System

Abstract

A system is provided.SOLUTION: A system comprising: means for detecting that a question has been posted; means for sending a request to a generative AI model; means for receiving and displaying an answer from the generative AI model on a question page; means for displaying a plurality of options to a questioner; and means for processing the answer of the generative AI model based on a selection of the questioner.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] Q&A services have a problem where many respondents try to answer simple questions, resulting in a waste of time and resources. Another issue is that questioners cannot get appropriate answers quickly. Furthermore, there are fewer opportunities for respondents to provide high-quality answers, which can lead to a decline in the overall quality of content. [Means for solving the problem]

[0005] To address this issue, the present invention provides a system that, when a question is posted, first uses a generative AI model to provide an answer to the question. This system includes a means for detecting that a question has been posted, a means for sending a request to the generative AI model, a means for receiving an answer from the generative AI model and displaying it on the question page, a means for displaying multiple options to the questioner, and a means for processing the answer from the generative AI model based on the questioner's selection. This allows the questioner to quickly obtain an answer and allows the answerer to focus on more advanced questions. Furthermore, this system includes a means for displaying a "View AI Answer" button on the page when a question is posted, allowing the answer from the generative AI model to be easily checked. Furthermore, by including a means for storing the answer from the generative AI model in a database and setting the answer as the best answer based on the questioner's selection, it is possible to provide a quick and appropriate answer.

[0006] "Means for detecting that a question has been posted" refers to a function for identifying and detecting a post when a user posts a question on the Q&A service platform.

[0007] "Means for sending a request to a generative AI model" refers to the function of sending the detected question data to a generative AI model and providing instructions for generating an answer based on the question content.

[0008] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to the functionality for receiving answer data returned from the generative AI model and visually displaying it on the question page.

[0009] "Means for displaying multiple options to the questioner" refers to a function that displays an options menu that allows the questioner to select an appropriate action in response to the answer provided by the generative AI model.

[0010] "Means for processing the generative AI model's answer based on the questioner's selection" refers to a function for performing processing such as setting the answer provided by the generative AI model as the best answer or not adding an answer, depending on the action selected by the questioner.

[0011] "A means to display a 'View AI response' button on the page when a question is posted" refers to a function that dynamically displays a button on the question page to check the response of the generative AI model when a new question is posted.

[0012] "Means for storing the answer of the generative AI model in a database and setting the answer as the best answer based on the selection of the questioner" refers to the function for recording the answer of the generative AI model in a database and managing that answer as the best answer based on the selection of the questioner. [Brief explanation of the drawings]

[0013] [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

[0014] 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.

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

[0016] 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).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

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

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0028] 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.

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

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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."

[0034] This invention is a system that uses a generative AI model to quickly provide answers to questions in a Q&A service. To implement this system, the following program processes and operations are required.

[0035] 1. Posting and discovering questions

[0036] A user enters a question on the Q&A service platform and clicks the "Post" button.

[0037] The terminal transmits the input question data to the server.

[0038] The server detects when a question is posted, analyzes the question data, and classifies it into the appropriate category.

[0039] 2. Display of "See AI's Answer" button

[0040] The server dynamically adds a "See AI's Answer" button to the question page.

[0041] The terminal redraws the updated page to the user, displaying the new button.

[0042] 3. Sending a request to the generative AI model

[0043] The user clicks the "See AI's answer" button.

[0044] The terminal detects this click event and sends the request to the server.

[0045] The server sends the question data as a request to the API of the generative AI model.

[0046] 4. Receiving and displaying answers from the generative AI model

[0047] The generative AI model analyzes the received question data and generates an appropriate answer.

[0048] The generative AI model returns the generated answer to the server.

[0049] The server formats the received answer data and adds it to the question page.

[0050] The device re-renders the updated page to the user, displaying the generative AI model's answer.

[0051] 5. Displaying options to the questioner

[0052] The server generates a UI that displays multiple options that the questioner can choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Do not add the AI ​​answer").

[0053] The device redraws its UI and displays it to the user.

[0054] 6. Processing Choice-Based Generative AI Model Answers

[0055] The user clicks on the appropriate option.

[0056] The terminal transmits the selection result to the server.

[0057] The server processes the generative AI model's answer based on the questioner's selection.

[0058] "Mark as Best Answer": Set the generative AI model's answer as the best answer in the database and display it on the page.

[0059] "Add to page without setting as best answer": The answer from the generative AI model is added to the database as an answer, and is displayed on the page without being set as the best answer.

[0060] "Don't add AI answer": The generated AI model's answer will not be added to the page or database.

[0061] Specific examples

[0062] Example 1: A simple question

[0063] A user posts a question to a Q&A service: "What is the appropriate amount of food for a cat?"

[0064] The server detects the question and displays a "See AI's Answer" button.

[0065] When the user clicks the button, the generative AI model generates and displays the answer: "For an adult cat, 2-3% of body weight is appropriate."

[0066] The user selects "make best answer" and the server sets the answer as the best answer.

[0067] Example 2: Advanced Questions

[0068] A user posts the question "What is the procedure for normalizing a wave function in quantum mechanics?"

[0069] The server detects the question and similarly displays a "See AI Answer" button.

[0070] When the user clicks the button, the generative AI model generates and displays the answer: "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1."

[0071] The user decides that the answer is insufficient and selects "Add to page without selecting as best answer." The server adds the generated AI model's answer to the page and expects other users to provide more detailed answers.

[0072] In this way, by utilizing a generative AI model, the system of the present invention allows questioners to receive quick and appropriate answers, allowing answerers to focus on asking more advanced questions.

[0073] The processing flow will be explained below.

[0074] Step 1:

[0075] The user enters a question into the input form on the Q&A service platform and clicks the "Post" button.

[0076] Step 2:

[0077] The terminal transmits the posted question data to the server.

[0078] Step 3:

[0079] The server stores the received question data in a database and detects that a question has been posted.

[0080] Step 4:

[0081] The server generates a script that dynamically adds a "View AI Answer" button to the question page.

[0082] Step 5:

[0083] The device receives the script sent from the server, redraws the question page, and displays the "See AI's Answer" button.

[0084] Step 6:

[0085] The user clicks the "See AI's answer" button.

[0086] Step 7:

[0087] The terminal detects the button click event and sends the request to the server.

[0088] Step 8:

[0089] The server sends the question data as a request to the API of the generative AI model.

[0090] Step 9:

[0091] A generative AI model analyzes the question data and generates appropriate answers.

[0092] Step 10:

[0093] The generative AI model returns the generated answer to the server.

[0094] Step 11:

[0095] The server receives the generated response data and formats it appropriately.

[0096] Step 12:

[0097] The server adds HTML containing the formatted answer to the question page.

[0098] Step 13:

[0099] The device renders the HTML received from the server and redraws the question page to display the answer from the generative AI model.

[0100] Step 14:

[0101] The server generates a UI to display multiple options for the answer of the generated AI model ("Make it the best answer," "Don't make it the best answer but add it to the page," "Don't add the AI ​​answer").

[0102] Step 15:

[0103] The device renders the UI with the options and redraws the question page to display to the user.

[0104] Step 16:

[0105] The user clicks on one of the choices.

[0106] Step 17:

[0107] The terminal detects the selection click event and sends the result to the server.

[0108] Step 18:

[0109] The server processes the answer of the generative AI model based on the selection results.

[0110] Step 19:

[0111] If you select "Set as best answer," the server will set the generated AI model's answer as the best answer in the database and display it on the question page.

[0112] Step 20:

[0113] If you select "Add to page without setting as best answer," the server will add the generated AI model's answer to the database and display it on the question page, but will not set it as the best answer.

[0114] Step 21:

[0115] If you select "Do not add AI answer," the server will not add the generated AI model's answer to the database or the question page, and will return the result to the device.

[0116] Step 22:

[0117] The terminal redraws the question page based on the results received from the server and displays the final state to the user.

[0118] The above is the specific program processing flow of this system.

[0119] Example 1

[0120] 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."

[0121] Conventional Q&A services have the problem that it takes a long time for questioners to receive an appropriate answer. Furthermore, because the quality and speed of the answer depends on the answerer, it is difficult to obtain an appropriate answer immediately to the question. Furthermore, the system's efficiency is low because question categorization and responses are done manually. Furthermore, conventional systems have limited options for automated answers using AI, leaving users with little flexibility in selecting AI answers.

[0122] 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.

[0123] In this invention, the server includes means for detecting that a question has been posted, means for analyzing the question data and classifying it into an appropriate category, means for dynamically adding a "See AI's Answer" button to the question page, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, and means for processing the answer of the generative AI model based on the choice made by the questioner. This allows the questioner to quickly obtain an appropriate answer, while allowing the answerer to focus on more advanced questions, improving the efficiency of the entire system.

[0124] The "means for detecting that a question has been posted" refers to a method or device for detecting that a user has entered a question on the Q&A service platform and clicked the post button.

[0125] "Means for analyzing question data and classifying it into appropriate categories" refers to a method or device that analyzes input question data using natural language processing (NLP) or the like, and classifies it into relevant categories based on its content.

[0126] The "means for dynamically adding a 'See AI's Answer' button to a question page" refers to a method or device for adding and displaying a 'See AI's Answer' button on a question page in real time after a question is posted.

[0127] "Means for sending a request to the generative AI model" refers to a method or device that sends the question data as a request to the API of the generative AI model when a user clicks the "View AI's answer" button.

[0128] "Means for receiving an answer from a generative AI model and displaying it on a question page" refers to a method or device for receiving an answer returned from a generative AI model and displaying it on a question page in a form that can be viewed by a user.

