System

Abstract

A system is provided.SOLUTION: Means for inputting a question by a user via a terminal, means for receiving the question by a server and transmitting the question to a generative model in an appropriate format, and means for generating a primary answer to the question by the generative model; A means for returning the primary response to the server, a means for transmitting the primary response from the server to the terminal of the user, a means for inputting a supplement or correction to the primary response by another user via the terminal, a means for receiving the supplement or correction information by the server and retransmitting the supplement or correction information to the generative model, a means for generating a response again by the generative model and returning the re-generated response to the server, and a means for transmitting the re-generated response from the server to the terminal of the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Conventional question-answering systems have the problem of lacking the accuracy of answers expected by users. Once an answer is generated, it is provided as is, and there is no mechanism to incorporate feedback from other users for clarification or correction. As a result, answers are often incomplete or inaccurate, which reduces user satisfaction. [Means for solving the problem]

[0005] The present invention provides a system including a means for a user to input a question via a terminal, a means for a server to receive the question and send it to a generative model in an appropriate format, a means for the generative model to generate a primary answer to the question and return the primary answer to the server, a means for the server to send the primary answer to the user's terminal, a means for another user to input a supplement or correction to the primary answer via a terminal, a means for the server to receive the supplement or correction information and resend it to the generative model, a means for the generative model to generate another answer and return the re-answer to the server, and a means for the server to send the re-answer to the user's terminal, thereby making it possible to provide highly accurate answers that reflect user feedback.

[0006] "User" refers to a person who operates a terminal in the system and posts a question.

[0007] A "terminal" is a device that allows a user to input and send a question, and refers to an electronic device such as a smartphone, PC, or tablet.

[0008] "Question" refers to the content or information that a user asks the system through a terminal.

[0009] A "server" refers to a computer system that receives questions from users, sends them to a generative model, and returns the generated answers to the user's device.

[0010] A "generative model" refers to an artificial intelligence algorithm that uses natural language processing to generate answers to user questions.

[0011] A "primary answer" refers to the answer that a generative model first generates in response to a user's question.

[0012] "Supplement or correction" refers to the act of inputting additional information or corrections to a generated answer by another user, and the content of such input.

[0013] A "re-answer" refers to an answer that is generated again by a generative model based on supplementary or corrective information.

[0014] The "final answer" refers to the answer that is finally generated after multiple re-answers. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] An embodiment for implementing the system of the present invention will be described below.

[0037] System Configuration

[0038] The present invention can be implemented in a network system including a user terminal, a server, and a generative model. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative model. The generative model is a program that uses an artificial intelligence algorithm to generate answers to questions.

[0039] User question input

[0040] A user inputs a question using a terminal. For example, the user inputs and sends a question such as "What recent books do you recommend?" through the terminal interface. This question is sent from the terminal to the server.

[0041] Submitting questions and generating answers

[0042] The server processes the received question and sends it to the generative model in the appropriate format. The generative model analyzes the question and generates an answer using natural language processing techniques. For example, the generative model generates an answer such as "The recommended book these days is 'Twilight'" and sends it back to the server.

[0043] Presenting the first answer

[0044] The server sends the primary answer received from the generative model to the user's terminal, where the user can check the primary answer.

[0045] Entering Supplementary and Corrective Information

[0046] Other users can add supplementary or corrective information to the primary answer. For example, another user can add supplementary information such as "Twilight is good, but I also recommend the recently published Attack on Titan," and send it to the server via their device.

[0047] Generate answers again

[0048] The server then sends additional or corrective information to the generative model. The generative model then generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'" and send it back to the server.

[0049] Providing a second answer

[0050] The server then sends the regenerated answer to the user's terminal, where the user can confirm the revised answer. In addition, other users can enter additional comments or corrections.

[0051] Third and final answers

[0052] By repeating the re-answering process multiple times as necessary, the generative model generates a final answer, such as "My recent recommended books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'"

[0053] In this way, the system based on the present invention can provide highly accurate answers to user questions while reflecting supplements and corrections from other users.

[0054] The processing flow will be explained below.

[0055] Specific processing steps of the program

[0056] Step 1:

[0057] The user inputs a question from the terminal.

[0058] Specific operation: The user inputs a question using the terminal interface and presses the "Send" button. At this time, the terminal converts the input question into a digital format and sends it to the server.

[0059] Step 2:

[0060] The terminal sends a question to the server.

[0061] Specific operation: The terminal sends the question entered by the user to the server using a pre-configured protocol (e.g., HTTP request).

[0062] Step 3:

[0063] The server receives the question and sends it to the generative model in the appropriate format.

[0064] Specific operation: The server receives the question, converts it into a format that the generative model can understand (e.g., JSON or XML), and sends it to the generative model.

[0065] Step 4:

[0066] The generative model generates a first-order answer to the question and sends it back to the server.

[0067] How it works: The generative model uses natural language processing techniques to analyze the question and generate a first-order answer, which is then sent back to the server.

[0068] Step 5:

[0069] The server sends the primary response to the user's terminal.

[0070] Specific operation: The server sends the initial answer received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0071] Step 6:

[0072] The user reviews the primary answer and adds any additions or corrections.

[0073] Specific operations: Other users read the primary response through their terminals, enter corrections or supplementary information, and press the "Submit" button once the supplements or corrections are complete.

[0074] Step 7:

[0075] The terminal sends the supplementary or corrective information to the server.

[0076] Specific operation: The device sends supplementary or corrective information to the server using a pre-configured protocol (e.g., HTTP request).

[0077] Step 8:

[0078] The server resubmits the supplementary or corrective information to the generative model.

[0079] Specific behavior: The server receives the supplementary or corrective information and sends it back to the generative model in an appropriate format (e.g., reformatted JSON data).

[0080] Step 9:

[0081] The generative model generates a re-answer and sends it back to the server.

[0082] What happens next: The generative model generates a new answer based on the new information, and this answer is sent back to the server.

[0083] Step 10:

[0084] The server sends a second response to the user's terminal.

[0085] Specific operation: The server sends the response received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0086] Step 11:

[0087] Further additions or corrections (if necessary).

[0088] Specific actions: Another user will check the answer again, enter additional or corrective information if necessary, and press the "Submit" button.

[0089] Step 12:

[0090] Repeat steps 6-10 above as needed.

[0091] Specific Action: Repeat the process of supplementing or correcting as necessary and continue the process until a final answer is produced.

[0092] Through these steps, a highly accurate final answer is obtained.

[0093] Example 1

[0094] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0095] In conventional question-answering systems, answers to user questions are often fixed, and when other users add supplementary or correction information, the answers are not automatically updated. As a result, answers to questions are not always optimal, making it difficult for users to obtain satisfactory answers, especially when current information or individual knowledge is required.

[0096] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0097] In this invention, the server includes a means for a user to input a question via a device, a means for a computer system to receive the question and send it to a generation algorithm in an appropriate format, and a means for the generation algorithm to generate an initial answer to the question and return the initial answer to the computer system. This makes it possible to obtain a highly accurate answer to the question that takes into account supplementary and correction information as needed. Furthermore, by generating and presenting new answers multiple times as needed, it is possible to ultimately provide an answer that satisfies the user.

[0098] "Device" means an electronic device that allows a user to input or receive information, including smartphones, computers, tablets, etc.

[0099] "Computer System" refers to a computer network system for receiving, processing, and transmitting data from a user to a generating algorithm.

[0100] "Generation algorithm" refers to a program or artificial intelligence technology that automatically generates answers to user questions.

[0101] "Entering a question" refers to the act of a user entering a question or request in text form through a device.

[0102] The "first answer" refers to the answer that the generation algorithm first generates, and refers to the information that is first presented in response to a question entered by a user.

[0103] "Entering supplementary or correction information" refers to the act of another user entering additional information or correction information in response to an initial response.

[0104] A "re-answer" refers to an answer that is regenerated by the generation algorithm based on supplementary or corrective information for the initial answer.

[0105] The "final answer" refers to a completed answer that is finally presented to the user after multiple answer generation processes.

[0106] MODE FOR CARRYING OUT THE INVENTION

[0107] System Overview

[0108] The system of the present invention is composed of a network system including a user terminal, a server, and a generative model. The user terminal refers to a device such as a smartphone, PC, or tablet, and provides an interface for inputting and receiving questions. The server is a computer system that receives questions from users and sends them to the generative algorithm. The generative algorithm is a program that automatically generates answers to questions, and uses natural language processing technology such as GPT-3.

[0109] Specific program description

[0110] Suppose a user inputs a question using a terminal. For example, "What recent books do you recommend?" This question is sent from the terminal to the server. The server receives this question and prepares it to be sent to the generation algorithm. Specifically, it converts the received text data into an appropriate format (for example, JSON format) and sends it to the generation algorithm using a REST API.

[0111] The generation algorithm analyzes the received question and generates an answer based on natural language processing technology. A generative AI model such as GPT-3 can be used to generate the answer. The generated answer is sent back to the server in the form of, for example, "The recommended book these days is 'Twilight.'"

[0112] The server receives the answer returned by the generation algorithm and sends it back to the user's device. The user can check the generated primary answer through the device interface. At this time, other users can also use their devices to enter supplementary or correction information for the primary answer. For example, supplementary information such as "Twilight is good, but I also recommend the recently published Attack on Titan."

[0113] When additional or corrective information is entered, the server receives this information again and sends it to the generation algorithm. The generation algorithm takes the additional information into account and generates a new answer. The regenerated answer will be in the form of, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'" The server sends this re-answer to the user's device so that the user can confirm it.

[0114] Specific examples

[0115] The system operates when the user enters the following prompt through the terminal:

[0116] "What recent books would you recommend?"

[0117] "Do you have any additional information about 'Twilight'?"

[0118] "Are there any other books you'd recommend?"

[0119] This system allows users to obtain highly accurate answers to their questions, reflecting supplementary and corrective information as it goes along, making it possible to provide satisfactory answers, especially in cases where current information or specialized knowledge is required.

[0120] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0121] Step 1:

[0122] The user inputs a question using the device interface. For example, the user inputs "What books do you recommend these days?" and presses the submit button. At this time, the input data is in text format, and the device captures the data.

[0123] Step 2:

[0124] The terminal sends a question from the user to the server. The data to be sent is in the form of an HTTP request and is transferred through a network path from the device to the server. The server then receives the question data.

[0125] Step 3:

[0126] The server processes the received question data into the appropriate format. Specifically, it converts the text data into JSON format and prepares to send a request to the generation algorithm's API. The input is text data, and the output is a JSON-formatted request.

[0127] Step 4:

[0128] The server sends the appropriately formatted question data to the generation algorithm, typically using a REST API, along with the API endpoint and any necessary authentication information.

[0129] Step 5:

[0130] The generation algorithm analyzes the received question data and generates an initial answer using natural language processing techniques. For example, a generative AI model such as GPT-3 can be used to generate an answer such as "The recommended book these days is 'Twilight.'" The input is the formatted question data, and the output is the generated answer text.

[0131] Step 6:

[0132] The generation algorithm returns the initial answer it generated to the server. The generated answer is returned to the server in JSON format, so the server receives the answer data. The input is the generated answer text, and the output is a JSON-formatted response.

[0133] Step 7:

[0134] The server sends the initial response data received from the generation algorithm to the user device. For example, it sends the data to the device as an HTTP response. At this time, the server sends the response data in an appropriate format to the user device. The input is the response data in JSON format, and the output is the response text sent to the user device.

[0135] Step 8:

[0136] The user checks their initial answer through the device interface and then inputs additional or corrective information based on it. For example, they might type, "I like 'Twilight,' but I also recommend the recently published 'Attack on Titan,'" and press the submit button. The input is supplementary information in text format.

[0137] Step 9:

[0138] The terminal sends supplementary or correction information from the user to the server. This supplementary information is sent in the form of an HTTP request, and the server receives the data. The input is the supplementary information in text format, and the output is the supplementary information sent to the server.

[0139] Step 10:

[0140] The server then processes the received supplemental information into the appropriate format and sends it to the generation algorithm. For example, it converts text data into JSON format and prepares it for resubmission. The input is the supplemental information in text format, and the output is the request in JSON format.

[0141] Step 11:

[0142] The generation algorithm generates a new answer based on the supplementary information. For example, it generates a new answer such as "Recommended books these days are 'Twilight' and 'Attack on Titan'. 'Our Little Sister' is also popular." The input is the formatted supplementary information, and the output is the generated new answer text.

[0143] Step 12:

[0144] The generation algorithm sends the regenerated answer back to the server, which receives the regenerated answer. The input is the regenerated answer text, and the output is a JSON-formatted response.

[0145] Step 13:

[0146] The server sends the re-answer to the user's device. This data is sent as an HTTP response, and the user can confirm the re-answer. The input is the re-answer data in JSON format, and the output is the re-answer text sent to the user's device.

[0147] Step 14:

[0148] The server and the generation algorithm repeat this process as many times as necessary until a final answer is generated, such as "Recommended recent books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'" The input is the repeatedly updated supplementary information, and the output is the final answer text.

[0149] In this way, a system is realized that provides highly accurate answers to user questions that sequentially reflect supplementary and corrective information.

[0150] (Application example 1)

[0151] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0152] In conventional online shopping systems, users must spend a lot of time obtaining detailed information about products, making it difficult to efficiently select appropriate products. Furthermore, if the answers to users' questions are inaccurate, there is a risk that their purchasing decisions will be affected. To solve this problem, a system is needed that provides users with quick and accurate answers to their questions and supports their purchasing decisions.

[0153] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0154] In this invention, the server includes a means for a user to input a question via a terminal, a means for the server to receive the question and transmit it to a generative model in an appropriate format, and a means for the generative model to generate a primary answer to the question and return the primary answer to the server, thereby enabling the user to obtain information about products in a virtual store in real time and make accurate and prompt purchasing decisions.

[0155] "Means for users to input questions via a terminal" refers to a function that allows users to input questions through an interface using a device such as a smartphone, tablet, or PC.

[0156] "Means for the server to receive and send in a format appropriate for the generative model" refers to the function by which the server receives a question from a user, converts it into a format that can be processed by the generative model, and sends it.

[0157] "Means for the generative model to generate a primary answer to a question and return the primary answer to the server" is a function that uses a generative algorithm to create an answer to a user's question and return the answer to the server.

[0158] The "means for the server to transmit the primary answer to the user's terminal" is a function for transmitting the primary answer received from the generative model to the user's device.

[0159] "Means for other users to enter supplementary information or corrections to the primary response via their device" refers to the ability for other users to use their own device to enter additional information or corrections to the response already provided.

[0160] The "means for the server to receive supplementary or corrective information and retransmit it to the generative model" is a function that allows the server to receive supplementary or corrective information and retransmit it to the generative model.

[0161] "Means for the generative model to generate an answer again and return the new answer to the server" is a function that generates a new answer based on supplementary or correction information and returns the new answer to the server.

[0162] The "means for the server to send a re-answer to the user's terminal" is a function for sending the re-answer received from the generative model again to the user's device.

[0163] "Means for allowing users to ask questions about products via a terminal within a virtual store" is a function that allows users to input questions about products they want through an interface within the virtual store.

[0164] "Means for the generative model to generate product information and comparative information based on the question" refers to a function in which the generative model creates detailed information about the product and comparative information with other products based on the content of the user's question.

[0165] "Means for presenting generated product information to the user" is a function that displays information about products and comparison information generated by the generative model on the user's device.

[0166] To implement this invention, a network system including a user terminal, a server, and a generative model must be constructed. The user terminal can be a smartphone, tablet, PC, or other device. The server sends user questions to the generative model in an appropriate format, receives the answers generated by the generative model, and returns them to the user terminal. The generative model can be a program using an advanced natural language processing algorithm, such as OpenAI's GPT-3.

[0167] System Configuration

[0168] 1. User Device

[0169] The user inputs a question via the terminal. This question is input through the interface, for example, "What smartphones do you recommend these days?" This question is then sent from the terminal to the server.

[0170] 2. Server

[0171] The server processes the received question and sends it to the generative model in the appropriate format, which is the prompt for the generative model to generate an answer in the following format:

[0172] User Question: What recent smartphones do you recommend?

[0173] answer:

[0174] 3. Generative Model

[0175] The generative model analyzes the question and generates an answer using natural language processing technology. An example of an answer generated by the generative model is, "The latest recommended smartphones include the iPhone 14 and the Samsung Galaxy S22. They each feature high-performance cameras and fast processors." This answer is then sent back to the server.

[0176] 4. Presenting the Answer

[0177] The server sends the primary answer received from the generative model to the user's device, where the user can check the primary answer.

[0178] 5. Entering Supplementary and Corrective Information

[0179] Other users can enter additional or corrective information to the primary answer. For example, they can enter, "The iPhone 14 is good, but if you prioritize cost-effectiveness, I also recommend the Pixel 6." This additional data is sent to the server.

[0180] 6. Generate answers again

[0181] The server then sends supplementary or corrective information to the generative model. The generative model then generates a new answer based on the new information. An example of a new answer might be, "Recommended smartphones are the iPhone 14 and the Samsung Galaxy S22. If cost-effectiveness is important, consider the Pixel 6." This new answer is also sent back to the server and then to the user's device.

[0182] This system configuration allows users to obtain fast and accurate information, which can greatly support purchasing decisions. Specific hardware requirements include a server and user devices (smartphones, PCs, tablets, etc.), and software requirements include Flask (a Python web framework), OpenAI API, and a generative model (GPT-3).

[0183] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0184] Step 1:

[0185] The user inputs a question via a terminal.

[0186] Users input questions using an interface from devices such as smartphones or PCs. The input questions are in the form of "What smartphones do you recommend these days?". The user's question is generated as input data.

[0187] Step 2:

[0188] The terminal sends a question to the server.

[0189] The terminal sends the question entered by the user to the server. When the server receives the question, it proceeds to the next processing step. The input includes the user's question data, and the output is the question data transferred to the server.

[0190] Step 3:

[0191] The server sends the received question to the generative model in the appropriate format.

[0192] The server analyzes the received question and formats it in a format that can be processed by the generative model. Specifically, the question is converted into a prompt sentence. It is converted into the format "User question: What smartphone do you recommend these days?\nAnswer: ". The input is the user's question data, and the output is a generated prompt sentence.

[0193] Step 4:

[0194] A generative model generates a first-order answer to the question.

[0195] A generative model (e.g., GPT-3) takes a prompt as input, analyzes it, generates it, and produces a first-order answer. For example, the model might generate an answer like, "The latest recommended smartphones are the iPhone 14 and the Samsung Galaxy S22. They each feature a high-performance camera and a fast processor." The input contains a prompt, and the output is a first-order answer.

[0196] Step 5:

[0197] The server sends the primary response to the user terminal.

[0198] The server sends the primary answer received from the generative model to the user's device, allowing the user to check the primary answer on their own device. The input is the primary answer from the generative model, and the output is the response data sent to the user's device.

[0199] Step 6:

[0200] Other users input supplements or corrections to the primary answer via their terminals.

[0201] Other users can add supplementary or correction information to the answers already provided. For example, "The iPhone 14 is good, but if cost performance is important, the Pixel 6 is also recommended." The input includes the initial answer and supplementary or correction information, and the output is the supplementary or correction data that is sent to the server.

[0202] Step 7:

[0203] The server receives the supplementary or corrective information and resubmits it to the generative model.

[0204] The server receives supplementary or correction information from other users and sends it back to the generative model. At this time, the supplementary or correction information is added to the prompt sentence. In this way, the server is ready to generate an answer again. The input is the supplementary or correction information, and the output is the data to be resubmitted to the generative model.

[0205] Step 8:

[0206] The generative model generates the answer again and sends the answer back to the server.

[0207] The generative model generates a new answer based on new information. For example, it generates a new answer in the form of "Recommended smartphones are the iPhone 14 and Samsung Galaxy S22. If cost-effectiveness is important, consider the Pixel 6." The input is a prompt sentence containing supplementary information and corrections, and the output is a new answer.

[0208] Step 9:

[0209] The server sends a second response to the user's terminal.

[0210] The server then sends the re-answer received from the generative model back to the user device, allowing the user to check the re-answer on their device. The input is the re-answer from the generative model, and the output is the re-answer data sent to the user device.

[0211] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0212] An embodiment for implementing the system of the present invention will be described below.

[0213] System Configuration

[0214] The present invention can be implemented in a network system including a user terminal, a server, a generative model, and an emotion engine. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative model and emotion engine. The generative model is a program that uses an artificial intelligence algorithm to generate answers to questions, and the emotion engine recognizes user emotions and adjusts the behavior of the generative model based on those emotions.

[0215] User question input

[0216] The user inputs a question using a terminal. For example, the user inputs and sends a question such as, "What recent books do you recommend?" The emotion engine analyzes the user's emotions along with the question, and sends the results to the server as additional information.

