System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting browsing histories, purchase histories, and review information of users; means for classifying personality and preference types of the users based on the collected information; means for training a generative AI model using the classified information; means for classifying all users into personality and preference types and generating user profiles; means for selecting a corresponding generative AI model based on the user profiles and generating personalized reviews; and means for displaying the generated reviews on user terminals.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 Internet environment, users are faced with a vast amount of content and product information. Effectively finding information that suits their preferences is an extremely difficult task. Furthermore, traditional reviews and ratings are heavily dependent on individual preferences, and general reviews have difficulty providing information that is appropriate for each individual user. This increases the risk that users will select products or services based on inappropriate information. Furthermore, companies are also faced with the challenge of reduced marketing efficiency and wasted resources due to their inability to provide appropriate review information. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means: A means for collecting users' browsing history, purchase history, and review information, and a means for classifying users' personalities and preference types based on the collected data. A means for training a generative AI model using the classified data and classifying all users into personality and preference types to generate user profiles is then provided. Furthermore, a system is provided that includes a means for selecting a corresponding generative AI model based on the user profile, generating personalized reviews, and a means for displaying the generated reviews on the user's device. This system enables users to efficiently obtain optimal review information based on their preferences, enabling companies to conduct optimal marketing.

[0006] "User" refers to an individual who uses the Internet to view, purchase, rate, review, or otherwise engage in activities such as viewing content or products.

[0007] "Browser history" refers to a record of content or products that a user has viewed on the Internet in the past.

[0008] "Purchase history" refers to a record of products and services a user has previously purchased over the Internet.

[0009] "Review information" refers to information such as ratings and comments made by users on content or products.

[0010] "Personality / preference type" refers to a category that classifies users based on their personality traits and preferences.

[0011] "Generative AI model" refers to an artificial intelligence model that uses machine learning technology to generate reviews that are specialized to a user's personality and preferences.

[0012] "Training data" refers to a data set used to train a generative AI model, and includes review information and ratings based on user personality and preference types.

[0013] "User profile" refers to information that integrates each user's personality, preferences, browsing history, purchase history, and review information.

[0014] "Personalized reviews" refer to review information optimized for a user's personality and preferences.

[0015] "Server" refers to a computing system that collects, processes, and stores data, and runs generative AI models.

[0016] A "terminal" is a device that a user uses to access the Internet, including a PC, smartphone, tablet, etc. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10]1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

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

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] This invention is a personalized AI review system that efficiently provides content suited to the interests and preferences of individual users on the Internet. This system collects users' browsing history, purchase history, and review information, and uses this information to train a generative AI model to provide optimal reviews to users.

[0039] Program processing flow

[0040] Step 1: Collect user data

[0041] The server collects browsing history, purchase information, and review information generated when users use e-commerce sites and content distribution services, and stores them in a database. Specifically, when users watch a movie, the server collects the viewing date and time, movie title, rating, and review content.

[0042] Step 2: Classify and train the sampled data

[0043] The server sends a questionnaire to some users to collect information about their personalities and preferences. Based on the collected questionnaire results, the server classifies users into personality and preference types, such as "action movie lover" and "comedy movie lover." This classified data is used as training data.

[0044] Step 3: Training the generative AI model

[0045] The server trains a generative AI model based on the classified training data. For example, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. Similarly, separate generative AI models are trained for other types.

[0046] Step 4: Classify all users into types

[0047] The server analyzes all user data, including purchase history, browsing history, and review information, and classifies each user into a personality type and preference type, laying the foundation for providing personalized reviews using a generative AI model that is optimal for each user.

[0048] Step 5: Generate a personalized review

[0049] When a user browses for new content or products, the device sends the request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses the appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0050] Specific examples

[0051] Example 1: Movie reviews

[0052] When User A watches Movie A, the device sends viewing information (viewing date and time, movie title, rating, review content) to the server. The server collects the data and stores it in a database. Later, when User A watches a new action movie B, the server checks the user profile and classifies them as an "action movie lover." It uses a generative AI model to generate a review of Action Movie B and sends it to the device. The device displays the review on Movie B's page.

[0053] Example 2: Product Review

[0054] User X purchases gadget Z and leaves a rating and review. This information is sent from the device to the server and stored in a database. Next, when User X browses for a new gadget Y, the server references the user profile and generates a personalized review using a generative AI model. The generated review is then displayed on User X's device.

[0055] This system allows users to easily obtain reviews that match their preferences, making information gathering and content selection more efficient. It also enables businesses to conduct marketing based on user preferences, enabling more effective advertising activities.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] Users use e-commerce sites and content distribution services to browse products, purchase them, and post reviews.

[0059] The device collects the user's browsing history, purchase information, and review information and sends it to the server.

[0060] The server stores the received data in a database.

[0061] Step 2:

[0062] The server sends a questionnaire to some users to collect information about their personalities and preferences.

[0063] The server analyzes the survey results and classifies users into personality and preference types.

[0064] Based on the survey, people are classified into categories such as "action movie lovers" and "comedy movie lovers."

[0065] Step 3:

[0066] The server extracts the classified user reviews and rating information as training data.

[0067] This training data is used to train generative AI models specialized for each personality and preference type.

[0068] For example, we will use user data of "action movie lovers" to train an AI model for generating action movie reviews.

[0069] Step 4:

[0070] The server analyzes all users' browsing history, purchase history, and review information.

[0071] Based on the analysis results, all users are classified into personality and preference types.

[0072] A user profile is generated and stored in a database.

[0073] Step 5:

[0074] When a user browses for new products or content, the device sends a request to the server.

[0075] The server references the user profile and checks the user's personality and preferences.

[0076] Select the appropriate generative AI model to generate personalized reviews.

[0077] Step 6:

[0078] The server sends the generated reviews to the device.

[0079] The terminal displays the received reviews to the user.

[0080] For example, when a user browses to a page about new action movie B, they will see reviews created using a generative AI model specialized for action movies.

[0081] Specific examples

[0082] Example 1: Movie reviews

[0083] 1. User A watches Movie A and posts a review.

[0084] 2. The device collects viewing information and sends it to the server.

[0085] 3. The server saves the data to a database.

[0086] 4. Using the survey results of User A, classify him as an "action movie lover."

[0087] 5. The server trains the generative AI model using User A's data.

[0088] 6. When user A wants to watch a new action movie B, the device sends a request.

[0089] 7. The server checks User A's profile and generates a personalized review.

[0090] 8. The generated review is displayed on User A's device.

[0091] Example 2: Product review

[0092] 1. User X buys gadget Z and posts a review.

[0093] 2. The device collects purchase information and sends it to the server.

[0094] 3. The server saves the data to a database.

[0095] 4. Using the survey results of User X, classify him / her as a "gadget enthusiast type."

[0096] 5. The server trains the generative AI model using User X's data.

[0097] 6. When user X views new gadget Y, the device sends a request.

[0098] 7. The server checks User X's profile and generates a personalized review.

[0099] 8. The generated review is displayed on User X's device.

[0100] Example 1

[0101] 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."

[0102] Many current e-commerce sites and content distribution services only provide users with general reviews and ratings, making it difficult to provide customized reviews that match the personalities and preferences of individual users. Furthermore, users lack the information they need to select the content and products that best suit them, preventing them from making efficient selections. To address these issues, a system is needed that provides reviews based on the individual preferences of users.

[0103] 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.

[0104] In this invention, the server includes means for collecting user operation history, purchase history, and evaluation information, means for classifying user personality and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into personality and preference types and generating user information, means for selecting a corresponding generative AI model based on the user information and generating customized reviews, and means for displaying the generated reviews on a user device. This allows users to easily obtain reviews that match their preferences, enabling them to efficiently collect information and select content.

[0105] "User operation history" refers to the record of operations performed by a user on an e-commerce site or content distribution service, and specifically includes the products and content viewed, the links clicked, and the user's behavior on the site.

[0106] "Purchase history" refers to a record of products and services that a user actually purchased on an e-commerce site, etc., and includes information such as the purchase date and time, product name, quantity, and price.

[0107] "Rating information" refers to records of ratings and reviews made by users on products, services, and content purchased by users, and includes feedback such as rating scores and comments.

[0108] A "generative artificial intelligence model" is an AI model that is generated by training a machine learning algorithm using collected data, and has the ability to predict and generate for specific tasks.

[0109] "User information" refers to the classification of a user's personality and preferences based on collected data, and the profile information based on that classification, including data regarding the user's interests and preferences.

[0110] "Customized reviews" refer to personalized reviews for individual users that are generated based on the user's personality and preferences, and unlike general reviews, contain content that reflects the user's specific needs and interests.

[0111] "User equipment" refers to a device used by a user through an Internet connection, and specifically includes smartphones, tablets, PCs, etc.

[0112] "Immediate generation" means that the generation process begins immediately after the user submits the request, without delay, and the results are provided within a short time.

[0113] This invention is a personalized AI review system for efficiently providing content suited to the interests and preferences of individual users on the Internet. This system collects users' operation history, purchase history, and rating information, and uses this information to train a generative AI model to provide optimal reviews to users.

[0114] The server collects operation history, purchase information, and rating information generated when users use e-commerce sites and content distribution services, and stores this information in a database. Specifically, the hardware and software that collects information such as the viewing date and time, movie title, rating, and review content when users watch a movie can stream data in real time using Apache Kafka, allowing data to be collected and stored efficiently.

[0115] The server sends a questionnaire to a subset of users to collect information about their personalities and preferences. Based on the survey results, users are classified into personality and preference types, such as "action movie lover" or "comedy movie lover." This classified data is used as training data. Data analysis is performed using data science languages ​​such as Python and R.

[0116] The server trains a generative AI model based on the classified training data. For example, a generative AI model specialized for action movies is created using user data for "action movie lovers." This training is performed using machine learning frameworks such as TensorFlow and PyTorch. The trained model is saved on the server and used for subsequent processing.

[0117] The server analyzes all user data and classifies each user into personality and preference types. This is done using an analytical algorithm based on the user's purchase history, operation history, and rating information. The classified user information is saved in a database as a profile.

[0118] When a user browses for new content or products, the device sends a request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses an appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0119] As a concrete example, when a user watches action movie B, the device sends viewing information (viewing date and time, movie title, rating, review content) to the server. The server collects the data and stores it in a database. Later, when the user views a new action movie C, the server checks the user profile and selects an appropriate generative AI model. The generated review is then displayed on the user's device.

[0120] Example prompt sentence:

[0121] 1. "Generate a personalized review of action movie B that user A is watching."

[0122] 2. "Generate a review for gadget Y by user X."

[0123] 3. "For users who like comedy movies, please write a review of a new comedy movie, C."

[0124] This system allows users to easily obtain reviews that match their preferences, enabling them to efficiently gather information and select content. It also enables businesses to conduct marketing based on user preferences, enabling more effective advertising activities.

[0125] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0126] Step 1:

[0127] The server collects operation history, purchase information, and rating information generated when a user uses an e-commerce site or content distribution service. Specifically, the device collects data such as the user's viewing date and time, movie title, rating, and review content, and sends this data to the server in real time. The server stores the received data in a database.

[0128] Input: User operation history, purchase information, rating information

[0129] Data processing / calculation: Real-time streaming, saving to database

[0130] Output: Saved data

[0131] Step 2:

[0132] The server sends a questionnaire to some users to collect information about their personalities and preferences. The users answer the questionnaire and send the results to the server. The server analyzes the questionnaire results and classifies users into categories such as "action movie lover" or "comedy movie lover."

[0133] Input: Survey response

[0134] Data processing / calculation: Analysis of survey results, classification of users

[0135] Output: Classified user data

[0136] Step 3:

[0137] The server trains a generative AI model based on the classified user data. For example, it uses user data for "action movie lovers" to create an AI model specialized for action movies. Specifically, it uses the collected data as training data using TensorFlow and PyTorch to train the generative AI model.

[0138] Input: Classified user data

[0139] Data processing / computation: training generative AI models

[0140] Output: A trained generative AI model

[0141] Step 4:

[0142] The server analyzes all user data and classifies each user into personality and preference types. This is done using an analytical algorithm based on the user's purchase history, operation history, and rating information. The classified user information is saved in a database as a profile.

[0143] Input: Purchase history, operation history, and rating information of all users

[0144] Data processing / calculation: Data analysis, user classification

[0145] Output: User profile

[0146] Step 5:

[0147] When a user browses for new content or products, the device sends a request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses an appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0148] Input: User request, user profile

[0149] Data processing / calculation: Profile reference, review generation using AI model

[0150] Output: Generated reviews

[0151] In this way, personalized reviews based on the user's preferences can be efficiently provided.

[0152] (Application example 1)

[0153] 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."

[0154] In today's Internet environment, users face the challenge of finding the content and products that best suit them from the vast amount of information available. Furthermore, current review systems often provide general opinions, and rarely provide information specific to individual users' preferences and characteristics. This means that users spend time and effort sifting through information to select the content and products that best suit them.

[0155] 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.

[0156] In this invention, the server includes means for collecting user browsing history, purchase history, and rating information, means for classifying user characteristics and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into characteristics and preference types and generating user profiles, means for selecting corresponding generative AI models based on the user profiles and generating personalized reviews, means for displaying the generated reviews on an information processing device, and means implemented as a smartphone application for providing optimal reviews for users in real time based on their viewing history and purchase history. This allows users to quickly obtain review information that matches their preferences and characteristics, making content and product selection efficient and effective.

[0157] A "user's browsing history" is a record of which web pages and content a user has viewed on the Internet.

[0158] "Purchase history" is a record of what products a user has purchased on the Internet.

[0159] "Rating information" refers to information about ratings and reviews of content or products that a user has viewed or purchased.

[0160] "Characteristics / preference type" refers to a type classified based on the user's preferences and personality traits.

[0161] A "user profile" is a detailed profile of personal information created based on data such as a user's characteristics, preferences, browsing history, and purchase history.

[0162] A "generative AI model" is an artificial intelligence model that uses machine learning to learn from a specific dataset and perform a specific task.

[0163] A "personalized review" is a customized review generated based on the individual characteristics and preferences of a user.

[0164] An "information processing device" is an electronic device that has the function of inputting, processing, and outputting data, and here it mainly refers to a smartphone.

[0165] A "smartphone application" is a software program that runs on a smartphone.

[0166] This invention is a system that collects users' browsing history, purchase history, and rating information, and generates personalized reviews based on the users' characteristics and preferences. This system is composed of a server, an information processing device (such as a smartphone), a generation AI model, etc. Specific embodiments are described below.

[0167] Hardware and software used

[0168] Server: Cloud server (e.g. Amazon Web Services EC2)

[0169] Database: Cloud database (e.g. Amazon RDS)

[0170] Generative AI models: models for natural language generation (e.g., OpenAI GPT)

[0171] Data processing device: Smartphone (e.g., Android or iOS app)

[0172] Data processing and calculation

[0173] Data collection and storage

[0174] The server collects browsing history, purchase history, and rating information generated when a user uses an information processing device (smartphone), and stores this data in a cloud database. For example, when a user watches a movie, the movie title, viewing date and time, rating, and review content are stored. This data serves as the basis for analyzing user preferences.

[0175] User Classification and Data Classification

[0176] Based on the collected data, the server categorizes users into characteristics and preferences. Specifically, based on their browsing and purchase history, users may be classified as those who like action movies or comedy movies. This categorization is stored in a cloud database and used as training data for the generative AI model.

[0177] Training an AI model

[0178] The server uses the classified data to train a generative AI model. For example, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. In this process, generative AI models such as OpenAI GPT are used.

[0179] Generate personalized reviews

[0180] When a user browses new content on an information processing device (smartphone), the server references the user profile and uses the corresponding generative AI model to generate a personalized review, which is generated in real time and displayed on the information processing device.

[0181] Specific examples

[0182] Example 1: Personalized movie reviews

[0183] When a user browses for a new movie using their smartphone, the server checks the user's profile and, if the user is classified as an "action movie lover," uses the generative AI model to generate a personalized review of that movie for action movie lovers. The generated review is displayed on the smartphone, allowing the user to use it as a reference when choosing a movie.

[0184] Prompt Sentence Examples

[0185] A user loved the action movie and gave the following review: It was packed with great action scenes and I really enjoyed it!

[0186] Example 2: Personalizing product reviews

[0187] When a user browses for a new gadget, the server checks the user's profile and uses a generative AI model to generate a personalized review of the gadget, which is then displayed on the user's smartphone to help inform their purchasing decision.

[0188] In this way, the invention provides personalized reviews based on the user's preferences, allowing the user to efficiently select the content and products that are best suited to them, thereby significantly improving the user experience.

[0189] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0190] Step 1:

[0191] Data collection

[0192] When a user browses or purchases using a device (smartphone), that data is collected. Specifically, when a user watches a movie, information such as the viewing date and time, movie title, rating, and review content is sent from the device to the server. The input data is viewing history, purchase history, and rating information, and the server stores this data in a cloud database. The output is the stored user data.

[0193] Step 2:

[0194] User Classification

[0195] The server classifies users into characteristics and preference types based on the collected data. The collected data (input) includes the user's browsing history and purchase history. The server analyzes this data to determine the genre of content the user prefers. For example, a user who watches a lot of action movies would be classified as an "action movie lover." This classification information (output) is stored in a cloud database.

[0196] Step 3:

[0197] Training generative AI models

[0198] The server uses the classified data (input) to train the generative AI model. Specifically, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. The generative AI model used in this process is OpenAI GPT, for example. Once training is complete, multiple generative AI models (output) according to their characteristics are obtained.

[0199] Step 4:

[0200] Creating a user profile

[0201] The server analyzes the data of all users, classifies each user into a characteristic and preference type, and generates a user profile. The input is the user's browsing history, purchase history, and rating information. The server analyzes this data and determines which preference type each user belongs to. Based on this determination, an individual user profile (output) is generated and stored in a cloud database.

[0202] Step 5:

[0203] Generate personalized reviews

[0204] When a user browses new content, a request is sent from the device to the server. The server references the user profile (input) and generates a personalized review using the corresponding generative AI model. Specifically, the server checks the user's preference type and generates a review using prompts appropriate for that type. The generated review (output) is sent to the device and displayed to the user.

[0205] Step 6:

[0206] View Reviews

[0207] The terminal displays the personalized reviews (input) received from the server to the user, allowing the user to select new content or products based on reviews that match their preferences. The output is the displayed reviews.

[0208] Specific examples

[0209] Example prompt sentence:

[0210] A user loved the action movie and gave the following review: It was packed with great action scenes and I really enjoyed it!

[0211] 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.

[0212] This invention combines a personalized AI review system that provides optimal content and product information based on a user's interests and preferences with an emotion engine that recognizes the user's emotions. This system can provide even more personalized reviews based on the user's emotional state at any given time.