[0129] "Means for displaying multiple options to the questioner" refers to a method or device that displays options for the questioner in response to the answer of the generative AI model, such as "Make this the best answer," "Add this to the page without making it the best answer," or "Do not add this to the AI ​​answer."

[0130] "Means for processing the answer of a generative AI model based on the questioner's selection" refers to a method or device that processes the answer of a generative AI model in accordance with the questioner's selection from the options presented, such as saving it in a database, displaying it, or discarding it.

[0131] This invention is a system that uses a generative AI model to provide quick answers to questions in a Q&A service. This system is realized by combining various means that operate mainly on a server, a terminal, and a user.

[0132] First, a user enters a question on the Q&A service platform and clicks the "Post" button. The device detects this action and sends the entered question data to the server. The server analyzes the received question data and detects that it is a question. The server then analyzes the question data using a natural language processing (NLP) library (e.g., spaCy) and classifies it into the appropriate category.

[0133] Next, the server generates HTML code to dynamically add a "See AI's Answer" button to the question page and sends it to the terminal. The terminal receives this code and redraws the user's browser screen to display the "See AI's Answer" button. When the user clicks this button, the terminal detects the click event and sends the request to the server.

[0134] Based on the received request, the server sends question data as a request to the generative AI model's API (e.g., OpenAI). The generative AI model analyzes the sent question data and generates an answer. The generative AI model returns the generated answer to the server. The server formats the received answer data, generates HTML code to add to the question page, and sends it to the device. The device receives this updated data and displays the generative AI model's answer by redrawing the user's browser screen.

[0135] Furthermore, the server generates a UI to display multiple options that the user can select from for the generated AI model's answer (e.g., "Make this the best answer," "Add to the page without making this the best answer," and "Don't add the AI ​​answer"). The device redraws this UI and displays it on the page. When the user clicks on an appropriate option, the device sends this selection to the server. The server processes the generated AI model's answer based on the user's selection, for example, saving it as the best answer in a database or updating the page.

[0136] Specific examples

[0137] Prompt statement example 1:

[0138] A user posts a question to a Q&A service: "What is the appropriate amount of food for a cat?" The server detects the question and displays a "See AI's Answer" button. When the user clicks the button, the generative AI model generates and displays the answer: "For an adult cat, 2-3% of its body weight is appropriate." The user selects "Make this the best answer," and the server adds the answer to the database as the best answer.

[0139] Prompt statement example 2:

[0140] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?" The server detects the question and similarly displays a "See AI's answer" button. When the user clicks the button, the generative AI model generates an answer: "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1," and displays it. The user decides that the answer is insufficient and selects "Add to page without marking as best answer." The server adds the generative AI model's answer to the page, and expects other users to provide more detailed answers.

[0141] In this way, the system utilizes generative AI models to enable users to get fast and relevant answers, allowing other users to focus on more advanced questions, improving the overall efficiency of the system and providing a better user experience by providing options for questioners.

[0142] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0143] Step 1: Post a question

[0144] Users enter their questions on the Q&A service platform and click the "Post" button.

[0145] Input: Text data of the user's question (e.g., "What is the appropriate amount of food for a cat?")

[0146] The terminal receives user input, converts the question data into JSON format, and sends it to the server.

[0147] Output: Question data in JSON format

[0148] Step 2: Question detection and parsing

[0149] The server receives the JSON data and detects that a question has been posted.

[0150] Input: JSON format question data sent from the terminal

[0151] The server uses an NLP library (e.g., spaCy) to parse the question data and classify it into appropriate categories.

[0152] Output: Question analysis results and category information (e.g., "Pet category")

[0153] Step 3: Display of the "See AI's Answer" button

[0154] The server generates HTML code to dynamically add a "View AI Answer" button to the question page.

[0155] Input: User question, analysis results, category information

[0156] The server sends a response containing the generated HTML code to the terminal.

[0157] The device uses the updated data received from the server to redraw the user's browser screen, which displays a new "See AI's Answer" button.

[0158] Output: Updated HTML code

[0159] Step 4: Sending a request to the generative AI model

[0160] The user clicks the "See AI's answer" button.

[0161] Input: User click event

[0162] The terminal detects the click event and sends a request along with the question data to the server.

[0163] Output: Request data to the server

[0164] The server sends the question data as a request to the API of the generative AI model (e.g., OpenAI).

[0165] Step 5: Receive and display the answer from the generative AI model

[0166] The server receives the response from the generative AI model.

[0167] Input: Answer data from the generative AI model

[0168] The generative AI model analyzes the question data and generates an appropriate answer (e.g., "For adult cats, 2-3% of body weight is appropriate").

[0169] The server formats the generated answer and generates HTML code to add to the question page.

[0170] Output: Updated HTML code and generated AI answer data

[0171] The device receives the updated page data and redraws the user's browser screen, which displays the answer from the generated AI model.

[0172] Step 6: Displaying the options

[0173] The server generates a UI to display multiple options for the user to choose from in response to the generated AI model's answer.

[0174] Input: Answer data for the generative AI model

[0175] The server transmits the generated UI data to the terminal.

[0176] The device uses the UI data to display the options to the user.

[0177] Output: Updated UI with choices

[0178] Step 7: Processing based on user selection

[0179] The user clicks on the appropriate option.

[0180] Input: User click event

[0181] The terminal transmits the selection result to the server.

[0182] Output: Request data to the server

[0183] The server processes the generative AI model's answer based on the questioner's selection.

[0184] "Mark as Best Answer": Set the generative AI model's answer as the best answer in the database and display it on the page.

[0185] "Add to page without setting as best answer": Add the generative AI model's answer and display it on the page without setting it as the best answer.

[0186] "Don't add AI answer": The generated AI model's answer will not be added to the page or database.

[0187] The above are the specific processing steps and operations of the program for this system.

[0188] (Application example 1)

[0189] 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."

[0190] In recent years, with the spread of electronic payment services, the number of inquiries from users has increased dramatically. This has resulted in delayed support responses, leading to problems that lead to a decline in user satisfaction. Furthermore, support staff are required to respond to a large number of inquiries, increasing their workload. To solve this problem, a system that can provide prompt and appropriate responses is needed. However, existing systems have difficulty in providing prompt and appropriate responses to inquiries.

[0191] 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.

[0192] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, means for processing the answer from the generative AI model based on the questioner's selection, means for providing a user interface for inputting questions about electronic payments, and means for receiving the answer from the generative AI model and displaying it on the question page related to electronic payments. This allows users to receive quick and accurate answers to inquiries about electronic payments, improving the efficiency of support responses and reducing workloads.

[0193] The "means for detecting that a question has been posted" is a mechanism by which the system determines that a user has entered and posted a question on the system.

[0194] The "means for sending a request to the generative AI model" is a mechanism for sending the posted question to the generative AI model and sending a request to generate an appropriate answer.

[0195] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to a mechanism for receiving answers returned from the generative AI model and displaying them on the user interface.

[0196] "Means for displaying multiple options to the questioner" refers to a mechanism that displays multiple options so that the user can select further actions in response to the answer from the generative AI model.

[0197] The "means for processing the generative AI model's answer based on the questioner's selection" refers to a mechanism that, after the user selects an option, stores or displays the generative AI model's answer in a database based on that selection.

[0198] The "means for providing a user interface for inputting questions about electronic payments" is a mechanism for providing an interface designed to allow users to intuitively input questions about electronic payments.

[0199] "Means for receiving the answer of the generative AI model and displaying it on the question page related to electronic payments" refers to a mechanism for receiving the answer related to electronic payments returned from the generative AI model and displaying it on the relevant question page.

[0200] This invention is a system for quickly and accurately answering questions about electronic payments. The system consists of posting and detecting questions, sending requests to a generative AI model, receiving and displaying answers, presenting options to the user, and processing the answers based on the selections.

[0201] The core of the system resides in the server, which has the following main functions:

[0202] 1. Post and discover questions:

[0203] When a user enters a question about electronic payments on the smartphone application and clicks the "Post" button, the device sends the question data to the server, which detects the post, analyzes the received question data, and classifies it into the appropriate category.

[0204] 2. Display of "See AI's Answer" button:

[0205] When a question is submitted, the server dynamically adds a "See AI's Answer" button to the question page. After this button is added to the page, the device redraws the page and displays the new button to the user.

[0206] 3. Sending a request to the generative AI model:

[0207] When a user clicks the "View AI's Answer" button, the click event is detected by the device and a request is sent to the server. The server then sends the received question data as a request to the API of the generative AI model.

[0208] 4. Receive and display answers from the generative AI model:

[0209] The generative AI model analyzes the received question data and generates an appropriate answer. The generated answer is returned to the server, which formats the answer data and adds it to the question page. The device redraws the updated page and displays the generative AI model's answer to the user.

[0210] 5. Displaying options to the questioner:

[0211] The server generates a UI to display multiple options for the questioner to choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Don't add the AI ​​answer"), and the device redraws the UI to display the options to the user.

[0212] 6. Processing the generative AI model's answers based on your choices:

[0213] When the user clicks on the appropriate option, the device sends the selection to the server, which then takes appropriate action on the generative AI model's answer based on the user's selection, such as setting it as the best answer, adding it to the page, or ignoring it.

[0214] The following hardware and software are used to realize this system.

[0215] 1. Smartphone application:

[0216] Platform: iOS and Android

[0217] Development language: Swift (iOS), Kotlin (Android)

[0218] 2. Server:

[0219] Platform: Linux-based server environment

[0220] Software used: Node.js, Django, Flask

[0221] 3. Generative AI Model:

[0222] Model Name: OpenAI GPT-3 or GPT-4

[0223] API used: OpenAI API

[0224] As a concrete example, if a question is posted asking "Where can I check my credit card statement?", the system will operate as follows:

[0225] 1. The server receives the question and adds a "See AI's Answer" button to the question page.

