system

Abstract

We provide the system. [Solution] A means of obtaining and analyzing the past activity records of posters who have evaluation information, A means for generating a highly-rated estimated avatar based on the activity record of the aforementioned poster, A means for forwarding the user's question to the aforementioned estimated avatar and generating an answer, A means of presenting the aforementioned answer to the user and collecting feedback, A means for analyzing the aforementioned feedback and using it to improve the response content of the estimated doppelganger, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] [In product reviews, it is difficult to obtain reliable information, and there is a problem that appropriate judgment cannot be made at the time of purchase. In addition, it is difficult for users to understand the extent to which the hobbies and preferences of the poster affect the review, and the current situation is that users have no choice but to rely on reviews with unclear substantial value. Due to such a situation, there is a problem that the reliability of purchase decisions based on reviews has decreased.]

Means for Solving the Problems

[0005] [This invention proposes providing a database system as a means for acquiring and analyzing the past activity records of posters who possess evaluation information, and using a generative AI model as a means for generating a highly-rated estimated avatar based on the poster's activity records. A natural language processing engine is utilized as a means for forwarding user questions to the estimated avatar and generating answers, and finally, a means for presenting the answers to the user and collecting feedback is provided. This makes it possible to analyze the feedback and improve the content of the estimated avatar's responses, thereby realizing a system that can provide users with more reliable information.]

[0006] "Rating information" refers to data on reliability and influence derived from past reviews and purchases made by the poster.

[0007] A "contributor" is [an individual or entity that has reviews or purchase history and provides information based on them].

[0008] "Activity log" refers to [the totality of the poster's review history, purchase history, and other related information].

[0009] A "presumed alter ego" is a virtual information provider created based on activity records, reflecting the poster's hobbies and preferences.

[0010] "Users" are consumers or end-users who seek information through the system.

[0011] A "generative AI model" is an artificial intelligence algorithm used to generate specific outputs based on data.

[0012] A "natural language processing engine" is a technology that understands, analyzes, and generates responses for human language. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of essential functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of essential functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of essential functions of a data processing device and a robot according to the fourth embodiment. [[ID=二十二]] [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), etc.

[0017] In the following embodiments, the labeled RAM (Random Access Memory) is a memory where information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, the labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0019] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), etc.

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the 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.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0027] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] This invention is a system that provides highly reliable information by utilizing the past activity records of posters who possess evaluation information. Specifically, it is implemented by a server, terminal, and user in the following manner.

[0035] The server will begin collecting data.

[0036] The server retrieves the poster's past activity records from the database. These activity records include reviews written by the poster, product purchase history, and review scores. The server analyzes the activity records using natural language processing technology to evaluate the poster's hobbies, preferences, and trustworthiness, and organizes the necessary data.

[0037] The server generates a hypothetical doppelganger.

[0038] Based on the collected data, the server uses a generative AI model to generate a simulated avatar that mimics the poster. This simulated avatar is configured to reflect the poster's tone, style, and evaluation information, enabling the provision of more realistic information.

[0039] The user interacts with an avatar to obtain information.

[0040] Users use a communication application via their terminal to input questions to their avatar. For example, if a user asks about the reliability of a particular home appliance, the server forwards the question to the appropriate avatar.

[0041] The server generates and provides answers based on the user's questions.

[0042] The server receives questions from users via a simulated avatar and analyzes the content of the questions. Based on the collected evaluation information and the poster's activity record, it generates a reliable answer. This answer includes specific product characteristics, benefits, or recommendations that address the user's question.

[0043] The device provides users with answers and facilitates feedback.

[0044] The generated responses are provided to the user via the device. User feedback is sent to the server via the device. This feedback is used to improve the accuracy and appeal of the estimated avatar's responses.

[0045] Specific example

[0046] For example, if a user wants to make a purchase decision based on reviews of a new smartphone's camera features, they might ask their virtual avatar via their device, "How good is the camera performance of this smartphone?" The server analyzes the activity records of posters who have previously written detailed reviews of that product, generates a response including a detailed evaluation of the camera features, and provides it to the user. This process allows the user to make a purchase decision based on reliable information.

[0047] The following describes the processing flow.

[0048] Step 1:

[0049] The server retrieves the poster's past activity records from the database. The data collected here includes reviews written by the poster, purchase history, and review scores.

[0050] Step 2:

[0051] The server uses natural language processing techniques to analyze the acquired activity logs. The analysis includes sentiment analysis of the review content, extraction of the poster's hobbies and preferences, and calculation of reliability.

[0052] Step 3:

[0053] The server uses the analysis results to generate an AI model that creates a simulated avatar of the poster. This simulated avatar is configured to provide realistic responses based on the poster's past activity records.

[0054] Step 4:

[0055] Users can use their devices to input and send questions about specific products through communication applications. For example, they can ask about product features or user experience.

[0056] Step 5:

[0057] The terminal forwards the user's input to the server. At this point, the question is recognized as an inquiry to the associated estimated avatar.

[0058] Step 6:

[0059] The server identifies relevant estimated avatars based on the received question and generates an answer. This utilizes insights gained from past activity logs and information based on review content.

[0060] Step 7:

[0061] The server sends the generated responses to the user's device for presentation. These responses include practical and specific product reviews and recommendations.

[0062] Step 8:

[0063] The terminal displays the response received from the server to the user and requests feedback on whether the response was useful.

[0064] Step 9:

[0065] Users enter feedback on the provided answers and send it back to the server via their device. This feedback is used to improve the estimated avatar.

[0066] Step 10:

[0067] The server analyzes the feedback it receives and incorporates the information to improve the accuracy of the estimated avatar's responses. It then optimizes the AI ​​model for future question-answering sessions.

[0068] (Example 1)

[0069] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0070] The reliability of information provided on the internet varies widely, making it difficult for users to select appropriate information when choosing products or services. In particular, when making decisions based on individual reviews and ratings, the criteria for judging the reliability of the poster are often ambiguous, potentially leading users to make incorrect choices. To address this problem, a system is needed that provides reliable information based on the poster's past activity records.

[0071] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0072] In this invention, the server includes means for acquiring past activity records of posters who have evaluation information and organizing them using database queries; means for tokenizing the activity records using natural language processing technology and evaluating the reliability of the posters through sentiment analysis; and means for generating highly-rated estimated avatars of posters using a generative AI model based on the analysis data. This enables users to make decisions based on highly reliable information.

[0073] "Rating information" refers to data that serves as an indicator for judging the reliability and quality of information provided by a poster, and is usually based on the content and scores of reviews.

[0074] "Past activity records" refers to the collective data including reviews and comments made by the poster in the past, as well as related digital history.

[0075] A "database query" is a structured search request used to retrieve information stored in a database.

[0076] "Natural language processing technology" refers to techniques that enable computers to understand and analyze human language, and includes processes such as text tokenization and content analysis.

[0077] "Sentiment analysis" is a technology that automatically determines emotions such as positive, negative, and neutral expressed within text.

[0078] A "generative AI model" is a model that uses artificial intelligence algorithms to automatically generate new information or text from given data.

[0079] A "presumed alter ego" is a virtual avatar with a personality created by a generative AI model that mimics the poster's past activity records and characteristics.

[0080] A "prompt statement" is a command or question that is entered into an AI or system to elicit a specific action or response.

[0081] "User interface" refers to the design and structure that serve as the point of contact for users to interact with a system, and includes display screens and operating methods.

[0082] "Feedback" refers to users' opinions and evaluations regarding the use of a system, and is data used to improve and optimize the system.

[0083] The system for implementing this invention mainly consists of three main elements: a server, a terminal, and a user.

[0084] The server manages and analyzes the data.

[0085] The server accesses the poster's past activity records stored in the database and uses SQL queries to organize the necessary data. The server uses Python and the Pandas library to construct the data, and leverages natural language processing techniques such as spaCy and NLTK to tokenize the posts and perform sentiment analysis to determine the poster's trustworthiness. Based on this, a generative AI model (e.g., GPT-3® or BERT) is used to generate an estimated avatar that reflects the poster's tone and style. This estimated avatar is then tailored to provide information relevant to the user's questions.

[0086] The terminal provides an interface with the user.

[0087] A terminal is an information processing device such as a smartphone or personal computer, and serves as the primary means for users to interact with the system. The terminal has the function of receiving user questions through communication applications (e.g., Slack or Zoom) and forwarding them to the server. Furthermore, the terminal provides a user interface, displays answers sent from the server, and also plays a role in collecting user feedback.

[0088] Users interpret information and provide feedback.

[0089] The user takes the role of requesting information by entering prompts through the terminal. For example, by prompting "Please tell me your review of this product," the user receives a detailed evaluation of a specific product from a simulated avatar. The user then makes a decision based on the information presented and provides feedback on their experience in that process to the server via the terminal. This feedback is used to improve the accuracy of the simulated avatar's responses and the reliability of the information.

[0090] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0091] Step 1:

[0092] The server retrieves the poster's past activity records from the database. It receives an identifier associated with a specific poster as input and executes SQL queries to collect data such as review content, purchase history, and review scores from the database. The output formats this data into a structured format (e.g., a data frame).

[0093] Step 2:

[0094] The server analyzes the acquired activity logs. It uses the data frame obtained in step 1 as input. Specifically, it uses a Python natural language processing library (e.g., spaCy) to tokenize the text data, perform sentiment analysis, and identify the trustworthiness of the poster. The output is a feature vector that quantifies trustworthiness and other characteristics.

[0095] Step 3:

[0096] The server generates an estimated avatar using a generative AI model based on feature vectors. The feature vectors obtained in step 2 are used as input. Specifically, the data is input to a generative AI model such as OpenAI's GPT to generate a virtual avatar that reflects the poster's tone of voice and evaluation information. The output is an AI model with the capability to generate text from this estimated avatar.

[0097] Step 4:

[0098] The user enters a prompt message via the terminal to obtain information. This input is a user query, such as "Please provide a review for this product." The terminal then forwards this information to the server.

[0099] Step 5:

[0100] The server receives the user's prompt and generates a response using an estimated avatar. The input is the prompt received in step 4. The server parses the query using natural language processing, extracts the necessary information from the estimated avatar, and generates a response for the user. The output is the interpreted text information to be presented to the user.

[0101] Step 6:

[0102] The terminal receives the response from the server and provides it to the user. Furthermore, it prompts the collection of feedback through the user interface. The input is the text response generated in step 5. A UI (user interface) is generated that is displayed to the user and allows them to provide feedback. As output, the feedback data collected from the user is sent to the server.

[0103] Step 7:

[0104] The server receives feedback and analyzes it to improve the estimated avatar's response. The input is the feedback data collected in step 6. The server statistically analyzes the feedback and adjusts the estimated avatar's parameters to improve response accuracy and reliability. The output is the improved AI model.

[0105] (Application Example 1)

[0106] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0107] A challenge is that users cannot quickly and easily obtain reliable information when selecting products, making purchasing decisions difficult. In particular, with a large amount of evaluation information and reviews available, it is difficult to determine which information is reliable.

[0108] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0109] In this invention, the server includes means for acquiring and analyzing the past activity history of a provider having evaluation information, means for generating a simulated personality with high evaluation information based on the provider's activity history, and means for providing detailed and reliable information for product selection using a mobile information terminal. This enables users to make quick purchasing decisions based on reliable information.

[0110] "Rating information" is an indicator of the poster's trustworthiness, calculated based on factors such as past activity history and the quality of reviews.

[0111] A "provider" refers to an individual or organization that posts information about a product or service and has a history of activity.

[0112] "Activity history" refers to a collection of information such as reviews, ratings, and related purchase history that the provider has recorded in the past.

[0113] "Analysis" is the process of analyzing activity history to extract and understand the characteristics and evaluation information of the provider.

[0114] A "simulated personality" is a virtual character generated by reproducing the tone and style of the provider, enabling the provision of reliable information in interactions with users.

[0115] "Personal information terminals" refer to electronic devices such as smartphones and tablets, which are devices used by users to obtain information.

[0116] "Reliable information" refers to accurate and trustworthy data backed by evaluation information that supports users' decision-making.

[0117] This system provides reliable information through communication between the server, terminal, and user. The server acquires provider data, including evaluation information and past activity history, and analyzes it using natural language processing technology. This analysis includes analyzing the provider's review content and purchase history. Through this analysis, the system identifies the provider's evaluation information and preferences, and generates a simulated personality using a generative AI model based on this information. This simulated personality reflects the provider's tone and style and interacts with the user.

[0118] The device functions as a personal digital assistant (Smartphone or tablet) and receives questions sent by the user through a communication application. These questions are forwarded to a server and analyzed by a simulated personality. The response generated based on the evaluation information is returned to the device and displayed to the user.

[0119] This system requires a backend server using cloud services such as AWS® to run, and OpenAI's GPT-4® is recommended as the generative AI model. For natural language processing, libraries such as spaCy can be used.

[0120] As a concrete example, consider a scenario where a user wants to know the evaluation of a new camera. The user asks, "What do you think of the performance of this camera?" via their device. This question is sent to a server, where a simulated personality extracts reliable data from past review information and generates a response in the form of, "This camera is highly rated for its high image quality, and is particularly good at shooting in low light."

[0121] An example of a prompt message might be: "User question: 'How is the performance of this camera?' Based on past reviews, generate a specific and reliable answer."

[0122] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0123] Step 1:

[0124] The server retrieves the provider's past activity history data. It collects review content, purchase history, and evaluation scores from the database and analyzes them using natural language processing techniques. The analysis calculates the provider's preferences and trustworthiness scores. The input is the provider's activity history, and the output is the analyzed evaluation information.

[0125] Step 2:

[0126] The server generates a simulated personality using a generative AI model based on the analysis results. This uses prompts that mimic the tone and style of the provider. The generative AI model outputs a simulated personality that can interact with the user. The input here is the analyzed evaluation information, and the output is the profile information of the simulated personality.

[0127] Step 3:

[0128] The user sends a question via a communication application using their device. The user enters a question about a product they are considering purchasing, and that question is forwarded to the server. The input is the user's question, and the output is the data passed to the server as the question.

