System

Abstract

A system is provided.SOLUTION: The system includes means for collecting behavior history and attribute information of a user, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating attributes and interests of the user, means for generating a supplementary prompt based on the estimated information, means for providing the generated supplementary prompt to a generative artificial intelligence to generate an optimal response to the user, means for transmitting a response result by the generative artificial intelligence to a terminal of the user, and means for collecting feedback from the user, analyzing the feedback, and storing the feedback in the database.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] Existing user response systems using generative AI have limited user information, making it difficult to provide responses optimized for each individual user. Furthermore, due to the diverse user demographic, general prompts alone cannot adequately meet user needs. Furthermore, the lack of a mechanism for effectively collecting and utilizing user feedback hinders progress in system improvement. [Means for solving the problem]

[0005] The present invention relates to a system that collects a user's behavioral history and attribute information, generates follow-up prompts based on the collected information, and provides them to a generative artificial intelligence. Specifically, the system includes a means for collecting a user's behavioral history and attribute information, a means for storing the collected data in a database, a means for analyzing the stored data and estimating the user's attributes and interests, and a means for generating follow-up prompts based on the estimated information. The system also includes a means for providing the generated follow-up prompts to the generative artificial intelligence and causing it to generate an optimal response for the user, a means for transmitting the generative artificial intelligence's response results to the user's device, and a means for collecting, analyzing, and storing user feedback in a database. This configuration enables highly accurate, personalized responses based on the user's attributes and interests.

[0006] "User" refers to any individual or legal entity that uses the System.

[0007] "Behavioral history" refers to a record of a user's accesses, clicks, and other actions on websites and applications.

[0008] "Attribute information" refers to basic information about users (age, gender, region, etc.) and areas of interest.

[0009] "Means of collection" refers to the methods and devices used to obtain user behavioral history and attribute information as data.

[0010] "Database" refers to a system for storing and managing collected data.

[0011] "Means of analysis" refers to the methods and algorithms used to analyze data stored in a database and identify customer attributes and interests.

[0012] "Supplemental prompts" refer to detailed instructions generated based on the user's specific information.

[0013] "Generative artificial intelligence" refers to software or systems that use natural language processing and machine learning to generate responses through a user interface.

[0014] "Response results" refers to answers or suggestions generated by the generative artificial intelligence based on the user's follow-up prompts.

[0015] "Feedback" refers to the user's evaluation or reaction to the response provided by the system. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] This invention relates to a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the supplementary prompts to a generation artificial intelligence.

[0038] Data Collection Module

[0039] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0040] Data Storage Module

[0041] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0042] Data Analysis Module

[0043] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to estimate the user's attributes (age, gender, region, etc.) and areas of interest with high accuracy. The analysis results are then saved back into the database and used as the basis for generating supplementary prompts.

[0044] Prompt Generation Module

[0045] The server obtains the demographic information and interest analysis results associated with a specific user, and then generates supplemental prompts based on this information. The supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[0046] AI-enabled module

[0047] The server provides the generated supplemental prompts to the AI ​​generator, which generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information tailored to the user's needs and interests.

[0048] The server formats the generated AI's response and sends it to the device, where the user can receive the most appropriate information and suggestions.

[0049] Feedback Collection Module

[0050] The device tracks the user's reactions to the AI's responses (e.g., clicks, viewing time, and feedback input), and the collected feedback data is sent to a server in real time.

[0051] The server analyzes the feedback data and stores the results in a database. These analysis results are used to generate prompts and improve the accuracy of AI responses in the future.

[0052] Specific examples

[0053] For example, consider a case where a travel agency uses this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[0054] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[0055] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has no history of staying in Okinawa in the past 10 years."

[0056] 3. In the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[0057] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "new activities at a beach resort in Okinawa."

[0058] 5. In the feedback collection module, the device collects the user's responses to the suggestions (e.g., clicks and feedback) and sends them to the server, which allows the system to improve the accuracy of prompt generation in the future.

[0059] In this way, this invention enables generative AI to provide more accurate and personalized responses based on user attributes and interest information.

[0060] The processing flow will be explained below.

[0061] Step 1:

[0062] The device tracks user actions (e.g., page views, clicks, and form fills) on websites and applications. It embeds JavaScript code and an SDK to collect user interaction data in real time.

[0063] Step 2:

[0064] The device periodically sends the collected behavioral data to the server in batch processing or real-time streaming, where the data is encrypted and processed to protect user privacy.

[0065] Step 3:

[0066] The server temporarily stores the received data in a cache and performs data reformatting and cleansing (e.g., removing duplicate data and correcting outliers).

[0067] Step 4:

[0068] The server stores the formatted data in a database, where the behavioral history and attribute information for each user are stored in the appropriate tables.

[0069] Step 5:

[0070] The server analyzes the stored data and uses machine learning algorithms to estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history).

[0071] Step 6:

[0072] The server stores the analysis results back in a database, making them available for subsequent prompt generation and AI response processes.

[0073] Step 7:

[0074] The server retrieves analytics related to a particular user and generates follow-up prompts based on that information, including specific instructions such as "travel interests."

[0075] Step 8:

[0076] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates an optimal response for the user based on the supplemental prompts.

[0077] Step 9:

[0078] Generative AI analyzes the supplemental prompts and generates specific and useful information and suggestions for the user, such as creating detailed suggestions for travel plans.

[0079] Step 10:

[0080] The server formats the response generated by the AI ​​and sends it to the user's device. The response is sent in HTML or JSON format, making it easy for the user to understand.

[0081] Step 11:

[0082] The user receives a response from the generative AI through their device, and the user can take action based on the response and provide feedback.

[0083] Step 12:

[0084] The device tracks the user's reactions to the responses (e.g., clicks, time spent, feedback input) and sends them to the server.

[0085] Step 13:

[0086] The server analyzes the collected feedback data and stores the results in a database. These analysis results are used to generate follow-up prompts and improve the accuracy of AI responses in the future.

[0087] In this way, advanced personalization based on user attribute information and interests can be achieved, enabling highly accurate user responses using generative AI.

[0088] Example 1

[0089] 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."

[0090] Conventional systems have had difficulty effectively collecting and analyzing user behavioral history and attribute information to provide highly accurate, personalized responses to users. Furthermore, they lacked the real-time nature of collected data and the ability to continuously improve the system using feedback data.

[0091] 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.

[0092] In this invention, the server includes means for collecting user behavior history and attribute information, means for temporarily caching the collected data and storing it in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests using a machine learning algorithm, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation artificial intelligence to generate an optimal response for the user, means for formatting the generated response results and sending them to the user's terminal, and means for collecting feedback from the user, analyzing it, and storing it in a database. This makes it possible to provide highly accurate personalized responses based on the user's attribute and interest information, and to continuously improve the system by utilizing the feedback data.

[0093] "User behavior history" refers to the history of operations such as page views, clicks, and form input when a user uses a website or application.

[0094] "Attribute information" refers to information that indicates a user's individual characteristics, such as their age, gender, region, and areas of interest.

[0095] "Temporary cache" refers to a temporary storage area for storing data for a short period of time, used to improve the efficiency of real-time processing.

[0096] A "database" is an information management system used to store and manage data in a structured way.

[0097] A "machine learning algorithm" is a computational method that analyzes data and learns patterns to make predictions and classify unknown data.

[0098] A "supplemental prompt" is a detailed instruction that allows the generative artificial intelligence to generate the optimal response for a particular user.

[0099] "Generative AI" refers to an AI engine that generates optimal responses based on input prompts.

[0100] "Feedback" refers to the reaction a user makes to a provided response (e.g., clicks, view time, feedback input).

[0101] "Real-time" refers to data collection and processing occurring immediately, with minimal delay.

[0102] This invention is a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the supplementary prompts to a generation artificial intelligence.

[0103] Data Collection Module

[0104] The server first obtains the user's consent. At this time, a pop-up screen requesting consent is displayed on the user's device. After consent is obtained, the device tracks the user's behavioral history (e.g., page views, clicks, form entries, etc.) and attribute information (e.g., age, gender, region, etc.) within websites and applications. The collected data is sent to the server in real time, and the server temporarily stores this data in a cache.

[0105] Data Storage Module

[0106] The server periodically transfers the cached data to a database, where it systematically organizes and stores each user's behavioral history and attribute information. Organizing the data speeds up subsequent data retrieval and analysis processes.

[0107] Data Analysis Module

[0108] The server analyzes the behavioral history and attribute information stored in the database using machine learning algorithms. This analysis uses a variety of machine learning algorithms (e.g., clustering, classification, regression analysis, etc.) to estimate the user's attributes and areas of interest with high accuracy. The analysis results are saved back in the database and used as the basis for generating supplementary prompts.

[0109] Prompt Generation Module

[0110] The server obtains the attribute information and interest analysis results related to a specific user and generates a supplemental prompt based on that information. The supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, a travel agency might use this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the AI ​​generator.

[0111] AI-enabled module

[0112] The server provides the generated supplemental prompts to the AI ​​generator, which generates an optimal response for the user based on the supplemental prompts. The generated response contains information that is specific and relevant to the user's needs and interests. The server formats the generated response and sends it to the user's device.

[0113] Feedback Collection Module

[0114] Users can receive optimal information and suggestions through the generated responses. The device tracks the user's reactions to the responses from the AI ​​generator (e.g., clicks, viewing time, feedback input, etc.) and sends the feedback data to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0115] Specific examples

[0116] For example, consider a travel agency using this system to input a prompt such as "Create a perfect travel plan for the user!" into the AI ​​generator. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server. This data is stored in a database. The server then analyzes the past data using the data analysis module to estimate attribute information such as "male, 50s, beach resort lover, no history of visiting Okinawa in the last 10 years." Next, through the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "male, 50s, beach resort lover, has not been to Okinawa in the last 10 years." Through the AI ​​support module, the server provides the supplemental prompt to the AI ​​generator, suggesting specific travel plans to the user, such as "new activities at beach resorts in Okinawa." Through the feedback collection module, the device collects the user's responses to the suggestions (clicks and feedback) and sends them to the server. This improves the accuracy of prompt generation in the future.

[0117] This invention enables generative artificial intelligence to provide more accurate and personalized responses based on user attributes and interest information.

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

[0119] Step 1: Obtaining User Consent

[0120] To obtain permission to collect the user's behavioral history and attribute information, the server displays a pop-up consent screen on the user's device. This screen explains the details and purpose of the data to be collected. When the user clicks the "Allow" button, the server recognizes that permission has been obtained.

[0121] Input: User interaction from the terminal

[0122] Output: Collection permission status

[0123] Step 2: Collecting behavioral history and attribute information

[0124] After permission is granted, the device will begin tracking the user's actions, using JavaScript code to record web page browsing history, clicks, form entries, and other information. The collected data is then sent to a server in real time.

[0125] Input: User operation data

[0126] Output: Operational data collected in real time

[0127] Step 3: Temporarily Caching Data

[0128] The server temporarily stores the user operation data transmitted in real time in a cache, which is used for immediate or batch processing of data.

[0129] Input: Operational data transmitted in real time

[0130] Output: Temporary cached data

[0131] Step 4: Store in the database

[0132] The server periodically transfers the cached data to a database, where it is organized by user and stored in a structured format. This process updates the data index, streamlining subsequent data retrieval and analysis processes.

[0133] Input: Cached operation data

[0134] Output: Structured data stored in a database

[0135] Step 5: Data analysis

[0136] The server analyzes the user's behavioral history and attribute information stored in the database, and uses machine learning algorithms to accurately estimate the user's attributes (age, gender, region, etc.) and areas of interest.

[0137] Input: Structured data stored in a database

[0138] Output: Inferred attributes and interest data as analysis results

[0139] Step 6: Generate supplemental prompts

[0140] The server generates supplemental prompts based on the analysis results, which are detailed instructions to provide user-specific information to the AI, including content based on the user's attributes and interests.

[0141] Input: Inferred attributes and interest data as analysis results

[0142] Output: Generated supplemental prompts

[0143] Step 7: Prompt the generative AI

[0144] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates an optimal response for the user based on the prompts.

[0145] Input: Generated supplemental prompt

[0146] Output: A response generated by the generative artificial intelligence

[0147] Step 8: Providing a response to the user

[0148] The server formats the generated response and sends it to the device, through which the user receives the most relevant information and offers.

[0149] Input: A response generated by generative artificial intelligence

[0150] Output: The formatted response sent to the user's terminal

[0151] Step 9: Gather user responses

[0152] The device tracks the user's reactions to the generated AI's responses (clicks, viewing time, feedback input, etc.), and the collected feedback data is sent to the server in real time.

[0153] Input: User response data

[0154] Output: Feedback data sent to the server

[0155] Step 10: Feedback analysis and system improvement

[0156] The server analyzes the feedback data and stores the results in a database. These analysis results are used to generate prompts and improve the accuracy of AI responses in the future.

[0157] Input: Feedback data

[0158] Output: Database containing analysis results, improving system performance

[0159] (Application example 1)

[0160] 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."

[0161] Current online shopping sites often provide users with uniform product suggestions and campaign information, lacking personalized suggestions tailored to each user's interests and purchasing history. This results in lower user satisfaction and a decrease in willingness to purchase. Furthermore, the inaccuracy of suggestions means that user feedback cannot be effectively utilized.

[0162] 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.

[0163] In this invention, the server includes means for collecting user behavior history and attribute information, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for sending the generation AI's response results to the user's terminal, means for collecting user feedback, analyzing it, and storing it in a database, means for making personalized product suggestions based on the user's behavior history and attribute information, and means for improving suggestion accuracy based on user feedback. This makes it possible to provide individualized product suggestions to users, improve user satisfaction, and increase purchasing motivation.

[0164] A "user" is an individual or group that uses the system, and whose behavioral history and attribute information is collected.

[0165] "Behavioral history" refers to the series of operations or actions a user performs on a website or application, including page views, clicks, and form entries.

[0166] "Attribute information" refers to individual information such as a user's age, gender, region, and interests, and is data collected along with behavioral history.

[0167] A "database" is a system for storing and managing collected user behavioral history and attribute information, enabling data organization and search.

[0168] "Generative AI" refers to artificial intelligence technology that generates optimal responses for users based on collected data and generated prompts.

[0169] A "prompt" is a detailed instruction that provides user-specific information to the artificial intelligence generating the system, and is also called a supplemental prompt.

[0170] "Feedback" refers to the user's reaction to the response from the AI ​​generator, including clicks, viewing time, and feedback input.

[0171] "Personalized product suggestions" refers to suggesting product information and campaign information that is optimized for each individual user based on the user's behavioral history and attribute information.

[0172] "Recommendation accuracy" refers to the ability to offer products and services that match a user's interests, and is improved based on user feedback.

[0173] This invention is a system that collects user behavior history and attribute information, generates supplemental prompts based on the analysis results, and uses artificial intelligence to provide the user with the optimal response. Specific embodiments for implementing this invention are described below.

[0174] Data Collection Module

[0175] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0176] Data Storage Module

[0177] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0178] Data Analysis Module

[0179] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses machine learning algorithms such as Python's Scikit-learn and TensorFlow to estimate the user's attributes (age, gender, region, etc.) and areas of interest with high accuracy. The analysis results are then saved back into the database and used as the basis for generating supplementary prompts.

[0180] Prompt Generation Module

[0181] The server obtains the attribute information and interest analysis results related to a specific user. It then generates a supplemental prompt based on this information. The supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, it could say, "For a female user in her 30s who is interested in fashion, please suggest new fashion collections and sales information. She has a history of three online purchases in the past six months and browsing history as a guest user."

[0182] AI-enabled module

[0183] The server provides the generated supplemental prompts to the generation AI. The generation AI generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information adapted to the user's needs and interests. The server formats the generation AI's response result and sends it to the terminal. The user can receive optimal information and suggestions through this response.

[0184] Feedback Collection Module

[0185] The device tracks the user's reactions to the AI's responses (e.g., clicks, viewing time, and feedback input). The collected feedback data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0186] A specific example is a personalized shopping assistant application on an online shopping site. Based on the user's behavioral history and attribute information, this application generates prompts such as, "Please suggest new fashion collections and sale information to a female user in her 30s who is interested in fashion. She has a history of three online shopping trips in the past six months and browsing history as a guest user," and the artificial intelligence then makes optimal product suggestions.

[0187] In this way, this invention enables generative AI to provide more accurate and personalized responses based on user attributes and interest information.

[0188] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0189] Step 1:

[0190] The server collects the user's behavioral history and attribute information. To do so, the device displays a pop-up message to request the user's permission. If permission is granted, the device tracks the user's behavior on websites and applications (page views, clicks, form input, etc.) and sends this data to the server in real time. The input is user behavior data, and the output is tracking data sent to the server.

[0191] Step 2:

[0192] The server temporarily caches the received data and then stores it in a database. At this time, the data is organized by user. The input is tracking data, and the output is behavioral history and attribute information stored in the database.

[0193] Step 3:

[0194] The server analyzes the user's behavioral history and attribute information stored in the database. Specifically, it uses machine learning algorithms (e.g., Python's Scikit-learn or TensorFlow) to estimate the user's attributes (age, gender, region, etc.) and areas of interest. In this analysis process, data is input into the algorithm, and the output is estimated attribute information and interest information.

[0195] Step 4:

[0196] The server generates a supplementary prompt based on the estimated attribute information and interest information. The supplementary prompt serves as a detailed instruction for the generation AI. The input is the estimated attribute information and interest information, and the output is a generated prompt sentence. Specific operations include forming a sentence based on the generation rules.

[0197] Step 5:

[0198] The server provides the generated supplemental prompt to the AI ​​generator, which receives the supplemental prompt and generates an optimal response for the user. In this process, a prompt sentence is input and the output is a generated response. Specific operations include the execution of a natural language processing model.

[0199] Step 6:

[0200] The server formats the generated response and sends it to the terminal. The user's terminal receives the response and displays it on its screen. In this process, the input is the generated response and the output is the response displayed on the user's terminal.

[0201] Step 7:

[0202] The device tracks the user's reactions to the responses from the AI ​​generator (e.g., clicks, viewing time, feedback input). The input is the user's reaction data, and the output is the feedback data sent to the server in real time.

[0203] Step 8:

[0204] The server analyzes the collected feedback data and stores the results in a database. The analysis results are used to generate prompts and improve the accuracy of AI responses from the next time onwards. The input is the feedback data, and the output is the analyzed feedback information. Specific operations include the execution of a feedback analysis algorithm.

[0205] 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.

[0206] This invention relates to a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the generated prompts to a generative artificial intelligence. Furthermore, the invention aims to further improve the accuracy of responses by combining an emotion engine that recognizes the user's emotions.

[0207] Data Collection Module

[0208] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0209] Data Storage Module

[0210] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0211] Data Analysis Module

[0212] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history) with high accuracy. It also uses an emotion engine to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are saved back in the database and serve as the basis for generating supplementary prompts.