[0226] 2. When the user clicks this button, the question data is sent to the generative AI model.

[0227] 3. The generative AI model generates the answer, "You can check your credit card statement on the 'Statement' page of the official app or website," and returns it to the server.

[0228] 4. The server displays this answer on the question page and gives the user the options of "Make it the best answer," "Add it to the page without making it the best answer," or "Don't add the AI ​​answer."

[0229] 5. If the user selects "Mark as Best Answer," the answer is saved in the database and set as the best answer on the question page.

[0230] This system will enable users to quickly obtain information about electronic payments and is expected to reduce the burden on support staff.

[0231] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0232] Step 1: Posting and discovering questions

[0233] A user enters a question about electronic payment into a smartphone application and clicks the "Post" button. At this time, the question text entered by the user is generated. The device sends this question text to the server. The server analyzes the received question data and detects that a question has been posted. Next, it classifies the question data into appropriate categories. This data processing process organizes the question content into categories.

[0234] Input: The question text entered by the user

[0235] Data processing: categorization of question data

[0236] Output: Question data with categories

[0237] Step 2: Displaying the "See AI's Answer" button

[0238] The server dynamically adds a "See AI's Answer" button to the question page based on the question data it received. It then instructs the device to redraw the updated page and display the new button to the user. This operation allows the user to expect a quick answer from the AI.

[0239] Input: Question data with categories

[0240] Data processing: Update the question page and add a button

[0241] Output: Updated question page with a "See AI's answer" button

[0242] Step 3: Sending a request to the generative AI model

[0243] When the user clicks the "View AI Answer" button, the device detects this click event. Based on this detection, the device resends the question data to the server. The server then sends the received question data as a request to the generative AI model's API. This data transmission causes the generative AI model to begin analyzing the question content.

[0244] Input: Click event of the "View AI's answer" button and question data

[0245] Data processing: Sending a request to the generative AI model API

[0246] Output: The question request received by the generative AI model

[0247] Step 4: Receive and display the answer from the generative AI model

[0248] The generative AI model analyzes the received question data and generates an appropriate answer. This answer is returned to the server. The server formats the received answer data and adds it to the question page. The device then redraws the updated page and displays the generative AI model's answer, allowing the user to see the answer provided by the AI ​​model.

[0249] Input: Answer data generated by a generative AI model

[0250] Data processing: Formatting of response data

[0251] Output: Question page with formatted answer data

[0252] Step 5: Present the options to the questioner

[0253] The server generates a UI to display multiple options for the questioner to choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Don't add the AI ​​answer") The device redraws this UI and displays it to the user.

[0254] Input: Answer data for the generative AI model

[0255] Data processing: generating options and updating the UI

[0256] Output: Question page with updated UI

[0257] Step 6: Processing the generative AI model's answer based on your selection

[0258] When the user clicks on the appropriate option, the device detects this click event and sends the selection result to the server. Based on the selection, the server saves the generative AI model's answer in a database or displays it on the page. For example, if the user selects "Set as best answer," the answer is set as the best answer in the database and displayed on the question page.

[0259] Input: User selection

[0260] Data processing: Save answers to database and update page display

[0261] Output: Updated question page and database state

[0262] 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.

[0263] This invention is a system that provides quick and appropriate answers to questions in a Q&A service using a generative AI model and an emotion engine. The configuration of this system and the program processing flow are described in detail below.

[0264] System configuration

[0265] The system mainly consists of the following components:

[0266] 1. Question Post Detection

[0267] The server has a means to detect new questions posted by users.

[0268] 2. Sending a request to the generative AI model

[0269] The server has the ability to send the detected questions to the generative AI model.

[0270] 3. Receiving and displaying answers from the generative AI model

[0271] The server has a means to receive the answer from the generative AI model and display it on the question page.

[0272] 4. Emotion Engine

[0273] The server has the means to recognize the questioner's emotions and analyze that information.

[0274] The emotion engine adjusts the generative AI model's responses based on the questioner's emotional state.

[0275] 5. Display of questioner's options

[0276] The server has a means to display a UI that offers the questioner multiple options for the answer of the generated AI model.

[0277] 6. Processing Choice-Based Answers

[0278] The server has a means to process the generative AI model's answer based on the questioner's selection.

[0279] The server has the function of storing the emotion information recognized by the emotion engine in a database.

[0280] 7. Learning the Emotion Engine

[0281] The server has a learning mechanism to improve emotion recognition accuracy based on feedback from the questioner.

[0282] Program flow

[0283] Posting and discovering questions

[0284] A user enters a question into the Q&A service's input form and clicks the "Post" button. The device sends the entered question data to the server. The server saves the question in a database, detects that the question has been posted, and analyzes the questioner's emotional information using an emotion engine.

[0285] Display of "See AI's answer" button

[0286] The server generates a script that dynamically adds a "See AI's Answer" button to the question page. The device receives this script, redraws the question page, and displays the "See AI's Answer" button.

[0287] Linking generative AI models and emotion engines

[0288] The user clicks the "See AI's Answer" button. The device detects the click event and sends the request to the server. The server sends the question data to the generative AI model, which then generates an answer.

[0289] The emotion engine adjusts the generated answer based on the emotional state of the questioner. The server receives the adjusted answer and displays it on the question page.

[0290] Displaying options to the questioner

[0291] The server generates a UI for the questioner to choose from three options for the AI ​​model's answer: "Make it the best answer," "Add it to the page without making it the best answer," or "Don't add it to the AI ​​answer." The device redraws the page and displays the options to the questioner.

[0292] Processing questioner selections and answers

[0293] The user clicks on one of the options. The device sends the selection to the server, which processes the answer of the generative AI model based on the selection.

[0294] Mark as best answer: The server saves the generated AI model's answer in the database as the best answer and displays it on the page.

[0295] Add to page without marking as best answer: The server adds the generated AI model's answer to the page, making it visible to other users.

[0296] Do not add AI answer: The server does not add the generated AI model answer to the page or database, but leaves it as is.

[0297] Emotion Engine Learning

[0298] Based on the feedback from the questioner, the server performs a learning process to improve the emotion recognition accuracy of the emotion engine.

[0299] Specific examples

[0300] Example 1: A simple question

[0301] A user posts a question: "What is the appropriate amount of food for a cat?" The server detects the question and displays a "See AI's Answer" button. When the user clicks the button, the generative AI model generates the answer: "For an adult cat, 2-3% of its body weight is appropriate," and the emotion engine appropriately adjusts the answer based on the questioner's emotional information and displays it. The user selects "Make Best Answer," and the server sets the answer as the best answer.

[0302] Example 2: A more complex question

[0303] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?" The server detects the question and displays a "See AI's answer" button. When the user clicks the button, the generative AI model generates the answer: "The procedure is to integrate the wave function over the entire space and make the square of the absolute value of that result 1." The emotion engine analyzes the questioner's emotions and adjusts the answer. The user selects "Add to page without marking as best answer," and the server adds the generative AI model's answer to the page.

[0304] In this way, the system of the present invention uses a generative AI model and an emotion engine in combination, allowing the questioner to receive a quick and appropriate answer while allowing the answerer to focus on asking more advanced questions.

[0305] The processing flow will be explained below.

[0306] Step 1:

[0307] The user enters a question into the Q&A service's input form and clicks the "Post" button.

[0308] Step 2:

[0309] The terminal transmits the posted question data to the server.

[0310] Step 3:

[0311] The server stores the received question data in a database and detects that a question has been posted.

[0312] Step 4:

[0313] The server calls the emotion engine and analyzes the questioner's emotional information based on the question data.

[0314] Step 5:

[0315] The server stores the emotion analysis results in a database.

[0316] Step 6:

[0317] The server generates a script that dynamically adds a "View AI Answer" button to the question page.

[0318] Step 7:

[0319] The device receives the script sent from the server, redraws the question page, and displays the "See AI's Answer" button.

[0320] Step 8:

[0321] The user clicks the "See AI's answer" button.

[0322] Step 9:

[0323] The terminal detects the button click event and sends the request to the server.

[0324] Step 10:

[0325] The server sends the question data and emotion information from the emotion engine as a request to the API of the generative AI model.

[0326] Step 11:

[0327] The generative AI model analyzes question data and emotional information to generate appropriate answers.

[0328] Step 12:

[0329] The generative AI model returns the generated answer to the server.

[0330] Step 13:

[0331] The server receives the generated response data and formats it appropriately.

[0332] Step 14:

[0333] The server adds HTML containing the formatted answer to the question page.

[0334] Step 15:

[0335] The device renders the HTML received from the server and redraws the question page to display the answer from the generative AI model.

[0336] Step 16:

[0337] The server generates a UI that provides multiple options for the answer of the generated AI model.

[0338] Step 17:

[0339] The device renders the UI with the options and redraws the question page to display to the user.

[0340] Step 18:

[0341] The user can choose to either "Make this the best answer," "Add it to the page without making it the best answer," or "Do not add the AI ​​answer."

[0342] Step 19:

[0343] The terminal transmits the user's selection to the server.

[0344] Step 20:

[0345] The server processes the answer of the generative AI model based on the selection results.

[0346] Step 21:

[0347] If you select "Set as best answer," the server will set the generated AI model's answer as the best answer in the database and display it on the question page.

[0348] Step 22:

[0349] If you select "Add to page without setting as best answer," the server will add the generated AI model's answer to the database and display it on the question page, but will not set it as the best answer.

