

[0569] The server uses a generative AI model to generate compliments such as, "The dress you chose is very elegant and graceful!"

[0570] Hardware and software used

[0571] Server: The server on which the Python program runs (using a server framework such as Flask).

[0572] User device: Smartphone (including iOS and Android).

[0573] Examples of prompt statements

[0574] Here is an example of a prompt that can be fed into a generative AI model:

[0575] text

[0576] "Please tell me the algorithm for a system that recommends the most suitable outfit based on the user's preference information and clothing data, and generates compliments for the selected outfit. The user's preference information is "elegant style" and "blue." Please also take into account past purchase data and clothing data."

[0577] The above is a specific embodiment for carrying out the invention. This system allows a user to find the best outfit based on their own preferences and wear the outfit they have chosen with confidence.

[0578] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0579] Processing Step Description

[0580] Step 1:

[0581] User login and preference input

[0582] The user logs in to the device and inputs their preferences (style, color, design, etc.), which are then sent from the device to the server.

[0583] Input: User login information, preferences

[0584] Output: Preference information sent to the server

[0585] Step 2:

[0586] Retrieving user data from the database

[0587] The server accesses the database to obtain the user's preferences, past purchase history, and clothing data, thereby gathering all data related to the user.

[0588] Input: Database query, user identity

[0589] Output: Acquired user data, owned clothing data

[0590] Step 3:

[0591] Obtaining information on fashion items

[0592] The server retrieves information about all fashion items (item name, description, color, style, etc.) from the database and creates a list.

[0593] Input: Database query

[0594] Output: A list of retrieved fashion items

[0595] Step 4:

[0596] TF-IDF vectorization of data

[0597] The server converts the acquired user preference information and fashion item description data into TF-IDF vectors, which converts each piece of data into a numerical value and puts it in a format suitable for calculations.

[0598] Input: User preference information, fashion item description data

[0599] Output: Vectorized user preference information and item description data

[0600] Step 5:

[0601] Calculating cosine similarity

[0602] The server calculates the cosine similarity using the TF-IDF vectorized data, and selects the item that most closely matches the user's preferences.

[0603] Input: Vectorized user preference information and item description data

[0604] Output: A list of items with high similarity

[0605] Step 6:

[0606] Selection of recommended items

[0607] Based on the results of the cosine similarity calculation, the server selects several items with the highest similarity and generates a list of recommended items.

[0608] Input: List of highly similar items

[0609] Output: Recommended item list

[0610] Step 7:

[0611] Generating compliments

[0612] The server generates compliments for the selected recommended items using a generative AI model. The server inputs prompts to the model to obtain natural and positive compliments.

[0613] Input: Recommended item list, prompt text

[0614] Output: Compliment generation

[0615] Step 8:

[0616] Displaying the results

[0617] The server sends a list of recommended items and compliments to the user's device, which then displays them. The user can check the suggested outfit and the feedback.

[0618] Input: Recommended item list, compliments

[0619] Output: The result displayed on the user's terminal

[0620] Through the above processing steps, the user can easily find the best outfit based on their preferences and feel confident in their choice.

[0621] 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.

[0622] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide feedback based on the user's emotions. Specific embodiments of the system are described below.

[0623] System configuration

[0624] The system of the present invention comprises a user terminal, a server, a database, and an emotion engine. The user terminal provides an interface for the user to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and fashion item catalog information. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, input text, etc., and sends the results to the server.

[0625] System Operation

[0626] 1. Enter user information

[0627] A user logs in to the system and inputs their preference information (e.g., preferred style, color, design, etc.). The terminal sends this information to the server.

[0628] 2. Acquiring emotional information

[0629] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. This emotion information is sent to the server.

[0630] 3. Obtaining User Data

[0631] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database.

[0632] 4. Acquiring fashion item data

[0633] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[0634] 5. Running the recommendation algorithm

[0635] The server converts the descriptions of fashion items into TF-IDF vectors, and also converts the user's preference information into vectors in the same way.

[0636] The server calculates the cosine similarity between these vectors to identify the items that most closely match the user's preferences, and recommends the most similar items to the user.

[0637] 6. Generating compliments

[0638] The server generates compliments for each selected item, building feedback that praises its quality and appeal based on color and style.

[0639] The server can adjust the content and tone of the compliment based on the user's emotional information obtained from the emotion engine.

[0640] 7. Displaying the results

[0641] The server sends the generated list of recommended items and the compliments to the device.

[0642] The user terminal displays these results to the user, who can then check the suggested outfits and their feedback.

[0643] Specific examples

[0644] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[0645] 2. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The emotion information is sent to the server.

[0646] 3. The server retrieves data such as the user's preferences and past purchase history from the database.

[0647] 4. The server retrieves all fashion item information from the database and converts the item descriptions into TF-IDF vectors.

[0648] 5. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the item that best matches the user's preference.

[0649] 6. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!". Based on the emotional information that the user is pleased, the server adjusts the tone of the compliment to be more positive.

[0650] 7. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[0651] The above is an embodiment of the system of the present invention. This system allows users to find the best outfit based on their preferences and feel confident in their choice. Furthermore, by incorporating an emotion engine, it is possible to provide feedback according to the user's emotions.

[0652] The processing flow will be explained below.

[0653] Step 1:

[0654] A user logs in to the system using a terminal and inputs his / her preference information (e.g., preferred style, color, design, etc.). The terminal transmits the input data to the server.

[0655] Step 2:

[0656] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. This emotion information is then sent to the server.

[0657] Step 3:

[0658] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database. Specifically, it issues a DB query to retrieve related data.

[0659] Step 4:

[0660] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database. The retrieved data is then used in the recommendation algorithm.

[0661] Step 5:

[0662] The server converts the descriptions of fashion items into TF-IDF (Term Frequency-Inverse Document Frequency) vectors, which is a process of representing each item description as a numerical vector.

[0663] Step 6:

[0664] The server also converts the user's preference information into a TF-IDF vector, which allows the user's preferences to be expressed as a numerical vector.

[0665] Step 7:

[0666] The server calculates the cosine similarity between these vectors, and uses the cosine similarity to identify the fashion item that most closely matches the user's preferences.

[0667] Step 8:

[0668] The server selects the top five fashion items from the items with the highest similarity scores, and the selected items are recommended to the user.

[0669] Step 9:

[0670] The server generates compliments for each selected item, generates feedback praising its goodness and attractiveness based on its color and style, and adjusts the content and tone of the compliments based on the user's emotional information obtained from the emotion engine.

[0671] Step 10:

[0672] The server sends the generated list of recommended items and the compliments to the device.

[0673] Step 11:

[0674] The device will display the recommended outfits and feedback to the user, and the user can check the suggested outfits and their feedback.

[0675] Specific examples

[0676] If a user logs in and inputs that they like "elegant style" and the color "blue," the system sends this preference information to the server. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The emotion information is sent to the server. The server retrieves the user's preference information and data such as past purchase history from the database. The server retrieves information on all fashion items from the database and converts the item descriptions into TF-IDF vectors. The server also converts the user's preference information into vectors and calculates cosine similarity to identify the item that best matches the user's preferences. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!" Based on the emotional information that the user is happy, the system adjusts the tone of the compliment to be more positive. The server sends the recommended item list and the compliment to the terminal, which displays them. The user can review the results and wear the outfit they chose with confidence.

[0677] Example 2

[0678] 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."

[0679] Conventional fashion recommendation systems can only suggest outfits based on the user's preference information and clothing data, but they have the problem of not being able to provide feedback that takes into account the user's emotions. This means that the system is unable to fully improve the user's feelings and satisfaction with the suggested outfits, resulting in limited effectiveness.

[0680] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0681] In this invention, the server includes means for acquiring user preference information, means for acquiring owned clothing data, and means for recommending optimal clothing based on the user preference information and owned clothing data. This makes it possible to suggest optimal clothing based on the user preference information and owned clothing data, thereby improving the user's feelings and satisfaction with the suggested clothing.

[0682] In addition, in this invention, the server includes means for recognizing emotions from the user's facial expressions, tone of voice, input text, etc., means for adjusting the content and tone of the compliments based on the emotional information, and means for generating compliments for the recommended outfit. This makes it possible to provide feedback that takes into consideration the user's emotions, further increasing user satisfaction.

[0683] "User preference information" refers to information about personal preferences such as the user's preferred style, color, and design.

[0684] "Owned clothing data" is information about clothing owned by the user, and includes detailed data such as the type, color, style, and brand of the item.

[0685] "Recommendation means" is a system function that selects optimal clothing based on the user's preference information and clothing data and presents it to the user.

[0686] The "means for generating compliments" is a function of the system that generates feedback praising the goodness and attractiveness of the recommended outfit.

[0687] "Means for recognizing emotions" refers to the system's function of analyzing and determining the user's emotions from the user's facial expressions, tone of voice, input text, etc.

[0688] The "means for adjusting the content and tone of the compliment based on emotional information" is a system function for adjusting the content and tone of the generated compliment to the user's current emotional state based on information obtained from the means for recognizing emotions.

[0689] "TF-IDF vectorization means" is a system function that calculates the importance of words using text data and converts it into a vector representation.

[0690] The "means for calculating cosine similarity" is a system function that measures the similarity between vectorized data using cosine angles and calculates the similarity.

[0691] MODE FOR CARRYING OUT THE INVENTION

[0692] This invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide feedback based on the user's emotions. Specific embodiments of this system are described below.

[0693] System configuration

[0694] The system of the present invention comprises a user terminal, a server, a database, and an emotion engine. The user terminal provides an interface for the user to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and fashion item catalog information. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, input text, etc., and sends the results to the server.

[0695] Hardware and Software Use

[0696] User terminals include devices such as PCs, smartphones, and tablets. The server is a server computer with high-performance computing capabilities, and the database can use a commercial relational database system (e.g., MySQL or PostgreSQL).

[0697] The emotion engine is realized using a model that uses deep learning technology (e.g., an emotion analysis model). For example, a camera and image recognition technology are required to analyze a user's facial expressions, and a microphone and voice analysis technology are required to analyze the tone of voice. A deep learning framework (e.g., TensorFlow, PyTorch) is used to implement the emotion engine.

[0698] Natural language processing techniques are used for the recommendation algorithm and compliment generation. Specifically, "TF-IDF vectorization" and "cosine similarity calculation" are used to process text data. These processes can be performed using the Python "scikit-learn" library. In addition, generative AI models (e.g., GPT models) are used for natural language generation.

[0699] Specific examples

[0700] A specific example of the operation of the system is shown below.

[0701] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[0702] 2. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. This emotion information is sent to the server.

[0703] 3. The server retrieves data such as the user's preferences and past purchase history from the database.

[0704] 4. The server retrieves all fashion item information from the database and converts the item descriptions into TF-IDF vectors.

[0705] 5. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the most suitable item.

[0706] 6. For example, if an "elegant blue dress" is selected, the server generates a compliment for this item, such as "The dress you chose is very elegant and graceful!". Based on the emotional information that the user is pleased, the tone of the compliment is adjusted to be more positive.

[0707] 7. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[0708] Prompt Sentence Examples

[0709] "Recommend the most suitable fashion items based on the user's preferences and past purchase history. Also, generate compliments for the selected items and adjust the tone according to the user's emotions."