[0213] Prompt Generation Module

[0214] The server obtains the demographic information, interests, and sentiment analysis results associated with a specific user, and then generates supplemental prompts based on this information. The supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[0215] AI-enabled module

[0216] The server provides the generated supplemental prompts to the generation AI. The generation AI generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information in a form that is adapted to the user's needs, interests, and even recognized emotions. The server formats the generation AI's response result and sends it to the device. The user can receive optimal information and suggestions through this response.

[0217] Feedback Collection Module

[0218] The device tracks the user's reactions to the AI's responses (e.g., clicks, time spent, and feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0219] Specific examples

[0220] For example, consider a case where a travel agency uses this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[0221] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[0222] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has not stayed in Okinawa in the last 10 years," and then uses an emotion engine to identify the user's recent emotional state.

[0223] 3. In the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[0224] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "New activities at beach resorts in Okinawa." These suggestions also take into account the user's emotions, so if the user feels like relaxing, for example, the server can choose the best resort for that time of year.

[0225] 5. In the feedback collection module, the device collects the user's reactions to the suggestions (e.g., clicks and feedback) and their emotions at the time, and sends them to the server. This allows the system to improve the accuracy of prompt generation in the future.

[0226] In this way, this invention enables generative AI to provide more accurate and personalized responses based on the user's attribute information, interests, and emotional information.

[0227] The processing flow will be explained below.

[0228] Step 1:

[0229] The device tracks data about user behavior on websites and applications (e.g., page views, clicks, and inputs). This process uses JavaScript code and SDKs to collect user interaction data in real time.

[0230] Step 2:

[0231] The device encrypts the tracked behavioral data and transmits it to a server over a secure protocol, either in batches or in real-time streaming.

[0232] Step 3:

[0233] The server temporarily caches the received behavioral data and then cleanses the cached data, for example, removing duplicate data and detecting invalid data.

[0234] Step 4:

[0235] The server stores the cleansed data in a database, where each user's behavioral history and attribute information are organized and saved.

[0236] Step 5:

[0237] The server analyzes the user's behavioral history and attribute information stored in the database, using machine learning algorithms to accurately estimate the user's attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history).

[0238] Step 6:

[0239] The server uses an emotion engine to recognize emotions from user behavior and feedback data, for example, by identifying emotions through text tone and facial expression analysis.

[0240] Step 7:

[0241] The server stores the analysis results and emotion recognition results in a database, which is used in the subsequent supplemental prompt generation step.

[0242] Step 8:

[0243] The server retrieves demographic information, interests, and sentiment analysis results related to a specific user from a database, and uses this information to generate detailed follow-up prompts.

[0244] Step 9:

[0245] The server provides the generated supplemental prompts to the artificial intelligence generating system, which generates an optimal response for the user based on the supplemental prompts.

[0246] Step 10:

[0247] Generative AI analyzes the supplemental prompts and generates specific and useful information and suggestions for the user, for example, specific suggestions based on the user's interests and emotions, rather than random recommendations for travel plans.

[0248] Step 11:

[0249] The server formats the response generated by the AI ​​and sends it to the user's device in a format suitable for the user interface (e.g., HTML, JSON).

[0250] Step 12:

[0251] The user receives a response from the AI ​​via their device, checks the response, and takes action based on it.

[0252] Step 13:

[0253] The device tracks the user's reactions to the responses (e.g., clicks, dwell time, feedback input) and their emotions at the time.

[0254] Step 14:

[0255] The device sends the collected feedback and emotion data to a server, where the data is encrypted to protect privacy.

[0256] Step 15:

[0257] The server receives the feedback data and emotion data and analyzes them again. The results of this analysis are stored in a database and used to generate prompts and improve the accuracy of AI responses in the future.

[0258] For example, when a travel agency provides a user with a suitable travel plan, it will take into account the user's past travel history, current interests, and emotional state through each of the above steps to provide personalized suggestions, which will improve user satisfaction and provide more effective services.

[0259] Example 2

[0260] 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."

[0261] Conventional information provision systems have had difficulty in providing personalized responses that take into account not only the user's behavioral history and attribute information, but also their emotions at the time. This has resulted in the inability to optimally respond to user needs and a decline in the quality of responses.

[0262] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user behavior history and attribute information, means for temporarily caching the collected data and then storing it in a database, means for updating the data stored in the database and indexes, means for analyzing the data stored in the database using a machine learning algorithm and estimating the user's attributes, interests, and emotions, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for transmitting the response results generated by the generation AI to the user's terminal, and means for collecting feedback from the user, analyzing it including emotional information, and storing it in a database. This makes it possible to provide highly accurate personalized responses that also take the user's emotional state into consideration.

[0263] "User behavior history" is a record of a series of operations and actions a user performs on a website or application.

[0264] "Attribute information" refers to information that indicates a user's personal characteristics, such as age, gender, region, hobbies, and purchasing history.

[0265] "Means of collection" refers to software or hardware mechanisms for capturing and recording user behavioral history and attribute information.

[0266] A "cache" is a high-speed accessible storage area for temporarily storing data.

[0267] A "database" is a system for efficiently storing, managing, and searching large amounts of data.

[0268] The "means for updating indexes" is a mechanism for creating and updating indexes of data in a database in order to improve search performance for the data.

[0269] A "machine learning algorithm" is an algorithm that learns patterns and rules from past data and applies them to new data.

[0270] "Analysis" refers to analyzing collected data and extracting meaningful information.

[0271] A "supplementary prompt" is a detailed instruction given to the artificial intelligence that is generated based on analyzed user information.

[0272] "Generative AI" is an AI technology that generates natural language responses and suggestions based on given data and instructions.

[0273] "Feedback" is a record of a user's reactions and opinions to the content provided by the system.

[0274] "Emotional information" is data that indicates the user's current emotional state and its changes.

[0275] This invention is a system that collects a user's behavioral history and attribute information, generates supplementary prompts based on the analysis results, and provides them to a generative artificial intelligence to provide the user with the optimal response. This system also aims to improve the accuracy of responses by combining it with an emotion engine that recognizes the user's emotions.

[0276] Data Collection Module

[0277] To collect user behavior history and attribute information, the server first obtains the user's consent. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, and form entries). This data collection is performed using JavaScript libraries and SDKs such as Google Analytics and Mixpanel. The collected data is sent to the server in real time, where it is temporarily cached and then stored in a database.

[0278] Data Storage Module

[0279] The server uses a database (e.g., MySQL, MongoDB) to move the collected data from the temporary cache to the database, organize the data, and update the data index to improve the search speed for each user's behavior history and attribute information.

[0280] Data Analysis Module

[0281] The server analyzes the user's behavioral history and attribute information stored in the database using machine learning algorithms (e.g., scikit-learn, TensorFlow) to accurately estimate the user's attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history). It also uses an emotion engine (e.g., IBM Watson, Affectiva) to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are saved back in the database and serve as the basis for generating supplementary prompts.

[0282] Prompt Generation Module

[0283] The server obtains attribute information, interests, and sentiment analysis results related to a specific user and generates supplemental prompts based on this information. Supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[0284] AI-enabled module

[0285] The server provides the generated supplemental prompts to a generative AI (e.g., OpenAI GPT-4, Google BERT). The generative AI generates the optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information that is adapted to the user's needs, interests, and even recognized emotions. The server formats the generative AI's response and sends it to the device. The user can receive optimal information and suggestions through this response.

[0286] Feedback Collection Module

[0287] The device collects the user's reactions to the responses from the AI ​​generator (e.g., clicks, time spent, feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes this feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0288] Specific examples

[0289] For example, consider a case where a travel agency uses this system to input a prompt such as "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[0290] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[0291] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has not stayed in Okinawa in the last 10 years," and the emotion engine identifies the user's recent emotional state.

[0292] 3. In the prompt generation module, the server generates a supplementary prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[0293] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "New activities at beach resorts in Okinawa." These suggestions also take into account the user's emotions, so if the user feels like relaxing, for example, the server can choose the best resort for that time of year.

[0294] 5. In the feedback collection module, the device collects the user's reactions to the suggestions (e.g., clicks and feedback) and their emotions at the time, and sends them to the server. This allows the system to improve the accuracy of prompt generation in the future.

[0295] In this way, generative AI can provide more accurate and personalized responses based on the user's attribute information, interests, and emotional information.

[0296] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0297] Step 1:

[0298] Obtain user consent.

[0299] The server first presents the user with a privacy policy and data terms of use, and obtains consent to collect behavioral history and attribute information. The input is the user's consent data, and the output is a confirmation flag of consent. Specifically, when the user clicks the consent button, the information is sent to the server.

[0300] Step 2:

[0301] Collect user behavioral history and attribute information.

[0302] The device tracks user behavior on websites and applications (e.g., page views, clicks, and form entries). The input is user behavior data, and the output is tracked behavior history data. Specifically, data is collected using JavaScript libraries and SDKs (e.g., Google Analytics, Mixpanel) and sent to a server in real time.

[0303] Step 3:

[0304] The collected data is temporarily cached and then stored in a database.

[0305] The server temporarily stores the received data in a cache (for example, using Redis or Memcached), and stores the data collected within a certain period in a database (for example, MySQL or MongoDB) in batch processing. The input is the collected behavioral history and attribute information, and the output is the data stored in the database. Specifically, the data is stored in the cache the moment it arrives at the server, and is then moved to the database in batch processing every hour.

[0306] Step 4:

[0307] Update the data index.

[0308] The server creates an index for the data stored in the database, enabling quick searches of each user's behavioral history and attribute information. The input is the data stored in the database, and the output is the updated index. Specifically, the server creates an index for newly stored data and updates the index.

[0309] Step 5:

[0310] User behavioral history and attribute information are analyzed using machine learning algorithms.

[0311] The server uses machine learning algorithms (e.g., scikit-learn, TensorFlow) to analyze the user's behavioral history and attribute information, and estimates attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history) with high accuracy. It also uses an emotion engine (e.g., IBM Watson, Affectiva) to recognize the user's emotions and analyze the emotions in the feedback data. The input is the behavioral history and attribute information stored in the database, and the output is the analyzed attribute information and emotion data. Specifically, the server schedules data analysis jobs and restores the analysis results to the database.

[0312] Step 6:

[0313] Generate follow-up prompts based on the analysis results.

[0314] The server obtains attribute information, interests, and the results of sentiment analysis related to a specific user, and generates a supplemental prompt based on this. The input is the analyzed attribute information and sentiment data, and the output is the generated supplemental prompt. Specifically, it generates a supplemental prompt based on the information "male in his 50s, likes beach resorts, hasn't been to Okinawa in the last 10 years."

[0315] Step 7:

[0316] Supplemental prompts are provided to the generative artificial intelligence to generate an optimal response.

[0317] The server provides the generated supplemental prompts to a generative artificial intelligence (e.g., OpenAI GPT-4, Google BERT). The AI ​​generates the optimal response for the user based on the supplemental prompts. The input is the supplemental prompt, and the output is the generated response. Specifically, the AI ​​generates the response, "Here are three suggestions for the latest activities you can enjoy at a beach resort in Okinawa."

[0318] Step 8:

[0319] The generated response is sent to the user's terminal.

[0320] The server formats the generated AI response and sends it to the device. The input is the generated response, and the output is the response delivered to the user. The specific behavior is to display detailed information about the suggested activity on a web page or application.

[0321] Step 9:

[0322] Collect and analyze user feedback.

[0323] The device tracks the user's reactions to the response from the AI ​​generator (e.g., clicks, time spent, feedback input) and their emotions at the time, and sends the collected data to the server in real time. The input is the user's reaction data, and the output is analyzed feedback. Specifically, if the user shows interest in a suggestion and clicks, that information and the timing of the click are tracked and sent to the server.

[0324] Step 10:

[0325] The feedback data is stored in a database and used to generate prompts in the future.

[0326] The server analyzes the feedback data and saves the results in a database. The input is the feedback data, and the output is the analysis results that will be used to generate future prompts. Specifically, if the analysis results of the feedback data indicate that "users are highly interested in information related to beach resorts," this information will be reflected in the generation of future prompts.

[0327] In this way, this system has a processing step that collects and analyzes the user's behavioral history, attribute information, and emotional information, and uses generative AI to provide highly accurate personalized responses.

[0328] (Application example 2)

[0329] 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."

[0330] Traditional personalization systems based on user behavior history and attribute information struggle to consider the user's emotional state, resulting in content and advertisements that are often not adapted to the user's current state. As a result, responses and suggestions to users are insufficient, limiting the effectiveness of advertisements and suggestions. Furthermore, there is a lack of effective methods for collecting and analyzing user feedback, making it difficult to continuously improve the system's personalization accuracy.

[0331] The identification processing 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 user behavior history and attribute information, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for transmitting the response results of the generation AI to the user's terminal, means for collecting feedback from the user, analyzing the feedback, and storing it in a database, means for recognizing the user's emotional state, means for generating advertisements based on the user's behavior history, attribute information, and emotional state, means for delivering the generated advertisements to the user's terminal, and means for collecting the user's reactions to advertisements and their emotional state. This enables highly accurate personalized responses and advertisement delivery based on the user's behavior history, attribute information, and emotional state.

[0332] "User behavior" is a record of specific actions (e.g., page views, clicks, form fills, etc.) that a user takes on a website or application.

[0333] "Descriptive information" means personal characteristics or profile information associated with a particular user (e.g., age, gender, region, interests, etc.).

[0334] A "database" is a collection of information that is organized so that collected data can be efficiently stored, managed, and retrieved.

[0335] "Supplemental prompts" are detailed instructions provided to the generative AI model that are generated based on the user's specific behavioral history and attribute information.

[0336] "Generative AI" refers to an AI system that generates appropriate responses or content based on provided prompts.

[0337] An "emotional state" is the specific emotional state (e.g., joy, sadness, stress, etc.) that a user is experiencing at a particular moment.

[0338] "Advertisement" means promotional content that is intended to inform or attract the attention of users about a particular product or service.

[0339] "Feedback" refers to the reactions and opinions (e.g., clicks, time spent, direct comments, etc.) that users have toward generated AI and advertisements.

[0340] The following configurations are possible as embodiments of the present invention. The present invention is a system that collects and analyzes a user's behavioral history and attribute information, and generates and delivers optimal advertisements taking into account their emotional state. This system is based on the interaction between a server, a terminal, and the user.

[0341] Data collection

[0342] The server first obtains the user's consent to collect the user's behavioral history and attribute information. This allows the data to be collected lawfully while protecting the user's privacy. The device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and transmits the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0343] Data Storage

[0344] The server moves the collected data from the temporary cache to a database and organizes it. The database organizes and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0345] Data analysis

[0346] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to accurately estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history). It also uses an emotion engine to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are stored back in the database and serve as the basis for generating supplementary prompts.

[0347] Supplemental prompt generation

[0348] The server obtains demographic information, interests, and sentiment analysis results related to a specific user. It then generates a supplemental prompt based on this information. A supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, a prompt might read, "Generate an ad for this user, a 30-year-old male user, interested in technology and sports, whose current emotional state is stressed. For this user, please generate an ad introducing the latest gadgets that can help relieve stress."

[0349] Generative Artificial Intelligence

[0350] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates optimal advertisements and content for the user based on the supplemental prompts. The generated advertisements and content contain specific and useful information in a form adapted to the user's needs, interests, and even recognized emotions.

[0351] Ad serving

[0352] The server formats the response generated by the artificial intelligence and sends it to the user's device, where the user can receive the most suitable advertisements and suggestions.

[0353] Feedback collection

[0354] The device tracks the user's reactions to the AI's responses (e.g., clicks, time spent, and feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0355] This allows for highly accurate personalized responses and ad delivery based on users' behavioral history, attribute information, and emotional state, while also effectively collecting user feedback to continuously improve the system.

[0356] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0357] Step 1: Collect user data

[0358] Input: User behavior history and attribute information

[0359] How it works: To collect user behavior history and attribute information, the server first obtains the user's consent. The device then tracks the user's actions on websites and applications (e.g., page views, clicks, and form entries) in real time and sends the data to the server.

[0360] Output: Collected user behavior history and attribute information data

[0361] Step 2: Store the data

[0362] Input: Collected user data

[0363] Specific operation: The server temporarily caches the received user behavior history and attribute information, then organizes the data and stores it in a database. In the database, the data for each user is systematically stored.

[0364] Output: User data stored in the database

[0365] Step 3: Data analysis

[0366] Input: User data stored in a database

[0367] Specific operation: The server analyzes the user's behavioral history and attribute information stored in the database. Using machine learning algorithms, it estimates the user's attributes (age, gender, region, etc.) and areas of interest (hobbies, purchasing history, etc.) with high accuracy. It also uses an emotion engine to recognize the user's emotional state and analyze the emotions in the feedback data.

[0368] Output: Analysis results including user attributes, interests, and emotional state

[0369] Step 4: Generate supplemental prompts

[0370] Input: Analysis results (user attributes, interests, emotional state)

[0371] Specific behavior: The server generates a supplemental prompt based on the analysis results, including user-specific information. For example, a specific prompt might be generated: "A 30-year-old male user with an interest in technology and sports, whose current emotional state is stress. Please generate an advertisement for this user introducing the latest gadgets that can help relieve stress."

[0372] Output: Generated supplemental prompts

[0373] Step 5: Generative AI generates a response

[0374] Input: Supplementary prompt

[0375] Specific operation: The server provides the generated supplemental prompts to the AI ​​generator, which generates optimal advertisements and content for the user based on the provided prompts.

[0376] Output: Generated ads and content

[0377] Step 6: Sending the response results

[0378] Input: Generated ads and content

[0379] Specific operation: The server formats the generated advertisement and content and sends it to the user's device. The user receives the response and views it.

[0380] Output: Ads and content sent to the user's device

[0381] Step 7: Gather feedback

[0382] Input: User's reaction and emotional state

[0383] Specific operation: The device tracks the user's reactions to ads and responses from the generated AI (e.g., clicks, time spent, feedback input) and their emotional state at the time in real time, and sends the data to the server.

[0384] Output: Collected feedback and emotional state data

[0385] Step 8: Analyze the feedback data

[0386] Input: Feedback data and emotional state data

[0387] Specific operation: The server analyzes the collected feedback data and emotional state data and stores the results in a database. The analysis results are used to generate supplementary prompts and improve the accuracy of AI responses in the future.

[0388] Output: Parsed feedback data

[0389] 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.

[0390] 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.

[0391] 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.

[0392] [Second embodiment]

[0393] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0394] 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.

[0395] 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).

[0396] 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.

[0397] 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.

[0398] 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).

[0399] 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.

[0400] 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.

[0401] 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.

[0402] 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.

[0403] 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.

[0404] 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."

[0405] This invention relates to a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the supplementary prompts to a generation artificial intelligence.

[0406] Data Collection Module

[0407] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0408] Data Storage Module

[0409] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0410] Data Analysis Module

[0411] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to estimate the user's attributes (age, gender, region, etc.) and areas of interest with high accuracy. The analysis results are then saved back into the database and used as the basis for generating supplementary prompts.