[0350] Step 23:

[0351] If you select "Do not add AI answer", the server will not add the generated AI model's answer to the database or the question page.

[0352] Step 24:

[0353] The server sends feedback from the questioner to the emotion engine, which then learns to improve emotion recognition accuracy.

[0354] Step 25:

[0355] The terminal redraws the question page based on the final result received from the server and displays it to the user.

[0356] The above is the specific program processing flow of this system.

[0357] Example 2

[0358] 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."

[0359] With conventional Q&A services, it was difficult for questioners to obtain prompt and appropriate answers. Furthermore, the answers generated by the AI ​​model were not properly adjusted according to the questioner's feelings, resulting in low satisfaction with the answer. Furthermore, there was a lack of a way to clearly indicate which option the questioner should take in response to the question posted, which did not improve the user experience.

[0360] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0361] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for analyzing the emotion of the questioner, means for adjusting the content of the answer of the generative AI model based on the analyzed emotion, means for displaying multiple options to the questioner, means for processing the answer of the generative AI model based on the selection of the questioner, and means for storing the emotion information in a database. This allows the questioner to receive a quick and appropriate answer, and further enables the answer to be adjusted according to the questioner's emotion, thereby improving the user experience.

[0362] "Question detection" is the process by which the server automatically recognizes questions posted by users.

[0363] "Sending a request to the generative AI model" is a means of sending information from the server to the generative AI model to analyze the detected question.

[0364] "Receiving and displaying answers from the generative AI model" refers to the process in which the server receives the answers provided by the generative AI model and displays them on the question page.

[0365] "Emotion analysis" is a means of determining the questioner's emotional state using an emotion engine.

[0366] "Adjusting response content based on emotions" is a method of appropriately changing the response content provided by the generative AI model based on analyzed emotional information.

[0367] "Displaying options" is the process by which the server provides a user interface with multiple options that the questioner can select from in response to the answer provided by the generative AI model.

[0368] "Choice-based answer processing" is the process by which the server appropriately processes the answer of the generative AI model depending on the option selected by the questioner.

[0369] "Storing emotional information" is a method of storing the emotional information analyzed by the emotion engine in a database.

[0370] This invention is a system that provides quick and appropriate answers to questions in a Q&A service using a generative AI model and an emotion engine. The configuration of this system and the program processing flow are described in detail below.

[0371] System configuration

[0372] The system consists of the following main components:

[0373] 1. Server

[0374] The server has a means for receiving and detecting queries from users.

[0375] It has the ability to send question data to a generative AI model.

[0376] It has a means to receive answers from the generative AI model and display them on the question page.

[0377] It integrates an emotion engine to analyze the questioner's emotions.

[0378] It has the ability to adjust the answers of the generative AI model based on the analyzed emotional information.

[0379] It has a means to display multiple options in the user interface.

[0380] It has a means to process the generative AI model's answers based on the choices.

[0381] It has the function of storing emotional information in a database.

[0382] 2. Terminal

[0383] The terminal provides an interface where the user can enter a question and click a "Post" button.

[0384] It has a means for sending question data to the server.

[0385] It has the means to receive and execute a script to dynamically display a "View AI Answer" button on the question page.

[0386] It has the function of sending the selection results to the server.

[0387] 3. Users

[0388] Users post questions using the input form of the Q&A service.

[0389] Click the "See AI's Answer" button to see the answer from the generated AI model.

[0390] Make the appropriate choice from multiple options.

[0391] Specific examples

[0392] Example 1: A simple question

[0393] Example prompt sentence:

[0394] What is the appropriate amount of cat food?

[0395] A user posts a question: "What is the right amount of food for a cat?"

[0396] The server detects the question and analyzes the questioner's emotional information using an emotion engine.

[0397] The server sends a script to the terminal that displays a "See AI's answer" button.

[0398] When the user clicks the "See AI Answer" button, the server sends the question data to the generative AI model and receives the answer.

[0399] For example, the generative AI model generates an answer such as "2-3% of body weight is appropriate for an adult cat," which is received by the server.

[0400] The emotion engine takes into account the emotional state of the questioner and adjusts the answer accordingly.

[0401] The server displays the adjusted answer on the question page.

[0402] The user selects "Mark as Best Answer" and the server saves the answer as the best answer in the database.

[0403] Example 2: Complex Question

[0404] Example prompt sentence:

[0405] "What is the procedure for normalizing wave functions in quantum mechanics?"

[0406] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?"

[0407] The server detects the question and analyzes the questioner's emotional information using an emotion engine.

[0408] The server sends a script to the terminal that displays a "See AI's answer" button.

[0409] When the user clicks the "See AI Answer" button, the server sends the question data to the generative AI model and receives the answer.

[0410] For example, the generative AI model generates an answer such as, "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1," and the server receives this.

[0411] The emotion engine analyzes the questioner's emotions and adjusts the response appropriately.

[0412] The server displays the adjusted answer on the question page.

[0413] The user selects "Add to page without marking as best answer" and the server adds the answer to the page.

[0414] In this way, the system of the present invention can use generative AI models and emotion engines to provide users with fast and relevant answers.

[0415] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0416] Specific explanation of program processing

[0417] Step 1:

[0418] The user enters a question into the Q&A service's input form and clicks the "Post" button.

[0419] Input: The question text entered by the user.

[0420] Output: The query data is sent to the server.

[0421] Specific behavior:

[0422] The terminal obtains the contents of the user input form.

[0423] The terminal generates a request to transmit the question data to the server.

[0424] The device sends a request to the server.

[0425] Step 2:

[0426] The server stores the received question data in a database and detects when a new question has been posted.

[0427] The server sends the question to the emotion engine and analyzes the questioner's emotional information.

[0428] Input: Question data sent from the terminal.

[0429] Output: The question data is stored in a database and the sentiment analysis results are obtained.

[0430] Specific behavior:

[0431] The server writes the received question data to the database.

[0432] The server monitors the database to detect new question submissions.

[0433] The server sends the question to the emotion engine and receives the analysis results.

[0434] Step 3:

[0435] The server generates a script that adds a "See AI's Answer" button to the question page and sends it to the terminal.

[0436] Input: The event that a new question is posted.

[0437] Output: The script is sent to the terminal and the button is displayed.

[0438] Specific behavior:

[0439] The server generates script code to add a "See AI's Answer" button.

[0440] The server sends the generated script to the terminal.

[0441] Step 4:

[0442] The device receives the script, redraws the question page, and displays the "See AI's Answer" button.

[0443] Input: The script sent by the server.

[0444] Output: A "See AI's answer" button will be displayed on the question page.

[0445] Specific behavior:

[0446] The terminal parses the script received from the server.

[0447] The terminal executes the received script and redraws the question page.

[0448] The device will display a "See AI's Answer" button.

[0449] Step 5:

[0450] The user clicks the "See AI's answer" button.

[0451] Input: User click event.

[0452] Output: A click event is sent to the server.

[0453] Specific behavior:

[0454] The device detects a click event.

[0455] The device sends the click event information to the server.

[0456] Step 6:

[0457] The server sends the question data to the generative AI model, which then generates an answer.

[0458] Input: Click event information and original question data.

[0459] Output: Answer data from the generative AI model.

[0460] Specific behavior:

[0461] The server creates a request to send the original question data to the generative AI model.

[0462] The server calls the API of the generated AI model and sends the question data.

[0463] The generative AI model analyzes the question and generates an appropriate answer.

[0464] The answer data generated by the generative AI model is sent back to the server.

[0465] Step 7:

[0466] The emotion engine takes into account the emotional state of the questioner and adjusts the generated answer content.

[0467] Input: Response data from the generative AI model and pre-obtained sentiment analysis results.

[0468] Output: Adjusted response data.

[0469] Specific behavior:

[0470] The server sends the response data from the generative AI model and the emotion analysis results to the emotion engine.

[0471] The emotion engine adjusts the response content.

[0472] The emotion engine sends the adjusted response data back to the server.

[0473] Step 8:

[0474] The server receives the adjusted answer and displays it on the question page.

[0475] Input: Calibrated response data from the sentiment engine.

[0476] Output: The answer displayed on the question page.

[0477] Specific behavior:

[0478] The server receives the adjusted response data.

[0479] The server generates data for updating the question page and sends it to the terminal.

[0480] Step 9:

[0481] The device will redraw the question page and display the adjusted answer.

[0482] Input: Data to update from the server.

[0483] Output: Redrawn question page and displayed answer.

[0484] Specific behavior:

[0485] The terminal analyzes the data received from the server and redraws the question page.

[0486] The device will display the adjusted answer.

[0487] Step 10:

[0488] The server generates a UI containing three options that the questioner can choose from in response to the generated AI model's answer: "Make this the best answer," "Add it to the page without making it the best answer," or "Do not add the AI ​​answer," and sends this UI to the device.

[0489] Input: Adjusted response data.

[0490] Output: The UI shown to the questioner.

[0491] Specific behavior:

[0492] Create a script that the server uses to generate a UI with the options.

[0493] The server sends the generated script to the terminal.

[0494] Step 11:

[0495] The terminal will redraw and show the options to the questioner.

[0496] Input: Script from server.

[0497] Output: The choices presented to the questioner.

[0498] Specific behavior:

[0499] The terminal parses the script received from the server.

[0500] The device will redraw the question page and display the options.

[0501] Step 12:

[0502] The user clicks on one of the displayed options.

[0503] Input: User click event.

[0504] Output: Click event information is sent to the server.

[0505] Specific behavior:

[0506] The device detects a click event.