[0217] Submitting questions and generating answers

[0218] The server processes the received question along with the emotional information and sends it to the generative model in an appropriate format. The generative model takes the question and emotional information into account to generate an answer. For example, if the answer is "The book I recommend these days is 'Twilight,'" the answer is adjusted based on the emotional information.

[0219] Presenting the first answer

[0220] The server sends the primary answer received from the generative model to the user's device, where the user can check the primary answer.

[0221] Entering Supplementary and Corrective Information

[0222] Other users can add supplementary or correction information to the primary answer. For example, they can add supplementary information such as, "Twilight is good, but I also recommend the recently published Attack on Titan." In this case, the emotion engine also recognizes the emotion of the user who added the supplementary or correction information and sends that information to the server.

[0223] Generate answers again

[0224] The server sends supplementary or corrective information and the emotion information to the generative model. The generative model generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'"

[0225] Providing a second answer

[0226] The server sends the regenerated answer to the user's terminal, where the user can review the revised answer and other users can enter additional comments or corrections.

[0227] Third and final answers

[0228] By repeating the re-answering process multiple times as necessary, the generative model generates a final answer. For example, a final answer might be, "My recent recommended books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'" During this process, the emotion engine continuously analyzes the user's emotions and optimizes the answer.

[0229] In this way, the system according to the present invention can provide a highly accurate final answer that takes into account the user's emotional information.

[0230] The processing flow will be explained below.

[0231] Specific processing steps of the system

[0232] Step 1:

[0233] The user inputs a question from the terminal.

[0234] Specific operation: The user inputs a question using the device interface and presses the "Send" button. At this time, the emotion engine analyzes the user's tone of voice and facial expressions to extract emotional information. The device then transmits the input question and emotional information to the server.

[0235] Step 2:

[0236] The device sends the question and emotion information to the server.

[0237] Specific operation: The device sends the question entered by the user and the emotional information analyzed by the emotion engine to the server using a pre-configured protocol (e.g., HTTP request).

[0238] Step 3:

[0239] The server receives the question and emotion information and sends it to the generative model in an appropriate format.

[0240] Specific operation: The server receives the question and emotion information, converts it into a format that the generative model can understand (e.g., JSON or XML), and sends it to the generative model.

[0241] Step 4:

[0242] The generative model generates a first-order answer based on the question and emotion information and sends it back to the server.

[0243] How it works: The generative model analyzes the question using natural language processing techniques and generates a first-order answer. The tone and content of the answer are adjusted based on the emotional information. This first-order answer is then sent back to the server.

[0244] Step 5:

[0245] The server sends the primary response to the user's terminal.

[0246] Specific operation: The server sends the initial answer received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0247] Step 6:

[0248] The user reviews the primary answer and adds any additions or corrections.

[0249] Specific operation: Other users read the primary response through their devices and enter corrections or supplementary information. The emotion engine recognizes the user's emotions while they are entering and sending the response, adds that information, and sends it to the server.

[0250] Step 7:

[0251] The terminal transmits supplementary or corrective information and emotion information to the server.

[0252] Specific operation: The device sends supplementary or corrective information, as well as the emotion information analyzed by the emotion engine, to the server via a pre-configured protocol (e.g., HTTP request).

[0253] Step 8:

[0254] The server retransmits the supplementary or corrective information and the emotion information to the generative model.

[0255] Specific behavior: The server receives the supplementary or corrective information, as well as the emotion information, and sends it back to the generative model in an appropriate format (e.g., reformatted JSON data).

[0256] Step 9:

[0257] The generative model generates a re-answer and sends it back to the server.

[0258] What it does: The generative model generates a new answer based on the new information and emotional information. The emotional information adjusts the tone and content of the new answer. The new answer is then sent back to the server.

[0259] Step 10:

[0260] The server sends a second response to the user's terminal.

[0261] Specific operation: The server sends the response received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0262] Step 11:

[0263] Further additions or corrections (if necessary).

[0264] Specific operation: Another user checks the answer again, enters additional or corrective information as necessary, and presses the "Send" button. During input and submission, the emotion engine recognizes the user's emotions, adds that information, and sends it to the server.

[0265] Step 12:

[0266] Repeat steps 6-10 above as needed.

[0267] What it does: The process continues, supplementing or correcting as needed, until a final answer is produced. Emotional information is continually considered throughout the process.

[0268] Through these steps, a highly accurate final answer that reflects the user's emotional information is obtained.

[0269] Example 2

[0270] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0271] Conventional question-answering systems have had the problem of difficulty in providing appropriate answers based on the user's emotions. In particular, they have had problems with generating answers that ignore the user's emotions, resulting in poor answer quality and difficulty in achieving user satisfaction. Furthermore, when additional clarifications or corrections are added, there is a lack of a mechanism for generating appropriate re-answers that take these into account. To solve these problems, highly accurate answer generation that reflects the user's emotional information is required.

[0272] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a question via a terminal, a means for the server to receive the question and the user's emotional information and transmit them to the generative model in an appropriate format, and a means for the generative model to generate a primary answer based on the question and the emotional information and return the primary answer to the server. This enables the generation of a high-quality answer that takes into account the user's emotional information.

[0273] "User" refers to an entity that utilizes the system to enter questions and receive answers.

[0274] "Device" refers to the device used by a user to enter questions and check answers, such as a smartphone, computer, or tablet.

[0275] "Server" refers to the computer system that receives questions from users and sends information to the generative model and emotion engine.

[0276] "Question" refers to a question or request entered by a user via a terminal.

[0277] A "generative model" refers to a program that uses an artificial intelligence algorithm to generate answers based on questions and emotional information.

[0278] "Emotion engine" refers to software that identifies and analyzes a user's emotions.

[0279] "Emotion information" refers to data relating to the user's emotions analyzed by the emotion engine.

[0280] "Primary answer" refers to the first answer generated by the generative model.

[0281] "Supplement or correction" refers to additional information or corrections entered by other users in response to the primary answer.

[0282] A "re-answer" refers to a second or subsequent answer generated by the generative model based on supplementary or corrective information and emotional information.

[0283] "Final Answer" refers to the final answer generated through the repeated answering process.

[0284] The system of the present invention generates answers to questions from users via a network system and provides highly accurate answers that take into account the user's emotional information using an emotion engine. The system of the present invention mainly includes the following components: a user terminal, a server, a generative model, and an emotion engine.

[0285] System Configuration

[0286] User terminal

[0287] A user terminal is a device that allows a user to input and send questions, such as a smartphone, PC, or tablet. The user terminal includes an input interface and a display interface, sends the questions input by the user to the server, and displays the answers from the server.

[0288] server

[0289] The server is responsible for receiving questions from users and sending them to the generative model and emotion engine. The server sends the emotion information analyzed by the emotion engine, along with the question and any supplementary or correction information, to the generative model in an appropriate format.

[0290] Generative Model

[0291] A generative model is a program that uses artificial intelligence algorithms to generate answers based on questions and sentiment information. For example, a natural language processing model such as GPT-3 is used. A generative model generates a first answer, a follow-up answer, and a final answer.

[0292] Emotion Engine

[0293] The emotion engine is software that analyzes the user's emotions and generates emotional information to accompany the user's questions and supplementary / corrective information, allowing the generative model to generate answers based on emotions.

[0294] Specific examples

[0295] The user uses a terminal to input and send a question such as "What recent books do you recommend?" The server receives this question and analyzes the user's emotions using an emotion engine. If the emotion engine determines that the user is curious, the server sends this emotional information along with the question to the generative model.

[0296] The generative model generates an answer based on the question and emotional information. For example, a primary answer such as "The recommended book these days is 'Twilight.' Another popular book is 'Attack on Titan.'" is generated and sent to the user's device via the server.

[0297] When another user adds additional information, such as "Twilight is good, but I also recommend the recently published Our Little Sister," the emotion engine analyzes the user's emotions. The server sends this new emotional information to the generative model, which then generates a new answer.

[0298] As a follow-up answer, the following answer is generated: "My recent recommended books are 'Twilight' and 'Our Little Sister.' In addition, 'Demon Slayer: Kimetsu no Yaiba' is also highly rated," and is sent to the user's device via the server.

[0299] Prompt Sentence Examples

[0300] "The user asked for recent book recommendations. The sentiment engine determined that the user is curious. Please take this into account when generating an answer."

[0301] In this way, the system of the present invention can provide highly accurate answers that take into account the user's emotional information, thereby improving user satisfaction.

[0302] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0303] The flow of this system's program processing

[0304] Specific explanation of processing steps

[0305] Step 1:

[0306] The user inputs and sends a question via a terminal.

[0307] Input: The user types into the terminal, "What recent books have you recommended?"

[0308] Data processing: When the send button on the terminal is pressed, the question text is sent to the server using an HTTP POST request.

[0309] Output: The question is sent to the server.

[0310] Step 2:

[0311] The server receives the question and passes it to the emotion engine for analysis of emotion information.

[0312] Input: The user's question sent to the server: "What books have you recommended recently?"

[0313] Data processing: The emotion engine performs sentiment analysis on the text to determine the user's emotion (e.g., curious).

[0314] Output: Analysis results including emotional information.

[0315] Step 3:

[0316] The server sends the question and emotion information to the generative model.

[0317] Input: Parsed emotion information (e.g., curious) and user question.

[0318] Data processing: The question and emotion information are sent to the generative model server in a data format such as JSON.

[0319] Output: The generative model receives the question and sentiment information.

[0320] Step 4:

[0321] A generative model generates a first-order answer based on the question and sentiment information.

[0322] Input: The question received by the generative model (e.g., "What recent books do you recommend?") and sentiment information.

[0323] Data processing: A generative model (e.g., GPT-3) analyzes the input prompt and generates a text answer.

[0324] Output: Text of the primary answer (e.g., "The book I recommend these days is 'Twilight.' Another popular book is 'Attack on Titan.'").

[0325] Step 5:

[0326] The server sends the primary response to the user's terminal.

[0327] Input: The text of the primary answer returned by the generative model.

[0328] Data processing: The server sends the primary answer to the user terminal as an HTTP response.

[0329] Output: The primary answer is displayed on the user's terminal.

[0330] Step 6:

[0331] Other users input supplementary or corrective information to the primary answer via the terminal.

[0332] Input: Another user types into their device, "Twilight is good, but I also recommend the recently published Our Little Sister."

[0333] Data processing: The entered supplementary or corrective information is sent to the server using an HTTP POST request.

[0334] Output: Any supplementary or corrective information is sent to the server.

[0335] Step 7:

[0336] The server retransmits the supplementary or corrective information and the emotion information to the generative model.

[0337] Input: Supplementary or correction information and sentiment information received by the server.

[0338] Data processing: Supplementary or corrective information and emotion information are resubmitted to the generative model server in a data format such as JSON.

[0339] Output: The generative model receives new information needed to generate a new answer.

[0340] Step 8:

[0341] The generative model generates a re-answer.

[0342] Input: New supplementary or correction information received by the generative model (e.g., "Twilight is good, but I also recommend the recently published Our Little Sister") and sentiment information.

[0343] Data processing: The generative model generates a new answer based on new information (e.g., "Recommended books these days are 'Twilight' and 'Our Little Sister.' Other popular books include 'Demon Slayer: Kimetsu no Yaiba.'").

[0344] Output: The text of the re-answer.

[0345] Step 9:

[0346] The server sends a second response to the user's terminal.

[0347] Input: The text of the re-answer returned by the generative model.

[0348] Data processing: The server sends a second response to the user terminal as an HTTP response.

[0349] Output: The answer is displayed on the user's terminal.

[0350] Thus, the system takes a series of steps to take affective information into account and provide highly accurate answers to the user's questions.

[0351] (Application example 2)

[0352] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0353] Conventional content distribution services exist that generate appropriate answers to user questions. However, because they do not take the user's emotions into account, the answers often do not match the user's wishes or circumstances. Emotional information is particularly important in entertainment, and ignoring it reduces satisfaction. Furthermore, because feedback from other users cannot be utilized, the answers lack accuracy and adaptability. Therefore, a system that analyzes the user's emotional information and reflects it in answer generation is needed.

[0354] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0355] In this invention, the server includes: a means for a user to input a question via a terminal; a means for the server to receive the question and transmit it to the generative model in an appropriate format; a means for the generative model to generate a primary answer to the question and return the primary answer to the server; a means for the server to transmit the primary answer to the user's terminal; a means for another user to input a supplement or correction to the primary answer via the terminal; a means for the server to receive the supplement or correction information and retransmit it to the generative model; a means for the generative model to generate a new answer and return the new answer to the server; a means for the server to transmit the new answer to the user's terminal; a means for analyzing the user's emotions using an emotion engine and adjusting the behavior of the generative model based on the analysis; and a means for updating the recommendation content based on feedback from other users. This makes it possible to provide more appropriate and satisfying answers that reflect the user's emotional information.

[0356] "Means for users to input questions via a terminal" refers to a mechanism that provides an interface for users to input and send questions using devices such as smartphones, tablets, and PCs.

[0357] "Means for the server to receive and send in an appropriate format to the generative model" refers to a mechanism by which the server receives questions entered by users, converts them into a format that can be processed by the generative AI model, and sends them.

[0358] "Means for the generative model to generate a first answer to a question and return the first answer to the server" refers to a mechanism by which the generative AI model generates an initial answer to a user's question and sends that answer to the server.

[0359] "Means by which the server sends the primary answer to the user's device" refers to a mechanism by which the server receives the primary answer obtained from the generative AI model and forwards it to the user's device.

[0360] "Means for other users to input supplementary information or corrections to the primary response via a terminal" refers to a system that provides an interface for other users to input additional information or corrections to the primary response using a smartphone, tablet, PC, etc.

[0361] "Means for the server to receive supplementary or corrective information and retransmit it to the generative model" refers to a mechanism by which, after additional information or corrections are entered, the server receives that information and retransmits it to the generative AI model.

[0362] "Means for the generative model to generate a new answer and return the new answer to the server" refers to a mechanism by which the generative AI model generates a new answer based on supplements or corrections and sends that answer to the server.

[0363] "Means for the server to send the re-answer to the user's device" refers to a mechanism by which the server receives the re-answer obtained from the generative AI model and forwards it to the user's device.

[0364] "Means of analyzing user emotions using an emotion engine and adjusting the behavior of the generative model based on that" refers to a mechanism that uses emotion analysis technology to detect user emotions and optimizes the content and format of the generative AI model's responses based on the results.

[0365] "Means for updating recommendations based on feedback from other users" refers to a mechanism that improves the recommendations of the generative AI model by reflecting opinions and ratings provided by other users.

[0366] Specific embodiments for carrying out the present invention will be described below.

[0367] System Configuration

[0368] This invention is embodied in a network system including a user terminal, a server, a generative AI model, and an emotion engine. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative AI model and the emotion engine. The generative AI model is a program that uses an artificial intelligence algorithm to generate answers to questions, and the emotion engine recognizes the user's emotions and adjusts the behavior of the generative AI model based on those emotions.

[0369] User question input

[0370] A user inputs a question using a device such as a smartphone or tablet. For example, a user might input and send a question such as, "I've been looking for a heartwarming movie lately. Do you have any recommendations?" The emotion engine analyzes the user's emotions along with the question and sends the results to the server as additional information.

[0371] Submitting questions and generating answers

[0372] The server processes the received question along with the emotional information and sends it to the generative AI model in an appropriate format. The generative AI model takes the question and emotional information into consideration to generate an answer. For example, if the answer generated is "A recent heartwarming movie I recommend is 'Green Book,'" the content and format of the answer are adjusted based on the emotional information.

[0373] Presenting the first answer

[0374] The server sends the initial answer received from the generative AI model to the user's device, where the user can check the initial answer.

[0375] Entering Supplementary and Corrective Information

[0376] Other users can add supplementary or correction information to the primary answer. For example, they can add supplementary information such as, "Green Book is good, but I also recommend the recently released Shawshank Redemption." In this case, the emotion engine also recognizes the emotion of the user who added the supplementary or correction information and sends that information to the server.

[0377] Generate answers again

[0378] The server sends supplementary or corrective information and the emotional information to the generative AI model. The generative AI model generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended recent heartwarming movies are 'Green Book' and 'The Shawshank Redemption.' Other popular movies include 'Les Miserables.'"

[0379] Providing a second answer

[0380] The server sends the regenerated answer to the user's terminal, where the user can review the revised answer and other users can enter additional comments or corrections.

[0381] Generate the final answer

[0382] The generative AI model generates a final answer by repeating the process of re-answering multiple times as necessary. During this process, the emotion engine continuously analyzes the user's emotions and optimizes the answer. For example, the final answer might be, "Recommended recent heartwarming movies are 'Green Book,' 'The Shawshank Redemption,' 'Les Miserables,' and 'Twilight.'"

[0383] In this way, the system according to the present invention can provide a highly accurate final answer that takes into account the user's emotional information.

[0384] Hardware and software used

[0385] Hardware: Smartphones, tablets, computers

[0386] Software: The program is implemented in Python and utilizes APIs for emotion engines and generative AI models.

[0387] Emotion Engine API (e.g. https: / / emotion-api.example.com / analyze)

[0388] Generate AI Model API (e.g. https: / / ai-model-api.example.com / generate)

[0389] Examples of prompt statements

[0390] User Question: I've been looking for a heartwarming movie lately. Any recommendations?

[0391] Emotional information: Relaxing, Heartwarming, Comforting

[0392] Output of generative AI model: A recent recommended heartwarming movie is "Green Book," but other good choices are "The Shawshank Redemption" and "Les Miserables."

[0393] According to this invention, it is possible to improve the accuracy of answers by incorporating information about the user's emotions, and to recommend content with a higher degree of satisfaction.

[0394] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0395] Step 1:

[0396] The user inputs a question via the terminal. The user inputs a question into a smartphone or tablet and sends it. An example of input is, "I've been looking for a heartwarming movie lately. Do you have any recommendations?"

[0397] Step 2:

[0398] The terminal sends a question and the server receives it. The terminal sends the question entered by the user to the server. The server receives this question and proceeds to the next step.

[0399] Step 3:

[0400] The server sends the received question to the emotion engine to obtain emotion information. The server sends the question to the emotion engine (e.g. https: / / emotion-api.example.com / analyze). The input is the question text, and the output is emotion information. For example, the emotion information detected is "relaxing, heartwarming, comforting."

[0401] Step 4:

[0402] The server sends the question and emotion information to the generative AI model. After receiving the emotion information, the server sends it along with the question to the generative AI model (e.g., https: / / ai-model-api.example.com / generate). The input is the question and emotion information, and the generative AI model generates a first answer based on this.

[0403] Step 5:

[0404] The generative AI model generates a first answer and sends it back to the server. The generative AI model generates an answer based on the input question and emotional information. For example, it generates an answer such as "A recent recommended heartwarming movie is 'Green Book'." This answer is sent to the server.

[0405] Step 6:

[0406] The server sends the primary answer to the user's device. The server sends the primary answer received from the generative AI model to the user's device. The user can check the primary answer through their device.

[0407] Step 7:

[0408] Other users can input supplementary information or corrections to the primary answer via the terminal. Other users can input additional information or corrections to the primary answer. For example, they can add supplementary information such as, "'Green Book' is good, but I also recommend the recently released 'Shawshank Redemption'."

[0409] Step 8:

[0410] The terminal transmits supplementary or corrective information to the server, which receives it. The terminal transmits supplementary or corrective information from other users to the server, which receives it.

[0411] Step 9:

[0412] The server retransmits the supplementary or correction information and emotional information to the generative AI model. The server retransmits the received supplementary or correction information and the emotional information to the generative AI model. The input is the supplementary or correction information and emotional information.

[0413] Step 10:

[0414] The generative AI model generates a re-answer based on the new information and sends it back to the server. For example, it might generate a re-answer such as, "Recommended recent heartwarming movies are 'Green Book' and 'The Shawshank Redemption.' Other popular movies include 'Les Miserables.'" and send it to the server.

[0415] Step 11:

[0416] The server sends the re-answer to the user's device. The server sends the re-answer received from the generative AI model to the user's device, where the user can check the re-answer.

[0417] Step 12:

[0418] If necessary, other users can enter additional supplementary or correction information, which is then received by the server and sent to the generative AI model. Finally, the generative AI model generates a final answer. For example, the final answer might be, "Recommended recent heartwarming movies are 'Green Book,' 'The Shawshank Redemption,' 'Les Miserables,' and 'Twilight.'"

[0419] Step 13:

[0420] The server sends the final answer to the user's device. The server sends the final answer received from the generative AI model to the user's device, where the user can check the final answer.

[0421] This series of processes enables highly accurate content recommendations that reflect the user's emotional information.

[0422] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0423] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0424] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0425] [Second embodiment]

[0426] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0427] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0428] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0429] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0430] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0431] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0432] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0433] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0434] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0435] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0436] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0437] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0438] An embodiment for implementing the system of the present invention will be described below.