[0412] Prompt Generation Module

[0413] The server obtains the demographic information and interest analysis results associated with a specific user, and then generates supplemental prompts based on this information. The supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[0414] AI-enabled module

[0415] The server provides the generated supplemental prompts to the AI ​​generator, which generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information tailored to the user's needs and interests.

[0416] The server formats the generated AI's response and sends it to the device, where the user can receive the most appropriate information and suggestions.

[0417] Feedback Collection Module

[0418] The device tracks the user's reactions to the AI's responses (e.g., clicks, viewing time, and feedback input), and the collected feedback data is sent to a server in real time.

[0419] The server analyzes the feedback data and stores the results in a database. These analysis results are used to generate prompts and improve the accuracy of AI responses in the future.

[0420] Specific examples

[0421] For example, consider a case where a travel agency uses this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[0422] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[0423] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has no history of staying in Okinawa in the past 10 years."

[0424] 3. In the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[0425] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "new activities at a beach resort in Okinawa."

[0426] 5. In the feedback collection module, the device collects the user's responses to the suggestions (e.g., clicks and feedback) and sends them to the server, which allows the system to improve the accuracy of prompt generation in the future.

[0427] In this way, this invention enables generative AI to provide more accurate and personalized responses based on user attributes and interest information.

[0428] The processing flow will be explained below.

[0429] Step 1:

[0430] The device tracks user actions (e.g., page views, clicks, and form fills) on websites and applications. It embeds JavaScript code and an SDK to collect user interaction data in real time.

[0431] Step 2:

[0432] The device periodically sends the collected behavioral data to the server in batch processing or real-time streaming, where the data is encrypted and processed to protect user privacy.

[0433] Step 3:

[0434] The server temporarily stores the received data in a cache and performs data reformatting and cleansing (e.g., removing duplicate data and correcting outliers).

[0435] Step 4:

[0436] The server stores the formatted data in a database, where the behavioral history and attribute information for each user are stored in the appropriate tables.

[0437] Step 5:

[0438] The server analyzes the stored data and uses machine learning algorithms to estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history).

[0439] Step 6:

[0440] The server stores the analysis results back in a database, making them available for subsequent prompt generation and AI response processes.

[0441] Step 7:

[0442] The server retrieves analytics related to a particular user and generates follow-up prompts based on that information, including specific instructions such as "travel interests."

[0443] Step 8:

[0444] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates an optimal response for the user based on the supplemental prompts.

[0445] Step 9:

[0446] Generative AI analyzes the supplemental prompts and generates specific and useful information and suggestions for the user, such as creating detailed suggestions for travel plans.

[0447] Step 10:

[0448] The server formats the response generated by the AI ​​and sends it to the user's device. The response is sent in HTML or JSON format, making it easy for the user to understand.

[0449] Step 11:

[0450] The user receives a response from the generative AI through their device, and the user can take action based on the response and provide feedback.

[0451] Step 12:

[0452] The device tracks the user's reactions to the responses (e.g., clicks, time spent, feedback input) and sends them to the server.

[0453] Step 13:

[0454] The server analyzes the collected feedback data and stores the results in a database. These analysis results are used to generate follow-up prompts and improve the accuracy of AI responses in the future.

[0455] In this way, advanced personalization based on user attribute information and interests can be achieved, enabling highly accurate user responses using generative AI.

[0456] Example 1

[0457] 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."

[0458] Conventional systems have had difficulty effectively collecting and analyzing user behavioral history and attribute information to provide highly accurate, personalized responses to users. Furthermore, they lacked the real-time nature of collected data and the ability to continuously improve the system using feedback data.

[0459] 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.

[0460] In this invention, the server includes means for collecting user behavior history and attribute information, means for temporarily caching the collected data and storing it in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests using a machine learning algorithm, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation artificial intelligence to generate an optimal response for the user, means for formatting the generated response results and sending them to the user's terminal, and means for collecting feedback from the user, analyzing it, and storing it in a database. This makes it possible to provide highly accurate personalized responses based on the user's attribute and interest information, and to continuously improve the system by utilizing the feedback data.

[0461] "User behavior history" refers to the history of operations such as page views, clicks, and form input when a user uses a website or application.

[0462] "Attribute information" refers to information that indicates a user's individual characteristics, such as their age, gender, region, and areas of interest.

[0463] "Temporary cache" refers to a temporary storage area for storing data for a short period of time, used to improve the efficiency of real-time processing.

[0464] A "database" is an information management system used to store and manage data in a structured way.

[0465] A "machine learning algorithm" is a computational method that analyzes data and learns patterns to make predictions and classify unknown data.

[0466] A "supplemental prompt" is a detailed instruction that allows the generative artificial intelligence to generate the optimal response for a particular user.

[0467] "Generative AI" refers to an AI engine that generates optimal responses based on input prompts.

[0468] "Feedback" refers to the reaction a user makes to a provided response (e.g., clicks, view time, feedback input).

[0469] "Real-time" refers to data collection and processing occurring immediately, with minimal delay.

[0470] This invention is a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the supplementary prompts to a generation artificial intelligence.

[0471] Data Collection Module

[0472] The server first obtains the user's consent. At this time, a pop-up screen requesting consent is displayed on the user's device. After consent is obtained, the device tracks the user's behavioral history (e.g., page views, clicks, form entries, etc.) and attribute information (e.g., age, gender, region, etc.) within websites and applications. The collected data is sent to the server in real time, and the server temporarily stores this data in a cache.

[0473] Data Storage Module

[0474] The server periodically transfers the cached data to a database, where it systematically organizes and stores each user's behavioral history and attribute information. Organizing the data speeds up subsequent data retrieval and analysis processes.

[0475] Data Analysis Module

[0476] The server analyzes the behavioral history and attribute information stored in the database using machine learning algorithms. This analysis uses a variety of machine learning algorithms (e.g., clustering, classification, regression analysis, etc.) to estimate the user's attributes and areas of interest with high accuracy. The analysis results are saved back in the database and used as the basis for generating supplementary prompts.

[0477] Prompt Generation Module

[0478] The server obtains the attribute information and interest analysis results related to a specific user and generates a supplemental prompt based on that information. The supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, a travel agency might use this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the AI ​​generator.

[0479] AI-enabled module

[0480] The server provides the generated supplemental prompts to the AI ​​generator, which generates an optimal response for the user based on the supplemental prompts. The generated response contains information that is specific and relevant to the user's needs and interests. The server formats the generated response and sends it to the user's device.

[0481] Feedback Collection Module

[0482] Users can receive optimal information and suggestions through the generated responses. The device tracks the user's reactions to the responses from the AI ​​generator (e.g., clicks, viewing time, feedback input, etc.) and sends the feedback data to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0483] Specific examples

[0484] For example, consider a travel agency using this system to input a prompt such as "Create a perfect travel plan for the user!" into the AI ​​generator. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server. This data is stored in a database. The server then analyzes the past data using the data analysis module to estimate attribute information such as "male, 50s, beach resort lover, no history of visiting Okinawa in the last 10 years." Next, through the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "male, 50s, beach resort lover, has not been to Okinawa in the last 10 years." Through the AI ​​support module, the server provides the supplemental prompt to the AI ​​generator, suggesting specific travel plans to the user, such as "new activities at beach resorts in Okinawa." Through the feedback collection module, the device collects the user's responses to the suggestions (clicks and feedback) and sends them to the server. This improves the accuracy of prompt generation in the future.

[0485] This invention enables generative artificial intelligence to provide more accurate and personalized responses based on user attributes and interest information.

[0486] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0487] Step 1: Obtaining User Consent

[0488] To obtain permission to collect the user's behavioral history and attribute information, the server displays a pop-up consent screen on the user's device. This screen explains the details and purpose of the data to be collected. When the user clicks the "Allow" button, the server recognizes that permission has been obtained.

[0489] Input: User interaction from the terminal

[0490] Output: Collection permission status

[0491] Step 2: Collecting behavioral history and attribute information

[0492] After permission is granted, the device will begin tracking the user's actions, using JavaScript code to record web page browsing history, clicks, form entries, and other information. The collected data is then sent to a server in real time.

[0493] Input: User operation data

[0494] Output: Operational data collected in real time

[0495] Step 3: Temporarily Caching Data

[0496] The server temporarily stores the user operation data transmitted in real time in a cache, which is used for immediate or batch processing of data.

[0497] Input: Operational data transmitted in real time

[0498] Output: Temporary cached data

[0499] Step 4: Store in the database

[0500] The server periodically transfers the cached data to a database, where it is organized by user and stored in a structured format. This process updates the data index, streamlining subsequent data retrieval and analysis processes.

[0501] Input: Cached operation data

[0502] Output: Structured data stored in a database

[0503] Step 5: Data analysis

[0504] The server analyzes the user's behavioral history and attribute information stored in the database, and uses machine learning algorithms to accurately estimate the user's attributes (age, gender, region, etc.) and areas of interest.

[0505] Input: Structured data stored in a database

[0506] Output: Inferred attributes and interest data as analysis results

[0507] Step 6: Generate supplemental prompts

[0508] The server generates supplemental prompts based on the analysis results, which are detailed instructions to provide user-specific information to the AI, including content based on the user's attributes and interests.

[0509] Input: Inferred attributes and interest data as analysis results

[0510] Output: Generated supplemental prompts

[0511] Step 7: Prompt the generative AI

[0512] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates an optimal response for the user based on the prompts.

[0513] Input: Generated supplemental prompt

[0514] Output: A response generated by the generative artificial intelligence

[0515] Step 8: Providing a response to the user

[0516] The server formats the generated response and sends it to the device, through which the user receives the most relevant information and offers.

[0517] Input: A response generated by generative artificial intelligence

[0518] Output: The formatted response sent to the user's terminal

[0519] Step 9: Gather user responses

[0520] The device tracks the user's reactions to the generated AI's responses (clicks, viewing time, feedback input, etc.), and the collected feedback data is sent to the server in real time.

[0521] Input: User response data

[0522] Output: Feedback data sent to the server

[0523] Step 10: Feedback analysis and system improvement

[0524] The server analyzes the feedback data and stores the results in a database. These analysis results are used to generate prompts and improve the accuracy of AI responses in the future.

[0525] Input: Feedback data

[0526] Output: Database containing analysis results, improving system performance

[0527] (Application example 1)

[0528] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0529] Current online shopping sites often provide users with uniform product suggestions and campaign information, lacking personalized suggestions tailored to each user's interests and purchasing history. This results in lower user satisfaction and a decrease in willingness to purchase. Furthermore, the inaccuracy of suggestions means that user feedback cannot be effectively utilized.

[0530] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0531] In this invention, the server includes means for collecting user behavior history and attribute information, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for sending the generation AI's response results to the user's terminal, means for collecting user feedback, analyzing it, and storing it in a database, means for making personalized product suggestions based on the user's behavior history and attribute information, and means for improving suggestion accuracy based on user feedback. This makes it possible to provide individualized product suggestions to users, improve user satisfaction, and increase purchasing motivation.

[0532] A "user" is an individual or group that uses the system, and whose behavioral history and attribute information is collected.

[0533] "Behavioral history" refers to the series of operations or actions a user performs on a website or application, including page views, clicks, and form entries.

[0534] "Attribute information" refers to individual information such as a user's age, gender, region, and interests, and is data collected along with behavioral history.

[0535] A "database" is a system for storing and managing collected user behavioral history and attribute information, enabling data organization and search.

[0536] "Generative AI" refers to artificial intelligence technology that generates optimal responses for users based on collected data and generated prompts.

[0537] A "prompt" is a detailed instruction that provides user-specific information to the artificial intelligence generating the system, and is also called a supplemental prompt.

[0538] "Feedback" refers to the user's reaction to the response from the AI ​​generator, including clicks, viewing time, and feedback input.

[0539] "Personalized product suggestions" refers to suggesting product information and campaign information that is optimized for each individual user based on the user's behavioral history and attribute information.

[0540] "Recommendation accuracy" refers to the ability to offer products and services that match a user's interests, and is improved based on user feedback.

[0541] This invention is a system that collects user behavior history and attribute information, generates supplemental prompts based on the analysis results, and uses artificial intelligence to provide the user with the optimal response. Specific embodiments for implementing this invention are described below.

[0542] Data Collection Module

[0543] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0544] Data Storage Module

[0545] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0546] Data Analysis Module

[0547] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses machine learning algorithms such as Python's Scikit-learn and TensorFlow to estimate the user's attributes (age, gender, region, etc.) and areas of interest with high accuracy. The analysis results are then saved back into the database and used as the basis for generating supplementary prompts.

[0548] Prompt Generation Module

[0549] The server obtains the attribute information and interest analysis results related to a specific user. It then generates a supplemental prompt based on this information. The supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, it could say, "For a female user in her 30s who is interested in fashion, please suggest new fashion collections and sales information. She has a history of three online purchases in the past six months and browsing history as a guest user."

[0550] AI-enabled module

[0551] The server provides the generated supplemental prompts to the generation AI. The generation AI generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information adapted to the user's needs and interests. The server formats the generation AI's response result and sends it to the terminal. The user can receive optimal information and suggestions through this response.

[0552] Feedback Collection Module

[0553] The device tracks the user's reactions to the AI's responses (e.g., clicks, viewing time, and feedback input). The collected feedback data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0554] A specific example is a personalized shopping assistant application on an online shopping site. Based on the user's behavioral history and attribute information, this application generates prompts such as, "Please suggest new fashion collections and sale information to a female user in her 30s who is interested in fashion. She has a history of three online shopping trips in the past six months and browsing history as a guest user," and the artificial intelligence then makes optimal product suggestions.

[0555] In this way, this invention enables generative AI to provide more accurate and personalized responses based on user attributes and interest information.

[0556] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0557] Step 1:

[0558] The server collects the user's behavioral history and attribute information. To do so, the device displays a pop-up message to request the user's permission. If permission is granted, the device tracks the user's behavior on websites and applications (page views, clicks, form input, etc.) and sends this data to the server in real time. The input is user behavior data, and the output is tracking data sent to the server.

[0559] Step 2:

[0560] The server temporarily caches the received data and then stores it in a database. At this time, the data is organized by user. The input is tracking data, and the output is behavioral history and attribute information stored in the database.

[0561] Step 3:

[0562] The server analyzes the user's behavioral history and attribute information stored in the database. Specifically, it uses machine learning algorithms (e.g., Python's Scikit-learn or TensorFlow) to estimate the user's attributes (age, gender, region, etc.) and areas of interest. In this analysis process, data is input into the algorithm, and the output is estimated attribute information and interest information.

[0563] Step 4:

[0564] The server generates a supplementary prompt based on the estimated attribute information and interest information. The supplementary prompt serves as a detailed instruction for the generation AI. The input is the estimated attribute information and interest information, and the output is a generated prompt sentence. Specific operations include forming a sentence based on the generation rules.

[0565] Step 5:

[0566] The server provides the generated supplemental prompt to the AI ​​generator, which receives the supplemental prompt and generates an optimal response for the user. In this process, a prompt sentence is input and the output is a generated response. Specific operations include the execution of a natural language processing model.

[0567] Step 6:

[0568] The server formats the generated response and sends it to the terminal. The user's terminal receives the response and displays it on its screen. In this process, the input is the generated response and the output is the response displayed on the user's terminal.

[0569] Step 7:

[0570] The device tracks the user's reactions to the responses from the AI ​​generator (e.g., clicks, viewing time, feedback input). The input is the user's reaction data, and the output is the feedback data sent to the server in real time.

[0571] Step 8:

[0572] The server analyzes the collected feedback data and stores the results in a database. The analysis results are used to generate prompts and improve the accuracy of AI responses from the next time onwards. The input is the feedback data, and the output is the analyzed feedback information. Specific operations include the execution of a feedback analysis algorithm.

[0573] 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.

[0574] This invention relates to a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the generated prompts to a generative artificial intelligence. Furthermore, the invention aims to further improve the accuracy of responses by combining an emotion engine that recognizes the user's emotions.

[0575] Data Collection Module

[0576] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0577] Data Storage Module

[0578] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0579] Data Analysis Module

[0580] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history) with high accuracy. It also uses an emotion engine to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are saved back in the database and serve as the basis for generating supplementary prompts.

[0581] Prompt Generation Module

[0582] The server obtains the demographic information, interests, and sentiment analysis results associated with a specific user, and then generates supplemental prompts based on this information. The supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[0583] AI-enabled module

[0584] The server provides the generated supplemental prompts to the generation AI. The generation AI generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information in a form that is adapted to the user's needs, interests, and even recognized emotions. The server formats the generation AI's response result and sends it to the device. The user can receive optimal information and suggestions through this response.

[0585] Feedback Collection Module

[0586] The device tracks the user's reactions to the AI's responses (e.g., clicks, time spent, and feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0587] Specific examples

[0588] For example, consider a case where a travel agency uses this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[0589] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[0590] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has not stayed in Okinawa in the last 10 years," and then uses an emotion engine to identify the user's recent emotional state.

[0591] 3. In the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[0592] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "New activities at beach resorts in Okinawa." These suggestions also take into account the user's emotions, so if the user feels like relaxing, for example, the server can choose the best resort for that time of year.

[0593] 5. In the feedback collection module, the device collects the user's reactions to the suggestions (e.g., clicks and feedback) and their emotions at the time, and sends them to the server. This allows the system to improve the accuracy of prompt generation in the future.

[0594] In this way, this invention enables generative AI to provide more accurate and personalized responses based on the user's attribute information, interests, and emotional information.

[0595] The processing flow will be explained below.

[0596] Step 1:

[0597] The device tracks data about user behavior on websites and applications (e.g., page views, clicks, and inputs). This process uses JavaScript code and SDKs to collect user interaction data in real time.

[0598] Step 2:

[0599] The device encrypts the tracked behavioral data and transmits it to a server over a secure protocol, either in batches or in real-time streaming.

[0600] Step 3:

[0601] The server temporarily caches the received behavioral data and then cleanses the cached data, for example, removing duplicate data and detecting invalid data.

[0602] Step 4:

[0603] The server stores the cleansed data in a database, where each user's behavioral history and attribute information are organized and saved.

[0604] Step 5:

[0605] The server analyzes the user's behavioral history and attribute information stored in the database, using machine learning algorithms to accurately estimate the user's attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history).

[0606] Step 6:

[0607] The server uses an emotion engine to recognize emotions from user behavior and feedback data, for example, by identifying emotions through text tone and facial expression analysis.

[0608] Step 7:

[0609] The server stores the analysis results and emotion recognition results in a database, which is used in the subsequent supplemental prompt generation step.

[0610] Step 8:

[0611] The server retrieves demographic information, interests, and sentiment analysis results related to a specific user from a database, and uses this information to generate detailed follow-up prompts.