[0507] The device sends the click event information to the server.

[0508] Step 13:

[0509] The server processes the generative AI model's answer based on the selection results.

[0510] Input: Click event information.

[0511] Output: Processed response data.

[0512] Specific behavior:

[0513] The server appropriately processes the answer of the generative AI model based on the click event information.

[0514] If it is selected as the best answer, the server saves the generated AI model's answer as the best answer in the database and displays it on the page.

[0515] If you add the answer to a page without marking it as the best answer, the server adds the answer from the generated AI model to the page, making it available for other users to view.

[0516] If you do not add an AI answer, the server will not add the generated AI model's answer to the page or database, but will leave it as is.

[0517] Step 14:

[0518] The server collects feedback from the questioner and uses it as learning data for the emotion engine.

[0519] Input: Feedback from the questioner.

[0520] Output: Improved recognition accuracy through emotion engine training.

[0521] Specific behavior:

[0522] The server provides an interface for collecting feedback data from questioners.

[0523] The server sends the collected feedback data to the emotion engine.

[0524] The emotion engine learns from feedback data to improve emotion recognition accuracy.

[0525] (Application example 2)

[0526] 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."

[0527] Conventional Q&A services respond slowly to questions, and the answers are often inadequate. Furthermore, answers are provided without considering the feelings of the questioner, resulting in low user satisfaction. Furthermore, answers are provided one-way, making it difficult to improve the service based on user feedback. A system that can solve these problems and provide fast and appropriate answers is needed.

[0528] 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.

[0529] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, means for processing the answer of the generative AI model based on the choice of the questioner, means for analyzing emotional information of the questioner using emotion recognition means and adjusting the answer of the generative AI model based on that information, means for collecting feedback from the questioner on whether the answer to the question was appropriate, and means for training the emotion recognition means to improve emotion recognition accuracy. This makes it possible to provide quick and appropriate answers to questions, as well as adaptive answers that take the questioner's emotions into consideration, thereby improving user satisfaction.

[0530] The "means for detecting that a question has been posted" is a function that, when a user enters a question into the input form and clicks the submit button, sends the question data to the server and recognizes that a new question has been posted.

[0531] The "means for sending a request to a generative AI model" is a function for sending a detected question to a generative AI model and sending a request to obtain an appropriate answer to the question.

[0532] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to a function that receives answers generated by the generative AI model and displays them on the question page in a format that is easy for users to view.

[0533] "Means for displaying multiple options to the questioner" refers to a function that displays a user interface that provides multiple options for the questioner to select in response to the answer of the generative AI model.

[0534] "Means for processing the generative AI model's answer based on the questioner's selection" refers to a function that optimally processes the generative AI model's answer based on the option selected by the questioner.

[0535] "Means for analyzing the emotional information of the questioner using emotion recognition means and adjusting the response of the generative AI model based on that information" refers to a function that analyzes the emotional state of the questioner and appropriately adjusts the content of the response in accordance with that emotion.

[0536] "Means for collecting feedback from the questioner on whether the answer to the question was appropriate" is a function that collects direct feedback from the questioner on whether the answer provided by the generative AI model was appropriate.

[0537] "Means for training the emotion recognition means to improve the emotion recognition accuracy" is a function that improves and trains the accuracy of the emotion recognition engine based on feedback from the questioner.

[0538] This invention is a system that provides prompt and appropriate answers in a Q&A service by combining a generative AI model and emotion recognition means. The configuration of this system and the program processing flow are described in detail below.

[0539] System configuration

[0540] The system consists of the following components:

[0541] 1. How to detect question posts

[0542] When a user enters a question into the input form and clicks the send button, the terminal sends the entered question data to the server.

[0543] 2. How to send requests to the generative AI model

[0544] The server sends the detected question to a generative AI model, which then generates an answer to that question.

[0545] 3. How to receive and display answers from the generative AI model

[0546] This is a function where the server receives the answer from the generative AI model and displays it on the question page.

[0547] 4. Emotion recognition means

[0548] This function analyzes the emotions of the questioner and adjusts the response content of the generative AI model based on that. This method analyzes the emotional state of the questioner and adjusts the response content to be appropriate for the questioner.

[0549] 5. A way to present multiple options to the questioner

[0550] This function displays a user interface to the questioner in response to the answer generated by the server using the AI ​​model, with options such as "I'm satisfied with this answer," "Please tell me more," or "The answer is inappropriate."

[0551] 6. Feedback Collection Methods

[0552] This is a method of collecting data to improve emotion recognition accuracy based on feedback from questioners, which enables the emotion recognition method to perform more accurate emotion analysis.

[0553] Program processing description

[0554] Question Post Detection

[0555] When a user enters a question into the Q&A service's input form and clicks the "Send" button, the device sends the entered question data to the server. The server saves the question in a database and detects that the question has been posted. At this time, the question data is sent to the emotion recognition means, which analyzes the questioner's emotions.

[0556] Sending requests to generative AI models

[0557] The server sends the detected question to the generative AI model, which generates an answer to the question and returns the answer to the server.

[0558] Receiving and adjusting answers from generative AI models

[0559] The server receives the answer from the generative AI model and adjusts it using emotion recognition to take into account the emotional state of the questioner. The adjusted answer is then displayed on the question page.

[0560] Displaying options to the questioner

[0561] The server displays the adjusted answer to the questioner along with the options of "I'm satisfied with this answer," "Please tell me more," or "The answer is not appropriate." When the questioner selects an option, the generative AI model's answer is processed accordingly.

[0562] Gathering feedback and learning

[0563] The feedback provided by the questioner is collected by the server, and this data is used as training data to improve the accuracy of the emotion recognition means.

[0564] Specific examples

[0565] 1. Simple Questions

[0566] A user posts a question such as "What is this product made of?" The server detects the question and sends it to the generative AI model. The answer from the generative AI model, "This product is made of cotton," is adjusted by the emotion recognition means and displayed on the question page. When the user selects "I'm satisfied with this answer," feedback is collected that the answer was appropriate.

[0567] 2. Complex Questions

[0568] A user posts a question such as "Please tell me the detailed manufacturing process of this product." The server detects the question and sends it to the generative AI model. The answer from the generative AI model, "This product is manufactured through multiple processes," is adjusted using emotion recognition and displayed on the question page. The user can select "Tell me more" to provide further details.

[0569] Prompt Sentence Examples

[0570] The user enters a question such as:

[0571] "What is this item made of?"

[0572] Example prompts to send to the generative AI model:

[0573] "The user is asking about the materials used in this product. The materials used in this product are as follows:"

[0574] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0575] Step 1:

[0576] A user enters a question into the Q&A service's input form and clicks the "Submit" button. The input is the question text entered by the user. The terminal sends this question data to the server. The output is the question data sent to the server.

[0577] Step 2:

[0578] The server receives new question data and stores it in a database. The input is the question data sent from the device. The server detects that a question has been posted and analyzes the questioner's emotional information using emotion recognition means. The output is the question data and the analyzed emotional information.

[0579] Step 3:

[0580] The server sends question data to the generative AI model. The input is the question and emotion information detected and analyzed by the server. The server sends a request to the generative AI model, which then generates an answer to the question. The output is the answer from the generative AI model.

[0581] Step 4:

[0582] The server receives the answer from the generative AI model and adjusts the answer using the emotion recognition means. The input is the answer from the generative AI model and pre-analyzed emotional information. The emotion recognition means adjusts the answer according to the questioner's emotions, and the server obtains the final adjusted answer. The output is the adjusted answer.

[0583] Step 5:

[0584] The server displays the adjusted answer on the question page. The input is the adjusted answer. The server displays this answer on the question page and generates a user interface with the options "I'm happy with this answer," "Please tell me more," and "This answer is not appropriate." The output is the answer and options that the user can view.

[0585] Step 6:

[0586] The user clicks on one of the options. The input is the option selected by the user. The terminal sends the selection to the server. The output is the selection sent to the server.

[0587] Step 7:

[0588] The server processes the generative AI model's answer based on the user's selection and stores it in a database if necessary. The inputs are the user's selection and the generative AI model's answer. Based on the selection, the server saves the answer as the "best answer," adds it to the page, or does nothing and exits. The output is the result saved in the database or an update to the page.

[0589] Step 8:

[0590] The server starts training the emotion recognition means based on feedback from the user. The input is the user's feedback data. The server uses this to train the emotion recognition means to improve its accuracy. The output is an emotion recognition means with improved accuracy.

[0591] 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.

[0592] 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.

[0593] 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.

[0594] [Second embodiment]

[0595] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0596] 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.

[0597] 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).

[0598] 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.

[0599] 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.

[0600] 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).

[0601] 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.

[0602] 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.

[0603] 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.

[0604] 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.

[0605] In the smart glasses 214, 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.

[0606] 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."

[0607] This invention is a system that uses a generative AI model to quickly provide answers to questions in a Q&A service. To implement this system, the following program processes and operations are required.

[0608] 1. Posting and discovering questions

[0609] A user enters a question on the Q&A service platform and clicks the "Post" button.

[0610] The terminal transmits the input question data to the server.

[0611] The server detects when a question is posted, analyzes the question data, and classifies it into the appropriate category.

[0612] 2. Display of "See AI's Answer" button

[0613] The server dynamically adds a "See AI's Answer" button to the question page.

[0614] The terminal redraws the updated page to the user, displaying the new button.

[0615] 3. Sending a request to the generative AI model

[0616] The user clicks the "See AI's answer" button.

[0617] The terminal detects this click event and sends the request to the server.

[0618] The server sends the question data as a request to the API of the generative AI model.