[0213] Program processing flow

[0214] Step 1: Collect user data

[0215] The server collects browsing history, purchase information, and review information generated when a user uses an e-commerce site or content distribution service, and stores this information in a database. For example, when a user watches a movie, the server collects the viewing date and time, movie title, rating, and review content.

[0216] Step 2: Classify and train the sampled data

[0217] The server sends a questionnaire to a certain number of users to collect information about their personalities and preferences. The server analyzes the survey results and classifies users into personality and preference types. For example, it classifies users into "action movie lovers" and "comedy movie lovers." Based on this, it generates training data and trains the generative AI model.

[0218] Step 3: Training the generative AI model

[0219] The server trains generative AI models for each personality and preference type based on the classified training data. For example, a generative AI model specialized for action movies is created using user data for "action movie lovers." Individual generative AI models are similarly trained for other types.

[0220] Step 4: Classify all users into types

[0221] The server analyzes all user data, including purchase history, browsing history, and review information, and classifies each user into a personality and preference type. Based on the analysis results, a user profile is generated and stored in a database.

[0222] Step 5: Implementing the Emotion Engine

[0223] The device is equipped with an emotion engine that recognizes the user's emotions in real time. The emotion engine analyzes facial expressions and speech while the user is viewing content to recognize the user's emotional state.

[0224] Step 6: Generate a personalized review

[0225] When a user browses new products or content, the device sends a request and emotional information to the server, which then references the user profile and emotional information to select a corresponding generative AI model to generate a personalized review.

[0226] Step 7: View personalized reviews

[0227] The server sends the generated review to the device, which then displays the received review to the user. For example, when a user browses a page about a new action movie B, if the emotion engine recognizes that the user is excited, it generates and displays a review that focuses on the action scenes.

[0228] Specific examples

[0229] Example 1: Movie reviews

[0230] User A watches Movie A and posts a review. The device collects viewing information and sends it to the server. The server saves the data in a database. User A's emotional information is then analyzed using an emotion engine. The server creates an optimal review based on the user's personality, preference type, and emotional information, and sends it to the device. When User A watches Movie B, the emotion engine recognizes that User A is excited, and generates and displays a review that suits that situation.

[0231] Example 2: Product review

[0232] User X purchases gadget Z and posts a review. The device collects purchase information and sends it to the server. The server saves the data in a database. The emotion engine then analyzes User X's emotional information. The server creates an optimal review based on the user's personality, preference type, and emotional information, and sends it to the device. When User X browses new gadget Y, the emotion engine recognizes that User X is surprised, and generates and displays a review that reflects that emotion.

[0233] This system allows users to obtain the most appropriate reviews based on their preferences and current emotional state. This makes information gathering and content selection more efficient and personalized. It also enables businesses to conduct more detailed marketing based on users' emotions, which is expected to produce even greater results.

[0234] The processing flow will be explained below.

[0235] Step 1:

[0236] Users use e-commerce sites and content distribution services to browse, purchase, and post reviews of products.

[0237] The device collects the user's browsing history, purchase information, and review information and sends it to the server.

[0238] The server stores the received data in a database.

[0239] Step 2:

[0240] The server sends a questionnaire to some users to collect information about their personalities and preferences.

[0241] The server analyzes the survey results and classifies users into personality and preference types.

[0242] For example, classifications such as "action movie lover type" and "comedy movie lover type" are made.

[0243] Step 3:

[0244] The server extracts the classified user reviews and rating information as training data.

[0245] This training data is used to train generative AI models specialized for each personality and preference type.

[0246] For example, we will use user data of "action movie lovers" to train an AI model for generating action movie reviews.

[0247] Step 4:

[0248] The server analyzes all users' browsing history, purchase history, and review information.

[0249] Based on the analysis results, all users are classified into personality and preference types.

[0250] A user profile is generated and stored in a database.

[0251] Step 5:

[0252] The device is equipped with an emotion engine that allows it to recognize the user's emotions in real time.

[0253] The emotion engine analyzes the user's facial expressions and speech while viewing content, and recognizes the user's emotional state.

[0254] Step 6:

[0255] When a user browses for new products or content, the device sends the request and emotion information to the server.

[0256] The server refers to the user profile and sentiment information, selects a corresponding generative AI model, and generates a personalized review.

[0257] Step 7:

[0258] The server sends the generated reviews to the device.

[0259] The terminal displays the received reviews to the user.

[0260] For example, when a user browses a page about a new action movie B, if the emotion engine recognizes that the user is excited, it will generate a review that focuses on the action scenes and display it on the device.

[0261] Specific examples

[0262] Example 1: Movie reviews

[0263] 1. User A watches Movie A and posts a review.

[0264] 2. The device collects viewing information and sends it to the server.

[0265] 3. The server saves the data to a database.

[0266] 4. Using the survey results of User A, classify him as an "action movie lover."

[0267] 5. The server trains the generative AI model using User A's data.

[0268] 6. When user A watches movie B, the emotion engine analyzes user A's emotional information in real time.

[0269] 7. The device sends the real-time emotion information to the server.

[0270] 8. The server references user A's profile and sentiment information to generate the most appropriate review.

[0271] 9. The review generated is sent back to the device and displayed on the page for Movie B that User A is viewing.

[0272] Example 2: Product review

[0273] 1. User X buys gadget Z and posts a review.

[0274] 2. The device collects purchase information and sends it to the server.

[0275] 3. The server saves the data to a database.

[0276] 4. Use the survey results of User X to classify him / her into the "gadget enthusiast type."

[0277] 5. The server trains the generative AI model using User X's data.

[0278] 6. When user X browses new gadget Y, the emotion engine analyzes user X's emotion information in real time.

[0279] 7. The device sends the real-time emotion information to the server.

[0280] 8. The server references user X's profile and sentiment information to generate the most appropriate review.

[0281] 9. The device displays the generated review to User X, allowing User X to view a personalized review of Gadget Y in a state of amazement.

[0282] This system allows users to easily find the most suitable reviews based on their preferences and current emotional state, making information gathering and content selection more efficient and personalized. It also enables businesses to conduct more effective marketing based on user emotions, leading to greater success.

[0283] Example 2

[0284] 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."

[0285] Conventional review systems can provide personalized reviews based on a user's personality and preferences, but they cannot provide reviews that take into account the user's current emotional state. This makes it difficult for users to obtain the most appropriate information based on their emotions at any given time. It also makes it difficult for companies to conduct effective marketing based on users' emotions.

[0286] 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.

[0287] In this invention, the server includes means for collecting users' browsing history, purchase history, and review information; means for classifying users' personality and preference types based on the collected data; means for training a generative AI model using the classified data; means for classifying all users into personality and preference types and generating user profiles; means for the user terminal to recognize the user's emotional state in real time using a camera or microphone; means for selecting a corresponding generative AI model based on the user profile and emotional information and generating personalized reviews; and means for displaying the generated personalized reviews on the user terminal. This allows users to obtain optimal reviews based on their current emotional state as well as their own preferences. Companies can also conduct effective marketing based on users' emotions.

[0288] A "user's browsing history" is a record of the pages and content a user accesses on a website or application.

[0289] "Purchase history" is a list of products that a user has purchased in the past and related information.

[0290] "Review information" is data including evaluations and opinions given by users regarding products and services.

[0291] "Personality / Preference Type" is a classification based on a user's personal personality or preferences for specific content or products.

[0292] A "generative AI model" is an artificial intelligence model that generates personalized reviews based on a user's personality, preferences, and emotional state.

[0293] A "user profile" is a data set that integrates a user's personality, preferences, browsing history, purchase history, review information, etc.

[0294] "Means for recognizing emotional states in real time" refers to technology that detects the user's emotions at any given time from their facial expressions and voice.

[0295] "Personalized reviews" are the presentation of ratings and opinions that are individually optimized based on the user's personality, preference type, and emotional state.

[0296] A "user terminal" is a device operated by a user, and includes a personal computer, a smartphone, a tablet, etc.

[0297] This invention combines a personalized AI review system that provides optimal content and product information based on a user's interests and preferences with an emotion engine that recognizes the user's emotions. This system can provide even more personalized reviews based on the user's emotional state at any given time.

[0298] The server collects browsing history, purchase history, and review information generated when users use e-commerce sites and content distribution services. This data is periodically retrieved via API and inserted into a database such as MySQL or MongoDB.

[0299] The server then sends users a questionnaire to collect information about their personalities and preferences. The results of this questionnaire are analyzed using natural language processing (NLP) techniques, and algorithms such as K-means clustering are used to classify users into categories such as "action movie fans" and "comedy movie fans." This classification data is used to generate a training dataset and train a generative AI model. The generative AI model, trained using Python and libraries such as TensorFlow and PyTorch, is specialized for each preference type.

[0300] The server analyzes newly collected user data and classifies all users into preference types using a pre-trained generative AI model. Based on this result, a user profile is generated and stored in a database. To ensure real-time performance, batch processing and real-time data streaming technologies are used.

[0301] The device is equipped with an emotion engine that recognizes the user's emotions in real time. It collects facial expressions and voice data through the camera and microphone, and analyzes emotions using the OpenCV library and the Google Cloud Speech-to-Text API. The analysis results are sent to a server and integrated with the user's profile.

[0302] When a user browses for new products or content, their device sends their request and emotional information to the server. The server then selects a corresponding AI model based on the received user profile and emotional information to generate a personalized review. It uses natural language generation (NLG) technology to generate review text that reflects the user's emotions.

[0303] As a concrete example, consider a scenario in which User A watches Movie A and posts a review. The device collects viewing information and sends it to the server. The server stores the data in a database and analyzes User A's emotional information using an emotion engine. The server creates an optimal review based on the personality, preference type, and emotional information, and sends it to the device. When User A watches Movie B, the emotion engine recognizes that User A is excited, and generates and displays a review that suits that situation.

[0304] An example prompt is:

[0305] "The powerful scenes in this action movie will get you even more excited."

[0306] This system allows users to obtain the most appropriate reviews based on their preferences and current emotional state, making information gathering and content selection more efficient and personalized. It also enables companies to conduct effective marketing based on users' emotions.

[0307] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0308] Step 1:

[0309] The server collects user data. Specifically, it obtains user browsing history, purchase history, and review information via API. The input is user activity log data. The server stores the collected data in a MySQL or MongoDB database. The output is organized user data.

[0310] Step 2:

[0311] The server sends a questionnaire to a certain number of users to collect information on their personalities and preferences. The input is the questionnaire response data from the users. The server analyzes this using natural language processing (NLP) technology and classifies users into personality and preference types using algorithms such as K-means clustering. The output is data classifying the users' personalities and preference types.

[0312] Step 3:

[0313] The server uses the classified data to train a generative AI model. The input is training data categorized into personality and preference types. A generative AI model for each type is trained using Python and the TensorFlow or PyTorch library. The output is a trained generative AI model.

[0314] Step 4:

[0315] The server analyzes all user data. Specifically, it inputs purchase history, browsing history, and review information into a previously trained AI model to classify each user into a personality and preference type. The input is newly collected user data, and the output is an updated user profile.

[0316] Step 5:

[0317] The device uses an emotion engine to recognize the user's emotions in real time. Specifically, it analyzes facial and voice data acquired from the camera and microphone using OpenCV and the Google Cloud Speech-to-Text API to detect the user's emotional state. The input is the user's facial and voice data, and the output is the analyzed emotional information.

[0318] Step 6:

[0319] When a user browses for new products or content, the device sends the request and emotional information to the server. The server references the user profile and emotional information, selects a corresponding generative AI model, and generates a personalized review. The input is the request data, emotional information, and user profile, and the output is the generated review.

[0320] Step 7:

[0321] The server generates personalized reviews and sends them to the device. The device displays the reviews it receives on the user interface (UI). Specifically, the reviews are inserted into web pages or app UIs designed with HTML and CSS, and the display is updated in real time. The input is the reviews sent from the server, and the output is the reviews displayed to the user.

[0322] (Application example 2)

[0323] 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."

[0324] Conventional personalized review systems could provide personalization based on the user's basic personality and preferences, but they could not consider the user's emotional state at any given time, making it difficult to provide more detailed personalization. Furthermore, they could not generate reviews in real time that reflected the user's emotional information, making it impossible to provide feedback that was in line with the user's current emotions.

[0325] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting users' browsing history, purchase history, and review information, means for classifying users' personality and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into personality and preference types and generating user profiles, means for selecting corresponding generative AI models based on the user profiles and generating personalized reviews, emotion engine means for recognizing the user's emotional state in real time, and means for displaying personalized reviews generated based on requests including the user's emotional information on the user terminal. This makes it possible to provide personalized reviews in real time that are tailored to the user's basic personality and preferences as well as their emotional state at any given time.

[0326] "User browsing history" refers to historical information about pages and content that a user views when using a website or application.

[0327] "Purchase history" is historical information about products purchased by a user online or offline.

[0328] "Review information" refers to information about ratings and comments posted by users about products or content that they have viewed or purchased.

[0329] "Means for classifying user personality and preference types" refers to a method or system for classifying a user's personality and preferences into specific types based on collected user data.

[0330] A "means for training a generative AI model" is a method or system that uses classified user data to train a generative AI model that responds to a specific purpose.

[0331] A "user profile" is individual profile information generated by integrating a user's personality, preferences, browsing history, purchase history, and the like.

[0332] The "emotion engine" is a technology that analyzes data such as the user's facial expressions and voice, and recognizes their emotional state at any given time in real time.

[0333] A "personalized review" is a customized rating or comment on a specific product or content that takes into account a user's individual personality, preferences, and feelings.

[0334] A "user terminal" is a device that a user directly uses, and refers to hardware such as a smartphone, tablet, or computer.

[0335] MODE FOR CARRYING OUT THE INVENTION

[0336] System Overview

[0337] This invention is a system that generates personalized reviews based on a user's browsing history, purchase history, and review information. Its special feature is that it recognizes the user's current emotional state and provides reviews that correspond to that emotion at that time. This system consists of a user terminal equipped with an emotion engine and a server that analyzes and processes the data.

[0338] Hardware and software used

[0339] Hardware

[0340] User devices: smartphones, tablets, computers, etc.

[0341] Server: High-performance computer, cloud infrastructure

[0342] software

[0343] EmotionEngine: Software that recognizes a user's emotional state by analyzing their facial expressions and voice

[0344] AIReviewGenerator: A generative AI model that generates personalized reviews based on user profiles and sentiment information

[0345] Django or Flask: a framework for server-side data management and API serving.

[0346] Processing flow and specific examples

[0347] Step 1: Data collection

[0348] The server collects user data such as viewing history, purchase history, and review information from the user's device and stores it in a database. For example, when a user watches a particular movie, the viewing date and time, movie title, rating, and review content are recorded.

[0349] Step 2: Classify users

[0350] The server classifies the user's personality and preferences based on the collected data. This identifies user types such as "action movie lover" or "comedy movie lover." This classification is used to train a generative AI model.

[0351] Step 3: Introducing the Emotion Engine

[0352] The user device is equipped with an EmotionEngine that analyzes the user's facial expressions and voice to recognize their current emotional state. For example, it can analyze the user's reactions in real time while watching a movie to identify their emotional state.

[0353] Step 4: Generate a personalized review

[0354] When a user browses new products or content, the device sends the corresponding request and emotional information to the server, which then selects a corresponding generative AI model based on the user profile and emotional information to generate a personalized review for the specific content.

[0355] Step 5: View reviews

[0356] The generated review is sent to the user's device and displayed to the user. For example, when a user browses a page about a new action movie, if the emotion engine recognizes that the user is excited, a review specific to the action scenes of that movie will be generated and displayed to the user.

[0357] Specific examples

[0358] If the emotion engine determines that the user is excited after watching movie A, it will send the following prompt to the generative AI model.

[0359] Prompt statement:

[0360] User ID 123's emotional state is excited. Generate a personalized review for content he / she is interested in based on the following information: User Data: { "Action Movie Lover", "Recently Watched Movies": "Movie A", "Rating": "High"} Content ID: Movie B

[0361] In this way, a personalized review optimized for the user's current emotional state can be provided. The above-described system and process allow users to obtain reviews that are in line with their emotions, providing a more satisfying content viewing experience.

[0362] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0363] Step 1:

[0364] The server collects user data such as viewing history, purchase history, and review information from the user's device. The collected data is stored in a database. This data includes, for example, the date and time when the user watched a particular movie, the movie title, rating, and review content.

[0365] Input: Browsing history, purchase history, and review information from user devices

[0366] Output: User data stored in the database

[0367] Specific operation: When the server receives an HTTP request, it structures the data and stores it in a database.

[0368] Step 2:

[0369] The server analyzes the collected user data and classifies the user's personality and preferences. For example, it may classify them into types such as "likes action movies" or "likes comedy movies." Training data is generated based on the classification results, and the generative AI model is trained.

[0370] Input: User data stored in the database

[0371] Output: Classified user data, training data

[0372] How it works: The server uses machine learning algorithms to analyze user data and classify it into user types.

[0373] Step 3:

[0374] The server analyzes all user data and classifies each user into personality and preference types, based on which a user profile is created and stored in a database.

[0375] Input: All users' data

[0376] Output: User profile

[0377] Specific operation: The server uses the data classified in the previous step to generate a profile for each user based on their personality and preference types.

[0378] Step 4:

[0379] The user device uses the EmotionEngine to analyze the user's facial expressions and voice in real time to recognize their emotional state. For example, while watching a movie, the device can identify emotional states such as "excitement" or "joy" from the user's facial expressions and voice.

[0380] Input: User's facial expression data, voice data

[0381] Output: Perceived emotional state

[0382] Specific operation: The EmotionEngine in the device analyzes visual and audio data and outputs the emotional state.

[0383] Step 5:

[0384] When a user browses for new products or content, the terminal sends requests and emotion information to the server.

[0385] Input: User request, perceived emotional state

[0386] Output: Request and emotion information sent to the server

[0387] Specific operation: The device sends data to the server as an HTTP request.

[0388] Step 6:

[0389] The server refers to the user profile and emotion information, selects a corresponding generative AI model, and generates a personalized review.

[0390] Input: User profile, perceived emotional state

[0391] Output: The generated personalized review

[0392] Specific operation: The server generates a prompt using AI Review Generator, inputs it into the generative AI model, and generates a review.

[0393] Step 7:

[0394] The generated review is sent to the user's terminal and displayed to the user. For example, if the user is recognized as excited when browsing a page about a new action movie, a review specific to the action scenes of that movie is generated and displayed.

[0395] Input: Generated personalized review

[0396] Output: The review displayed on the user's device

[0397] Specific behavior: The server sends the generated review to the device as an HTTP response, and the device displays it.