[0612] Step 9:

[0613] The server provides the generated supplemental prompts to the artificial intelligence generating system, which generates an optimal response for the user based on the supplemental prompts.

[0614] Step 10:

[0615] Generative AI analyzes the supplemental prompts and generates specific and useful information and suggestions for the user, for example, specific suggestions based on the user's interests and emotions, rather than random recommendations for travel plans.

[0616] Step 11:

[0617] The server formats the response generated by the AI ​​and sends it to the user's device in a format suitable for the user interface (e.g., HTML, JSON).

[0618] Step 12:

[0619] The user receives a response from the AI ​​via their device, checks the response, and takes action based on it.

[0620] Step 13:

[0621] The device tracks the user's reactions to the responses (e.g., clicks, dwell time, feedback input) and their emotions at the time.

[0622] Step 14:

[0623] The device sends the collected feedback and emotion data to a server, where the data is encrypted to protect privacy.

[0624] Step 15:

[0625] The server receives the feedback data and emotion data and analyzes them again. The results of this analysis are stored in a database and used to generate prompts and improve the accuracy of AI responses in the future.

[0626] For example, when a travel agency provides a user with a suitable travel plan, it will take into account the user's past travel history, current interests, and emotional state through each of the above steps to provide personalized suggestions, which will improve user satisfaction and provide more effective services.

[0627] Example 2

[0628] 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."

[0629] Conventional information provision systems have had difficulty in providing personalized responses that take into account not only the user's behavioral history and attribute information, but also their emotions at the time. This has resulted in the inability to optimally respond to user needs and a decline in the quality of responses.

[0630] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user behavior history and attribute information, means for temporarily caching the collected data and then storing it in a database, means for updating the data stored in the database and indexes, means for analyzing the data stored in the database using a machine learning algorithm and estimating the user's attributes, interests, and emotions, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for transmitting the response results generated by the generation AI to the user's terminal, and means for collecting feedback from the user, analyzing it including emotional information, and storing it in a database. This makes it possible to provide highly accurate personalized responses that also take the user's emotional state into consideration.

[0631] "User behavior history" is a record of a series of operations and actions a user performs on a website or application.

[0632] "Attribute information" refers to information that indicates a user's personal characteristics, such as age, gender, region, hobbies, and purchasing history.

[0633] "Means of collection" refers to software or hardware mechanisms for capturing and recording user behavioral history and attribute information.

[0634] A "cache" is a high-speed accessible storage area for temporarily storing data.

[0635] A "database" is a system for efficiently storing, managing, and searching large amounts of data.

[0636] The "means for updating indexes" is a mechanism for creating and updating indexes of data in a database in order to improve search performance for the data.

[0637] A "machine learning algorithm" is an algorithm that learns patterns and rules from past data and applies them to new data.

[0638] "Analysis" refers to analyzing collected data and extracting meaningful information.

[0639] A "supplementary prompt" is a detailed instruction given to the artificial intelligence that is generated based on analyzed user information.

[0640] "Generative AI" is an AI technology that generates natural language responses and suggestions based on given data and instructions.

[0641] "Feedback" is a record of a user's reactions and opinions to the content provided by the system.

[0642] "Emotional information" is data that indicates the user's current emotional state and its changes.

[0643] This invention is a system that collects a user's behavioral history and attribute information, generates supplementary prompts based on the analysis results, and provides them to a generative artificial intelligence to provide the user with the optimal response. This system also aims to improve the accuracy of responses by combining it with an emotion engine that recognizes the user's emotions.

[0644] Data Collection Module

[0645] To collect user behavior history and attribute information, the server first obtains the user's consent. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, and form entries). This data collection is performed using JavaScript libraries and SDKs such as Google Analytics and Mixpanel. The collected data is sent to the server in real time, where it is temporarily cached and then stored in a database.

[0646] Data Storage Module

[0647] The server uses a database (e.g., MySQL, MongoDB) to move the collected data from the temporary cache to the database, organize the data, and update the data index to improve the search speed for each user's behavior history and attribute information.

[0648] Data Analysis Module

[0649] The server analyzes the user's behavioral history and attribute information stored in the database using machine learning algorithms (e.g., scikit-learn, TensorFlow) to accurately estimate the user's attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history). It also uses an emotion engine (e.g., IBM Watson, Affectiva) to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are saved back in the database and serve as the basis for generating supplementary prompts.

[0650] Prompt Generation Module

[0651] The server obtains attribute information, interests, and sentiment analysis results related to a specific user and generates supplemental prompts based on this information. Supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[0652] AI-enabled module

[0653] The server provides the generated supplemental prompts to a generative AI (e.g., OpenAI GPT-4, Google BERT). The generative AI generates the optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information that is adapted to the user's needs, interests, and even recognized emotions. The server formats the generative AI's response and sends it to the device. The user can receive optimal information and suggestions through this response.

[0654] Feedback Collection Module

[0655] The device collects the user's reactions to the responses from the AI ​​generator (e.g., clicks, time spent, feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes this feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0656] Specific examples

[0657] For example, consider a case where a travel agency uses this system to input a prompt such as "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[0658] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[0659] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has not stayed in Okinawa in the last 10 years," and the emotion engine identifies the user's recent emotional state.

[0660] 3. In the prompt generation module, the server generates a supplementary prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[0661] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "New activities at beach resorts in Okinawa." These suggestions also take into account the user's emotions, so if the user feels like relaxing, for example, the server can choose the best resort for that time of year.

[0662] 5. In the feedback collection module, the device collects the user's reactions to the suggestions (e.g., clicks and feedback) and their emotions at the time, and sends them to the server. This allows the system to improve the accuracy of prompt generation in the future.

[0663] In this way, generative AI can provide more accurate and personalized responses based on the user's attribute information, interests, and emotional information.

[0664] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0665] Step 1:

[0666] Obtain user consent.

[0667] The server first presents the user with a privacy policy and data terms of use, and obtains consent to collect behavioral history and attribute information. The input is the user's consent data, and the output is a confirmation flag of consent. Specifically, when the user clicks the consent button, the information is sent to the server.

[0668] Step 2:

[0669] Collect user behavioral history and attribute information.

[0670] The device tracks user behavior on websites and applications (e.g., page views, clicks, and form entries). The input is user behavior data, and the output is tracked behavior history data. Specifically, data is collected using JavaScript libraries and SDKs (e.g., Google Analytics, Mixpanel) and sent to a server in real time.

[0671] Step 3:

[0672] The collected data is temporarily cached and then stored in a database.

[0673] The server temporarily stores the received data in a cache (for example, using Redis or Memcached), and stores the data collected within a certain period in a database (for example, MySQL or MongoDB) in batch processing. The input is the collected behavioral history and attribute information, and the output is the data stored in the database. Specifically, the data is stored in the cache the moment it arrives at the server, and is then moved to the database in batch processing every hour.

[0674] Step 4:

[0675] Update the data index.

[0676] The server creates an index for the data stored in the database, enabling quick searches of each user's behavioral history and attribute information. The input is the data stored in the database, and the output is the updated index. Specifically, the server creates an index for newly stored data and updates the index.

[0677] Step 5:

[0678] User behavioral history and attribute information are analyzed using machine learning algorithms.

[0679] The server uses machine learning algorithms (e.g., scikit-learn, TensorFlow) to analyze the user's behavioral history and attribute information, and estimates attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history) with high accuracy. It also uses an emotion engine (e.g., IBM Watson, Affectiva) to recognize the user's emotions and analyze the emotions in the feedback data. The input is the behavioral history and attribute information stored in the database, and the output is the analyzed attribute information and emotion data. Specifically, the server schedules data analysis jobs and restores the analysis results to the database.

[0680] Step 6:

[0681] Generate follow-up prompts based on the analysis results.

[0682] The server obtains attribute information, interests, and the results of sentiment analysis related to a specific user, and generates a supplemental prompt based on this. The input is the analyzed attribute information and sentiment data, and the output is the generated supplemental prompt. Specifically, it generates a supplemental prompt based on the information "male in his 50s, likes beach resorts, hasn't been to Okinawa in the last 10 years."

[0683] Step 7:

[0684] Supplemental prompts are provided to the generative artificial intelligence to generate an optimal response.

[0685] The server provides the generated supplemental prompts to a generative artificial intelligence (e.g., OpenAI GPT-4, Google BERT). The AI ​​generates the optimal response for the user based on the supplemental prompts. The input is the supplemental prompt, and the output is the generated response. Specifically, the AI ​​generates the response, "Here are three suggestions for the latest activities you can enjoy at a beach resort in Okinawa."

[0686] Step 8:

[0687] The generated response is sent to the user's terminal.

[0688] The server formats the generated AI response and sends it to the device. The input is the generated response, and the output is the response delivered to the user. The specific behavior is to display detailed information about the suggested activity on a web page or application.

[0689] Step 9:

[0690] Collect and analyze user feedback.

[0691] The device tracks the user's reactions to the response from the AI ​​generator (e.g., clicks, time spent, feedback input) and their emotions at the time, and sends the collected data to the server in real time. The input is the user's reaction data, and the output is analyzed feedback. Specifically, if the user shows interest in a suggestion and clicks, that information and the timing of the click are tracked and sent to the server.

[0692] Step 10:

[0693] The feedback data is stored in a database and used to generate prompts in the future.

[0694] The server analyzes the feedback data and saves the results in a database. The input is the feedback data, and the output is the analysis results that will be used to generate future prompts. Specifically, if the analysis results of the feedback data indicate that "users are highly interested in information related to beach resorts," this information will be reflected in the generation of future prompts.

[0695] In this way, this system has a processing step that collects and analyzes the user's behavioral history, attribute information, and emotional information, and uses generative AI to provide highly accurate personalized responses.

[0696] (Application example 2)

[0697] 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."

[0698] Traditional personalization systems based on user behavior history and attribute information struggle to consider the user's emotional state, resulting in content and advertisements that are often not adapted to the user's current state. As a result, responses and suggestions to users are insufficient, limiting the effectiveness of advertisements and suggestions. Furthermore, there is a lack of effective methods for collecting and analyzing user feedback, making it difficult to continuously improve the system's personalization accuracy.

[0699] The identification processing 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 user behavior history and attribute information, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for transmitting the response results of the generation AI to the user's terminal, means for collecting feedback from the user, analyzing the feedback, and storing it in a database, means for recognizing the user's emotional state, means for generating advertisements based on the user's behavior history, attribute information, and emotional state, means for delivering the generated advertisements to the user's terminal, and means for collecting the user's reactions to advertisements and their emotional state. This enables highly accurate personalized responses and advertisement delivery based on the user's behavior history, attribute information, and emotional state.

[0700] "User behavior" is a record of specific actions (e.g., page views, clicks, form fills, etc.) that a user takes on a website or application.

[0701] "Descriptive information" means personal characteristics or profile information associated with a particular user (e.g., age, gender, region, interests, etc.).

[0702] A "database" is a collection of information that is organized so that collected data can be efficiently stored, managed, and retrieved.

[0703] "Supplemental prompts" are detailed instructions provided to the generative AI model that are generated based on the user's specific behavioral history and attribute information.

[0704] "Generative AI" refers to an AI system that generates appropriate responses or content based on provided prompts.

[0705] An "emotional state" is the specific emotional state (e.g., joy, sadness, stress, etc.) that a user is experiencing at a particular moment.

[0706] "Advertisement" means promotional content that is intended to inform or attract the attention of users about a particular product or service.

[0707] "Feedback" refers to the reactions and opinions (e.g., clicks, time spent, direct comments, etc.) that users have toward generated AI and advertisements.

[0708] The following configurations are possible as embodiments of the present invention. The present invention is a system that collects and analyzes a user's behavioral history and attribute information, and generates and delivers optimal advertisements taking into account their emotional state. This system is based on the interaction between a server, a terminal, and the user.

[0709] Data collection

[0710] The server first obtains the user's consent to collect the user's behavioral history and attribute information. This allows the data to be collected lawfully while protecting the user's privacy. The device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and transmits the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0711] Data Storage

[0712] The server moves the collected data from the temporary cache to a database and organizes it. The database organizes and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0713] Data analysis

[0714] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to accurately estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history). It also uses an emotion engine to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are stored back in the database and serve as the basis for generating supplementary prompts.

[0715] Supplemental prompt generation

[0716] The server obtains demographic information, interests, and sentiment analysis results related to a specific user. It then generates a supplemental prompt based on this information. A supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, a prompt might read, "Generate an ad for this user, a 30-year-old male user, interested in technology and sports, whose current emotional state is stressed. For this user, please generate an ad introducing the latest gadgets that can help relieve stress."

[0717] Generative Artificial Intelligence

[0718] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates optimal advertisements and content for the user based on the supplemental prompts. The generated advertisements and content contain specific and useful information in a form adapted to the user's needs, interests, and even recognized emotions.

[0719] Ad serving

[0720] The server formats the response generated by the artificial intelligence and sends it to the user's device, where the user can receive the most suitable advertisements and suggestions.

[0721] Feedback collection

[0722] The device tracks the user's reactions to the AI's responses (e.g., clicks, time spent, and feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0723] This allows for highly accurate personalized responses and ad delivery based on users' behavioral history, attribute information, and emotional state, while also effectively collecting user feedback to continuously improve the system.

[0724] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0725] Step 1: Collect user data

[0726] Input: User behavior history and attribute information

[0727] How it works: To collect user behavior history and attribute information, the server first obtains the user's consent. The device then tracks the user's actions on websites and applications (e.g., page views, clicks, and form entries) in real time and sends the data to the server.

[0728] Output: Collected user behavior history and attribute information data

[0729] Step 2: Store the data

[0730] Input: Collected user data

[0731] Specific operation: The server temporarily caches the received user behavior history and attribute information, then organizes the data and stores it in a database. In the database, the data for each user is systematically stored.

[0732] Output: User data stored in the database

[0733] Step 3: Data analysis

[0734] Input: User data stored in a database

[0735] Specific operation: The server analyzes the user's behavioral history and attribute information stored in the database. Using machine learning algorithms, it estimates the user's attributes (age, gender, region, etc.) and areas of interest (hobbies, purchasing history, etc.) with high accuracy. It also uses an emotion engine to recognize the user's emotional state and analyze the emotions in the feedback data.

[0736] Output: Analysis results including user attributes, interests, and emotional state

[0737] Step 4: Generate supplemental prompts

[0738] Input: Analysis results (user attributes, interests, emotional state)

[0739] Specific behavior: The server generates a supplemental prompt based on the analysis results, including user-specific information. For example, a specific prompt might be generated: "A 30-year-old male user with an interest in technology and sports, whose current emotional state is stress. Please generate an advertisement for this user introducing the latest gadgets that can help relieve stress."

[0740] Output: Generated supplemental prompts

[0741] Step 5: Generative AI generates a response

[0742] Input: Supplementary prompt

[0743] Specific operation: The server provides the generated supplemental prompts to the AI ​​generator, which generates optimal advertisements and content for the user based on the provided prompts.

[0744] Output: Generated ads and content

[0745] Step 6: Sending the response results

[0746] Input: Generated ads and content

[0747] Specific operation: The server formats the generated advertisement and content and sends it to the user's device. The user receives the response and views it.

[0748] Output: Ads and content sent to the user's device

[0749] Step 7: Gather feedback

[0750] Input: User's reaction and emotional state

[0751] Specific operation: The device tracks the user's reactions to ads and responses from the generated AI (e.g., clicks, time spent, feedback input) and their emotional state at the time in real time, and sends the data to the server.

[0752] Output: Collected feedback and emotional state data

[0753] Step 8: Analyze the feedback data

[0754] Input: Feedback data and emotional state data

[0755] Specific operation: The server analyzes the collected feedback data and emotional state data and stores the results in a database. The analysis results are used to generate supplementary prompts and improve the accuracy of AI responses in the future.

[0756] Output: Parsed feedback data

[0757] 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.

[0758] 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.

[0759] 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.

[0760] [Third embodiment]

[0761] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0762] 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.

[0763] 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).

[0764] 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.

[0765] 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.

[0766] 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).

[0767] 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.

[0768] 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.

[0769] 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.

[0770] 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.

[0771] 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.

[0772] 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."

[0773] This invention relates to a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the supplementary prompts to a generation artificial intelligence.

[0774] Data Collection Module

[0775] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0776] Data Storage Module

[0777] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0778] Data Analysis Module

[0779] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to estimate the user's attributes (age, gender, region, etc.) and areas of interest with high accuracy. The analysis results are then saved back into the database and used as the basis for generating supplementary prompts.

[0780] Prompt Generation Module

[0781] The server obtains the demographic information and interest analysis results associated with a specific user, and then generates supplemental prompts based on this information. The supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[0782] AI-enabled module

[0783] The server provides the generated supplemental prompts to the AI ​​generator, which generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information tailored to the user's needs and interests.

[0784] The server formats the generated AI's response and sends it to the device, where the user can receive the most appropriate information and suggestions.

[0785] Feedback Collection Module

[0786] The device tracks the user's reactions to the AI's responses (e.g., clicks, viewing time, and feedback input), and the collected feedback data is sent to a server in real time.

[0787] The server analyzes the feedback data and stores the results in a database. These analysis results are used to generate prompts and improve the accuracy of AI responses in the future.

[0788] Specific examples

[0789] For example, consider a case where a travel agency uses this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[0790] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[0791] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has no history of staying in Okinawa in the past 10 years."

[0792] 3. In the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[0793] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "new activities at a beach resort in Okinawa."

[0794] 5. In the feedback collection module, the device collects the user's responses to the suggestions (e.g., clicks and feedback) and sends them to the server, which allows the system to improve the accuracy of prompt generation in the future.

[0795] In this way, this invention enables generative AI to provide more accurate and personalized responses based on user attributes and interest information.

[0796] The processing flow will be explained below.

[0797] Step 1:

[0798] The device tracks user actions (e.g., page views, clicks, and form fills) on websites and applications. It embeds JavaScript code and an SDK to collect user interaction data in real time.

[0799] Step 2:

[0800] The device periodically sends the collected behavioral data to the server in batch processing or real-time streaming, where the data is encrypted and processed to protect user privacy.

[0801] Step 3:

[0802] The server temporarily stores the received data in a cache and performs data reformatting and cleansing (e.g., removing duplicate data and correcting outliers).

[0803] Step 4:

[0804] The server stores the formatted data in a database, where the behavioral history and attribute information for each user are stored in the appropriate tables.

[0805] Step 5:

[0806] The server analyzes the stored data and uses machine learning algorithms to estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history).

[0807] Step 6:

[0808] The server stores the analysis results back in a database, making them available for subsequent prompt generation and AI response processes.

[0809] Step 7:

[0810] The server retrieves analytics related to a particular user and generates follow-up prompts based on that information, including specific instructions such as "travel interests."

[0811] Step 8:

[0812] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates an optimal response for the user based on the supplemental prompts.

[0813] Step 9:

[0814] Generative AI analyzes the supplemental prompts and generates specific and useful information and suggestions for the user, such as creating detailed suggestions for travel plans.