[0710] As a result, this system utilizes the user's preference information and emotional information to recommend the most suitable outfits and provide the user with a high level of satisfaction.

[0711] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0712] Step 1:

[0713] A user logs in to the system. The user enters login information, which the terminal sends to the server. The server then refers to a database to authenticate the login information and authenticates the user. If authentication is successful, a screen for entering the user's preferences is displayed.

[0714] Step 2:

[0715] The user inputs their preference information. For example, they select "elegant style" and "blue color." The terminal sends this preference information to the server. The server stores the input preference information in a database. This preference information becomes the input for subsequent processing.

[0716] Step 3:

[0717] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. For example, facial expression data is acquired from a camera and analyzed using a deep learning model to recognize "joy." This emotion information is sent to the server, which temporarily stores the received emotion information.

[0718] Step 4:

[0719] The server retrieves basic information about the user (preferences, past purchase history, etc.) from a database. The server issues SQL queries to extract this information from the database and stores it in memory. This basic information becomes the input for the recommendation algorithm.

[0720] Step 5:

[0721] The server retrieves all fashion item information from the database. Specifically, detailed data such as item name, description, color, and style are extracted using SQL queries. The server receives this information and prepares to convert each item description into a TF-IDF vector.

[0722] Step 6:

[0723] The server converts the fashion item descriptions into TF-IDF vectors. Using the Python "scikit-learn" library, the item descriptions are converted into numerical vectors. User preference information is also vectorized in the same way. These vectors are used as inputs for the cosine similarity calculation.

[0724] Step 7:

[0725] The server calculates the cosine similarity and identifies the item that most closely matches the user's preferences. It outputs a similarity score by calculating the cosine similarity between vectors. It selects the item with the highest score and creates a list of recommended items.

[0726] Step 8:

[0727] The server generates compliments for the selected item. Using a generative AI model (e.g., a GPT model), it generates text that praises the good points and attractiveness of the selected item. It includes specific feedback using color and style information. It also adjusts the tone of the compliments based on emotional information from an emotion engine.

[0728] Step 9:

[0729] The server sends the generated list of recommended items and compliments to the device. The generated text data and item information are sent to the device in JSON format. The device displays these results to the user. The user can check the suggested outfits and their feedback and wear the outfits they have chosen with confidence.

[0730] (Application example 2)

[0731] 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."

[0732] Conventional fashion recommendation systems can recommend clothing based on a user's preferences, but they are unable to provide emotional feedback on the selection. This leaves users with little confidence in the recommended clothing, resulting in low satisfaction with their selection. Furthermore, because previous systems do not take the user's emotional state into account, they are unable to provide appropriate feedback that reflects the user's psychological state at any given time. Therefore, the objective of this invention is to provide a system that takes into account the user's preference information and emotional information, allowing the user to select fashion items with greater satisfaction and confidence.

[0733] 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.

[0734] In this invention, the server includes means for acquiring user preference information, means for acquiring owned clothing data, means for recommending optimal clothing based on the user preference information and the owned clothing data, means for generating compliments for the recommended clothing, means for acquiring user emotion information and adjusting the tone of the compliments based on the emotion information, and means for displaying the optimal clothing after it has been recommended. This makes it possible not only to recommend optimal clothing based on the user preference information, but also to generate compliments for the selected clothing and provide feedback according to the user's emotions.

[0735] "User preference information" is data relating to the user's preferences such as style, color, and design.

[0736] "Owned clothing data" is data about clothing owned by the user, and includes information such as item name, color, style, and description.

[0737] The "means for recommending optimal clothing" is a means for identifying and recommending the most suitable clothing for a user based on the user's preference information and clothing data owned by the user.

[0738] The "means for generating compliments" is a means for generating feedback praising the goodness and attractiveness of the recommended outfit.

[0739] "User emotion information" is data related to emotions analyzed from the user's facial expression, tone of voice, input text, and the like.

[0740] The "means for adjusting the tone of the compliment" is a means for adjusting the content and tone of the generated compliment based on the acquired emotional information of the user.

[0741] The "means for displaying the optimal outfit after it has been recommended" is a means for visually presenting information about the selected outfit and feedback thereon to the user.

[0742] This system recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it provides feedback based on the user's emotions. This system can be implemented as an application that runs on smart glasses, head-mounted displays, and smartphones.

[0743] Hardware and software used

[0744] Hardware:

[0745] Smart Glasses

[0746] head-mounted display

[0747] Smartphone

[0748] software:

[0749] Python environment

[0750] TF-IDF Vectorizer (scikit-learn)

[0751] Cosine similarity calculation module (scikit-learn)

[0752] Sentiment analysis engine (AWS Rekognition, Google Cloud Vision, or TextBlob)

[0753] Data processing and calculation

[0754] 1. Data Vectorization:

[0755] The server converts the user's preference information and the description of the clothing data they own into a TF-IDF vector, which allows the data features to be expressed as a numerical vector and made comparable.

[0756] 2. Similarity calculation:

[0757] The server calculates the cosine similarity between the vectorized user preference information and the clothing data, and identifies the items that best match the preferences.

[0758] 3. Compliment Generation:

[0759] The server generates feedback for the selected item, praising its merits and attractiveness, and adjusts the content and tone of the praise based on the user's emotional information obtained from the emotion engine.

[0760] 4. Displaying the results:

[0761] The user device visually presents the recommended item list and compliments received from the server to the user, allowing the user to check the suggested outfit and receive feedback on it.

[0762] Specific examples

[0763] For example, if a user inputs that they like "elegant style" and "blue," the system processes this preference information. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The server vectorizes the user's preference information and clothing data, calculates the cosine similarity, and identifies the item that best matches the user's preference.

[0764] For a selected item (e.g., "an elegant blue dress"), the server generates a compliment such as "The dress you chose is very elegant and graceful!" Furthermore, based on the emotional information that the user is pleased, the tone of the compliment is adjusted to be more positive.

[0765] Prompt Sentence Examples

[0766] Here are some example input prompts for a generative AI model:

[0767] "A system that recommends the most suitable outfit based on the user's preference information and clothing data, and generates compliments for the selected outfit. Furthermore, please combine it with an emotion engine that recognizes the user's emotions and provides emotion-based feedback."

[0768] The above is a specific embodiment of the present invention. This system allows users to find the perfect outfit based on their preferences and feel confident in their choice. Furthermore, by incorporating an emotion engine, it is possible to provide feedback based on the user's emotions.

[0769] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0770] Step 1:

[0771] The user logs in to the device and inputs their preferences (e.g., preferred style, color, design, etc.). The user's input information is acquired by the device and sent to the server. The input data includes the user's preferred style and color, as well as past purchase history. The server stores this information in a database.

[0772] Step 2:

[0773] The server activates an emotion engine and analyzes the user's facial expressions, tone of voice, input text, etc. to obtain emotional information. Specifically, the user's facial expressions and voice data obtained using a camera or microphone are sent to the emotion analysis engine, which extracts emotional information as positive, negative, or neutral. This emotional information is then sent to the server and stored in a database.

[0774] Step 3:

[0775] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database. At the same time, it also retrieves data about all fashion items from the database. This data includes item names, descriptions, colors, styles, etc.

[0776] Step 4:

[0777] The server converts the descriptions of fashion items into TF-IDF vectors. Then, it converts the user's preference information into TF-IDF vectors in the same way. This converts the string data into numerical vectors, making it possible to perform calculations and comparisons.

[0778] Step 5:

[0779] The server uses the vectorized data to calculate the cosine similarity, which identifies items that are closest to the user's preferences. Items with a high cosine similarity are deemed to be suitable for the user's preferences.

[0780] Step 6:

[0781] The server recommends the most similar items to the user by listing them in descending order of cosine similarity, and selecting the top few items. Detailed information about these items (such as names, descriptions, and images) is then retrieved.

[0782] Step 7:

[0783] The server generates compliments for each selected item. The generated compliments highlight the item's quality and attractiveness based on its color and style. Furthermore, the tone of the compliments is adjusted based on the user's emotional information obtained from the emotion engine. For example, if the user is in a positive state, the tone of the compliments will be brighter and more positive.

[0784] Step 8:

[0785] The server sends the generated list of recommended items and compliments to the device. Specifically, it includes the list of recommended items, compliments for each item, and detailed information about each item (image, description, price, etc.). The device visually presents this information to the user so that the user can confirm it.

[0786] The above is the processing flow based on a specific embodiment of the present invention. This system allows users to find the best outfit based on their personal preferences and feel confident about the outfit they choose. It is also possible to provide feedback based on the user's emotions through an emotion engine.

[0787] 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.

[0788] 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.

[0789] 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.

[0790] [Third embodiment]

[0791] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0792] 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.

[0793] 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).

[0794] 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.

[0795] 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.

[0796] 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).

[0797] 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.

[0798] 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.

[0799] 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.

[0800] 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.

[0801] 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.

[0802] 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."

[0803] The present invention relates to a system for recommending optimal clothing based on a user's preference information and clothing data, and generating compliments for the selected clothing. Specific embodiments of the system are described below.

[0804] System configuration

[0805] The system of the present invention comprises a user terminal, a server, and a database. The user terminal provides an interface for users to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and catalog information of fashion items.

[0806] System Operation

[0807] 1. Enter user information

[0808] The user logs in to the system and inputs their preferences (preferred style, color, design, etc.), which are then sent to the server by the terminal.

[0809] 2. Obtaining User Data

[0810] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database.

[0811] 3. Acquiring fashion item data

[0812] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[0813] 4. Running the recommendation algorithm

[0814] The server converts the descriptions of fashion items into TF-IDF vectors, and also converts the user's preference information into vectors in the same way.

[0815] The server calculates the cosine similarity between these vectors to identify the items that most closely match the user's preferences, and recommends the most similar items to the user.

[0816] 5. Generating compliments

[0817] The server generates compliments for each selected item, building feedback that praises its quality and appeal based on color and style.

[0818] 6. Displaying the results

[0819] The server sends the generated list of recommended items and the compliments to the device.

[0820] The user terminal displays these results to the user, who can then check the suggested outfits and their feedback.

[0821] Specific examples

[0822] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[0823] 2. The server retrieves data such as the user's preferences and past purchase history from the database.

[0824] 3. The server retrieves information on all fashion items from the database and converts the item descriptions into TF-IDF vectors.

[0825] 4. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the item that best matches the user's preference.

[0826] 5. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!"

[0827] 6. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[0828] The above is an embodiment of the system of the present invention, which allows users to find the best outfit based on their preferences and feel confident in their choice.

[0829] The processing flow will be explained below.

[0830] Step 1:

[0831] A user logs in to the system using a terminal and inputs their preference information (e.g., preferred style, color, design), and the terminal transmits the input data to the server.

[0832] Step 2:

[0833] The server retrieves basic information about the user from the database, such as preference information and past purchase history corresponding to the user ID.

[0834] Step 3:

[0835] The server retrieves all the fashion item information from the database, including the name, description, color, style, etc. of the outfit.

[0836] Step 4:

[0837] The server converts the descriptions of fashion items into TF-IDF vectors, which is a process for expressing clothing features as numerical vectors.

[0838] Step 5:

[0839] The server also converts the user's preference information into a TF-IDF vector, which allows the user's preferences to be expressed as a numerical vector.