[0398] 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.

[0399] 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.

[0400] 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.

[0401] [Second embodiment]

[0402] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0403] 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.

[0404] 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).

[0405] 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.

[0406] 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.

[0407] 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).

[0408] 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.

[0409] 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.

[0410] 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.

[0411] 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.

[0412] 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.

[0413] 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."

[0414] This invention is a personalized AI review system that efficiently provides content suited to the interests and preferences of individual users on the Internet. This system collects users' browsing history, purchase history, and review information, and uses this information to train a generative AI model to provide optimal reviews to users.

[0415] Program processing flow

[0416] Step 1: Collect user data

[0417] The server collects browsing history, purchase information, and review information generated when users use e-commerce sites and content distribution services, and stores them in a database. Specifically, when users watch a movie, the server collects the viewing date and time, movie title, rating, and review content.

[0418] Step 2: Classify and train the sampled data

[0419] The server sends a questionnaire to some users to collect information about their personalities and preferences. Based on the collected questionnaire results, the server classifies users into personality and preference types, such as "action movie lover" and "comedy movie lover." This classified data is used as training data.

[0420] Step 3: Training the generative AI model

[0421] The server trains a generative AI model based on the classified training data. For example, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. Similarly, separate generative AI models are trained for other types.

[0422] Step 4: Classify all users into types

[0423] The server analyzes all user data, including purchase history, browsing history, and review information, and classifies each user into a personality type and preference type, laying the foundation for providing personalized reviews using a generative AI model that is optimal for each user.

[0424] Step 5: Generate a personalized review

[0425] When a user browses for new content or products, the device sends the request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses the appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0426] Specific examples

[0427] Example 1: Movie reviews

[0428] When User A watches Movie A, the device sends viewing information (viewing date and time, movie title, rating, review content) to the server. The server collects the data and stores it in a database. Later, when User A watches a new action movie B, the server checks the user profile and classifies them as an "action movie lover." It uses a generative AI model to generate a review of Action Movie B and sends it to the device. The device displays the review on Movie B's page.

[0429] Example 2: Product Review

[0430] User X purchases gadget Z and leaves a rating and review. This information is sent from the device to the server and stored in a database. Next, when User X browses for a new gadget Y, the server references the user profile and generates a personalized review using a generative AI model. The generated review is then displayed on User X's device.

[0431] This system allows users to easily obtain reviews that match their preferences, making information gathering and content selection more efficient. It also enables businesses to conduct marketing based on user preferences, enabling more effective advertising activities.

[0432] The processing flow will be explained below.

[0433] Step 1:

[0434] Users use e-commerce sites and content distribution services to browse products, purchase them, and post reviews.

[0435] The device collects the user's browsing history, purchase information, and review information and sends it to the server.

[0436] The server stores the received data in a database.

[0437] Step 2:

[0438] The server sends a questionnaire to some users to collect information about their personalities and preferences.

[0439] The server analyzes the survey results and classifies users into personality and preference types.

[0440] Based on the survey, people are classified into categories such as "action movie lovers" and "comedy movie lovers."

[0441] Step 3:

[0442] The server extracts the classified user reviews and rating information as training data.

[0443] This training data is used to train generative AI models specialized for each personality and preference type.

[0444] For example, we will use user data of "action movie lovers" to train an AI model for generating action movie reviews.

[0445] Step 4:

[0446] The server analyzes all users' browsing history, purchase history, and review information.

[0447] Based on the analysis results, all users are classified into personality and preference types.

[0448] A user profile is generated and stored in a database.

[0449] Step 5:

[0450] When a user browses for new products or content, the device sends a request to the server.

[0451] The server references the user profile and checks the user's personality and preferences.

[0452] Select the appropriate generative AI model to generate personalized reviews.

[0453] Step 6:

[0454] The server sends the generated reviews to the device.

[0455] The terminal displays the received reviews to the user.

[0456] For example, when a user browses to a page about new action movie B, they will see reviews created using a generative AI model specialized for action movies.

[0457] Specific examples

[0458] Example 1: Movie reviews

[0459] 1. User A watches Movie A and posts a review.

[0460] 2. The device collects viewing information and sends it to the server.

[0461] 3. The server saves the data to a database.

[0462] 4. Using the survey results of User A, classify him as an "action movie lover."

[0463] 5. The server trains the generative AI model using User A's data.

[0464] 6. When user A wants to watch a new action movie B, the device sends a request.

[0465] 7. The server checks User A's profile and generates a personalized review.

[0466] 8. The generated review is displayed on User A's device.

[0467] Example 2: Product review

[0468] 1. User X buys gadget Z and posts a review.

[0469] 2. The device collects purchase information and sends it to the server.

[0470] 3. The server saves the data to a database.

[0471] 4. Using the survey results of User X, classify him / her as a "gadget enthusiast type."

[0472] 5. The server trains the generative AI model using User X's data.

[0473] 6. When user X views new gadget Y, the device sends a request.

[0474] 7. The server checks User X's profile and generates a personalized review.

[0475] 8. The generated review is displayed on User X's device.

[0476] Example 1

[0477] 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."

[0478] Many current e-commerce sites and content distribution services only provide users with general reviews and ratings, making it difficult to provide customized reviews that match the personalities and preferences of individual users. Furthermore, users lack the information they need to select the content and products that best suit them, preventing them from making efficient selections. To address these issues, a system is needed that provides reviews based on the individual preferences of users.

[0479] 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.

[0480] In this invention, the server includes means for collecting user operation history, purchase history, and evaluation information, means for classifying user personality and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into personality and preference types and generating user information, means for selecting a corresponding generative AI model based on the user information and generating customized reviews, and means for displaying the generated reviews on a user device. This allows users to easily obtain reviews that match their preferences, enabling them to efficiently collect information and select content.

[0481] "User operation history" refers to the record of operations performed by a user on an e-commerce site or content distribution service, and specifically includes the products and content viewed, the links clicked, and the user's behavior on the site.

[0482] "Purchase history" refers to a record of products and services that a user actually purchased on an e-commerce site, etc., and includes information such as the purchase date and time, product name, quantity, and price.

[0483] "Rating information" refers to records of ratings and reviews made by users on products, services, and content purchased by users, and includes feedback such as rating scores and comments.

[0484] A "generative artificial intelligence model" is an AI model that is generated by training a machine learning algorithm using collected data, and has the ability to predict and generate for specific tasks.

[0485] "User information" refers to the classification of a user's personality and preferences based on collected data, and the profile information based on that classification, including data regarding the user's interests and preferences.

[0486] "Customized reviews" refer to personalized reviews for individual users that are generated based on the user's personality and preferences, and unlike general reviews, contain content that reflects the user's specific needs and interests.

[0487] "User equipment" refers to a device used by a user through an Internet connection, and specifically includes smartphones, tablets, PCs, etc.

[0488] "Immediate generation" means that the generation process begins immediately after the user submits the request, without delay, and the results are provided within a short time.

[0489] This invention is a personalized AI review system for efficiently providing content suited to the interests and preferences of individual users on the Internet. This system collects users' operation history, purchase history, and rating information, and uses this information to train a generative AI model to provide optimal reviews to users.

[0490] The server collects operation history, purchase information, and rating information generated when users use e-commerce sites and content distribution services, and stores this information in a database. Specifically, the hardware and software that collects information such as the viewing date and time, movie title, rating, and review content when users watch a movie can stream data in real time using Apache Kafka, allowing data to be collected and stored efficiently.

[0491] The server sends a questionnaire to a subset of users to collect information about their personalities and preferences. Based on the survey results, users are classified into personality and preference types, such as "action movie lover" or "comedy movie lover." This classified data is used as training data. Data analysis is performed using data science languages ​​such as Python and R.

[0492] The server trains a generative AI model based on the classified training data. For example, a generative AI model specialized for action movies is created using user data for "action movie lovers." This training is performed using machine learning frameworks such as TensorFlow and PyTorch. The trained model is saved on the server and used for subsequent processing.

[0493] The server analyzes all user data and classifies each user into personality and preference types. This is done using an analytical algorithm based on the user's purchase history, operation history, and rating information. The classified user information is saved in a database as a profile.

[0494] When a user browses for new content or products, the device sends a request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses an appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0495] As a concrete example, when a user watches action movie B, the device sends viewing information (viewing date and time, movie title, rating, review content) to the server. The server collects the data and stores it in a database. Later, when the user views a new action movie C, the server checks the user profile and selects an appropriate generative AI model. The generated review is then displayed on the user's device.

[0496] Example prompt sentence:

[0497] 1. "Generate a personalized review of action movie B that user A is watching."

[0498] 2. "Generate a review for gadget Y by user X."

[0499] 3. "For users who like comedy movies, please write a review of a new comedy movie, C."

[0500] This system allows users to easily obtain reviews that match their preferences, enabling them to efficiently gather information and select content. It also enables businesses to conduct marketing based on user preferences, enabling more effective advertising activities.

[0501] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0502] Step 1:

[0503] The server collects operation history, purchase information, and rating information generated when a user uses an e-commerce site or content distribution service. Specifically, the device collects data such as the user's viewing date and time, movie title, rating, and review content, and sends this data to the server in real time. The server stores the received data in a database.

[0504] Input: User operation history, purchase information, rating information

[0505] Data processing / calculation: Real-time streaming, saving to database

[0506] Output: Saved data

[0507] Step 2:

[0508] The server sends a questionnaire to some users to collect information about their personalities and preferences. The users answer the questionnaire and send the results to the server. The server analyzes the questionnaire results and classifies users into categories such as "action movie lover" or "comedy movie lover."

[0509] Input: Survey response

[0510] Data processing / calculation: Analysis of survey results, classification of users

[0511] Output: Classified user data

[0512] Step 3:

[0513] The server trains a generative AI model based on the classified user data. For example, it uses user data for "action movie lovers" to create an AI model specialized for action movies. Specifically, it uses the collected data as training data using TensorFlow and PyTorch to train the generative AI model.

[0514] Input: Classified user data

[0515] Data processing / computation: training generative AI models

[0516] Output: A trained generative AI model

[0517] Step 4:

[0518] The server analyzes all user data and classifies each user into personality and preference types. This is done using an analytical algorithm based on the user's purchase history, operation history, and rating information. The classified user information is saved in a database as a profile.

[0519] Input: Purchase history, operation history, and rating information of all users

[0520] Data processing / calculation: Data analysis, user classification

[0521] Output: User profile

[0522] Step 5:

[0523] When a user browses for new content or products, the device sends a request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses an appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0524] Input: User request, user profile

[0525] Data processing / calculation: Profile reference, review generation using AI model

[0526] Output: Generated reviews

[0527] In this way, personalized reviews based on the user's preferences can be efficiently provided.

[0528] (Application example 1)

[0529] 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."

[0530] In today's Internet environment, users face the challenge of finding the content and products that best suit them from the vast amount of information available. Furthermore, current review systems often provide general opinions, and rarely provide information specific to individual users' preferences and characteristics. This means that users spend time and effort sifting through information to select the content and products that best suit them.

[0531] 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.

[0532] In this invention, the server includes means for collecting user browsing history, purchase history, and rating information, means for classifying user characteristics and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into characteristics and preference types and generating user profiles, means for selecting corresponding generative AI models based on the user profiles and generating personalized reviews, means for displaying the generated reviews on an information processing device, and means implemented as a smartphone application for providing optimal reviews for users in real time based on their viewing history and purchase history. This allows users to quickly obtain review information that matches their preferences and characteristics, making content and product selection efficient and effective.

[0533] A "user's browsing history" is a record of which web pages and content a user has viewed on the Internet.

[0534] "Purchase history" is a record of what products a user has purchased on the Internet.

[0535] "Rating information" refers to information about ratings and reviews of content or products that a user has viewed or purchased.

[0536] "Characteristics / preference type" refers to a type classified based on the user's preferences and personality traits.

[0537] A "user profile" is a detailed profile of personal information created based on data such as a user's characteristics, preferences, browsing history, and purchase history.

[0538] A "generative AI model" is an artificial intelligence model that uses machine learning to learn from a specific dataset and perform a specific task.

[0539] A "personalized review" is a customized review generated based on the individual characteristics and preferences of a user.

[0540] An "information processing device" is an electronic device that has the function of inputting, processing, and outputting data, and here it mainly refers to a smartphone.

[0541] A "smartphone application" is a software program that runs on a smartphone.

[0542] This invention is a system that collects users' browsing history, purchase history, and rating information, and generates personalized reviews based on the users' characteristics and preferences. This system is composed of a server, an information processing device (such as a smartphone), a generation AI model, etc. Specific embodiments are described below.

[0543] Hardware and software used

[0544] Server: Cloud server (e.g. Amazon Web Services EC2)

[0545] Database: Cloud database (e.g. Amazon RDS)

[0546] Generative AI models: models for natural language generation (e.g., OpenAI GPT)

[0547] Data processing device: Smartphone (e.g., Android or iOS app)

[0548] Data processing and calculation

[0549] Data collection and storage

[0550] The server collects browsing history, purchase history, and rating information generated when a user uses an information processing device (smartphone), and stores this data in a cloud database. For example, when a user watches a movie, the movie title, viewing date and time, rating, and review content are stored. This data serves as the basis for analyzing user preferences.

[0551] User Classification and Data Classification

[0552] Based on the collected data, the server categorizes users into characteristics and preferences. Specifically, based on their browsing and purchase history, users may be classified as those who like action movies or comedy movies. This categorization is stored in a cloud database and used as training data for the generative AI model.

[0553] Training an AI model

[0554] The server uses the classified data to train a generative AI model. For example, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. In this process, generative AI models such as OpenAI GPT are used.

[0555] Generate personalized reviews

[0556] When a user browses new content on an information processing device (smartphone), the server references the user profile and uses the corresponding generative AI model to generate a personalized review, which is generated in real time and displayed on the information processing device.

[0557] Specific examples

[0558] Example 1: Personalized movie reviews

[0559] When a user browses for a new movie using their smartphone, the server checks the user's profile and, if the user is classified as an "action movie lover," uses the generative AI model to generate a personalized review of that movie for action movie lovers. The generated review is displayed on the smartphone, allowing the user to use it as a reference when choosing a movie.

[0560] Prompt Sentence Examples

[0561] A user loved the action movie and gave the following review: It was packed with great action scenes and I really enjoyed it!

[0562] Example 2: Personalizing product reviews

[0563] When a user browses for a new gadget, the server checks the user's profile and uses a generative AI model to generate a personalized review of the gadget, which is then displayed on the user's smartphone to help inform their purchasing decision.

[0564] In this way, the invention provides personalized reviews based on the user's preferences, allowing the user to efficiently select the content and products that are best suited to them, thereby significantly improving the user experience.

[0565] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0566] Step 1:

[0567] Data collection

[0568] When a user browses or purchases using a device (smartphone), that data is collected. Specifically, when a user watches a movie, information such as the viewing date and time, movie title, rating, and review content is sent from the device to the server. The input data is viewing history, purchase history, and rating information, and the server stores this data in a cloud database. The output is the stored user data.

[0569] Step 2:

[0570] User Classification

[0571] The server classifies users into characteristics and preference types based on the collected data. The collected data (input) includes the user's browsing history and purchase history. The server analyzes this data to determine the genre of content the user prefers. For example, a user who watches a lot of action movies would be classified as an "action movie lover." This classification information (output) is stored in a cloud database.

[0572] Step 3:

[0573] Training generative AI models

[0574] The server uses the classified data (input) to train the generative AI model. Specifically, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. The generative AI model used in this process is OpenAI GPT, for example. Once training is complete, multiple generative AI models (output) according to their characteristics are obtained.

[0575] Step 4:

[0576] Creating a user profile

[0577] The server analyzes the data of all users, classifies each user into a characteristic and preference type, and generates a user profile. The input is the user's browsing history, purchase history, and rating information. The server analyzes this data and determines which preference type each user belongs to. Based on this determination, an individual user profile (output) is generated and stored in a cloud database.

[0578] Step 5:

[0579] Generate personalized reviews

[0580] When a user browses new content, a request is sent from the device to the server. The server references the user profile (input) and generates a personalized review using the corresponding generative AI model. Specifically, the server checks the user's preference type and generates a review using prompts appropriate for that type. The generated review (output) is sent to the device and displayed to the user.

[0581] Step 6:

[0582] View Reviews

[0583] The terminal displays the personalized reviews (input) received from the server to the user, allowing the user to select new content or products based on reviews that match their preferences. The output is the displayed reviews.

[0584] Specific examples

[0585] Example prompt sentence:

[0586] A user loved the action movie and gave the following review: It was packed with great action scenes and I really enjoyed it!

[0587] 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.

[0588] This invention combines a personalized AI review system that provides optimal content and product information based on a user's interests and preferences with an emotion engine that recognizes the user's emotions. This system can provide even more personalized reviews based on the user's emotional state at any given time.

[0589] Program processing flow

[0590] Step 1: Collect user data

[0591] The server collects browsing history, purchase information, and review information generated when a user uses an e-commerce site or content distribution service, and stores this information in a database. For example, when a user watches a movie, the server collects the viewing date and time, movie title, rating, and review content.

[0592] Step 2: Classify and train the sampled data

[0593] The server sends a questionnaire to a certain number of users to collect information about their personalities and preferences. The server analyzes the survey results and classifies users into personality and preference types. For example, it classifies users into "action movie lovers" and "comedy movie lovers." Based on this, it generates training data and trains the generative AI model.

[0594] Step 3: Training the generative AI model

[0595] The server trains generative AI models for each personality and preference type based on the classified training data. For example, a generative AI model specialized for action movies is created using user data for "action movie lovers." Individual generative AI models are similarly trained for other types.

[0596] Step 4: Classify all users into types

[0597] The server analyzes all user data, including purchase history, browsing history, and review information, and classifies each user into a personality and preference type. Based on the analysis results, a user profile is generated and stored in a database.

[0598] Step 5: Implementing the Emotion Engine

[0599] The device is equipped with an emotion engine that recognizes the user's emotions in real time. The emotion engine analyzes facial expressions and speech while the user is viewing content to recognize the user's emotional state.

[0600] Step 6: Generate a personalized review

[0601] When a user browses new products or content, the device sends a request and emotional information to the server, which then references the user profile and emotional information to select a corresponding generative AI model to generate a personalized review.

[0602] Step 7: View personalized reviews

[0603] The server sends the generated review to the device, which then displays the received review to the user. For example, when a user browses a page about a new action movie B, if the emotion engine recognizes that the user is excited, it generates and displays a review that focuses on the action scenes.