[0815] Step 10:

[0816] The server formats the response generated by the AI ​​and sends it to the user's device. The response is sent in HTML or JSON format, making it easy for the user to understand.

[0817] Step 11:

[0818] The user receives a response from the generative AI through their device, and the user can take action based on the response and provide feedback.

[0819] Step 12:

[0820] The device tracks the user's reactions to the responses (e.g., clicks, time spent, feedback input) and sends them to the server.

[0821] Step 13:

[0822] The server analyzes the collected feedback data and stores the results in a database. These analysis results are used to generate follow-up prompts and improve the accuracy of AI responses in the future.

[0823] In this way, advanced personalization based on user attribute information and interests can be achieved, enabling highly accurate user responses using generative AI.

[0824] Example 1

[0825] 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."

[0826] Conventional systems have had difficulty effectively collecting and analyzing user behavioral history and attribute information to provide highly accurate, personalized responses to users. Furthermore, they lacked the real-time nature of collected data and the ability to continuously improve the system using feedback data.

[0827] 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.

[0828] In this invention, the server includes means for collecting user behavior history and attribute information, means for temporarily caching the collected data and storing it in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests using a machine learning algorithm, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation artificial intelligence to generate an optimal response for the user, means for formatting the generated response results and sending them to the user's terminal, and means for collecting feedback from the user, analyzing it, and storing it in a database. This makes it possible to provide highly accurate personalized responses based on the user's attribute and interest information, and to continuously improve the system by utilizing the feedback data.

[0829] "User behavior history" refers to the history of operations such as page views, clicks, and form input when a user uses a website or application.

[0830] "Attribute information" refers to information that indicates a user's individual characteristics, such as their age, gender, region, and areas of interest.

[0831] "Temporary cache" refers to a temporary storage area for storing data for a short period of time, used to improve the efficiency of real-time processing.

[0832] A "database" is an information management system used to store and manage data in a structured way.

[0833] A "machine learning algorithm" is a computational method that analyzes data and learns patterns to make predictions and classify unknown data.

[0834] A "supplemental prompt" is a detailed instruction that allows the generative artificial intelligence to generate the optimal response for a particular user.

[0835] "Generative AI" refers to an AI engine that generates optimal responses based on input prompts.

[0836] "Feedback" refers to the reaction a user makes to a provided response (e.g., clicks, view time, feedback input).

[0837] "Real-time" refers to data collection and processing occurring immediately, with minimal delay.

[0838] This invention is a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the supplementary prompts to a generation artificial intelligence.

[0839] Data Collection Module

[0840] The server first obtains the user's consent. At this time, a pop-up screen requesting consent is displayed on the user's device. After consent is obtained, the device tracks the user's behavioral history (e.g., page views, clicks, form entries, etc.) and attribute information (e.g., age, gender, region, etc.) within websites and applications. The collected data is sent to the server in real time, and the server temporarily stores this data in a cache.

[0841] Data Storage Module

[0842] The server periodically transfers the cached data to a database, where it systematically organizes and stores each user's behavioral history and attribute information. Organizing the data speeds up subsequent data retrieval and analysis processes.

[0843] Data Analysis Module

[0844] The server analyzes the behavioral history and attribute information stored in the database using machine learning algorithms. This analysis uses a variety of machine learning algorithms (e.g., clustering, classification, regression analysis, etc.) to estimate the user's attributes and areas of interest with high accuracy. The analysis results are saved back in the database and used as the basis for generating supplementary prompts.

[0845] Prompt Generation Module

[0846] The server obtains the attribute information and interest analysis results related to a specific user and generates a supplemental prompt based on that information. The supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, a travel agency might use this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the AI ​​generator.

[0847] AI-enabled module

[0848] The server provides the generated supplemental prompts to the AI ​​generator, which generates an optimal response for the user based on the supplemental prompts. The generated response contains information that is specific and relevant to the user's needs and interests. The server formats the generated response and sends it to the user's device.

[0849] Feedback Collection Module

[0850] Users can receive optimal information and suggestions through the generated responses. The device tracks the user's reactions to the responses from the AI ​​generator (e.g., clicks, viewing time, feedback input, etc.) and sends the feedback data to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0851] Specific examples

[0852] For example, consider a travel agency using this system to input a prompt such as "Create a perfect travel plan for the user!" into the AI ​​generator. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server. This data is stored in a database. The server then analyzes the past data using the data analysis module to estimate attribute information such as "male, 50s, beach resort lover, no history of visiting Okinawa in the last 10 years." Next, through the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "male, 50s, beach resort lover, has not been to Okinawa in the last 10 years." Through the AI ​​support module, the server provides the supplemental prompt to the AI ​​generator, suggesting specific travel plans to the user, such as "new activities at beach resorts in Okinawa." Through the feedback collection module, the device collects the user's responses to the suggestions (clicks and feedback) and sends them to the server. This improves the accuracy of prompt generation in the future.

[0853] This invention enables generative artificial intelligence to provide more accurate and personalized responses based on user attributes and interest information.

[0854] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0855] Step 1: Obtaining User Consent

[0856] To obtain permission to collect the user's behavioral history and attribute information, the server displays a pop-up consent screen on the user's device. This screen explains the details and purpose of the data to be collected. When the user clicks the "Allow" button, the server recognizes that permission has been obtained.

[0857] Input: User interaction from the terminal

[0858] Output: Collection permission status

[0859] Step 2: Collecting behavioral history and attribute information

[0860] After permission is granted, the device will begin tracking the user's actions, using JavaScript code to record web page browsing history, clicks, form entries, and other information. The collected data is then sent to a server in real time.

[0861] Input: User operation data

[0862] Output: Operational data collected in real time

[0863] Step 3: Temporarily Caching Data

[0864] The server temporarily stores the user operation data transmitted in real time in a cache, which is used for immediate or batch processing of data.

[0865] Input: Operational data transmitted in real time

[0866] Output: Temporary cached data

[0867] Step 4: Store in the database

[0868] The server periodically transfers the cached data to a database, where it is organized by user and stored in a structured format. This process updates the data index, streamlining subsequent data retrieval and analysis processes.

[0869] Input: Cached operation data

[0870] Output: Structured data stored in a database

[0871] Step 5: Data analysis

[0872] The server analyzes the user's behavioral history and attribute information stored in the database, and uses machine learning algorithms to accurately estimate the user's attributes (age, gender, region, etc.) and areas of interest.

[0873] Input: Structured data stored in a database

[0874] Output: Inferred attributes and interest data as analysis results

[0875] Step 6: Generate supplemental prompts

[0876] The server generates supplemental prompts based on the analysis results, which are detailed instructions to provide user-specific information to the AI, including content based on the user's attributes and interests.

[0877] Input: Inferred attributes and interest data as analysis results

[0878] Output: Generated supplemental prompts

[0879] Step 7: Prompt the generative AI

[0880] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates an optimal response for the user based on the prompts.

[0881] Input: Generated supplemental prompt

[0882] Output: A response generated by the generative artificial intelligence

[0883] Step 8: Providing a response to the user

[0884] The server formats the generated response and sends it to the device, through which the user receives the most relevant information and offers.

[0885] Input: A response generated by generative artificial intelligence

[0886] Output: The formatted response sent to the user's terminal

[0887] Step 9: Gather user responses

[0888] The device tracks the user's reactions to the generated AI's responses (clicks, viewing time, feedback input, etc.), and the collected feedback data is sent to the server in real time.

[0889] Input: User response data

[0890] Output: Feedback data sent to the server

[0891] Step 10: Feedback analysis and system improvement

[0892] The server analyzes the feedback data and stores the results in a database. These analysis results are used to generate prompts and improve the accuracy of AI responses in the future.

[0893] Input: Feedback data

[0894] Output: Database containing analysis results, improving system performance

[0895] (Application example 1)

[0896] 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."

[0897] Current online shopping sites often provide users with uniform product suggestions and campaign information, lacking personalized suggestions tailored to each user's interests and purchasing history. This results in lower user satisfaction and a decrease in willingness to purchase. Furthermore, the inaccuracy of suggestions means that user feedback cannot be effectively utilized.

[0898] 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.

[0899] In this invention, the server includes means for collecting user behavior history and attribute information, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for sending the generation AI's response results to the user's terminal, means for collecting user feedback, analyzing it, and storing it in a database, means for making personalized product suggestions based on the user's behavior history and attribute information, and means for improving suggestion accuracy based on user feedback. This makes it possible to provide individualized product suggestions to users, improve user satisfaction, and increase purchasing motivation.

[0900] A "user" is an individual or group that uses the system, and whose behavioral history and attribute information is collected.

[0901] "Behavioral history" refers to the series of operations or actions a user performs on a website or application, including page views, clicks, and form entries.

[0902] "Attribute information" refers to individual information such as a user's age, gender, region, and interests, and is data collected along with behavioral history.

[0903] A "database" is a system for storing and managing collected user behavioral history and attribute information, enabling data organization and search.

[0904] "Generative AI" refers to artificial intelligence technology that generates optimal responses for users based on collected data and generated prompts.

[0905] A "prompt" is a detailed instruction that provides user-specific information to the artificial intelligence generating the system, and is also called a supplemental prompt.

[0906] "Feedback" refers to the user's reaction to the response from the AI ​​generator, including clicks, viewing time, and feedback input.

[0907] "Personalized product suggestions" refers to suggesting product information and campaign information that is optimized for each individual user based on the user's behavioral history and attribute information.

[0908] "Recommendation accuracy" refers to the ability to offer products and services that match a user's interests, and is improved based on user feedback.

[0909] This invention is a system that collects user behavior history and attribute information, generates supplemental prompts based on the analysis results, and uses artificial intelligence to provide the user with the optimal response. Specific embodiments for implementing this invention are described below.

[0910] Data Collection Module

[0911] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0912] Data Storage Module

[0913] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0914] Data Analysis Module

[0915] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses machine learning algorithms such as Python's Scikit-learn and TensorFlow to estimate the user's attributes (age, gender, region, etc.) and areas of interest with high accuracy. The analysis results are then saved back into the database and used as the basis for generating supplementary prompts.

[0916] Prompt Generation Module

[0917] The server obtains the attribute information and interest analysis results related to a specific user. It then generates a supplemental prompt based on this information. The supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, it could say, "For a female user in her 30s who is interested in fashion, please suggest new fashion collections and sales information. She has a history of three online purchases in the past six months and browsing history as a guest user."

[0918] AI-enabled module

[0919] The server provides the generated supplemental prompts to the generation AI. The generation AI generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information adapted to the user's needs and interests. The server formats the generation AI's response result and sends it to the terminal. The user can receive optimal information and suggestions through this response.

[0920] Feedback Collection Module

[0921] The device tracks the user's reactions to the AI's responses (e.g., clicks, viewing time, and feedback input). The collected feedback data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0922] A specific example is a personalized shopping assistant application on an online shopping site. Based on the user's behavioral history and attribute information, this application generates prompts such as, "Please suggest new fashion collections and sale information to a female user in her 30s who is interested in fashion. She has a history of three online shopping trips in the past six months and browsing history as a guest user," and the artificial intelligence then makes optimal product suggestions.

[0923] In this way, this invention enables generative AI to provide more accurate and personalized responses based on user attributes and interest information.

[0924] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0925] Step 1:

[0926] The server collects the user's behavioral history and attribute information. To do so, the device displays a pop-up message to request the user's permission. If permission is granted, the device tracks the user's behavior on websites and applications (page views, clicks, form input, etc.) and sends this data to the server in real time. The input is user behavior data, and the output is tracking data sent to the server.

[0927] Step 2:

[0928] The server temporarily caches the received data and then stores it in a database. At this time, the data is organized by user. The input is tracking data, and the output is behavioral history and attribute information stored in the database.

[0929] Step 3:

[0930] The server analyzes the user's behavioral history and attribute information stored in the database. Specifically, it uses machine learning algorithms (e.g., Python's Scikit-learn or TensorFlow) to estimate the user's attributes (age, gender, region, etc.) and areas of interest. In this analysis process, data is input into the algorithm, and the output is estimated attribute information and interest information.

[0931] Step 4:

[0932] The server generates a supplementary prompt based on the estimated attribute information and interest information. The supplementary prompt serves as a detailed instruction for the generation AI. The input is the estimated attribute information and interest information, and the output is a generated prompt sentence. Specific operations include forming a sentence based on the generation rules.

[0933] Step 5:

[0934] The server provides the generated supplemental prompt to the AI ​​generator, which receives the supplemental prompt and generates an optimal response for the user. In this process, a prompt sentence is input and the output is a generated response. Specific operations include the execution of a natural language processing model.

[0935] Step 6:

[0936] The server formats the generated response and sends it to the terminal. The user's terminal receives the response and displays it on its screen. In this process, the input is the generated response and the output is the response displayed on the user's terminal.

[0937] Step 7:

[0938] The device tracks the user's reactions to the responses from the AI ​​generator (e.g., clicks, viewing time, feedback input). The input is the user's reaction data, and the output is the feedback data sent to the server in real time.

[0939] Step 8:

[0940] The server analyzes the collected feedback data and stores the results in a database. The analysis results are used to generate prompts and improve the accuracy of AI responses from the next time onwards. The input is the feedback data, and the output is the analyzed feedback information. Specific operations include the execution of a feedback analysis algorithm.

[0941] 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.

[0942] This invention relates to a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the generated prompts to a generative artificial intelligence. Furthermore, the invention aims to further improve the accuracy of responses by combining an emotion engine that recognizes the user's emotions.

[0943] Data Collection Module

[0944] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[0945] Data Storage Module

[0946] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[0947] Data Analysis Module

[0948] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history) with high accuracy. It also uses an emotion engine to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are saved back in the database and serve as the basis for generating supplementary prompts.

[0949] Prompt Generation Module

[0950] The server obtains the demographic information, interests, and sentiment analysis results associated with a specific user, and then generates supplemental prompts based on this information. The supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[0951] AI-enabled module

[0952] The server provides the generated supplemental prompts to the generation AI. The generation AI generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information in a form that is adapted to the user's needs, interests, and even recognized emotions. The server formats the generation AI's response result and sends it to the device. The user can receive optimal information and suggestions through this response.

[0953] Feedback Collection Module

[0954] The device tracks the user's reactions to the AI's responses (e.g., clicks, time spent, and feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[0955] Specific examples

[0956] For example, consider a case where a travel agency uses this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[0957] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[0958] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has not stayed in Okinawa in the last 10 years," and then uses an emotion engine to identify the user's recent emotional state.

[0959] 3. In the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[0960] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "New activities at beach resorts in Okinawa." These suggestions also take into account the user's emotions, so if the user feels like relaxing, for example, the server can choose the best resort for that time of year.

[0961] 5. In the feedback collection module, the device collects the user's reactions to the suggestions (e.g., clicks and feedback) and their emotions at the time, and sends them to the server. This allows the system to improve the accuracy of prompt generation in the future.

[0962] In this way, this invention enables generative AI to provide more accurate and personalized responses based on the user's attribute information, interests, and emotional information.

[0963] The processing flow will be explained below.

[0964] Step 1:

[0965] The device tracks data about user behavior on websites and applications (e.g., page views, clicks, and inputs). This process uses JavaScript code and SDKs to collect user interaction data in real time.

[0966] Step 2:

[0967] The device encrypts the tracked behavioral data and transmits it to a server over a secure protocol, either in batches or in real-time streaming.

[0968] Step 3:

[0969] The server temporarily caches the received behavioral data and then cleanses the cached data, for example, removing duplicate data and detecting invalid data.

[0970] Step 4:

[0971] The server stores the cleansed data in a database, where each user's behavioral history and attribute information are organized and saved.

[0972] Step 5:

[0973] The server analyzes the user's behavioral history and attribute information stored in the database, using machine learning algorithms to accurately estimate the user's attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history).

[0974] Step 6:

[0975] The server uses an emotion engine to recognize emotions from user behavior and feedback data, for example, by identifying emotions through text tone and facial expression analysis.

[0976] Step 7:

[0977] The server stores the analysis results and emotion recognition results in a database, which is used in the subsequent supplemental prompt generation step.

[0978] Step 8:

[0979] The server retrieves demographic information, interests, and sentiment analysis results related to a specific user from a database, and uses this information to generate detailed follow-up prompts.

[0980] Step 9:

[0981] The server provides the generated supplemental prompts to the artificial intelligence generating system, which generates an optimal response for the user based on the supplemental prompts.

[0982] Step 10:

[0983] Generative AI analyzes the supplemental prompts and generates specific and useful information and suggestions for the user, for example, specific suggestions based on the user's interests and emotions, rather than random recommendations for travel plans.

[0984] Step 11:

[0985] The server formats the response generated by the AI ​​and sends it to the user's device in a format suitable for the user interface (e.g., HTML, JSON).

[0986] Step 12:

[0987] The user receives a response from the AI ​​via their device, checks the response, and takes action based on it.

[0988] Step 13:

[0989] The device tracks the user's reactions to the responses (e.g., clicks, dwell time, feedback input) and their emotions at the time.

[0990] Step 14:

[0991] The device sends the collected feedback and emotion data to a server, where the data is encrypted to protect privacy.

[0992] Step 15:

[0993] The server receives the feedback data and emotion data and analyzes them again. The results of this analysis are stored in a database and used to generate prompts and improve the accuracy of AI responses in the future.

[0994] For example, when a travel agency provides a user with a suitable travel plan, it will take into account the user's past travel history, current interests, and emotional state through each of the above steps to provide personalized suggestions, which will improve user satisfaction and provide more effective services.

[0995] Example 2

[0996] 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."

[0997] Conventional information provision systems have had difficulty in providing personalized responses that take into account not only the user's behavioral history and attribute information, but also their emotions at the time. This has resulted in the inability to optimally respond to user needs and a decline in the quality of responses.

[0998] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user behavior history and attribute information, means for temporarily caching the collected data and then storing it in a database, means for updating the data stored in the database and indexes, means for analyzing the data stored in the database using a machine learning algorithm and estimating the user's attributes, interests, and emotions, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for transmitting the response results generated by the generation AI to the user's terminal, and means for collecting feedback from the user, analyzing it including emotional information, and storing it in a database. This makes it possible to provide highly accurate personalized responses that also take the user's emotional state into consideration.

[0999] "User behavior history" is a record of a series of operations and actions a user performs on a website or application.

[1000] "Attribute information" refers to information that indicates a user's personal characteristics, such as age, gender, region, hobbies, and purchasing history.

[1001] "Means of collection" refers to software or hardware mechanisms for capturing and recording user behavioral history and attribute information.

[1002] A "cache" is a high-speed accessible storage area for temporarily storing data.

[1003] A "database" is a system for efficiently storing, managing, and searching large amounts of data.

[1004] The "means for updating indexes" is a mechanism for creating and updating indexes of data in a database in order to improve search performance for the data.