[0129] Step 4:

[0130] The server receives user questions and generates answers through a simulated personality. It utilizes the provider's past evaluation information to create reliable answers. The input is the user's question and the simulated personality's profile information, and the output is an answer that includes detailed product information.

[0131] Step 5:

[0132] The terminal presents the user with the response received from the server. The user reviews the detailed information about the product and evaluates its reliability. The input is the response from the server, and the output is the information used by the user to make a purchasing decision.

[0133] Step 6:

[0134] The user sends feedback on the response via the terminal. The server receives this feedback, analyzes it as the initial input, and uses it to improve the accuracy of the simulated personality's responses. This feedback processing improves the accuracy of future responses. The input is the user's feedback, and the output is the updated profile of the simulated personality.

[0135] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0136] This invention is a system that generates a simulated avatar based on the poster's past activity record and combines it with an emotion engine that recognizes the user's emotions. The aim of this system is to provide more appropriate and personalized advice when users ask questions about products and services. Specific embodiments involving the server, terminal, and user are described below.

[0137] Data collection and analysis

[0138] The server collects the poster's past activity records from a database. This data includes review content, purchase history, and the poster's attribute information. The server analyzes this data using natural language processing technology to identify the poster's hobbies and preferences, and to calculate a reliability score.

[0139] Generation of a Presumed Doppelganger

[0140] The server generates an estimated avatar using a generative AI model based on the analyzed data. The estimated avatar reflects the poster's tone and style and has the ability to provide the user with the most relevant information. In addition, an emotion engine recognizes the user's emotions and reflects them in the avatar's responses, thereby improving the quality of the responses.

[0141] User interaction

[0142] Users can send questions to their virtual avatar through a communication application using their device. For example, if a user wants to know about a specific feature of a product, they can type and send that question. The server receives this and uses an emotion engine to determine the user's emotions from what they say.

[0143] Response generation and provision

[0144] The server generates an appropriate response through an estimated avatar based on the user's question and perceived emotions. The response is tailored to the user's preferences, taking into account the poster's rating and preference information. The generated response is delivered to the user via the terminal and includes content that matches the user's emotions.

[0145] Collecting and utilizing feedback

[0146] Users can provide feedback on the responses they receive. This feedback is sent from the device to the server, where it is analyzed by an emotion engine and used as data to improve the quality of future responses.

[0147] Specific example

[0148] For example, if a user asks, "I'm curious about the night photography performance of the new camera," the emotion engine recognizes the user's level of interest and expectations. The server analyzes data from past posters who have highly rated similar cameras and, through an estimated avatar, provides specific information tailored to the user's emotions, such as, "This camera has the ability to take clear photos even in low-light environments." This allows the user to receive a more appropriate and satisfying answer to their question.

[0149] The following describes the processing flow.

[0150] Step 1:

[0151] The server retrieves the poster's past activity records from the database. This data includes review text, rating scores, and information about purchased items.

[0152] Step 2:

[0153] The server analyzes the acquired activity records using natural language processing technology to identify the poster's evaluation information, hobbies, and preferences. This reveals which products the poster has detailed knowledge of and can be trusted.

[0154] Step 3:

[0155] The server uses an AI model based on the analysis results to generate an estimated avatar that reflects the poster's style and trustworthiness. This estimated avatar is a virtual information provider that possesses the poster's characteristics and is capable of realistic dialogue.

[0156] Step 4:

[0157] Users input and submit questions about specific products using a communication application on their device. The questions are free-form and can be specific, asking about the product's performance and user experience.

[0158] Step 5:

[0159] The terminal forwards the user's question to the server. At this time, the sent question is processed as an inquiry to a highly relevant estimated avatar.

[0160] Step 6:

[0161] The server uses an emotion engine to analyze the emotions expressed in the user's questions. This analysis identifies the user's level of interest, expectations, and anxieties.

[0162] Step 7:

[0163] The server reflects perceived emotions and generates answers to user questions through an estimated avatar. The responses are tailored to the user's emotions and interests, providing specific and practical information.

[0164] Step 8:

[0165] The device presents the generated response to the user. This response is provided in a format that is most helpful to the user, taking into account the results of the sentiment engine's analysis.

[0166] Step 9:

[0167] Users provide feedback on the answers they receive. By sending feedback to the server via their device, information that contributes to future improvements is provided.

[0168] Step 10:

[0169] The server analyzes the feedback and updates information to improve the accuracy of the estimated avatar's responses and emotion recognition. This improves the quality of question responses in subsequent sessions.

[0170] (Example 2)

[0171] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0172] Traditional information delivery systems faced the challenge of providing personalized responses to users. In particular, responses to user questions tended to be generic, failing to offer specific and useful information tailored to the user's emotions and preferences. Furthermore, mechanisms for effectively utilizing feedback to improve responses were insufficient.

[0173] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0174] In this invention, the server includes means for acquiring and analyzing the past behavioral history of an information provider having evaluation information; means for generating a virtual representation of the information provider with high evaluation information based on the information provider's behavioral history; and means for analyzing the user's emotions using emotion analysis technology and reflecting that in the response. This makes it possible to provide a personalized response that matches the user's emotions and preferences.

[0175] "Evaluation information" refers to an index that represents reliability and value calculated based on the information provider's past actions and evaluations.

[0176] "Information provider" refers to an individual or organization whose past activity history is used in order to provide information to users.

[0177] "Past activity history" refers to a record of the information provider's past activities and actions, including review content and purchase history.

[0178] A "virtual representation" refers to a simulated personality or profile created using artificial intelligence technology based on the information provider's past behavior and trustworthiness.

[0179] "Natural language processing technology" refers to all technologies that process and analyze human language using computers.

[0180] "Emotion analysis technology" is a technology that identifies and understands emotions from a user's words and actions.

[0181] "Generative AI technology" refers to technology that uses artificial intelligence algorithms to automatically generate text or content.

[0182] A "prompt statement" refers to an instruction or input statement given to an AI model to generate a specific output.

[0183] This invention is a system that provides personalized information to users by utilizing the past behavioral history of information providers. The detailed configuration for implementing this system is described below.

[0184] Data collection and analysis

[0185] The server uses a database management system to retrieve the past behavioral history of information providers. This data includes the content of reviews and purchase history submitted by the information providers. The server uses natural language processing libraries (e.g., NLTK and spaCy) to analyze the collected data and extract the information providers' preferences and reliability scores.

[0186] Generation of virtual representations

[0187] The server uses generative AI technology (e.g., large-scale language models) to generate a virtual representation based on the analyzed data. This virtual representation reflects the information provider's language style, interests, and preferences, and has the ability to provide information highly relevant to the user's question. Prompt statements are crucial for generation; a concrete example is "Generate a response in the same style as reviews previously conducted by [Information Provider Name]."

[0188] User-system interaction

[0189] Users can send questions to the system using a communication program on their device. The device transmits the user's question to the server, which uses sentiment analysis technology to analyze the user's emotions based on the content of the question.

[0190] Response generation and provision

[0191] The server constructs an appropriate response based on the generated virtual representation and the user's sentiment information. Generative AI technology leverages prompt text to refine the response and provide the most relevant information to the user. The response is sent to the user via the terminal and can be viewed on the user's screen.

[0192] Gathering and applying feedback

[0193] Users can provide feedback on the responses they receive on their devices. This feedback is sent to the server, which analyzes it to improve the quality of the responses. Specifically, the feedback information is used to adjust the AI ​​model that generates the responses and to improve the natural language processing algorithms.

[0194] In this way, by effectively combining generative AI technology and natural language processing technology while utilizing the behavioral history of information providers, it is possible to realize a system that provides users with personalized and valuable information.

[0195] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0196] Step 1:

[0197] The server retrieves the information provider's past behavioral history from the database. It receives the information provider's identification information as input, and based on this information, extracts review content, purchase history, and attribute information from the database using queries. As output, it formats this data into a dataset for analysis.

[0198] Step 2:

[0199] The server uses natural language processing technology to analyze the acquired behavioral history. Using the dataset from Step 1 as input, it tokenizes text data, extracts keywords, and performs sentiment analysis to identify the information provider's preferences and trustworthiness score. As output, metadata representing the analysis results is generated.

[0200] Step 3:

[0201] The server generates a virtual representation using a generative AI model. It receives the analysis results from step 2 as input and feeds the data into the AI ​​model using pre-configured prompts. The output is a virtual representation that reflects the information provider's language style and preferences.

[0202] Step 4:

[0203] The user sends a question to the system using a communication program on their terminal. The user inputs a specific question into the terminal as input, and the terminal transmits this to the server. The question arrives at the server as output.

[0204] Step 5:

[0205] The server analyzes the user's question using sentiment analysis technology. The input is the user's question text, and the server identifies the emotional state based on language patterns and keywords. The output generates data indicating the user's emotions.

[0206] Step 6:

[0207] The server constructs a response using a generative AI model based on virtual representations and user sentiment data. Virtual representations and sentiment analysis results are used as input, and prompts are used to instruct the AI ​​model to generate a response. The output is a response that is highly relevant to the user and matches their emotions.

[0208] Step 7:

[0209] The server sends the generated response to the terminal and presents it to the user. The input is the response generated in step 6, and its contents are displayed to the user through the terminal. The output is that the user can view the response.

[0210] Step 8:

[0211] The user provides feedback on the response provided via their device and sends it to the server. The input is the feedback content entered via the device and sent as data to the server. The output is the feedback data stored on the server.

[0212] Step 9:

[0213] The server analyzes feedback to improve response quality. It uses user feedback data as input and analyzes the feedback content using sentiment analysis techniques and machine learning algorithms. The output provides insights for improving future responses.

[0214] (Application Example 2)

[0215] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0216] Modern users have diverse tastes and emotions, and there is a demand for information tailored to these needs. However, conventional systems struggle to provide personalized information that responds to users' emotions. Therefore, there is a need for a system that can provide optimized information tailored to each user's individual emotional state and preferences.

[0217] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0218] In this invention, the server includes means for acquiring and analyzing the past activity records of posters who have evaluation information; means for generating a highly-rated avatar based on the poster's activity records and providing personalized advice that corresponds to the user's emotions using emotion analysis technology; and means for transferring the user's questions to the avatar and generating answers that take the user's emotions into consideration. This makes it possible to provide optimized information based on the individual user's emotions and preferences.

[0219] "Evaluation information" refers to an index that indicates the reliability and value of a poster, calculated based on the poster's past activities and acquired data.

[0220] "Past activity records" refers to the collective behavioral history of the poster, including reviews, purchase history, and attribute information.

[0221] A "presumed alter ego" is a virtual character created based on the poster's past activity records, tone of voice, and style, for the purpose of interacting with users.

[0222] "Sentiment analysis technology" is a technology that recognizes emotions from a user's text and actions, and uses that information to derive an appropriate response.

[0223] "Personalized advice" means taking into account the user's preferences and emotional state to provide the information and suggestions that are most suitable for that individual.

[0224] "Information terminal device" is a general term for communication-capable devices such as smartphones and tablets that users use to send questions.

[0225] A "generative AI model" is an artificial intelligence model that learns from large amounts of data to generate and provide optimal information about the user.

[0226] A "prompt message" is a text instruction given to an AI to tell it to generate specific information or respond in a particular way.

[0227] This invention is a system that provides personalized information to users. The program for realizing this system is constructed by combining specific methods and technologies.

[0228] First, the server uses a high-performance data processing server as hardware, and for software, it uses Python, natural language processing libraries (e.g., spaCy), and sentiment analysis libraries (e.g., TextBlob). This allows the server to retrieve the poster's past activity records and analyze data including rating information. This data includes the poster's reviews and past purchase history.

[0229] Next, the server generates a simulated avatar using a generative AI model based on the analyzed data. This AI model is built using TENSORFLOW®. The simulated avatar is a virtual character that reflects the poster's tone and style, and is useful for interacting with users. Utilizing sentiment analysis technology, the avatar recognizes the user's emotions and generates responses that correspond to those emotions.

[0230] The user submits a question using an information terminal device (e.g., smartphone, tablet). The terminal forwards the user's question to the server via a communication application with a simple interface. The server analyzes the received question, performs sentiment analysis, and identifies the user's emotional state.

[0231] Ultimately, the server generates an appropriate response based on the user's question and emotional state, and presents it to the user through an information terminal. The user receives the response and returns feedback to the server corresponding to their emotions. This feedback is analyzed to improve the quality of future responses.

[0232] For example, if a user asks, "I'm looking for a candle with a relaxing scent," the emotion engine recognizes the user's relaxed state and provides optimized advice such as, "This aromatherapy candle is perfect for relaxation time and has received high ratings." An example of a prompt used in this process would be, "Recognize the user's emotion as relaxed, please suggest a scenario for using the scented candle."

[0233] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0234] Step 1:

[0235] The server collects data from a database of the poster's past activity records. This data includes the poster's review content, purchase history, and attribute information. The server analyzes the data using Python and natural language processing libraries to calculate a reliability score. This allows the server to identify the poster's hobbies and preferences and calculate a reliability score based on the input activity records.

[0236] Step 2:

[0237] The server generates an estimated avatar using a generative AI model based on the data analyzed in Step 1. TensorFlow is used for the generative AI model. It receives the analysis results as input, generates the estimated avatar, and records the avatar information in the database. The generated estimated avatar reflects the poster's tone and style.

[0238] Step 3:

[0239] The user sends a question to the server using a communication application on their device. The device receives the entered question text and sends it to the server. Once the user's question is transferred from the device to the server, the process moves on to the next step.

[0240] Step 4:

[0241] The server analyzes the received question text using a sentiment analysis library to identify the user's emotions. The input is the question text, and the output is the user's emotional state (e.g., relaxed, excited, interested). This allows the server to recognize the user's emotions along with the question content.

[0242] Step 5:

[0243] The server generates appropriate responses through an estimated avatar based on the question and the perceived emotions. It receives the generation AI model and emotion analysis results as input, and constructs a personalized response by creating a prompt and presenting it to the AI ​​model. The response is then delivered to the user via the terminal.

[0244] Step 6:

[0245] The user submits feedback on the provided response. The device sends the text of the feedback to the server. The server collects the feedback and analyzes it using an emotion engine. Based on the analysis results, the server helps improve the content of the estimated avatar's responses. This makes it possible to use the feedback for future improvements.