[0840] Step 6:

[0841] The server calculates the cosine similarity between these vectors, and uses the cosine similarity to identify the fashion item that most closely matches the user's preferences.

[0842] Step 7:

[0843] The server selects the top five fashion items from among the items with the highest similarity scores, thereby recommending the most suitable outfit for the user.

[0844] Step 8:

[0845] The server generates compliments for each selected item, generating feedback that praises its quality and appeal based on color and style.

[0846] Step 9:

[0847] The server sends the generated list of recommended items and the compliments to the device.

[0848] Step 10:

[0849] The device will display the recommended outfits and feedback to the user, and the user can check the suggested outfits and their feedback.

[0850] Example 1

[0851] 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."

[0852] In recent years, interest in fashion has increased, and many people want to choose clothes that suit their own tastes. However, choosing the best outfit from the vast number of fashion items available is not easy, and the process takes time and effort. It is also difficult to be confident in one's choices. Furthermore, existing recommendation systems may not be accurate enough to recommend items that precisely match a user's tastes.

[0853] 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.

[0854] In this invention, the server includes means for acquiring user preference information, means for acquiring user clothing data, means for recommending optimal clothing based on the user preference information and the user clothing data, and means for generating compliments for the recommended clothing using a generative AI model. This makes it possible to select optimal clothing based on the user's preferences and generate compliments to give the user confidence.

[0855] "User preference information" is information about personal fashion preferences such as the user's preferred style, color, and design.

[0856] "Owned clothing data" is information about clothing currently owned by the user, including the item name, description, color, style, and so on.

[0857] The "means for recommending optimal clothing" is a function for identifying and recommending the most suitable clothing for a user using the user's preference information and clothing data.

[0858] A "generative AI model" refers to an artificial intelligence algorithm that generates a response in natural language in response to a certain input (prompt), such as a generative language model.

[0859] The "means for generating compliments" is a function that uses a generative AI model to create feedback praising the good qualities and attractiveness of the selected outfit.

[0860] MODE FOR CARRYING OUT THE INVENTION

[0861] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Specific embodiments of the system are described below.

[0862] System configuration

[0863] This system consists of a user terminal, a server, and a database. The user terminal provides an interface for users to access the system and input their preference information. The server processes the user's preference information and clothing data and recommends the most suitable outfits. The database stores user information, clothing data, and catalog information of fashion items.

[0864] System Operation

[0865] 1. Enter user information

[0866] The user logs into the system using a terminal and inputs preference information (preferred style, color, design, etc.).

[0867] Example: A user inputs that they like "elegant style" and the color "blue." The device sends this information to the server through the interface.

[0868] 2. Obtaining User Data

[0869] The server retrieves basic information about the user (preferences, past purchase history, etc.) from a database. This information is retrieved using an SQL database (e.g., MySQL).

[0870] Example: The server executes an SQL query to retrieve user preferences such as "elegant" and "blue."

[0871] 3. Acquiring fashion item data

[0872] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[0873] Example: The server executes an SQL query to retrieve item information such as "an elegant blue dress" from a database.

[0874] 4. Running the recommendation algorithm

[0875] The server converts the descriptions of fashion items into TF-IDF vectors and also converts the user's preference information into vectors. This is done using Python and natural language processing libraries (e.g., scikit-learn, NLTK).

[0876] Cosine similarity is calculated to identify items that most closely match the user's preferences.

[0877] Example: The description of each item is vectorized using TF-IDF vectorization, and the cosine similarity is calculated to select dresses that are "elegant" and "blue."

[0878] 5. Generating compliments

[0879] The server generates a compliment for each selected item. To do this, it uses a generative AI model (e.g., GPT-3). Specifically, it generates a compliment for the selected item by inputting a prompt sentence into the generative AI model.

[0880] Example prompt: "Generate a compliment about this elegant blue dress."

[0881] Example: A generative AI model can generate compliments such as, "The dress you chose is very elegant and graceful!"

[0882] 6. Displaying the results

[0883] The server transmits the generated list of recommended items and the compliments to the terminal, which displays them.

[0884] Example: The server sends the generated item list and compliments in JSON format, and the web browser or smartphone app on the user's device displays "The blue dress you chose is very elegant and wonderful!"

[0885] In this way, the present invention can recommend optimal outfits based on a user's preferences, providing confidence in the choice. Furthermore, by using a generative AI model, the generation of compliments can be automated, providing users with rapid positive feedback.

[0886] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0887] Step 1: Enter your information

[0888] Users log in to the system using a terminal and enter their preference information. Specifically, the user enters their authentication information on the login screen and transitions to the customer preference interface. There, the user enters their preferred style, color, design, etc. into the input form.

[0889] Input: Authentication information, preference information (preferred style, color, design, etc.)

[0890] Output: User preference information is sent to the server

[0891] Step 2: Retrieving User Data

[0892] The server executes an SQL query to retrieve basic user information from the database, including past purchase history and registered preference information.

[0893] Input: User ID

[0894] Output: Basic information of the user (preference information, past purchase history, etc.)

[0895] Step 3: Obtaining fashion item data

[0896] The server retrieves all the fashion item information from the database. Specifically, the server executes SQL queries to retrieve the name, description, color, style, etc. of each fashion item.

[0897] Input: None

[0898] Output: Fashion item information (name, description, color, style, etc.)

[0899] Step 4: Run the recommendation algorithm

[0900] The server converts the descriptions of fashion items and user preference information into TF-IDF vectors. Specifically, it uses Python and natural language processing libraries (e.g., scikit-learn, NLTK) to vectorize the text data and calculate the cosine similarity between them to identify the most suitable items.

[0901] Input: Description of fashion item, user preference information

[0902] Output: A list of items that best fit the user's preferences

[0903] Step 5: Generate compliments

[0904] The server generates a compliment for the selected item using a generative AI model (e.g., GPT-3). Specifically, the server inputs a prompt sentence into the generative AI model to generate feedback.

[0905] Input: Recommended item

[0906] Output: Generated compliment

[0907] Step 6: View the results

[0908] The server sends the generated list of recommended items and compliments to the device, which then displays them. Specifically, the device parses the received JSON data and displays it to the user in a web browser or application.

[0909] Input: Recommended item list, compliments

[0910] Output: A list of recommended items and a compliment displayed to the user

[0911] (Application example 1)

[0912] 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."

[0913] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. In particular, conventional systems require users to select a vast number of fashion items themselves, and this has led to issues with insufficient appropriate recommendations and feedback. There is a need for systems that can improve user satisfaction and recommend optimal clothing in a rational and timely manner.

[0914] 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.

[0915] In this invention, the server includes a means for acquiring user preference information, a means for acquiring user clothing data, a means for recommending optimal clothing based on the user's preference information and the user's clothing data, a means for generating compliments for the recommended clothing, and a means for displaying these results. This allows users to easily find optimal clothing based on their preferences and feel confident in their selection. Furthermore, a recommendation algorithm using TF-IDF vectorization and cosine similarity calculations achieves highly accurate recommendations. Furthermore, by using prompt sentences that generate compliments using a generative AI model, more natural and persuasive feedback can be provided.

[0916] definition statement

[0917] "User preference information" is information about fashion such as the style, color, and design that the user prefers.

[0918] "Owned clothing data" is detailed information about clothing items owned by the user, including the name, description, color, style, etc. of each item.

[0919] "Means for recommending optimal clothing" refers to a method or device for selecting the most suitable clothing for a user using an algorithm based on the user's preference information and clothing data.

[0920] "Compliment generator" refers to an algorithm or mechanism for generating positive feedback on a recommended outfit.

[0921] "Means for displaying these results" refers to a method or device for displaying the recommended outfits and compliments on the outfits on the user terminal.

[0922] "TF-IDF vectorization" is a technique for quantifying text data as vectors, and is a technology that converts it into a vector representation by taking into account the importance of each word.

[0923] "Cosine similarity" is an index for measuring the similarity between two vectors, and is calculated based on the angle of the vectors.

[0924] "Generative AI model" refers to an algorithm or system that uses artificial intelligence to automatically create text or other artifacts.

[0925] A "prompt sentence" is an input sentence used to elicit a specific response from a generative AI model.

[0926] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Specific embodiments of this system are described below.

[0927] System configuration

[0928] The system of the present invention consists of the following main parts:

[0929] User terminal

[0930] server

[0931] Database

[0932] These parts work together to provide optimal outfit recommendations and compliment generation for the user.

[0933] User device functions

[0934] The user terminal provides an interface that receives input from the user and displays the results. The user logs in using the terminal and inputs preference information (preferred style, color, design, etc.). The terminal sends this information to the server.

[0935] Server Features

[0936] The server has the following functions:

[0937] 1. Data Acquisition:

[0938] The server obtains the user's preference information and clothing data from the database.

[0939] 2. Data Processing:

[0940] The server converts preference information and clothing description data into TF-IDF vectors and calculates the cosine similarity to select the most suitable outfit.

[0941] 3. Compliment Generation:

[0942] The server generates natural and complimenting words for the selected outfit, using a generative AI model to input prompts and create feedback accordingly.

[0943] Database

[0944] The database stores user information, clothing data, and fashion item catalog information, allowing the server to quickly and efficiently retrieve the required information.

[0945] Example of operation

[0946] 1. Log in and enter your preferences:

[0947] The user logs in and inputs his / her preference information of "elegant style" and "blue."

[0948] 2. Retrieving data from the database:

[0949] The server retrieves the user's preference information, past purchase history, and clothing data from the database.

[0950] 3. Clothing recommendations:

[0951] The server uses TF-IDF vectorization and cosine similarity calculation to select the item that best matches the user's preferences, such as an "elegant blue dress."

[0952] 4. Generating compliments:

[0953] The server uses a generative AI model to generate compliments such as, "The dress you chose is very elegant and graceful!"

[0954] Hardware and software used

[0955] Server: The server on which the Python program runs (using a server framework such as Flask).

[0956] User device: Smartphone (including iOS and Android).

[0957] Examples of prompt statements

[0958] Here is an example of a prompt that can be fed into a generative AI model:

[0959] text

[0960] "Please tell me the algorithm for a system that recommends the most suitable outfit based on the user's preference information and clothing data, and generates compliments for the selected outfit. The user's preference information is "elegant style" and "blue." Please also take into account past purchase data and clothing data."

[0961] The above is a specific embodiment for carrying out the invention. This system allows a user to find the best outfit based on their own preferences and wear the outfit they have chosen with confidence.

[0962] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0963] Processing Step Description

[0964] Step 1:

[0965] User login and preference input

[0966] The user logs in to the device and inputs their preferences (style, color, design, etc.), which are then sent from the device to the server.

[0967] Input: User login information, preferences

[0968] Output: Preference information sent to the server

[0969] Step 2:

[0970] Retrieving user data from the database

[0971] The server accesses the database to obtain the user's preferences, past purchase history, and clothing data, thereby gathering all data related to the user.

[0972] Input: Database query, user identity

[0973] Output: Acquired user data, owned clothing data

[0974] Step 3:

[0975] Obtaining information on fashion items

[0976] The server retrieves information about all fashion items (item name, description, color, style, etc.) from the database and creates a list.

[0977] Input: Database query

[0978] Output: A list of retrieved fashion items

[0979] Step 4:

[0980] TF-IDF vectorization of data

[0981] The server converts the acquired user preference information and fashion item description data into TF-IDF vectors, which converts each piece of data into a numerical value and puts it in a format suitable for calculations.