[0439] System Configuration

[0440] The present invention can be implemented in a network system including a user terminal, a server, and a generative model. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative model. The generative model is a program that uses an artificial intelligence algorithm to generate answers to questions.

[0441] User question input

[0442] A user inputs a question using a terminal. For example, the user inputs and sends a question such as "What recent books do you recommend?" through the terminal interface. This question is sent from the terminal to the server.

[0443] Submitting questions and generating answers

[0444] The server processes the received question and sends it to the generative model in the appropriate format. The generative model analyzes the question and generates an answer using natural language processing techniques. For example, the generative model generates an answer such as "The recommended book these days is 'Twilight'" and sends it back to the server.

[0445] Presenting the first answer

[0446] The server sends the primary answer received from the generative model to the user's terminal, where the user can check the primary answer.

[0447] Entering Supplementary and Corrective Information

[0448] Other users can add supplementary or corrective information to the primary answer. For example, another user can add supplementary information such as "Twilight is good, but I also recommend the recently published Attack on Titan," and send it to the server via their device.

[0449] Generate answers again

[0450] The server then sends additional or corrective information to the generative model. The generative model then generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'" and send it back to the server.

[0451] Providing a second answer

[0452] The server then sends the regenerated answer to the user's terminal, where the user can confirm the revised answer. In addition, other users can enter additional comments or corrections.

[0453] Third and final answers

[0454] By repeating the re-answering process multiple times as necessary, the generative model generates a final answer, such as "My recent recommended books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'"

[0455] In this way, the system based on the present invention can provide highly accurate answers to user questions while reflecting supplements and corrections from other users.

[0456] The processing flow will be explained below.

[0457] Specific processing steps of the program

[0458] Step 1:

[0459] The user inputs a question from the terminal.

[0460] Specific operation: The user inputs a question using the terminal interface and presses the "Send" button. At this time, the terminal converts the input question into a digital format and sends it to the server.

[0461] Step 2:

[0462] The terminal sends a question to the server.

[0463] Specific operation: The terminal sends the question entered by the user to the server using a pre-configured protocol (e.g., HTTP request).

[0464] Step 3:

[0465] The server receives the question and sends it to the generative model in the appropriate format.

[0466] Specific operation: The server receives the question, converts it into a format that the generative model can understand (e.g., JSON or XML), and sends it to the generative model.

[0467] Step 4:

[0468] The generative model generates a first-order answer to the question and sends it back to the server.

[0469] How it works: The generative model uses natural language processing techniques to analyze the question and generate a first-order answer, which is then sent back to the server.

[0470] Step 5:

[0471] The server sends the primary response to the user's terminal.

[0472] Specific operation: The server sends the initial answer received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0473] Step 6:

[0474] The user reviews the primary answer and adds any additions or corrections.

[0475] Specific operations: Other users read the primary response through their terminals, enter corrections or supplementary information, and press the "Submit" button once the supplements or corrections are complete.

[0476] Step 7:

[0477] The terminal sends the supplementary or corrective information to the server.

[0478] Specific operation: The device sends supplementary or corrective information to the server using a pre-configured protocol (e.g., HTTP request).

[0479] Step 8:

[0480] The server resubmits the supplementary or corrective information to the generative model.

[0481] Specific behavior: The server receives the supplementary or corrective information and sends it back to the generative model in an appropriate format (e.g., reformatted JSON data).

[0482] Step 9:

[0483] The generative model generates a re-answer and sends it back to the server.

[0484] What happens next: The generative model generates a new answer based on the new information, and this answer is sent back to the server.

[0485] Step 10:

[0486] The server sends a second response to the user's terminal.

[0487] Specific operation: The server sends the response received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0488] Step 11:

[0489] Further additions or corrections (if necessary).

[0490] Specific actions: Another user will check the answer again, enter additional or corrective information if necessary, and press the "Submit" button.

[0491] Step 12:

[0492] Repeat steps 6-10 above as needed.

[0493] Specific Action: Repeat the process of supplementing or correcting as necessary and continue the process until a final answer is produced.

[0494] Through these steps, a highly accurate final answer is obtained.

[0495] Example 1

[0496] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0497] In conventional question-answering systems, answers to user questions are often fixed, and when other users add supplementary or correction information, the answers are not automatically updated. As a result, answers to questions are not always optimal, making it difficult for users to obtain satisfactory answers, especially when current information or individual knowledge is required.

[0498] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0499] In this invention, the server includes a means for a user to input a question via a device, a means for a computer system to receive the question and send it to a generation algorithm in an appropriate format, and a means for the generation algorithm to generate an initial answer to the question and return the initial answer to the computer system. This makes it possible to obtain a highly accurate answer to the question that takes into account supplementary and correction information as needed. Furthermore, by generating and presenting new answers multiple times as needed, it is possible to ultimately provide an answer that satisfies the user.

[0500] "Device" means an electronic device that allows a user to input or receive information, including smartphones, computers, tablets, etc.

[0501] "Computer System" refers to a computer network system for receiving, processing, and transmitting data from a user to a generating algorithm.

[0502] "Generation algorithm" refers to a program or artificial intelligence technology that automatically generates answers to user questions.

[0503] "Entering a question" refers to the act of a user entering a question or request in text form through a device.

[0504] The "first answer" refers to the answer that the generation algorithm first generates, and refers to the information that is first presented in response to a question entered by a user.

[0505] "Entering supplementary or correction information" refers to the act of another user entering additional information or correction information in response to an initial response.

[0506] A "re-answer" refers to an answer that is regenerated by the generation algorithm based on supplementary or corrective information for the initial answer.

[0507] The "final answer" refers to a completed answer that is finally presented to the user after multiple answer generation processes.

[0508] MODE FOR CARRYING OUT THE INVENTION

[0509] System Overview

[0510] The system of the present invention is composed of a network system including a user terminal, a server, and a generative model. The user terminal refers to a device such as a smartphone, PC, or tablet, and provides an interface for inputting and receiving questions. The server is a computer system that receives questions from users and sends them to the generative algorithm. The generative algorithm is a program that automatically generates answers to questions, and uses natural language processing technology such as GPT-3.

[0511] Specific program description

[0512] Suppose a user inputs a question using a terminal. For example, "What recent books do you recommend?" This question is sent from the terminal to the server. The server receives this question and prepares it to be sent to the generation algorithm. Specifically, it converts the received text data into an appropriate format (for example, JSON format) and sends it to the generation algorithm using a REST API.

[0513] The generation algorithm analyzes the received question and generates an answer based on natural language processing technology. A generative AI model such as GPT-3 can be used to generate the answer. The generated answer is sent back to the server in the form of, for example, "The recommended book these days is 'Twilight.'"

[0514] The server receives the answer returned by the generation algorithm and sends it back to the user's device. The user can check the generated primary answer through the device interface. At this time, other users can also use their devices to enter supplementary or correction information for the primary answer. For example, supplementary information such as "Twilight is good, but I also recommend the recently published Attack on Titan."

[0515] When additional or corrective information is entered, the server receives this information again and sends it to the generation algorithm. The generation algorithm takes the additional information into account and generates a new answer. The regenerated answer will be in the form of, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'" The server sends this re-answer to the user's device so that the user can confirm it.

[0516] Specific examples

[0517] The system operates when the user enters the following prompt through the terminal:

[0518] "What recent books would you recommend?"

[0519] "Do you have any additional information about 'Twilight'?"

[0520] "Are there any other books you'd recommend?"

[0521] This system allows users to obtain highly accurate answers to their questions, reflecting supplementary and corrective information as it goes along, making it possible to provide satisfactory answers, especially in cases where current information or specialized knowledge is required.

[0522] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0523] Step 1:

[0524] The user inputs a question using the device interface. For example, the user inputs "What books do you recommend these days?" and presses the submit button. At this time, the input data is in text format, and the device captures the data.

[0525] Step 2:

[0526] The terminal sends a question from the user to the server. The data to be sent is in the form of an HTTP request and is transferred through a network path from the device to the server. The server then receives the question data.

[0527] Step 3:

[0528] The server processes the received question data into the appropriate format. Specifically, it converts the text data into JSON format and prepares to send a request to the generation algorithm's API. The input is text data, and the output is a JSON-formatted request.

[0529] Step 4:

[0530] The server sends the appropriately formatted question data to the generation algorithm, typically using a REST API, along with the API endpoint and any necessary authentication information.

[0531] Step 5:

[0532] The generation algorithm analyzes the received question data and generates an initial answer using natural language processing techniques. For example, a generative AI model such as GPT-3 can be used to generate an answer such as "The recommended book these days is 'Twilight.'" The input is the formatted question data, and the output is the generated answer text.

[0533] Step 6:

[0534] The generation algorithm returns the initial answer it generated to the server. The generated answer is returned to the server in JSON format, so the server receives the answer data. The input is the generated answer text, and the output is a JSON-formatted response.

[0535] Step 7:

[0536] The server sends the initial response data received from the generation algorithm to the user device. For example, it sends the data to the device as an HTTP response. At this time, the server sends the response data in an appropriate format to the user device. The input is the response data in JSON format, and the output is the response text sent to the user device.

[0537] Step 8:

[0538] The user checks their initial answer through the device interface and then inputs additional or corrective information based on it. For example, they might type, "I like 'Twilight,' but I also recommend the recently published 'Attack on Titan,'" and press the submit button. The input is supplementary information in text format.

[0539] Step 9:

[0540] The terminal sends supplementary or correction information from the user to the server. This supplementary information is sent in the form of an HTTP request, and the server receives the data. The input is the supplementary information in text format, and the output is the supplementary information sent to the server.

[0541] Step 10:

[0542] The server then processes the received supplemental information into the appropriate format and sends it to the generation algorithm. For example, it converts text data into JSON format and prepares it for resubmission. The input is the supplemental information in text format, and the output is the request in JSON format.

[0543] Step 11:

[0544] The generation algorithm generates a new answer based on the supplementary information. For example, it generates a new answer such as "Recommended books these days are 'Twilight' and 'Attack on Titan'. 'Our Little Sister' is also popular." The input is the formatted supplementary information, and the output is the generated new answer text.

[0545] Step 12:

[0546] The generation algorithm sends the regenerated answer back to the server, which receives the regenerated answer. The input is the regenerated answer text, and the output is a JSON-formatted response.

[0547] Step 13:

[0548] The server sends the re-answer to the user's device. This data is sent as an HTTP response, and the user can confirm the re-answer. The input is the re-answer data in JSON format, and the output is the re-answer text sent to the user's device.

[0549] Step 14:

[0550] The server and the generation algorithm repeat this process as many times as necessary until a final answer is generated, such as "Recommended recent books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'" The input is the repeatedly updated supplementary information, and the output is the final answer text.

[0551] In this way, a system is realized that provides highly accurate answers to user questions that sequentially reflect supplementary and corrective information.

[0552] (Application example 1)

[0553] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0554] In conventional online shopping systems, users must spend a lot of time obtaining detailed information about products, making it difficult to efficiently select appropriate products. Furthermore, if the answers to users' questions are inaccurate, there is a risk that their purchasing decisions will be affected. To solve this problem, a system is needed that provides users with quick and accurate answers to their questions and supports their purchasing decisions.

[0555] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0556] In this invention, the server includes a means for a user to input a question via a terminal, a means for the server to receive the question and transmit it to a generative model in an appropriate format, and a means for the generative model to generate a primary answer to the question and return the primary answer to the server, thereby enabling the user to obtain information about products in a virtual store in real time and make accurate and prompt purchasing decisions.

[0557] "Means for users to input questions via a terminal" refers to a function that allows users to input questions through an interface using a device such as a smartphone, tablet, or PC.

[0558] "Means for the server to receive and send in a format appropriate for the generative model" refers to the function by which the server receives a question from a user, converts it into a format that can be processed by the generative model, and sends it.

[0559] "Means for the generative model to generate a primary answer to a question and return the primary answer to the server" is a function that uses a generative algorithm to create an answer to a user's question and return the answer to the server.

[0560] The "means for the server to transmit the primary answer to the user's terminal" is a function for transmitting the primary answer received from the generative model to the user's device.

[0561] "Means for other users to enter supplementary information or corrections to the primary response via their device" refers to the ability for other users to use their own device to enter additional information or corrections to the response already provided.

[0562] The "means for the server to receive supplementary or corrective information and retransmit it to the generative model" is a function that allows the server to receive supplementary or corrective information and retransmit it to the generative model.

[0563] "Means for the generative model to generate an answer again and return the new answer to the server" is a function that generates a new answer based on supplementary or correction information and returns the new answer to the server.

[0564] The "means for the server to send a re-answer to the user's terminal" is a function for sending the re-answer received from the generative model again to the user's device.

[0565] "Means for allowing users to ask questions about products via a terminal within a virtual store" is a function that allows users to input questions about products they want through an interface within the virtual store.

[0566] "Means for the generative model to generate product information and comparative information based on the question" refers to a function in which the generative model creates detailed information about the product and comparative information with other products based on the content of the user's question.

[0567] "Means for presenting generated product information to the user" is a function that displays information about products and comparison information generated by the generative model on the user's device.

[0568] To implement this invention, a network system including a user terminal, a server, and a generative model must be constructed. The user terminal can be a smartphone, tablet, PC, or other device. The server sends user questions to the generative model in an appropriate format, receives the answers generated by the generative model, and returns them to the user terminal. The generative model can be a program using an advanced natural language processing algorithm, such as OpenAI's GPT-3.

[0569] System Configuration

[0570] 1. User Device

[0571] The user inputs a question via the terminal. This question is input through the interface, for example, "What smartphones do you recommend these days?" This question is then sent from the terminal to the server.

[0572] 2. Server

[0573] The server processes the received question and sends it to the generative model in the appropriate format, which is the prompt for the generative model to generate an answer in the following format:

[0574] User Question: What recent smartphones do you recommend?

[0575] answer:

[0576] 3. Generative Model

[0577] The generative model analyzes the question and generates an answer using natural language processing technology. An example of an answer generated by the generative model is, "The latest recommended smartphones include the iPhone 14 and the Samsung Galaxy S22. They each feature high-performance cameras and fast processors." This answer is then sent back to the server.

[0578] 4. Presenting the Answer

[0579] The server sends the primary answer received from the generative model to the user's device, where the user can check the primary answer.

[0580] 5. Entering Supplementary and Corrective Information

[0581] Other users can enter additional or corrective information to the primary answer. For example, they can enter, "The iPhone 14 is good, but if you prioritize cost-effectiveness, I also recommend the Pixel 6." This additional data is sent to the server.

[0582] 6. Generate answers again

[0583] The server then sends supplementary or corrective information to the generative model. The generative model then generates a new answer based on the new information. An example of a new answer might be, "Recommended smartphones are the iPhone 14 and the Samsung Galaxy S22. If cost-effectiveness is important, consider the Pixel 6." This new answer is also sent back to the server and then to the user's device.

[0584] This system configuration allows users to obtain fast and accurate information, which can greatly support purchasing decisions. Specific hardware requirements include a server and user devices (smartphones, PCs, tablets, etc.), and software requirements include Flask (a Python web framework), OpenAI API, and a generative model (GPT-3).

[0585] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0586] Step 1:

[0587] The user inputs a question via a terminal.

[0588] Users input questions using an interface from devices such as smartphones or PCs. The input questions are in the form of "What smartphones do you recommend these days?". The user's question is generated as input data.

[0589] Step 2:

[0590] The terminal sends a question to the server.

[0591] The terminal sends the question entered by the user to the server. When the server receives the question, it proceeds to the next processing step. The input includes the user's question data, and the output is the question data transferred to the server.

[0592] Step 3:

[0593] The server sends the received question to the generative model in the appropriate format.

[0594] The server analyzes the received question and formats it in a format that can be processed by the generative model. Specifically, the question is converted into a prompt sentence. It is converted into the format "User question: What smartphone do you recommend these days?\nAnswer: ". The input is the user's question data, and the output is a generated prompt sentence.

[0595] Step 4:

[0596] A generative model generates a first-order answer to the question.

[0597] A generative model (e.g., GPT-3) takes a prompt as input, analyzes it, generates it, and produces a first-order answer. For example, the model might generate an answer like, "The latest recommended smartphones are the iPhone 14 and the Samsung Galaxy S22. They each feature a high-performance camera and a fast processor." The input contains a prompt, and the output is a first-order answer.

[0598] Step 5:

[0599] The server sends the primary response to the user terminal.

[0600] The server sends the primary answer received from the generative model to the user's device, allowing the user to check the primary answer on their own device. The input is the primary answer from the generative model, and the output is the response data sent to the user's device.

[0601] Step 6:

[0602] Other users input supplements or corrections to the primary answer via their terminals.

[0603] Other users can add supplementary or correction information to the answers already provided. For example, "The iPhone 14 is good, but if cost performance is important, the Pixel 6 is also recommended." The input includes the initial answer and supplementary or correction information, and the output is the supplementary or correction data that is sent to the server.

[0604] Step 7:

[0605] The server receives the supplementary or corrective information and resubmits it to the generative model.

[0606] The server receives supplementary or correction information from other users and sends it back to the generative model. At this time, the supplementary or correction information is added to the prompt sentence. In this way, the server is ready to generate an answer again. The input is the supplementary or correction information, and the output is the data to be resubmitted to the generative model.

[0607] Step 8:

[0608] The generative model generates the answer again and sends the answer back to the server.

[0609] The generative model generates a new answer based on new information. For example, it generates a new answer in the form of "Recommended smartphones are the iPhone 14 and Samsung Galaxy S22. If cost-effectiveness is important, consider the Pixel 6." The input is a prompt sentence containing supplementary information and corrections, and the output is a new answer.

[0610] Step 9:

[0611] The server sends a second response to the user's terminal.

[0612] The server then sends the re-answer received from the generative model back to the user device, allowing the user to check the re-answer on their device. The input is the re-answer from the generative model, and the output is the re-answer data sent to the user device.

[0613] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0614] An embodiment for implementing the system of the present invention will be described below.

[0615] System Configuration

[0616] The present invention can be implemented in a network system including a user terminal, a server, a generative model, and an emotion engine. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative model and emotion engine. The generative model is a program that uses an artificial intelligence algorithm to generate answers to questions, and the emotion engine recognizes user emotions and adjusts the behavior of the generative model based on those emotions.

[0617] User question input

[0618] The user inputs a question using a terminal. For example, the user inputs and sends a question such as, "What recent books do you recommend?" The emotion engine analyzes the user's emotions along with the question, and sends the results to the server as additional information.

[0619] Submitting questions and generating answers

[0620] The server processes the received question along with the emotional information and sends it to the generative model in an appropriate format. The generative model takes the question and emotional information into account to generate an answer. For example, if the answer is "The book I recommend these days is 'Twilight,'" the answer is adjusted based on the emotional information.

[0621] Presenting the first answer

[0622] The server sends the primary answer received from the generative model to the user's device, where the user can check the primary answer.

[0623] Entering Supplementary and Corrective Information

[0624] Other users can add supplementary or correction information to the primary answer. For example, they can add supplementary information such as, "Twilight is good, but I also recommend the recently published Attack on Titan." In this case, the emotion engine also recognizes the emotion of the user who added the supplementary or correction information and sends that information to the server.

[0625] Generate answers again

[0626] The server sends supplementary or corrective information and the emotion information to the generative model. The generative model generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'"

[0627] Providing a second answer

[0628] The server sends the regenerated answer to the user's terminal, where the user can review the revised answer and other users can enter additional comments or corrections.

[0629] Third and final answers

[0630] By repeating the re-answering process multiple times as necessary, the generative model generates a final answer. For example, a final answer might be, "My recent recommended books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'" During this process, the emotion engine continuously analyzes the user's emotions and optimizes the answer.

[0631] In this way, the system according to the present invention can provide a highly accurate final answer that takes into account the user's emotional information.

[0632] The processing flow will be explained below.

[0633] Specific processing steps of the system

[0634] Step 1:

[0635] The user inputs a question from the terminal.

[0636] Specific operation: The user inputs a question using the device interface and presses the "Send" button. At this time, the emotion engine analyzes the user's tone of voice and facial expressions to extract emotional information. The device then transmits the input question and emotional information to the server.

[0637] Step 2:

[0638] The device sends the question and emotion information to the server.

[0639] Specific operation: The device sends the question entered by the user and the emotional information analyzed by the emotion engine to the server using a pre-configured protocol (e.g., HTTP request).

[0640] Step 3:

[0641] The server receives the question and emotion information and sends it to the generative model in an appropriate format.

[0642] Specific operation: The server receives the question and emotion information, converts it into a format that the generative model can understand (e.g., JSON or XML), and sends it to the generative model.

[0643] Step 4:

[0644] The generative model generates a first-order answer based on the question and emotion information and sends it back to the server.