[0246] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0247] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0248] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0249] [Second Embodiment]

[0250] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0251] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0252] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0253] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0254] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0255] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0256] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0257] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0258] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0259] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0260] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0261] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0262] This invention is a system that provides highly reliable information by utilizing the past activity records of posters who possess evaluation information. Specifically, it is implemented by a server, terminal, and user in the following manner.

[0263] The server will begin collecting data.

[0264] The server retrieves the poster's past activity records from the database. These activity records include reviews written by the poster, product purchase history, and review scores. The server analyzes the activity records using natural language processing technology to evaluate the poster's hobbies, preferences, and trustworthiness, and organizes the necessary data.

[0265] The server generates a hypothetical doppelganger.

[0266] Based on the collected data, the server uses a generative AI model to generate a simulated avatar that mimics the poster. This simulated avatar is configured to reflect the poster's tone, style, and evaluation information, enabling the provision of more realistic information.

[0267] The user interacts with an avatar to obtain information.

[0268] Users use a communication application via their terminal to input questions to their avatar. For example, if a user asks about the reliability of a particular home appliance, the server forwards the question to the appropriate avatar.

[0269] The server generates and provides answers based on the user's questions.

[0270] The server receives questions from users via a simulated avatar and analyzes the content of the questions. Based on the collected evaluation information and the poster's activity record, it generates a reliable answer. This answer includes specific product characteristics, benefits, or recommendations that address the user's question.

[0271] The device provides users with answers and facilitates feedback.

[0272] The generated responses are provided to the user via the device. User feedback is sent to the server via the device. This feedback is used to improve the accuracy and appeal of the estimated avatar's responses.

[0273] Specific example

[0274] For example, when a user wants to make a purchase decision based on a review of the camera function of a new smartphone, the user asks the virtual twin through the terminal, "How good is the camera performance of this smartphone?" The server analyzes the activity records of the posters who have previously reviewed the product in detail, generates an answer including a detailed evaluation of the camera function based on that information, and provides it to the user. Through this process, the user can make a purchase decision based on reliable information.

[0275] The following describes the processing flow.

[0276] Step 1:

[0277] The server obtains the past activity records of the posters from the database. The data collected here includes the reviews written by the posters, the history of the products purchased, and the review scores.

[0278] Step 2:

[0279] The server uses natural language processing technology to analyze the obtained activity records. In the analysis, sentiment analysis of the review content, the hobbies and preferences of the posters are extracted, and the reliability is calculated.

[0280] Step 3:

[0281] The server uses the generated AI model based on the analysis results to generate a virtual twin that mimics the poster. This virtual twin is configured to give a realistic response based on the past activity records of the poster.

[0282] Step 4:

[0283] The user uses the terminal to input and send a question about a specific product from the communication application. For example, questions can be asked about the features and usage feelings of the product.

[0284] Step 5:

[0285] The terminal transfers the question input by the user to the server. At this time, the question is recognized as an inquiry to the relevant estimated avatar.

[0286] Step 6:

[0287] Based on the received question, the server identifies the relevant estimated avatar and generates an answer. Here, information based on the knowledge and review content obtained from past activity records is utilized.

[0288] Step 7:

[0289] The server sends the generated answer to the terminal to present it to the user. The answer includes practical and specific product reviews and recommendations.

[0290] Step 8:

[0291] The terminal displays the answer received from the server to the user and requests feedback on whether the answer was useful.

[0292] Step 9:

[0293] The user inputs feedback on the provided answer and returns it to the server through the terminal. This feedback is used to improve the estimated avatar.

[0294] Step 10:

[0295] The server analyzes the obtained feedback and reflects the information to improve the response accuracy of the estimated avatar. The AI model is optimized for future question responses.

[0296] (Example 1)

[0297] Next, Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0298] The reliability of information provided on the internet varies widely, making it difficult for users to select appropriate information when choosing products or services. In particular, when making decisions based on individual reviews and ratings, the criteria for judging the reliability of the poster are often ambiguous, potentially leading users to make incorrect choices. To address this problem, a system is needed that provides reliable information based on the poster's past activity records.

[0299] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0300] In this invention, the server includes means for acquiring past activity records of posters who have evaluation information and organizing them using database queries; means for tokenizing the activity records using natural language processing technology and evaluating the reliability of the posters through sentiment analysis; and means for generating highly-rated estimated avatars of posters using a generative AI model based on the analysis data. This enables users to make decisions based on highly reliable information.

[0301] "Rating information" refers to data that serves as an indicator for judging the reliability and quality of information provided by a poster, and is usually based on the content and scores of reviews.

[0302] "Past activity records" refers to the collective data including reviews and comments made by the poster in the past, as well as related digital history.

[0303] A "database query" is a structured search request used to retrieve information stored in a database.

[0304] "Natural language processing technology" refers to techniques that enable computers to understand and analyze human language, and includes processes such as text tokenization and content analysis.

[0305] "Sentiment analysis" is a technology that automatically determines the sentiment in a text, such as positive, negative, or neutral.

[0306] A "generative AI model" is a model that uses artificial intelligence algorithms to automatically generate new information or text from given data.

[0307] An "estimated avatar" is an avatar with a virtual personality created by a generative AI model that mimics the past activity records and characteristics of the poster.

[0308] A "prompt sentence" is an instruction sentence or question sentence input to an AI or system to elicit a specific action or response.

[0309] A "user interface" refers to the design and structure that serves as the contact point for users to interact with the system, including the display screen and operation methods.

[0310] "Feedback" is the user's opinions and evaluations regarding the use of the system, and is data used for system improvement and optimization.

[0311] The system for implementing this invention mainly consists of three main elements: a server, a terminal, and a user.

[0312] The server manages and analyzes data

[0313] The server accesses the past activity records of the poster stored in the database and organizes the necessary data using SQL queries. The server constructs the data using Python and the Pandas library, and utilizes natural language processing technologies such as spaCy and NLTK to tokenize the posted content and perform sentiment analysis to identify the reliability of the poster. Thereby, an estimated avatar reflecting the tone and style of the poster is generated using a generative AI model (e.g., GPT-3 or BERT). This estimated avatar is adjusted to provide information suitable for the user's questions.

[0314] The terminal provides an interface with the user.

[0315] A terminal is an information processing device such as a smartphone or personal computer, and serves as the primary means for users to interact with the system. The terminal has the function of receiving user questions through communication applications (e.g., Slack or Zoom) and forwarding them to the server. Furthermore, the terminal provides a user interface, displays answers sent from the server, and also plays a role in collecting user feedback.

[0316] Users interpret information and provide feedback.

[0317] The user takes the role of requesting information by entering prompts through the terminal. For example, by prompting "Please tell me your review of this product," the user receives a detailed evaluation of a specific product from a simulated avatar. The user then makes a decision based on the information presented and provides feedback on their experience in that process to the server via the terminal. This feedback is used to improve the accuracy of the simulated avatar's responses and the reliability of the information.

[0318] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0319] Step 1:

[0320] The server retrieves the poster's past activity records from the database. It receives an identifier associated with a specific poster as input and executes SQL queries to collect data such as review content, purchase history, and review scores from the database. The output formats this data into a structured format (e.g., a data frame).

[0321] Step 2:

[0322] The server analyzes the acquired activity logs. It uses the data frame obtained in step 1 as input. Specifically, it uses a Python natural language processing library (e.g., spaCy) to tokenize the text data, perform sentiment analysis, and identify the trustworthiness of the poster. The output is a feature vector that quantifies trustworthiness and other characteristics.

[0323] Step 3:

[0324] The server generates an estimated avatar using a generative AI model based on feature vectors. The feature vectors obtained in step 2 are used as input. Specifically, the data is input to a generative AI model such as OpenAI's GPT to generate a virtual avatar that reflects the poster's tone of voice and evaluation information. The output is an AI model with the ability to generate text from this estimated avatar.

[0325] Step 4:

[0326] The user enters a prompt message via the terminal to obtain information. This input is a user query, such as "Please provide a review for this product." The terminal then forwards this information to the server.

[0327] Step 5:

[0328] The server receives the user's prompt and generates a response using an estimated avatar. The input is the prompt received in step 4. The server parses the query using natural language processing, extracts the necessary information from the estimated avatar, and generates a response for the user. The output is the interpreted text information to be presented to the user.

[0329] Step 6:

[0330] The terminal receives the response from the server and provides it to the user. Furthermore, it prompts the collection of feedback through the user interface. The input is the text response generated in step 5. A UI (user interface) is generated that is displayed to the user and allows them to provide feedback. As output, the feedback data collected from the user is sent to the server.

[0331] Step 7:

[0332] The server receives feedback and analyzes it to improve the estimated avatar's response. The input is the feedback data collected in step 6. The server statistically analyzes the feedback and adjusts the estimated avatar's parameters to improve response accuracy and reliability. The output is the improved AI model.

[0333] (Application Example 1)

[0334] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0335] A challenge is that users cannot quickly and easily obtain reliable information when selecting products, making purchasing decisions difficult. In particular, with a large amount of evaluation information and reviews available, it is difficult to determine which information is reliable.

[0336] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0337] In this invention, the server includes means for acquiring and analyzing the past activity history of a provider having evaluation information, means for generating a simulated personality with high evaluation information based on the provider's activity history, and means for providing detailed and reliable information for product selection using a mobile information terminal. This enables users to make quick purchasing decisions based on reliable information.

[0338] "Rating information" is an indicator of the poster's trustworthiness, calculated based on factors such as past activity history and the quality of reviews.

[0339] A "provider" refers to an individual or organization that posts information about a product or service and has a history of activity.

[0340] "Activity history" refers to a collection of information such as reviews, ratings, and related purchase history that the provider has recorded in the past.

[0341] "Analysis" is the process of analyzing activity history to extract and understand the characteristics and evaluation information of the provider.

[0342] A "simulated personality" is a virtual character generated by reproducing the tone and style of the provider, enabling the provision of reliable information in interactions with users.

[0343] "Personal information terminals" refer to electronic devices such as smartphones and tablets, which are devices used by users to obtain information.

[0344] "Reliable information" refers to accurate and trustworthy data backed by evaluation information that supports users' decision-making.

[0345] This system provides reliable information through communication between the server, terminal, and user. The server acquires provider data, including evaluation information and past activity history, and analyzes it using natural language processing technology. This analysis includes analyzing the provider's review content and purchase history. Through this analysis, the system identifies the provider's evaluation information and preferences, and generates a simulated personality using a generative AI model based on this information. This simulated personality reflects the provider's tone and style and interacts with the user.

[0346] The device functions as a personal digital assistant (Smartphone or tablet) and receives questions sent by the user through a communication application. These questions are forwarded to a server and analyzed by a simulated personality. The response generated based on the evaluation information is returned to the device and displayed to the user.

[0347] This system requires a backend server using a cloud service such as AWS to run, and OpenAI's GPT-4 is recommended as the generative AI model. For natural language processing, libraries such as spaCy can be used.

[0348] As a concrete example, consider a scenario where a user wants to know the evaluation of a new camera. The user asks, "What do you think of the performance of this camera?" via their device. This question is sent to a server, where a simulated personality extracts reliable data from past review information and generates a response in the form of, "This camera is highly rated for its high image quality, and is particularly good at shooting in low light."

[0349] An example of a prompt message might be: "User question: 'How is the performance of this camera?' Based on past reviews, generate a specific and reliable answer."

[0350] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0351] Step 1:

[0352] The server retrieves the provider's past activity history data. It collects review content, purchase history, and evaluation scores from the database and analyzes them using natural language processing techniques. The analysis calculates the provider's preferences and trustworthiness scores. The input is the provider's activity history, and the output is the analyzed evaluation information.

[0353] Step 2:

[0354] The server generates a simulated personality using a generative AI model based on the analysis results. This uses prompts that mimic the tone and style of the provider. The generative AI model outputs a simulated personality that can interact with the user. The input here is the analyzed evaluation information, and the output is the profile information of the simulated personality.

[0355] Step 3:

[0356] The user sends a question via a communication application using their device. The user enters a question about a product they are considering purchasing, and that question is forwarded to the server. The input is the user's question, and the output is the data passed to the server as the question.

[0357] Step 4:

[0358] The server receives user questions and generates answers through a simulated personality. It utilizes the provider's past evaluation information to create reliable answers. The input is the user's question and the simulated personality's profile information, and the output is an answer that includes detailed product information.

[0359] Step 5:

[0360] The terminal presents the user with the response received from the server. The user reviews the detailed information about the product and evaluates its reliability. The input is the response from the server, and the output is the information used by the user to make a purchasing decision.

[0361] Step 6:

[0362] The user sends feedback on the response via the terminal. The server receives this feedback, analyzes it as the initial input, and uses it to improve the accuracy of the simulated personality's responses. This feedback processing improves the accuracy of future responses. The input is the user's feedback, and the output is the updated profile of the simulated personality.

[0363] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0364] This invention is a system that generates a simulated avatar based on the poster's past activity record and combines it with an emotion engine that recognizes the user's emotions. The aim of this system is to provide more appropriate and personalized advice when users ask questions about products and services. Specific embodiments involving the server, terminal, and user are described below.

[0365] Data collection and analysis

[0366] The server collects the poster's past activity records from a database. This data includes review content, purchase history, and the poster's attribute information. The server analyzes this data using natural language processing technology to identify the poster's hobbies and preferences, and to calculate a reliability score.

[0367] Generation of a Presumed Doppelganger

[0368] The server generates an estimated avatar using a generative AI model based on the analyzed data. The estimated avatar reflects the poster's tone and style and has the ability to provide the user with the most relevant information. In addition, an emotion engine recognizes the user's emotions and reflects them in the avatar's responses, thereby improving the quality of the responses.

[0369] User interaction

[0370] Users can send questions to their virtual avatar through a communication application using their device. For example, if a user wants to know about a specific feature of a product, they can type and send that question. The server receives this and uses an emotion engine to determine the user's emotions from what they say.

[0371] Response generation and provision

[0372] The server generates an appropriate response through an estimated avatar based on the user's question and perceived emotions. The response is tailored to the user's preferences, taking into account the poster's rating and preference information. The generated response is delivered to the user via the terminal and includes content that matches the user's emotions.