[0982] Input: User preference information, fashion item description data

[0983] Output: Vectorized user preference information and item description data

[0984] Step 5:

[0985] Calculating cosine similarity

[0986] The server calculates the cosine similarity using the TF-IDF vectorized data, and selects the item that most closely matches the user's preferences.

[0987] Input: Vectorized user preference information and item description data

[0988] Output: A list of items with high similarity

[0989] Step 6:

[0990] Selection of recommended items

[0991] Based on the results of the cosine similarity calculation, the server selects several items with the highest similarity and generates a list of recommended items.

[0992] Input: List of highly similar items

[0993] Output: Recommended item list

[0994] Step 7:

[0995] Generating compliments

[0996] The server generates compliments for the selected recommended items using a generative AI model. The server inputs prompts to the model to obtain natural and positive compliments.

[0997] Input: Recommended item list, prompt text

[0998] Output: Compliment generation

[0999] Step 8:

[1000] Displaying the results

[1001] The server sends a list of recommended items and compliments to the user's device, which then displays them. The user can check the suggested outfit and the feedback.

[1002] Input: Recommended item list, compliments

[1003] Output: The result displayed on the user's terminal

[1004] Through the above processing steps, the user can easily find the best outfit based on their preferences and feel confident in their choice.

[1005] 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.

[1006] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide feedback based on the user's emotions. Specific embodiments of the system are described below.

[1007] System configuration

[1008] The system of the present invention comprises a user terminal, a server, a database, and an emotion engine. The user terminal provides an interface for the user to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and fashion item catalog information. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, input text, etc., and sends the results to the server.

[1009] System Operation

[1010] 1. Enter user information

[1011] A user logs in to the system and inputs their preference information (e.g., preferred style, color, design, etc.). The terminal sends this information to the server.

[1012] 2. Acquiring emotional information

[1013] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. This emotion information is sent to the server.

[1014] 3. Obtaining User Data

[1015] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database.

[1016] 4. Acquiring fashion item data

[1017] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[1018] 5. Running the recommendation algorithm

[1019] The server converts the descriptions of fashion items into TF-IDF vectors, and also converts the user's preference information into vectors in the same way.

[1020] The server calculates the cosine similarity between these vectors to identify the items that most closely match the user's preferences, and recommends the most similar items to the user.

[1021] 6. Generating compliments

[1022] The server generates compliments for each selected item, building feedback that praises its quality and appeal based on color and style.

[1023] The server can adjust the content and tone of the compliment based on the user's emotional information obtained from the emotion engine.

[1024] 7. Displaying the results

[1025] The server sends the generated list of recommended items and the compliments to the device.

[1026] The user terminal displays these results to the user, who can then check the suggested outfits and their feedback.

[1027] Specific examples

[1028] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[1029] 2. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The emotion information is sent to the server.

[1030] 3. The server retrieves data such as the user's preferences and past purchase history from the database.

[1031] 4. The server retrieves all fashion item information from the database and converts the item descriptions into TF-IDF vectors.

[1032] 5. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the item that best matches the user's preference.

[1033] 6. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!". Based on the emotional information that the user is pleased, the server adjusts the tone of the compliment to be more positive.

[1034] 7. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[1035] The above is an embodiment of the system of the present invention. This system allows users to find the best outfit based on their preferences and feel confident in their choice. Furthermore, by incorporating an emotion engine, it is possible to provide feedback according to the user's emotions.

[1036] The processing flow will be explained below.

[1037] Step 1:

[1038] A user logs in to the system using a terminal and inputs his / her preference information (e.g., preferred style, color, design, etc.). The terminal transmits the input data to the server.

[1039] Step 2:

[1040] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. This emotion information is then sent to the server.

[1041] Step 3:

[1042] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database. Specifically, it issues a DB query to retrieve related data.

[1043] Step 4:

[1044] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database. The retrieved data is then used in the recommendation algorithm.

[1045] Step 5:

[1046] The server converts the descriptions of fashion items into TF-IDF (Term Frequency-Inverse Document Frequency) vectors, which is a process of representing each item description as a numerical vector.

[1047] Step 6:

[1048] The server also converts the user's preference information into a TF-IDF vector, which allows the user's preferences to be expressed as a numerical vector.

[1049] Step 7:

[1050] The server calculates the cosine similarity between these vectors, and uses the cosine similarity to identify the fashion item that most closely matches the user's preferences.

[1051] Step 8:

[1052] The server selects the top five fashion items from the items with the highest similarity scores, and the selected items are recommended to the user.

[1053] Step 9:

[1054] The server generates compliments for each selected item, generates feedback praising its goodness and attractiveness based on its color and style, and adjusts the content and tone of the compliments based on the user's emotional information obtained from the emotion engine.

[1055] Step 10:

[1056] The server sends the generated list of recommended items and the compliments to the device.

[1057] Step 11:

[1058] The device will display the recommended outfits and feedback to the user, and the user can check the suggested outfits and their feedback.

[1059] Specific examples

[1060] If a user logs in and inputs that they like "elegant style" and the color "blue," the system sends this preference information to the server. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The emotion information is sent to the server. The server retrieves the user's preference information and data such as past purchase history from the database. The server retrieves information on all fashion items from the database and converts the item descriptions into TF-IDF vectors. The server also converts the user's preference information into vectors and calculates cosine similarity to identify the item that best matches the user's preferences. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!" Based on the emotional information that the user is happy, the system adjusts the tone of the compliment to be more positive. The server sends the recommended item list and the compliment to the terminal, which displays them. The user can review the results and wear the outfit they chose with confidence.

[1061] Example 2

[1062] 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."

[1063] Conventional fashion recommendation systems can only suggest outfits based on the user's preference information and clothing data, but they have the problem of not being able to provide feedback that takes into account the user's emotions. This means that the system is unable to fully improve the user's feelings and satisfaction with the suggested outfits, resulting in limited effectiveness.

[1064] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1065] In this invention, the server includes means for acquiring user preference information, means for acquiring owned clothing data, and means for recommending optimal clothing based on the user preference information and owned clothing data. This makes it possible to suggest optimal clothing based on the user preference information and owned clothing data, thereby improving the user's feelings and satisfaction with the suggested clothing.

[1066] In addition, in this invention, the server includes means for recognizing emotions from the user's facial expressions, tone of voice, input text, etc., means for adjusting the content and tone of the compliments based on the emotional information, and means for generating compliments for the recommended outfit. This makes it possible to provide feedback that takes into consideration the user's emotions, further increasing user satisfaction.

[1067] "User preference information" refers to information about personal preferences such as the user's preferred style, color, and design.

[1068] "Owned clothing data" is information about clothing owned by the user, and includes detailed data such as the type, color, style, and brand of the item.

[1069] "Recommendation means" is a system function that selects optimal clothing based on the user's preference information and clothing data and presents it to the user.

[1070] The "means for generating compliments" is a function of the system that generates feedback praising the goodness and attractiveness of the recommended outfit.

[1071] "Means for recognizing emotions" refers to the system's function of analyzing and determining the user's emotions from the user's facial expressions, tone of voice, input text, etc.

[1072] The "means for adjusting the content and tone of the compliment based on emotional information" is a system function for adjusting the content and tone of the generated compliment to the user's current emotional state based on information obtained from the means for recognizing emotions.

[1073] "TF-IDF vectorization means" is a system function that calculates the importance of words using text data and converts it into a vector representation.

[1074] The "means for calculating cosine similarity" is a system function that measures the similarity between vectorized data using cosine angles and calculates the similarity.

[1075] MODE FOR CARRYING OUT THE INVENTION

[1076] This invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide feedback based on the user's emotions. Specific embodiments of this system are described below.

[1077] System configuration

[1078] The system of the present invention comprises a user terminal, a server, a database, and an emotion engine. The user terminal provides an interface for the user to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and fashion item catalog information. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, input text, etc., and sends the results to the server.

[1079] Hardware and Software Use

[1080] User terminals include devices such as PCs, smartphones, and tablets. The server is a server computer with high-performance computing capabilities, and the database can use a commercial relational database system (e.g., MySQL or PostgreSQL).

[1081] The emotion engine is realized using a model that uses deep learning technology (e.g., an emotion analysis model). For example, a camera and image recognition technology are required to analyze a user's facial expressions, and a microphone and voice analysis technology are required to analyze the tone of voice. A deep learning framework (e.g., TensorFlow, PyTorch) is used to implement the emotion engine.

[1082] Natural language processing techniques are used for the recommendation algorithm and compliment generation. Specifically, "TF-IDF vectorization" and "cosine similarity calculation" are used to process text data. These processes can be performed using the Python "scikit-learn" library. In addition, generative AI models (e.g., GPT models) are used for natural language generation.

[1083] Specific examples

[1084] A specific example of the operation of the system is shown below.

[1085] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[1086] 2. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. This emotion information is sent to the server.

[1087] 3. The server retrieves data such as the user's preferences and past purchase history from the database.

[1088] 4. The server retrieves all fashion item information from the database and converts the item descriptions into TF-IDF vectors.

[1089] 5. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the most suitable item.

[1090] 6. For example, if an "elegant blue dress" is selected, the server generates a compliment for this item, such as "The dress you chose is very elegant and graceful!". Based on the emotional information that the user is pleased, the tone of the compliment is adjusted to be more positive.

[1091] 7. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[1092] Prompt Sentence Examples

[1093] "Recommend the most suitable fashion items based on the user's preferences and past purchase history. Also, generate compliments for the selected items and adjust the tone according to the user's emotions."

[1094] As a result, this system utilizes the user's preference information and emotional information to recommend the most suitable outfits and provide the user with a high level of satisfaction.

[1095] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1096] Step 1:

[1097] A user logs in to the system. The user enters login information, which the terminal sends to the server. The server then refers to a database to authenticate the login information and authenticates the user. If authentication is successful, a screen for entering the user's preferences is displayed.

[1098] Step 2:

[1099] The user inputs their preference information. For example, they select "elegant style" and "blue color." The terminal sends this preference information to the server. The server stores the input preference information in a database. This preference information becomes the input for subsequent processing.

[1100] Step 3:

[1101] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. For example, facial expression data is acquired from a camera and analyzed using a deep learning model to recognize "joy." This emotion information is sent to the server, which temporarily stores the received emotion information.

[1102] Step 4:

[1103] The server retrieves basic information about the user (preferences, past purchase history, etc.) from a database. The server issues SQL queries to extract this information from the database and stores it in memory. This basic information becomes the input for the recommendation algorithm.

[1104] Step 5:

[1105] The server retrieves all fashion item information from the database. Specifically, detailed data such as item name, description, color, and style are extracted using SQL queries. The server receives this information and prepares to convert each item description into a TF-IDF vector.

[1106] Step 6:

[1107] The server converts the fashion item descriptions into TF-IDF vectors. Using the Python "scikit-learn" library, the item descriptions are converted into numerical vectors. User preference information is also vectorized in the same way. These vectors are used as inputs for the cosine similarity calculation.

[1108] Step 7:

[1109] The server calculates the cosine similarity and identifies the item that most closely matches the user's preferences. It outputs a similarity score by calculating the cosine similarity between vectors. It selects the item with the highest score and creates a list of recommended items.