[1005] A "machine learning algorithm" is an algorithm that learns patterns and rules from past data and applies them to new data.

[1006] "Analysis" refers to analyzing collected data and extracting meaningful information.

[1007] A "supplementary prompt" is a detailed instruction given to the artificial intelligence that is generated based on analyzed user information.

[1008] "Generative AI" is an AI technology that generates natural language responses and suggestions based on given data and instructions.

[1009] "Feedback" is a record of a user's reactions and opinions to the content provided by the system.

[1010] "Emotional information" is data that indicates the user's current emotional state and its changes.

[1011] This invention is a system that collects a user's behavioral history and attribute information, generates supplementary prompts based on the analysis results, and provides them to a generative artificial intelligence to provide the user with the optimal response. This system also aims to improve the accuracy of responses by combining it with an emotion engine that recognizes the user's emotions.

[1012] Data Collection Module

[1013] To collect user behavior history and attribute information, the server first obtains the user's consent. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, and form entries). This data collection is performed using JavaScript libraries and SDKs such as Google Analytics and Mixpanel. The collected data is sent to the server in real time, where it is temporarily cached and then stored in a database.

[1014] Data Storage Module

[1015] The server uses a database (e.g., MySQL, MongoDB) to move the collected data from the temporary cache to the database, organize the data, and update the data index to improve the search speed for each user's behavior history and attribute information.

[1016] Data Analysis Module

[1017] The server analyzes the user's behavioral history and attribute information stored in the database using machine learning algorithms (e.g., scikit-learn, TensorFlow) to accurately estimate the user's attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history). It also uses an emotion engine (e.g., IBM Watson, Affectiva) to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are saved back in the database and serve as the basis for generating supplementary prompts.

[1018] Prompt Generation Module

[1019] The server obtains attribute information, interests, and sentiment analysis results related to a specific user and generates supplemental prompts based on this information. Supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[1020] AI-enabled module

[1021] The server provides the generated supplemental prompts to a generative AI (e.g., OpenAI GPT-4, Google BERT). The generative AI generates the optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information that is adapted to the user's needs, interests, and even recognized emotions. The server formats the generative AI's response and sends it to the device. The user can receive optimal information and suggestions through this response.

[1022] Feedback Collection Module

[1023] The device collects the user's reactions to the responses from the AI ​​generator (e.g., clicks, time spent, feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes this feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[1024] Specific examples

[1025] For example, consider a case where a travel agency uses this system to input a prompt such as "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[1026] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[1027] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has not stayed in Okinawa in the last 10 years," and the emotion engine identifies the user's recent emotional state.

[1028] 3. In the prompt generation module, the server generates a supplementary prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[1029] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "New activities at beach resorts in Okinawa." These suggestions also take into account the user's emotions, so if the user feels like relaxing, for example, the server can choose the best resort for that time of year.

[1030] 5. In the feedback collection module, the device collects the user's reactions to the suggestions (e.g., clicks and feedback) and their emotions at the time, and sends them to the server. This allows the system to improve the accuracy of prompt generation in the future.

[1031] In this way, generative AI can provide more accurate and personalized responses based on the user's attribute information, interests, and emotional information.

[1032] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1033] Step 1:

[1034] Obtain user consent.

[1035] The server first presents the user with a privacy policy and data terms of use, and obtains consent to collect behavioral history and attribute information. The input is the user's consent data, and the output is a confirmation flag of consent. Specifically, when the user clicks the consent button, the information is sent to the server.

[1036] Step 2:

[1037] Collect user behavioral history and attribute information.

[1038] The device tracks user behavior on websites and applications (e.g., page views, clicks, and form entries). The input is user behavior data, and the output is tracked behavior history data. Specifically, data is collected using JavaScript libraries and SDKs (e.g., Google Analytics, Mixpanel) and sent to a server in real time.

[1039] Step 3:

[1040] The collected data is temporarily cached and then stored in a database.

[1041] The server temporarily stores the received data in a cache (for example, using Redis or Memcached), and stores the data collected within a certain period in a database (for example, MySQL or MongoDB) in batch processing. The input is the collected behavioral history and attribute information, and the output is the data stored in the database. Specifically, the data is stored in the cache the moment it arrives at the server, and is then moved to the database in batch processing every hour.

[1042] Step 4:

[1043] Update the data index.

[1044] The server creates an index for the data stored in the database, enabling quick searches of each user's behavioral history and attribute information. The input is the data stored in the database, and the output is the updated index. Specifically, the server creates an index for newly stored data and updates the index.

[1045] Step 5:

[1046] User behavioral history and attribute information are analyzed using machine learning algorithms.

[1047] The server uses machine learning algorithms (e.g., scikit-learn, TensorFlow) to analyze the user's behavioral history and attribute information, and estimates attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history) with high accuracy. It also uses an emotion engine (e.g., IBM Watson, Affectiva) to recognize the user's emotions and analyze the emotions in the feedback data. The input is the behavioral history and attribute information stored in the database, and the output is the analyzed attribute information and emotion data. Specifically, the server schedules data analysis jobs and restores the analysis results to the database.

[1048] Step 6:

[1049] Generate follow-up prompts based on the analysis results.

[1050] The server obtains attribute information, interests, and the results of sentiment analysis related to a specific user, and generates a supplemental prompt based on this. The input is the analyzed attribute information and sentiment data, and the output is the generated supplemental prompt. Specifically, it generates a supplemental prompt based on the information "male in his 50s, likes beach resorts, hasn't been to Okinawa in the last 10 years."

[1051] Step 7:

[1052] Supplemental prompts are provided to the generative artificial intelligence to generate an optimal response.

[1053] The server provides the generated supplemental prompts to a generative artificial intelligence (e.g., OpenAI GPT-4, Google BERT). The AI ​​generates the optimal response for the user based on the supplemental prompts. The input is the supplemental prompt, and the output is the generated response. Specifically, the AI ​​generates the response, "Here are three suggestions for the latest activities you can enjoy at a beach resort in Okinawa."

[1054] Step 8:

[1055] The generated response is sent to the user's terminal.

[1056] The server formats the generated AI response and sends it to the device. The input is the generated response, and the output is the response delivered to the user. The specific behavior is to display detailed information about the suggested activity on a web page or application.

[1057] Step 9:

[1058] Collect and analyze user feedback.

[1059] The device tracks the user's reactions to the response from the AI ​​generator (e.g., clicks, time spent, feedback input) and their emotions at the time, and sends the collected data to the server in real time. The input is the user's reaction data, and the output is analyzed feedback. Specifically, if the user shows interest in a suggestion and clicks, that information and the timing of the click are tracked and sent to the server.

[1060] Step 10:

[1061] The feedback data is stored in a database and used to generate prompts in the future.

[1062] The server analyzes the feedback data and saves the results in a database. The input is the feedback data, and the output is the analysis results that will be used to generate future prompts. Specifically, if the analysis results of the feedback data indicate that "users are highly interested in information related to beach resorts," this information will be reflected in the generation of future prompts.

[1063] In this way, this system has a processing step that collects and analyzes the user's behavioral history, attribute information, and emotional information, and uses generative AI to provide highly accurate personalized responses.

[1064] (Application example 2)

[1065] 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."

[1066] Traditional personalization systems based on user behavior history and attribute information struggle to consider the user's emotional state, resulting in content and advertisements that are often not adapted to the user's current state. As a result, responses and suggestions to users are insufficient, limiting the effectiveness of advertisements and suggestions. Furthermore, there is a lack of effective methods for collecting and analyzing user feedback, making it difficult to continuously improve the system's personalization accuracy.

[1067] The identification processing 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 user behavior history and attribute information, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for transmitting the response results of the generation AI to the user's terminal, means for collecting feedback from the user, analyzing the feedback, and storing it in a database, means for recognizing the user's emotional state, means for generating advertisements based on the user's behavior history, attribute information, and emotional state, means for delivering the generated advertisements to the user's terminal, and means for collecting the user's reactions to advertisements and their emotional state. This enables highly accurate personalized responses and advertisement delivery based on the user's behavior history, attribute information, and emotional state.

[1068] "User behavior" is a record of specific actions (e.g., page views, clicks, form fills, etc.) that a user takes on a website or application.

[1069] "Descriptive information" means personal characteristics or profile information associated with a particular user (e.g., age, gender, region, interests, etc.).

[1070] A "database" is a collection of information that is organized so that collected data can be efficiently stored, managed, and retrieved.

[1071] "Supplemental prompts" are detailed instructions provided to the generative AI model that are generated based on the user's specific behavioral history and attribute information.

[1072] "Generative AI" refers to an AI system that generates appropriate responses or content based on provided prompts.

[1073] An "emotional state" is the specific emotional state (e.g., joy, sadness, stress, etc.) that a user is experiencing at a particular moment.

[1074] "Advertisement" means promotional content that is intended to inform or attract the attention of users about a particular product or service.

[1075] "Feedback" refers to the reactions and opinions (e.g., clicks, time spent, direct comments, etc.) that users have toward generated AI and advertisements.

[1076] The following configurations are possible as embodiments of the present invention. The present invention is a system that collects and analyzes a user's behavioral history and attribute information, and generates and delivers optimal advertisements taking into account their emotional state. This system is based on the interaction between a server, a terminal, and the user.

[1077] Data collection

[1078] The server first obtains the user's consent to collect the user's behavioral history and attribute information. This allows the data to be collected lawfully while protecting the user's privacy. The device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and transmits the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[1079] Data Storage

[1080] The server moves the collected data from the temporary cache to a database and organizes it. The database organizes and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[1081] Data analysis

[1082] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to accurately estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history). It also uses an emotion engine to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are stored back in the database and serve as the basis for generating supplementary prompts.

[1083] Supplemental prompt generation

[1084] The server obtains demographic information, interests, and sentiment analysis results related to a specific user. It then generates a supplemental prompt based on this information. A supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, a prompt might read, "Generate an ad for this user, a 30-year-old male user, interested in technology and sports, whose current emotional state is stressed. For this user, please generate an ad introducing the latest gadgets that can help relieve stress."

[1085] Generative Artificial Intelligence

[1086] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates optimal advertisements and content for the user based on the supplemental prompts. The generated advertisements and content contain specific and useful information in a form adapted to the user's needs, interests, and even recognized emotions.

[1087] Ad serving

[1088] The server formats the response generated by the artificial intelligence and sends it to the user's device, where the user can receive the most suitable advertisements and suggestions.

[1089] Feedback collection

[1090] The device tracks the user's reactions to the AI's responses (e.g., clicks, time spent, and feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[1091] This allows for highly accurate personalized responses and ad delivery based on users' behavioral history, attribute information, and emotional state, while also effectively collecting user feedback to continuously improve the system.

[1092] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1093] Step 1: Collect user data

[1094] Input: User behavior history and attribute information

[1095] How it works: To collect user behavior history and attribute information, the server first obtains the user's consent. The device then tracks the user's actions on websites and applications (e.g., page views, clicks, and form entries) in real time and sends the data to the server.

[1096] Output: Collected user behavior history and attribute information data

[1097] Step 2: Store the data

[1098] Input: Collected user data

[1099] Specific operation: The server temporarily caches the received user behavior history and attribute information, then organizes the data and stores it in a database. In the database, the data for each user is systematically stored.

[1100] Output: User data stored in the database

[1101] Step 3: Data analysis

[1102] Input: User data stored in a database

[1103] Specific operation: The server analyzes the user's behavioral history and attribute information stored in the database. Using machine learning algorithms, it estimates the user's attributes (age, gender, region, etc.) and areas of interest (hobbies, purchasing history, etc.) with high accuracy. It also uses an emotion engine to recognize the user's emotional state and analyze the emotions in the feedback data.

[1104] Output: Analysis results including user attributes, interests, and emotional state

[1105] Step 4: Generate supplemental prompts

[1106] Input: Analysis results (user attributes, interests, emotional state)

[1107] Specific behavior: The server generates a supplemental prompt based on the analysis results, including user-specific information. For example, a specific prompt might be generated: "A 30-year-old male user with an interest in technology and sports, whose current emotional state is stress. Please generate an advertisement for this user introducing the latest gadgets that can help relieve stress."

[1108] Output: Generated supplemental prompts

[1109] Step 5: Generative AI generates a response

[1110] Input: Supplementary prompt

[1111] Specific operation: The server provides the generated supplemental prompts to the AI ​​generator, which generates optimal advertisements and content for the user based on the provided prompts.

[1112] Output: Generated ads and content

[1113] Step 6: Sending the response results

[1114] Input: Generated ads and content

[1115] Specific operation: The server formats the generated advertisement and content and sends it to the user's device. The user receives the response and views it.

[1116] Output: Ads and content sent to the user's device

[1117] Step 7: Gather feedback

[1118] Input: User's reaction and emotional state

[1119] Specific operation: The device tracks the user's reactions to ads and responses from the generated AI (e.g., clicks, time spent, feedback input) and their emotional state at the time in real time, and sends the data to the server.

[1120] Output: Collected feedback and emotional state data

[1121] Step 8: Analyze the feedback data

[1122] Input: Feedback data and emotional state data

[1123] Specific operation: The server analyzes the collected feedback data and emotional state data and stores the results in a database. The analysis results are used to generate supplementary prompts and improve the accuracy of AI responses in the future.

[1124] Output: Parsed feedback data

[1125] 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.

[1126] 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.

[1127] 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.

[1128] [Fourth embodiment]

[1129] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1130] 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.

[1131] 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).

[1132] 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.

[1133] 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.

[1134] 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).

[1135] 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.

[1136] 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.

[1137] 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.

[1138] 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.

[1139] 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.

[1140] 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.

[1141] 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."

[1142] This invention relates to a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the supplementary prompts to a generation artificial intelligence.

[1143] Data Collection Module

[1144] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[1145] Data Storage Module

[1146] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[1147] Data Analysis Module

[1148] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to estimate the user's attributes (age, gender, region, etc.) and areas of interest with high accuracy. The analysis results are then saved back into the database and used as the basis for generating supplementary prompts.

[1149] Prompt Generation Module

[1150] The server obtains the demographic information and interest analysis results associated with a specific user, and then generates supplemental prompts based on this information. The supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[1151] AI-enabled module

[1152] The server provides the generated supplemental prompts to the AI ​​generator, which generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information tailored to the user's needs and interests.

[1153] The server formats the generated AI's response and sends it to the device, where the user can receive the most appropriate information and suggestions.

[1154] Feedback Collection Module

[1155] The device tracks the user's reactions to the AI's responses (e.g., clicks, viewing time, and feedback input), and the collected feedback data is sent to a server in real time.

[1156] The server analyzes the feedback data and stores the results in a database. These analysis results are used to generate prompts and improve the accuracy of AI responses in the future.

[1157] Specific examples

[1158] For example, consider a case where a travel agency uses this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[1159] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[1160] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has no history of staying in Okinawa in the past 10 years."

[1161] 3. In the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[1162] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "new activities at a beach resort in Okinawa."

[1163] 5. In the feedback collection module, the device collects the user's responses to the suggestions (e.g., clicks and feedback) and sends them to the server, which allows the system to improve the accuracy of prompt generation in the future.

[1164] In this way, this invention enables generative AI to provide more accurate and personalized responses based on user attributes and interest information.

[1165] The processing flow will be explained below.

[1166] Step 1:

[1167] The device tracks user actions (e.g., page views, clicks, and form fills) on websites and applications. It embeds JavaScript code and an SDK to collect user interaction data in real time.

[1168] Step 2:

[1169] The device periodically sends the collected behavioral data to the server in batch processing or real-time streaming, where the data is encrypted and processed to protect user privacy.

[1170] Step 3:

[1171] The server temporarily stores the received data in a cache and performs data reformatting and cleansing (e.g., removing duplicate data and correcting outliers).

[1172] Step 4:

[1173] The server stores the formatted data in a database, where the behavioral history and attribute information for each user are stored in the appropriate tables.

[1174] Step 5:

[1175] The server analyzes the stored data and uses machine learning algorithms to estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history).

[1176] Step 6:

[1177] The server stores the analysis results back in a database, making them available for subsequent prompt generation and AI response processes.

[1178] Step 7:

[1179] The server retrieves analytics related to a particular user and generates follow-up prompts based on that information, including specific instructions such as "travel interests."

[1180] Step 8:

[1181] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates an optimal response for the user based on the supplemental prompts.

[1182] Step 9:

[1183] Generative AI analyzes the supplemental prompts and generates specific and useful information and suggestions for the user, such as creating detailed suggestions for travel plans.

[1184] Step 10:

[1185] The server formats the response generated by the AI ​​and sends it to the user's device. The response is sent in HTML or JSON format, making it easy for the user to understand.

[1186] Step 11:

[1187] The user receives a response from the generative AI through their device, and the user can take action based on the response and provide feedback.

[1188] Step 12:

[1189] The device tracks the user's reactions to the responses (e.g., clicks, time spent, feedback input) and sends them to the server.

[1190] Step 13:

[1191] The server analyzes the collected feedback data and stores the results in a database. These analysis results are used to generate follow-up prompts and improve the accuracy of AI responses in the future.

[1192] In this way, advanced personalization based on user attribute information and interests can be achieved, enabling highly accurate user responses using generative AI.

[1193] Example 1

[1194] 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."

[1195] Conventional systems have had difficulty effectively collecting and analyzing user behavioral history and attribute information to provide highly accurate, personalized responses to users. Furthermore, they lacked the real-time nature of collected data and the ability to continuously improve the system using feedback data.

[1196] 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.

[1197] In this invention, the server includes means for collecting user behavior history and attribute information, means for temporarily caching the collected data and storing it in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests using a machine learning algorithm, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation artificial intelligence to generate an optimal response for the user, means for formatting the generated response results and sending them to the user's terminal, and means for collecting feedback from the user, analyzing it, and storing it in a database. This makes it possible to provide highly accurate personalized responses based on the user's attribute and interest information, and to continuously improve the system by utilizing the feedback data.

[1198] "User behavior history" refers to the history of operations such as page views, clicks, and form input when a user uses a website or application.

[1199] "Attribute information" refers to information that indicates a user's individual characteristics, such as their age, gender, region, and areas of interest.

[1200] "Temporary cache" refers to a temporary storage area for storing data for a short period of time, used to improve the efficiency of real-time processing.

[1201] A "database" is an information management system used to store and manage data in a structured way.

[1202] A "machine learning algorithm" is a computational method that analyzes data and learns patterns to make predictions and classify unknown data.

[1203] A "supplemental prompt" is a detailed instruction that allows the generative artificial intelligence to generate the optimal response for a particular user.

[1204] "Generative AI" refers to an AI engine that generates optimal responses based on input prompts.

[1205] "Feedback" refers to the reaction a user makes to a provided response (e.g., clicks, view time, feedback input).

[1206] "Real-time" refers to data collection and processing occurring immediately, with minimal delay.

[1207] This invention is a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the supplementary prompts to a generation artificial intelligence.