[0645] How it works: The generative model analyzes the question using natural language processing techniques and generates a first-order answer. The tone and content of the answer are adjusted based on the emotional information. This first-order answer is then sent back to the server.

[0646] Step 5:

[0647] The server sends the primary response to the user's terminal.

[0648] Specific operation: The server sends the initial answer received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0649] Step 6:

[0650] The user reviews the primary answer and adds any additions or corrections.

[0651] Specific operation: Other users read the primary response through their devices and enter corrections or supplementary information. The emotion engine recognizes the user's emotions while they are entering and sending the response, adds that information, and sends it to the server.

[0652] Step 7:

[0653] The terminal transmits supplementary or corrective information and emotion information to the server.

[0654] Specific operation: The device sends supplementary or corrective information, as well as the emotion information analyzed by the emotion engine, to the server via a pre-configured protocol (e.g., HTTP request).

[0655] Step 8:

[0656] The server retransmits the supplementary or corrective information and the emotion information to the generative model.

[0657] Specific behavior: The server receives the supplementary or corrective information, as well as the emotion information, and sends it back to the generative model in an appropriate format (e.g., reformatted JSON data).

[0658] Step 9:

[0659] The generative model generates a re-answer and sends it back to the server.

[0660] What it does: The generative model generates a new answer based on the new information and emotional information. The emotional information adjusts the tone and content of the new answer. The new answer is then sent back to the server.

[0661] Step 10:

[0662] The server sends a second response to the user's terminal.

[0663] Specific operation: The server sends the response received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0664] Step 11:

[0665] Further additions or corrections (if necessary).

[0666] Specific operation: Another user checks the answer again, enters additional or corrective information as necessary, and presses the "Send" button. During input and submission, the emotion engine recognizes the user's emotions, adds that information, and sends it to the server.

[0667] Step 12:

[0668] Repeat steps 6-10 above as needed.

[0669] What it does: The process continues, supplementing or correcting as needed, until a final answer is produced. Emotional information is continually considered throughout the process.

[0670] Through these steps, a highly accurate final answer that reflects the user's emotional information is obtained.

[0671] Example 2

[0672] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0673] Conventional question-answering systems have had the problem of difficulty in providing appropriate answers based on the user's emotions. In particular, they have had problems with generating answers that ignore the user's emotions, resulting in poor answer quality and difficulty in achieving user satisfaction. Furthermore, when additional clarifications or corrections are added, there is a lack of a mechanism for generating appropriate re-answers that take these into account. To solve these problems, highly accurate answer generation that reflects the user's emotional information is required.

[0674] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a question via a terminal, a means for the server to receive the question and the user's emotional information and transmit them to the generative model in an appropriate format, and a means for the generative model to generate a primary answer based on the question and the emotional information and return the primary answer to the server. This enables the generation of a high-quality answer that takes into account the user's emotional information.

[0675] "User" refers to an entity that utilizes the system to enter questions and receive answers.

[0676] "Device" refers to the device used by a user to enter questions and check answers, such as a smartphone, computer, or tablet.

[0677] "Server" refers to the computer system that receives questions from users and sends information to the generative model and emotion engine.

[0678] "Question" refers to a question or request entered by a user via a terminal.

[0679] A "generative model" refers to a program that uses an artificial intelligence algorithm to generate answers based on questions and emotional information.

[0680] "Emotion engine" refers to software that identifies and analyzes a user's emotions.

[0681] "Emotion information" refers to data relating to the user's emotions analyzed by the emotion engine.

[0682] "Primary answer" refers to the first answer generated by the generative model.

[0683] "Supplement or correction" refers to additional information or corrections entered by other users in response to the primary answer.

[0684] A "re-answer" refers to a second or subsequent answer generated by the generative model based on supplementary or corrective information and emotional information.

[0685] "Final Answer" refers to the final answer generated through the repeated answering process.

[0686] The system of the present invention generates answers to questions from users via a network system and provides highly accurate answers that take into account the user's emotional information using an emotion engine. The system of the present invention mainly includes the following components: a user terminal, a server, a generative model, and an emotion engine.

[0687] System Configuration

[0688] User terminal

[0689] A user terminal is a device that allows a user to input and send questions, such as a smartphone, PC, or tablet. The user terminal includes an input interface and a display interface, sends the questions input by the user to the server, and displays the answers from the server.

[0690] server

[0691] The server is responsible for receiving questions from users and sending them to the generative model and emotion engine. The server sends the emotion information analyzed by the emotion engine, along with the question and any supplementary or correction information, to the generative model in an appropriate format.

[0692] Generative Model

[0693] A generative model is a program that uses artificial intelligence algorithms to generate answers based on questions and sentiment information. For example, a natural language processing model such as GPT-3 is used. A generative model generates a first answer, a follow-up answer, and a final answer.

[0694] Emotion Engine

[0695] The emotion engine is software that analyzes the user's emotions and generates emotional information to accompany the user's questions and supplementary / corrective information, allowing the generative model to generate answers based on emotions.

[0696] Specific examples

[0697] The user uses a terminal to input and send a question such as "What recent books do you recommend?" The server receives this question and analyzes the user's emotions using an emotion engine. If the emotion engine determines that the user is curious, the server sends this emotional information along with the question to the generative model.

[0698] The generative model generates an answer based on the question and emotional information. For example, a primary answer such as "The recommended book these days is 'Twilight.' Another popular book is 'Attack on Titan.'" is generated and sent to the user's device via the server.

[0699] When another user adds additional information, such as "Twilight is good, but I also recommend the recently published Our Little Sister," the emotion engine analyzes the user's emotions. The server sends this new emotional information to the generative model, which then generates a new answer.

[0700] As a follow-up answer, the following answer is generated: "My recent recommended books are 'Twilight' and 'Our Little Sister.' In addition, 'Demon Slayer: Kimetsu no Yaiba' is also highly rated," and is sent to the user's device via the server.

[0701] Prompt Sentence Examples

[0702] "The user asked for recent book recommendations. The sentiment engine determined that the user is curious. Please take this into account when generating an answer."

[0703] In this way, the system of the present invention can provide highly accurate answers that take into account the user's emotional information, thereby improving user satisfaction.

[0704] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0705] The flow of this system's program processing

[0706] Specific explanation of processing steps

[0707] Step 1:

[0708] The user inputs and sends a question via a terminal.

[0709] Input: The user types into the terminal, "What recent books have you recommended?"

[0710] Data processing: When the send button on the terminal is pressed, the question text is sent to the server using an HTTP POST request.

[0711] Output: The question is sent to the server.

[0712] Step 2:

[0713] The server receives the question and passes it to the emotion engine for analysis of emotion information.

[0714] Input: The user's question sent to the server: "What books have you recommended recently?"

[0715] Data processing: The emotion engine performs sentiment analysis on the text to determine the user's emotion (e.g., curious).

[0716] Output: Analysis results including emotional information.

[0717] Step 3:

[0718] The server sends the question and emotion information to the generative model.

[0719] Input: Parsed emotion information (e.g., curious) and user question.

[0720] Data processing: The question and emotion information are sent to the generative model server in a data format such as JSON.

[0721] Output: The generative model receives the question and sentiment information.

[0722] Step 4:

[0723] A generative model generates a first-order answer based on the question and sentiment information.

[0724] Input: The question received by the generative model (e.g., "What recent books do you recommend?") and sentiment information.

[0725] Data processing: A generative model (e.g., GPT-3) analyzes the input prompt and generates a text answer.

[0726] Output: Text of the primary answer (e.g., "The book I recommend these days is 'Twilight.' Another popular book is 'Attack on Titan.'").

[0727] Step 5:

[0728] The server sends the primary response to the user's terminal.

[0729] Input: The text of the primary answer returned by the generative model.

[0730] Data processing: The server sends the primary answer to the user terminal as an HTTP response.

[0731] Output: The primary answer is displayed on the user's terminal.

[0732] Step 6:

[0733] Other users input supplementary or corrective information to the primary answer via the terminal.

[0734] Input: Another user types into their device, "Twilight is good, but I also recommend the recently published Our Little Sister."

[0735] Data processing: The entered supplementary or corrective information is sent to the server using an HTTP POST request.

[0736] Output: Any supplementary or corrective information is sent to the server.

[0737] Step 7:

[0738] The server retransmits the supplementary or corrective information and the emotion information to the generative model.

[0739] Input: Supplementary or correction information and sentiment information received by the server.

[0740] Data processing: Supplementary or corrective information and emotion information are resubmitted to the generative model server in a data format such as JSON.

[0741] Output: The generative model receives new information needed to generate a new answer.

[0742] Step 8:

[0743] The generative model generates a re-answer.

[0744] Input: New supplementary or correction information received by the generative model (e.g., "Twilight is good, but I also recommend the recently published Our Little Sister") and sentiment information.

[0745] Data processing: The generative model generates a new answer based on new information (e.g., "Recommended books these days are 'Twilight' and 'Our Little Sister.' Other popular books include 'Demon Slayer: Kimetsu no Yaiba.'").

[0746] Output: The text of the re-answer.

[0747] Step 9:

[0748] The server sends a second response to the user's terminal.

[0749] Input: The text of the re-answer returned by the generative model.

[0750] Data processing: The server sends a second response to the user terminal as an HTTP response.

[0751] Output: The answer is displayed on the user's terminal.

[0752] Thus, the system takes a series of steps to take affective information into account and provide highly accurate answers to the user's questions.

[0753] (Application example 2)

[0754] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0755] Conventional content distribution services exist that generate appropriate answers to user questions. However, because they do not take the user's emotions into account, the answers often do not match the user's wishes or circumstances. Emotional information is particularly important in entertainment, and ignoring it reduces satisfaction. Furthermore, because feedback from other users cannot be utilized, the answers lack accuracy and adaptability. Therefore, a system that analyzes the user's emotional information and reflects it in answer generation is needed.

[0756] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0757] In this invention, the server includes: a means for a user to input a question via a terminal; a means for the server to receive the question and transmit it to the generative model in an appropriate format; a means for the generative model to generate a primary answer to the question and return the primary answer to the server; a means for the server to transmit the primary answer to the user's terminal; a means for another user to input a supplement or correction to the primary answer via the terminal; a means for the server to receive the supplement or correction information and retransmit it to the generative model; a means for the generative model to generate a new answer and return the new answer to the server; a means for the server to transmit the new answer to the user's terminal; a means for analyzing the user's emotions using an emotion engine and adjusting the behavior of the generative model based on the analysis; and a means for updating the recommendation content based on feedback from other users. This makes it possible to provide more appropriate and satisfying answers that reflect the user's emotional information.

[0758] "Means for users to input questions via a terminal" refers to a mechanism that provides an interface for users to input and send questions using devices such as smartphones, tablets, and PCs.

[0759] "Means for the server to receive and send in an appropriate format to the generative model" refers to a mechanism by which the server receives questions entered by users, converts them into a format that can be processed by the generative AI model, and sends them.

[0760] "Means for the generative model to generate a first answer to a question and return the first answer to the server" refers to a mechanism by which the generative AI model generates an initial answer to a user's question and sends that answer to the server.

[0761] "Means by which the server sends the primary answer to the user's device" refers to a mechanism by which the server receives the primary answer obtained from the generative AI model and forwards it to the user's device.

[0762] "Means for other users to input supplementary information or corrections to the primary response via a terminal" refers to a system that provides an interface for other users to input additional information or corrections to the primary response using a smartphone, tablet, PC, etc.

[0763] "Means for the server to receive supplementary or corrective information and retransmit it to the generative model" refers to a mechanism by which, after additional information or corrections are entered, the server receives that information and retransmits it to the generative AI model.

[0764] "Means for the generative model to generate a new answer and return the new answer to the server" refers to a mechanism by which the generative AI model generates a new answer based on supplements or corrections and sends that answer to the server.

[0765] "Means for the server to send the re-answer to the user's device" refers to a mechanism by which the server receives the re-answer obtained from the generative AI model and forwards it to the user's device.

[0766] "Means of analyzing user emotions using an emotion engine and adjusting the behavior of the generative model based on that" refers to a mechanism that uses emotion analysis technology to detect user emotions and optimizes the content and format of the generative AI model's responses based on the results.

[0767] "Means for updating recommendations based on feedback from other users" refers to a mechanism that improves the recommendations of the generative AI model by reflecting opinions and ratings provided by other users.

[0768] Specific embodiments for carrying out the present invention will be described below.

[0769] System Configuration

[0770] This invention is embodied in a network system including a user terminal, a server, a generative AI model, and an emotion engine. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative AI model and the emotion engine. The generative AI model is a program that uses an artificial intelligence algorithm to generate answers to questions, and the emotion engine recognizes the user's emotions and adjusts the behavior of the generative AI model based on those emotions.

[0771] User question input

[0772] A user inputs a question using a device such as a smartphone or tablet. For example, a user might input and send a question such as, "I've been looking for a heartwarming movie lately. Do you have any recommendations?" The emotion engine analyzes the user's emotions along with the question and sends the results to the server as additional information.

[0773] Submitting questions and generating answers

[0774] The server processes the received question along with the emotional information and sends it to the generative AI model in an appropriate format. The generative AI model takes the question and emotional information into consideration to generate an answer. For example, if the answer generated is "A recent heartwarming movie I recommend is 'Green Book,'" the content and format of the answer are adjusted based on the emotional information.

[0775] Presenting the first answer

[0776] The server sends the initial answer received from the generative AI model to the user's device, where the user can check the initial answer.

[0777] Entering Supplementary and Corrective Information

[0778] Other users can add supplementary or correction information to the primary answer. For example, they can add supplementary information such as, "Green Book is good, but I also recommend the recently released Shawshank Redemption." In this case, the emotion engine also recognizes the emotion of the user who added the supplementary or correction information and sends that information to the server.

[0779] Generate answers again

[0780] The server sends supplementary or corrective information and the emotional information to the generative AI model. The generative AI model generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended recent heartwarming movies are 'Green Book' and 'The Shawshank Redemption.' Other popular movies include 'Les Miserables.'"

[0781] Providing a second answer

[0782] The server sends the regenerated answer to the user's terminal, where the user can review the revised answer and other users can enter additional comments or corrections.

[0783] Generate the final answer

[0784] The generative AI model generates a final answer by repeating the process of re-answering multiple times as necessary. During this process, the emotion engine continuously analyzes the user's emotions and optimizes the answer. For example, the final answer might be, "Recommended recent heartwarming movies are 'Green Book,' 'The Shawshank Redemption,' 'Les Miserables,' and 'Twilight.'"

[0785] In this way, the system according to the present invention can provide a highly accurate final answer that takes into account the user's emotional information.

[0786] Hardware and software used

[0787] Hardware: Smartphones, tablets, computers

[0788] Software: The program is implemented in Python and utilizes APIs for emotion engines and generative AI models.

[0789] Emotion Engine API (e.g. https: / / emotion-api.example.com / analyze)

[0790] Generate AI Model API (e.g. https: / / ai-model-api.example.com / generate)

[0791] Examples of prompt statements

[0792] User Question: I've been looking for a heartwarming movie lately. Any recommendations?

[0793] Emotional information: Relaxing, Heartwarming, Comforting

[0794] Output of generative AI model: A recent recommended heartwarming movie is "Green Book," but other good choices are "The Shawshank Redemption" and "Les Miserables."

[0795] According to this invention, it is possible to improve the accuracy of answers by incorporating information about the user's emotions, and to recommend content with a higher degree of satisfaction.

[0796] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0797] Step 1:

[0798] The user inputs a question via the terminal. The user inputs a question into a smartphone or tablet and sends it. An example of input is, "I've been looking for a heartwarming movie lately. Do you have any recommendations?"

[0799] Step 2:

[0800] The terminal sends a question and the server receives it. The terminal sends the question entered by the user to the server. The server receives this question and proceeds to the next step.

[0801] Step 3:

[0802] The server sends the received question to the emotion engine to obtain emotion information. The server sends the question to the emotion engine (e.g. https: / / emotion-api.example.com / analyze). The input is the question text, and the output is emotion information. For example, the emotion information detected is "relaxing, heartwarming, comforting."

[0803] Step 4:

[0804] The server sends the question and emotion information to the generative AI model. After receiving the emotion information, the server sends it along with the question to the generative AI model (e.g., https: / / ai-model-api.example.com / generate). The input is the question and emotion information, and the generative AI model generates a first answer based on this.

[0805] Step 5:

[0806] The generative AI model generates a first answer and sends it back to the server. The generative AI model generates an answer based on the input question and emotional information. For example, it generates an answer such as "A recent recommended heartwarming movie is 'Green Book'." This answer is sent to the server.

[0807] Step 6:

[0808] The server sends the primary answer to the user's device. The server sends the primary answer received from the generative AI model to the user's device. The user can check the primary answer through their device.

[0809] Step 7:

[0810] Other users can input supplementary information or corrections to the primary answer via the terminal. Other users can input additional information or corrections to the primary answer. For example, they can add supplementary information such as, "'Green Book' is good, but I also recommend the recently released 'Shawshank Redemption'."

[0811] Step 8:

[0812] The terminal transmits supplementary or corrective information to the server, which receives it. The terminal transmits supplementary or corrective information from other users to the server, which receives it.

[0813] Step 9:

[0814] The server retransmits the supplementary or correction information and emotional information to the generative AI model. The server retransmits the received supplementary or correction information and the emotional information to the generative AI model. The input is the supplementary or correction information and emotional information.

[0815] Step 10:

[0816] The generative AI model generates a re-answer based on the new information and sends it back to the server. For example, it might generate a re-answer such as, "Recommended recent heartwarming movies are 'Green Book' and 'The Shawshank Redemption.' Other popular movies include 'Les Miserables.'" and send it to the server.

[0817] Step 11:

[0818] The server sends the re-answer to the user's device. The server sends the re-answer received from the generative AI model to the user's device, where the user can check the re-answer.

[0819] Step 12:

[0820] If necessary, other users can enter additional supplementary or correction information, which is then received by the server and sent to the generative AI model. Finally, the generative AI model generates a final answer. For example, the final answer might be, "Recommended recent heartwarming movies are 'Green Book,' 'The Shawshank Redemption,' 'Les Miserables,' and 'Twilight.'"

[0821] Step 13:

[0822] The server sends the final answer to the user's device. The server sends the final answer received from the generative AI model to the user's device, where the user can check the final answer.

[0823] This series of processes enables highly accurate content recommendations that reflect the user's emotional information.

[0824] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0825] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0826] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0827] [Third embodiment]

[0828] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0829] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0830] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0831] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0832] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0833] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0834] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0835] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0836] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0837] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0838] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0839] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0840] An embodiment for implementing the system of the present invention will be described below.

[0841] System Configuration

[0842] The present invention can be implemented in a network system including a user terminal, a server, and a generative model. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative model. The generative model is a program that uses an artificial intelligence algorithm to generate answers to questions.

[0843] User question input

[0844] A user inputs a question using a terminal. For example, the user inputs and sends a question such as "What recent books do you recommend?" through the terminal interface. This question is sent from the terminal to the server.

[0845] Submitting questions and generating answers

[0846] The server processes the received question and sends it to the generative model in the appropriate format. The generative model analyzes the question and generates an answer using natural language processing techniques. For example, the generative model generates an answer such as "The recommended book these days is 'Twilight'" and sends it back to the server.

[0847] Presenting the first answer

[0848] The server sends the primary answer received from the generative model to the user's terminal, where the user can check the primary answer.

[0849] Entering Supplementary and Corrective Information

[0850] Other users can add supplementary or corrective information to the primary answer. For example, another user can add supplementary information such as "Twilight is good, but I also recommend the recently published Attack on Titan," and send it to the server via their device.

[0851] Generate answers again

[0852] The server then sends additional or corrective information to the generative model. The generative model then generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'" and send it back to the server.

[0853] Providing a second answer

[0854] The server then sends the regenerated answer to the user's terminal, where the user can confirm the revised answer. In addition, other users can enter additional comments or corrections.

[0855] Third and final answers

[0856] By repeating the re-answering process multiple times as necessary, the generative model generates a final answer, such as "My recent recommended books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'"

[0857] In this way, the system based on the present invention can provide highly accurate answers to user questions while reflecting supplements and corrections from other users.

[0858] The processing flow will be explained below.

[0859] Specific processing steps of the program

[0860] Step 1:

[0861] The user inputs a question from the terminal.

[0862] Specific operation: The user inputs a question using the terminal interface and presses the "Send" button. At this time, the terminal converts the input question into a digital format and sends it to the server.

[0863] Step 2:

[0864] The terminal sends a question to the server.

[0865] Specific operation: The terminal sends the question entered by the user to the server using a pre-configured protocol (e.g., HTTP request).

[0866] Step 3:

[0867] The server receives the question and sends it to the generative model in the appropriate format.