[0604] Specific examples

[0605] Example 1: Movie reviews

[0606] User A watches Movie A and posts a review. The device collects viewing information and sends it to the server. The server saves the data in a database. User A's emotional information is then analyzed using an emotion engine. The server creates an optimal review based on the user's personality, preference type, and emotional information, and sends it to the device. When User A watches Movie B, the emotion engine recognizes that User A is excited, and generates and displays a review that suits that situation.

[0607] Example 2: Product review

[0608] User X purchases gadget Z and posts a review. The device collects purchase information and sends it to the server. The server saves the data in a database. The emotion engine then analyzes User X's emotional information. The server creates an optimal review based on the user's personality, preference type, and emotional information, and sends it to the device. When User X browses new gadget Y, the emotion engine recognizes that User X is surprised, and generates and displays a review that reflects that emotion.

[0609] This system allows users to obtain the most appropriate reviews based on their preferences and current emotional state. This makes information gathering and content selection more efficient and personalized. It also enables businesses to conduct more detailed marketing based on users' emotions, which is expected to produce even greater results.

[0610] The processing flow will be explained below.

[0611] Step 1:

[0612] Users use e-commerce sites and content distribution services to browse, purchase, and post reviews of products.

[0613] The device collects the user's browsing history, purchase information, and review information and sends it to the server.

[0614] The server stores the received data in a database.

[0615] Step 2:

[0616] The server sends a questionnaire to some users to collect information about their personalities and preferences.

[0617] The server analyzes the survey results and classifies users into personality and preference types.

[0618] For example, classifications such as "action movie lover type" and "comedy movie lover type" are made.

[0619] Step 3:

[0620] The server extracts the classified user reviews and rating information as training data.

[0621] This training data is used to train generative AI models specialized for each personality and preference type.

[0622] For example, we will use user data of "action movie lovers" to train an AI model for generating action movie reviews.

[0623] Step 4:

[0624] The server analyzes all users' browsing history, purchase history, and review information.

[0625] Based on the analysis results, all users are classified into personality and preference types.

[0626] A user profile is generated and stored in a database.

[0627] Step 5:

[0628] The device is equipped with an emotion engine that allows it to recognize the user's emotions in real time.

[0629] The emotion engine analyzes the user's facial expressions and speech while viewing content, and recognizes the user's emotional state.

[0630] Step 6:

[0631] When a user browses for new products or content, the device sends the request and emotion information to the server.

[0632] The server refers to the user profile and sentiment information, selects a corresponding generative AI model, and generates a personalized review.

[0633] Step 7:

[0634] The server sends the generated reviews to the device.

[0635] The terminal displays the received reviews to the user.

[0636] For example, when a user browses a page about a new action movie B, if the emotion engine recognizes that the user is excited, it will generate a review that focuses on the action scenes and display it on the device.

[0637] Specific examples

[0638] Example 1: Movie reviews

[0639] 1. User A watches Movie A and posts a review.

[0640] 2. The device collects viewing information and sends it to the server.

[0641] 3. The server saves the data to a database.

[0642] 4. Using the survey results of User A, classify him as an "action movie lover."

[0643] 5. The server trains the generative AI model using User A's data.

[0644] 6. When user A watches movie B, the emotion engine analyzes user A's emotional information in real time.

[0645] 7. The device sends the real-time emotion information to the server.

[0646] 8. The server references user A's profile and sentiment information to generate the most appropriate review.

[0647] 9. The review generated is sent back to the device and displayed on the page for Movie B that User A is viewing.

[0648] Example 2: Product review

[0649] 1. User X buys gadget Z and posts a review.

[0650] 2. The device collects purchase information and sends it to the server.

[0651] 3. The server saves the data to a database.

[0652] 4. Use the survey results of User X to classify him / her into the "gadget enthusiast type."

[0653] 5. The server trains the generative AI model using User X's data.

[0654] 6. When user X browses new gadget Y, the emotion engine analyzes user X's emotion information in real time.

[0655] 7. The device sends the real-time emotion information to the server.

[0656] 8. The server references user X's profile and sentiment information to generate the most appropriate review.

[0657] 9. The device displays the generated review to User X, allowing User X to view a personalized review of Gadget Y in a state of amazement.

[0658] This system allows users to easily find the most suitable reviews based on their preferences and current emotional state, making information gathering and content selection more efficient and personalized. It also enables businesses to conduct more effective marketing based on user emotions, leading to greater success.

[0659] Example 2

[0660] 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."

[0661] Conventional review systems can provide personalized reviews based on a user's personality and preferences, but they cannot provide reviews that take into account the user's current emotional state. This makes it difficult for users to obtain the most appropriate information based on their emotions at any given time. It also makes it difficult for companies to conduct effective marketing based on users' emotions.

[0662] 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.

[0663] In this invention, the server includes means for collecting users' browsing history, purchase history, and review information; means for classifying users' personality and preference types based on the collected data; means for training a generative AI model using the classified data; means for classifying all users into personality and preference types and generating user profiles; means for the user terminal to recognize the user's emotional state in real time using a camera or microphone; means for selecting a corresponding generative AI model based on the user profile and emotional information and generating personalized reviews; and means for displaying the generated personalized reviews on the user terminal. This allows users to obtain optimal reviews based on their current emotional state as well as their own preferences. Companies can also conduct effective marketing based on users' emotions.

[0664] A "user's browsing history" is a record of the pages and content a user accesses on a website or application.

[0665] "Purchase history" is a list of products that a user has purchased in the past and related information.

[0666] "Review information" is data including evaluations and opinions given by users regarding products and services.

[0667] "Personality / Preference Type" is a classification based on a user's personal personality or preferences for specific content or products.

[0668] A "generative AI model" is an artificial intelligence model that generates personalized reviews based on a user's personality, preferences, and emotional state.

[0669] A "user profile" is a data set that integrates a user's personality, preferences, browsing history, purchase history, review information, etc.

[0670] "Means for recognizing emotional states in real time" refers to technology that detects the user's emotions at any given time from their facial expressions and voice.

[0671] "Personalized reviews" are the presentation of ratings and opinions that are individually optimized based on the user's personality, preference type, and emotional state.

[0672] A "user terminal" is a device operated by a user, and includes a personal computer, a smartphone, a tablet, etc.

[0673] This invention combines a personalized AI review system that provides optimal content and product information based on a user's interests and preferences with an emotion engine that recognizes the user's emotions. This system can provide even more personalized reviews based on the user's emotional state at any given time.

[0674] The server collects browsing history, purchase history, and review information generated when users use e-commerce sites and content distribution services. This data is periodically retrieved via API and inserted into a database such as MySQL or MongoDB.

[0675] The server then sends users a questionnaire to collect information about their personalities and preferences. The results of this questionnaire are analyzed using natural language processing (NLP) techniques, and algorithms such as K-means clustering are used to classify users into categories such as "action movie fans" and "comedy movie fans." This classification data is used to generate a training dataset and train a generative AI model. The generative AI model, trained using Python and libraries such as TensorFlow and PyTorch, is specialized for each preference type.

[0676] The server analyzes newly collected user data and classifies all users into preference types using a pre-trained generative AI model. Based on this result, a user profile is generated and stored in a database. To ensure real-time performance, batch processing and real-time data streaming technologies are used.

[0677] The device is equipped with an emotion engine that recognizes the user's emotions in real time. It collects facial expressions and voice data through the camera and microphone, and analyzes emotions using the OpenCV library and the Google Cloud Speech-to-Text API. The analysis results are sent to a server and integrated with the user's profile.

[0678] When a user browses for new products or content, their device sends their request and emotional information to the server. The server then selects a corresponding AI model based on the received user profile and emotional information to generate a personalized review. It uses natural language generation (NLG) technology to generate review text that reflects the user's emotions.

[0679] As a concrete example, consider a scenario in which User A watches Movie A and posts a review. The device collects viewing information and sends it to the server. The server stores the data in a database and analyzes User A's emotional information using an emotion engine. The server creates an optimal review based on the personality, preference type, and emotional information, and sends it to the device. When User A watches Movie B, the emotion engine recognizes that User A is excited, and generates and displays a review that suits that situation.

[0680] An example prompt is:

[0681] "The powerful scenes in this action movie will get you even more excited."

[0682] This system allows users to obtain the most appropriate reviews based on their preferences and current emotional state, making information gathering and content selection more efficient and personalized. It also enables companies to conduct effective marketing based on users' emotions.

[0683] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0684] Step 1:

[0685] The server collects user data. Specifically, it obtains user browsing history, purchase history, and review information via API. The input is user activity log data. The server stores the collected data in a MySQL or MongoDB database. The output is organized user data.

[0686] Step 2:

[0687] The server sends a questionnaire to a certain number of users to collect information on their personalities and preferences. The input is the questionnaire response data from the users. The server analyzes this using natural language processing (NLP) technology and classifies users into personality and preference types using algorithms such as K-means clustering. The output is data classifying the users' personalities and preference types.

[0688] Step 3:

[0689] The server uses the classified data to train a generative AI model. The input is training data categorized into personality and preference types. A generative AI model for each type is trained using Python and the TensorFlow or PyTorch library. The output is a trained generative AI model.

[0690] Step 4:

[0691] The server analyzes all user data. Specifically, it inputs purchase history, browsing history, and review information into a previously trained AI model to classify each user into a personality and preference type. The input is newly collected user data, and the output is an updated user profile.

[0692] Step 5:

[0693] The device uses an emotion engine to recognize the user's emotions in real time. Specifically, it analyzes facial and voice data acquired from the camera and microphone using OpenCV and the Google Cloud Speech-to-Text API to detect the user's emotional state. The input is the user's facial and voice data, and the output is the analyzed emotional information.

[0694] Step 6:

[0695] When a user browses for new products or content, the device sends the request and emotional information to the server. The server references the user profile and emotional information, selects a corresponding generative AI model, and generates a personalized review. The input is the request data, emotional information, and user profile, and the output is the generated review.

[0696] Step 7:

[0697] The server generates personalized reviews and sends them to the device. The device displays the reviews it receives on the user interface (UI). Specifically, the reviews are inserted into web pages or app UIs designed with HTML and CSS, and the display is updated in real time. The input is the reviews sent from the server, and the output is the reviews displayed to the user.

[0698] (Application example 2)

[0699] 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."

[0700] Conventional personalized review systems could provide personalization based on the user's basic personality and preferences, but they could not consider the user's emotional state at any given time, making it difficult to provide more detailed personalization. Furthermore, they could not generate reviews in real time that reflected the user's emotional information, making it impossible to provide feedback that was in line with the user's current emotions.

[0701] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting users' browsing history, purchase history, and review information, means for classifying users' personality and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into personality and preference types and generating user profiles, means for selecting corresponding generative AI models based on the user profiles and generating personalized reviews, emotion engine means for recognizing the user's emotional state in real time, and means for displaying personalized reviews generated based on requests including the user's emotional information on the user terminal. This makes it possible to provide personalized reviews in real time that are tailored to the user's basic personality and preferences as well as their emotional state at any given time.

[0702] "User browsing history" refers to historical information about pages and content that a user views when using a website or application.

[0703] "Purchase history" is historical information about products purchased by a user online or offline.

[0704] "Review information" refers to information about ratings and comments posted by users about products or content that they have viewed or purchased.

[0705] "Means for classifying user personality and preference types" refers to a method or system for classifying a user's personality and preferences into specific types based on collected user data.

[0706] A "means for training a generative AI model" is a method or system that uses classified user data to train a generative AI model that responds to a specific purpose.

[0707] A "user profile" is individual profile information generated by integrating a user's personality, preferences, browsing history, purchase history, and the like.

[0708] The "emotion engine" is a technology that analyzes data such as the user's facial expressions and voice, and recognizes their emotional state at any given time in real time.

[0709] A "personalized review" is a customized rating or comment on a specific product or content that takes into account a user's individual personality, preferences, and feelings.

[0710] A "user terminal" is a device that a user directly uses, and refers to hardware such as a smartphone, tablet, or computer.

[0711] MODE FOR CARRYING OUT THE INVENTION

[0712] System Overview

[0713] This invention is a system that generates personalized reviews based on a user's browsing history, purchase history, and review information. Its special feature is that it recognizes the user's current emotional state and provides reviews that correspond to that emotion at that time. This system consists of a user terminal equipped with an emotion engine and a server that analyzes and processes the data.

[0714] Hardware and software used

[0715] Hardware

[0716] User devices: smartphones, tablets, computers, etc.

[0717] Server: High-performance computer, cloud infrastructure

[0718] software

[0719] EmotionEngine: Software that recognizes a user's emotional state by analyzing their facial expressions and voice

[0720] AIReviewGenerator: A generative AI model that generates personalized reviews based on user profiles and sentiment information

[0721] Django or Flask: a framework for server-side data management and API serving.

[0722] Processing flow and specific examples

[0723] Step 1: Data collection

[0724] The server collects user data such as viewing history, purchase history, and review information from the user's device and stores it in a database. For example, when a user watches a particular movie, the viewing date and time, movie title, rating, and review content are recorded.

[0725] Step 2: Classify users

[0726] The server classifies the user's personality and preferences based on the collected data. This identifies user types such as "action movie lover" or "comedy movie lover." This classification is used to train a generative AI model.

[0727] Step 3: Introducing the Emotion Engine

[0728] The user device is equipped with an EmotionEngine that analyzes the user's facial expressions and voice to recognize their current emotional state. For example, it can analyze the user's reactions in real time while watching a movie to identify their emotional state.

[0729] Step 4: Generate a personalized review

[0730] When a user browses new products or content, the device sends the corresponding request and emotional information to the server, which then selects a corresponding generative AI model based on the user profile and emotional information to generate a personalized review for the specific content.

[0731] Step 5: View reviews

[0732] The generated review is sent to the user's device and displayed to the user. For example, when a user browses a page about a new action movie, if the emotion engine recognizes that the user is excited, a review specific to the action scenes of that movie will be generated and displayed to the user.

[0733] Specific examples

[0734] If the emotion engine determines that the user is excited after watching movie A, it will send the following prompt to the generative AI model.

[0735] Prompt statement:

[0736] User ID 123's emotional state is excited. Generate a personalized review for content he / she is interested in based on the following information: User Data: { "Action Movie Lover", "Recently Watched Movies": "Movie A", "Rating": "High"} Content ID: Movie B

[0737] In this way, a personalized review optimized for the user's current emotional state can be provided. The above-described system and process allow users to obtain reviews that are in line with their emotions, providing a more satisfying content viewing experience.

[0738] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0739] Step 1:

[0740] The server collects user data such as viewing history, purchase history, and review information from the user's device. The collected data is stored in a database. This data includes, for example, the date and time when the user watched a particular movie, the movie title, rating, and review content.

[0741] Input: Browsing history, purchase history, and review information from user devices

[0742] Output: User data stored in the database

[0743] Specific operation: When the server receives an HTTP request, it structures the data and stores it in a database.

[0744] Step 2:

[0745] The server analyzes the collected user data and classifies the user's personality and preferences. For example, it may classify them into types such as "likes action movies" or "likes comedy movies." Training data is generated based on the classification results, and the generative AI model is trained.

[0746] Input: User data stored in the database

[0747] Output: Classified user data, training data

[0748] How it works: The server uses machine learning algorithms to analyze user data and classify it into user types.

[0749] Step 3:

[0750] The server analyzes all user data and classifies each user into personality and preference types, based on which a user profile is created and stored in a database.

[0751] Input: All users' data

[0752] Output: User profile

[0753] Specific operation: The server uses the data classified in the previous step to generate a profile for each user based on their personality and preference types.

[0754] Step 4:

[0755] The user device uses the EmotionEngine to analyze the user's facial expressions and voice in real time to recognize their emotional state. For example, while watching a movie, the device can identify emotional states such as "excitement" or "joy" from the user's facial expressions and voice.

[0756] Input: User's facial expression data, voice data

[0757] Output: Perceived emotional state

[0758] Specific operation: The EmotionEngine in the device analyzes visual and audio data and outputs the emotional state.

[0759] Step 5:

[0760] When a user browses for new products or content, the terminal sends requests and emotion information to the server.

[0761] Input: User request, perceived emotional state

[0762] Output: Request and emotion information sent to the server

[0763] Specific operation: The device sends data to the server as an HTTP request.

[0764] Step 6:

[0765] The server refers to the user profile and emotion information, selects a corresponding generative AI model, and generates a personalized review.

[0766] Input: User profile, perceived emotional state

[0767] Output: The generated personalized review

[0768] Specific operation: The server generates a prompt using AI Review Generator, inputs it into the generative AI model, and generates a review.

[0769] Step 7:

[0770] The generated review is sent to the user's terminal and displayed to the user. For example, if the user is recognized as excited when browsing a page about a new action movie, a review specific to the action scenes of that movie is generated and displayed.

[0771] Input: Generated personalized review

[0772] Output: The review displayed on the user's device

[0773] Specific behavior: The server sends the generated review to the device as an HTTP response, and the device displays it.

[0774] 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.

[0775] 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.

[0776] 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.

[0777] [Third embodiment]

[0778] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0779] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0780] 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).

[0781] 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.

[0782] 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.

[0783] 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).

[0784] 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.

[0785] 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.

[0786] 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.

[0787] 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.

[0788] 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.

[0789] 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."

[0790] This invention is a personalized AI review system that efficiently provides content suited to the interests and preferences of individual users on the Internet. This system collects users' browsing history, purchase history, and review information, and uses this information to train a generative AI model to provide optimal reviews to users.

[0791] Program processing flow

[0792] Step 1: Collect user data

[0793] The server collects browsing history, purchase information, and review information generated when users use e-commerce sites and content distribution services, and stores them in a database. Specifically, when users watch a movie, the server collects the viewing date and time, movie title, rating, and review content.

[0794] Step 2: Classify and train the sampled data

[0795] The server sends a questionnaire to some users to collect information about their personalities and preferences. Based on the collected questionnaire results, the server classifies users into personality and preference types, such as "action movie lover" and "comedy movie lover." This classified data is used as training data.

[0796] Step 3: Training the generative AI model

[0797] The server trains a generative AI model based on the classified training data. For example, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. Similarly, separate generative AI models are trained for other types.

[0798] Step 4: Classify all users into types

[0799] The server analyzes all user data, including purchase history, browsing history, and review information, and classifies each user into a personality type and preference type, laying the foundation for providing personalized reviews using a generative AI model that is optimal for each user.

[0800] Step 5: Generate a personalized review

[0801] When a user browses for new content or products, the device sends the request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses the appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0802] Specific examples

[0803] Example 1: Movie reviews

[0804] When User A watches Movie A, the device sends viewing information (viewing date and time, movie title, rating, review content) to the server. The server collects the data and stores it in a database. Later, when User A watches a new action movie B, the server checks the user profile and classifies them as an "action movie lover." It uses a generative AI model to generate a review of Action Movie B and sends it to the device. The device displays the review on Movie B's page.