[1208] Data Collection Module

[1209] The server first obtains the user's consent. At this time, a pop-up screen requesting consent is displayed on the user's device. After consent is obtained, the device tracks the user's behavioral history (e.g., page views, clicks, form entries, etc.) and attribute information (e.g., age, gender, region, etc.) within websites and applications. The collected data is sent to the server in real time, and the server temporarily stores this data in a cache.

[1210] Data Storage Module

[1211] The server periodically transfers the cached data to a database, where it systematically organizes and stores each user's behavioral history and attribute information. Organizing the data speeds up subsequent data retrieval and analysis processes.

[1212] Data Analysis Module

[1213] The server analyzes the behavioral history and attribute information stored in the database using machine learning algorithms. This analysis uses a variety of machine learning algorithms (e.g., clustering, classification, regression analysis, etc.) to estimate the user's attributes and areas of interest with high accuracy. The analysis results are saved back in the database and used as the basis for generating supplementary prompts.

[1214] Prompt Generation Module

[1215] The server obtains the attribute information and interest analysis results related to a specific user and generates a supplemental prompt based on that information. The supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, a travel agency might use this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the AI ​​generator.

[1216] AI-enabled module

[1217] The server provides the generated supplemental prompts to the AI ​​generator, which generates an optimal response for the user based on the supplemental prompts. The generated response contains information that is specific and relevant to the user's needs and interests. The server formats the generated response and sends it to the user's device.

[1218] Feedback Collection Module

[1219] Users can receive optimal information and suggestions through the generated responses. The device tracks the user's reactions to the responses from the AI ​​generator (e.g., clicks, viewing time, feedback input, etc.) and sends the feedback data to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[1220] Specific examples

[1221] For example, consider a travel agency using this system to input a prompt such as "Create a perfect travel plan for the user!" into the AI ​​generator. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server. This data is stored in a database. The server then analyzes the past data using the data analysis module to estimate attribute information such as "male, 50s, beach resort lover, no history of visiting Okinawa in the last 10 years." Next, through the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "male, 50s, beach resort lover, has not been to Okinawa in the last 10 years." Through the AI ​​support module, the server provides the supplemental prompt to the AI ​​generator, suggesting specific travel plans to the user, such as "new activities at beach resorts in Okinawa." Through the feedback collection module, the device collects the user's responses to the suggestions (clicks and feedback) and sends them to the server. This improves the accuracy of prompt generation in the future.

[1222] This invention enables generative artificial intelligence to provide more accurate and personalized responses based on user attributes and interest information.

[1223] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1224] Step 1: Obtaining User Consent

[1225] To obtain permission to collect the user's behavioral history and attribute information, the server displays a pop-up consent screen on the user's device. This screen explains the details and purpose of the data to be collected. When the user clicks the "Allow" button, the server recognizes that permission has been obtained.

[1226] Input: User interaction from the terminal

[1227] Output: Collection permission status

[1228] Step 2: Collecting behavioral history and attribute information

[1229] After permission is granted, the device will begin tracking the user's actions, using JavaScript code to record web page browsing history, clicks, form entries, and other information. The collected data is then sent to a server in real time.

[1230] Input: User operation data

[1231] Output: Operational data collected in real time

[1232] Step 3: Temporarily Caching Data

[1233] The server temporarily stores the user operation data transmitted in real time in a cache, which is used for immediate or batch processing of data.

[1234] Input: Operational data transmitted in real time

[1235] Output: Temporary cached data

[1236] Step 4: Store in the database

[1237] The server periodically transfers the cached data to a database, where it is organized by user and stored in a structured format. This process updates the data index, streamlining subsequent data retrieval and analysis processes.

[1238] Input: Cached operation data

[1239] Output: Structured data stored in a database

[1240] Step 5: Data analysis

[1241] The server analyzes the user's behavioral history and attribute information stored in the database, and uses machine learning algorithms to accurately estimate the user's attributes (age, gender, region, etc.) and areas of interest.

[1242] Input: Structured data stored in a database

[1243] Output: Inferred attributes and interest data as analysis results

[1244] Step 6: Generate supplemental prompts

[1245] The server generates supplemental prompts based on the analysis results, which are detailed instructions to provide user-specific information to the AI, including content based on the user's attributes and interests.

[1246] Input: Inferred attributes and interest data as analysis results

[1247] Output: Generated supplemental prompts

[1248] Step 7: Prompt the generative AI

[1249] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates an optimal response for the user based on the prompts.

[1250] Input: Generated supplemental prompt

[1251] Output: A response generated by the generative artificial intelligence

[1252] Step 8: Providing a response to the user

[1253] The server formats the generated response and sends it to the device, through which the user receives the most relevant information and offers.

[1254] Input: A response generated by generative artificial intelligence

[1255] Output: The formatted response sent to the user's terminal

[1256] Step 9: Gather user responses

[1257] The device tracks the user's reactions to the generated AI's responses (clicks, viewing time, feedback input, etc.), and the collected feedback data is sent to the server in real time.

[1258] Input: User response data

[1259] Output: Feedback data sent to the server

[1260] Step 10: Feedback analysis and system improvement

[1261] The server analyzes the feedback data and stores the results in a database. These analysis results are used to generate prompts and improve the accuracy of AI responses in the future.

[1262] Input: Feedback data

[1263] Output: Database containing analysis results, improving system performance

[1264] (Application example 1)

[1265] 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."

[1266] Current online shopping sites often provide users with uniform product suggestions and campaign information, lacking personalized suggestions tailored to each user's interests and purchasing history. This results in lower user satisfaction and a decrease in willingness to purchase. Furthermore, the inaccuracy of suggestions means that user feedback cannot be effectively utilized.

[1267] 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.

[1268] In this invention, the server includes means for collecting user behavior history and attribute information, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for sending the generation AI's response results to the user's terminal, means for collecting user feedback, analyzing it, and storing it in a database, means for making personalized product suggestions based on the user's behavior history and attribute information, and means for improving suggestion accuracy based on user feedback. This makes it possible to provide individualized product suggestions to users, improve user satisfaction, and increase purchasing motivation.

[1269] A "user" is an individual or group that uses the system, and whose behavioral history and attribute information is collected.

[1270] "Behavioral history" refers to the series of operations or actions a user performs on a website or application, including page views, clicks, and form entries.

[1271] "Attribute information" refers to individual information such as a user's age, gender, region, and interests, and is data collected along with behavioral history.

[1272] A "database" is a system for storing and managing collected user behavioral history and attribute information, enabling data organization and search.

[1273] "Generative AI" refers to artificial intelligence technology that generates optimal responses for users based on collected data and generated prompts.

[1274] A "prompt" is a detailed instruction that provides user-specific information to the artificial intelligence generating the system, and is also called a supplemental prompt.

[1275] "Feedback" refers to the user's reaction to the response from the AI ​​generator, including clicks, viewing time, and feedback input.

[1276] "Personalized product suggestions" refers to suggesting product information and campaign information that is optimized for each individual user based on the user's behavioral history and attribute information.

[1277] "Recommendation accuracy" refers to the ability to offer products and services that match a user's interests, and is improved based on user feedback.

[1278] This invention is a system that collects user behavior history and attribute information, generates supplemental prompts based on the analysis results, and uses artificial intelligence to provide the user with the optimal response. Specific embodiments for implementing this invention are described below.

[1279] Data Collection Module

[1280] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[1281] Data Storage Module

[1282] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[1283] Data Analysis Module

[1284] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses machine learning algorithms such as Python's Scikit-learn and TensorFlow to estimate the user's attributes (age, gender, region, etc.) and areas of interest with high accuracy. The analysis results are then saved back into the database and used as the basis for generating supplementary prompts.

[1285] Prompt Generation Module

[1286] The server obtains the attribute information and interest analysis results related to a specific user. It then generates a supplemental prompt based on this information. The supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, it could say, "For a female user in her 30s who is interested in fashion, please suggest new fashion collections and sales information. She has a history of three online purchases in the past six months and browsing history as a guest user."

[1287] AI-enabled module

[1288] The server provides the generated supplemental prompts to the generation AI. The generation AI generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information adapted to the user's needs and interests. The server formats the generation AI's response result and sends it to the terminal. The user can receive optimal information and suggestions through this response.

[1289] Feedback Collection Module

[1290] The device tracks the user's reactions to the AI's responses (e.g., clicks, viewing time, and feedback input). The collected feedback data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[1291] A specific example is a personalized shopping assistant application on an online shopping site. Based on the user's behavioral history and attribute information, this application generates prompts such as, "Please suggest new fashion collections and sale information to a female user in her 30s who is interested in fashion. She has a history of three online shopping trips in the past six months and browsing history as a guest user," and the artificial intelligence then makes optimal product suggestions.

[1292] In this way, this invention enables generative AI to provide more accurate and personalized responses based on user attributes and interest information.

[1293] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1294] Step 1:

[1295] The server collects the user's behavioral history and attribute information. To do so, the device displays a pop-up message to request the user's permission. If permission is granted, the device tracks the user's behavior on websites and applications (page views, clicks, form input, etc.) and sends this data to the server in real time. The input is user behavior data, and the output is tracking data sent to the server.

[1296] Step 2:

[1297] The server temporarily caches the received data and then stores it in a database. At this time, the data is organized by user. The input is tracking data, and the output is behavioral history and attribute information stored in the database.

[1298] Step 3:

[1299] The server analyzes the user's behavioral history and attribute information stored in the database. Specifically, it uses machine learning algorithms (e.g., Python's Scikit-learn or TensorFlow) to estimate the user's attributes (age, gender, region, etc.) and areas of interest. In this analysis process, data is input into the algorithm, and the output is estimated attribute information and interest information.

[1300] Step 4:

[1301] The server generates a supplementary prompt based on the estimated attribute information and interest information. The supplementary prompt serves as a detailed instruction for the generation AI. The input is the estimated attribute information and interest information, and the output is a generated prompt sentence. Specific operations include forming a sentence based on the generation rules.

[1302] Step 5:

[1303] The server provides the generated supplemental prompt to the AI ​​generator, which receives the supplemental prompt and generates an optimal response for the user. In this process, a prompt sentence is input and the output is a generated response. Specific operations include the execution of a natural language processing model.

[1304] Step 6:

[1305] The server formats the generated response and sends it to the terminal. The user's terminal receives the response and displays it on its screen. In this process, the input is the generated response and the output is the response displayed on the user's terminal.

[1306] Step 7:

[1307] The device tracks the user's reactions to the responses from the AI ​​generator (e.g., clicks, viewing time, feedback input). The input is the user's reaction data, and the output is the feedback data sent to the server in real time.

[1308] Step 8:

[1309] The server analyzes the collected feedback data and stores the results in a database. The analysis results are used to generate prompts and improve the accuracy of AI responses from the next time onwards. The input is the feedback data, and the output is the analyzed feedback information. Specific operations include the execution of a feedback analysis algorithm.

[1310] 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.

[1311] This invention relates to a system that provides optimal responses to users by collecting user behavior history and attribute information, generating supplementary prompts based on the analysis results, and providing the generated prompts to a generative artificial intelligence. Furthermore, the invention aims to further improve the accuracy of responses by combining an emotion engine that recognizes the user's emotions.

[1312] Data Collection Module

[1313] The server first obtains the user's consent to collect the user's behavioral history and attribute information. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and sends the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[1314] Data Storage Module

[1315] The server moves the collected data from the temporary cache to a database, organizes the data, and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[1316] Data Analysis Module

[1317] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history) with high accuracy. It also uses an emotion engine to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are saved back in the database and serve as the basis for generating supplementary prompts.

[1318] Prompt Generation Module

[1319] The server obtains the demographic information, interests, and sentiment analysis results associated with a specific user, and then generates supplemental prompts based on this information. The supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[1320] AI-enabled module

[1321] The server provides the generated supplemental prompts to the generation AI. The generation AI generates an optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information in a form that is adapted to the user's needs, interests, and even recognized emotions. The server formats the generation AI's response result and sends it to the device. The user can receive optimal information and suggestions through this response.

[1322] Feedback Collection Module

[1323] The device tracks the user's reactions to the AI's responses (e.g., clicks, time spent, and feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[1324] Specific examples

[1325] For example, consider a case where a travel agency uses this system to input a prompt such as, "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[1326] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[1327] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has not stayed in Okinawa in the last 10 years," and then uses an emotion engine to identify the user's recent emotional state.

[1328] 3. In the supplemental prompt generation module, the server generates a supplemental prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[1329] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "New activities at beach resorts in Okinawa." These suggestions also take into account the user's emotions, so if the user feels like relaxing, for example, the server can choose the best resort for that time of year.

[1330] 5. In the feedback collection module, the device collects the user's reactions to the suggestions (e.g., clicks and feedback) and their emotions at the time, and sends them to the server. This allows the system to improve the accuracy of prompt generation in the future.

[1331] In this way, this invention enables generative AI to provide more accurate and personalized responses based on the user's attribute information, interests, and emotional information.

[1332] The processing flow will be explained below.

[1333] Step 1:

[1334] The device tracks data about user behavior on websites and applications (e.g., page views, clicks, and inputs). This process uses JavaScript code and SDKs to collect user interaction data in real time.

[1335] Step 2:

[1336] The device encrypts the tracked behavioral data and transmits it to a server over a secure protocol, either in batches or in real-time streaming.

[1337] Step 3:

[1338] The server temporarily caches the received behavioral data and then cleanses the cached data, for example, removing duplicate data and detecting invalid data.

[1339] Step 4:

[1340] The server stores the cleansed data in a database, where each user's behavioral history and attribute information are organized and saved.

[1341] Step 5:

[1342] The server analyzes the user's behavioral history and attribute information stored in the database, using machine learning algorithms to accurately estimate the user's attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history).

[1343] Step 6:

[1344] The server uses an emotion engine to recognize emotions from user behavior and feedback data, for example, by identifying emotions through text tone and facial expression analysis.

[1345] Step 7:

[1346] The server stores the analysis results and emotion recognition results in a database, which is used in the subsequent supplemental prompt generation step.

[1347] Step 8:

[1348] The server retrieves demographic information, interests, and sentiment analysis results related to a specific user from a database, and uses this information to generate detailed follow-up prompts.

[1349] Step 9:

[1350] The server provides the generated supplemental prompts to the artificial intelligence generating system, which generates an optimal response for the user based on the supplemental prompts.

[1351] Step 10:

[1352] Generative AI analyzes the supplemental prompts and generates specific and useful information and suggestions for the user, for example, specific suggestions based on the user's interests and emotions, rather than random recommendations for travel plans.

[1353] Step 11:

[1354] The server formats the response generated by the AI ​​and sends it to the user's device in a format suitable for the user interface (e.g., HTML, JSON).

[1355] Step 12:

[1356] The user receives a response from the AI ​​via their device, checks the response, and takes action based on it.

[1357] Step 13:

[1358] The device tracks the user's reactions to the responses (e.g., clicks, dwell time, feedback input) and their emotions at the time.

[1359] Step 14:

[1360] The device sends the collected feedback and emotion data to a server, where the data is encrypted to protect privacy.

[1361] Step 15:

[1362] The server receives the feedback data and emotion data and analyzes them again. The results of this analysis are stored in a database and used to generate prompts and improve the accuracy of AI responses in the future.

[1363] For example, when a travel agency provides a user with a suitable travel plan, it will take into account the user's past travel history, current interests, and emotional state through each of the above steps to provide personalized suggestions, which will improve user satisfaction and provide more effective services.

[1364] Example 2

[1365] 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."

[1366] Conventional information provision systems have had difficulty in providing personalized responses that take into account not only the user's behavioral history and attribute information, but also their emotions at the time. This has resulted in the inability to optimally respond to user needs and a decline in the quality of responses.

[1367] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user behavior history and attribute information, means for temporarily caching the collected data and then storing it in a database, means for updating the data stored in the database and indexes, means for analyzing the data stored in the database using a machine learning algorithm and estimating the user's attributes, interests, and emotions, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for transmitting the response results generated by the generation AI to the user's terminal, and means for collecting feedback from the user, analyzing it including emotional information, and storing it in a database. This makes it possible to provide highly accurate personalized responses that also take the user's emotional state into consideration.

[1368] "User behavior history" is a record of a series of operations and actions a user performs on a website or application.

[1369] "Attribute information" refers to information that indicates a user's personal characteristics, such as age, gender, region, hobbies, and purchasing history.

[1370] "Means of collection" refers to software or hardware mechanisms for capturing and recording user behavioral history and attribute information.

[1371] A "cache" is a high-speed accessible storage area for temporarily storing data.

[1372] A "database" is a system for efficiently storing, managing, and searching large amounts of data.

[1373] The "means for updating indexes" is a mechanism for creating and updating indexes of data in a database in order to improve search performance for the data.

[1374] A "machine learning algorithm" is an algorithm that learns patterns and rules from past data and applies them to new data.

[1375] "Analysis" refers to analyzing collected data and extracting meaningful information.

[1376] A "supplementary prompt" is a detailed instruction given to the artificial intelligence that is generated based on analyzed user information.

[1377] "Generative AI" is an AI technology that generates natural language responses and suggestions based on given data and instructions.

[1378] "Feedback" is a record of a user's reactions and opinions to the content provided by the system.

[1379] "Emotional information" is data that indicates the user's current emotional state and its changes.

[1380] This invention is a system that collects a user's behavioral history and attribute information, generates supplementary prompts based on the analysis results, and provides them to a generative artificial intelligence to provide the user with the optimal response. This system also aims to improve the accuracy of responses by combining it with an emotion engine that recognizes the user's emotions.

[1381] Data Collection Module

[1382] To collect user behavior history and attribute information, the server first obtains the user's consent. After obtaining consent, the device tracks the user's behavior on websites and applications (e.g., page views, clicks, and form entries). This data collection is performed using JavaScript libraries and SDKs such as Google Analytics and Mixpanel. The collected data is sent to the server in real time, where it is temporarily cached and then stored in a database.

[1383] Data Storage Module

[1384] The server uses a database (e.g., MySQL, MongoDB) to move the collected data from the temporary cache to the database, organize the data, and update the data index to improve the search speed for each user's behavior history and attribute information.

[1385] Data Analysis Module

[1386] The server analyzes the user's behavioral history and attribute information stored in the database using machine learning algorithms (e.g., scikit-learn, TensorFlow) to accurately estimate the user's attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history). It also uses an emotion engine (e.g., IBM Watson, Affectiva) to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are saved back in the database and serve as the basis for generating supplementary prompts.

[1387] Prompt Generation Module

[1388] The server obtains attribute information, interests, and sentiment analysis results related to a specific user and generates supplemental prompts based on this information. Supplemental prompts are detailed instructions that provide user-specific information to the AI ​​generator.

[1389] AI-enabled module

[1390] The server provides the generated supplemental prompts to a generative AI (e.g., OpenAI GPT-4, Google BERT). The generative AI generates the optimal response for the user based on the supplemental prompts. The generated response contains specific and useful information that is adapted to the user's needs, interests, and even recognized emotions. The server formats the generative AI's response and sends it to the device. The user can receive optimal information and suggestions through this response.