[0868] Specific operation: The server receives the question, converts it into a format that the generative model can understand (e.g., JSON or XML), and sends it to the generative model.

[0869] Step 4:

[0870] The generative model generates a first-order answer to the question and sends it back to the server.

[0871] How it works: The generative model uses natural language processing techniques to analyze the question and generate a first-order answer, which is then sent back to the server.

[0872] Step 5:

[0873] The server sends the primary response to the user's terminal.

[0874] Specific operation: The server sends the initial answer received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0875] Step 6:

[0876] The user reviews the primary answer and adds any additions or corrections.

[0877] Specific operations: Other users read the primary response through their terminals, enter corrections or supplementary information, and press the "Submit" button once the supplements or corrections are complete.

[0878] Step 7:

[0879] The terminal sends the supplementary or corrective information to the server.

[0880] Specific operation: The device sends supplementary or corrective information to the server using a pre-configured protocol (e.g., HTTP request).

[0881] Step 8:

[0882] The server resubmits the supplementary or corrective information to the generative model.

[0883] Specific behavior: The server receives the supplementary or corrective information and sends it back to the generative model in an appropriate format (e.g., reformatted JSON data).

[0884] Step 9:

[0885] The generative model generates a re-answer and sends it back to the server.

[0886] What happens next: The generative model generates a new answer based on the new information, and this answer is sent back to the server.

[0887] Step 10:

[0888] The server sends a second response to the user's terminal.

[0889] Specific operation: The server sends the response received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[0890] Step 11:

[0891] Further additions or corrections (if necessary).

[0892] Specific actions: Another user will check the answer again, enter additional or corrective information if necessary, and press the "Submit" button.

[0893] Step 12:

[0894] Repeat steps 6-10 above as needed.

[0895] Specific Action: Repeat the process of supplementing or correcting as necessary and continue the process until a final answer is produced.

[0896] Through these steps, a highly accurate final answer is obtained.

[0897] Example 1

[0898] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0899] In conventional question-answering systems, answers to user questions are often fixed, and when other users add supplementary or correction information, the answers are not automatically updated. As a result, answers to questions are not always optimal, making it difficult for users to obtain satisfactory answers, especially when current information or individual knowledge is required.

[0900] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0901] In this invention, the server includes a means for a user to input a question via a device, a means for a computer system to receive the question and send it to a generation algorithm in an appropriate format, and a means for the generation algorithm to generate an initial answer to the question and return the initial answer to the computer system. This makes it possible to obtain a highly accurate answer to the question that takes into account supplementary and correction information as needed. Furthermore, by generating and presenting new answers multiple times as needed, it is possible to ultimately provide an answer that satisfies the user.

[0902] "Device" means an electronic device that allows a user to input or receive information, including a smartphone, computer, tablet, etc.

[0903] "Computer System" refers to a computer network system for receiving, processing, and transmitting data from a user to a generating algorithm.

[0904] "Generation algorithm" refers to a program or artificial intelligence technology that automatically generates answers to user questions.

[0905] "Entering a question" refers to the act of a user entering a question or request in text form through a device.

[0906] The "first answer" refers to the answer that the generation algorithm first generates, and refers to the information that is first presented in response to a question entered by a user.

[0907] "Entering supplementary or correction information" refers to the act of another user entering additional information or correction information in response to an initial response.

[0908] A "re-answer" refers to an answer that is regenerated by the generation algorithm based on supplementary or corrective information for the initial answer.

[0909] The "final answer" refers to a completed answer that is finally presented to the user after multiple answer generation processes.

[0910] MODE FOR CARRYING OUT THE INVENTION

[0911] System Overview

[0912] The system of the present invention is composed of a network system including a user terminal, a server, and a generative model. The user terminal refers to a device such as a smartphone, PC, or tablet, and provides an interface for inputting and receiving questions. The server is a computer system that receives questions from users and sends them to the generative algorithm. The generative algorithm is a program that automatically generates answers to questions, and uses natural language processing technology such as GPT-3.

[0913] Specific program description

[0914] Suppose a user inputs a question using a terminal. For example, "What recent books do you recommend?" This question is sent from the terminal to the server. The server receives this question and prepares it to be sent to the generation algorithm. Specifically, it converts the received text data into an appropriate format (for example, JSON format) and sends it to the generation algorithm using a REST API.

[0915] The generation algorithm analyzes the received question and generates an answer based on natural language processing technology. A generative AI model such as GPT-3 can be used to generate the answer. The generated answer is sent back to the server in the form of, for example, "The recommended book these days is 'Twilight.'"

[0916] The server receives the answer returned by the generation algorithm and sends it back to the user's device. The user can check the generated primary answer through the device interface. At this time, other users can also use their devices to enter supplementary or correction information for the primary answer. For example, supplementary information such as "Twilight is good, but I also recommend the recently published Attack on Titan."

[0917] When additional or corrective information is entered, the server receives this information again and sends it to the generation algorithm. The generation algorithm takes the additional information into account and generates a new answer. The regenerated answer will be in the form of, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'" The server sends this re-answer to the user's device so that the user can confirm it.

[0918] Specific examples

[0919] The system operates when the user enters the following prompt through the terminal:

[0920] "What recent books would you recommend?"

[0921] "Do you have any additional information about 'Twilight'?"

[0922] "Are there any other books you'd recommend?"

[0923] This system allows users to obtain highly accurate answers to their questions, reflecting supplementary and corrective information as it goes along, making it possible to provide satisfactory answers, especially in cases where current information or specialized knowledge is required.

[0924] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0925] Step 1:

[0926] The user inputs a question using the device interface. For example, the user inputs "What books do you recommend these days?" and presses the submit button. At this time, the input data is in text format, and the device captures the data.

[0927] Step 2:

[0928] The terminal sends a question from the user to the server. The data to be sent is in the form of an HTTP request and is transferred through a network path from the device to the server. The server then receives the question data.

[0929] Step 3:

[0930] The server processes the received question data into the appropriate format. Specifically, it converts the text data into JSON format and prepares to send a request to the generation algorithm's API. The input is text data, and the output is a JSON-formatted request.

[0931] Step 4:

[0932] The server sends the appropriately formatted question data to the generation algorithm, typically using a REST API, along with the API endpoint and any necessary authentication information.

[0933] Step 5:

[0934] The generation algorithm analyzes the received question data and generates an initial answer using natural language processing techniques. For example, a generative AI model such as GPT-3 can be used to generate an answer such as "The recommended book these days is 'Twilight.'" The input is the formatted question data, and the output is the generated answer text.

[0935] Step 6:

[0936] The generation algorithm returns the initial answer it generated to the server. The generated answer is returned to the server in JSON format, so the server receives the answer data. The input is the generated answer text, and the output is a JSON-formatted response.

[0937] Step 7:

[0938] The server sends the initial response data received from the generation algorithm to the user device. For example, it sends the data to the device as an HTTP response. At this time, the server sends the response data in an appropriate format to the user device. The input is the response data in JSON format, and the output is the response text sent to the user device.

[0939] Step 8:

[0940] The user checks their initial answer through the device interface and then inputs supplementary or correction information based on it. For example, they might type, "I like 'Twilight,' but I also recommend the recently published 'Attack on Titan,'" and press the submit button. The input is supplementary information in text format.

[0941] Step 9:

[0942] The terminal sends supplementary or correction information from the user to the server. This supplementary information is sent in the form of an HTTP request, and the server receives the data. The input is the supplementary information in text format, and the output is the supplementary information sent to the server.

[0943] Step 10:

[0944] The server then processes the received supplemental information into the appropriate format and sends it to the generation algorithm. For example, it converts text data into JSON format and prepares it for resubmission. The input is the supplemental information in text format, and the output is the request in JSON format.

[0945] Step 11:

[0946] The generation algorithm generates a new answer based on the supplementary information. For example, it generates a new answer such as "Recommended books these days are 'Twilight' and 'Attack on Titan'. 'Our Little Sister' is also popular." The input is the formatted supplementary information, and the output is the generated new answer text.

[0947] Step 12:

[0948] The generation algorithm sends the regenerated answer back to the server, which receives the regenerated answer. The input is the regenerated answer text, and the output is a JSON-formatted response.

[0949] Step 13:

[0950] The server sends the re-answer to the user's device. This data is sent as an HTTP response, and the user can confirm the re-answer. The input is the re-answer data in JSON format, and the output is the re-answer text sent to the user's device.

[0951] Step 14:

[0952] The server and the generation algorithm repeat this process as many times as necessary until a final answer is generated, such as "Recommended recent books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'" The input is the repeatedly updated supplementary information, and the output is the final answer text.

[0953] In this way, a system is realized that provides highly accurate answers to user questions that sequentially reflect supplementary and corrective information.

[0954] (Application example 1)

[0955] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0956] In conventional online shopping systems, users must spend a lot of time obtaining detailed information about products, making it difficult to efficiently select appropriate products. Furthermore, if the answers to users' questions are inaccurate, there is a risk that their purchasing decisions will be affected. To solve this problem, a system is needed that provides users with quick and accurate answers to their questions and supports their purchasing decisions.

[0957] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0958] In this invention, the server includes a means for a user to input a question via a terminal, a means for the server to receive the question and transmit it to a generative model in an appropriate format, and a means for the generative model to generate a primary answer to the question and return the primary answer to the server, thereby enabling the user to obtain information about products in a virtual store in real time and make accurate and prompt purchasing decisions.

[0959] "Means for users to input questions via a terminal" refers to a function that allows users to input questions through an interface using a device such as a smartphone, tablet, or PC.

[0960] "Means for the server to receive and send in a format appropriate for the generative model" refers to the function by which the server receives a question from a user, converts it into a format that can be processed by the generative model, and sends it.

[0961] "Means for the generative model to generate a primary answer to a question and return the primary answer to the server" is a function that uses a generative algorithm to create an answer to a user's question and return the answer to the server.

[0962] The "means for the server to transmit the primary answer to the user's terminal" is a function for transmitting the primary answer received from the generative model to the user's device.

[0963] "Means for other users to enter supplementary information or corrections to the primary response via their device" refers to the ability for other users to use their own device to enter additional information or corrections to the response already provided.

[0964] "Means for the server to receive supplementary or corrective information and retransmit it to the generative model" is a function that allows the server to receive supplementary or corrective information and retransmit it to the generative model.

[0965] "Means for the generative model to generate an answer again and return the new answer to the server" is a function that generates a new answer based on supplementary or correction information and returns the new answer to the server.

[0966] The "means for the server to send a re-answer to the user's terminal" is a function for sending the re-answer received from the generative model again to the user's device.

[0967] "Means for allowing users to ask questions about products via a terminal within a virtual store" is a function that allows users to input questions about products they want through an interface within the virtual store.

[0968] "Means for the generative model to generate product information and comparative information based on the question" refers to a function in which the generative model creates detailed information about the product and comparative information with other products based on the content of the user's question.

[0969] "Means for presenting generated product information to the user" is a function that displays information about products and comparison information generated by the generative model on the user's device.

[0970] To implement this invention, a network system including a user terminal, a server, and a generative model must be constructed. The user terminal can be a smartphone, tablet, PC, or other device. The server sends user questions to the generative model in an appropriate format, receives the answers generated by the generative model, and returns them to the user terminal. The generative model can be a program using an advanced natural language processing algorithm, such as OpenAI's GPT-3.

[0971] System Configuration

[0972] 1. User Device

[0973] The user inputs a question via the terminal. This question is input through the interface, for example, "What smartphones do you recommend these days?" This question is then sent from the terminal to the server.

[0974] 2. Server

[0975] The server processes the received question and sends it to the generative model in the appropriate format, which is the prompt for the generative model to generate an answer in the following format:

[0976] User Question: What recent smartphones do you recommend?

[0977] answer:

[0978] 3. Generative Model

[0979] The generative model analyzes the question and generates an answer using natural language processing technology. An example of an answer generated by the generative model is, "The latest recommended smartphones include the iPhone 14 and the Samsung Galaxy S22. They each feature high-performance cameras and fast processors." This answer is then sent back to the server.

[0980] 4. Presenting the Answer

[0981] The server sends the primary answer received from the generative model to the user's device, where the user can check the primary answer.

[0982] 5. Entering Supplementary and Corrective Information

[0983] Other users can enter additional or corrective information to the primary answer. For example, they can enter, "The iPhone 14 is good, but if you prioritize cost-effectiveness, I also recommend the Pixel 6." This additional data is sent to the server.

[0984] 6. Generate answers again

[0985] The server then sends supplementary or corrective information to the generative model. The generative model then generates a new answer based on the new information. An example of a new answer might be, "Recommended smartphones are the iPhone 14 and the Samsung Galaxy S22. If cost-effectiveness is important, consider the Pixel 6." This new answer is also sent back to the server and then to the user's device.

[0986] This system configuration allows users to obtain fast and accurate information, which can greatly support purchasing decisions. Specific hardware requirements include a server and user devices (smartphones, PCs, tablets, etc.), and software requirements include Flask (a Python web framework), OpenAI API, and a generative model (GPT-3).

[0987] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0988] Step 1:

[0989] The user inputs a question via a terminal.

[0990] Users input questions using an interface from devices such as smartphones or PCs. The input questions are in the form of "What smartphones do you recommend these days?" The user's question is generated as input data.

[0991] Step 2:

[0992] The terminal sends a question to the server.

[0993] The terminal sends the question entered by the user to the server. When the server receives the question, it proceeds to the next processing step. The input includes the user's question data, and the output is the question data transferred to the server.

[0994] Step 3:

[0995] The server sends the received question to the generative model in the appropriate format.

[0996] The server analyzes the received question and formats it in a format that can be processed by the generative model. Specifically, the question is converted into a prompt sentence. It is converted into the format "User question: What smartphone do you recommend these days?\nAnswer: ". The input is the user's question data, and the output is a generated prompt sentence.

[0997] Step 4:

[0998] A generative model generates a first-order answer to the question.

[0999] A generative model (e.g., GPT-3) takes a prompt as input, analyzes it, generates it, and produces a first-order answer. For example, the model might generate an answer like, "The latest recommended smartphones are the iPhone 14 and the Samsung Galaxy S22. They each feature a high-performance camera and a fast processor." The input contains a prompt, and the output is a first-order answer.

[1000] Step 5:

[1001] The server sends the primary response to the user terminal.

[1002] The server sends the primary answer received from the generative model to the user's device, allowing the user to check the primary answer on their own device. The primary answer from the generative model is the input, and the answer data sent to the user's device is the output.

[1003] Step 6:

[1004] Other users input supplements or corrections to the primary answer via their terminals.

[1005] Other users can add supplementary or correction information to the answers already provided. For example, "The iPhone 14 is good, but if cost performance is important, the Pixel 6 is also recommended." The input includes the initial answer and supplementary or correction information, and the output is the supplementary or correction data that is sent to the server.

[1006] Step 7:

[1007] The server receives the supplementary or corrective information and resubmits it to the generative model.

[1008] The server receives supplementary or correction information from other users and sends it back to the generative model. At this time, the supplementary or correction information is added to the prompt sentence. In this way, the server is ready to generate an answer again. The input is the supplementary or correction information, and the output is the data to be resubmitted to the generative model.

[1009] Step 8:

[1010] The generative model generates the answer again and sends the answer back to the server.

[1011] The generative model generates a new answer based on new information. For example, it generates a new answer in the form of "Recommended smartphones are the iPhone 14 and Samsung Galaxy S22. If cost-effectiveness is important, consider the Pixel 6." The input is a prompt sentence containing supplementary information and corrections, and the output is a new answer.

[1012] Step 9:

[1013] The server sends a second response to the user's terminal.

[1014] The server then sends the re-answer received from the generative model back to the user device, allowing the user to check the re-answer on their device. The input is the re-answer from the generative model, and the output is the re-answer data sent to the user device.

[1015] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1016] An embodiment for implementing the system of the present invention will be described below.

[1017] System Configuration

[1018] The present invention can be implemented in a network system including a user terminal, a server, a generative model, and an emotion engine. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative model and emotion engine. The generative model is a program that uses an artificial intelligence algorithm to generate answers to questions, and the emotion engine recognizes user emotions and adjusts the behavior of the generative model based on those emotions.

[1019] User question input

[1020] The user inputs a question using a terminal. For example, the user inputs and sends a question such as, "What recent books do you recommend?" The emotion engine analyzes the user's emotions along with the question, and sends the results to the server as additional information.

[1021] Submitting questions and generating answers

[1022] The server processes the received question along with the emotional information and sends it to the generative model in an appropriate format. The generative model takes the question and emotional information into account to generate an answer. For example, if the answer is "The book I recommend these days is 'Twilight,'" the answer is adjusted based on the emotional information.

[1023] Presenting the first answer

[1024] The server sends the primary answer received from the generative model to the user's device, where the user can check the primary answer.

[1025] Entering Supplementary and Corrective Information

[1026] Other users can add supplementary or correction information to the primary answer. For example, they can add supplementary information such as, "Twilight is good, but I also recommend the recently published Attack on Titan." In this case, the emotion engine also recognizes the emotion of the user who added the supplementary or correction information and sends that information to the server.

[1027] Generate answers again

[1028] The server sends supplementary or corrective information and the emotion information to the generative model. The generative model generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'"

[1029] Providing a second answer

[1030] The server sends the regenerated answer to the user's terminal, where the user can review the revised answer and other users can enter additional comments or corrections.

[1031] Third and final answers

[1032] By repeating the re-answering process multiple times as necessary, the generative model generates a final answer. For example, a final answer might be, "My recent recommended books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'" During this process, the emotion engine continuously analyzes the user's emotions and optimizes the answer.

[1033] In this way, the system according to the present invention can provide a highly accurate final answer that takes into account the user's emotional information.

[1034] The processing flow will be explained below.

[1035] Specific processing steps of the system

[1036] Step 1:

[1037] The user inputs a question from the terminal.

[1038] Specific operation: The user inputs a question using the device interface and presses the "Send" button. At this time, the emotion engine analyzes the user's tone of voice and facial expressions to extract emotional information. The device then transmits the input question and emotional information to the server.

[1039] Step 2:

[1040] The device sends the question and emotion information to the server.

[1041] Specific operation: The device sends the question entered by the user and the emotional information analyzed by the emotion engine to the server using a pre-configured protocol (e.g., HTTP request).

[1042] Step 3:

[1043] The server receives the question and emotion information and sends it to the generative model in an appropriate format.

[1044] Specific operation: The server receives the question and emotion information, converts it into a format that the generative model can understand (e.g., JSON or XML), and sends it to the generative model.

[1045] Step 4:

[1046] The generative model generates a first-order answer based on the question and emotion information and sends it back to the server.

[1047] How it works: The generative model analyzes the question using natural language processing techniques and generates a first-order answer. The tone and content of the answer are adjusted based on the emotional information. This first-order answer is then sent back to the server.

[1048] Step 5:

[1049] The server sends the primary response to the user's terminal.

[1050] Specific operation: The server sends the initial answer received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[1051] Step 6:

[1052] The user reviews the primary answer and adds any additions or corrections.

[1053] Specific operation: Other users read the primary response through their devices and enter corrections or supplementary information. The emotion engine recognizes the user's emotions while they are entering and sending the response, adds that information, and sends it to the server.

[1054] Step 7:

[1055] The terminal transmits supplementary or corrective information and emotion information to the server.

[1056] Specific operation: The device sends supplementary or corrective information, as well as the emotion information analyzed by the emotion engine, to the server via a pre-configured protocol (e.g., HTTP request).

[1057] Step 8:

[1058] The server retransmits the supplementary or corrective information and the emotion information to the generative model.

[1059] Specific behavior: The server receives the supplementary or corrective information, as well as the emotion information, and sends it back to the generative model in an appropriate format (e.g., reformatted JSON data).

[1060] Step 9:

[1061] The generative model generates a re-answer and sends it back to the server.

[1062] What it does: The generative model generates a new answer based on the new information and emotional information. The emotional information adjusts the tone and content of the new answer. The new answer is then sent back to the server.

[1063] Step 10:

[1064] The server sends a second response to the user's terminal.

[1065] Specific operation: The server sends the response received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[1066] Step 11:

[1067] Further additions or corrections (if necessary).

[1068] Specific operation: Another user checks the answer again, enters additional or corrective information as necessary, and presses the "Send" button. During input and submission, the emotion engine recognizes the user's emotions, adds that information, and sends it to the server.

[1069] Step 12:

[1070] Repeat steps 6-10 above as needed.

[1071] What it does: The process continues, supplementing or correcting as needed, until a final answer is produced. Emotional information is continually considered throughout the process.

[1072] Through these steps, a highly accurate final answer that reflects the user's emotional information is obtained.