[0619] 4. Receiving and displaying answers from the generative AI model

[0620] The generative AI model analyzes the received question data and generates an appropriate answer.

[0621] The generative AI model returns the generated answer to the server.

[0622] The server formats the received answer data and adds it to the question page.

[0623] The device re-renders the updated page to the user, displaying the generative AI model's answer.

[0624] 5. Displaying options to the questioner

[0625] The server generates a UI that displays multiple options that the questioner can choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Do not add the AI ​​answer").

[0626] The device redraws its UI and displays it to the user.

[0627] 6. Processing Choice-Based Generative AI Model Answers

[0628] The user clicks on the appropriate option.

[0629] The terminal transmits the selection result to the server.

[0630] The server processes the generative AI model's answer based on the questioner's selection.

[0631] "Mark as Best Answer": Set the generative AI model's answer as the best answer in the database and display it on the page.

[0632] "Add to page without setting as best answer": The answer from the generative AI model is added to the database as an answer, and is displayed on the page without being set as the best answer.

[0633] "Don't add AI answer": The generated AI model's answer will not be added to the page or database.

[0634] Specific examples

[0635] Example 1: A simple question

[0636] A user posts a question to a Q&A service: "What is the appropriate amount of food for a cat?"

[0637] The server detects the question and displays a "See AI's Answer" button.

[0638] When the user clicks the button, the generative AI model generates and displays the answer: "For an adult cat, 2-3% of body weight is appropriate."

[0639] The user selects "make best answer" and the server sets the answer as the best answer.

[0640] Example 2: Advanced Questions

[0641] A user posts the question "What is the procedure for normalizing a wave function in quantum mechanics?"

[0642] The server detects the question and similarly displays a "See AI Answer" button.

[0643] When the user clicks the button, the generative AI model generates and displays the answer: "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1."

[0644] The user decides that the answer is insufficient and selects "Add to page without selecting as best answer." The server adds the generated AI model's answer to the page and expects other users to provide more detailed answers.

[0645] In this way, by utilizing a generative AI model, the system of the present invention allows questioners to receive quick and appropriate answers, allowing answerers to focus on asking more advanced questions.

[0646] The processing flow will be explained below.

[0647] Step 1:

[0648] The user enters a question into the input form on the Q&A service platform and clicks the "Post" button.

[0649] Step 2:

[0650] The terminal transmits the posted question data to the server.

[0651] Step 3:

[0652] The server stores the received question data in a database and detects that a question has been posted.

[0653] Step 4:

[0654] The server generates a script that dynamically adds a "View AI Answer" button to the question page.

[0655] Step 5:

[0656] The device receives the script sent from the server, redraws the question page, and displays the "See AI's Answer" button.

[0657] Step 6:

[0658] The user clicks the "See AI's answer" button.

[0659] Step 7:

[0660] The terminal detects the button click event and sends the request to the server.

[0661] Step 8:

[0662] The server sends the question data as a request to the API of the generative AI model.

[0663] Step 9:

[0664] A generative AI model analyzes the question data and generates appropriate answers.

[0665] Step 10:

[0666] The generative AI model returns the generated answer to the server.

[0667] Step 11:

[0668] The server receives the generated response data and formats it appropriately.

[0669] Step 12:

[0670] The server adds HTML containing the formatted answer to the question page.

[0671] Step 13:

[0672] The device renders the HTML received from the server and redraws the question page to display the answer from the generative AI model.

[0673] Step 14:

[0674] The server generates a UI to display multiple options for the answer of the generated AI model ("Make it the best answer," "Don't make it the best answer but add it to the page," "Don't add the AI ​​answer").

[0675] Step 15:

[0676] The device renders the UI with the options and redraws the question page to display to the user.

[0677] Step 16:

[0678] The user clicks on one of the choices.

[0679] Step 17:

[0680] The terminal detects the selection click event and sends the result to the server.

[0681] Step 18:

[0682] The server processes the answer of the generative AI model based on the selection results.

[0683] Step 19:

[0684] If you select "Set as best answer," the server will set the generated AI model's answer as the best answer in the database and display it on the question page.

[0685] Step 20:

[0686] If you select "Add to page without setting as best answer," the server will add the generated AI model's answer to the database and display it on the question page, but will not set it as the best answer.

[0687] Step 21:

[0688] If you select "Do not add AI answer," the server will not add the generated AI model's answer to the database or the question page, and will return the result to the device.

[0689] Step 22:

[0690] The terminal redraws the question page based on the results received from the server and displays the final state to the user.

[0691] The above is the specific program processing flow of this system.

[0692] Example 1

[0693] 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."

[0694] Conventional Q&A services have the problem that it takes a long time for questioners to receive an appropriate answer. Furthermore, because the quality and speed of the answer depends on the answerer, it is difficult to obtain an appropriate answer immediately to the question. Furthermore, the system's efficiency is low because question categorization and responses are done manually. Furthermore, conventional systems have limited options for automated answers using AI, leaving users with little flexibility in selecting AI answers.

[0695] 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.

[0696] In this invention, the server includes means for detecting that a question has been posted, means for analyzing the question data and classifying it into an appropriate category, means for dynamically adding a "See AI's Answer" button to the question page, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, and means for processing the answer of the generative AI model based on the choice made by the questioner. This allows the questioner to quickly obtain an appropriate answer, while allowing the answerer to focus on more advanced questions, improving the efficiency of the entire system.

[0697] The "means for detecting that a question has been posted" refers to a method or device for detecting that a user has entered a question on the Q&A service platform and clicked the post button.

[0698] "Means for analyzing question data and classifying it into appropriate categories" refers to a method or device that analyzes input question data using natural language processing (NLP) or the like, and classifies it into relevant categories based on its content.

[0699] The "means for dynamically adding a 'See AI's Answer' button to a question page" refers to a method or device for adding and displaying a 'See AI's Answer' button on a question page in real time after a question is posted.

[0700] "Means for sending a request to the generative AI model" refers to a method or device that sends the question data as a request to the API of the generative AI model when a user clicks the "View AI's answer" button.

[0701] "Means for receiving an answer from a generative AI model and displaying it on a question page" refers to a method or device for receiving an answer returned from a generative AI model and displaying it on a question page in a form that can be viewed by a user.

[0702] "Means for displaying multiple options to the questioner" refers to a method or device that displays options for the questioner in response to the answer of the generative AI model, such as "Make this the best answer," "Add this to the page without making it the best answer," or "Do not add this to the AI ​​answer."

[0703] "Means for processing the answer of a generative AI model based on the questioner's selection" refers to a method or device that processes the answer of a generative AI model in accordance with the questioner's selection from the options presented, such as saving it in a database, displaying it, or discarding it.

[0704] This invention is a system that uses a generative AI model to provide quick answers to questions in a Q&A service. This system is realized by combining various means that operate mainly on a server, a terminal, and a user.

[0705] First, a user enters a question on the Q&A service platform and clicks the "Post" button. The device detects this action and sends the entered question data to the server. The server analyzes the received question data and detects that it is a question. The server then analyzes the question data using a natural language processing (NLP) library (e.g., spaCy) and classifies it into the appropriate category.

[0706] Next, the server generates HTML code to dynamically add a "See AI's Answer" button to the question page and sends it to the terminal. The terminal receives this code and redraws the user's browser screen to display the "See AI's Answer" button. When the user clicks this button, the terminal detects the click event and sends the request to the server.

[0707] Based on the received request, the server sends question data as a request to the generative AI model's API (e.g., OpenAI). The generative AI model analyzes the sent question data and generates an answer. The generative AI model returns the generated answer to the server. The server formats the received answer data, generates HTML code to add to the question page, and sends it to the device. The device receives this updated data and displays the generative AI model's answer by redrawing the user's browser screen.

[0708] Furthermore, the server generates a UI to display multiple options that the user can select from for the generated AI model's answer (e.g., "Make this the best answer," "Add to the page without making this the best answer," and "Don't add the AI ​​answer"). The device redraws this UI and displays it on the page. When the user clicks on an appropriate option, the device sends this selection to the server. The server processes the generated AI model's answer based on the user's selection, for example, saving it as the best answer in a database or updating the page.

[0709] Specific examples

[0710] Prompt statement example 1:

[0711] A user posts a question to a Q&A service: "What is the appropriate amount of food for a cat?" The server detects the question and displays a "See AI's Answer" button. When the user clicks the button, the generative AI model generates and displays the answer: "For an adult cat, 2-3% of its body weight is appropriate." The user selects "Make this the best answer," and the server adds the answer to the database as the best answer.

[0712] Prompt statement example 2:

[0713] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?" The server detects the question and similarly displays a "See AI's answer" button. When the user clicks the button, the generative AI model generates an answer: "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1," and displays it. The user decides that the answer is insufficient and selects "Add to page without marking as best answer." The server adds the generative AI model's answer to the page, and expects other users to provide more detailed answers.

[0714] In this way, the system utilizes generative AI models to enable users to get fast and relevant answers, allowing other users to focus on more advanced questions, improving the overall efficiency of the system and providing a better user experience by providing options for questioners.

[0715] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0716] Step 1: Post a question

[0717] Users enter their questions on the Q&A service platform and click the "Post" button.

[0718] Input: Text data of the user's question (e.g., "What is the appropriate amount of food for a cat?")

[0719] The terminal receives user input, converts the question data into JSON format, and sends it to the server.

[0720] Output: Question data in JSON format

[0721] Step 2: Question detection and parsing

[0722] The server receives the JSON data and detects that a question has been posted.

[0723] Input: JSON format question data sent from the terminal

[0724] The server uses an NLP library (e.g., spaCy) to parse the question data and classify it into appropriate categories.