[1110] Step 8:

[1111] The server generates compliments for the selected item. Using a generative AI model (e.g., a GPT model), it generates text that praises the good points and attractiveness of the selected item. It includes specific feedback using color and style information. It also adjusts the tone of the compliments based on emotional information from an emotion engine.

[1112] Step 9:

[1113] The server sends the generated list of recommended items and compliments to the device. The generated text data and item information are sent to the device in JSON format. The device displays these results to the user. The user can check the suggested outfits and their feedback and wear the outfits they have chosen with confidence.

[1114] (Application example 2)

[1115] 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."

[1116] Conventional fashion recommendation systems can recommend clothing based on a user's preferences, but they are unable to provide emotional feedback on the selection. This leaves users with little confidence in the recommended clothing, resulting in low satisfaction with their selection. Furthermore, because previous systems do not take the user's emotional state into account, they are unable to provide appropriate feedback that reflects the user's psychological state at any given time. Therefore, the objective of this invention is to provide a system that takes into account the user's preference information and emotional information, allowing the user to select fashion items with greater satisfaction and confidence.

[1117] 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.

[1118] In this invention, the server includes means for acquiring user preference information, means for acquiring owned clothing data, means for recommending optimal clothing based on the user preference information and the owned clothing data, means for generating compliments for the recommended clothing, means for acquiring user emotion information and adjusting the tone of the compliments based on the emotion information, and means for displaying the optimal clothing after it has been recommended. This makes it possible not only to recommend optimal clothing based on the user preference information, but also to generate compliments for the selected clothing and provide feedback according to the user's emotions.

[1119] "User preference information" is data relating to the user's preferences such as style, color, and design.

[1120] "Owned clothing data" is data about clothing owned by the user, and includes information such as item name, color, style, and description.

[1121] The "means for recommending optimal clothing" is a means for identifying and recommending the most suitable clothing for a user based on the user's preference information and clothing data owned by the user.

[1122] The "means for generating compliments" is a means for generating feedback praising the goodness and attractiveness of the recommended outfit.

[1123] "User emotion information" is data related to emotions analyzed from the user's facial expression, tone of voice, input text, and the like.

[1124] The "means for adjusting the tone of the compliment" is a means for adjusting the content and tone of the generated compliment based on the acquired emotional information of the user.

[1125] The "means for displaying the optimal outfit after it has been recommended" is a means for visually presenting information about the selected outfit and feedback thereon to the user.

[1126] This system recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it provides feedback based on the user's emotions. This system can be implemented as an application that runs on smart glasses, head-mounted displays, and smartphones.

[1127] Hardware and software used

[1128] Hardware:

[1129] Smart Glasses

[1130] head-mounted display

[1131] Smartphone

[1132] software:

[1133] Python environment

[1134] TF-IDF Vectorizer (scikit-learn)

[1135] Cosine similarity calculation module (scikit-learn)

[1136] Sentiment analysis engine (AWS Rekognition, Google Cloud Vision, or TextBlob)

[1137] Data processing and calculation

[1138] 1. Data Vectorization:

[1139] The server converts the user's preference information and the description of the clothing data they own into a TF-IDF vector, which allows the data features to be expressed as a numerical vector and made comparable.

[1140] 2. Similarity calculation:

[1141] The server calculates the cosine similarity between the vectorized user preference information and the clothing data, and identifies the items that best match the preferences.

[1142] 3. Compliment Generation:

[1143] The server generates feedback for the selected item, praising its merits and attractiveness, and adjusts the content and tone of the praise based on the user's emotional information obtained from the emotion engine.

[1144] 4. Displaying the results:

[1145] The user device visually presents the recommended item list and compliments received from the server to the user, allowing the user to check the suggested outfit and receive feedback on it.

[1146] Specific examples

[1147] For example, if a user inputs that they like "elegant style" and "blue," the system processes this preference information. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The server vectorizes the user's preference information and clothing data, calculates the cosine similarity, and identifies the item that best matches the user's preference.

[1148] For a selected item (e.g., "an elegant blue dress"), the server generates a compliment such as "The dress you chose is very elegant and graceful!" Furthermore, based on the emotional information that the user is pleased, the tone of the compliment is adjusted to be more positive.

[1149] Prompt Sentence Examples

[1150] Here are some example input prompts for a generative AI model:

[1151] "A system that recommends the most suitable outfit based on the user's preference information and clothing data, and generates compliments for the selected outfit. Furthermore, please combine it with an emotion engine that recognizes the user's emotions and provides emotion-based feedback."

[1152] The above is a specific embodiment of the present invention. This system allows users to find the perfect outfit based on their preferences and feel confident in their choice. Furthermore, by incorporating an emotion engine, it is possible to provide feedback based on the user's emotions.

[1153] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1154] Step 1:

[1155] The user logs in to the device and inputs their preferences (e.g., preferred style, color, design, etc.). The user's input information is acquired by the device and sent to the server. The input data includes the user's preferred style and color, as well as past purchase history. The server stores this information in a database.

[1156] Step 2:

[1157] The server activates an emotion engine and analyzes the user's facial expressions, tone of voice, input text, etc. to obtain emotional information. Specifically, the user's facial expressions and voice data obtained using a camera or microphone are sent to the emotion analysis engine, which extracts emotional information as positive, negative, or neutral. This emotional information is then sent to the server and stored in a database.

[1158] Step 3:

[1159] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database. At the same time, it also retrieves data about all fashion items from the database. This data includes item names, descriptions, colors, styles, etc.

[1160] Step 4:

[1161] The server converts the descriptions of fashion items into TF-IDF vectors. Then, it converts the user's preference information into TF-IDF vectors in the same way. This converts the string data into numerical vectors, making it possible to perform calculations and comparisons.

[1162] Step 5:

[1163] The server uses the vectorized data to calculate the cosine similarity, which identifies items that are closest to the user's preferences. Items with a high cosine similarity are deemed to be suitable for the user's preferences.

[1164] Step 6:

[1165] The server recommends the most similar items to the user by listing them in descending order of cosine similarity, and selecting the top few items. Detailed information about these items (such as names, descriptions, and images) is then retrieved.

[1166] Step 7:

[1167] The server generates compliments for each selected item. The generated compliments highlight the item's quality and attractiveness based on its color and style. Furthermore, the tone of the compliments is adjusted based on the user's emotional information obtained from the emotion engine. For example, if the user is in a positive state, the tone of the compliments will be brighter and more positive.

[1168] Step 8:

[1169] The server sends the generated list of recommended items and compliments to the device. Specifically, it includes the list of recommended items, compliments for each item, and detailed information about each item (image, description, price, etc.). The device visually presents this information to the user so that the user can confirm it.

[1170] The above is the processing flow based on a specific embodiment of the present invention. This system allows users to find the best outfit based on their personal preferences and feel confident about the outfit they choose. It is also possible to provide feedback based on the user's emotions through an emotion engine.

[1171] 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.

[1172] 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.

[1173] 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.

[1174] [Fourth embodiment]

[1175] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1176] 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.

[1177] 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).

[1178] 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.

[1179] 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.

[1180] 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).

[1181] 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.

[1182] 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.

[1183] 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.

[1184] 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.

[1185] 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.

[1186] 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.

[1187] 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."

[1188] The present invention relates to a system for recommending optimal clothing based on a user's preference information and clothing data, and generating compliments for the selected clothing. Specific embodiments of the system are described below.

[1189] System configuration

[1190] The system of the present invention comprises a user terminal, a server, and a database. The user terminal provides an interface for users to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and catalog information of fashion items.

[1191] System Operation

[1192] 1. Enter user information

[1193] The user logs in to the system and inputs their preferences (preferred style, color, design, etc.), which are then sent to the server by the terminal.

[1194] 2. Obtaining User Data

[1195] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database.

[1196] 3. Acquiring fashion item data

[1197] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[1198] 4. Running the recommendation algorithm

[1199] The server converts the descriptions of fashion items into TF-IDF vectors, and also converts the user's preference information into vectors in the same way.

[1200] The server calculates the cosine similarity between these vectors to identify the items that most closely match the user's preferences, and recommends the most similar items to the user.

[1201] 5. Generating compliments

[1202] The server generates compliments for each selected item, building feedback that praises its quality and appeal based on color and style.

[1203] 6. Displaying the results

[1204] The server sends the generated list of recommended items and the compliments to the device.

[1205] The user terminal displays these results to the user, who can then check the suggested outfits and their feedback.

[1206] Specific examples

[1207] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[1208] 2. The server retrieves data such as the user's preferences and past purchase history from the database.

[1209] 3. The server retrieves information on all fashion items from the database and converts the item descriptions into TF-IDF vectors.

[1210] 4. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the item that best matches the user's preference.

[1211] 5. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!"

[1212] 6. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[1213] The above is an embodiment of the system of the present invention, which allows users to find the best outfit based on their preferences and feel confident in their choice.

[1214] The processing flow will be explained below.

[1215] Step 1:

[1216] A user logs in to the system using a terminal and inputs their preference information (e.g., preferred style, color, design), and the terminal transmits the input data to the server.

[1217] Step 2:

[1218] The server retrieves basic information about the user from the database, such as preference information and past purchase history corresponding to the user ID.

[1219] Step 3:

[1220] The server retrieves all the fashion item information from the database, including the name, description, color, style, etc. of the outfit.

[1221] Step 4:

[1222] The server converts the descriptions of fashion items into TF-IDF vectors, which is a process for expressing clothing features as numerical vectors.

[1223] Step 5:

[1224] The server also converts the user's preference information into a TF-IDF vector, which allows the user's preferences to be expressed as a numerical vector.

[1225] Step 6:

[1226] The server calculates the cosine similarity between these vectors, and uses the cosine similarity to identify the fashion item that most closely matches the user's preferences.

[1227] Step 7:

[1228] The server selects the top five fashion items from among the items with the highest similarity scores, thereby recommending the most suitable outfit for the user.

[1229] Step 8:

[1230] The server generates compliments for each selected item, generating feedback that praises its quality and appeal based on color and style.

[1231] Step 9:

[1232] The server sends the generated list of recommended items and the compliments to the device.

[1233] Step 10:

[1234] The device will display the recommended outfits and feedback to the user, and the user can check the suggested outfits and their feedback.

[1235] Example 1

[1236] 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."

[1237] In recent years, interest in fashion has increased, and many people want to choose clothes that suit their own tastes. However, choosing the best outfit from the vast number of fashion items available is not easy, and the process takes time and effort. It is also difficult to be confident in one's choices. Furthermore, existing recommendation systems may not be accurate enough to recommend items that precisely match a user's tastes.

[1238] 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.

[1239] In this invention, the server includes means for acquiring user preference information, means for acquiring user clothing data, means for recommending optimal clothing based on the user preference information and the user clothing data, and means for generating compliments for the recommended clothing using a generative AI model. This makes it possible to select optimal clothing based on the user's preferences and generate compliments to give the user confidence.

[1240] "User preference information" is information about personal fashion preferences such as the user's preferred style, color, and design.

[1241] "Owned clothing data" is information about clothing currently owned by the user, including the item name, description, color, style, and so on.

[1242] The "means for recommending optimal clothing" is a function for identifying and recommending the most suitable clothing for a user using the user's preference information and clothing data.