[0373] Collecting and utilizing feedback

[0374] Users can provide feedback on the responses they receive. This feedback is sent from the device to the server, where it is analyzed by an emotion engine and used as data to improve the quality of future responses.

[0375] Specific example

[0376] For example, if a user asks, "I'm curious about the night photography performance of the new camera," the emotion engine recognizes the user's level of interest and expectations. The server analyzes data from past posters who have highly rated similar cameras and, through an estimated avatar, provides specific information tailored to the user's emotions, such as, "This camera has the ability to take clear photos even in low-light environments." This allows the user to receive a more appropriate and satisfying answer to their question.

[0377] The following describes the processing flow.

[0378] Step 1:

[0379] The server retrieves the poster's past activity records from the database. This data includes review text, rating scores, and information about purchased items.

[0380] Step 2:

[0381] The server analyzes the acquired activity records using natural language processing technology to identify the poster's evaluation information, hobbies, and preferences. This reveals which products the poster has detailed knowledge of and can be trusted.

[0382] Step 3:

[0383] The server uses an AI model based on the analysis results to generate an estimated avatar that reflects the poster's style and trustworthiness. This estimated avatar is a virtual information provider that possesses the poster's characteristics and is capable of realistic dialogue.

[0384] Step 4:

[0385] Users input and submit questions about specific products using a communication application on their device. The questions are free-form and can be specific, asking about the product's performance and user experience.

[0386] Step 5:

[0387] The terminal forwards the user's question to the server. At this time, the sent question is processed as an inquiry to a highly relevant estimated avatar.

[0388] Step 6:

[0389] The server uses an emotion engine to analyze the emotions expressed in the user's questions. This analysis identifies the user's level of interest, expectations, and anxieties.

[0390] Step 7:

[0391] The server reflects perceived emotions and generates answers to user questions through an estimated avatar. The responses are tailored to the user's emotions and interests, providing specific and practical information.

[0392] Step 8:

[0393] The device presents the generated response to the user. This response is provided in a format that is most helpful to the user, taking into account the results of the sentiment engine's analysis.

[0394] Step 9:

[0395] Users provide feedback on the answers they receive. By sending feedback to the server via their device, information that contributes to future improvements is provided.

[0396] Step 10:

[0397] The server analyzes the feedback and updates information to improve the accuracy of the estimated avatar's responses and emotion recognition. This improves the quality of question responses in subsequent sessions.

[0398] (Example 2)

[0399] Next, we will describe Example 2. 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".

[0400] Traditional information delivery systems faced the challenge of providing personalized responses to users. In particular, responses to user questions tended to be generic, failing to offer specific and useful information tailored to the user's emotions and preferences. Furthermore, mechanisms for effectively utilizing feedback to improve responses were insufficient.

[0401] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0402] In this invention, the server includes means for acquiring and analyzing the past behavioral history of an information provider having evaluation information; means for generating a virtual representation of the information provider with high evaluation information based on the information provider's behavioral history; and means for analyzing the user's emotions using emotion analysis technology and reflecting that in the response. This makes it possible to provide a personalized response that matches the user's emotions and preferences.

[0403] "Evaluation information" refers to an index that represents reliability and value calculated based on the information provider's past actions and evaluations.

[0404] "Information provider" refers to an individual or organization whose past activity history is used in order to provide information to users.

[0405] "Past activity history" refers to a record of the information provider's past activities and actions, including review content and purchase history.

[0406] A "virtual representation" refers to a simulated personality or profile created using artificial intelligence technology based on the information provider's past behavior and trustworthiness.

[0407] "Natural language processing technology" refers to all technologies that process and analyze human language using computers.

[0408] "Emotion analysis technology" is a technology that identifies and understands emotions from a user's words and actions.

[0409] "Generative AI technology" refers to technology that uses artificial intelligence algorithms to automatically generate text or content.

[0410] A "prompt statement" refers to an instruction or input statement given to an AI model to generate a specific output.

[0411] This invention is a system that provides personalized information to users by utilizing the past behavioral history of information providers. The detailed configuration for implementing this system is described below.

[0412] Data collection and analysis

[0413] The server uses a database management system to retrieve the past behavioral history of information providers. This data includes the content of reviews and purchase history submitted by the information providers. The server uses natural language processing libraries (e.g., NLTK and spaCy) to analyze the collected data and extract the information providers' preferences and reliability scores.

[0414] Generation of virtual representations

[0415] The server uses generative AI technology (e.g., large-scale language models) to generate a virtual representation based on the analyzed data. This virtual representation reflects the information provider's language style, interests, and preferences, and has the ability to provide information highly relevant to the user's question. Prompt statements are crucial for generation; a concrete example is "Generate a response in the same style as reviews previously conducted by [Information Provider Name]."

[0416] User-system interaction

[0417] Users can send questions to the system using a communication program on their device. The device transmits the user's question to the server, which uses sentiment analysis technology to analyze the user's emotions based on the content of the question.

[0418] Response generation and provision

[0419] The server constructs an appropriate response based on the generated virtual representation and the user's sentiment information. Generative AI technology leverages prompt text to refine the response and provide the most relevant information to the user. The response is sent to the user via the terminal and can be viewed on the user's screen.

[0420] Gathering and applying feedback

[0421] Users can provide feedback on the responses they receive on their devices. This feedback is sent to the server, which analyzes it to improve the quality of the responses. Specifically, the feedback information is used to adjust the AI ​​model that generates the responses and to improve the natural language processing algorithms.

[0422] In this way, by effectively combining generative AI technology and natural language processing technology while utilizing the behavioral history of information providers, it is possible to realize a system that provides users with personalized and valuable information.

[0423] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0424] Step 1:

[0425] The server retrieves the information provider's past behavioral history from the database. It receives the information provider's identification information as input, and based on this information, extracts review content, purchase history, and attribute information from the database using queries. As output, it formats this data into a dataset for analysis.

[0426] Step 2:

[0427] The server uses natural language processing technology to analyze the acquired behavioral history. Using the dataset from Step 1 as input, it tokenizes text data, extracts keywords, and performs sentiment analysis to identify the information provider's preferences and trustworthiness score. As output, metadata representing the analysis results is generated.

[0428] Step 3:

[0429] The server generates a virtual representation using a generative AI model. It receives the analysis results from step 2 as input and feeds the data into the AI ​​model using pre-configured prompts. The output is a virtual representation that reflects the information provider's language style and preferences.

[0430] Step 4:

[0431] The user sends a question to the system using a communication program on their terminal. The user inputs a specific question into the terminal as input, and the terminal transmits this to the server. The question arrives at the server as output.

[0432] Step 5:

[0433] The server analyzes the user's question using sentiment analysis technology. The input is the user's question text, and the server identifies the emotional state based on language patterns and keywords. The output generates data indicating the user's emotions.

[0434] Step 6:

[0435] The server constructs a response using a generative AI model based on virtual representations and user sentiment data. Virtual representations and sentiment analysis results are used as input, and prompts are used to instruct the AI ​​model to generate a response. The output is a response that is highly relevant to the user and matches their emotions.

[0436] Step 7:

[0437] The server sends the generated response to the terminal and presents it to the user. The input is the response generated in step 6, and its contents are displayed to the user through the terminal. The output is that the user can view the response.

[0438] Step 8:

[0439] The user provides feedback on the response provided via their device and sends it to the server. The input is the feedback content entered via the device and sent as data to the server. The output is the feedback data stored on the server.

[0440] Step 9:

[0441] The server analyzes feedback to improve response quality. It uses user feedback data as input and analyzes the feedback content using sentiment analysis techniques and machine learning algorithms. The output provides insights for improving future responses.

[0442] (Application Example 2)

[0443] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0444] Modern users have diverse tastes and emotions, and there is a demand for information tailored to these needs. However, conventional systems struggle to provide personalized information that responds to users' emotions. Therefore, there is a need for a system that can provide optimized information tailored to each user's individual emotional state and preferences.

[0445] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0446] In this invention, the server includes means for acquiring and analyzing the past activity records of posters who have evaluation information; means for generating a highly-rated avatar based on the poster's activity records and providing personalized advice that corresponds to the user's emotions using emotion analysis technology; and means for transferring the user's questions to the avatar and generating answers that take the user's emotions into consideration. This makes it possible to provide optimized information based on the individual user's emotions and preferences.

[0447] "Evaluation information" refers to an index that indicates the reliability and value of a poster, calculated based on the poster's past activities and acquired data.

[0448] "Past activity records" refers to the collective behavioral history of the poster, including reviews, purchase history, and attribute information.

[0449] A "presumed alter ego" is a virtual character created based on the poster's past activity records, tone of voice, and style, for the purpose of interacting with users.

[0450] "Sentiment analysis technology" is a technology that recognizes emotions from a user's text and actions, and uses that information to derive an appropriate response.

[0451] "Personalized advice" means taking into account the user's preferences and emotional state to provide the information and suggestions that are most suitable for that individual.

[0452] "Information terminal device" is a general term for communication-capable devices such as smartphones and tablets that users use to send questions.

[0453] A "generative AI model" is an artificial intelligence model that learns from large amounts of data to generate and provide optimal information about the user.

[0454] A "prompt message" is a text instruction given to an AI to tell it to generate specific information or respond in a particular way.

[0455] This invention is a system that provides personalized information to users. The program for realizing this system is constructed by combining specific methods and technologies.

[0456] First, the server uses a high-performance data processing server as hardware, and for software, it uses Python, natural language processing libraries (e.g., spaCy), and sentiment analysis libraries (e.g., TextBlob). This allows the server to retrieve the poster's past activity records and analyze data including rating information. This data includes the poster's reviews and past purchase history.

[0457] Next, the server generates a simulated avatar using a generative AI model based on the analyzed data. This AI model is built using TensorFlow. The simulated avatar is a virtual character that reflects the poster's tone and style, and is useful for interacting with the user. Utilizing sentiment analysis technology, the avatar recognizes the user's emotions and generates responses that correspond to those emotions.

[0458] The user submits a question using an information terminal device (e.g., smartphone, tablet). The terminal forwards the user's question to the server via a communication application with a simple interface. The server analyzes the received question, performs sentiment analysis, and identifies the user's emotional state.

[0459] Ultimately, the server generates an appropriate response based on the user's question and emotional state, and presents it to the user through an information terminal. The user receives the response and returns feedback to the server corresponding to their emotions. This feedback is analyzed to improve the quality of future responses.

[0460] For example, if a user asks, "I'm looking for a candle with a relaxing scent," the emotion engine recognizes the user's relaxed state and provides optimized advice such as, "This aromatherapy candle is perfect for relaxation time and has received high ratings." An example of a prompt used in this process would be, "Recognize the user's emotion as relaxed, please suggest a scenario for using the scented candle."

[0461] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0462] Step 1:

[0463] The server collects data from a database of the poster's past activity records. This data includes the poster's review content, purchase history, and attribute information. The server analyzes the data using Python and natural language processing libraries to calculate a reliability score. This allows the server to identify the poster's hobbies and preferences and calculate a reliability score based on the input activity records.

[0464] Step 2:

[0465] The server generates an estimated avatar using a generative AI model based on the data analyzed in Step 1. TensorFlow is used for the generative AI model. It receives the analysis results as input, generates the estimated avatar, and records the avatar information in the database. The generated estimated avatar reflects the poster's tone and style.

[0466] Step 3:

[0467] The user sends a question to the server using a communication application on their device. The device receives the entered question text and sends it to the server. Once the user's question is transferred from the device to the server, the process moves on to the next step.

[0468] Step 4:

[0469] The server analyzes the received question text using a sentiment analysis library to identify the user's emotions. The input is the question text, and the output is the user's emotional state (e.g., relaxed, excited, interested). This allows the server to recognize the user's emotions along with the question content.

[0470] Step 5:

[0471] The server generates appropriate responses through an estimated avatar based on the question and the perceived emotions. It receives the generation AI model and emotion analysis results as input, and constructs a personalized response by creating a prompt and presenting it to the AI ​​model. The response is then delivered to the user via the terminal.

[0472] Step 6:

[0473] The user submits feedback on the provided response. The device sends the text of the feedback to the server. The server collects the feedback and analyzes it using an emotion engine. Based on the analysis results, the server helps improve the content of the estimated avatar's responses. This makes it possible to use the feedback for future improvements.

[0474] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0475] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0476] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0477] [Third Embodiment]

[0478] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0479] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0480] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0481] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0482] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0483] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0484] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0485] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0486] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0487] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0488] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0489] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0490] This invention is a system that provides highly reliable information by utilizing the past activity records of posters who possess evaluation information. Specifically, it is implemented by a server, terminal, and user in the following manner.

[0491] The server will begin collecting data.

[0492] The server retrieves the poster's past activity records from the database. These activity records include reviews written by the poster, product purchase history, and review scores. The server analyzes the activity records using natural language processing technology to evaluate the poster's hobbies, preferences, and trustworthiness, and organizes the necessary data.

[0493] The server generates a hypothetical doppelganger.

[0494] Based on the collected data, the server uses a generative AI model to generate a simulated avatar that mimics the poster. This simulated avatar is configured to reflect the poster's tone, style, and evaluation information, enabling the provision of more realistic information.

[0495] The user interacts with an avatar to obtain information.

[0496] Users use a communication application via their terminal to input questions to their avatar. For example, if a user asks about the reliability of a particular home appliance, the server forwards the question to the appropriate avatar.

[0497] The server generates and provides answers based on the user's questions.

[0498] The server receives questions from users via a simulated avatar and analyzes the content of the questions. Based on the collected evaluation information and the poster's activity record, it generates a reliable answer. This answer includes specific product characteristics, benefits, or recommendations that address the user's question.

[0499] The device provides users with answers and facilitates feedback.

[0500] The generated responses are provided to the user via the device. User feedback is sent to the server via the device. This feedback is used to improve the accuracy and appeal of the estimated avatar's responses.

[0501] Specific example

[0502] For example, if a user wants to make a purchase decision based on reviews of a new smartphone's camera features, they might ask their virtual avatar via their device, "How good is the camera performance of this smartphone?" The server analyzes the activity records of posters who have previously written detailed reviews of that product, generates a response including a detailed evaluation of the camera features, and provides it to the user. This process allows the user to make a purchase decision based on reliable information.