[0805] Example 2: Product Review

[0806] User X purchases gadget Z and leaves a rating and review. This information is sent from the device to the server and stored in a database. Next, when User X browses for a new gadget Y, the server references the user profile and generates a personalized review using a generative AI model. The generated review is then displayed on User X's device.

[0807] This system allows users to easily obtain reviews that match their preferences, making information gathering and content selection more efficient. It also enables businesses to conduct marketing based on user preferences, enabling more effective advertising activities.

[0808] The processing flow will be explained below.

[0809] Step 1:

[0810] Users use e-commerce sites and content distribution services to browse products, purchase them, and post reviews.

[0811] The device collects the user's browsing history, purchase information, and review information and sends it to the server.

[0812] The server stores the received data in a database.

[0813] Step 2:

[0814] The server sends a questionnaire to some users to collect information about their personalities and preferences.

[0815] The server analyzes the survey results and classifies users into personality and preference types.

[0816] Based on the survey, people are classified into categories such as "action movie lovers" and "comedy movie lovers."

[0817] Step 3:

[0818] The server extracts the classified user reviews and rating information as training data.

[0819] This training data is used to train generative AI models specialized for each personality and preference type.

[0820] For example, we will use user data of "action movie lovers" to train an AI model for generating action movie reviews.

[0821] Step 4:

[0822] The server analyzes all users' browsing history, purchase history, and review information.

[0823] Based on the analysis results, all users are classified into personality and preference types.

[0824] A user profile is generated and stored in a database.

[0825] Step 5:

[0826] When a user browses for new products or content, the device sends a request to the server.

[0827] The server references the user profile and checks the user's personality and preferences.

[0828] Select the appropriate generative AI model to generate personalized reviews.

[0829] Step 6:

[0830] The server sends the generated reviews to the device.

[0831] The terminal displays the received reviews to the user.

[0832] For example, when a user browses to a page about new action movie B, they will see reviews created using a generative AI model specialized for action movies.

[0833] Specific examples

[0834] Example 1: Movie reviews

[0835] 1. User A watches Movie A and posts a review.

[0836] 2. The device collects viewing information and sends it to the server.

[0837] 3. The server saves the data to a database.

[0838] 4. Using the survey results of User A, classify him as an "action movie lover."

[0839] 5. The server trains the generative AI model using User A's data.

[0840] 6. When user A wants to watch a new action movie B, the device sends a request.

[0841] 7. The server checks User A's profile and generates a personalized review.

[0842] 8. The generated review is displayed on User A's device.

[0843] Example 2: Product review

[0844] 1. User X buys gadget Z and posts a review.

[0845] 2. The device collects purchase information and sends it to the server.

[0846] 3. The server saves the data to a database.

[0847] 4. Using the survey results of User X, classify him / her as a "gadget enthusiast type."

[0848] 5. The server trains the generative AI model using User X's data.

[0849] 6. When user X views new gadget Y, the device sends a request.

[0850] 7. The server checks User X's profile and generates a personalized review.

[0851] 8. The generated review is displayed on User X's device.

[0852] Example 1

[0853] 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."

[0854] Many current e-commerce sites and content distribution services only provide users with general reviews and ratings, making it difficult to provide customized reviews that match the personalities and preferences of individual users. Furthermore, users lack the information they need to select the content and products that best suit them, preventing them from making efficient selections. To address these issues, a system is needed that provides reviews based on the individual preferences of users.

[0855] 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.

[0856] In this invention, the server includes means for collecting user operation history, purchase history, and evaluation information, means for classifying user personality and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into personality and preference types and generating user information, means for selecting a corresponding generative AI model based on the user information and generating customized reviews, and means for displaying the generated reviews on a user device. This allows users to easily obtain reviews that match their preferences, enabling them to efficiently collect information and select content.

[0857] "User operation history" refers to the record of operations performed by a user on an e-commerce site or content distribution service, and specifically includes the products and content viewed, the links clicked, and the user's behavior on the site.

[0858] "Purchase history" refers to a record of products and services that a user actually purchased on an e-commerce site, etc., and includes information such as the purchase date and time, product name, quantity, and price.

[0859] "Rating information" refers to records of ratings and reviews made by users on products, services, and content purchased by users, and includes feedback such as rating scores and comments.

[0860] A "generative artificial intelligence model" is an AI model that is generated by training a machine learning algorithm using collected data, and has the ability to predict and generate for specific tasks.

[0861] "User information" refers to the classification of a user's personality and preferences based on collected data, and the profile information based on that classification, including data regarding the user's interests and preferences.

[0862] "Customized reviews" refer to personalized reviews for individual users that are generated based on the user's personality and preferences, and unlike general reviews, contain content that reflects the user's specific needs and interests.

[0863] "User equipment" refers to a device used by a user through an Internet connection, and specifically includes smartphones, tablets, PCs, etc.

[0864] "Immediate generation" means that the generation process begins immediately after the user submits the request, without delay, and the results are provided within a short time.

[0865] This invention is a personalized AI review system for efficiently providing content suited to the interests and preferences of individual users on the Internet. This system collects users' operation history, purchase history, and rating information, and uses this information to train a generative AI model to provide optimal reviews to users.

[0866] The server collects operation history, purchase information, and rating information generated when users use e-commerce sites and content distribution services, and stores this information in a database. Specifically, the hardware and software that collects information such as the viewing date and time, movie title, rating, and review content when users watch a movie can stream data in real time using Apache Kafka, allowing data to be collected and stored efficiently.

[0867] The server sends a questionnaire to a subset of users to collect information about their personalities and preferences. Based on the survey results, users are classified into personality and preference types, such as "action movie lover" or "comedy movie lover." This classified data is used as training data. Data analysis is performed using data science languages ​​such as Python and R.

[0868] The server trains a generative AI model based on the classified training data. For example, a generative AI model specialized for action movies is created using user data for "action movie lovers." This training is performed using machine learning frameworks such as TensorFlow and PyTorch. The trained model is saved on the server and used for subsequent processing.

[0869] The server analyzes all user data and classifies each user into personality and preference types. This is done using an analytical algorithm based on the user's purchase history, operation history, and rating information. The classified user information is saved in a database as a profile.

[0870] When a user browses for new content or products, the device sends a request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses an appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0871] As a concrete example, when a user watches action movie B, the device sends viewing information (viewing date and time, movie title, rating, review content) to the server. The server collects the data and stores it in a database. Later, when the user views a new action movie C, the server checks the user profile and selects an appropriate generative AI model. The generated review is then displayed on the user's device.

[0872] Example prompt sentence:

[0873] 1. "Generate a personalized review of action movie B that user A is watching."

[0874] 2. "Generate a review for gadget Y by user X."

[0875] 3. "For users who like comedy movies, please write a review of a new comedy movie, C."

[0876] This system allows users to easily obtain reviews that match their preferences, enabling them to efficiently gather information and select content. It also enables businesses to conduct marketing based on user preferences, enabling more effective advertising activities.

[0877] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0878] Step 1:

[0879] The server collects operation history, purchase information, and rating information generated when a user uses an e-commerce site or content distribution service. Specifically, the device collects data such as the user's viewing date and time, movie title, rating, and review content, and sends this data to the server in real time. The server stores the received data in a database.

[0880] Input: User operation history, purchase information, rating information

[0881] Data processing / calculation: Real-time streaming, saving to database

[0882] Output: Saved data

[0883] Step 2:

[0884] The server sends a questionnaire to some users to collect information about their personalities and preferences. The users answer the questionnaire and send the results to the server. The server analyzes the questionnaire results and classifies users into categories such as "action movie lover" or "comedy movie lover."

[0885] Input: Survey response

[0886] Data processing / calculation: Analysis of survey results, classification of users

[0887] Output: Classified user data

[0888] Step 3:

[0889] The server trains a generative AI model based on the classified user data. For example, it uses user data for "action movie lovers" to create an AI model specialized for action movies. Specifically, it uses the collected data as training data using TensorFlow and PyTorch to train the generative AI model.

[0890] Input: Classified user data

[0891] Data processing / computation: training generative AI models

[0892] Output: A trained generative AI model

[0893] Step 4:

[0894] The server analyzes all user data and classifies each user into personality and preference types. This is done using an analytical algorithm based on the user's purchase history, operation history, and rating information. The classified user information is saved in a database as a profile.

[0895] Input: Purchase history, operation history, and rating information of all users

[0896] Data processing / calculation: Data analysis, user classification

[0897] Output: User profile

[0898] Step 5:

[0899] When a user browses for new content or products, the device sends a request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses an appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[0900] Input: User request, user profile

[0901] Data processing / calculation: Profile reference, review generation using AI model

[0902] Output: Generated reviews

[0903] In this way, personalized reviews based on the user's preferences can be efficiently provided.

[0904] (Application example 1)

[0905] 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."

[0906] In today's Internet environment, users face the challenge of finding the content and products that best suit them from the vast amount of information available. Furthermore, current review systems often provide general opinions, and rarely provide information specific to individual users' preferences and characteristics. This means that users spend time and effort sifting through information to select the content and products that best suit them.

[0907] 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.

[0908] In this invention, the server includes means for collecting user browsing history, purchase history, and rating information, means for classifying user characteristics and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into characteristics and preference types and generating user profiles, means for selecting corresponding generative AI models based on the user profiles and generating personalized reviews, means for displaying the generated reviews on an information processing device, and means implemented as a smartphone application for providing optimal reviews for users in real time based on their viewing history and purchase history. This allows users to quickly obtain review information that matches their preferences and characteristics, making content and product selection efficient and effective.

[0909] A "user's browsing history" is a record of which web pages and content a user has viewed on the Internet.

[0910] "Purchase history" is a record of what products a user has purchased on the Internet.

[0911] "Rating information" refers to information about ratings and reviews of content or products that a user has viewed or purchased.

[0912] "Characteristics / preference type" refers to a type classified based on the user's preferences and personality traits.

[0913] A "user profile" is a detailed profile of personal information created based on data such as a user's characteristics, preferences, browsing history, and purchase history.

[0914] A "generative AI model" is an artificial intelligence model that uses machine learning to learn from a specific dataset and perform a specific task.

[0915] A "personalized review" is a customized review generated based on the individual characteristics and preferences of a user.

[0916] An "information processing device" is an electronic device that has the function of inputting, processing, and outputting data, and here it mainly refers to a smartphone.

[0917] A "smartphone application" is a software program that runs on a smartphone.

[0918] This invention is a system that collects users' browsing history, purchase history, and rating information, and generates personalized reviews based on the users' characteristics and preferences. This system is composed of a server, an information processing device (such as a smartphone), a generation AI model, etc. Specific embodiments are described below.

[0919] Hardware and software used

[0920] Server: Cloud server (e.g. Amazon Web Services EC2)

[0921] Database: Cloud database (e.g. Amazon RDS)

[0922] Generative AI models: models for natural language generation (e.g., OpenAI GPT)

[0923] Data processing device: Smartphone (e.g., Android or iOS app)

[0924] Data processing and calculation

[0925] Data collection and storage

[0926] The server collects browsing history, purchase history, and rating information generated when a user uses an information processing device (smartphone), and stores this data in a cloud database. For example, when a user watches a movie, the movie title, viewing date and time, rating, and review content are stored. This data serves as the basis for analyzing user preferences.

[0927] User Classification and Data Classification

[0928] Based on the collected data, the server categorizes users into characteristics and preferences. Specifically, based on their browsing and purchase history, users may be classified as those who like action movies or comedy movies. This categorization is stored in a cloud database and used as training data for the generative AI model.

[0929] Training an AI model

[0930] The server uses the classified data to train a generative AI model. For example, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. In this process, generative AI models such as OpenAI GPT are used.

[0931] Generate personalized reviews

[0932] When a user browses new content on an information processing device (smartphone), the server references the user profile and uses the corresponding generative AI model to generate a personalized review, which is generated in real time and displayed on the information processing device.

[0933] Specific examples

[0934] Example 1: Personalized movie reviews

[0935] When a user browses for a new movie using their smartphone, the server checks the user's profile and, if the user is classified as an "action movie lover," uses the generative AI model to generate a personalized review of that movie for action movie lovers. The generated review is displayed on the smartphone, allowing the user to use it as a reference when choosing a movie.

[0936] Prompt Sentence Examples

[0937] A user loved the action movie and gave the following review: It was packed with great action scenes and I really enjoyed it!

[0938] Example 2: Personalizing product reviews

[0939] When a user browses for a new gadget, the server checks the user's profile and uses a generative AI model to generate a personalized review of the gadget, which is then displayed on the user's smartphone to help inform their purchasing decision.

[0940] In this way, the invention provides personalized reviews based on the user's preferences, allowing the user to efficiently select the content and products that are best suited to them, thereby significantly improving the user experience.

[0941] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0942] Step 1:

[0943] Data collection

[0944] When a user browses or purchases using a device (smartphone), that data is collected. Specifically, when a user watches a movie, information such as the viewing date and time, movie title, rating, and review content is sent from the device to the server. The input data is viewing history, purchase history, and rating information, and the server stores this data in a cloud database. The output is the stored user data.

[0945] Step 2:

[0946] User Classification

[0947] The server classifies users into characteristics and preference types based on the collected data. The collected data (input) includes the user's browsing history and purchase history. The server analyzes this data to determine the genre of content the user prefers. For example, a user who watches a lot of action movies would be classified as an "action movie lover." This classification information (output) is stored in a cloud database.

[0948] Step 3:

[0949] Training generative AI models

[0950] The server uses the classified data (input) to train the generative AI model. Specifically, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. The generative AI model used in this process is OpenAI GPT, for example. Once training is complete, multiple generative AI models (output) according to their characteristics are obtained.

[0951] Step 4:

[0952] Creating a user profile

[0953] The server analyzes the data of all users, classifies each user into a characteristic and preference type, and generates a user profile. The input is the user's browsing history, purchase history, and rating information. The server analyzes this data and determines which preference type each user belongs to. Based on this determination, an individual user profile (output) is generated and stored in a cloud database.

[0954] Step 5:

[0955] Generate personalized reviews

[0956] When a user browses new content, a request is sent from the device to the server. The server references the user profile (input) and generates a personalized review using the corresponding generative AI model. Specifically, the server checks the user's preference type and generates a review using prompts appropriate for that type. The generated review (output) is sent to the device and displayed to the user.

[0957] Step 6:

[0958] View Reviews

[0959] The terminal displays the personalized reviews (input) received from the server to the user, allowing the user to select new content or products based on reviews that match their preferences. The output is the displayed reviews.

[0960] Specific examples

[0961] Example prompt sentence:

[0962] A user loved the action movie and gave the following review: It was packed with great action scenes and I really enjoyed it!

[0963] 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.

[0964] This invention combines a personalized AI review system that provides optimal content and product information based on a user's interests and preferences with an emotion engine that recognizes the user's emotions. This system can provide even more personalized reviews based on the user's emotional state at any given time.

[0965] Program processing flow

[0966] Step 1: Collect user data

[0967] The server collects browsing history, purchase information, and review information generated when a user uses an e-commerce site or content distribution service, and stores this information in a database. For example, when a user watches a movie, the server collects the viewing date and time, movie title, rating, and review content.

[0968] Step 2: Classify and train the sampled data

[0969] The server sends a questionnaire to a certain number of users to collect information about their personalities and preferences. The server analyzes the survey results and classifies users into personality and preference types. For example, it classifies users into "action movie lovers" and "comedy movie lovers." Based on this, it generates training data and trains the generative AI model.

[0970] Step 3: Training the generative AI model

[0971] The server trains generative AI models for each personality and preference type based on the classified training data. For example, a generative AI model specialized for action movies is created using user data for "action movie lovers." Individual generative AI models are similarly trained for other types.

[0972] Step 4: Classify all users into types

[0973] The server analyzes all user data, including purchase history, browsing history, and review information, and classifies each user into a personality and preference type. Based on the analysis results, a user profile is generated and stored in a database.

[0974] Step 5: Implementing the Emotion Engine

[0975] The device is equipped with an emotion engine that recognizes the user's emotions in real time. The emotion engine analyzes facial expressions and speech while the user is viewing content to recognize the user's emotional state.

[0976] Step 6: Generate a personalized review

[0977] When a user browses new products or content, the device sends a request and emotional information to the server, which then references the user profile and emotional information to select a corresponding generative AI model to generate a personalized review.

[0978] Step 7: View personalized reviews

[0979] The server sends the generated review to the device, which then displays the received review to the user. For example, when a user browses a page about a new action movie B, if the emotion engine recognizes that the user is excited, it generates and displays a review that focuses on the action scenes.

[0980] Specific examples

[0981] Example 1: Movie reviews

[0982] User A watches Movie A and posts a review. The device collects viewing information and sends it to the server. The server saves the data in a database. User A's emotional information is then analyzed using an emotion engine. The server creates an optimal review based on the user's personality, preference type, and emotional information, and sends it to the device. When User A watches Movie B, the emotion engine recognizes that User A is excited, and generates and displays a review that suits that situation.

[0983] Example 2: Product review

[0984] User X purchases gadget Z and posts a review. The device collects purchase information and sends it to the server. The server saves the data in a database. The emotion engine then analyzes User X's emotional information. The server creates an optimal review based on the user's personality, preference type, and emotional information, and sends it to the device. When User X browses new gadget Y, the emotion engine recognizes that User X is surprised, and generates and displays a review that reflects that emotion.

[0985] This system allows users to obtain the most appropriate reviews based on their preferences and current emotional state. This makes information gathering and content selection more efficient and personalized. It also enables businesses to conduct more detailed marketing based on users' emotions, which is expected to produce even greater results.

[0986] The processing flow will be explained below.

[0987] Step 1:

[0988] Users use e-commerce sites and content distribution services to browse, purchase, and post reviews of products.

[0989] The device collects the user's browsing history, purchase information, and review information and sends it to the server.

[0990] The server stores the received data in a database.

[0991] Step 2:

[0992] The server sends a questionnaire to some users to collect information about their personalities and preferences.

[0993] The server analyzes the survey results and classifies users into personality and preference types.

[0994] For example, classifications such as "action movie lover type" and "comedy movie lover type" are made.

[0995] Step 3:

[0996] The server extracts the classified user reviews and rating information as training data.

[0997] This training data is used to train generative AI models specialized for each personality and preference type.

[0998] For example, we will use user data of "action movie lovers" to train an AI model for generating action movie reviews.

[0999] Step 4:

[1000] The server analyzes all users' browsing history, purchase history, and review information.

[1001] Based on the analysis results, all users are classified into personality and preference types.

[1002] A user profile is generated and stored in a database.