[1073] Example 2

[1074] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1075] Conventional question-answering systems have had the problem of difficulty in providing appropriate answers based on the user's emotions. In particular, they have had problems with generating answers that ignore the user's emotions, resulting in poor answer quality and difficulty in achieving user satisfaction. Furthermore, when additional clarifications or corrections are added, there is a lack of a mechanism for generating appropriate re-answers that take these into account. To solve these problems, highly accurate answer generation that reflects the user's emotional information is required.

[1076] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a question via a terminal, a means for the server to receive the question and the user's emotional information and transmit them to the generative model in an appropriate format, and a means for the generative model to generate a primary answer based on the question and the emotional information and return the primary answer to the server. This enables the generation of a high-quality answer that takes into account the user's emotional information.

[1077] "User" refers to an entity that utilizes the system to enter questions and receive answers.

[1078] "Device" refers to the device used by a user to enter questions and check answers, such as a smartphone, computer, or tablet.

[1079] "Server" refers to the computer system that receives questions from users and sends information to the generative model and emotion engine.

[1080] "Question" refers to a question or request entered by a user via a terminal.

[1081] A "generative model" refers to a program that uses an artificial intelligence algorithm to generate answers based on questions and emotional information.

[1082] "Emotion engine" refers to software that identifies and analyzes a user's emotions.

[1083] "Emotion information" refers to data relating to the user's emotions analyzed by the emotion engine.

[1084] "Primary answer" refers to the first answer generated by the generative model.

[1085] "Supplement or correction" refers to additional information or corrections entered by other users in response to the primary answer.

[1086] A "re-answer" refers to a second or subsequent answer generated by the generative model based on supplementary or corrective information and emotional information.

[1087] "Final Answer" refers to the final answer generated through the repeated answering process.

[1088] The system of the present invention generates answers to questions from users via a network system and provides highly accurate answers that take into account the user's emotional information using an emotion engine. The system of the present invention mainly includes the following components: a user terminal, a server, a generative model, and an emotion engine.

[1089] System Configuration

[1090] User terminal

[1091] A user terminal is a device that allows a user to input and send questions, such as a smartphone, PC, or tablet. The user terminal includes an input interface and a display interface, sends the questions input by the user to the server, and displays the answers from the server.

[1092] server

[1093] The server is responsible for receiving questions from users and sending them to the generative model and emotion engine. The server sends the emotion information analyzed by the emotion engine, along with the question and any supplementary or correction information, to the generative model in an appropriate format.

[1094] Generative Model

[1095] A generative model is a program that uses artificial intelligence algorithms to generate answers based on questions and sentiment information. For example, a natural language processing model such as GPT-3 is used. A generative model generates a first answer, a follow-up answer, and a final answer.

[1096] Emotion Engine

[1097] The emotion engine is software that analyzes the user's emotions and generates emotional information to accompany the user's questions and supplementary / corrective information, allowing the generative model to generate answers based on emotions.

[1098] Specific examples

[1099] The user uses a terminal to input and send a question such as "What recent books do you recommend?" The server receives this question and analyzes the user's emotions using an emotion engine. If the emotion engine determines that the user is curious, the server sends this emotional information along with the question to the generative model.

[1100] The generative model generates an answer based on the question and emotional information. For example, a primary answer such as "The book I recommend these days is 'Twilight.' Another popular book is 'Attack on Titan.'" is generated and sent to the user's device via the server.

[1101] When another user adds additional information, such as "Twilight is good, but I also recommend the recently published Our Little Sister," the emotion engine analyzes the user's emotions. The server sends this new emotional information to the generative model, which then generates a new answer.

[1102] As a follow-up answer, the following answer is generated: "My recent recommended books are 'Twilight' and 'Our Little Sister.' In addition, 'Demon Slayer: Kimetsu no Yaiba' is also highly rated," and is sent to the user's device via the server.

[1103] Prompt Sentence Examples

[1104] "The user asked for recent book recommendations. The sentiment engine determined that the user is curious. Please take this into account when generating an answer."

[1105] In this way, the system of the present invention can provide highly accurate answers that take into account the user's emotional information, thereby improving user satisfaction.

[1106] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1107] The flow of this system's program processing

[1108] Specific explanation of processing steps

[1109] Step 1:

[1110] The user inputs and sends a question via a terminal.

[1111] Input: The user types into the terminal, "What recent books have you recommended?"

[1112] Data processing: When the send button on the terminal is pressed, the question text is sent to the server using an HTTP POST request.

[1113] Output: The question is sent to the server.

[1114] Step 2:

[1115] The server receives the question and passes it to the emotion engine for analysis of emotion information.

[1116] Input: The user's question sent to the server: "What books have you recommended recently?"

[1117] Data processing: The emotion engine performs sentiment analysis on the text to determine the user's emotion (e.g., curious).

[1118] Output: Analysis results including emotional information.

[1119] Step 3:

[1120] The server sends the question and emotion information to the generative model.

[1121] Input: Parsed emotion information (e.g., curious) and user question.

[1122] Data processing: The question and emotion information are sent to the generative model server in a data format such as JSON.

[1123] Output: The generative model receives the question and sentiment information.

[1124] Step 4:

[1125] A generative model generates a first-order answer based on the question and sentiment information.

[1126] Input: The question received by the generative model (e.g., "What recent books do you recommend?") and sentiment information.

[1127] Data processing: A generative model (e.g., GPT-3) analyzes the input prompt and generates a text answer.

[1128] Output: Text of the primary answer (e.g., "The book I recommend these days is 'Twilight.' Another popular book is 'Attack on Titan.'").

[1129] Step 5:

[1130] The server sends the primary response to the user's terminal.

[1131] Input: The text of the primary answer returned by the generative model.

[1132] Data processing: The server sends the primary answer to the user terminal as an HTTP response.

[1133] Output: The primary answer is displayed on the user's terminal.

[1134] Step 6:

[1135] Other users input supplementary or corrective information to the primary answer via the terminal.

[1136] Input: Another user types into their device, "Twilight is good, but I also recommend the recently published Our Little Sister."

[1137] Data processing: The entered supplementary or corrective information is sent to the server using an HTTP POST request.

[1138] Output: Any supplementary or corrective information is sent to the server.

[1139] Step 7:

[1140] The server retransmits the supplementary or corrective information and the emotion information to the generative model.

[1141] Input: Supplementary or correction information and sentiment information received by the server.

[1142] Data processing: Supplementary or corrective information and emotion information are resubmitted to the generative model server in a data format such as JSON.

[1143] Output: The generative model receives new information needed to generate a new answer.

[1144] Step 8:

[1145] The generative model generates a re-answer.

[1146] Input: New supplementary or correction information received by the generative model (e.g., "Twilight is good, but I also recommend the recently published Our Little Sister") and sentiment information.

[1147] Data processing: The generative model generates a new answer based on new information (e.g., "Recommended books these days are 'Twilight' and 'Our Little Sister.' Other popular books include 'Demon Slayer: Kimetsu no Yaiba.'").

[1148] Output: The text of the re-answer.

[1149] Step 9:

[1150] The server sends a second response to the user's terminal.

[1151] Input: The text of the re-answer returned by the generative model.

[1152] Data processing: The server sends a second response to the user terminal as an HTTP response.

[1153] Output: The answer is displayed on the user's terminal.

[1154] Thus, the system takes a series of steps to take affective information into account and provide highly accurate answers to the user's questions.

[1155] (Application example 2)

[1156] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1157] Conventional content distribution services exist that generate appropriate answers to user questions. However, because they do not take the user's emotions into account, the answers often do not match the user's wishes or circumstances. Emotional information is particularly important in entertainment, and ignoring it reduces satisfaction. Furthermore, because feedback from other users cannot be utilized, the answers lack accuracy and adaptability. Therefore, a system that analyzes the user's emotional information and reflects it in answer generation is needed.

[1158] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1159] In this invention, the server includes: a means for a user to input a question via a terminal; a means for the server to receive the question and transmit it to the generative model in an appropriate format; a means for the generative model to generate a primary answer to the question and return the primary answer to the server; a means for the server to transmit the primary answer to the user's terminal; a means for another user to input a supplement or correction to the primary answer via the terminal; a means for the server to receive the supplement or correction information and retransmit it to the generative model; a means for the generative model to generate a new answer and return the new answer to the server; a means for the server to transmit the new answer to the user's terminal; a means for analyzing the user's emotions using an emotion engine and adjusting the behavior of the generative model based on the analysis; and a means for updating the recommendation content based on feedback from other users. This makes it possible to provide more appropriate and satisfying answers that reflect the user's emotional information.

[1160] "Means for users to input questions via a terminal" refers to a mechanism that provides an interface for users to input and send questions using devices such as smartphones, tablets, and PCs.

[1161] "Means for the server to receive and send in an appropriate format to the generative model" refers to a mechanism by which the server receives questions entered by users, converts them into a format that can be processed by the generative AI model, and sends them.

[1162] "Means for the generative model to generate a first answer to a question and return the first answer to the server" refers to a mechanism by which the generative AI model generates an initial answer to a user's question and sends that answer to the server.

[1163] "Means by which the server sends the primary answer to the user's device" refers to a mechanism by which the server receives the primary answer obtained from the generative AI model and forwards it to the user's device.

[1164] "Means for other users to input supplementary information or corrections to the primary response via a terminal" refers to a system that provides an interface for other users to input additional information or corrections to the primary response using a smartphone, tablet, PC, etc.

[1165] "Means for the server to receive supplementary or corrective information and retransmit it to the generative model" refers to a mechanism by which, after additional information or corrections are entered, the server receives that information and retransmits it to the generative AI model.

[1166] "Means for the generative model to generate a new answer and return the new answer to the server" refers to a mechanism by which the generative AI model generates a new answer based on supplements or corrections and sends that answer to the server.

[1167] "Means for the server to send the re-answer to the user's device" refers to a mechanism by which the server receives the re-answer obtained from the generative AI model and forwards it to the user's device.

[1168] "Means for analyzing user emotions using an emotion engine and adjusting the behavior of the generative model based on that" refers to a mechanism that uses emotion analysis technology to detect user emotions and optimizes the content and format of the generative AI model's responses based on the results.

[1169] "Means for updating recommendations based on feedback from other users" refers to a mechanism for improving the recommendations of the generative AI model by reflecting opinions and ratings provided by other users.

[1170] Specific embodiments for carrying out the present invention will be described below.

[1171] System Configuration

[1172] This invention is embodied in a network system including a user terminal, a server, a generative AI model, and an emotion engine. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative AI model and the emotion engine. The generative AI model is a program that uses an artificial intelligence algorithm to generate answers to questions, and the emotion engine recognizes the user's emotions and adjusts the behavior of the generative AI model based on those emotions.

[1173] User question input

[1174] A user inputs a question using a device such as a smartphone or tablet. For example, a user might input and send a question such as, "I've been looking for a heartwarming movie lately. Do you have any recommendations?" The emotion engine analyzes the user's emotions along with the question and sends the results to the server as additional information.

[1175] Submitting questions and generating answers

[1176] The server processes the received question along with the emotional information and sends it to the generative AI model in an appropriate format. The generative AI model takes the question and emotional information into consideration to generate an answer. For example, if the answer generated is "A recent heartwarming movie I recommend is 'Green Book,'" the content and format of the answer are adjusted based on the emotional information.

[1177] Presenting the first answer

[1178] The server sends the initial answer received from the generative AI model to the user's device, where the user can check the initial answer.

[1179] Entering Supplementary and Corrective Information

[1180] Other users can add supplementary or correction information to the primary answer. For example, they can add supplementary information such as, "Green Book is good, but I also recommend the recently released Shawshank Redemption." In this case, the emotion engine also recognizes the emotion of the user who added the supplementary or correction information and sends that information to the server.

[1181] Generate answers again

[1182] The server sends supplementary or corrective information and the emotional information to the generative AI model. The generative AI model generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended recent heartwarming movies are 'Green Book' and 'The Shawshank Redemption.' Other popular movies include 'Les Miserables.'"

[1183] Providing a second answer

[1184] The server sends the regenerated answer to the user's terminal, where the user can review the revised answer and other users can enter additional comments or corrections.

[1185] Generate the final answer

[1186] The generative AI model generates a final answer by repeating the process of re-answering multiple times as necessary. During this process, the emotion engine continuously analyzes the user's emotions and optimizes the answer. For example, the final answer might be, "Recommended recent heartwarming movies are 'Green Book,' 'The Shawshank Redemption,' 'Les Miserables,' and 'Twilight.'"

[1187] In this way, the system according to the present invention can provide a highly accurate final answer that takes into account the user's emotional information.

[1188] Hardware and software used

[1189] Hardware: Smartphones, tablets, computers

[1190] Software: The program is implemented in Python and utilizes APIs for emotion engines and generative AI models.

[1191] Emotion Engine API (e.g. https: / / emotion-api.example.com / analyze)

[1192] Generate AI Model API (e.g. https: / / ai-model-api.example.com / generate)

[1193] Examples of prompt statements

[1194] User Question: I've been looking for a heartwarming movie lately. Any recommendations?

[1195] Emotional information: Relaxing, Heartwarming, Comforting

[1196] Output of generative AI model: A recent recommended heartwarming movie is "Green Book," but other good choices are "The Shawshank Redemption" and "Les Miserables."

[1197] According to this invention, it is possible to improve the accuracy of answers by incorporating information about the user's emotions, and to recommend content with a higher degree of satisfaction.

[1198] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1199] Step 1:

[1200] The user inputs a question via the terminal. The user inputs a question into a smartphone or tablet and sends it. An example of input is, "I've been looking for a heartwarming movie lately. Do you have any recommendations?"

[1201] Step 2:

[1202] The terminal sends a question and the server receives it. The terminal sends the question entered by the user to the server. The server receives this question and proceeds to the next step.

[1203] Step 3:

[1204] The server sends the received question to the emotion engine to obtain emotion information. The server sends the question to the emotion engine (e.g. https: / / emotion-api.example.com / analyze). The input is the question text, and the output is emotion information. For example, the emotion information detected is "relaxing, heartwarming, comforting."

[1205] Step 4:

[1206] The server sends the question and emotion information to the generative AI model. After receiving the emotion information, the server sends it along with the question to the generative AI model (e.g., https: / / ai-model-api.example.com / generate). The input is the question and emotion information, and the generative AI model generates a first answer based on this.

[1207] Step 5:

[1208] The generative AI model generates a first answer and sends it back to the server. The generative AI model generates an answer based on the input question and emotional information. For example, it generates an answer such as "A recent recommended heartwarming movie is 'Green Book'." This answer is sent to the server.

[1209] Step 6:

[1210] The server sends the primary answer to the user's device. The server sends the primary answer received from the generative AI model to the user's device. The user can check the primary answer through their device.

[1211] Step 7:

[1212] Other users can input supplementary information or corrections to the primary answer via the terminal. Other users can input additional information or corrections to the primary answer. For example, they can add supplementary information such as, "'Green Book' is good, but I also recommend the recently released 'Shawshank Redemption'."

[1213] Step 8:

[1214] The terminal transmits supplementary or corrective information to the server, which receives it. The terminal transmits supplementary or corrective information from other users to the server, which receives it.

[1215] Step 9:

[1216] The server retransmits the supplementary or correction information and emotional information to the generative AI model. The server retransmits the received supplementary or correction information and the emotional information to the generative AI model. The input is the supplementary or correction information and emotional information.

[1217] Step 10:

[1218] The generative AI model generates a re-answer based on the new information and sends it back to the server. For example, it might generate a re-answer such as, "Recommended recent heartwarming movies are 'Green Book' and 'The Shawshank Redemption.' Other popular movies include 'Les Miserables.'" and send it to the server.

[1219] Step 11:

[1220] The server sends the re-answer to the user's device. The server sends the re-answer received from the generative AI model to the user's device, where the user can check the re-answer.

[1221] Step 12:

[1222] If necessary, other users can enter additional supplementary or correction information, which is then received by the server and sent to the generative AI model. Finally, the generative AI model generates a final answer. For example, the final answer might be, "Recommended recent heartwarming movies are 'Green Book,' 'The Shawshank Redemption,' 'Les Miserables,' and 'Twilight.'"

[1223] Step 13:

[1224] The server sends the final answer to the user's device. The server sends the final answer received from the generative AI model to the user's device, where the user can check the final answer.

[1225] This series of processes enables highly accurate content recommendations that reflect the user's emotional information.

[1226] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1227] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1228] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1229] [Fourth embodiment]

[1230] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1231] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1232] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1233] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1234] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1235] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1236] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1237] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1238] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1239] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1240] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1241] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1242] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1243] An embodiment for implementing the system of the present invention will be described below.

[1244] System Configuration

[1245] The present invention can be implemented in a network system including a user terminal, a server, and a generative model. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative model. The generative model is a program that uses an artificial intelligence algorithm to generate answers to questions.

[1246] User question input

[1247] A user inputs a question using a terminal. For example, the user inputs and sends a question such as "What recent books do you recommend?" through the terminal interface. This question is sent from the terminal to the server.

[1248] Submitting questions and generating answers

[1249] The server processes the received question and sends it to the generative model in the appropriate format. The generative model analyzes the question and generates an answer using natural language processing techniques. For example, the generative model generates an answer such as "The recommended book these days is 'Twilight'" and sends it back to the server.

[1250] Presenting the first answer

[1251] The server sends the primary answer received from the generative model to the user's terminal, where the user can check the primary answer.

[1252] Entering Supplementary and Corrective Information

[1253] Other users can add supplementary or corrective information to the primary answer. For example, another user can add supplementary information such as "Twilight is good, but I also recommend the recently published Attack on Titan," and send it to the server via their device.

[1254] Generate answers again

[1255] The server then sends additional or corrective information to the generative model. The generative model then generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'" and send it back to the server.

[1256] Providing a second answer

[1257] The server then sends the regenerated answer to the user's terminal, where the user can confirm the revised answer. In addition, other users can enter additional comments or corrections.

[1258] Third and final answers

[1259] By repeating the re-answering process multiple times as necessary, the generative model generates a final answer, such as "My recent recommended books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'"

[1260] In this way, the system based on the present invention can provide highly accurate answers to user questions while reflecting supplements and corrections from other users.

[1261] The processing flow will be explained below.

[1262] Specific processing steps of the program

[1263] Step 1:

[1264] The user inputs a question from the terminal.

[1265] Specific operation: The user inputs a question using the terminal interface and presses the "Send" button. At this time, the terminal converts the input question into a digital format and sends it to the server.

[1266] Step 2:

[1267] The terminal sends a question to the server.

[1268] Specific operation: The terminal sends the question entered by the user to the server using a pre-configured protocol (e.g., HTTP request).

[1269] Step 3:

[1270] The server receives the question and sends it to the generative model in the appropriate format.

[1271] Specific operation: The server receives the question, converts it into a format that the generative model can understand (e.g., JSON or XML), and sends it to the generative model.

[1272] Step 4:

[1273] The generative model generates a first-order answer to the question and sends it back to the server.

[1274] How it works: The generative model uses natural language processing techniques to analyze the question and generate a first-order answer, which is then sent back to the server.

[1275] Step 5:

[1276] The server sends the primary response to the user's terminal.

[1277] Specific operation: The server sends the initial answer received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[1278] Step 6:

[1279] The user reviews the primary answer and adds any additions or corrections.

[1280] Specific operations: Other users read the primary response through their terminals, enter corrections or supplementary information, and press the "Submit" button once the supplements or corrections are complete.

[1281] Step 7:

[1282] The terminal sends the supplementary or corrective information to the server.

[1283] Specific operation: The device sends supplementary or corrective information to the server using a pre-configured protocol (e.g., HTTP request).

[1284] Step 8:

[1285] The server resubmits the supplementary or corrective information to the generative model.

[1286] Specific behavior: The server receives the supplementary or corrective information and sends it back to the generative model in an appropriate format (e.g., reformatted JSON data).

[1287] Step 9:

[1288] The generative model generates a re-answer and sends it back to the server.

[1289] What happens next: The generative model generates a new answer based on the new information, and this answer is sent back to the server.

[1290] Step 10:

[1291] The server sends a second response to the user's terminal.

[1292] Specific operation: The server sends the response received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[1293] Step 11:

[1294] Further additions or corrections (if necessary).

[1295] Specific actions: Another user will check the answer again, enter additional or corrective information if necessary, and press the "Submit" button.

[1296] Step 12:

[1297] Repeat steps 6-10 above as needed.

[1298] Specific Action: Repeat the process of supplementing or correcting as necessary and continue the process until a final answer is produced.

[1299] Through these steps, a highly accurate final answer is obtained.

[1300] Example 1

[1301] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1302] In conventional question-answering systems, answers to user questions are often fixed, and when other users add supplementary or correction information, the answers are not automatically updated. As a result, answers to questions are not always optimal, making it difficult for users to obtain satisfactory answers, especially when current information or individual knowledge is required.