[0503] The following describes the processing flow.

[0504] Step 1:

[0505] The server retrieves the poster's past activity records from the database. The data collected here includes reviews written by the poster, purchase history, and review scores.

[0506] Step 2:

[0507] The server uses natural language processing techniques to analyze the acquired activity logs. The analysis includes sentiment analysis of the review content, extraction of the poster's hobbies and preferences, and calculation of reliability.

[0508] Step 3:

[0509] The server uses the analysis results to generate an AI model that creates a simulated avatar of the poster. This simulated avatar is configured to provide realistic responses based on the poster's past activity records.

[0510] Step 4:

[0511] Users can use their devices to input and send questions about specific products through communication applications. For example, they can ask about product features or user experience.

[0512] Step 5:

[0513] The terminal forwards the user's input to the server. At this point, the question is recognized as an inquiry to the associated estimated avatar.

[0514] Step 6:

[0515] The server identifies relevant estimated avatars based on the received question and generates an answer. This utilizes insights gained from past activity logs and information based on review content.

[0516] Step 7:

[0517] The server sends the generated responses to the user's device for presentation. These responses include practical and specific product reviews and recommendations.

[0518] Step 8:

[0519] The terminal displays the response received from the server to the user and requests feedback on whether the response was useful.

[0520] Step 9:

[0521] Users enter feedback on the provided answers and send it back to the server via their device. This feedback is used to improve the estimated avatar.

[0522] Step 10:

[0523] The server analyzes the feedback it receives and incorporates the information to improve the accuracy of the estimated avatar's responses. It then optimizes the AI ​​model for future question-answering sessions.

[0524] (Example 1)

[0525] Next, we will describe Example 1. 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."

[0526] The reliability of information provided on the internet varies widely, making it difficult for users to select appropriate information when choosing products or services. In particular, when making decisions based on individual reviews and ratings, the criteria for judging the reliability of the poster are often ambiguous, potentially leading users to make incorrect choices. To address this problem, a system is needed that provides reliable information based on the poster's past activity records.

[0527] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0528] In this invention, the server includes means for acquiring past activity records of posters who have evaluation information and organizing them using database queries; means for tokenizing the activity records using natural language processing technology and evaluating the reliability of the posters through sentiment analysis; and means for generating highly-rated estimated avatars of posters using a generative AI model based on the analysis data. This enables users to make decisions based on highly reliable information.

[0529] "Rating information" refers to data that serves as an indicator for judging the reliability and quality of information provided by a poster, and is usually based on the content and scores of reviews.

[0530] "Past activity records" refers to the collective data including reviews and comments made by the poster in the past, as well as related digital history.

[0531] A "database query" is a structured search request used to retrieve information stored in a database.

[0532] "Natural language processing technology" refers to techniques that enable computers to understand and analyze human language, and includes processes such as text tokenization and content analysis.

[0533] "Sentiment analysis" is a technology that automatically determines emotions such as positive, negative, and neutral expressed within text.

[0534] A "generative AI model" is a model that uses artificial intelligence algorithms to automatically generate new information or text from given data.

[0535] A "presumed alter ego" is a virtual avatar with a personality created by a generative AI model that mimics the poster's past activity records and characteristics.

[0536] A "prompt statement" is a command or question that is entered into an AI or system to elicit a specific action or response.

[0537] "User interface" refers to the design and structure that serve as the point of contact for users to interact with a system, and includes display screens and operating methods.

[0538] "Feedback" refers to users' opinions and evaluations regarding the use of a system, and is data used to improve and optimize the system.

[0539] The system for implementing this invention mainly consists of three main elements: a server, a terminal, and a user.

[0540] The server manages and analyzes the data.

[0541] The server accesses the poster's past activity records stored in the database and uses SQL queries to organize the necessary data. The server uses Python and the Pandas library to construct the data, and leverages natural language processing techniques such as spaCy and NLTK to tokenize the posts and perform sentiment analysis to determine the poster's trustworthiness. Based on this, a generative AI model (e.g., GPT-3 or BERT) is used to generate an estimated avatar that reflects the poster's tone and style. This estimated avatar is then tailored to provide information relevant to the user's questions.

[0542] The terminal provides an interface with the user.

[0543] A terminal is an information processing device such as a smartphone or personal computer, and serves as the primary means for users to interact with the system. The terminal has the function of receiving user questions through communication applications (e.g., Slack or Zoom) and forwarding them to the server. Furthermore, the terminal provides a user interface, displays answers sent from the server, and also plays a role in collecting user feedback.

[0544] Users interpret information and provide feedback.

[0545] The user takes the role of requesting information by entering prompts through the terminal. For example, by prompting "Please tell me your review of this product," the user receives a detailed evaluation of a specific product from a simulated avatar. The user then makes a decision based on the information presented and provides feedback on their experience in that process to the server via the terminal. This feedback is used to improve the accuracy of the simulated avatar's responses and the reliability of the information.

[0546] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0547] Step 1:

[0548] The server retrieves the poster's past activity records from the database. It receives an identifier associated with a specific poster as input and executes SQL queries to collect data such as review content, purchase history, and review scores from the database. The output formats this data into a structured format (e.g., a data frame).

[0549] Step 2:

[0550] The server analyzes the acquired activity logs. It uses the data frame obtained in step 1 as input. Specifically, it uses a Python natural language processing library (e.g., spaCy) to tokenize the text data, perform sentiment analysis, and identify the trustworthiness of the poster. The output is a feature vector that quantifies trustworthiness and other characteristics.

[0551] Step 3:

[0552] The server generates an estimated avatar using a generative AI model based on feature vectors. The feature vectors obtained in step 2 are used as input. Specifically, the data is input to a generative AI model such as OpenAI's GPT to generate a virtual avatar that reflects the poster's tone of voice and evaluation information. The output is an AI model with the ability to generate text from this estimated avatar.

[0553] Step 4:

[0554] The user enters a prompt message via the terminal to obtain information. This input is a user query, such as "Please provide a review for this product." The terminal then forwards this information to the server.

[0555] Step 5:

[0556] The server receives the user's prompt and generates a response using an estimated avatar. The input is the prompt received in step 4. The server parses the query using natural language processing, extracts the necessary information from the estimated avatar, and generates a response for the user. The output is the interpreted text information to be presented to the user.

[0557] Step 6:

[0558] The terminal receives the response from the server and provides it to the user. Furthermore, it prompts the collection of feedback through the user interface. The input is the text response generated in step 5. A UI (user interface) is generated that is displayed to the user and allows them to provide feedback. As output, the feedback data collected from the user is sent to the server.

[0559] Step 7:

[0560] The server receives feedback and analyzes it to improve the estimated avatar's response. The input is the feedback data collected in step 6. The server statistically analyzes the feedback and adjusts the estimated avatar's parameters to improve response accuracy and reliability. The output is the improved AI model.

[0561] (Application Example 1)

[0562] Next, we will explain Application Example 1. In the following explanation, 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."

[0563] A challenge is that users cannot quickly and easily obtain reliable information when selecting products, making purchasing decisions difficult. In particular, with a large amount of evaluation information and reviews available, it is difficult to determine which information is reliable.

[0564] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0565] In this invention, the server includes means for acquiring and analyzing the past activity history of a provider having evaluation information, means for generating a simulated personality with high evaluation information based on the provider's activity history, and means for providing detailed and reliable information for product selection using a mobile information terminal. This enables users to make quick purchasing decisions based on reliable information.

[0566] "Rating information" is an indicator of the poster's trustworthiness, calculated based on factors such as past activity history and the quality of reviews.

[0567] A "provider" refers to an individual or organization that posts information about a product or service and has a history of activity.

[0568] "Activity history" refers to a collection of information such as reviews, ratings, and related purchase history that the provider has recorded in the past.

[0569] "Analysis" is the process of analyzing activity history to extract and understand the characteristics and evaluation information of the provider.

[0570] A "simulated personality" is a virtual character generated by reproducing the tone and style of the provider, enabling the provision of reliable information in interactions with users.

[0571] "Personal information terminals" refer to electronic devices such as smartphones and tablets, which are devices used by users to obtain information.

[0572] "Reliable information" refers to accurate and trustworthy data backed by evaluation information that supports users' decision-making.

[0573] This system provides reliable information through communication between the server, terminal, and user. The server acquires provider data, including evaluation information and past activity history, and analyzes it using natural language processing technology. This analysis includes analyzing the provider's review content and purchase history. Through this analysis, the system identifies the provider's evaluation information and preferences, and generates a simulated personality using a generative AI model based on this information. This simulated personality reflects the provider's tone and style and interacts with the user.

[0574] The device functions as a personal digital assistant (Smartphone or tablet) and receives questions sent by the user through a communication application. These questions are forwarded to a server and analyzed by a simulated personality. The response generated based on the evaluation information is returned to the device and displayed to the user.

[0575] This system requires a backend server using a cloud service such as AWS to run, and OpenAI's GPT-4 is recommended as the generative AI model. For natural language processing, libraries such as spaCy can be used.

[0576] As a concrete example, consider a scenario where a user wants to know the evaluation of a new camera. The user asks, "What do you think of the performance of this camera?" via their device. This question is sent to a server, where a simulated personality extracts reliable data from past review information and generates a response in the form of, "This camera is highly rated for its high image quality, and is particularly good at shooting in low light."

[0577] An example of a prompt message might be: "User question: 'How is the performance of this camera?' Based on past reviews, generate a specific and reliable answer."

[0578] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0579] Step 1:

[0580] The server retrieves the provider's past activity history data. It collects review content, purchase history, and evaluation scores from the database and analyzes them using natural language processing techniques. The analysis calculates the provider's preferences and trustworthiness scores. The input is the provider's activity history, and the output is the analyzed evaluation information.

[0581] Step 2:

[0582] The server generates a simulated personality using a generative AI model based on the analysis results. This uses prompts that mimic the tone and style of the provider. The generative AI model outputs a simulated personality that can interact with the user. The input here is the analyzed evaluation information, and the output is the profile information of the simulated personality.

[0583] Step 3:

[0584] The user sends a question via a communication application using their device. The user enters a question about a product they are considering purchasing, and that question is forwarded to the server. The input is the user's question, and the output is the data passed to the server as the question.

[0585] Step 4:

[0586] The server receives user questions and generates answers through a simulated personality. It utilizes the provider's past evaluation information to create reliable answers. The input is the user's question and the simulated personality's profile information, and the output is an answer that includes detailed product information.

[0587] Step 5:

[0588] The terminal presents the user with the response received from the server. The user reviews the detailed information about the product and evaluates its reliability. The input is the response from the server, and the output is the information used by the user to make a purchasing decision.

[0589] Step 6:

[0590] The user sends feedback on the response via the terminal. The server receives this feedback, analyzes it as the initial input, and uses it to improve the accuracy of the simulated personality's responses. This feedback processing improves the accuracy of future responses. The input is the user's feedback, and the output is the updated profile of the simulated personality.

[0591] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0592] This invention is a system that generates a simulated avatar based on the poster's past activity record and combines it with an emotion engine that recognizes the user's emotions. The aim of this system is to provide more appropriate and personalized advice when users ask questions about products and services. Specific embodiments involving the server, terminal, and user are described below.

[0593] Data collection and analysis

[0594] The server collects the poster's past activity records from a database. This data includes review content, purchase history, and the poster's attribute information. The server analyzes this data using natural language processing technology to identify the poster's hobbies and preferences, and to calculate a reliability score.

[0595] Generation of a Presumed Doppelganger

[0596] The server generates an estimated avatar using a generative AI model based on the analyzed data. The estimated avatar reflects the poster's tone and style and has the ability to provide the user with the most relevant information. In addition, an emotion engine recognizes the user's emotions and reflects them in the avatar's responses, thereby improving the quality of the responses.

[0597] User interaction

[0598] Users can send questions to their virtual avatar through a communication application using their device. For example, if a user wants to know about a specific feature of a product, they can type and send that question. The server receives this and uses an emotion engine to determine the user's emotions from what they say.

[0599] Response generation and provision

[0600] The server generates an appropriate response through an estimated avatar based on the user's question and perceived emotions. The response is tailored to the user's preferences, taking into account the poster's rating and preference information. The generated response is delivered to the user via the terminal and includes content that matches the user's emotions.

[0601] Collecting and utilizing feedback

[0602] Users can provide feedback on the responses they receive. This feedback is sent from the device to the server, where it is analyzed by an emotion engine and used as data to improve the quality of future responses.

[0603] Specific example

[0604] For example, if a user asks, "I'm curious about the night photography performance of the new camera," the emotion engine recognizes the user's level of interest and expectations. The server analyzes data from past posters who have highly rated similar cameras and, through an estimated avatar, provides specific information tailored to the user's emotions, such as, "This camera has the ability to take clear photos even in low-light environments." This allows the user to receive a more appropriate and satisfying answer to their question.

[0605] The following describes the processing flow.

[0606] Step 1:

[0607] The server retrieves the poster's past activity records from the database. This data includes review text, rating scores, and information about purchased items.

[0608] Step 2:

[0609] The server analyzes the acquired activity records using natural language processing technology to identify the poster's evaluation information, hobbies, and preferences. This reveals which products the poster has detailed knowledge of and can be trusted.

[0610] Step 3:

[0611] The server uses an AI model based on the analysis results to generate an estimated avatar that reflects the poster's style and trustworthiness. This estimated avatar is a virtual information provider that possesses the poster's characteristics and is capable of realistic dialogue.

[0612] Step 4:

[0613] Users input and submit questions about specific products using a communication application on their device. The questions are free-form and can be specific, asking about the product's performance and user experience.

[0614] Step 5:

[0615] The terminal forwards the user's question to the server. At this time, the sent question is processed as an inquiry to a highly relevant estimated avatar.

[0616] Step 6:

[0617] The server uses an emotion engine to analyze the emotions expressed in the user's questions. This analysis identifies the user's level of interest, expectations, and anxieties.