[1003] Step 5:

[1004] The device is equipped with an emotion engine that allows it to recognize the user's emotions in real time.

[1005] The emotion engine analyzes the user's facial expressions and speech while viewing content, and recognizes the user's emotional state.

[1006] Step 6:

[1007] When a user browses for new products or content, the device sends the request and emotion information to the server.

[1008] The server refers to the user profile and sentiment information, selects a corresponding generative AI model, and generates a personalized review.

[1009] Step 7:

[1010] The server sends the generated reviews to the device.

[1011] The terminal displays the received reviews to the user.

[1012] For example, when a user browses a page about a new action movie B, if the emotion engine recognizes that the user is excited, it will generate a review that focuses on the action scenes and display it on the device.

[1013] Specific examples

[1014] Example 1: Movie reviews

[1015] 1. User A watches Movie A and posts a review.

[1016] 2. The device collects viewing information and sends it to the server.

[1017] 3. The server saves the data to a database.

[1018] 4. Using the survey results of User A, classify him as an "action movie lover."

[1019] 5. The server trains the generative AI model using User A's data.

[1020] 6. When user A watches movie B, the emotion engine analyzes user A's emotional information in real time.

[1021] 7. The device sends the real-time emotion information to the server.

[1022] 8. The server references user A's profile and sentiment information to generate the most appropriate review.

[1023] 9. The review generated is sent back to the device and displayed on the page for Movie B that User A is viewing.

[1024] Example 2: Product review

[1025] 1. User X buys gadget Z and posts a review.

[1026] 2. The device collects purchase information and sends it to the server.

[1027] 3. The server saves the data to a database.

[1028] 4. Use the survey results of User X to classify him / her into the "gadget enthusiast type."

[1029] 5. The server trains the generative AI model using User X's data.

[1030] 6. When user X browses new gadget Y, the emotion engine analyzes user X's emotion information in real time.

[1031] 7. The device sends the real-time emotion information to the server.

[1032] 8. The server references user X's profile and sentiment information to generate the most appropriate review.

[1033] 9. The device displays the generated review to User X, allowing User X to view a personalized review of Gadget Y in a state of amazement.

[1034] This system allows users to easily find the most suitable reviews based on their preferences and current emotional state, making information gathering and content selection more efficient and personalized. It also enables businesses to conduct more effective marketing based on user emotions, leading to greater success.

[1035] Example 2

[1036] 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."

[1037] Conventional review systems can provide personalized reviews based on a user's personality and preferences, but they cannot provide reviews that take into account the user's current emotional state. This makes it difficult for users to obtain the most appropriate information based on their emotions at any given time. It also makes it difficult for companies to conduct effective marketing based on users' emotions.

[1038] 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.

[1039] In this invention, the server includes means for collecting users' browsing history, purchase history, and review information; means for classifying users' personality and preference types based on the collected data; means for training a generative AI model using the classified data; means for classifying all users into personality and preference types and generating user profiles; means for the user terminal to recognize the user's emotional state in real time using a camera or microphone; means for selecting a corresponding generative AI model based on the user profile and emotional information and generating personalized reviews; and means for displaying the generated personalized reviews on the user terminal. This allows users to obtain optimal reviews based on their current emotional state as well as their own preferences. Companies can also conduct effective marketing based on users' emotions.

[1040] A "user's browsing history" is a record of the pages and content a user accesses on a website or application.

[1041] "Purchase history" is a list of products that a user has purchased in the past and related information.

[1042] "Review information" is data including evaluations and opinions given by users regarding products and services.

[1043] "Personality / Preference Type" is a classification based on a user's personal personality or preferences for specific content or products.

[1044] A "generative AI model" is an artificial intelligence model that generates personalized reviews based on a user's personality, preferences, and emotional state.

[1045] A "user profile" is a data set that integrates a user's personality, preferences, browsing history, purchase history, review information, etc.

[1046] "Means for recognizing emotional states in real time" refers to technology that detects the user's emotions at any given time from their facial expressions and voice.

[1047] "Personalized reviews" are the presentation of ratings and opinions that are individually optimized based on the user's personality, preference type, and emotional state.

[1048] A "user terminal" is a device operated by a user, and includes a personal computer, a smartphone, a tablet, etc.

[1049] This invention combines a personalized AI review system that provides optimal content and product information based on a user's interests and preferences with an emotion engine that recognizes the user's emotions. This system can provide even more personalized reviews based on the user's emotional state at any given time.

[1050] The server collects browsing history, purchase history, and review information generated when users use e-commerce sites and content distribution services. This data is periodically retrieved via API and inserted into a database such as MySQL or MongoDB.

[1051] The server then sends users a questionnaire to collect information about their personalities and preferences. The results of this questionnaire are analyzed using natural language processing (NLP) techniques, and algorithms such as K-means clustering are used to classify users into categories such as "action movie fans" and "comedy movie fans." This classification data is used to generate a training dataset and train a generative AI model. The generative AI model, trained using Python and libraries such as TensorFlow and PyTorch, is specialized for each preference type.

[1052] The server analyzes newly collected user data and classifies all users into preference types using a pre-trained generative AI model. Based on this result, a user profile is generated and stored in a database. To ensure real-time performance, batch processing and real-time data streaming technologies are used.

[1053] The device is equipped with an emotion engine that recognizes the user's emotions in real time. It collects facial expressions and voice data through the camera and microphone, and analyzes emotions using the OpenCV library and the Google Cloud Speech-to-Text API. The analysis results are sent to a server and integrated with the user's profile.

[1054] When a user browses for new products or content, their device sends their request and emotional information to the server. The server then selects a corresponding AI model based on the received user profile and emotional information to generate a personalized review. It uses natural language generation (NLG) technology to generate review text that reflects the user's emotions.

[1055] As a concrete example, consider a scenario in which User A watches Movie A and posts a review. The device collects viewing information and sends it to the server. The server stores the data in a database and analyzes User A's emotional information using an emotion engine. The server creates an optimal review based on the personality, preference type, and emotional information, and sends it to the device. When User A watches Movie B, the emotion engine recognizes that User A is excited, and generates and displays a review that suits that situation.

[1056] An example prompt is:

[1057] "The powerful scenes in this action movie will get you even more excited."

[1058] This system allows users to obtain the most appropriate reviews based on their preferences and current emotional state, making information gathering and content selection more efficient and personalized. It also enables companies to conduct effective marketing based on users' emotions.

[1059] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1060] Step 1:

[1061] The server collects user data. Specifically, it obtains user browsing history, purchase history, and review information via API. The input is user activity log data. The server stores the collected data in a MySQL or MongoDB database. The output is organized user data.

[1062] Step 2:

[1063] The server sends a questionnaire to a certain number of users to collect information on their personalities and preferences. The input is the questionnaire response data from the users. The server analyzes this using natural language processing (NLP) technology and classifies users into personality and preference types using algorithms such as K-means clustering. The output is data classifying the users' personalities and preference types.

[1064] Step 3:

[1065] The server uses the classified data to train a generative AI model. The input is training data categorized into personality and preference types. A generative AI model for each type is trained using Python and the TensorFlow or PyTorch library. The output is a trained generative AI model.

[1066] Step 4:

[1067] The server analyzes all user data. Specifically, it inputs purchase history, browsing history, and review information into a previously trained AI model to classify each user into a personality and preference type. The input is newly collected user data, and the output is an updated user profile.

[1068] Step 5:

[1069] The device uses an emotion engine to recognize the user's emotions in real time. Specifically, it analyzes facial and voice data acquired from the camera and microphone using OpenCV and the Google Cloud Speech-to-Text API to detect the user's emotional state. The input is the user's facial and voice data, and the output is the analyzed emotional information.

[1070] Step 6:

[1071] When a user browses for new products or content, the device sends the request and emotional information to the server. The server references the user profile and emotional information, selects a corresponding generative AI model, and generates a personalized review. The input is the request data, emotional information, and user profile, and the output is the generated review.

[1072] Step 7:

[1073] The server generates personalized reviews and sends them to the device. The device displays the reviews it receives on the user interface (UI). Specifically, the reviews are inserted into web pages or app UIs designed with HTML and CSS, and the display is updated in real time. The input is the reviews sent from the server, and the output is the reviews displayed to the user.

[1074] (Application example 2)

[1075] 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."

[1076] Conventional personalized review systems could provide personalization based on the user's basic personality and preferences, but they could not consider the user's emotional state at any given time, making it difficult to provide more detailed personalization. Furthermore, they could not generate reviews in real time that reflected the user's emotional information, making it impossible to provide feedback that was in line with the user's current emotions.

[1077] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting users' browsing history, purchase history, and review information, means for classifying users' personality and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into personality and preference types and generating user profiles, means for selecting corresponding generative AI models based on the user profiles and generating personalized reviews, emotion engine means for recognizing the user's emotional state in real time, and means for displaying personalized reviews generated based on requests including the user's emotional information on the user terminal. This makes it possible to provide personalized reviews in real time that are tailored to the user's basic personality and preferences as well as their emotional state at any given time.

[1078] "User browsing history" refers to historical information about pages and content that a user views when using a website or application.

[1079] "Purchase history" is historical information about products purchased by a user online or offline.

[1080] "Review information" refers to information about ratings and comments posted by users about products or content that they have viewed or purchased.

[1081] "Means for classifying user personality and preference types" refers to a method or system for classifying a user's personality and preferences into specific types based on collected user data.

[1082] A "means for training a generative AI model" is a method or system that uses classified user data to train a generative AI model that responds to a specific purpose.

[1083] A "user profile" is individual profile information generated by integrating a user's personality, preferences, browsing history, purchase history, and the like.

[1084] The "emotion engine" is a technology that analyzes data such as the user's facial expressions and voice, and recognizes their emotional state at any given time in real time.

[1085] A "personalized review" is a customized rating or comment on a specific product or content that takes into account a user's individual personality, preferences, and feelings.

[1086] A "user terminal" is a device that a user directly uses, and refers to hardware such as a smartphone, tablet, or computer.

[1087] MODE FOR CARRYING OUT THE INVENTION

[1088] System Overview

[1089] This invention is a system that generates personalized reviews based on a user's browsing history, purchase history, and review information. Its special feature is that it recognizes the user's current emotional state and provides reviews that correspond to that emotion at that time. This system consists of a user terminal equipped with an emotion engine and a server that analyzes and processes the data.

[1090] Hardware and software used

[1091] Hardware

[1092] User devices: smartphones, tablets, computers, etc.

[1093] Server: High-performance computer, cloud infrastructure

[1094] software

[1095] EmotionEngine: Software that recognizes a user's emotional state by analyzing their facial expressions and voice

[1096] AIReviewGenerator: A generative AI model that generates personalized reviews based on user profiles and sentiment information

[1097] Django or Flask: a framework for server-side data management and API serving.

[1098] Processing flow and specific examples

[1099] Step 1: Data collection

[1100] The server collects user data such as viewing history, purchase history, and review information from the user's device and stores it in a database. For example, when a user watches a particular movie, the viewing date and time, movie title, rating, and review content are recorded.

[1101] Step 2: Classify users

[1102] The server classifies the user's personality and preferences based on the collected data. This identifies user types such as "action movie lover" or "comedy movie lover." This classification is used to train a generative AI model.

[1103] Step 3: Introducing the Emotion Engine

[1104] The user device is equipped with an EmotionEngine that analyzes the user's facial expressions and voice to recognize their current emotional state. For example, it can analyze the user's reactions in real time while watching a movie to identify their emotional state.

[1105] Step 4: Generate a personalized review

[1106] When a user browses new products or content, the device sends the corresponding request and emotional information to the server, which then selects a corresponding generative AI model based on the user profile and emotional information to generate a personalized review for the specific content.

[1107] Step 5: View reviews

[1108] The generated review is sent to the user's device and displayed to the user. For example, when a user browses a page about a new action movie, if the emotion engine recognizes that the user is excited, a review specific to the action scenes of that movie will be generated and displayed to the user.

[1109] Specific examples

[1110] If the emotion engine determines that the user is excited after watching movie A, it will send the following prompt to the generative AI model.

[1111] Prompt statement:

[1112] User ID 123's emotional state is excited. Generate a personalized review for content he / she is interested in based on the following information: User Data: { "Action Movie Lover", "Recently Watched Movies": "Movie A", "Rating": "High"} Content ID: Movie B

[1113] In this way, a personalized review optimized for the user's current emotional state can be provided. The above-described system and process allow users to obtain reviews that are in line with their emotions, providing a more satisfying content viewing experience.

[1114] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1115] Step 1:

[1116] The server collects user data such as viewing history, purchase history, and review information from the user's device. The collected data is stored in a database. This data includes, for example, the date and time when the user watched a particular movie, the movie title, rating, and review content.

[1117] Input: Browsing history, purchase history, and review information from user devices

[1118] Output: User data stored in the database

[1119] Specific operation: When the server receives an HTTP request, it structures the data and stores it in a database.

[1120] Step 2:

[1121] The server analyzes the collected user data and classifies the user's personality and preferences. For example, it may classify them into types such as "likes action movies" or "likes comedy movies." Training data is generated based on the classification results, and the generative AI model is trained.

[1122] Input: User data stored in the database

[1123] Output: Classified user data, training data

[1124] How it works: The server uses machine learning algorithms to analyze user data and classify it into user types.

[1125] Step 3:

[1126] The server analyzes all user data and classifies each user into personality and preference types, based on which a user profile is created and stored in a database.

[1127] Input: All users' data

[1128] Output: User profile

[1129] Specific operation: The server uses the data classified in the previous step to generate a profile for each user based on their personality and preference types.

[1130] Step 4:

[1131] The user device uses the EmotionEngine to analyze the user's facial expressions and voice in real time to recognize their emotional state. For example, while watching a movie, the device can identify emotional states such as "excitement" or "joy" from the user's facial expressions and voice.

[1132] Input: User's facial expression data, voice data

[1133] Output: Perceived emotional state

[1134] Specific operation: The EmotionEngine in the device analyzes visual and audio data and outputs the emotional state.

[1135] Step 5:

[1136] When a user browses for new products or content, the terminal sends requests and emotion information to the server.

[1137] Input: User request, perceived emotional state

[1138] Output: Request and emotion information sent to the server

[1139] Specific operation: The device sends data to the server as an HTTP request.

[1140] Step 6:

[1141] The server refers to the user profile and emotion information, selects a corresponding generative AI model, and generates a personalized review.

[1142] Input: User profile, perceived emotional state

[1143] Output: The generated personalized review

[1144] Specific operation: The server generates a prompt using AI Review Generator, inputs it into the generative AI model, and generates a review.

[1145] Step 7:

[1146] The generated review is sent to the user's terminal and displayed to the user. For example, if the user is recognized as excited when browsing a page about a new action movie, a review specific to the action scenes of that movie is generated and displayed.

[1147] Input: Generated personalized review

[1148] Output: The review displayed on the user's device

[1149] Specific behavior: The server sends the generated review to the device as an HTTP response, and the device displays it.

[1150] 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.

[1151] 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.

[1152] 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.

[1153] [Fourth embodiment]

[1154] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1155] 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.

[1156] 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).

[1157] 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.

[1158] 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.

[1159] 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).

[1160] 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.

[1161] 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.

[1162] 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.

[1163] 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.

[1164] 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.

[1165] 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.

[1166] 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."

[1167] This invention is a personalized AI review system that efficiently provides content suited to the interests and preferences of individual users on the Internet. This system collects users' browsing history, purchase history, and review information, and uses this information to train a generative AI model to provide optimal reviews to users.

[1168] Program processing flow

[1169] Step 1: Collect user data

[1170] The server collects browsing history, purchase information, and review information generated when users use e-commerce sites and content distribution services, and stores them in a database. Specifically, when users watch a movie, the server collects the viewing date and time, movie title, rating, and review content.

[1171] Step 2: Classify and train the sampled data

[1172] The server sends a questionnaire to some users to collect information about their personalities and preferences. Based on the collected questionnaire results, the server classifies users into personality and preference types, such as "action movie lover" and "comedy movie lover." This classified data is used as training data.

[1173] Step 3: Training the generative AI model

[1174] The server trains a generative AI model based on the classified training data. For example, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. Similarly, separate generative AI models are trained for other types.

[1175] Step 4: Classify all users into types

[1176] The server analyzes all user data, including purchase history, browsing history, and review information, and classifies each user into a personality type and preference type, laying the foundation for providing personalized reviews using a generative AI model that is optimal for each user.

[1177] Step 5: Generate a personalized review

[1178] When a user browses for new content or products, the device sends the request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses the appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[1179] Specific examples

[1180] Example 1: Movie reviews

[1181] When User A watches Movie A, the device sends viewing information (viewing date and time, movie title, rating, review content) to the server. The server collects the data and stores it in a database. Later, when User A watches a new action movie B, the server checks the user profile and classifies them as an "action movie lover." It uses a generative AI model to generate a review of Action Movie B and sends it to the device. The device displays the review on Movie B's page.

[1182] Example 2: Product Review

[1183] User X purchases gadget Z and leaves a rating and review. This information is sent from the device to the server and stored in a database. Next, when User X browses for a new gadget Y, the server references the user profile and generates a personalized review using a generative AI model. The generated review is then displayed on User X's device.

[1184] This system allows users to easily obtain reviews that match their preferences, making information gathering and content selection more efficient. It also enables businesses to conduct marketing based on user preferences, enabling more effective advertising activities.

[1185] The processing flow will be explained below.

[1186] Step 1:

[1187] Users use e-commerce sites and content distribution services to browse products, purchase them, and post reviews.

[1188] The device collects the user's browsing history, purchase information, and review information and sends it to the server.

[1189] The server stores the received data in a database.

[1190] Step 2:

[1191] The server sends a questionnaire to some users to collect information about their personalities and preferences.

[1192] The server analyzes the survey results and classifies users into personality and preference types.

[1193] Based on the survey, people are classified into categories such as "action movie lovers" and "comedy movie lovers."

[1194] Step 3:

[1195] The server extracts the classified user reviews and rating information as training data.

[1196] This training data is used to train generative AI models specialized for each personality and preference type.

[1197] For example, we will use user data of "action movie lovers" to train an AI model for generating action movie reviews.

[1198] Step 4:

[1199] The server analyzes all users' browsing history, purchase history, and review information.

[1200] Based on the analysis results, all users are classified into personality and preference types.

[1201] A user profile is generated and stored in a database.

[1202] Step 5:

[1203] When a user browses for new products or content, the device sends a request to the server.

[1204] The server references the user profile and checks the user's personality and preferences.

[1205] Select the appropriate generative AI model to generate personalized reviews.

[1206] Step 6:

[1207] The server sends the generated reviews to the device.