[1303] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1304] In this invention, the server includes a means for a user to input a question via a device, a means for a computer system to receive the question and send it to a generation algorithm in an appropriate format, and a means for the generation algorithm to generate an initial answer to the question and return the initial answer to the computer system. This makes it possible to obtain a highly accurate answer to the question that takes into account supplementary and correction information as needed. Furthermore, by generating and presenting new answers multiple times as needed, it is possible to ultimately provide an answer that satisfies the user.

[1305] "Device" means an electronic device that allows a user to input or receive information, including a smartphone, computer, tablet, etc.

[1306] "Computer System" refers to a computer network system for receiving, processing, and transmitting data from a user to a generating algorithm.

[1307] "Generation algorithm" refers to a program or artificial intelligence technology that automatically generates answers to user questions.

[1308] "Entering a question" refers to the act of a user entering a question or request in text form through a device.

[1309] The "first answer" refers to the answer that the generation algorithm first generates, and refers to the information that is first presented in response to a question entered by a user.

[1310] "Entering supplementary or correction information" refers to the act of another user entering additional information or correction information in response to an initial response.

[1311] A "re-answer" refers to an answer that is regenerated by the generation algorithm based on supplementary or corrective information for the initial answer.

[1312] The "final answer" refers to a completed answer that is finally presented to the user after multiple answer generation processes.

[1313] MODE FOR CARRYING OUT THE INVENTION

[1314] System Overview

[1315] The system of the present invention is composed of a network system including a user terminal, a server, and a generative model. The user terminal refers to a device such as a smartphone, PC, or tablet, and provides an interface for inputting and receiving questions. The server is a computer system that receives questions from users and sends them to the generative algorithm. The generative algorithm is a program that automatically generates answers to questions, and uses natural language processing technology such as GPT-3.

[1316] Specific program description

[1317] Suppose a user inputs a question using a terminal. For example, "What recent books do you recommend?" This question is sent from the terminal to the server. The server receives this question and prepares it to be sent to the generation algorithm. Specifically, it converts the received text data into an appropriate format (for example, JSON format) and sends it to the generation algorithm using a REST API.

[1318] The generation algorithm analyzes the received question and generates an answer based on natural language processing technology. A generative AI model such as GPT-3 can be used to generate the answer. The generated answer is sent back to the server in the form of, for example, "The recommended book these days is 'Twilight.'"

[1319] The server receives the answer returned by the generation algorithm and sends it back to the user's device. The user can check the generated primary answer through the device interface. At this time, other users can also use their devices to enter supplementary or correction information for the primary answer. For example, supplementary information such as "Twilight is good, but I also recommend the recently published Attack on Titan."

[1320] When additional or corrective information is entered, the server receives this information again and sends it to the generation algorithm. The generation algorithm takes the additional information into account and generates a new answer. The regenerated answer will be in the form of, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'" The server sends this re-answer to the user's device so that the user can confirm it.

[1321] Specific examples

[1322] The system operates when the user enters the following prompt through the terminal:

[1323] "What recent books would you recommend?"

[1324] "Do you have any additional information about 'Twilight'?"

[1325] "Are there any other books you'd recommend?"

[1326] This system allows users to obtain highly accurate answers to their questions, reflecting supplementary and corrective information as it goes along, making it possible to provide satisfactory answers, especially in cases where current information or specialized knowledge is required.

[1327] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1328] Step 1:

[1329] The user inputs a question using the device interface. For example, the user inputs "What books do you recommend these days?" and presses the submit button. At this time, the input data is in text format, and the device captures the data.

[1330] Step 2:

[1331] The terminal sends a question from the user to the server. The data to be sent is in the form of an HTTP request and is transferred through a network path from the device to the server. The server then receives the question data.

[1332] Step 3:

[1333] The server processes the received question data into the appropriate format. Specifically, it converts the text data into JSON format and prepares to send a request to the generation algorithm's API. The input is text data, and the output is a JSON-formatted request.

[1334] Step 4:

[1335] The server sends the appropriately formatted question data to the generation algorithm, typically using a REST API, along with the API endpoint and any necessary authentication information.

[1336] Step 5:

[1337] The generation algorithm analyzes the received question data and generates an initial answer using natural language processing techniques. For example, a generative AI model such as GPT-3 can be used to generate an answer such as "The recommended book these days is 'Twilight.'" The input is the formatted question data, and the output is the generated answer text.

[1338] Step 6:

[1339] The generation algorithm returns the initial answer it generated to the server. The generated answer is returned to the server in JSON format, so the server receives the answer data. The input is the generated answer text, and the output is a JSON-formatted response.

[1340] Step 7:

[1341] The server sends the initial response data received from the generation algorithm to the user device. For example, it sends the data to the device as an HTTP response. At this time, the server sends the response data in an appropriate format to the user device. The input is the response data in JSON format, and the output is the response text sent to the user device.

[1342] Step 8:

[1343] The user checks their initial answer through the device interface and then inputs supplementary or correction information based on it. For example, they might type, "I like 'Twilight,' but I also recommend the recently published 'Attack on Titan,'" and press the submit button. The input is supplementary information in text format.

[1344] Step 9:

[1345] The terminal sends supplementary or correction information from the user to the server. This supplementary information is sent in the form of an HTTP request, and the server receives the data. The input is the supplementary information in text format, and the output is the supplementary information sent to the server.

[1346] Step 10:

[1347] The server then processes the received supplemental information into the appropriate format and sends it to the generation algorithm. For example, it converts text data into JSON format and prepares it for resubmission. The input is the supplemental information in text format, and the output is the request in JSON format.

[1348] Step 11:

[1349] The generation algorithm generates a new answer based on the supplementary information. For example, it generates a new answer such as "Recommended books these days are 'Twilight' and 'Attack on Titan'. 'Our Little Sister' is also popular." The input is the formatted supplementary information, and the output is the generated new answer text.

[1350] Step 12:

[1351] The generation algorithm sends the regenerated answer back to the server, which receives the regenerated answer. The input is the regenerated answer text, and the output is a JSON-formatted response.

[1352] Step 13:

[1353] The server sends the re-answer to the user's device. This data is sent as an HTTP response, and the user can confirm the re-answer. The input is the re-answer data in JSON format, and the output is the re-answer text sent to the user's device.

[1354] Step 14:

[1355] The server and the generation algorithm repeat this process as many times as necessary until a final answer is generated, such as "Recommended recent books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'" The input is the repeatedly updated supplementary information, and the output is the final answer text.

[1356] In this way, a system is realized that provides highly accurate answers to user questions that sequentially reflect supplementary and corrective information.

[1357] (Application example 1)

[1358] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1359] In conventional online shopping systems, users must spend a lot of time obtaining detailed information about products, making it difficult to efficiently select appropriate products. Furthermore, if the answers to users' questions are inaccurate, there is a risk that their purchasing decisions will be affected. To solve this problem, a system is needed that provides users with quick and accurate answers to their questions and supports their purchasing decisions.

[1360] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1361] In this invention, the server includes a means for a user to input a question via a terminal, a means for the server to receive the question and transmit it to a generative model in an appropriate format, and a means for the generative model to generate a primary answer to the question and return the primary answer to the server, thereby enabling the user to obtain information about products in a virtual store in real time and make accurate and prompt purchasing decisions.

[1362] "Means for users to input questions via a terminal" refers to a function that allows users to input questions through an interface using a device such as a smartphone, tablet, or PC.

[1363] "Means for the server to receive and send in a format appropriate for the generative model" refers to the function by which the server receives a question from a user, converts it into a format that can be processed by the generative model, and sends it.

[1364] "Means for the generative model to generate a primary answer to a question and return the primary answer to the server" is a function that uses a generative algorithm to create an answer to a user's question and return the answer to the server.

[1365] The "means for the server to transmit the primary answer to the user's terminal" is a function for transmitting the primary answer received from the generative model to the user's device.

[1366] "Means for other users to enter supplementary information or corrections to the primary response via their device" refers to the ability for other users to use their own device to enter additional information or corrections to the response already provided.

[1367] "Means for the server to receive supplementary or corrective information and retransmit it to the generative model" is a function that allows the server to receive supplementary or corrective information and retransmit it to the generative model.

[1368] "Means for the generative model to generate an answer again and return the new answer to the server" is a function that generates a new answer based on supplementary or correction information and returns the new answer to the server.

[1369] The "means for the server to send a re-answer to the user's terminal" is a function for sending the re-answer received from the generative model again to the user's device.

[1370] "Means for allowing users to ask questions about products via a terminal within a virtual store" is a function that allows users to input questions about products they want through an interface within the virtual store.

[1371] "Means for the generative model to generate product information and comparative information based on the question" refers to a function in which the generative model creates detailed information about the product and comparative information with other products based on the content of the user's question.

[1372] "Means for presenting generated product information to the user" is a function that displays information about products and comparison information generated by the generative model on the user's device.

[1373] To implement this invention, a network system including a user terminal, a server, and a generative model must be constructed. The user terminal can be a smartphone, tablet, PC, or other device. The server sends user questions to the generative model in an appropriate format, receives the answers generated by the generative model, and returns them to the user terminal. The generative model can be a program using an advanced natural language processing algorithm, such as OpenAI's GPT-3.

[1374] System Configuration

[1375] 1. User Device

[1376] The user inputs a question via the terminal. This question is input through the interface, for example, "What smartphones do you recommend these days?" This question is then sent from the terminal to the server.

[1377] 2. Server

[1378] The server processes the received question and sends it to the generative model in the appropriate format, which is the prompt for the generative model to generate an answer in the following format:

[1379] User Question: What recent smartphones do you recommend?

[1380] answer:

[1381] 3. Generative Model

[1382] The generative model analyzes the question and generates an answer using natural language processing technology. An example of an answer generated by the generative model is, "The latest recommended smartphones include the iPhone 14 and the Samsung Galaxy S22. They each feature high-performance cameras and fast processors." This answer is then sent back to the server.

[1383] 4. Presenting the Answer

[1384] The server sends the primary answer received from the generative model to the user's device, where the user can check the primary answer.

[1385] 5. Entering Supplementary and Corrective Information

[1386] Other users can enter additional or corrective information to the primary answer. For example, they can enter, "The iPhone 14 is good, but if you prioritize cost-effectiveness, I also recommend the Pixel 6." This additional data is sent to the server.

[1387] 6. Generate answers again

[1388] The server then sends supplementary or corrective information to the generative model. The generative model then generates a new answer based on the new information. An example of a new answer might be, "Recommended smartphones are the iPhone 14 and the Samsung Galaxy S22. If cost-effectiveness is important, consider the Pixel 6." This new answer is also sent back to the server and then to the user's device.

[1389] This system configuration allows users to obtain fast and accurate information, which can greatly support purchasing decisions. Specific hardware requirements include a server and user devices (smartphones, PCs, tablets, etc.), and software requirements include Flask (a Python web framework), OpenAI API, and a generative model (GPT-3).

[1390] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1391] Step 1:

[1392] The user inputs a question via a terminal.

[1393] Users input questions using an interface from devices such as smartphones or PCs. The input questions are in the form of "What smartphones do you recommend these days?" The user's question is generated as input data.

[1394] Step 2:

[1395] The terminal sends a question to the server.

[1396] The terminal sends the question entered by the user to the server. When the server receives the question, it proceeds to the next processing step. The input includes the user's question data, and the output is the question data transferred to the server.

[1397] Step 3:

[1398] The server sends the received question to the generative model in the appropriate format.

[1399] The server analyzes the received question and formats it in a format that can be processed by the generative model. Specifically, the question is converted into a prompt sentence. It is converted into the format "User question: What smartphone do you recommend these days?\nAnswer: ". The input is the user's question data, and the output is a generated prompt sentence.

[1400] Step 4:

[1401] A generative model generates a first-order answer to the question.

[1402] A generative model (e.g., GPT-3) takes a prompt as input, analyzes it, generates it, and produces a first-order answer. For example, the model might generate an answer like, "The latest recommended smartphones are the iPhone 14 and the Samsung Galaxy S22. They each feature a high-performance camera and a fast processor." The input contains a prompt, and the output is a first-order answer.

[1403] Step 5:

[1404] The server sends the primary response to the user terminal.

[1405] The server sends the primary answer received from the generative model to the user's device, allowing the user to check the primary answer on their own device. The primary answer from the generative model is the input, and the answer data sent to the user's device is the output.

[1406] Step 6:

[1407] Other users input supplements or corrections to the primary answer via their terminals.

[1408] Other users can add supplementary or correction information to the answers already provided. For example, "The iPhone 14 is good, but if cost performance is important, the Pixel 6 is also recommended." The input includes the initial answer and supplementary or correction information, and the output is the supplementary or correction data that is sent to the server.

[1409] Step 7:

[1410] The server receives the supplementary or corrective information and resubmits it to the generative model.

[1411] The server receives supplementary or correction information from other users and sends it back to the generative model. At this time, the supplementary or correction information is added to the prompt sentence. In this way, the server is ready to generate an answer again. The input is the supplementary or correction information, and the output is the data to be resubmitted to the generative model.

[1412] Step 8:

[1413] The generative model generates the answer again and sends the answer back to the server.

[1414] The generative model generates a new answer based on new information. For example, it generates a new answer in the form of "Recommended smartphones are the iPhone 14 and Samsung Galaxy S22. If cost-effectiveness is important, consider the Pixel 6." The input is a prompt sentence containing supplementary information and corrections, and the output is a new answer.

[1415] Step 9:

[1416] The server sends a second response to the user's terminal.

[1417] The server then sends the re-answer received from the generative model back to the user device, allowing the user to check the re-answer on their device. The input is the re-answer from the generative model, and the output is the re-answer data sent to the user device.

[1418] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1419] An embodiment for implementing the system of the present invention will be described below.

[1420] System Configuration

[1421] The present invention can be implemented in a network system including a user terminal, a server, a generative model, and an emotion engine. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative model and emotion engine. The generative model is a program that uses an artificial intelligence algorithm to generate answers to questions, and the emotion engine recognizes user emotions and adjusts the behavior of the generative model based on those emotions.

[1422] User question input

[1423] The user inputs a question using a terminal. For example, the user inputs and sends a question such as, "What recent books do you recommend?" The emotion engine analyzes the user's emotions along with the question, and sends the results to the server as additional information.

[1424] Submitting questions and generating answers

[1425] The server processes the received question along with the emotional information and sends it to the generative model in an appropriate format. The generative model takes the question and emotional information into account to generate an answer. For example, if the answer is "The book I recommend these days is 'Twilight,'" the answer is adjusted based on the emotional information.

[1426] Presenting the first answer

[1427] The server sends the primary answer received from the generative model to the user's device, where the user can check the primary answer.

[1428] Entering Supplementary and Corrective Information

[1429] Other users can add supplementary or correction information to the primary answer. For example, they can add supplementary information such as, "Twilight is good, but I also recommend the recently published Attack on Titan." In this case, the emotion engine also recognizes the emotion of the user who added the supplementary or correction information and sends that information to the server.

[1430] Generate answers again

[1431] The server sends supplementary or corrective information and the emotion information to the generative model. The generative model generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended books these days are 'Twilight' and 'Attack on Titan.' Other popular books include 'Our Little Sister.'"

[1432] Providing a second answer

[1433] The server sends the regenerated answer to the user's terminal, where the user can review the revised answer and other users can enter additional comments or corrections.

[1434] Third and final answers

[1435] By repeating the re-answering process multiple times as necessary, the generative model generates a final answer. For example, a final answer might be, "My recent recommended books are 'Twilight,' 'Attack on Titan,' 'Our Little Sister,' and 'Demon Slayer: Kimetsu no Yaiba.'" During this process, the emotion engine continuously analyzes the user's emotions and optimizes the answer.

[1436] In this way, the system according to the present invention can provide a highly accurate final answer that takes into account the user's emotional information.

[1437] The processing flow will be explained below.

[1438] Specific processing steps of the system

[1439] Step 1:

[1440] The user inputs a question from the terminal.

[1441] Specific operation: The user inputs a question using the device interface and presses the "Send" button. At this time, the emotion engine analyzes the user's tone of voice and facial expressions to extract emotional information. The device then transmits the input question and emotional information to the server.

[1442] Step 2:

[1443] The device sends the question and emotion information to the server.

[1444] Specific operation: The device sends the question entered by the user and the emotional information analyzed by the emotion engine to the server using a pre-configured protocol (e.g., HTTP request).

[1445] Step 3:

[1446] The server receives the question and emotion information and sends it to the generative model in an appropriate format.

[1447] Specific operation: The server receives the question and emotion information, converts it into a format that the generative model can understand (e.g., JSON or XML), and sends it to the generative model.

[1448] Step 4:

[1449] The generative model generates a first-order answer based on the question and emotion information and sends it back to the server.

[1450] How it works: The generative model analyzes the question using natural language processing techniques and generates a first-order answer. The tone and content of the answer are adjusted based on the emotional information. This first-order answer is then sent back to the server.

[1451] Step 5:

[1452] The server sends the primary response to the user's terminal.

[1453] Specific operation: The server sends the initial answer received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[1454] Step 6:

[1455] The user reviews the primary answer and adds any additions or corrections.

[1456] Specific operation: Other users read the primary response through their devices and enter corrections or supplementary information. The emotion engine recognizes the user's emotions while they are entering and sending the response, adds that information, and sends it to the server.

[1457] Step 7:

[1458] The terminal transmits supplementary or corrective information and emotion information to the server.

[1459] Specific operation: The device sends supplementary or corrective information, as well as the emotion information analyzed by the emotion engine, to the server via a pre-configured protocol (e.g., HTTP request).

[1460] Step 8:

[1461] The server retransmits the supplementary or corrective information and the emotion information to the generative model.

[1462] Specific behavior: The server receives the supplementary or corrective information, as well as the emotion information, and sends it back to the generative model in an appropriate format (e.g., reformatted JSON data).

[1463] Step 9:

[1464] The generative model generates a re-answer and sends it back to the server.

[1465] What it does: The generative model generates a new answer based on the new information and emotional information. The emotional information adjusts the tone and content of the new answer. The new answer is then sent back to the server.

[1466] Step 10:

[1467] The server sends a second response to the user's terminal.

[1468] Specific operation: The server sends the response received from the generative model to the user's device using the appropriate protocol (e.g., HTTP response).

[1469] Step 11:

[1470] Further additions or corrections (if necessary).

[1471] Specific operation: Another user checks the answer again, enters additional or corrective information as necessary, and presses the "Send" button. During input and submission, the emotion engine recognizes the user's emotions, adds that information, and sends it to the server.

[1472] Step 12:

[1473] Repeat steps 6-10 above as needed.

[1474] What it does: The process continues, supplementing or correcting as needed, until a final answer is produced. Emotional information is continually considered throughout the process.

[1475] Through these steps, a highly accurate final answer that reflects the user's emotional information is obtained.

[1476] Example 2

[1477] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1478] Conventional question-answering systems have had the problem of difficulty in providing appropriate answers based on the user's emotions. In particular, they have had problems with generating answers that ignore the user's emotions, resulting in poor answer quality and difficulty in achieving user satisfaction. Furthermore, when additional clarifications or corrections are added, there is a lack of a mechanism for generating appropriate re-answers that take these into account. To solve these problems, highly accurate answer generation that reflects the user's emotional information is required.

[1479] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a question via a terminal, a means for the server to receive the question and the user's emotional information and transmit them to the generative model in an appropriate format, and a means for the generative model to generate a primary answer based on the question and the emotional information and return the primary answer to the server. This enables the generation of a high-quality answer that takes into account the user's emotional information.

[1480] "User" refers to an entity that utilizes the system to enter questions and receive answers.

[1481] "Device" refers to the device used by a user to enter questions and check answers, such as a smartphone, computer, or tablet.

[1482] "Server" refers to the computer system that receives questions from users and sends information to the generative model and emotion engine.

[1483] "Question" refers to a question or request entered by a user via a terminal.

[1484] A "generative model" refers to a program that uses an artificial intelligence algorithm to generate answers based on questions and emotional information.

[1485] "Emotion engine" refers to software that identifies and analyzes a user's emotions.

[1486] "Emotion information" refers to data relating to the user's emotions analyzed by the emotion engine.

[1487] "Primary answer" refers to the first answer generated by the generative model.

[1488] "Supplement or correction" refers to additional information or corrections entered by other users in response to the primary answer.

[1489] A "re-answer" refers to a second or subsequent answer generated by the generative model based on supplementary or corrective information and emotional information.

[1490] "Final Answer" refers to the final answer generated through the repeated answering process.

[1491] The system of the present invention generates answers to questions from users via a network system and provides highly accurate answers that take into account the user's emotional information using an emotion engine. The system of the present invention mainly includes the following components: a user terminal, a server, a generative model, and an emotion engine.

[1492] System Configuration

[1493] User terminal

[1494] A user terminal is a device that allows a user to input and send questions, such as a smartphone, PC, or tablet. The user terminal includes an input interface and a display interface, sends the questions input by the user to the server, and displays the answers from the server.