[0618] Step 7:

[0619] The server reflects perceived emotions and generates answers to user questions through an estimated avatar. The responses are tailored to the user's emotions and interests, providing specific and practical information.

[0620] Step 8:

[0621] The device presents the generated response to the user. This response is provided in a format that is most helpful to the user, taking into account the results of the sentiment engine's analysis.

[0622] Step 9:

[0623] Users provide feedback on the answers they receive. By sending feedback to the server via their device, information that contributes to future improvements is provided.

[0624] Step 10:

[0625] The server analyzes the feedback and updates information to improve the accuracy of the estimated avatar's responses and emotion recognition. This improves the quality of question responses in subsequent sessions.

[0626] (Example 2)

[0627] Next, we will describe Example 2. 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."

[0628] Traditional information delivery systems faced the challenge of providing personalized responses to users. In particular, responses to user questions tended to be generic, failing to offer specific and useful information tailored to the user's emotions and preferences. Furthermore, mechanisms for effectively utilizing feedback to improve responses were insufficient.

[0629] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0630] In this invention, the server includes means for acquiring and analyzing the past behavioral history of an information provider having evaluation information; means for generating a virtual representation of the information provider with high evaluation information based on the information provider's behavioral history; and means for analyzing the user's emotions using emotion analysis technology and reflecting that in the response. This makes it possible to provide a personalized response that matches the user's emotions and preferences.

[0631] "Evaluation information" refers to an index that represents reliability and value calculated based on the information provider's past actions and evaluations.

[0632] "Information provider" refers to an individual or organization whose past activity history is used in order to provide information to users.

[0633] "Past activity history" refers to a record of the information provider's past activities and actions, including review content and purchase history.

[0634] A "virtual representation" refers to a simulated personality or profile created using artificial intelligence technology based on the information provider's past behavior and trustworthiness.

[0635] "Natural language processing technology" refers to all technologies that process and analyze human language using computers.

[0636] "Emotion analysis technology" is a technology that identifies and understands emotions from a user's words and actions.

[0637] "Generative AI technology" refers to technology that uses artificial intelligence algorithms to automatically generate text or content.

[0638] A "prompt statement" refers to an instruction or input statement given to an AI model to generate a specific output.

[0639] This invention is a system that provides personalized information to users by utilizing the past behavioral history of information providers. The detailed configuration for implementing this system is described below.

[0640] Data collection and analysis

[0641] The server uses a database management system to retrieve the past behavioral history of information providers. This data includes the content of reviews and purchase history submitted by the information providers. The server uses natural language processing libraries (e.g., NLTK and spaCy) to analyze the collected data and extract the information providers' preferences and reliability scores.

[0642] Generation of virtual representations

[0643] The server uses generative AI technology (e.g., large-scale language models) to generate a virtual representation based on the analyzed data. This virtual representation reflects the information provider's language style, interests, and preferences, and has the ability to provide information highly relevant to the user's question. Prompt statements are crucial for generation; a concrete example is "Generate a response in the same style as reviews previously conducted by [Information Provider Name]."

[0644] User-system interaction

[0645] Users can send questions to the system using a communication program on their device. The device transmits the user's question to the server, which uses sentiment analysis technology to analyze the user's emotions based on the content of the question.

[0646] Response generation and provision

[0647] The server constructs an appropriate response based on the generated virtual representation and the user's sentiment information. Generative AI technology leverages prompt text to refine the response and provide the most relevant information to the user. The response is sent to the user via the terminal and can be viewed on the user's screen.

[0648] Gathering and applying feedback

[0649] Users can provide feedback on the responses they receive on their devices. This feedback is sent to the server, which analyzes it to improve the quality of the responses. Specifically, the feedback information is used to adjust the AI ​​model that generates the responses and to improve the natural language processing algorithms.

[0650] In this way, by effectively combining generative AI technology and natural language processing technology while utilizing the behavioral history of information providers, it is possible to realize a system that provides users with personalized and valuable information.

[0651] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0652] Step 1:

[0653] The server retrieves the information provider's past behavioral history from the database. It receives the information provider's identification information as input, and based on this information, extracts review content, purchase history, and attribute information from the database using queries. As output, it formats this data into a dataset for analysis.

[0654] Step 2:

[0655] The server uses natural language processing technology to analyze the acquired behavioral history. Using the dataset from Step 1 as input, it tokenizes text data, extracts keywords, and performs sentiment analysis to identify the information provider's preferences and trustworthiness score. As output, metadata representing the analysis results is generated.

[0656] Step 3:

[0657] The server generates a virtual representation using a generative AI model. It receives the analysis results from step 2 as input and feeds the data into the AI ​​model using pre-configured prompts. The output is a virtual representation that reflects the information provider's language style and preferences.

[0658] Step 4:

[0659] The user sends a question to the system using a communication program on their terminal. The user inputs a specific question into the terminal as input, and the terminal transmits this to the server. The question arrives at the server as output.

[0660] Step 5:

[0661] The server analyzes the user's question using sentiment analysis technology. The input is the user's question text, and the server identifies the emotional state based on language patterns and keywords. The output generates data indicating the user's emotions.

[0662] Step 6:

[0663] The server constructs a response using a generative AI model based on virtual representations and user sentiment data. Virtual representations and sentiment analysis results are used as input, and prompts are used to instruct the AI ​​model to generate a response. The output is a response that is highly relevant to the user and matches their emotions.

[0664] Step 7:

[0665] The server sends the generated response to the terminal and presents it to the user. The input is the response generated in step 6, and its contents are displayed to the user through the terminal. The output is that the user can view the response.

[0666] Step 8:

[0667] The user provides feedback on the response provided via their device and sends it to the server. The input is the feedback content entered via the device and sent as data to the server. The output is the feedback data stored on the server.

[0668] Step 9:

[0669] The server analyzes feedback to improve response quality. It uses user feedback data as input and analyzes the feedback content using sentiment analysis techniques and machine learning algorithms. The output provides insights for improving future responses.

[0670] (Application Example 2)

[0671] Next, we will explain application example 2. In the following explanation, 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."

[0672] Modern users have diverse tastes and emotions, and there is a demand for information tailored to these needs. However, conventional systems struggle to provide personalized information that responds to users' emotions. Therefore, there is a need for a system that can provide optimized information tailored to each user's individual emotional state and preferences.

[0673] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0674] In this invention, the server includes means for acquiring and analyzing the past activity records of posters who have evaluation information; means for generating a highly-rated avatar based on the poster's activity records and providing personalized advice that corresponds to the user's emotions using emotion analysis technology; and means for transferring the user's questions to the avatar and generating answers that take the user's emotions into consideration. This makes it possible to provide optimized information based on the individual user's emotions and preferences.

[0675] "Evaluation information" refers to an index that indicates the reliability and value of a poster, calculated based on the poster's past activities and acquired data.

[0676] "Past activity records" refers to the collective behavioral history of the poster, including reviews, purchase history, and attribute information.

[0677] A "presumed alter ego" is a virtual character created based on the poster's past activity records, tone of voice, and style, for the purpose of interacting with users.

[0678] "Sentiment analysis technology" is a technology that recognizes emotions from a user's text and actions, and uses that information to derive an appropriate response.

[0679] "Personalized advice" means taking into account the user's preferences and emotional state to provide the information and suggestions that are most suitable for that individual.

[0680] "Information terminal device" is a general term for communication-capable devices such as smartphones and tablets that users use to send questions.

[0681] A "generative AI model" is an artificial intelligence model that learns from large amounts of data to generate and provide optimal information about the user.

[0682] A "prompt message" is a text instruction given to an AI to tell it to generate specific information or respond in a particular way.

[0683] This invention is a system that provides personalized information to users. The program for realizing this system is constructed by combining specific methods and technologies.

[0684] First, the server uses a high-performance data processing server as hardware, and for software, it uses Python, natural language processing libraries (e.g., spaCy), and sentiment analysis libraries (e.g., TextBlob). This allows the server to retrieve the poster's past activity records and analyze data including rating information. This data includes the poster's reviews and past purchase history.

[0685] Next, the server generates a simulated avatar using a generative AI model based on the analyzed data. This AI model is built using TensorFlow. The simulated avatar is a virtual character that reflects the poster's tone and style, and is useful for interacting with the user. Utilizing sentiment analysis technology, the avatar recognizes the user's emotions and generates responses that correspond to those emotions.

[0686] The user submits a question using an information terminal device (e.g., smartphone, tablet). The terminal forwards the user's question to the server via a communication application with a simple interface. The server analyzes the received question, performs sentiment analysis, and identifies the user's emotional state.

[0687] Ultimately, the server generates an appropriate response based on the user's question and emotional state, and presents it to the user through an information terminal. The user receives the response and returns feedback to the server corresponding to their emotions. This feedback is analyzed to improve the quality of future responses.

[0688] For example, if a user asks, "I'm looking for a candle with a relaxing scent," the emotion engine recognizes the user's relaxed state and provides optimized advice such as, "This aromatherapy candle is perfect for relaxation time and has received high ratings." An example of a prompt used in this process would be, "Recognize the user's emotion as relaxed, please suggest a scenario for using the scented candle."

[0689] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0690] Step 1:

[0691] The server collects data from a database of the poster's past activity records. This data includes the poster's review content, purchase history, and attribute information. The server analyzes the data using Python and natural language processing libraries to calculate a reliability score. This allows the server to identify the poster's hobbies and preferences and calculate a reliability score based on the input activity records.

[0692] Step 2:

[0693] The server generates an estimated avatar using a generative AI model based on the data analyzed in Step 1. TensorFlow is used for the generative AI model. It receives the analysis results as input, generates the estimated avatar, and records the avatar information in the database. The generated estimated avatar reflects the poster's tone and style.

[0694] Step 3:

[0695] The user sends a question to the server using a communication application on their device. The device receives the entered question text and sends it to the server. Once the user's question is transferred from the device to the server, the process moves on to the next step.

[0696] Step 4:

[0697] The server analyzes the received question text using a sentiment analysis library to identify the user's emotions. The input is the question text, and the output is the user's emotional state (e.g., relaxed, excited, interested). This allows the server to recognize the user's emotions along with the question content.

[0698] Step 5:

[0699] The server generates appropriate responses through an estimated avatar based on the question and the perceived emotions. It receives the generation AI model and emotion analysis results as input, and constructs a personalized response by creating a prompt and presenting it to the AI ​​model. The response is then delivered to the user via the terminal.

[0700] Step 6:

[0701] The user submits feedback on the provided response. The device sends the text of the feedback to the server. The server collects the feedback and analyzes it using an emotion engine. Based on the analysis results, the server helps improve the content of the estimated avatar's responses. This makes it possible to use the feedback for future improvements.

[0702] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0703] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0704] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0705] [Fourth Embodiment]

[0706] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0707] As shown in Figure 7, the 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.

[0708] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0709] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0710] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0711] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0712] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0713] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0714] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0715] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0716] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0717] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0718] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0719] This invention is a system that provides highly reliable information by utilizing the past activity records of posters who possess evaluation information. Specifically, it is implemented by a server, terminal, and user in the following manner.

[0720] The server will begin collecting data.

[0721] The server retrieves the poster's past activity records from the database. These activity records include reviews written by the poster, product purchase history, and review scores. The server analyzes the activity records using natural language processing technology to evaluate the poster's hobbies, preferences, and trustworthiness, and organizes the necessary data.

[0722] The server generates a hypothetical doppelganger.

[0723] Based on the collected data, the server uses a generative AI model to generate a simulated avatar that mimics the poster. This simulated avatar is configured to reflect the poster's tone, style, and evaluation information, enabling the provision of more realistic information.

[0724] The user interacts with an avatar to obtain information.

[0725] Users use a communication application via their terminal to input questions to their avatar. For example, if a user asks about the reliability of a particular home appliance, the server forwards the question to the appropriate avatar.

[0726] The server generates and provides answers based on the user's questions.

[0727] The server receives questions from users via a simulated avatar and analyzes the content of the questions. Based on the collected evaluation information and the poster's activity record, it generates a reliable answer. This answer includes specific product characteristics, benefits, or recommendations that address the user's question.

[0728] The device provides users with answers and facilitates feedback.

[0729] The generated responses are provided to the user via the device. User feedback is sent to the server via the device. This feedback is used to improve the accuracy and appeal of the estimated avatar's responses.

[0730] Specific example

[0731] For example, if a user wants to make a purchase decision based on reviews of a new smartphone's camera features, they might ask their virtual avatar via their device, "How good is the camera performance of this smartphone?" The server analyzes the activity records of posters who have previously written detailed reviews of that product, generates a response including a detailed evaluation of the camera features, and provides it to the user. This process allows the user to make a purchase decision based on reliable information.

[0732] The following describes the processing flow.

[0733] Step 1:

[0734] The server retrieves the poster's past activity records from the database. The data collected here includes reviews written by the poster, purchase history, and review scores.

[0735] Step 2:

[0736] The server uses natural language processing techniques to analyze the acquired activity logs. The analysis includes sentiment analysis of the review content, extraction of the poster's hobbies and preferences, and calculation of reliability.

[0737] Step 3:

[0738] The server uses the analysis results to generate an AI model that creates a simulated avatar of the poster. This simulated avatar is configured to provide realistic responses based on the poster's past activity records.

[0739] Step 4:

[0740] Users can use their devices to input and send questions about specific products through communication applications. For example, they can ask about product features or user experience.

[0741] Step 5:

[0742] The terminal forwards the user's input to the server. At this point, the question is recognized as an inquiry to the associated estimated avatar.

[0743] Step 6:

[0744] The server identifies relevant estimated avatars based on the received question and generates an answer. This utilizes insights gained from past activity logs and information based on review content.

[0745] Step 7:

[0746] The server sends the generated responses to the user's device for presentation. These responses include practical and specific product reviews and recommendations.

[0747] Step 8:

[0748] The terminal displays the response received from the server to the user and requests feedback on whether the response was useful.

[0749] Step 9:

[0750] Users enter feedback on the provided answers and send it back to the server via their device. This feedback is used to improve the estimated avatar.

[0751] Step 10:

[0752] The server analyzes the feedback it receives and incorporates the information to improve the accuracy of the estimated avatar's responses. It then optimizes the AI ​​model for future question-answering sessions.