[1208] The terminal displays the received reviews to the user.

[1209] For example, when a user browses to a page about new action movie B, they will see reviews created using a generative AI model specialized for action movies.

[1210] Specific examples

[1211] Example 1: Movie reviews

[1212] 1. User A watches Movie A and posts a review.

[1213] 2. The device collects viewing information and sends it to the server.

[1214] 3. The server saves the data to a database.

[1215] 4. Using the survey results of User A, classify him as an "action movie lover."

[1216] 5. The server trains the generative AI model using User A's data.

[1217] 6. When user A wants to watch a new action movie B, the device sends a request.

[1218] 7. The server checks User A's profile and generates a personalized review.

[1219] 8. The generated review is displayed on User A's device.

[1220] Example 2: Product review

[1221] 1. User X buys gadget Z and posts a review.

[1222] 2. The device collects purchase information and sends it to the server.

[1223] 3. The server saves the data to a database.

[1224] 4. Using the survey results of User X, classify him / her as a "gadget enthusiast type."

[1225] 5. The server trains the generative AI model using User X's data.

[1226] 6. When user X views new gadget Y, the device sends a request.

[1227] 7. The server checks User X's profile and generates a personalized review.

[1228] 8. The generated review is displayed on User X's device.

[1229] Example 1

[1230] 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."

[1231] Many current e-commerce sites and content distribution services only provide users with general reviews and ratings, making it difficult to provide customized reviews that match the personalities and preferences of individual users. Furthermore, users lack the information they need to select the content and products that best suit them, preventing them from making efficient selections. To address these issues, a system is needed that provides reviews based on the individual preferences of users.

[1232] 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.

[1233] In this invention, the server includes means for collecting user operation history, purchase history, and evaluation information, means for classifying user personality and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into personality and preference types and generating user information, means for selecting a corresponding generative AI model based on the user information and generating customized reviews, and means for displaying the generated reviews on a user device. This allows users to easily obtain reviews that match their preferences, enabling them to efficiently collect information and select content.

[1234] "User operation history" refers to the record of operations performed by a user on an e-commerce site or content distribution service, and specifically includes the products and content viewed, the links clicked, and the user's behavior on the site.

[1235] "Purchase history" refers to a record of products and services that a user actually purchased on an e-commerce site, etc., and includes information such as the purchase date and time, product name, quantity, and price.

[1236] "Rating information" refers to records of ratings and reviews made by users on products, services, and content purchased by users, and includes feedback such as rating scores and comments.

[1237] A "generative artificial intelligence model" is an AI model that is generated by training a machine learning algorithm using collected data, and has the ability to predict and generate for specific tasks.

[1238] "User information" refers to the classification of a user's personality and preferences based on collected data, and the profile information based on that classification, including data regarding the user's interests and preferences.

[1239] "Customized reviews" refer to personalized reviews for individual users that are generated based on the user's personality and preferences, and unlike general reviews, contain content that reflects the user's specific needs and interests.

[1240] "User equipment" refers to a device used by a user through an Internet connection, and specifically includes smartphones, tablets, PCs, etc.

[1241] "Immediate generation" means that the generation process begins immediately after the user submits the request, without delay, and the results are provided within a short time.

[1242] This invention is a personalized AI review system for efficiently providing content suited to the interests and preferences of individual users on the Internet. This system collects users' operation history, purchase history, and rating information, and uses this information to train a generative AI model to provide optimal reviews to users.

[1243] The server collects operation history, purchase information, and rating information generated when users use e-commerce sites and content distribution services, and stores this information in a database. Specifically, the hardware and software that collects information such as the viewing date and time, movie title, rating, and review content when users watch a movie can stream data in real time using Apache Kafka, allowing data to be collected and stored efficiently.

[1244] The server sends a questionnaire to a subset of users to collect information about their personalities and preferences. Based on the survey results, users are classified into personality and preference types, such as "action movie lover" or "comedy movie lover." This classified data is used as training data. Data analysis is performed using data science languages ​​such as Python and R.

[1245] The server trains a generative AI model based on the classified training data. For example, a generative AI model specialized for action movies is created using user data for "action movie lovers." This training is performed using machine learning frameworks such as TensorFlow and PyTorch. The trained model is saved on the server and used for subsequent processing.

[1246] The server analyzes all user data and classifies each user into personality and preference types. This is done using an analytical algorithm based on the user's purchase history, operation history, and rating information. The classified user information is saved in a database as a profile.

[1247] When a user browses for new content or products, the device sends a request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses an appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[1248] As a concrete example, when a user watches action movie B, the device sends viewing information (viewing date and time, movie title, rating, review content) to the server. The server collects the data and stores it in a database. Later, when the user views a new action movie C, the server checks the user profile and selects an appropriate generative AI model. The generated review is then displayed on the user's device.

[1249] Example prompt sentence:

[1250] 1. "Generate a personalized review of action movie B that user A is watching."

[1251] 2. "Generate a review for gadget Y by user X."

[1252] 3. "For users who like comedy movies, please write a review of a new comedy movie, C."

[1253] This system allows users to easily obtain reviews that match their preferences, enabling them to efficiently gather information and select content. It also enables businesses to conduct marketing based on user preferences, enabling more effective advertising activities.

[1254] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1255] Step 1:

[1256] The server collects operation history, purchase information, and rating information generated when a user uses an e-commerce site or content distribution service. Specifically, the device collects data such as the user's viewing date and time, movie title, rating, and review content, and sends this data to the server in real time. The server stores the received data in a database.

[1257] Input: User operation history, purchase information, rating information

[1258] Data processing / calculation: Real-time streaming, saving to database

[1259] Output: Saved data

[1260] Step 2:

[1261] The server sends a questionnaire to some users to collect information about their personalities and preferences. The users answer the questionnaire and send the results to the server. The server analyzes the questionnaire results and classifies users into categories such as "action movie lover" or "comedy movie lover."

[1262] Input: Survey response

[1263] Data processing / calculation: Analysis of survey results, classification of users

[1264] Output: Classified user data

[1265] Step 3:

[1266] The server trains a generative AI model based on the classified user data. For example, it uses user data for "action movie lovers" to create an AI model specialized for action movies. Specifically, it uses the collected data as training data using TensorFlow and PyTorch to train the generative AI model.

[1267] Input: Classified user data

[1268] Data processing / computation: training generative AI models

[1269] Output: A trained generative AI model

[1270] Step 4:

[1271] The server analyzes all user data and classifies each user into personality and preference types. This is done using an analytical algorithm based on the user's purchase history, operation history, and rating information. The classified user information is saved in a database as a profile.

[1272] Input: Purchase history, operation history, and rating information of all users

[1273] Data processing / calculation: Data analysis, user classification

[1274] Output: User profile

[1275] Step 5:

[1276] When a user browses for new content or products, the device sends a request to the server. The server references the user profile to determine which personality and preference type the user falls into. It then uses an appropriate generative AI model to generate a personalized review and sends it back to the device. The device then displays the received review to the user.

[1277] Input: User request, user profile

[1278] Data processing / calculation: Profile reference, review generation using AI model

[1279] Output: Generated reviews

[1280] In this way, personalized reviews based on the user's preferences can be efficiently provided.

[1281] (Application example 1)

[1282] 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."

[1283] In today's Internet environment, users face the challenge of finding the content and products that best suit them from the vast amount of information available. Furthermore, current review systems often provide general opinions, and rarely provide information specific to individual users' preferences and characteristics. This means that users spend time and effort sifting through information to select the content and products that best suit them.

[1284] 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.

[1285] In this invention, the server includes means for collecting user browsing history, purchase history, and rating information, means for classifying user characteristics and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into characteristics and preference types and generating user profiles, means for selecting corresponding generative AI models based on the user profiles and generating personalized reviews, means for displaying the generated reviews on an information processing device, and means implemented as a smartphone application for providing optimal reviews for users in real time based on their viewing history and purchase history. This allows users to quickly obtain review information that matches their preferences and characteristics, making content and product selection efficient and effective.

[1286] A "user's browsing history" is a record of which web pages and content a user has viewed on the Internet.

[1287] "Purchase history" is a record of what products a user has purchased on the Internet.

[1288] "Rating information" refers to information about ratings and reviews of content or products that a user has viewed or purchased.

[1289] "Characteristics / preference type" refers to a type classified based on the user's preferences and personality traits.

[1290] A "user profile" is a detailed profile of personal information created based on data such as a user's characteristics, preferences, browsing history, and purchase history.

[1291] A "generative AI model" is an artificial intelligence model that uses machine learning to learn from a specific dataset and perform a specific task.

[1292] A "personalized review" is a customized review generated based on the individual characteristics and preferences of a user.

[1293] An "information processing device" is an electronic device that has the function of inputting, processing, and outputting data, and here it mainly refers to a smartphone.

[1294] A "smartphone application" is a software program that runs on a smartphone.

[1295] This invention is a system that collects users' browsing history, purchase history, and rating information, and generates personalized reviews based on the users' characteristics and preferences. This system is composed of a server, an information processing device (such as a smartphone), a generation AI model, etc. Specific embodiments are described below.

[1296] Hardware and software used

[1297] Server: Cloud server (e.g. Amazon Web Services EC2)

[1298] Database: Cloud database (e.g. Amazon RDS)

[1299] Generative AI models: models for natural language generation (e.g., OpenAI GPT)

[1300] Data processing device: Smartphone (e.g., Android or iOS app)

[1301] Data processing and calculation

[1302] Data collection and storage

[1303] The server collects browsing history, purchase history, and rating information generated when a user uses an information processing device (smartphone), and stores this data in a cloud database. For example, when a user watches a movie, the movie title, viewing date and time, rating, and review content are stored. This data serves as the basis for analyzing user preferences.

[1304] User Classification and Data Classification

[1305] Based on the collected data, the server categorizes users into characteristics and preferences. Specifically, based on their browsing and purchase history, users may be classified as those who like action movies or comedy movies. This categorization is stored in a cloud database and used as training data for the generative AI model.

[1306] Training an AI model

[1307] The server uses the classified data to train a generative AI model. For example, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. In this process, generative AI models such as OpenAI GPT are used.

[1308] Generate personalized reviews

[1309] When a user browses new content on an information processing device (smartphone), the server references the user profile and uses the corresponding generative AI model to generate a personalized review, which is generated in real time and displayed on the information processing device.

[1310] Specific examples

[1311] Example 1: Personalized movie reviews

[1312] When a user browses for a new movie using their smartphone, the server checks the user's profile and, if the user is classified as an "action movie lover," uses the generative AI model to generate a personalized review of that movie for action movie lovers. The generated review is displayed on the smartphone, allowing the user to use it as a reference when choosing a movie.

[1313] Prompt Sentence Examples

[1314] A user loved the action movie and gave the following review: It was packed with great action scenes and I really enjoyed it!

[1315] Example 2: Personalizing product reviews

[1316] When a user browses for a new gadget, the server checks the user's profile and uses a generative AI model to generate a personalized review of the gadget, which is then displayed on the user's smartphone to help inform their purchasing decision.

[1317] In this way, the invention provides personalized reviews based on the user's preferences, allowing the user to efficiently select the content and products that are best suited to them, thereby significantly improving the user experience.

[1318] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1319] Step 1:

[1320] Data collection

[1321] When a user browses or purchases using a device (smartphone), that data is collected. Specifically, when a user watches a movie, information such as the viewing date and time, movie title, rating, and review content is sent from the device to the server. The input data is viewing history, purchase history, and rating information, and the server stores this data in a cloud database. The output is the stored user data.

[1322] Step 2:

[1323] User Classification

[1324] The server classifies users into characteristics and preference types based on the collected data. The collected data (input) includes the user's browsing history and purchase history. The server analyzes this data to determine the genre of content the user prefers. For example, a user who watches a lot of action movies would be classified as an "action movie lover." This classification information (output) is stored in a cloud database.

[1325] Step 3:

[1326] Training generative AI models

[1327] The server uses the classified data (input) to train the generative AI model. Specifically, it uses user data for "action movie lovers" to create a generative AI model specialized for action movies. The generative AI model used in this process is OpenAI GPT, for example. Once training is complete, multiple generative AI models (output) according to their characteristics are obtained.

[1328] Step 4:

[1329] Creating a user profile

[1330] The server analyzes the data of all users, classifies each user into a characteristic and preference type, and generates a user profile. The input is the user's browsing history, purchase history, and rating information. The server analyzes this data and determines which preference type each user belongs to. Based on this determination, an individual user profile (output) is generated and stored in a cloud database.

[1331] Step 5:

[1332] Generate personalized reviews

[1333] When a user browses new content, a request is sent from the device to the server. The server references the user profile (input) and generates a personalized review using the corresponding generative AI model. Specifically, the server checks the user's preference type and generates a review using prompts appropriate for that type. The generated review (output) is sent to the device and displayed to the user.

[1334] Step 6:

[1335] View Reviews

[1336] The terminal displays the personalized reviews (input) received from the server to the user, allowing the user to select new content or products based on reviews that match their preferences. The output is the displayed reviews.

[1337] Specific examples

[1338] Example prompt sentence:

[1339] A user loved the action movie and gave the following review: It was packed with great action scenes and I really enjoyed it!

[1340] 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.

[1341] This invention combines a personalized AI review system that provides optimal content and product information based on a user's interests and preferences with an emotion engine that recognizes the user's emotions. This system can provide even more personalized reviews based on the user's emotional state at any given time.

[1342] Program processing flow

[1343] Step 1: Collect user data

[1344] The server collects browsing history, purchase information, and review information generated when a user uses an e-commerce site or content distribution service, and stores this information in a database. For example, when a user watches a movie, the server collects the viewing date and time, movie title, rating, and review content.

[1345] Step 2: Classify and train the sampled data

[1346] The server sends a questionnaire to a certain number of users to collect information about their personalities and preferences. The server analyzes the survey results and classifies users into personality and preference types. For example, it classifies users into "action movie lovers" and "comedy movie lovers." Based on this, it generates training data and trains the generative AI model.

[1347] Step 3: Training the generative AI model

[1348] The server trains generative AI models for each personality and preference type based on the classified training data. For example, a generative AI model specialized for action movies is created using user data for "action movie lovers." Individual generative AI models are similarly trained for other types.

[1349] Step 4: Classify all users into types

[1350] The server analyzes all user data, including purchase history, browsing history, and review information, and classifies each user into a personality and preference type. Based on the analysis results, a user profile is generated and stored in a database.

[1351] Step 5: Implementing the Emotion Engine

[1352] The device is equipped with an emotion engine that recognizes the user's emotions in real time. The emotion engine analyzes facial expressions and speech while the user is viewing content to recognize the user's emotional state.

[1353] Step 6: Generate a personalized review

[1354] When a user browses new products or content, the device sends a request and emotional information to the server, which then references the user profile and emotional information to select a corresponding generative AI model to generate a personalized review.

[1355] Step 7: View personalized reviews

[1356] The server sends the generated review to the device, which then displays the received review to the user. For example, when a user browses a page about a new action movie B, if the emotion engine recognizes that the user is excited, it generates and displays a review that focuses on the action scenes.

[1357] Specific examples

[1358] Example 1: Movie reviews

[1359] User A watches Movie A and posts a review. The device collects viewing information and sends it to the server. The server saves the data in a database. User A's emotional information is then analyzed using an emotion engine. The server creates an optimal review based on the user's personality, preference type, and emotional information, and sends it to the device. When User A watches Movie B, the emotion engine recognizes that User A is excited, and generates and displays a review that suits that situation.

[1360] Example 2: Product review

[1361] User X purchases gadget Z and posts a review. The device collects purchase information and sends it to the server. The server saves the data in a database. The emotion engine then analyzes User X's emotional information. The server creates an optimal review based on the user's personality, preference type, and emotional information, and sends it to the device. When User X browses new gadget Y, the emotion engine recognizes that User X is surprised, and generates and displays a review that reflects that emotion.

[1362] This system allows users to obtain the most appropriate reviews based on their preferences and current emotional state. This makes information gathering and content selection more efficient and personalized. It also enables businesses to conduct more detailed marketing based on users' emotions, which is expected to produce even greater results.

[1363] The processing flow will be explained below.

[1364] Step 1:

[1365] Users use e-commerce sites and content distribution services to browse, purchase, and post reviews of products.

[1366] The device collects the user's browsing history, purchase information, and review information and sends it to the server.

[1367] The server stores the received data in a database.

[1368] Step 2:

[1369] The server sends a questionnaire to some users to collect information about their personalities and preferences.

[1370] The server analyzes the survey results and classifies users into personality and preference types.

[1371] For example, classifications such as "action movie lover type" and "comedy movie lover type" are made.

[1372] Step 3:

[1373] The server extracts the classified user reviews and rating information as training data.

[1374] This training data is used to train generative AI models specialized for each personality and preference type.

[1375] For example, we will use user data of "action movie lovers" to train an AI model for generating action movie reviews.

[1376] Step 4:

[1377] The server analyzes all users' browsing history, purchase history, and review information.

[1378] Based on the analysis results, all users are classified into personality and preference types.

[1379] A user profile is generated and stored in a database.

[1380] Step 5:

[1381] The device is equipped with an emotion engine that allows it to recognize the user's emotions in real time.

[1382] The emotion engine analyzes the user's facial expressions and speech while viewing content, and recognizes the user's emotional state.

[1383] Step 6:

[1384] When a user browses for new products or content, the device sends the request and emotion information to the server.

[1385] The server refers to the user profile and sentiment information, selects a corresponding generative AI model, and generates a personalized review.

[1386] Step 7:

[1387] The server sends the generated reviews to the device.

[1388] The terminal displays the received reviews to the user.

[1389] For example, when a user browses a page about a new action movie B, if the emotion engine recognizes that the user is excited, it will generate a review that focuses on the action scenes and display it on the device.

[1390] Specific examples

[1391] Example 1: Movie reviews

[1392] 1. User A watches Movie A and posts a review.

[1393] 2. The device collects viewing information and sends it to the server.

[1394] 3. The server saves the data to a database.

[1395] 4. Using the survey results of User A, classify him as an "action movie lover."

[1396] 5. The server trains the generative AI model using User A's data.

[1397] 6. When user A watches movie B, the emotion engine analyzes user A's emotional information in real time.

[1398] 7. The device sends the real-time emotion information to the server.

[1399] 8. The server references user A's profile and sentiment information to generate the most appropriate review.

[1400] 9. The review generated is sent back to the device and displayed on the page for Movie B that User A is viewing.

[1401] Example 2: Product review

[1402] 1. User X buys gadget Z and posts a review.

[1403] 2. The device collects purchase information and sends it to the server.

[1404] 3. The server saves the data to a database.

[1405] 4. Use the survey results of User X to classify him / her into the "gadget enthusiast type."