[1495] server

[1496] The server is responsible for receiving questions from users and sending them to the generative model and emotion engine. The server sends the emotion information analyzed by the emotion engine, along with the question and any supplementary or correction information, to the generative model in an appropriate format.

[1497] Generative Model

[1498] A generative model is a program that uses artificial intelligence algorithms to generate answers based on questions and sentiment information. For example, a natural language processing model such as GPT-3 is used. A generative model generates a first answer, a follow-up answer, and a final answer.

[1499] Emotion Engine

[1500] The emotion engine is software that analyzes the user's emotions and generates emotional information to accompany the user's questions and supplementary / corrective information, allowing the generative model to generate answers based on emotions.

[1501] Specific examples

[1502] The user uses a terminal to input and send a question such as "What recent books do you recommend?" The server receives this question and analyzes the user's emotions using an emotion engine. If the emotion engine determines that the user is curious, the server sends this emotional information along with the question to the generative model.

[1503] The generative model generates an answer based on the question and emotional information. For example, a primary answer such as "The book I recommend these days is 'Twilight.' Another popular book is 'Attack on Titan.'" is generated and sent to the user's device via the server.

[1504] When another user adds additional information, such as "Twilight is good, but I also recommend the recently published Our Little Sister," the emotion engine analyzes the user's emotions. The server sends this new emotional information to the generative model, which then generates a new answer.

[1505] As a follow-up answer, the following answer is generated: "My recent recommended books are 'Twilight' and 'Our Little Sister.' In addition, 'Demon Slayer: Kimetsu no Yaiba' is also highly rated," and is sent to the user's device via the server.

[1506] Prompt Sentence Examples

[1507] "The user asked for recent book recommendations. The sentiment engine determined that the user is curious. Please take this into account when generating an answer."

[1508] In this way, the system of the present invention can provide highly accurate answers that take into account the user's emotional information, thereby improving user satisfaction.

[1509] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1510] The flow of this system's program processing

[1511] Specific explanation of processing steps

[1512] Step 1:

[1513] The user inputs and sends a question via a terminal.

[1514] Input: The user types into the terminal, "What recent books have you recommended?"

[1515] Data processing: When the send button on the terminal is pressed, the question text is sent to the server using an HTTP POST request.

[1516] Output: The question is sent to the server.

[1517] Step 2:

[1518] The server receives the question and passes it to the emotion engine for analysis of emotion information.

[1519] Input: The user's question sent to the server: "What books have you recommended recently?"

[1520] Data processing: The emotion engine performs sentiment analysis on the text to determine the user's emotion (e.g., curious).

[1521] Output: Analysis results including emotional information.

[1522] Step 3:

[1523] The server sends the question and emotion information to the generative model.

[1524] Input: Parsed emotion information (e.g., curious) and user question.

[1525] Data processing: The question and emotion information are sent to the generative model server in a data format such as JSON.

[1526] Output: The generative model receives the question and sentiment information.

[1527] Step 4:

[1528] A generative model generates a first-order answer based on the question and sentiment information.

[1529] Input: The question received by the generative model (e.g., "What recent books do you recommend?") and sentiment information.

[1530] Data processing: A generative model (e.g., GPT-3) analyzes the input prompt and generates a text answer.

[1531] Output: Text of the primary answer (e.g., "The book I recommend these days is 'Twilight.' Another popular book is 'Attack on Titan.'").

[1532] Step 5:

[1533] The server sends the primary response to the user's terminal.

[1534] Input: The text of the primary answer returned by the generative model.

[1535] Data processing: The server sends the primary answer to the user terminal as an HTTP response.

[1536] Output: The primary answer is displayed on the user's terminal.

[1537] Step 6:

[1538] Other users input supplementary or corrective information to the primary answer via the terminal.

[1539] Input: Another user types into their device, "Twilight is good, but I also recommend the recently published Our Little Sister."

[1540] Data processing: The entered supplementary or corrective information is sent to the server using an HTTP POST request.

[1541] Output: Any supplementary or corrective information is sent to the server.

[1542] Step 7:

[1543] The server retransmits the supplementary or corrective information and the emotion information to the generative model.

[1544] Input: Supplementary or correction information and sentiment information received by the server.

[1545] Data processing: Supplementary or corrective information and emotion information are resubmitted to the generative model server in a data format such as JSON.

[1546] Output: The generative model receives new information needed to generate a new answer.

[1547] Step 8:

[1548] The generative model generates a re-answer.

[1549] Input: New supplementary or correction information received by the generative model (e.g., "Twilight is good, but I also recommend the recently published Our Little Sister") and sentiment information.

[1550] Data processing: The generative model generates a new answer based on new information (e.g., "Recommended books these days are 'Twilight' and 'Our Little Sister.' Other popular books include 'Demon Slayer: Kimetsu no Yaiba.'").

[1551] Output: The text of the re-answer.

[1552] Step 9:

[1553] The server sends a second response to the user's terminal.

[1554] Input: The text of the re-answer returned by the generative model.

[1555] Data processing: The server sends a second response to the user terminal as an HTTP response.

[1556] Output: The answer is displayed on the user's terminal.

[1557] Thus, the system takes a series of steps to take affective information into account and provide highly accurate answers to the user's questions.

[1558] (Application example 2)

[1559] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1560] Conventional content distribution services exist that generate appropriate answers to user questions. However, because they do not take the user's emotions into account, the answers often do not match the user's wishes or circumstances. Emotional information is particularly important in entertainment, and ignoring it reduces satisfaction. Furthermore, because feedback from other users cannot be utilized, the answers lack accuracy and adaptability. Therefore, a system that analyzes the user's emotional information and reflects it in answer generation is needed.

[1561] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1562] In this invention, the server includes: a means for a user to input a question via a terminal; a means for the server to receive the question and transmit it to the generative model in an appropriate format; a means for the generative model to generate a primary answer to the question and return the primary answer to the server; a means for the server to transmit the primary answer to the user's terminal; a means for another user to input a supplement or correction to the primary answer via the terminal; a means for the server to receive the supplement or correction information and retransmit it to the generative model; a means for the generative model to generate a new answer and return the new answer to the server; a means for the server to transmit the new answer to the user's terminal; a means for analyzing the user's emotions using an emotion engine and adjusting the behavior of the generative model based on the analysis; and a means for updating the recommendation content based on feedback from other users. This makes it possible to provide more appropriate and satisfying answers that reflect the user's emotional information.

[1563] "Means for users to input questions via a terminal" refers to a mechanism that provides an interface for users to input and send questions using devices such as smartphones, tablets, and PCs.

[1564] "Means for the server to receive and send in an appropriate format to the generative model" refers to a mechanism by which the server receives questions entered by users, converts them into a format that can be processed by the generative AI model, and sends them.

[1565] "Means for the generative model to generate a first answer to a question and return the first answer to the server" refers to a mechanism by which the generative AI model generates an initial answer to a user's question and sends that answer to the server.

[1566] "Means by which the server sends the primary answer to the user's device" refers to a mechanism by which the server receives the primary answer obtained from the generative AI model and forwards it to the user's device.

[1567] "Means for other users to input supplementary information or corrections to the primary response via a terminal" refers to a system that provides an interface for other users to input additional information or corrections to the primary response using a smartphone, tablet, PC, etc.

[1568] "Means for the server to receive supplementary or corrective information and retransmit it to the generative model" refers to a mechanism by which, after additional information or corrections are entered, the server receives that information and retransmits it to the generative AI model.

[1569] "Means for the generative model to generate a new answer and return the new answer to the server" refers to a mechanism by which the generative AI model generates a new answer based on supplements or corrections and sends that answer to the server.

[1570] "Means for the server to send the re-answer to the user's device" refers to a mechanism by which the server receives the re-answer obtained from the generative AI model and forwards it to the user's device.

[1571] "Means for analyzing user emotions using an emotion engine and adjusting the behavior of the generative model based on that" refers to a mechanism that uses emotion analysis technology to detect user emotions and optimizes the content and format of the generative AI model's responses based on the results.

[1572] "Means for updating recommendations based on feedback from other users" refers to a mechanism for improving the recommendations of the generative AI model by reflecting opinions and ratings provided by other users.

[1573] Specific embodiments for carrying out the present invention will be described below.

[1574] System Configuration

[1575] This invention is embodied in a network system including a user terminal, a server, a generative AI model, and an emotion engine. The user terminal is a device for inputting and sending questions, such as a smartphone, PC, or tablet. The server is a computer system that receives questions from users and sends them to the generative AI model and the emotion engine. The generative AI model is a program that uses an artificial intelligence algorithm to generate answers to questions, and the emotion engine recognizes the user's emotions and adjusts the behavior of the generative AI model based on those emotions.

[1576] User question input

[1577] A user inputs a question using a device such as a smartphone or tablet. For example, a user might input and send a question such as, "I've been looking for a heartwarming movie lately. Do you have any recommendations?" The emotion engine analyzes the user's emotions along with the question and sends the results to the server as additional information.

[1578] Submitting questions and generating answers

[1579] The server processes the received question along with the emotional information and sends it to the generative AI model in an appropriate format. The generative AI model takes the question and emotional information into consideration to generate an answer. For example, if the answer generated is "A recent heartwarming movie I recommend is 'Green Book,'" the content and format of the answer are adjusted based on the emotional information.

[1580] Presenting the first answer

[1581] The server sends the initial answer received from the generative AI model to the user's device, where the user can check the initial answer.

[1582] Entering Supplementary and Corrective Information

[1583] Other users can add supplementary or correction information to the primary answer. For example, they can add supplementary information such as, "Green Book is good, but I also recommend the recently released Shawshank Redemption." In this case, the emotion engine also recognizes the emotion of the user who added the supplementary or correction information and sends that information to the server.

[1584] Generate answers again

[1585] The server sends supplementary or corrective information and the emotional information to the generative AI model. The generative AI model generates a new answer based on this new information. For example, it might generate a new answer such as, "Recommended recent heartwarming movies are 'Green Book' and 'The Shawshank Redemption.' Other popular movies include 'Les Miserables.'"

[1586] Providing a second answer

[1587] The server sends the regenerated answer to the user's terminal, where the user can review the revised answer and other users can enter additional comments or corrections.

[1588] Generate the final answer

[1589] The generative AI model generates a final answer by repeating the process of re-answering multiple times as necessary. During this process, the emotion engine continuously analyzes the user's emotions and optimizes the answer. For example, the final answer might be, "Recommended recent heartwarming movies are 'Green Book,' 'The Shawshank Redemption,' 'Les Miserables,' and 'Twilight.'"

[1590] In this way, the system according to the present invention can provide a highly accurate final answer that takes into account the user's emotional information.

[1591] Hardware and software used

[1592] Hardware: Smartphones, tablets, computers

[1593] Software: The program is implemented in Python and utilizes APIs for emotion engines and generative AI models.

[1594] Emotion Engine API (e.g. https: / / emotion-api.example.com / analyze)

[1595] Generate AI Model API (e.g. https: / / ai-model-api.example.com / generate)

[1596] Examples of prompt statements

[1597] User Question: I've been looking for a heartwarming movie lately. Any recommendations?

[1598] Emotional information: Relaxing, Heartwarming, Comforting

[1599] Output of generative AI model: A recent recommended heartwarming movie is "Green Book," but other good choices are "The Shawshank Redemption" and "Les Miserables."

[1600] According to this invention, it is possible to improve the accuracy of answers by incorporating information about the user's emotions, and to recommend content with a higher degree of satisfaction.

[1601] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1602] Step 1:

[1603] The user inputs a question via the terminal. The user inputs a question into a smartphone or tablet and sends it. An example of input is, "I've been looking for a heartwarming movie lately. Do you have any recommendations?"

[1604] Step 2:

[1605] The terminal sends a question and the server receives it. The terminal sends the question entered by the user to the server. The server receives this question and proceeds to the next step.

[1606] Step 3:

[1607] The server sends the received question to the emotion engine to obtain emotion information. The server sends the question to the emotion engine (e.g. https: / / emotion-api.example.com / analyze). The input is the question text, and the output is emotion information. For example, the emotion information detected is "relaxing, heartwarming, comforting."

[1608] Step 4:

[1609] The server sends the question and emotion information to the generative AI model. After receiving the emotion information, the server sends it along with the question to the generative AI model (e.g., https: / / ai-model-api.example.com / generate). The input is the question and emotion information, and the generative AI model generates a first answer based on this.

[1610] Step 5:

[1611] The generative AI model generates a first answer and sends it back to the server. The generative AI model generates an answer based on the input question and emotional information. For example, it generates an answer such as "A recent recommended heartwarming movie is 'Green Book'." This answer is sent to the server.

[1612] Step 6:

[1613] The server sends the primary answer to the user's device. The server sends the primary answer received from the generative AI model to the user's device. The user can check the primary answer through their device.

[1614] Step 7:

[1615] Other users can input supplementary information or corrections to the primary answer via the terminal. Other users can input additional information or corrections to the primary answer. For example, they can add supplementary information such as, "'Green Book' is good, but I also recommend the recently released 'Shawshank Redemption'."

[1616] Step 8:

[1617] The terminal transmits supplementary or corrective information to the server, which receives it. The terminal transmits supplementary or corrective information from other users to the server, which receives it.

[1618] Step 9:

[1619] The server retransmits the supplementary or correction information and emotional information to the generative AI model. The server retransmits the received supplementary or correction information and the emotional information to the generative AI model. The input is the supplementary or correction information and emotional information.

[1620] Step 10:

[1621] The generative AI model generates a re-answer based on the new information and sends it back to the server. For example, it might generate a re-answer such as, "Recommended recent heartwarming movies are 'Green Book' and 'The Shawshank Redemption.' Other popular movies include 'Les Miserables.'" and send it to the server.

[1622] Step 11:

[1623] The server sends the re-answer to the user's device. The server sends the re-answer received from the generative AI model to the user's device, where the user can check the re-answer.

[1624] Step 12:

[1625] If necessary, other users can enter additional supplementary or correction information, which is then received by the server and sent to the generative AI model. Finally, the generative AI model generates a final answer. For example, the final answer might be, "Recommended recent heartwarming movies are 'Green Book,' 'The Shawshank Redemption,' 'Les Miserables,' and 'Twilight.'"

[1626] Step 13:

[1627] The server sends the final answer to the user's device. The server sends the final answer received from the generative AI model to the user's device, where the user can check the final answer.

[1628] This series of processes enables highly accurate content recommendations that reflect the user's emotional information.

[1629] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1630] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1631] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1632] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1633] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1634] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1635] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1636] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1637] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1638] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1639] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1640] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1641] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1642] 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.

[1643] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1644] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1645] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1646] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1647] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1648] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1649] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1650] The following is further disclosed regarding the above embodiment.

[1651] (Claim 1)

[1652] a means for a user to input a question via a terminal;

[1653] means for receiving the question at the server and transmitting it to the generative model in a format appropriate for the question;

[1654] means for generating a first-order answer to the question using the generative model and returning the first-order answer to the server;

[1655] A means for the server to transmit a primary response to the user's terminal;

[1656] a means for other users to input supplements or corrections to the primary answers via the terminal;

[1657] a means for the server to receive and retransmit supplemental or corrective information to the generative model;

[1658] A means for the generative model to generate an answer again and return the answer again to the server;

[1659] A means for the server to send a re-answer to the user's terminal;

[1660] A system including:

[1661] (Claim 2)

[1662] 10. The system of claim 1, further comprising means for generating a re-answer more than once for the generative model.

[1663] (Claim 3)

[1664] 10. The system of claim 1, further comprising means for the generative model to generate a final answer based on further supplemental or corrective input.

[1665] "Example 1"

[1666] (Claim 1)

[1667] a means for a user to input a question via the device;

[1668] means for receiving said question by a computer system and transmitting it in a format suitable for a generating algorithm;

[1669] means for the generation algorithm to generate an initial answer to the question and to return the initial answer to the computer system;

[1670] means for the computer system to transmit the initial response to the user's device;

[1671] a means for other users to enter supplements or corrections to the initial answers via the device;

[1672] means for the computer system to receive and retransmit supplemental or corrective information to the generating algorithm;

[1673] a generating algorithm for generating a new answer and sending the new answer back to the computer system;

[1674] means for the computer system to transmit the re-answer to the user's device;

[1675] means for generating re-answers multiple times until a final answer is generated;

[1676] A system including:

[1677] (Claim 2)

[1678] 10. The system of claim 1, wherein the generating algorithm further comprises means for generating the re-answer multiple times.

[1679] (Claim 3)

[1680] 10. The system of claim 1, further comprising means for the generating algorithm to generate a final answer based on further supplemental or corrective input.

[1681] "Application Example 1"

[1682] (Claim 1)

[1683] a means for a user to input a question via a terminal;

[1684] means for receiving the question at the server and transmitting it to the generative model in a format appropriate for the question;

[1685] means for generating a first-order answer to the question using the generative model and returning the first-order answer to the server;

[1686] A means for the server to transmit a primary response to the user's terminal;

[1687] a means for other users to input supplements or corrections to the primary answers via the terminal;

[1688] a means for the server to receive and retransmit supplemental or corrective information to the generative model;

[1689] A means for the generative model to generate an answer again and return the answer again to the server;

[1690] A means for the server to send a re-answer to the user's terminal;

[1691] A means for allowing users to ask questions about product information via a terminal in the virtual store;

[1692] A means for generating product information and comparison information using a generative model based on the question;

[1693] means for presenting the generated product information to a user;

[1694] A system including:

[1695] (Claim 2)

[1696] 10. The system of claim 1, further comprising means for generating a re-answer more than once for the generative model.

[1697] (Claim 3)

[1698] 10. The system of claim 1, further comprising means for the generative model to generate a final answer based on further supplemental or corrective input.

[1699] "Example 2: Combining Emotion Engines"

[1700] (Claim 1)

[1701] a means for a user to input a question via a terminal;

[1702] A server receives the question and the user's emotion information and transmits them to the generative model in a format appropriate for the server;

[1703] a means for generating a first answer based on the question and the emotion information by the generative model and returning the first answer to the server;

[1704] A means for the server to transmit a primary response to the user's terminal;

[1705] a means for other users to input supplements or corrections to the primary answers via the terminal;

[1706] a means for analyzing emotion information using an emotion engine when the supplement or correction is input, and a server receiving the supplement or correction information and the emotion information and retransmitting them to the generative model;

[1707] a means for the generative model to generate a new answer based on the supplementary or corrective information and the emotion information, and to return the new answer to the server;

[1708] A means for the server to send a second response to the user's terminal;

[1709] A system including:

[1710] (Claim 2)

[1711] 10. The system of claim 1, wherein the generative model includes means for generating a re-answer two or more times based on new supplemental or corrective information and sentiment information.

[1712] (Claim 3)

[1713] 10. The system of claim 1, wherein the generative model further comprises means for generating a final answer using an emotion engine.

[1714] "Application example 2 when combining emotion engines"

[1715] (Claim 1)

[1716] a means for a user to input a question via a terminal;

[1717] means for receiving the question at the server and transmitting it to the generative model in a format appropriate for the question;

[1718] means for generating a first-order answer to the question using the generative model and returning the first-order answer to the server;

[1719] A means for the server to transmit a primary response to the user's terminal;

[1720] a means for other users to input supplements or corrections to the primary answers via the terminal;

[1721] a means for the server to receive and retransmit supplemental or corrective information to the generative model;

[1722] A means for the generative model to generate an answer again and return the answer again to the server;

[1723] A means for the server to send a re-answer to the user's terminal;

[1724] a means for analyzing a user's emotions using an emotion engine and adjusting the behavior of the generative model based on the user's emotions;

[1725] A way to update your recommendations based on feedback from other users;

[1726] A system including:

[1727] (Claim 2)

[1728] 2. The system according to claim 1, further comprising means for generating a re-answer twice or more times by the generative model, and for regenerating the answer content in consideration of the emotion analysis result by the emotion engine.

[1729] (Claim 3)

[1730] 2. The system of claim 1, further comprising means for the generative model to generate a final answer according to the emotion information analyzed by the emotion engine based on the input of further supplementation or correction. [Explanation of symbols]

[1731] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for a user to input a question via a terminal; means for receiving the question at the server and transmitting it to the generative model in a format appropriate for the question; means for generating a first-order answer to the question using the generative model and returning the first-order answer to the server; A means for the server to transmit a primary response to the user's terminal; a means for other users to input supplements or corrections to the primary response via a terminal; a means for the server to receive and retransmit supplemental or corrective information to the generative model; A means for the generative model to generate an answer again and return the answer again to the server; A means for the server to send a re-answer to the user's terminal; A system including:

2. The system of claim 1 , further comprising means for generating a re-answer more than once for the generative model.

3. The system of claim 1 further comprising means for the generative model to generate a final answer based on further supplemental or corrective input.

Images