[1243] A "generative AI model" refers to an artificial intelligence algorithm that generates a response in natural language in response to a certain input (prompt), such as a generative language model.

[1244] The "means for generating compliments" is a function that uses a generative AI model to create feedback praising the good qualities and attractiveness of the selected outfit.

[1245] MODE FOR CARRYING OUT THE INVENTION

[1246] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Specific embodiments of the system are described below.

[1247] System configuration

[1248] This system consists of a user terminal, a server, and a database. The user terminal provides an interface for users to access the system and input their preference information. The server processes the user's preference information and clothing data and recommends the most suitable outfits. The database stores user information, clothing data, and catalog information of fashion items.

[1249] System Operation

[1250] 1. Enter user information

[1251] The user logs into the system using a terminal and inputs preference information (preferred style, color, design, etc.).

[1252] Example: A user inputs that they like "elegant style" and the color "blue." The device sends this information to the server through the interface.

[1253] 2. Obtaining User Data

[1254] The server retrieves basic information about the user (preferences, past purchase history, etc.) from a database. This information is retrieved using an SQL database (e.g., MySQL).

[1255] Example: The server executes an SQL query to retrieve user preferences such as "elegant" and "blue."

[1256] 3. Acquiring fashion item data

[1257] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[1258] Example: The server executes an SQL query to retrieve item information such as "an elegant blue dress" from a database.

[1259] 4. Running the recommendation algorithm

[1260] The server converts the descriptions of fashion items into TF-IDF vectors and also converts the user's preference information into vectors. This is done using Python and natural language processing libraries (e.g., scikit-learn, NLTK).

[1261] Cosine similarity is calculated to identify items that most closely match the user's preferences.

[1262] Example: The description of each item is vectorized using TF-IDF vectorization, and the cosine similarity is calculated to select dresses that are "elegant" and "blue."

[1263] 5. Generating compliments

[1264] The server generates a compliment for each selected item. To do this, it uses a generative AI model (e.g., GPT-3). Specifically, it generates a compliment for the selected item by inputting a prompt sentence into the generative AI model.

[1265] Example prompt: "Generate a compliment about this elegant blue dress."

[1266] Example: A generative AI model can generate compliments such as, "The dress you chose is very elegant and graceful!"

[1267] 6. Displaying the results

[1268] The server transmits the generated list of recommended items and the compliments to the terminal, which displays them.

[1269] Example: The server sends the generated item list and compliments in JSON format, and the web browser or smartphone app on the user's device displays "The blue dress you chose is very elegant and wonderful!"

[1270] In this way, the present invention can recommend optimal outfits based on a user's preferences, providing confidence in the choice. Furthermore, by using a generative AI model, the generation of compliments can be automated, providing users with rapid positive feedback.

[1271] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1272] Step 1: Enter your information

[1273] Users log in to the system using a terminal and enter their preference information. Specifically, the user enters their authentication information on the login screen and transitions to the customer preference interface. There, the user enters their preferred style, color, design, etc. into the input form.

[1274] Input: Authentication information, preference information (preferred style, color, design, etc.)

[1275] Output: User preference information is sent to the server

[1276] Step 2: Retrieving User Data

[1277] The server executes an SQL query to retrieve basic user information from the database, including past purchase history and registered preference information.

[1278] Input: User ID

[1279] Output: Basic information of the user (preference information, past purchase history, etc.)

[1280] Step 3: Obtaining fashion item data

[1281] The server retrieves all the fashion item information from the database. Specifically, the server executes SQL queries to retrieve the name, description, color, style, etc. of each fashion item.

[1282] Input: None

[1283] Output: Fashion item information (name, description, color, style, etc.)

[1284] Step 4: Run the recommendation algorithm

[1285] The server converts the descriptions of fashion items and user preference information into TF-IDF vectors. Specifically, it uses Python and natural language processing libraries (e.g., scikit-learn, NLTK) to vectorize the text data and calculate the cosine similarity between them to identify the most suitable items.

[1286] Input: Description of fashion item, user preference information

[1287] Output: A list of items that best fit the user's preferences

[1288] Step 5: Generate compliments

[1289] The server generates a compliment for the selected item using a generative AI model (e.g., GPT-3). Specifically, the server inputs a prompt sentence into the generative AI model to generate feedback.

[1290] Input: Recommended item

[1291] Output: Generated compliment

[1292] Step 6: View the results

[1293] The server sends the generated list of recommended items and compliments to the device, which then displays them. Specifically, the device parses the received JSON data and displays it to the user in a web browser or application.

[1294] Input: Recommended item list, compliments

[1295] Output: A list of recommended items and a compliment displayed to the user

[1296] (Application example 1)

[1297] 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."

[1298] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. In particular, conventional systems require users to select a vast number of fashion items themselves, and this has led to issues with insufficient appropriate recommendations and feedback. There is a need for systems that can improve user satisfaction and recommend optimal clothing in a rational and timely manner.

[1299] 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.

[1300] In this invention, the server includes a means for acquiring user preference information, a means for acquiring user clothing data, a means for recommending optimal clothing based on the user's preference information and the user's clothing data, a means for generating compliments for the recommended clothing, and a means for displaying these results. This allows users to easily find optimal clothing based on their preferences and feel confident in their selection. Furthermore, a recommendation algorithm using TF-IDF vectorization and cosine similarity calculations achieves highly accurate recommendations. Furthermore, by using prompt sentences that generate compliments using a generative AI model, more natural and persuasive feedback can be provided.

[1301] definition statement

[1302] "User preference information" is information about fashion such as the style, color, and design that the user prefers.

[1303] "Owned clothing data" is detailed information about clothing items owned by the user, including the name, description, color, style, etc. of each item.

[1304] "Means for recommending optimal clothing" refers to a method or device for selecting the most suitable clothing for a user using an algorithm based on the user's preference information and clothing data.

[1305] "Compliment generator" refers to an algorithm or mechanism for generating positive feedback on a recommended outfit.

[1306] "Means for displaying these results" refers to a method or device for displaying the recommended outfits and compliments on the outfits on the user terminal.

[1307] "TF-IDF vectorization" is a technique for quantifying text data as vectors, and is a technology that converts it into a vector representation by taking into account the importance of each word.

[1308] "Cosine similarity" is an index for measuring the similarity between two vectors, and is calculated based on the angle of the vectors.

[1309] "Generative AI model" refers to an algorithm or system that uses artificial intelligence to automatically create text or other artifacts.

[1310] A "prompt sentence" is an input sentence used to elicit a specific response from a generative AI model.

[1311] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Specific embodiments of this system are described below.

[1312] System configuration

[1313] The system of the present invention consists of the following main parts:

[1314] User terminal

[1315] server

[1316] Database

[1317] These parts work together to provide optimal outfit recommendations and compliment generation for the user.

[1318] User device functions

[1319] The user terminal provides an interface that receives input from the user and displays the results. The user logs in using the terminal and inputs preference information (preferred style, color, design, etc.). The terminal sends this information to the server.

[1320] Server Features

[1321] The server has the following functions:

[1322] 1. Data Acquisition:

[1323] The server obtains the user's preference information and clothing data from the database.

[1324] 2. Data Processing:

[1325] The server converts preference information and clothing description data into TF-IDF vectors and calculates the cosine similarity to select the most suitable outfit.

[1326] 3. Compliment Generation:

[1327] The server generates natural and complimenting words for the selected outfit, using a generative AI model to input prompts and create feedback accordingly.

[1328] Database

[1329] The database stores user information, clothing data, and fashion item catalog information, allowing the server to quickly and efficiently retrieve the required information.

[1330] Example of operation

[1331] 1. Log in and enter your preferences:

[1332] The user logs in and inputs his / her preference information of "elegant style" and "blue."

[1333] 2. Retrieving data from the database:

[1334] The server retrieves the user's preference information, past purchase history, and clothing data from the database.

[1335] 3. Clothing recommendations:

[1336] The server uses TF-IDF vectorization and cosine similarity calculation to select the item that best matches the user's preferences, such as an "elegant blue dress."

[1337] 4. Generating compliments:

[1338] The server uses a generative AI model to generate compliments such as, "The dress you chose is very elegant and graceful!"

[1339] Hardware and software used

[1340] Server: The server on which the Python program runs (using a server framework such as Flask).

[1341] User device: Smartphone (including iOS and Android).

[1342] Examples of prompt statements

[1343] Here is an example of a prompt that can be fed into a generative AI model:

[1344] text

[1345] "Please tell me the algorithm for a system that recommends the most suitable outfit based on the user's preference information and clothing data, and generates compliments for the selected outfit. The user's preference information is "elegant style" and "blue." Please also take into account past purchase data and clothing data."

[1346] The above is a specific embodiment for carrying out the invention. This system allows a user to find the best outfit based on their own preferences and wear the outfit they have chosen with confidence.

[1347] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1348] Processing Step Description

[1349] Step 1:

[1350] User login and preference input

[1351] The user logs in to the device and inputs their preferences (style, color, design, etc.), which are then sent from the device to the server.

[1352] Input: User login information, preferences

[1353] Output: Preference information sent to the server

[1354] Step 2:

[1355] Retrieving user data from the database

[1356] The server accesses the database to obtain the user's preferences, past purchase history, and clothing data, thereby gathering all data related to the user.

[1357] Input: Database query, user identity

[1358] Output: Acquired user data, owned clothing data

[1359] Step 3:

[1360] Obtaining information on fashion items

[1361] The server retrieves information about all fashion items (item name, description, color, style, etc.) from the database and creates a list.

[1362] Input: Database query

[1363] Output: A list of retrieved fashion items

[1364] Step 4:

[1365] TF-IDF vectorization of data

[1366] The server converts the acquired user preference information and fashion item description data into TF-IDF vectors, which converts each piece of data into a numerical value and puts it in a format suitable for calculations.

[1367] Input: User preference information, fashion item description data

[1368] Output: Vectorized user preference information and item description data

[1369] Step 5:

[1370] Calculating cosine similarity

[1371] The server calculates the cosine similarity using the TF-IDF vectorized data, and selects the item that most closely matches the user's preferences.

[1372] Input: Vectorized user preference information and item description data

[1373] Output: A list of items with high similarity

[1374] Step 6:

[1375] Selection of recommended items

[1376] Based on the results of the cosine similarity calculation, the server selects several items with the highest similarity and generates a list of recommended items.

[1377] Input: List of highly similar items

[1378] Output: Recommended item list

[1379] Step 7:

[1380] Generating compliments

[1381] The server generates compliments for the selected recommended items using a generative AI model. The server inputs prompts to the model to obtain natural and positive compliments.

[1382] Input: Recommended item list, prompt text

[1383] Output: Compliment generation

[1384] Step 8:

[1385] Displaying the results

[1386] The server sends a list of recommended items and compliments to the user's device, which then displays them. The user can check the suggested outfit and the feedback.

[1387] Input: Recommended item list, compliments

[1388] Output: The result displayed on the user's terminal

[1389] Through the above processing steps, the user can easily find the best outfit based on their preferences and feel confident in their choice.

[1390] 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.

[1391] The present invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide feedback based on the user's emotions. Specific embodiments of the system are described below.