[0725] Output: Question analysis results and category information (e.g., "Pet category")

[0726] Step 3: Display of the "See AI's Answer" button

[0727] The server generates HTML code to dynamically add a "View AI Answer" button to the question page.

[0728] Input: User question, analysis results, category information

[0729] The server sends a response containing the generated HTML code to the terminal.

[0730] The device uses the updated data received from the server to redraw the user's browser screen, which displays a new "See AI's Answer" button.

[0731] Output: Updated HTML code

[0732] Step 4: Sending a request to the generative AI model

[0733] The user clicks the "See AI's answer" button.

[0734] Input: User click event

[0735] The terminal detects the click event and sends a request along with the question data to the server.

[0736] Output: Request data to the server

[0737] The server sends the question data as a request to the API of the generative AI model (e.g., OpenAI).

[0738] Step 5: Receive and display the answer from the generative AI model

[0739] The server receives the response from the generative AI model.

[0740] Input: Answer data from the generative AI model

[0741] The generative AI model analyzes the question data and generates an appropriate answer (e.g., "For adult cats, 2-3% of body weight is appropriate").

[0742] The server formats the generated answer and generates HTML code to add to the question page.

[0743] Output: Updated HTML code and generated AI answer data

[0744] The device receives the updated page data and redraws the user's browser screen, which displays the answer from the generated AI model.

[0745] Step 6: Displaying the options

[0746] The server generates a UI to display multiple options for the user to choose from in response to the generated AI model's answer.

[0747] Input: Answer data for the generative AI model

[0748] The server transmits the generated UI data to the terminal.

[0749] The device uses the UI data to display the options to the user.

[0750] Output: Updated UI with choices

[0751] Step 7: Processing based on user selection

[0752] The user clicks on the appropriate option.

[0753] Input: User click event

[0754] The terminal transmits the selection result to the server.

[0755] Output: Request data to the server

[0756] The server processes the generative AI model's answer based on the questioner's selection.

[0757] "Mark as Best Answer": Set the generative AI model's answer as the best answer in the database and display it on the page.

[0758] "Add to page without setting as best answer": Add the generative AI model's answer and display it on the page without setting it as the best answer.

[0759] "Don't add AI answer": The generated AI model's answer will not be added to the page or database.

[0760] The above are the specific processing steps and operations of the program for this system.

[0761] (Application example 1)

[0762] 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."

[0763] In recent years, with the spread of electronic payment services, the number of inquiries from users has increased dramatically. This has resulted in delayed support responses, leading to problems that lead to a decline in user satisfaction. Furthermore, support staff are required to respond to a large number of inquiries, increasing their workload. To solve this problem, a system that can provide prompt and appropriate responses is needed. However, existing systems have difficulty in providing prompt and appropriate responses to inquiries.

[0764] 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.

[0765] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, means for processing the answer from the generative AI model based on the questioner's selection, means for providing a user interface for inputting questions about electronic payments, and means for receiving the answer from the generative AI model and displaying it on the question page related to electronic payments. This allows users to receive quick and accurate answers to inquiries about electronic payments, improving the efficiency of support responses and reducing workloads.

[0766] The "means for detecting that a question has been posted" is a mechanism by which the system determines that a user has entered and posted a question on the system.

[0767] The "means for sending a request to the generative AI model" is a mechanism for sending the posted question to the generative AI model and sending a request to generate an appropriate answer.

[0768] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to a mechanism for receiving answers returned from the generative AI model and displaying them on the user interface.

[0769] "Means for displaying multiple options to the questioner" refers to a mechanism that displays multiple options so that the user can select further actions in response to the answer from the generative AI model.

[0770] The "means for processing the generative AI model's answer based on the questioner's selection" refers to a mechanism that, after the user selects an option, stores or displays the generative AI model's answer in a database based on that selection.

[0771] The "means for providing a user interface for inputting questions about electronic payments" is a mechanism for providing an interface designed to allow users to intuitively input questions about electronic payments.

[0772] "Means for receiving the answer of the generative AI model and displaying it on the question page related to electronic payments" refers to a mechanism for receiving the answer related to electronic payments returned from the generative AI model and displaying it on the relevant question page.

[0773] This invention is a system for quickly and accurately answering questions about electronic payments. The system consists of posting and detecting questions, sending requests to a generative AI model, receiving and displaying answers, presenting options to the user, and processing the answers based on the selections.

[0774] The core of the system resides in the server, which has the following main functions:

[0775] 1. Post and discover questions:

[0776] When a user enters a question about electronic payments on the smartphone application and clicks the "Post" button, the device sends the question data to the server, which detects the post, analyzes the received question data, and classifies it into the appropriate category.

[0777] 2. Display of "See AI's Answer" button:

[0778] When a question is submitted, the server dynamically adds a "See AI's Answer" button to the question page. After this button is added to the page, the device redraws the page and displays the new button to the user.

[0779] 3. Sending a request to the generative AI model:

[0780] When a user clicks the "View AI's Answer" button, the click event is detected by the device and a request is sent to the server. The server then sends the received question data as a request to the API of the generative AI model.

[0781] 4. Receive and display answers from the generative AI model:

[0782] The generative AI model analyzes the received question data and generates an appropriate answer. The generated answer is returned to the server, which formats the answer data and adds it to the question page. The device redraws the updated page and displays the generative AI model's answer to the user.

[0783] 5. Displaying options to the questioner:

[0784] The server generates a UI to display multiple options for the questioner to choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Don't add the AI ​​answer"), and the device redraws the UI to display the options to the user.

[0785] 6. Processing the generative AI model's answers based on your choices:

[0786] When the user clicks on the appropriate option, the device sends the selection to the server, which then takes appropriate action on the generative AI model's answer based on the user's selection, such as setting it as the best answer, adding it to the page, or ignoring it.

[0787] The following hardware and software are used to realize this system.

[0788] 1. Smartphone application:

[0789] Platform: iOS and Android

[0790] Development language: Swift (iOS), Kotlin (Android)

[0791] 2. Server:

[0792] Platform: Linux-based server environment

[0793] Software used: Node.js, Django, Flask

[0794] 3. Generative AI Model:

[0795] Model Name: OpenAI GPT-3 or GPT-4

[0796] API used: OpenAI API

[0797] As a concrete example, if a question is posted asking "Where can I check my credit card statement?", the system will operate as follows:

[0798] 1. The server receives the question and adds a "See AI's Answer" button to the question page.

[0799] 2. When the user clicks this button, the question data is sent to the generative AI model.

[0800] 3. The generative AI model generates the answer, "You can check your credit card statement on the 'Statement' page of the official app or website," and returns it to the server.

[0801] 4. The server displays this answer on the question page and gives the user the options of "Make it the best answer," "Add it to the page without making it the best answer," or "Don't add the AI ​​answer."

[0802] 5. If the user selects "Mark as Best Answer," the answer is saved in the database and set as the best answer on the question page.

[0803] This system will enable users to quickly obtain information about electronic payments and is expected to reduce the burden on support staff.

[0804] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0805] Step 1: Posting and discovering questions

[0806] A user enters a question about electronic payment into a smartphone application and clicks the "Post" button. At this time, the question text entered by the user is generated. The device sends this question text to the server. The server analyzes the received question data and detects that a question has been posted. Next, it classifies the question data into appropriate categories. This data processing process organizes the question content into categories.

[0807] Input: The question text entered by the user

[0808] Data processing: categorization of question data

[0809] Output: Question data with categories

[0810] Step 2: Displaying the "See AI's Answer" button

[0811] The server dynamically adds a "See AI's Answer" button to the question page based on the question data it received. It then instructs the device to redraw the updated page and display the new button to the user. This operation allows the user to expect a quick answer from the AI.

[0812] Input: Question data with categories

[0813] Data processing: Update the question page and add a button

[0814] Output: Updated question page with a "See AI's answer" button

[0815] Step 3: Sending a request to the generative AI model

[0816] When the user clicks the "View AI Answer" button, the device detects this click event. Based on this detection, the device resends the question data to the server. The server then sends the received question data as a request to the generative AI model's API. This data transmission causes the generative AI model to begin analyzing the question content.

[0817] Input: Click event of the "View AI's answer" button and question data

[0818] Data processing: Sending a request to the generative AI model API

[0819] Output: The question request received by the generative AI model

[0820] Step 4: Receive and display the answer from the generative AI model

[0821] The generative AI model analyzes the received question data and generates an appropriate answer. This answer is returned to the server. The server formats the received answer data and adds it to the question page. The device then redraws the updated page and displays the generative AI model's answer, allowing the user to see the answer provided by the AI ​​model.

[0822] Input: Answer data generated by a generative AI model

[0823] Data processing: Formatting of response data

[0824] Output: Question page with formatted answer data

[0825] Step 5: Present the options to the questioner

[0826] The server generates a UI to display multiple options for the questioner to choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Don't add the AI ​​answer") The device redraws this UI and displays it to the user.

[0827] Input: Answer data for the generative AI model

[0828] Data processing: generating options and updating the UI

[0829] Output: Question page with updated UI

[0830] Step 6: Processing the generative AI model's answer based on your selection

[0831] When the user clicks on the appropriate option, the device detects this click event and sends the selection result to the server. Based on the selection, the server saves the generative AI model's answer in a database or displays it on the page. For example, if the user selects "Set as best answer," the answer is set as the best answer in the database and displayed on the question page.

[0832] Input: User selection

[0833] Data processing: Save answers to database and update page display

[0834] Output: Updated question page and database state

[0835] 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.