[1391] Feedback Collection Module

[1392] The device collects the user's reactions to the responses from the AI ​​generator (e.g., clicks, time spent, feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes this feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[1393] Specific examples

[1394] For example, consider a case where a travel agency uses this system to input a prompt such as "Create a travel plan that's perfect for the user!" into the artificial intelligence generator.

[1395] 1. Through the data collection module, the device collects the user's past travel history and interests and sends them to the server, where they are stored in a database.

[1396] 2. In the data analysis module, the server analyzes past data and estimates attribute information such as "male, in his 50s, likes beach resorts, has not stayed in Okinawa in the last 10 years," and the emotion engine identifies the user's recent emotional state.

[1397] 3. In the prompt generation module, the server generates a supplementary prompt based on the analysis results: "Male in his 50s, likes beach resorts, has not been to Okinawa in the last 10 years."

[1398] 4. In the AI-enabled module, the server provides supplementary prompts to the AI ​​generator, suggesting specific travel plans to the user, such as "New activities at beach resorts in Okinawa." These suggestions also take into account the user's emotions, so if the user feels like relaxing, for example, the server can choose the best resort for that time of year.

[1399] 5. In the feedback collection module, the device collects the user's reactions to the suggestions (e.g., clicks and feedback) and their emotions at the time, and sends them to the server. This allows the system to improve the accuracy of prompt generation in the future.

[1400] In this way, generative AI can provide more accurate and personalized responses based on the user's attribute information, interests, and emotional information.

[1401] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1402] Step 1:

[1403] Obtain user consent.

[1404] The server first presents the user with a privacy policy and data terms of use, and obtains consent to collect behavioral history and attribute information. The input is the user's consent data, and the output is a confirmation flag of consent. Specifically, when the user clicks the consent button, the information is sent to the server.

[1405] Step 2:

[1406] Collect user behavioral history and attribute information.

[1407] The device tracks user behavior on websites and applications (e.g., page views, clicks, and form entries). The input is user behavior data, and the output is tracked behavior history data. Specifically, data is collected using JavaScript libraries and SDKs (e.g., Google Analytics, Mixpanel) and sent to a server in real time.

[1408] Step 3:

[1409] The collected data is temporarily cached and then stored in a database.

[1410] The server temporarily stores the received data in a cache (for example, using Redis or Memcached), and stores the data collected within a certain period in a database (for example, MySQL or MongoDB) in batch processing. The input is the collected behavioral history and attribute information, and the output is the data stored in the database. Specifically, the data is stored in the cache the moment it arrives at the server, and is then moved to the database in batch processing every hour.

[1411] Step 4:

[1412] Update the data index.

[1413] The server creates an index for the data stored in the database, enabling quick searches of each user's behavioral history and attribute information. The input is the data stored in the database, and the output is the updated index. Specifically, the server creates an index for newly stored data and updates the index.

[1414] Step 5:

[1415] User behavioral history and attribute information are analyzed using machine learning algorithms.

[1416] The server uses machine learning algorithms (e.g., scikit-learn, TensorFlow) to analyze the user's behavioral history and attribute information, and estimates attributes (e.g., age, gender, region) and interests (e.g., hobbies, purchasing history) with high accuracy. It also uses an emotion engine (e.g., IBM Watson, Affectiva) to recognize the user's emotions and analyze the emotions in the feedback data. The input is the behavioral history and attribute information stored in the database, and the output is the analyzed attribute information and emotion data. Specifically, the server schedules data analysis jobs and restores the analysis results to the database.

[1417] Step 6:

[1418] Generate follow-up prompts based on the analysis results.

[1419] The server obtains attribute information, interests, and the results of sentiment analysis related to a specific user, and generates a supplemental prompt based on this. The input is the analyzed attribute information and sentiment data, and the output is the generated supplemental prompt. Specifically, it generates a supplemental prompt based on the information "male in his 50s, likes beach resorts, hasn't been to Okinawa in the last 10 years."

[1420] Step 7:

[1421] Supplemental prompts are provided to the generative artificial intelligence to generate an optimal response.

[1422] The server provides the generated supplemental prompts to a generative artificial intelligence (e.g., OpenAI GPT-4, Google BERT). The AI ​​generates the optimal response for the user based on the supplemental prompts. The input is the supplemental prompt, and the output is the generated response. Specifically, the AI ​​generates the response, "Here are three suggestions for the latest activities you can enjoy at a beach resort in Okinawa."

[1423] Step 8:

[1424] The generated response is sent to the user's terminal.

[1425] The server formats the generated AI response and sends it to the device. The input is the generated response, and the output is the response delivered to the user. The specific behavior is to display detailed information about the suggested activity on a web page or application.

[1426] Step 9:

[1427] Collect and analyze user feedback.

[1428] The device tracks the user's reactions to the response from the AI ​​generator (e.g., clicks, time spent, feedback input) and their emotions at the time, and sends the collected data to the server in real time. The input is the user's reaction data, and the output is analyzed feedback. Specifically, if the user shows interest in a suggestion and clicks, that information and the timing of the click are tracked and sent to the server.

[1429] Step 10:

[1430] The feedback data is stored in a database and used to generate prompts in the future.

[1431] The server analyzes the feedback data and saves the results in a database. The input is the feedback data, and the output is the analysis results that will be used to generate future prompts. Specifically, if the analysis results of the feedback data indicate that "users are highly interested in information related to beach resorts," this information will be reflected in the generation of future prompts.

[1432] In this way, this system has a processing step that collects and analyzes the user's behavioral history, attribute information, and emotional information, and uses generative AI to provide highly accurate personalized responses.

[1433] (Application example 2)

[1434] 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."

[1435] Traditional personalization systems based on user behavior history and attribute information struggle to consider the user's emotional state, resulting in content and advertisements that are often not adapted to the user's current state. As a result, responses and suggestions to users are insufficient, limiting the effectiveness of advertisements and suggestions. Furthermore, there is a lack of effective methods for collecting and analyzing user feedback, making it difficult to continuously improve the system's personalization accuracy.

[1436] The identification processing 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 user behavior history and attribute information, means for storing the collected data in a database, means for analyzing the data stored in the database and estimating the user's attributes and interests, means for generating follow-up prompts based on the estimated information, means for providing the generated follow-up prompts to a generation AI to generate an optimal response for the user, means for transmitting the response results of the generation AI to the user's terminal, means for collecting feedback from the user, analyzing the feedback, and storing it in a database, means for recognizing the user's emotional state, means for generating advertisements based on the user's behavior history, attribute information, and emotional state, means for delivering the generated advertisements to the user's terminal, and means for collecting the user's reactions to advertisements and their emotional state. This enables highly accurate personalized responses and advertisement delivery based on the user's behavior history, attribute information, and emotional state.

[1437] "User behavior" is a record of specific actions (e.g., page views, clicks, form fills, etc.) that a user takes on a website or application.

[1438] "Descriptive information" means personal characteristics or profile information associated with a particular user (e.g., age, gender, region, interests, etc.).

[1439] A "database" is a collection of information that is organized so that collected data can be efficiently stored, managed, and retrieved.

[1440] "Supplemental prompts" are detailed instructions provided to the generative AI model that are generated based on the user's specific behavioral history and attribute information.

[1441] "Generative AI" refers to an AI system that generates appropriate responses or content based on provided prompts.

[1442] An "emotional state" is the specific emotional state (e.g., joy, sadness, stress, etc.) that a user is experiencing at a particular moment.

[1443] "Advertisement" means promotional content that is intended to inform or attract the attention of users about a particular product or service.

[1444] "Feedback" refers to the reactions and opinions (e.g., clicks, time spent, direct comments, etc.) that users have toward generated AI and advertisements.

[1445] The following configurations are possible as embodiments of the present invention. The present invention is a system that collects and analyzes a user's behavioral history and attribute information, and generates and delivers optimal advertisements taking into account their emotional state. This system is based on the interaction between a server, a terminal, and the user.

[1446] Data collection

[1447] The server first obtains the user's consent to collect the user's behavioral history and attribute information. This allows the data to be collected lawfully while protecting the user's privacy. The device tracks the user's behavior on websites and applications (e.g., page views, clicks, form entries) and transmits the acquired data to the server in real time. The server temporarily caches the received data and then stores it in a database.

[1448] Data Storage

[1449] The server moves the collected data from the temporary cache to a database and organizes it. The database organizes and stores the behavioral history and attribute information for each user. It also updates the data index to speed up subsequent data retrieval and analysis processes.

[1450] Data analysis

[1451] The server analyzes the user's behavioral history and attribute information stored in the database. This analysis uses a machine learning algorithm to accurately estimate the user's attributes (e.g., age, gender, region) and areas of interest (e.g., hobbies, purchasing history). It also uses an emotion engine to recognize the user's emotions and analyze the emotions in the feedback data. The analysis results are stored back in the database and serve as the basis for generating supplementary prompts.

[1452] Supplemental prompt generation

[1453] The server obtains demographic information, interests, and sentiment analysis results related to a specific user. It then generates a supplemental prompt based on this information. A supplemental prompt is a detailed instruction that provides user-specific information to the AI ​​generator. For example, a prompt might read, "Generate an ad for this user, a 30-year-old male user, interested in technology and sports, whose current emotional state is stressed. For this user, please generate an ad introducing the latest gadgets that can help relieve stress."

[1454] Generative Artificial Intelligence

[1455] The server provides the generated supplemental prompts to the artificial intelligence generator, which generates optimal advertisements and content for the user based on the supplemental prompts. The generated advertisements and content contain specific and useful information in a form adapted to the user's needs, interests, and even recognized emotions.

[1456] Ad serving

[1457] The server formats the response generated by the artificial intelligence and sends it to the user's device, where the user can receive the most suitable advertisements and suggestions.

[1458] Feedback collection

[1459] The device tracks the user's reactions to the AI's responses (e.g., clicks, time spent, and feedback input) and their emotions at the time. The collected feedback and emotion data is sent to the server in real time. The server analyzes the feedback data and stores the results in a database. The analysis results are used to generate future prompts and improve the accuracy of the AI's responses.

[1460] This allows for highly accurate personalized responses and ad delivery based on users' behavioral history, attribute information, and emotional state, while also effectively collecting user feedback to continuously improve the system.

[1461] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1462] Step 1: Collect user data

[1463] Input: User behavior history and attribute information

[1464] How it works: To collect user behavior history and attribute information, the server first obtains the user's consent. The device then tracks the user's actions on websites and applications (e.g., page views, clicks, and form entries) in real time and sends the data to the server.

[1465] Output: Collected user behavior history and attribute information data

[1466] Step 2: Store the data

[1467] Input: Collected user data

[1468] Specific operation: The server temporarily caches the received user behavior history and attribute information, then organizes the data and stores it in a database. In the database, the data for each user is systematically stored.

[1469] Output: User data stored in the database

[1470] Step 3: Data analysis

[1471] Input: User data stored in a database

[1472] Specific operation: The server analyzes the user's behavioral history and attribute information stored in the database. Using machine learning algorithms, it estimates the user's attributes (age, gender, region, etc.) and areas of interest (hobbies, purchasing history, etc.) with high accuracy. It also uses an emotion engine to recognize the user's emotional state and analyze the emotions in the feedback data.

[1473] Output: Analysis results including user attributes, interests, and emotional state

[1474] Step 4: Generate supplemental prompts

[1475] Input: Analysis results (user attributes, interests, emotional state)

[1476] Specific behavior: The server generates a supplemental prompt based on the analysis results, including user-specific information. For example, a specific prompt might be generated: "A 30-year-old male user with an interest in technology and sports, whose current emotional state is stress. Please generate an advertisement for this user introducing the latest gadgets that can help relieve stress."

[1477] Output: Generated supplemental prompts

[1478] Step 5: Generative AI generates a response

[1479] Input: Supplementary prompt

[1480] Specific operation: The server provides the generated supplemental prompts to the AI ​​generator, which generates optimal advertisements and content for the user based on the provided prompts.

[1481] Output: Generated ads and content

[1482] Step 6: Sending the response results

[1483] Input: Generated ads and content

[1484] Specific operation: The server formats the generated advertisement and content and sends it to the user's device. The user receives the response and views it.

[1485] Output: Ads and content sent to the user's device

[1486] Step 7: Gather feedback

[1487] Input: User's reaction and emotional state

[1488] Specific operation: The device tracks the user's reactions to ads and responses from the generated AI (e.g., clicks, time spent, feedback input) and their emotional state at the time in real time, and sends the data to the server.

[1489] Output: Collected feedback and emotional state data

[1490] Step 8: Analyze the feedback data

[1491] Input: Feedback data and emotional state data

[1492] Specific operation: The server analyzes the collected feedback data and emotional state data and stores the results in a database. The analysis results are used to generate supplementary prompts and improve the accuracy of AI responses in the future.

[1493] Output: Parsed feedback data

[1494] 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.

[1495] 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.

[1496] 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.

[1497] 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.

[1498] 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.

[1499] 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.

[1500] 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).

[1501] 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.

[1502] 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."

[1503] 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.

[1504] 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).

[1505] 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.

[1506] 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.

[1507] 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.

[1508] 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.

[1509] 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.

[1510] 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.

[1511] 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.

[1512] 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.

[1513] 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.

[1514] 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.

[1515] The following is further disclosed regarding the above embodiment.

[1516] (Claim 1)

[1517] A means of collecting user behavior history and attribute information;

[1518] means for storing the collected data in a database;

[1519] A means for analyzing data stored in a database and estimating user attributes and interests;

[1520] a means for generating a follow-up prompt based on the estimated information;

[1521] A means for providing the generated supplemental prompt to a generating artificial intelligence to generate an optimal response for the user;

[1522] A means for transmitting a response result from the generating artificial intelligence to a user's terminal;

[1523] A system that includes a means of collecting user feedback, analyzing it, and storing it in a database.

[1524] (Claim 2)

[1525] 2. The system according to claim 1, further comprising means for collecting behavioral history and attribute information with the user's consent.

[1526] (Claim 3)

[1527] 10. The system of claim 1, further comprising means for estimating user attributes and interests using machine learning algorithms.

[1528] "Example 1"

[1529] (Claim 1)

[1530] A means of collecting user behavior history and attribute information;

[1531] A means for temporarily caching the collected data and storing it in a database;

[1532] A means for analyzing data stored in a database and using machine learning algorithms to estimate user attributes and interests;

[1533] a means for generating a follow-up prompt based on the estimated information;

[1534] A means for providing the generated supplemental prompt to a generating artificial intelligence to generate an optimal response for the user;

[1535] A means for formatting the response result of the generating artificial intelligence and transmitting it to the user's terminal;

[1536] A system that includes a means of collecting user feedback, analyzing it, and storing it in a database.

[1537] (Claim 2)

[1538] 2. The system according to claim 1, further comprising means for collecting behavioral history and attribute information with the user's consent.

[1539] (Claim 3)

[1540] 2. The system of claim 1, further comprising means for analyzing the feedback data and utilizing the analyzed data to generate subsequent prompts and improve response accuracy.

[1541] "Application Example 1"

[1542] (Claim 1)

[1543] A means of collecting user behavior history and attribute information;

[1544] means for storing the collected data in a database;

[1545] A means for analyzing data stored in a database and estimating user attributes and interests;

[1546] a means for generating a follow-up prompt based on the estimated information;

[1547] A means for providing the generated supplemental prompt to a generating artificial intelligence to generate an optimal response for the user;

[1548] A means for transmitting a response result from the generating artificial intelligence to a user's terminal;

[1549] A means of collecting user feedback, analyzing it and storing it in a database;

[1550] A means for making personalized product suggestions based on the user's behavioral history and attribute information;

[1551] A means to improve the accuracy of suggestions based on user feedback, and

[1552] A system including:

[1553] (Claim 2)

[1554] 2. The system according to claim 1, further comprising means for collecting behavioral history and attribute information with the user's consent.

[1555] (Claim 3)

[1556] 10. The system of claim 1, further comprising means for estimating user attributes and interests using machine learning algorithms.

[1557] "Example 2: Combining Emotion Engines"

[1558] (Claim 1)

[1559] A means of collecting user behavior history and attribute information;

[1560] a means for temporarily caching the collected data and subsequently storing it in a database;

[1561] a means for updating the data and indexes stored in the database;

[1562] A means for analyzing data stored in a database using a machine learning algorithm to estimate a user's attributes, interests, and emotions;

[1563] a means for generating a follow-up prompt based on the estimated information;

[1564] A means for providing the generated supplemental prompt to a generating artificial intelligence to generate an optimal response for the user;

[1565] A means for transmitting a response result from the generating artificial intelligence to a user's terminal;

[1566] A system that includes a means to collect feedback from users, analyze it, including emotional information, and store it in a database.

[1567] (Claim 2)

[1568] 2. The system according to claim 1, further comprising means for collecting behavioral history and attribute information with the user's consent.

[1569] (Claim 3)

[1570] 10. The system of claim 1, further comprising means for estimating user attributes, interests, and emotions using machine learning algorithms.

[1571] "Application example 2 when combining emotion engines"

[1572] (Claim 1)

[1573] A means of collecting user behavior history and attribute information;

[1574] means for storing the collected data in a database;

[1575] A means for analyzing data stored in a database and estimating user attributes and interests;

[1576] a means for generating a follow-up prompt based on the estimated information;

[1577] A means for providing the generated supplemental prompt to a generating artificial intelligence to generate an optimal response for the user;

[1578] A means for transmitting a response result from the generating artificial intelligence to a user's terminal;

[1579] A means of collecting user feedback, analyzing it and storing it in a database;

[1580] a means for recognizing the emotional state of a user;

[1581] A means for generating advertisements based on a user's behavioral history, attribute information, and emotional state;

[1582] A means for delivering the generated advertisement to a user's device;

[1583] A means of collecting users' reactions to advertisements and their emotional state

[1584] A system including:

[1585] (Claim 2)

[1586] 2. The system according to claim 1, further comprising means for collecting behavioral history and attribute information with the user's consent.

[1587] (Claim 3)

[1588] 10. The system of claim 1, further comprising means for estimating user attributes and interests using machine learning algorithms. [Explanation of symbols]

[1589] 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 of collecting user behavior history and attribute information; means for storing the collected data in a database; A means for analyzing data stored in a database and estimating user attributes and interests; a means for generating a follow-up prompt based on the estimated information; A means for providing the generated supplemental prompt to a generating artificial intelligence to generate an optimal response for the user; A means for transmitting a response result from the generating artificial intelligence to a user's terminal; A system that includes a means of collecting user feedback, analyzing it, and storing it in a database.

2. The system according to claim 1 , further comprising means for collecting behavioral history and attribute information with the user's consent.

3. The system of claim 1 , further comprising means for estimating user attributes and interests using a machine learning algorithm.

Images