[1406] 5. The server trains the generative AI model using User X's data.

[1407] 6. When user X browses new gadget Y, the emotion engine analyzes user X's emotion information in real time.

[1408] 7. The device sends the real-time emotion information to the server.

[1409] 8. The server references user X's profile and sentiment information to generate the most appropriate review.

[1410] 9. The device displays the generated review to User X, allowing User X to view a personalized review of Gadget Y in a state of amazement.

[1411] This system allows users to easily find the most suitable reviews based on their preferences and current emotional state, making information gathering and content selection more efficient and personalized. It also enables businesses to conduct more effective marketing based on user emotions, leading to greater success.

[1412] Example 2

[1413] 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."

[1414] Conventional review systems can provide personalized reviews based on a user's personality and preferences, but they cannot provide reviews that take into account the user's current emotional state. This makes it difficult for users to obtain the most appropriate information based on their emotions at any given time. It also makes it difficult for companies to conduct effective marketing based on users' emotions.

[1415] 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.

[1416] In this invention, the server includes means for collecting users' browsing history, purchase history, and review information; means for classifying users' personality and preference types based on the collected data; means for training a generative AI model using the classified data; means for classifying all users into personality and preference types and generating user profiles; means for the user terminal to recognize the user's emotional state in real time using a camera or microphone; means for selecting a corresponding generative AI model based on the user profile and emotional information and generating personalized reviews; and means for displaying the generated personalized reviews on the user terminal. This allows users to obtain optimal reviews based on their current emotional state as well as their own preferences. Companies can also conduct effective marketing based on users' emotions.

[1417] A "user's browsing history" is a record of the pages and content a user accesses on a website or application.

[1418] "Purchase history" is a list of products that a user has purchased in the past and related information.

[1419] "Review information" is data including evaluations and opinions given by users regarding products and services.

[1420] "Personality / Preference Type" is a classification based on a user's personal personality or preferences for specific content or products.

[1421] A "generative AI model" is an artificial intelligence model that generates personalized reviews based on a user's personality, preferences, and emotional state.

[1422] A "user profile" is a data set that integrates a user's personality, preferences, browsing history, purchase history, review information, etc.

[1423] "Means for recognizing emotional states in real time" refers to technology that detects the user's emotions at any given time from their facial expressions and voice.

[1424] "Personalized reviews" are the presentation of ratings and opinions that are individually optimized based on the user's personality, preference type, and emotional state.

[1425] A "user terminal" is a device operated by a user, and includes a personal computer, a smartphone, a tablet, etc.

[1426] This invention combines a personalized AI review system that provides optimal content and product information based on a user's interests and preferences with an emotion engine that recognizes the user's emotions. This system can provide even more personalized reviews based on the user's emotional state at any given time.

[1427] The server collects browsing history, purchase history, and review information generated when users use e-commerce sites and content distribution services. This data is periodically retrieved via API and inserted into a database such as MySQL or MongoDB.

[1428] The server then sends users a questionnaire to collect information about their personalities and preferences. The results of this questionnaire are analyzed using natural language processing (NLP) techniques, and algorithms such as K-means clustering are used to classify users into categories such as "action movie fans" and "comedy movie fans." This classification data is used to generate a training dataset and train a generative AI model. The generative AI model, trained using Python and libraries such as TensorFlow and PyTorch, is specialized for each preference type.

[1429] The server analyzes newly collected user data and classifies all users into preference types using a pre-trained generative AI model. Based on this result, a user profile is generated and stored in a database. To ensure real-time performance, batch processing and real-time data streaming technologies are used.

[1430] The device is equipped with an emotion engine that recognizes the user's emotions in real time. It collects facial expressions and voice data through the camera and microphone, and analyzes emotions using the OpenCV library and the Google Cloud Speech-to-Text API. The analysis results are sent to a server and integrated with the user's profile.

[1431] When a user browses for new products or content, their device sends their request and emotional information to the server. The server then selects a corresponding AI model based on the received user profile and emotional information to generate a personalized review. It uses natural language generation (NLG) technology to generate review text that reflects the user's emotions.

[1432] As a concrete example, consider a scenario in which User A watches Movie A and posts a review. The device collects viewing information and sends it to the server. The server stores the data in a database and analyzes User A's emotional information using an emotion engine. The server creates an optimal review based on the personality, preference type, and emotional information, and sends it to the device. When User A watches Movie B, the emotion engine recognizes that User A is excited, and generates and displays a review that suits that situation.

[1433] An example prompt is:

[1434] "The powerful scenes in this action movie will get you even more excited."

[1435] This system allows users to obtain the most appropriate reviews based on their preferences and current emotional state, making information gathering and content selection more efficient and personalized. It also enables companies to conduct effective marketing based on users' emotions.

[1436] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1437] Step 1:

[1438] The server collects user data. Specifically, it obtains user browsing history, purchase history, and review information via API. The input is user activity log data. The server stores the collected data in a MySQL or MongoDB database. The output is organized user data.

[1439] Step 2:

[1440] The server sends a questionnaire to a certain number of users to collect information on their personalities and preferences. The input is the questionnaire response data from the users. The server analyzes this using natural language processing (NLP) technology and classifies users into personality and preference types using algorithms such as K-means clustering. The output is data classifying the users' personalities and preference types.

[1441] Step 3:

[1442] The server uses the classified data to train a generative AI model. The input is training data categorized into personality and preference types. A generative AI model for each type is trained using Python and the TensorFlow or PyTorch library. The output is a trained generative AI model.

[1443] Step 4:

[1444] The server analyzes all user data. Specifically, it inputs purchase history, browsing history, and review information into a previously trained AI model to classify each user into a personality and preference type. The input is newly collected user data, and the output is an updated user profile.

[1445] Step 5:

[1446] The device uses an emotion engine to recognize the user's emotions in real time. Specifically, it analyzes facial and voice data acquired from the camera and microphone using OpenCV and the Google Cloud Speech-to-Text API to detect the user's emotional state. The input is the user's facial and voice data, and the output is the analyzed emotional information.

[1447] Step 6:

[1448] When a user browses for new products or content, the device sends the request and emotional information to the server. The server references the user profile and emotional information, selects a corresponding generative AI model, and generates a personalized review. The input is the request data, emotional information, and user profile, and the output is the generated review.

[1449] Step 7:

[1450] The server generates personalized reviews and sends them to the device. The device displays the reviews it receives on the user interface (UI). Specifically, the reviews are inserted into web pages or app UIs designed with HTML and CSS, and the display is updated in real time. The input is the reviews sent from the server, and the output is the reviews displayed to the user.

[1451] (Application example 2)

[1452] 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."

[1453] Conventional personalized review systems could provide personalization based on the user's basic personality and preferences, but they could not consider the user's emotional state at any given time, making it difficult to provide more detailed personalization. Furthermore, they could not generate reviews in real time that reflected the user's emotional information, making it impossible to provide feedback that was in line with the user's current emotions.

[1454] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting users' browsing history, purchase history, and review information, means for classifying users' personality and preference types based on the collected data, means for training a generative AI model using the classified data, means for classifying all users into personality and preference types and generating user profiles, means for selecting corresponding generative AI models based on the user profiles and generating personalized reviews, emotion engine means for recognizing the user's emotional state in real time, and means for displaying personalized reviews generated based on requests including the user's emotional information on the user terminal. This makes it possible to provide personalized reviews in real time that are tailored to the user's basic personality and preferences as well as their emotional state at any given time.

[1455] "User browsing history" refers to historical information about pages and content that a user views when using a website or application.

[1456] "Purchase history" is historical information about products purchased by a user online or offline.

[1457] "Review information" refers to information about ratings and comments posted by users about products or content that they have viewed or purchased.

[1458] "Means for classifying user personality and preference types" refers to a method or system for classifying a user's personality and preferences into specific types based on collected user data.

[1459] A "means for training a generative AI model" is a method or system that uses classified user data to train a generative AI model that responds to a specific purpose.

[1460] A "user profile" is individual profile information generated by integrating a user's personality, preferences, browsing history, purchase history, and the like.

[1461] The "emotion engine" is a technology that analyzes data such as the user's facial expressions and voice, and recognizes their emotional state at any given time in real time.

[1462] A "personalized review" is a customized rating or comment on a specific product or content that takes into account a user's individual personality, preferences, and feelings.

[1463] A "user terminal" is a device that a user directly uses, and refers to hardware such as a smartphone, tablet, or computer.

[1464] MODE FOR CARRYING OUT THE INVENTION

[1465] System Overview

[1466] This invention is a system that generates personalized reviews based on a user's browsing history, purchase history, and review information. Its special feature is that it recognizes the user's current emotional state and provides reviews that correspond to that emotion at that time. This system consists of a user terminal equipped with an emotion engine and a server that analyzes and processes the data.

[1467] Hardware and software used

[1468] Hardware

[1469] User devices: smartphones, tablets, computers, etc.

[1470] Server: High-performance computer, cloud infrastructure

[1471] software

[1472] EmotionEngine: Software that recognizes a user's emotional state by analyzing their facial expressions and voice

[1473] AIReviewGenerator: A generative AI model that generates personalized reviews based on user profiles and sentiment information

[1474] Django or Flask: a framework for server-side data management and API serving.

[1475] Processing flow and specific examples

[1476] Step 1: Data collection

[1477] The server collects user data such as viewing history, purchase history, and review information from the user's device and stores it in a database. For example, when a user watches a particular movie, the viewing date and time, movie title, rating, and review content are recorded.

[1478] Step 2: Classify users

[1479] The server classifies the user's personality and preferences based on the collected data. This identifies user types such as "action movie lover" or "comedy movie lover." This classification is used to train a generative AI model.

[1480] Step 3: Introducing the Emotion Engine

[1481] The user device is equipped with an EmotionEngine that analyzes the user's facial expressions and voice to recognize their current emotional state. For example, it can analyze the user's reactions in real time while watching a movie to identify their emotional state.

[1482] Step 4: Generate a personalized review

[1483] When a user browses new products or content, the device sends the corresponding request and emotional information to the server, which then selects a corresponding generative AI model based on the user profile and emotional information to generate a personalized review for the specific content.

[1484] Step 5: View reviews

[1485] The generated review is sent to the user's device and displayed to the user. For example, when a user browses a page about a new action movie, if the emotion engine recognizes that the user is excited, a review specific to the action scenes of that movie will be generated and displayed to the user.

[1486] Specific examples

[1487] If the emotion engine determines that the user is excited after watching movie A, it will send the following prompt to the generative AI model.

[1488] Prompt statement:

[1489] User ID 123's emotional state is excited. Generate a personalized review for content he / she is interested in based on the following information: User Data: { "Action Movie Lover", "Recently Watched Movies": "Movie A", "Rating": "High"} Content ID: Movie B

[1490] In this way, a personalized review optimized for the user's current emotional state can be provided. The above-described system and process allow users to obtain reviews that are in line with their emotions, providing a more satisfying content viewing experience.

[1491] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1492] Step 1:

[1493] The server collects user data such as viewing history, purchase history, and review information from the user's device. The collected data is stored in a database. This data includes, for example, the date and time when the user watched a particular movie, the movie title, rating, and review content.

[1494] Input: Browsing history, purchase history, and review information from user devices

[1495] Output: User data stored in the database

[1496] Specific operation: When the server receives an HTTP request, it structures the data and stores it in a database.

[1497] Step 2:

[1498] The server analyzes the collected user data and classifies the user's personality and preferences. For example, it may classify them into types such as "likes action movies" or "likes comedy movies." Training data is generated based on the classification results, and the generative AI model is trained.

[1499] Input: User data stored in the database

[1500] Output: Classified user data, training data

[1501] How it works: The server uses machine learning algorithms to analyze user data and classify it into user types.

[1502] Step 3:

[1503] The server analyzes all user data and classifies each user into personality and preference types, based on which a user profile is created and stored in a database.

[1504] Input: All users' data

[1505] Output: User profile

[1506] Specific operation: The server uses the data classified in the previous step to generate a profile for each user based on their personality and preference types.

[1507] Step 4:

[1508] The user device uses the EmotionEngine to analyze the user's facial expressions and voice in real time to recognize their emotional state. For example, while watching a movie, the device can identify emotional states such as "excitement" or "joy" from the user's facial expressions and voice.

[1509] Input: User's facial expression data, voice data

[1510] Output: Perceived emotional state

[1511] Specific operation: The EmotionEngine in the device analyzes visual and audio data and outputs the emotional state.

[1512] Step 5:

[1513] When a user browses for new products or content, the terminal sends requests and emotion information to the server.

[1514] Input: User request, perceived emotional state

[1515] Output: Request and emotion information sent to the server

[1516] Specific operation: The device sends data to the server as an HTTP request.

[1517] Step 6:

[1518] The server refers to the user profile and emotion information, selects a corresponding generative AI model, and generates a personalized review.

[1519] Input: User profile, perceived emotional state

[1520] Output: The generated personalized review

[1521] Specific operation: The server generates a prompt using AI Review Generator, inputs it into the generative AI model, and generates a review.

[1522] Step 7:

[1523] The generated review is sent to the user's terminal and displayed to the user. For example, if the user is recognized as excited when browsing a page about a new action movie, a review specific to the action scenes of that movie is generated and displayed.

[1524] Input: Generated personalized review

[1525] Output: The review displayed on the user's device

[1526] Specific behavior: The server sends the generated review to the device as an HTTP response, and the device displays it.

[1527] 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.

[1528] 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.

[1529] 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.

[1530] 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.

[1531] 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.

[1532] 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.

[1533] 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).

[1534] 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.

[1535] 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."

[1536] 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.

[1537] 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).

[1538] 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.

[1539] 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.

[1540] 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.

[1541] 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.

[1542] 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.

[1543] 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.

[1544] 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.

[1545] 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.

[1546] 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.

[1547] 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.

[1548] The following is further disclosed regarding the above embodiment.

[1549] (Claim 1)

[1550] A means for collecting user browsing history, purchase history, and review information;

[1551] A means for classifying user personality and preference types based on collected data;

[1552] a means for training a generative AI model using the classified data; and

[1553] A means for classifying all users into personality and preference types and generating user profiles;

[1554] A means for selecting a corresponding generative AI model based on a user profile to generate a personalized review;

[1555] The system includes means for displaying the generated reviews on a user terminal.

[1556] (Claim 2)

[1557] A means for collecting questionnaires from users and classifying their personality and preference types;

[1558] 10. The system of claim 1, further comprising means for generating training data using survey results.

[1559] (Claim 3)

[1560] 10. The system of claim 1, further comprising means for generating personalized reviews of products or content in real time in response to a user request.

[1561] "Example 1"

[1562] (Claim 1)

[1563] A means for collecting user operation history, purchase history, and evaluation information;

[1564] A means for classifying user personality and preference types based on collected data;

[1565] a means for training a generative artificial intelligence model using the classified data;

[1566] A means for classifying all users into personality and preference types and generating user information;

[1567] A means for selecting a corresponding artificial intelligence model based on user information to generate a customized review;

[1568] The system includes means for displaying the generated reviews on a user device.

[1569] (Claim 2)

[1570] A means of collecting survey data from users and classifying their personality and preference types;

[1571] 10. The system of claim 1, further comprising means for generating training data using the survey results.

[1572] (Claim 3)

[1573] 10. The system of claim 1, further comprising means for instantly generating a customized review of a product or content in response to a user request.

[1574] "Application Example 1"

[1575] (Claim 1)

[1576] A means for collecting user browsing history, purchase history, and evaluation information;

[1577] A means for classifying user characteristics and preference types based on collected data;

[1578] a means for training a generative AI model using the classified data; and

[1579] A means for classifying all users into characteristic / preference types and generating user profiles;

[1580] A means for selecting a corresponding generative AI model based on a user profile to generate a personalized review;

[1581] a means for displaying the generated reviews on an information processing device;

[1582] Implemented as a smartphone application, it provides users with the most appropriate reviews in real time based on their viewing and purchasing history.

[1583] A system including:

[1584] (Claim 2)

[1585] A means for collecting questionnaires from users and classifying their characteristics and preferences;

[1586] 10. The system of claim 1, further comprising means for generating training data using survey results.

[1587] (Claim 3)

[1588] 10. The system of claim 1, further comprising means for generating personalized reviews of content offerings or products in real time in response to a user request.

[1589] "Example 2: Combining Emotion Engines"

[1590] (Claim 1)

[1591] A means for collecting user browsing history, purchase history, and review information;

[1592] A means for classifying user personality and preference types based on collected data;

[1593] a means for training a generative AI model using the classified data; and

[1594] A means for classifying all users into personality and preference types and generating user profiles;

[1595] A means for the user terminal to recognize the user's emotional state in real time using a camera or microphone;

[1596] A means for selecting a corresponding generative AI model based on the user profile and emotion information to generate a personalized review;

[1597] The system includes a means for displaying the generated personalized review on a user terminal.

[1598] (Claim 2)

[1599] A means for collecting questionnaires from users and classifying their personality and preference types;

[1600] 10. The system of claim 1, further comprising means for generating training data using survey results.

[1601] (Claim 3)

[1602] 10. The system of claim 1, further comprising means for generating personalized reviews of products or content in real time in response to a user request.

[1603] "Application example 2 when combining emotion engines"

[1604] (Claim 1)

[1605] A means for collecting user browsing history, purchase history, and review information;

[1606] A means for classifying user personality and preference types based on collected data;

[1607] a means for training a generative AI model using the classified data; and

[1608] A means for classifying all users into personality and preference types and generating user profiles;

[1609] A means for selecting a corresponding generative AI model based on a user profile to generate a personalized review;

[1610] emotion engine means for recognizing the user's emotional state in real time;

[1611] A system including a means for displaying, on a user terminal, a personalized review generated based on a request including emotional information of the user.

[1612] (Claim 2)

[1613] A means for collecting questionnaires from users and classifying their personality and preference types;

[1614] 10. The system of claim 1, further comprising means for generating training data using survey results.

[1615] (Claim 3)

[1616] 10. The system of claim 1, further comprising means for generating personalized reviews of products or content in real time in response to a user request. [Explanation of symbols]

[1617] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for collecting user browsing history, purchase history, and review information; A means for classifying the user's personality and preference type based on the collected data; a means for training a generative AI model using the classified data; and A means for classifying all users into personality and preference types and generating user profiles; A means for selecting a corresponding generative AI model based on a user profile to generate a personalized review; The system includes means for displaying the generated reviews on a user terminal.

2. A means for collecting questionnaires from users and classifying their personality and preference types; The system of claim 1 further comprising means for generating training data using survey results.

3. 10. The system of claim 1, further comprising means for generating personalized reviews of products or content in real time in response to a user request.

Images