[0753] (Example 1)

[0754] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0755] The reliability of information provided on the internet varies widely, making it difficult for users to select appropriate information when choosing products or services. In particular, when making decisions based on individual reviews and ratings, the criteria for judging the reliability of the poster are often ambiguous, potentially leading users to make incorrect choices. To address this problem, a system is needed that provides reliable information based on the poster's past activity records.

[0756] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0757] In this invention, the server includes means for acquiring past activity records of posters who have evaluation information and organizing them using database queries; means for tokenizing the activity records using natural language processing technology and evaluating the reliability of the posters through sentiment analysis; and means for generating highly-rated estimated avatars of posters using a generative AI model based on the analysis data. This enables users to make decisions based on highly reliable information.

[0758] "Rating information" refers to data that serves as an indicator for judging the reliability and quality of information provided by a poster, and is usually based on the content and scores of reviews.

[0759] "Past activity records" refers to the collective data including reviews and comments made by the poster in the past, as well as related digital history.

[0760] A "database query" is a structured search request used to retrieve information stored in a database.

[0761] "Natural language processing technology" refers to techniques that enable computers to understand and analyze human language, and includes processes such as text tokenization and content analysis.

[0762] "Sentiment analysis" is a technology that automatically determines emotions such as positive, negative, and neutral expressed within text.

[0763] A "generative AI model" is a model that uses artificial intelligence algorithms to automatically generate new information or text from given data.

[0764] A "presumed alter ego" is a virtual avatar with a personality created by a generative AI model that mimics the poster's past activity records and characteristics.

[0765] A "prompt statement" is a command or question that is entered into an AI or system to elicit a specific action or response.

[0766] "User interface" refers to the design and structure that serve as the point of contact for users to interact with a system, and includes display screens and operating methods.

[0767] "Feedback" refers to users' opinions and evaluations regarding the use of a system, and is data used to improve and optimize the system.

[0768] The system for implementing this invention mainly consists of three main elements: a server, a terminal, and a user.

[0769] The server manages and analyzes the data.

[0770] The server accesses the poster's past activity records stored in the database and uses SQL queries to organize the necessary data. The server uses Python and the Pandas library to construct the data, and leverages natural language processing techniques such as spaCy and NLTK to tokenize the posts and perform sentiment analysis to determine the poster's trustworthiness. Based on this, a generative AI model (e.g., GPT-3 or BERT) is used to generate an estimated avatar that reflects the poster's tone and style. This estimated avatar is then tailored to provide information relevant to the user's questions.

[0771] The terminal provides an interface with the user.

[0772] A terminal is an information processing device such as a smartphone or personal computer, and serves as the primary means for users to interact with the system. The terminal has the function of receiving user questions through communication applications (e.g., Slack or Zoom) and forwarding them to the server. Furthermore, the terminal provides a user interface, displays answers sent from the server, and also plays a role in collecting user feedback.

[0773] Users interpret information and provide feedback.

[0774] The user takes the role of requesting information by entering prompts through the terminal. For example, by prompting "Please tell me your review of this product," the user receives a detailed evaluation of a specific product from a simulated avatar. The user then makes a decision based on the information presented and provides feedback on their experience in that process to the server via the terminal. This feedback is used to improve the accuracy of the simulated avatar's responses and the reliability of the information.

[0775] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0776] Step 1:

[0777] The server retrieves the poster's past activity records from the database. It receives an identifier associated with a specific poster as input and executes SQL queries to collect data such as review content, purchase history, and review scores from the database. The output formats this data into a structured format (e.g., a data frame).

[0778] Step 2:

[0779] The server analyzes the acquired activity logs. It uses the data frame obtained in step 1 as input. Specifically, it uses a Python natural language processing library (e.g., spaCy) to tokenize the text data, perform sentiment analysis, and identify the trustworthiness of the poster. The output is a feature vector that quantifies trustworthiness and other characteristics.

[0780] Step 3:

[0781] The server generates an estimated avatar using a generative AI model based on feature vectors. The feature vectors obtained in step 2 are used as input. Specifically, the data is input to a generative AI model such as OpenAI's GPT to generate a virtual avatar that reflects the poster's tone of voice and evaluation information. The output is an AI model with the ability to generate text from this estimated avatar.

[0782] Step 4:

[0783] The user enters a prompt message via the terminal to obtain information. This input is a user query, such as "Please provide a review for this product." The terminal then forwards this information to the server.

[0784] Step 5:

[0785] The server receives the user's prompt and generates a response using an estimated avatar. The input is the prompt received in step 4. The server parses the query using natural language processing, extracts the necessary information from the estimated avatar, and generates a response for the user. The output is the interpreted text information to be presented to the user.

[0786] Step 6:

[0787] The terminal receives the response from the server and provides it to the user. Furthermore, it prompts the collection of feedback through the user interface. The input is the text response generated in step 5. A UI (user interface) is generated that is displayed to the user and allows them to provide feedback. As output, the feedback data collected from the user is sent to the server.

[0788] Step 7:

[0789] The server receives feedback and analyzes it to improve the estimated avatar's response. The input is the feedback data collected in step 6. The server statistically analyzes the feedback and adjusts the estimated avatar's parameters to improve response accuracy and reliability. The output is the improved AI model.

[0790] (Application Example 1)

[0791] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0792] A challenge is that users cannot quickly and easily obtain reliable information when selecting products, making purchasing decisions difficult. In particular, with a large amount of evaluation information and reviews available, it is difficult to determine which information is reliable.

[0793] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0794] In this invention, the server includes means for acquiring and analyzing the past activity history of a provider having evaluation information, means for generating a simulated personality with high evaluation information based on the provider's activity history, and means for providing detailed and reliable information for product selection using a mobile information terminal. This enables users to make quick purchasing decisions based on reliable information.

[0795] "Rating information" is an indicator of the poster's trustworthiness, calculated based on factors such as past activity history and the quality of reviews.

[0796] A "provider" refers to an individual or organization that posts information about a product or service and has a history of activity.

[0797] "Activity history" refers to a collection of information such as reviews, ratings, and related purchase history that the provider has recorded in the past.

[0798] "Analysis" is the process of analyzing activity history to extract and understand the characteristics and evaluation information of the provider.

[0799] A "simulated personality" is a virtual character generated by reproducing the tone and style of the provider, enabling the provision of reliable information in interactions with users.

[0800] "Personal information terminals" refer to electronic devices such as smartphones and tablets, which are devices used by users to obtain information.

[0801] "Reliable information" refers to accurate and trustworthy data backed by evaluation information that supports users' decision-making.

[0802] This system provides reliable information through communication between the server, terminal, and user. The server acquires provider data, including evaluation information and past activity history, and analyzes it using natural language processing technology. This analysis includes analyzing the provider's review content and purchase history. Through this analysis, the system identifies the provider's evaluation information and preferences, and generates a simulated personality using a generative AI model based on this information. This simulated personality reflects the provider's tone and style and interacts with the user.

[0803] The device functions as a personal digital assistant (Smartphone or tablet) and receives questions sent by the user through a communication application. These questions are forwarded to a server and analyzed by a simulated personality. The response generated based on the evaluation information is returned to the device and displayed to the user.

[0804] This system requires a backend server using a cloud service such as AWS to run, and OpenAI's GPT-4 is recommended as the generative AI model. For natural language processing, libraries such as spaCy can be used.

[0805] As a concrete example, consider a scenario where a user wants to know the evaluation of a new camera. The user asks, "What do you think of the performance of this camera?" via their device. This question is sent to a server, where a simulated personality extracts reliable data from past review information and generates a response in the form of, "This camera is highly rated for its high image quality, and is particularly good at shooting in low light."

[0806] An example of a prompt message might be: "User question: 'How is the performance of this camera?' Based on past reviews, generate a specific and reliable answer."

[0807] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0808] Step 1:

[0809] The server retrieves the provider's past activity history data. It collects review content, purchase history, and evaluation scores from the database and analyzes them using natural language processing techniques. The analysis calculates the provider's preferences and trustworthiness scores. The input is the provider's activity history, and the output is the analyzed evaluation information.

[0810] Step 2:

[0811] The server generates a simulated personality using a generative AI model based on the analysis results. This uses prompts that mimic the tone and style of the provider. The generative AI model outputs a simulated personality that can interact with the user. The input here is the analyzed evaluation information, and the output is the profile information of the simulated personality.

[0812] Step 3:

[0813] The user sends a question via a communication application using their device. The user enters a question about a product they are considering purchasing, and that question is forwarded to the server. The input is the user's question, and the output is the data passed to the server as the question.

[0814] Step 4:

[0815] The server receives user questions and generates answers through a simulated personality. It utilizes the provider's past evaluation information to create reliable answers. The input is the user's question and the simulated personality's profile information, and the output is an answer that includes detailed product information.

[0816] Step 5:

[0817] The terminal presents the user with the response received from the server. The user reviews the detailed information about the product and evaluates its reliability. The input is the response from the server, and the output is the information used by the user to make a purchasing decision.

[0818] Step 6:

[0819] The user sends feedback on the response via the terminal. The server receives this feedback, analyzes it as the initial input, and uses it to improve the accuracy of the simulated personality's responses. This feedback processing improves the accuracy of future responses. The input is the user's feedback, and the output is the updated profile of the simulated personality.

[0820] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0821] This invention is a system that generates a simulated avatar based on the poster's past activity record and combines it with an emotion engine that recognizes the user's emotions. The aim of this system is to provide more appropriate and personalized advice when users ask questions about products and services. Specific embodiments involving the server, terminal, and user are described below.

[0822] Data collection and analysis

[0823] The server collects the poster's past activity records from a database. This data includes review content, purchase history, and the poster's attribute information. The server analyzes this data using natural language processing technology to identify the poster's hobbies and preferences, and to calculate a reliability score.

[0824] Generation of a Presumed Doppelganger

[0825] The server generates an estimated avatar using a generative AI model based on the analyzed data. The estimated avatar reflects the poster's tone and style and has the ability to provide the user with the most relevant information. In addition, an emotion engine recognizes the user's emotions and reflects them in the avatar's responses, thereby improving the quality of the responses.

[0826] User interaction

[0827] Users can send questions to their virtual avatar through a communication application using their device. For example, if a user wants to know about a specific feature of a product, they can type and send that question. The server receives this and uses an emotion engine to determine the user's emotions from what they say.

[0828] Response generation and provision

[0829] The server generates an appropriate response through an estimated avatar based on the user's question and perceived emotions. The response is tailored to the user's preferences, taking into account the poster's rating and preference information. The generated response is delivered to the user via the terminal and includes content that matches the user's emotions.

[0830] Collecting and utilizing feedback

[0831] Users can provide feedback on the responses they receive. This feedback is sent from the device to the server, where it is analyzed by an emotion engine and used as data to improve the quality of future responses.

[0832] Specific example

[0833] For example, if a user asks, "I'm curious about the night photography performance of the new camera," the emotion engine recognizes the user's level of interest and expectations. The server analyzes data from past posters who have highly rated similar cameras and, through an estimated avatar, provides specific information tailored to the user's emotions, such as, "This camera has the ability to take clear photos even in low-light environments." This allows the user to receive a more appropriate and satisfying answer to their question.

[0834] The following describes the processing flow.

[0835] Step 1:

[0836] The server retrieves the poster's past activity records from the database. This data includes review text, rating scores, and information about purchased items.

[0837] Step 2:

[0838] The server analyzes the acquired activity records using natural language processing technology to identify the poster's evaluation information, hobbies, and preferences. This reveals which products the poster has detailed knowledge of and can be trusted.

[0839] Step 3:

[0840] The server uses an AI model based on the analysis results to generate an estimated avatar that reflects the poster's style and trustworthiness. This estimated avatar is a virtual information provider that possesses the poster's characteristics and is capable of realistic dialogue.

[0841] Step 4:

[0842] Users input and submit questions about specific products using a communication application on their device. The questions are free-form and can be specific, asking about the product's performance and user experience.

[0843] Step 5:

[0844] The terminal forwards the user's question to the server. At this time, the sent question is processed as an inquiry to a highly relevant estimated avatar.

[0845] Step 6:

[0846] The server uses an emotion engine to analyze the emotions expressed in the user's questions. This analysis identifies the user's level of interest, expectations, and anxieties.

[0847] Step 7:

[0848] The server reflects perceived emotions and generates answers to user questions through an estimated avatar. The responses are tailored to the user's emotions and interests, providing specific and practical information.

[0849] Step 8:

[0850] The device presents the generated response to the user. This response is provided in a format that is most helpful to the user, taking into account the results of the sentiment engine's analysis.

[0851] Step 9:

[0852] Users provide feedback on the answers they receive. By sending feedback to the server via their device, information that contributes to future improvements is provided.

[0853] Step 10:

[0854] The server analyzes the feedback and updates information to improve the accuracy of the estimated avatar's responses and emotion recognition. This improves the quality of question responses in subsequent sessions.

[0855] (Example 2)

[0856] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0857] Traditional information delivery systems faced the challenge of providing personalized responses to users. In particular, responses to user questions tended to be generic, failing to offer specific and useful information tailored to the user's emotions and preferences. Furthermore, mechanisms for effectively utilizing feedback to improve responses were insufficient.

[0858] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0859] In this invention, the server includes means for acquiring and analyzing the past behavioral history of an information provider having evaluation information; means for generating a virtual representation of the information provider with high evaluation information based on the information provider's behavioral history; and means for analyzing the user's emotions using emotion analysis technology and reflecting that in the response. This makes it possible to provide a personalized response that matches the user's emotions and preferences.

[0860] "Evaluation information" refers to an index that represents reliability and value calculated based on the information provider's past actions and evaluations.

[0861] "Information provider" refers to an individual or organization whose past activity history is used in order to provide information to users.

[0862] "Past activity history" refers to a record of the information provider's past activities and actions, including review content and purchase history.

[0863] A "virtual representation" refers to a simulated personality or profile created using artificial intelligence technology based on the information provider's past behavior and trustworthiness.

[0864] "Natural language processing technology" refers to all technologies that process and analyze human language using computers.

[0865] "Emotion analysis technology" is a technology that identifies and understands emotions from a user's words and actions.