[1392] System configuration

[1393] The system of the present invention comprises a user terminal, a server, a database, and an emotion engine. The user terminal provides an interface for the user to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and fashion item catalog information. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, input text, etc., and sends the results to the server.

[1394] System Operation

[1395] 1. Enter user information

[1396] A user logs in to the system and inputs their preference information (e.g., preferred style, color, design, etc.). The terminal sends this information to the server.

[1397] 2. Acquiring emotional information

[1398] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. This emotion information is sent to the server.

[1399] 3. Obtaining User Data

[1400] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database.

[1401] 4. Acquiring fashion item data

[1402] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database.

[1403] 5. Running the recommendation algorithm

[1404] The server converts the descriptions of fashion items into TF-IDF vectors, and also converts the user's preference information into vectors in the same way.

[1405] The server calculates the cosine similarity between these vectors to identify the items that most closely match the user's preferences, and recommends the most similar items to the user.

[1406] 6. Generating compliments

[1407] The server generates compliments for each selected item, building feedback that praises its quality and appeal based on color and style.

[1408] The server can adjust the content and tone of the compliment based on the user's emotional information obtained from the emotion engine.

[1409] 7. Displaying the results

[1410] The server sends the generated list of recommended items and the compliments to the device.

[1411] The user terminal displays these results to the user, who can then check the suggested outfits and their feedback.

[1412] Specific examples

[1413] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[1414] 2. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The emotion information is sent to the server.

[1415] 3. The server retrieves data such as the user's preferences and past purchase history from the database.

[1416] 4. The server retrieves all fashion item information from the database and converts the item descriptions into TF-IDF vectors.

[1417] 5. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the item that best matches the user's preference.

[1418] 6. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!". Based on the emotional information that the user is pleased, the server adjusts the tone of the compliment to be more positive.

[1419] 7. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[1420] The above is an embodiment of the system of the present invention. This system allows users to find the best outfit based on their preferences and feel confident in their choice. Furthermore, by incorporating an emotion engine, it is possible to provide feedback according to the user's emotions.

[1421] The processing flow will be explained below.

[1422] Step 1:

[1423] A user logs in to the system using a terminal and inputs his / her preference information (e.g., preferred style, color, design, etc.). The terminal transmits the input data to the server.

[1424] Step 2:

[1425] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. This emotion information is then sent to the server.

[1426] Step 3:

[1427] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database. Specifically, it issues a DB query to retrieve related data.

[1428] Step 4:

[1429] The server retrieves all fashion item information (clothing name, description, color, style, etc.) from the database. The retrieved data is then used in the recommendation algorithm.

[1430] Step 5:

[1431] The server converts the descriptions of fashion items into TF-IDF (Term Frequency-Inverse Document Frequency) vectors, which is a process of representing each item description as a numerical vector.

[1432] Step 6:

[1433] The server also converts the user's preference information into a TF-IDF vector, which allows the user's preferences to be expressed as a numerical vector.

[1434] Step 7:

[1435] The server calculates the cosine similarity between these vectors, and uses the cosine similarity to identify the fashion item that most closely matches the user's preferences.

[1436] Step 8:

[1437] The server selects the top five fashion items from the items with the highest similarity scores, and the selected items are recommended to the user.

[1438] Step 9:

[1439] The server generates compliments for each selected item, generates feedback praising its goodness and attractiveness based on its color and style, and adjusts the content and tone of the compliments based on the user's emotional information obtained from the emotion engine.

[1440] Step 10:

[1441] The server sends the generated list of recommended items and the compliments to the device.

[1442] Step 11:

[1443] The device will display the recommended outfits and feedback to the user, and the user can check the suggested outfits and their feedback.

[1444] Specific examples

[1445] If a user logs in and inputs that they like "elegant style" and the color "blue," the system sends this preference information to the server. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The emotion information is sent to the server. The server retrieves the user's preference information and data such as past purchase history from the database. The server retrieves information on all fashion items from the database and converts the item descriptions into TF-IDF vectors. The server also converts the user's preference information into vectors and calculates cosine similarity to identify the item that best matches the user's preferences. For example, if an "elegant blue dress" is selected, the server generates a compliment about the item, such as "The dress you chose is very elegant and graceful!" Based on the emotional information that the user is happy, the system adjusts the tone of the compliment to be more positive. The server sends the recommended item list and the compliment to the terminal, which displays them. The user can review the results and wear the outfit they chose with confidence.

[1446] Example 2

[1447] 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."

[1448] Conventional fashion recommendation systems can only suggest outfits based on the user's preference information and clothing data, but they have the problem of not being able to provide feedback that takes into account the user's emotions. This means that the system is unable to fully improve the user's feelings and satisfaction with the suggested outfits, resulting in limited effectiveness.

[1449] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1450] In this invention, the server includes means for acquiring user preference information, means for acquiring owned clothing data, and means for recommending optimal clothing based on the user preference information and owned clothing data. This makes it possible to suggest optimal clothing based on the user preference information and owned clothing data, thereby improving the user's feelings and satisfaction with the suggested clothing.

[1451] In addition, in this invention, the server includes means for recognizing emotions from the user's facial expressions, tone of voice, input text, etc., means for adjusting the content and tone of the compliments based on the emotional information, and means for generating compliments for the recommended outfit. This makes it possible to provide feedback that takes into consideration the user's emotions, further increasing user satisfaction.

[1452] "User preference information" refers to information about personal preferences such as the user's preferred style, color, and design.

[1453] "Owned clothing data" is information about clothing owned by the user, and includes detailed data such as the type, color, style, and brand of the item.

[1454] "Recommendation means" is a system function that selects optimal clothing based on the user's preference information and clothing data and presents it to the user.

[1455] The "means for generating compliments" is a function of the system that generates feedback praising the goodness and attractiveness of the recommended outfit.

[1456] "Means for recognizing emotions" refers to the system's function of analyzing and determining the user's emotions from the user's facial expressions, tone of voice, input text, etc.

[1457] The "means for adjusting the content and tone of the compliment based on emotional information" is a system function for adjusting the content and tone of the generated compliment to the user's current emotional state based on information obtained from the means for recognizing emotions.

[1458] "TF-IDF vectorization means" is a system function that calculates the importance of words using text data and converts it into a vector representation.

[1459] The "means for calculating cosine similarity" is a system function that measures the similarity between vectorized data using cosine angles and calculates the similarity.

[1460] MODE FOR CARRYING OUT THE INVENTION

[1461] This invention relates to a system that recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide feedback based on the user's emotions. Specific embodiments of this system are described below.

[1462] System configuration

[1463] The system of the present invention comprises a user terminal, a server, a database, and an emotion engine. The user terminal provides an interface for the user to access the system and input their own preference information. The server processes the user's preference information and clothing data and recommends optimal outfits. The database stores user information, clothing data, and fashion item catalog information. The emotion engine recognizes emotions from the user's facial expressions, tone of voice, input text, etc., and sends the results to the server.

[1464] Hardware and Software Use

[1465] User terminals include devices such as PCs, smartphones, and tablets. The server is a server computer with high-performance computing capabilities, and the database can use a commercial relational database system (e.g., MySQL or PostgreSQL).

[1466] The emotion engine is realized using a model that uses deep learning technology (e.g., an emotion analysis model). For example, a camera and image recognition technology are required to analyze a user's facial expressions, and a microphone and voice analysis technology are required to analyze the tone of voice. A deep learning framework (e.g., TensorFlow, PyTorch) is used to implement the emotion engine.

[1467] Natural language processing techniques are used for the recommendation algorithm and compliment generation. Specifically, "TF-IDF vectorization" and "cosine similarity calculation" are used to process text data. These processes can be performed using the Python "scikit-learn" library. In addition, generative AI models (e.g., GPT models) are used for natural language generation.

[1468] Specific examples

[1469] A specific example of the operation of the system is shown below.

[1470] 1. If a user logs in and inputs that they like "elegant style" and the color "blue," the system will send this preference information to the server.

[1471] 2. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. This emotion information is sent to the server.

[1472] 3. The server retrieves data such as the user's preferences and past purchase history from the database.

[1473] 4. The server retrieves all fashion item information from the database and converts the item descriptions into TF-IDF vectors.

[1474] 5. The server similarly vectorizes the user's preference information and calculates the cosine similarity to identify the most suitable item.

[1475] 6. For example, if an "elegant blue dress" is selected, the server generates a compliment for this item, such as "The dress you chose is very elegant and graceful!". Based on the emotional information that the user is pleased, the tone of the compliment is adjusted to be more positive.

[1476] 7. The server sends the recommended items and compliments to the device, which displays them. The user can then check the results and wear the outfit they chose with confidence.

[1477] Prompt Sentence Examples

[1478] "Recommend the most suitable fashion items based on the user's preferences and past purchase history. Also, generate compliments for the selected items and adjust the tone according to the user's emotions."

[1479] As a result, this system utilizes the user's preference information and emotional information to recommend the most suitable outfits and provide the user with a high level of satisfaction.

[1480] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1481] Step 1:

[1482] A user logs in to the system. The user enters login information, which the terminal sends to the server. The server then refers to a database to authenticate the login information and authenticates the user. If authentication is successful, a screen for entering the user's preferences is displayed.

[1483] Step 2:

[1484] The user inputs their preference information. For example, they select "elegant style" and "blue color." The terminal sends this preference information to the server. The server stores the input preference information in a database. This preference information becomes the input for subsequent processing.

[1485] Step 3:

[1486] The emotion engine analyzes the user's facial expressions, tone of voice, input text, etc. to recognize the user's emotions. For example, facial expression data is acquired from a camera and analyzed using a deep learning model to recognize "joy." This emotion information is sent to the server, which temporarily stores the received emotion information.

[1487] Step 4:

[1488] The server retrieves basic information about the user (preferences, past purchase history, etc.) from a database. The server issues SQL queries to extract this information from the database and stores it in memory. This basic information becomes the input for the recommendation algorithm.

[1489] Step 5:

[1490] The server retrieves all fashion item information from the database. Specifically, detailed data such as item name, description, color, and style are extracted using SQL queries. The server receives this information and prepares to convert each item description into a TF-IDF vector.

[1491] Step 6:

[1492] The server converts the fashion item descriptions into TF-IDF vectors. Using the Python "scikit-learn" library, the item descriptions are converted into numerical vectors. User preference information is also vectorized in the same way. These vectors are used as inputs for the cosine similarity calculation.

[1493] Step 7:

[1494] The server calculates the cosine similarity and identifies the item that most closely matches the user's preferences. It outputs a similarity score by calculating the cosine similarity between vectors. It selects the item with the highest score and creates a list of recommended items.

[1495] Step 8:

[1496] The server generates compliments for the selected item. Using a generative AI model (e.g., a GPT model), it generates text that praises the good points and attractiveness of the selected item. It includes specific feedback using color and style information. It also adjusts the tone of the compliments based on emotional information from an emotion engine.

[1497] Step 9:

[1498] The server sends the generated list of recommended items and compliments to the device. The generated text data and item information are sent to the device in JSON format. The device displays these results to the user. The user can check the suggested outfits and their feedback and wear the outfits they have chosen with confidence.

[1499] (Application example 2)

[1500] 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."