[0866] "Generative AI technology" refers to technology that uses artificial intelligence algorithms to automatically generate text or content.

[0867] A "prompt statement" refers to an instruction or input statement given to an AI model to generate a specific output.

[0868] This invention is a system that provides personalized information to users by utilizing the past behavioral history of information providers. The detailed configuration for implementing this system is described below.

[0869] Data collection and analysis

[0870] The server uses a database management system to retrieve the past behavioral history of information providers. This data includes the content of reviews and purchase history submitted by the information providers. The server uses natural language processing libraries (e.g., NLTK and spaCy) to analyze the collected data and extract the information providers' preferences and reliability scores.

[0871] Generation of virtual representations

[0872] The server uses generative AI technology (e.g., large-scale language models) to generate a virtual representation based on the analyzed data. This virtual representation reflects the information provider's language style, interests, and preferences, and has the ability to provide information highly relevant to the user's question. Prompt statements are crucial for generation; a concrete example is "Generate a response in the same style as reviews previously conducted by [Information Provider Name]."

[0873] User-system interaction

[0874] Users can send questions to the system using a communication program on their device. The device transmits the user's question to the server, which uses sentiment analysis technology to analyze the user's emotions based on the content of the question.

[0875] Response generation and provision

[0876] The server constructs an appropriate response based on the generated virtual representation and the user's sentiment information. Generative AI technology leverages prompt text to refine the response and provide the most relevant information to the user. The response is sent to the user via the terminal and can be viewed on the user's screen.

[0877] Gathering and applying feedback

[0878] Users can provide feedback on the responses they receive on their devices. This feedback is sent to the server, which analyzes it to improve the quality of the responses. Specifically, the feedback information is used to adjust the AI ​​model that generates the responses and to improve the natural language processing algorithms.

[0879] In this way, by effectively combining generative AI technology and natural language processing technology while utilizing the behavioral history of information providers, it is possible to realize a system that provides users with personalized and valuable information.

[0880] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0881] Step 1:

[0882] The server retrieves the information provider's past behavioral history from the database. It receives the information provider's identification information as input, and based on this information, extracts review content, purchase history, and attribute information from the database using queries. As output, it formats this data into a dataset for analysis.

[0883] Step 2:

[0884] The server uses natural language processing technology to analyze the acquired behavioral history. Using the dataset from Step 1 as input, it tokenizes text data, extracts keywords, and performs sentiment analysis to identify the information provider's preferences and trustworthiness score. As output, metadata representing the analysis results is generated.

[0885] Step 3:

[0886] The server generates a virtual representation using a generative AI model. It receives the analysis results from step 2 as input and feeds the data into the AI ​​model using pre-configured prompts. The output is a virtual representation that reflects the information provider's language style and preferences.

[0887] Step 4:

[0888] The user sends a question to the system using a communication program on their terminal. The user inputs a specific question into the terminal as input, and the terminal transmits this to the server. The question arrives at the server as output.

[0889] Step 5:

[0890] The server analyzes the user's question using sentiment analysis technology. The input is the user's question text, and the server identifies the emotional state based on language patterns and keywords. The output generates data indicating the user's emotions.

[0891] Step 6:

[0892] The server constructs a response using a generative AI model based on virtual representations and user sentiment data. Virtual representations and sentiment analysis results are used as input, and prompts are used to instruct the AI ​​model to generate a response. The output is a response that is highly relevant to the user and matches their emotions.

[0893] Step 7:

[0894] The server sends the generated response to the terminal and presents it to the user. The input is the response generated in step 6, and its contents are displayed to the user through the terminal. The output is that the user can view the response.

[0895] Step 8:

[0896] The user provides feedback on the response provided via their device and sends it to the server. The input is the feedback content entered via the device and sent as data to the server. The output is the feedback data stored on the server.

[0897] Step 9:

[0898] The server analyzes feedback to improve response quality. It uses user feedback data as input and analyzes the feedback content using sentiment analysis techniques and machine learning algorithms. The output provides insights for improving future responses.

[0899] (Application Example 2)

[0900] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0901] Modern users have diverse tastes and emotions, and there is a demand for information tailored to these needs. However, conventional systems struggle to provide personalized information that responds to users' emotions. Therefore, there is a need for a system that can provide optimized information tailored to each user's individual emotional state and preferences.

[0902] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0903] In this invention, the server includes means for acquiring and analyzing the past activity records of posters who have evaluation information; means for generating a highly-rated avatar based on the poster's activity records and providing personalized advice that corresponds to the user's emotions using emotion analysis technology; and means for transferring the user's questions to the avatar and generating answers that take the user's emotions into consideration. This makes it possible to provide optimized information based on the individual user's emotions and preferences.

[0904] "Evaluation information" refers to an index that indicates the reliability and value of a poster, calculated based on the poster's past activities and acquired data.

[0905] "Past activity records" refers to the collective behavioral history of the poster, including reviews, purchase history, and attribute information.

[0906] A "presumed alter ego" is a virtual character created based on the poster's past activity records, tone of voice, and style, for the purpose of interacting with users.

[0907] "Sentiment analysis technology" is a technology that recognizes emotions from a user's text and actions, and uses that information to derive an appropriate response.

[0908] "Personalized advice" means taking into account the user's preferences and emotional state to provide the information and suggestions that are most suitable for that individual.

[0909] "Information terminal device" is a general term for communication-capable devices such as smartphones and tablets that users use to send questions.

[0910] A "generative AI model" is an artificial intelligence model that learns from large amounts of data to generate and provide optimal information about the user.

[0911] A "prompt message" is a text instruction given to an AI to tell it to generate specific information or respond in a particular way.

[0912] This invention is a system that provides personalized information to users. The program for realizing this system is constructed by combining specific methods and technologies.

[0913] First, the server uses a high-performance data processing server as hardware, and for software, it uses Python, natural language processing libraries (e.g., spaCy), and sentiment analysis libraries (e.g., TextBlob). This allows the server to retrieve the poster's past activity records and analyze data including rating information. This data includes the poster's reviews and past purchase history.

[0914] Next, the server generates a simulated avatar using a generative AI model based on the analyzed data. This AI model is built using TensorFlow. The simulated avatar is a virtual character that reflects the poster's tone and style, and is useful for interacting with the user. Utilizing sentiment analysis technology, the avatar recognizes the user's emotions and generates responses that correspond to those emotions.

[0915] The user submits a question using an information terminal device (e.g., smartphone, tablet). The terminal forwards the user's question to the server via a communication application with a simple interface. The server analyzes the received question, performs sentiment analysis, and identifies the user's emotional state.

[0916] Ultimately, the server generates an appropriate response based on the user's question and emotional state, and presents it to the user through an information terminal. The user receives the response and returns feedback to the server corresponding to their emotions. This feedback is analyzed to improve the quality of future responses.

[0917] For example, if a user asks, "I'm looking for a candle with a relaxing scent," the emotion engine recognizes the user's relaxed state and provides optimized advice such as, "This aromatherapy candle is perfect for relaxation time and has received high ratings." An example of a prompt used in this process would be, "Recognize the user's emotion as relaxed, please suggest a scenario for using the scented candle."

[0918] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0919] Step 1:

[0920] The server collects data from a database of the poster's past activity records. This data includes the poster's review content, purchase history, and attribute information. The server analyzes the data using Python and natural language processing libraries to calculate a reliability score. This allows the server to identify the poster's hobbies and preferences and calculate a reliability score based on the input activity records.

[0921] Step 2:

[0922] The server generates an estimated avatar using a generative AI model based on the data analyzed in Step 1. TensorFlow is used for the generative AI model. It receives the analysis results as input, generates the estimated avatar, and records the avatar information in the database. The generated estimated avatar reflects the poster's tone and style.

[0923] Step 3:

[0924] The user sends a question to the server using a communication application on their device. The device receives the entered question text and sends it to the server. Once the user's question is transferred from the device to the server, the process moves on to the next step.

[0925] Step 4:

[0926] The server analyzes the received question text using a sentiment analysis library to identify the user's emotions. The input is the question text, and the output is the user's emotional state (e.g., relaxed, excited, interested). This allows the server to recognize the user's emotions along with the question content.

[0927] Step 5:

[0928] The server generates appropriate responses through an estimated avatar based on the question and the perceived emotions. It receives the generation AI model and emotion analysis results as input, and constructs a personalized response by creating a prompt and presenting it to the AI ​​model. The response is then delivered to the user via the terminal.

[0929] Step 6:

[0930] The user submits feedback on the provided response. The device sends the text of the feedback to the server. The server collects the feedback and analyzes it using an emotion engine. Based on the analysis results, the server helps improve the content of the estimated avatar's responses. This makes it possible to use the feedback for future improvements.

[0931] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0932] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0933] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0934] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0935] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0936] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0937] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0938] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0939] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0940] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0941] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0942] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0943] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0944] 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.

[0945] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0946] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0947] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0948] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0949] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0950] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0951] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0952] The following is further disclosed regarding the embodiments described above.

[0953] (Claim 1)

[0954] [Methods for obtaining and analyzing past activity records of posters who have evaluation information,

[0955] [Method for generating a highly-rated estimated avatar based on the activity record of the aforementioned poster,

[0956] [Methods for forwarding the user's question to the aforementioned estimated avatar and generating an answer,

[0957] [Means of presenting the above answer to the user and collecting feedback,

[0958] [Means for analyzing the aforementioned feedback and using it to improve the response content of the estimated doppelganger,

[0959] A system that includes this.

[0960] (Claim 2)

[0961] The system according to claim 1, which enables a user to send a question using a communication application.

[0962] (Claim 3)

[0963] [The system according to claim 1, which uses natural language processing technology to analyze the activity record and identify the poster's hobbies and preferences.

[0964] "Example 1"

[0965] (Claim 1)

[0966] [Methods for obtaining past activity records of posters with evaluation information and organizing them using database queries,

[0967] [A means of tokenizing the activity record using natural language processing technology and evaluating the trustworthiness of the poster through sentiment analysis,

[0968] [Methods for generating an estimated avatar with high evaluation information using a generation AI model based on the aforementioned analysis data,

[0969] [Methods for forwarding a prompt message sent by the user via a terminal to the estimated avatar and generating a response,

[0970] [A user interface is provided to present the above answer to the user, and means are used to collect feedback,

[0971] [Means for statistically analyzing the feedback and using it to improve the response content of the estimated doppelganger,

[0972] A system that includes this.

[0973] (Claim 2)

[0974] The system according to claim 1, which enables a user to send a question using an information processing application on a terminal.

[0975] (Claim 3)

[0976] [The system according to claim 1, which uses natural language processing technology to analyze the activity record and identify the poster's hobbies, preferences, and trustworthiness.

[0977] "Application Example 1"

[0978] (Claim 1)

[0979] [Means for obtaining and analyzing the past activity history of providers who possess evaluation information,

[0980] [Methods for generating a simulated personality with high evaluation information based on the activity history of the aforementioned provider,

[0981] [Means for forwarding user inquiries to the aforementioned simulated personality and generating responses,

[0982] [Means for showing the above response to the user and collecting the response,

[0983] [Means used to analyze the aforementioned reactions and improve the content of the simulated personality's responses,

[0984] [Means of providing detailed and reliable information for product selection using mobile information terminals,

[0985] A system that includes this.

[0986] (Claim 2)

[0987] The system according to claim 1, which enables a user to send an inquiry using a communication application.

[0988] (Claim 3)

[0989] [The system according to claim 1, which uses natural language processing technology to analyze the activity history and identify the provider's preferences.

[0990] "Example 2 of combining an emotion engine"

[0991] (Claim 1)

[0992] [Means for obtaining and analyzing the past behavioral history of information providers who possess evaluation information,

[0993] [Methods for generating a highly-rated virtual representation of the information provider based on their behavioral history,

[0994] [Means for forwarding user inquiries to the aforementioned virtual representation and generating responses,

[0995] [Means for presenting the above response to the user and collecting their evaluation,

[0996] [Means used to analyze the above evaluation and improve the response content of the virtual representation,

[0997] [Methods for analyzing the aforementioned behavioral history using natural language processing technology and identifying the preferences of the information provider,

[0998] [Methods for analyzing user emotions using emotion analysis technology and reflecting them in responses,

[0999] A system that includes this.

[1000] (Claim 2)

[1001] [The system according to claim 1, which enables a user to send an inquiry using a communication program.

[1002] (Claim 3)

[1003] [The system according to claim 1, which utilizes prompt statements when creating the virtual representation using generative AI technology.

[1004] "Application example 2 of combining emotional engines"

[1005] (Claim 1)

[1006] [Methods for obtaining and analyzing past activity records of posters who have evaluation information,

[1007] [A means for generating a highly-rated estimated avatar based on the activity record of the aforementioned poster, and providing personalized advice that corresponds to the user's emotions using sentiment analysis technology,

[1008] [Methods for forwarding user questions to the aforementioned estimated avatar and generating answers that take the user's emotions into consideration,

[1009] [Means for presenting the above response to the user, collecting feedback, and using it to improve the response content of the estimated avatar by analyzing it with an emotion engine,

[1010] A system that includes this.

[1011] (Claim 2)

[1012] [The system according to claim 1, which enables a user to send a question to an estimated avatar based on a generated AI model using an information terminal device, and to provide the user with information optimized for the user using prompt sentences designed by the AI.

[1013] (Claim 3)

[1014] [The system according to claim 1, which uses natural language processing technology and sentiment analysis to analyze the activity record, identify the poster's hobbies, preferences and emotional state, and recommend optimized products and services based on this. [Explanation of Symbols]

[1015] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of obtaining and analyzing the past activity records of posters who have evaluation information, A means for generating a highly-rated estimated avatar based on the activity record of the aforementioned poster, A means for forwarding the user's question to the aforementioned estimated avatar and generating an answer, A means of presenting the aforementioned answer to the user and collecting feedback, A means for analyzing the aforementioned feedback and using it to improve the response content of the estimated doppelganger, A system that includes this.

2. The system according to claim 1, which enables a user to send a question using a communication application.

3. The system according to claim 1, which uses natural language processing technology to analyze the activity record and identify the poster's hobbies and preferences.

Images