[1501] Conventional fashion recommendation systems can recommend clothing based on a user's preferences, but they are unable to provide emotional feedback on the selection. This leaves users with little confidence in the recommended clothing, resulting in low satisfaction with their selection. Furthermore, because previous systems do not take the user's emotional state into account, they are unable to provide appropriate feedback that reflects the user's psychological state at any given time. Therefore, the objective of this invention is to provide a system that takes into account the user's preference information and emotional information, allowing the user to select fashion items with greater satisfaction and confidence.

[1502] 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.

[1503] In this invention, the server includes means for acquiring user preference information, means for acquiring owned clothing data, means for recommending optimal clothing based on the user preference information and the owned clothing data, means for generating compliments for the recommended clothing, means for acquiring user emotion information and adjusting the tone of the compliments based on the emotion information, and means for displaying the optimal clothing after it has been recommended. This makes it possible not only to recommend optimal clothing based on the user preference information, but also to generate compliments for the selected clothing and provide feedback according to the user's emotions.

[1504] "User preference information" is data relating to the user's preferences such as style, color, and design.

[1505] "Owned clothing data" is data about clothing owned by the user, and includes information such as item name, color, style, and description.

[1506] The "means for recommending optimal clothing" is a means for identifying and recommending the most suitable clothing for a user based on the user's preference information and clothing data owned by the user.

[1507] The "means for generating compliments" is a means for generating feedback praising the goodness and attractiveness of the recommended outfit.

[1508] "User emotion information" is data related to emotions analyzed from the user's facial expression, tone of voice, input text, and the like.

[1509] The "means for adjusting the tone of the compliment" is a means for adjusting the content and tone of the generated compliment based on the acquired emotional information of the user.

[1510] The "means for displaying the optimal outfit after it has been recommended" is a means for visually presenting information about the selected outfit and feedback thereon to the user.

[1511] This system recommends optimal clothing based on a user's preference information and clothing data, and generates compliments for the selected clothing. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it provides feedback based on the user's emotions. This system can be implemented as an application that runs on smart glasses, head-mounted displays, and smartphones.

[1512] Hardware and software used

[1513] Hardware:

[1514] Smart Glasses

[1515] head-mounted display

[1516] Smartphone

[1517] software:

[1518] Python environment

[1519] TF-IDF Vectorizer (scikit-learn)

[1520] Cosine similarity calculation module (scikit-learn)

[1521] Sentiment analysis engine (AWS Rekognition, Google Cloud Vision, or TextBlob)

[1522] Data processing and calculation

[1523] 1. Data Vectorization:

[1524] The server converts the user's preference information and the description of the clothing data they own into a TF-IDF vector, which allows the data features to be expressed as a numerical vector and made comparable.

[1525] 2. Similarity calculation:

[1526] The server calculates the cosine similarity between the vectorized user preference information and the clothing data, and identifies the items that best match the preferences.

[1527] 3. Compliment Generation:

[1528] The server generates feedback for the selected item, praising its merits and attractiveness, and adjusts the content and tone of the praise based on the user's emotional information obtained from the emotion engine.

[1529] 4. Displaying the results:

[1530] The user device visually presents the recommended item list and compliments received from the server to the user, allowing the user to check the suggested outfit and receive feedback on it.

[1531] Specific examples

[1532] For example, if a user inputs that they like "elegant style" and "blue," the system processes this preference information. The emotion engine analyzes the user's facial expression and recognizes, for example, that the user is happy. The server vectorizes the user's preference information and clothing data, calculates the cosine similarity, and identifies the item that best matches the user's preference.

[1533] For a selected item (e.g., "an elegant blue dress"), the server generates a compliment such as "The dress you chose is very elegant and graceful!" Furthermore, based on the emotional information that the user is pleased, the tone of the compliment is adjusted to be more positive.

[1534] Prompt Sentence Examples

[1535] Here are some example input prompts for a generative AI model:

[1536] "A system that recommends the most suitable outfit based on the user's preference information and clothing data, and generates compliments for the selected outfit. Furthermore, please combine it with an emotion engine that recognizes the user's emotions and provides emotion-based feedback."

[1537] The above is a specific embodiment of the present invention. This system allows users to find the perfect outfit based on their preferences and feel confident in their choice. Furthermore, by incorporating an emotion engine, it is possible to provide feedback based on the user's emotions.

[1538] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1539] Step 1:

[1540] The user logs in to the device and inputs their preferences (e.g., preferred style, color, design, etc.). The user's input information is acquired by the device and sent to the server. The input data includes the user's preferred style and color, as well as past purchase history. The server stores this information in a database.

[1541] Step 2:

[1542] The server activates an emotion engine and analyzes the user's facial expressions, tone of voice, input text, etc. to obtain emotional information. Specifically, the user's facial expressions and voice data obtained using a camera or microphone are sent to the emotion analysis engine, which extracts emotional information as positive, negative, or neutral. This emotional information is then sent to the server and stored in a database.

[1543] Step 3:

[1544] The server retrieves basic information about the user (preferences, past purchase history, etc.) from the database. At the same time, it also retrieves data about all fashion items from the database. This data includes item names, descriptions, colors, styles, etc.

[1545] Step 4:

[1546] The server converts the descriptions of fashion items into TF-IDF vectors. Then, it converts the user's preference information into TF-IDF vectors in the same way. This converts the string data into numerical vectors, making it possible to perform calculations and comparisons.

[1547] Step 5:

[1548] The server uses the vectorized data to calculate the cosine similarity, which identifies items that are closest to the user's preferences. Items with a high cosine similarity are deemed to be suitable for the user's preferences.

[1549] Step 6:

[1550] The server recommends the most similar items to the user by listing them in descending order of cosine similarity, and selecting the top few items. Detailed information about these items (such as names, descriptions, and images) is then retrieved.

[1551] Step 7:

[1552] The server generates compliments for each selected item. The generated compliments highlight the item's quality and attractiveness based on its color and style. Furthermore, the tone of the compliments is adjusted based on the user's emotional information obtained from the emotion engine. For example, if the user is in a positive state, the tone of the compliments will be brighter and more positive.

[1553] Step 8:

[1554] The server sends the generated list of recommended items and compliments to the device. Specifically, it includes the list of recommended items, compliments for each item, and detailed information about each item (image, description, price, etc.). The device visually presents this information to the user so that the user can confirm it.

[1555] The above is the processing flow based on a specific embodiment of the present invention. This system allows users to find the best outfit based on their personal preferences and feel confident about the outfit they choose. It is also possible to provide feedback based on the user's emotions through an emotion engine.

[1556] 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.

[1557] 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.

[1558] 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.

[1559] 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.

[1560] 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.

[1561] 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.

[1562] 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).

[1563] 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.

[1564] 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."

[1565] 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.

[1566] 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).

[1567] 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.

[1568] 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.

[1569] 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.

[1570] 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.

[1571] 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.

[1572] 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.

[1573] 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.

[1574] 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.

[1575] 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.

[1576] 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.

[1577] The following is further disclosed regarding the above embodiment.

[1578] (Claim 1)

[1579] A means for acquiring user preference information;

[1580] A means for acquiring owned clothing data;

[1581] A means for recommending optimal clothing based on the user's preference information and clothing data;

[1582] means for generating compliments for the recommended outfit;

[1583] A system including:

[1584] (Claim 2)

[1585] A means for converting user preference information into a TF-IDF vector;

[1586] A means to convert the clothing description data into TF-IDF vectors;

[1587] A means for calculating cosine similarity and selecting the most suitable outfit is included.

[1588] 10. The system of claim 1.

[1589] (Claim 3)

[1590] means for generating compliments based on clothing color and style;

[1591] 10. The system of claim 1.

[1592] "Example 1"

[1593] (Claim 1)

[1594] A means for acquiring user preference information;

[1595] A means for acquiring owned clothing data;

[1596] A means for recommending optimal clothing based on the user's preference information and clothing data;

[1597] A means for generating compliments for the recommended outfit using a generative AI model; and

[1598] A system including:

[1599] (Claim 2)

[1600] A means for converting user preference information into a TF-IDF vector;

[1601] A means to convert the clothing description data into TF-IDF vectors;

[1602] A means for calculating cosine similarity and selecting the most suitable outfit is included.

[1603] 10. The system of claim 1.

[1604] (Claim 3)

[1605] means for generating compliments using a generative AI model based on clothing color and style;

[1606] 10. The system of claim 1.

[1607] "Application Example 1"

[1608] (Claim 1)

[1609] A means for acquiring user preference information;

[1610] A means for acquiring owned clothing data;

[1611] A means for recommending optimal clothing based on the user's preference information and clothing data;

[1612] means for generating compliments for the recommended outfit;

[1613] a means for displaying these results;

[1614] A system including:

[1615] (Claim 2)

[1616] A means for converting user preference information into a TF-IDF vector;

[1617] A means to convert the clothing description data into TF-IDF vectors;

[1618] A means for calculating cosine similarity and selecting the most suitable outfit is included.

[1619] 10. The system of claim 1.

[1620] (Claim 3)

[1621] a means for generating compliments based on clothing color and style;

[1622] Including prompts generated using AI models,

[1623] 10. The system of claim 1.

[1624] "Example 2: Combining Emotion Engines"

[1625] (Claim 1)

[1626] A means for acquiring user preference information;

[1627] A means for acquiring owned clothing data;

[1628] A means for recommending optimal clothing based on the user's preference information and clothing data;

[1629] means for generating compliments for the recommended outfit;

[1630] A means of recognizing emotions from the user's facial expressions, tone of voice, input text, etc.;

[1631] A means to adjust the content and tone of compliments based on emotional information;

[1632] A system including:

[1633] (Claim 2)

[1634] A means for converting user preference information into a TF-IDF vector;

[1635] A means to convert the clothing description data into TF-IDF vectors;

[1636] A means for calculating cosine similarity and selecting the most suitable outfit is included.

[1637] 10. The system of claim 1.

[1638] (Claim 3)

[1639] means for generating compliments based on clothing color and style;

[1640] 10. The system of claim 1.

[1641] "Application example 2 when combining emotion engines"

[1642] (Claim 1)

[1643] A means for acquiring user preference information;

[1644] A means for acquiring owned clothing data;

[1645] A means for recommending optimal clothing based on the user's preference information and clothing data;

[1646] means for generating compliments for the recommended outfit;

[1647] means for acquiring emotional information of the user and adjusting the tone of the compliment based on the emotional information;

[1648] Provide a way to display the best outfits after they are recommended

[1649] A system including:

[1650] (Claim 2)

[1651] A means for converting user preference information into a TF-IDF vector;

[1652] A means to convert the clothing description data into TF-IDF vectors;

[1653] A means for calculating cosine similarity and selecting the most suitable outfit is included.

[1654] 10. The system of claim 1.

[1655] (Claim 3)

[1656] a means for generating compliments based on clothing color and style;

[1657] means for adjusting the tone of the compliment based on the user's emotional information;

[1658] 10. The system of claim 1. [Explanation of symbols]

[1659] 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 acquiring user preference information; A means for acquiring owned clothing data; A means for recommending optimal clothing based on the user's preference information and clothing data; means for generating compliments for the recommended outfit; A system including:

2. A means for converting user preference information into a TF-IDF vector; A means to convert the clothing description data into TF-IDF vectors; A means for calculating cosine similarity and selecting the most suitable outfit is included. The system of claim 1 .

3. means for generating compliments based on clothing color and style; The system of claim 1 .

Images