[0836] This invention is a system that provides quick and appropriate answers to questions in a Q&A service using a generative AI model and an emotion engine. The configuration of this system and the program processing flow are described in detail below.

[0837] System configuration

[0838] The system mainly consists of the following components:

[0839] 1. Question Post Detection

[0840] The server has a means to detect new questions posted by users.

[0841] 2. Sending a request to the generative AI model

[0842] The server has the ability to send the detected questions to the generative AI model.

[0843] 3. Receiving and displaying answers from the generative AI model

[0844] The server has a means to receive the answer from the generative AI model and display it on the question page.

[0845] 4. Emotion Engine

[0846] The server has the means to recognize the questioner's emotions and analyze that information.

[0847] The emotion engine adjusts the generative AI model's responses based on the questioner's emotional state.

[0848] 5. Display of questioner's options

[0849] The server has a means to display a UI that offers the questioner multiple options for the answer of the generated AI model.

[0850] 6. Processing Choice-Based Answers

[0851] The server has a means to process the generative AI model's answer based on the questioner's selection.

[0852] The server has the function of storing the emotion information recognized by the emotion engine in a database.

[0853] 7. Learning the Emotion Engine

[0854] The server has a learning mechanism to improve emotion recognition accuracy based on feedback from the questioner.

[0855] Program flow

[0856] Posting and discovering questions

[0857] A user enters a question into the Q&A service's input form and clicks the "Post" button. The device sends the entered question data to the server. The server saves the question in a database, detects that the question has been posted, and analyzes the questioner's emotional information using an emotion engine.

[0858] Display of "See AI's answer" button

[0859] The server generates a script that dynamically adds a "See AI's Answer" button to the question page. The device receives this script, redraws the question page, and displays the "See AI's Answer" button.

[0860] Linking generative AI models and emotion engines

[0861] The user clicks the "See AI's Answer" button. The device detects the click event and sends the request to the server. The server sends the question data to the generative AI model, which then generates an answer.

[0862] The emotion engine adjusts the generated answer based on the emotional state of the questioner. The server receives the adjusted answer and displays it on the question page.

[0863] Displaying options to the questioner

[0864] The server generates a UI for the questioner to choose from three options for the AI ​​model's answer: "Make it the best answer," "Add it to the page without making it the best answer," or "Don't add it to the AI ​​answer." The device redraws the page and displays the options to the questioner.

[0865] Processing questioner selections and answers

[0866] The user clicks on one of the options. The device sends the selection to the server, which processes the answer of the generative AI model based on the selection.

[0867] Mark as best answer: The server saves the generated AI model's answer in the database as the best answer and displays it on the page.

[0868] Add to page without marking as best answer: The server adds the generated AI model's answer to the page, making it visible to other users.

[0869] Do not add AI answer: The server does not add the generated AI model answer to the page or database, but leaves it as is.

[0870] Emotion Engine Learning

[0871] Based on the feedback from the questioner, the server performs a learning process to improve the emotion recognition accuracy of the emotion engine.

[0872] Specific examples

[0873] Example 1: A simple question

[0874] A user posts a question: "What is the appropriate amount of food for a cat?" The server detects the question and displays a "See AI's Answer" button. When the user clicks the button, the generative AI model generates the answer: "For an adult cat, 2-3% of its body weight is appropriate," and the emotion engine appropriately adjusts the answer based on the questioner's emotional information and displays it. The user selects "Make Best Answer," and the server sets the answer as the best answer.

[0875] Example 2: A more complex question

[0876] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?" The server detects the question and displays a "See AI's answer" button. When the user clicks the button, the generative AI model generates the answer: "The procedure is to integrate the wave function over the entire space and make the square of the absolute value of that result 1." The emotion engine analyzes the questioner's emotions and adjusts the answer. The user selects "Add to page without marking as best answer," and the server adds the generative AI model's answer to the page.

[0877] In this way, the system of the present invention uses a generative AI model and an emotion engine in combination, allowing the questioner to receive a quick and appropriate answer while allowing the answerer to focus on asking more advanced questions.

[0878] The processing flow will be explained below.

[0879] Step 1:

[0880] The user enters a question into the Q&A service's input form and clicks the "Post" button.

[0881] Step 2:

[0882] The terminal transmits the posted question data to the server.

[0883] Step 3:

[0884] The server stores the received question data in a database and detects that a question has been posted.

[0885] Step 4:

[0886] The server calls the emotion engine and analyzes the questioner's emotional information based on the question data.

[0887] Step 5:

[0888] The server stores the emotion analysis results in a database.

[0889] Step 6:

[0890] The server generates a script that dynamically adds a "View AI Answer" button to the question page.

[0891] Step 7:

[0892] The device receives the script sent from the server, redraws the question page, and displays the "See AI's Answer" button.

[0893] Step 8:

[0894] The user clicks the "See AI's answer" button.

[0895] Step 9:

[0896] The terminal detects the button click event and sends the request to the server.

[0897] Step 10:

[0898] The server sends the question data and emotion information from the emotion engine as a request to the API of the generative AI model.

[0899] Step 11:

[0900] The generative AI model analyzes question data and emotional information to generate appropriate answers.

[0901] Step 12:

[0902] The generative AI model returns the generated answer to the server.

[0903] Step 13:

[0904] The server receives the generated response data and formats it appropriately.

[0905] Step 14:

[0906] The server adds HTML containing the formatted answer to the question page.

[0907] Step 15:

[0908] The device renders the HTML received from the server and redraws the question page to display the answer from the generative AI model.

[0909] Step 16:

[0910] The server generates a UI that provides multiple options for the answer of the generated AI model.

[0911] Step 17:

[0912] The device renders the UI with the options and redraws the question page to display to the user.

[0913] Step 18:

[0914] The user can choose to either "Make this the best answer," "Add it to the page without making it the best answer," or "Do not add the AI ​​answer."

[0915] Step 19:

[0916] The terminal transmits the user's selection to the server.

[0917] Step 20:

[0918] The server processes the answer of the generative AI model based on the selection results.

[0919] Step 21:

[0920] If you select "Set as best answer," the server will set the generated AI model's answer as the best answer in the database and display it on the question page.

[0921] Step 22:

[0922] If you select "Add to page without setting as best answer," the server will add the generated AI model's answer to the database and display it on the question page, but will not set it as the best answer.

[0923] Step 23:

[0924] If you select "Do not add AI answer", the server will not add the generated AI model's answer to the database or the question page.

[0925] Step 24:

[0926] The server sends feedback from the questioner to the emotion engine, which then learns to improve emotion recognition accuracy.

[0927] Step 25:

[0928] The terminal redraws the question page based on the final result received from the server and displays it to the user.

[0929] The above is the specific program processing flow of this system.

[0930] Example 2

[0931] 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."

[0932] With conventional Q&A services, it was difficult for questioners to obtain prompt and appropriate answers. Furthermore, the answers generated by the AI ​​model were not properly adjusted according to the questioner's feelings, resulting in low satisfaction with the answer. Furthermore, there was a lack of a way to clearly indicate which option the questioner should take in response to the question posted, which did not improve the user experience.

[0933] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0934] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for analyzing the emotion of the questioner, means for adjusting the content of the answer of the generative AI model based on the analyzed emotion, means for displaying multiple options to the questioner, means for processing the answer of the generative AI model based on the selection of the questioner, and means for storing the emotion information in a database. This allows the questioner to receive a quick and appropriate answer, and further enables the answer to be adjusted according to the questioner's emotion, thereby improving the user experience.

[0935] "Question detection" is the process by which the server automatically recognizes questions posted by users.

[0936] "Sending a request to the generative AI model" is a means of sending information from the server to the generative AI model to analyze the detected question.

[0937] "Receiving and displaying answers from the generative AI model" refers to the process in which the server receives the answers provided by the generative AI model and displays them on the question page.

[0938] "Emotion analysis" is a means of determining the questioner's emotional state using an emotion engine.

[0939] "Adjusting response content based on emotions" is a method of appropriately changing the response content provided by the generative AI model based on analyzed emotional information.

[0940] "Displaying options" is the process by which the server provides a user interface with multiple options that the questioner can select from in response to the answer provided by the generative AI model.

[0941] "Choice-based answer processing" is the process by which the server appropriately processes the answer of the generative AI model depending on the option selected by the questioner.

[0942] "Storing emotional information" is a method of storing the emotional information analyzed by the emotion engine in a database.

[0943] This invention is a system that provides quick and appropriate answers to questions in a Q&A service using a generative AI model and an emotion engine. The configuration of this system and the program processing flow are described in detail below.

[0944] System configuration

[0945] The system consists of the following main components:

[0946] 1. Server

[0947] The server has a means for receiving and detecting queries from users.

[0948] It has the ability to send question data to a generative AI model.

[0949] It has a means to receive answers from the generative AI model and display them on the question page.

[0950] It integrates an emotion engine to analyze the questioner's emotions.

[0951] It has the ability to adjust the answers of the generative AI model based on the analyzed emotional information.

[0952] It has a means to display multiple options in the user interface.

[0953] It has a means to process the generative AI model's answers based on the choices.

[0954] It has the function of storing emotional information in a database.

[0955] 2. Terminal

[0956] The terminal provides an interface where the user can enter a question and click a "Post" button.

[0957] It has a means for sending question data to the server.

[0958] It has the means to receive and execute a script to dynamically display a "View AI Answer" button on the question page.

[0959] It has the function of sending the selection results to the server.

[0960] 3. Users

[0961] Users post questions using the input form of the Q&A service.

[0962] Click the "See AI's Answer" button to see the answer from the generated AI model.

[0963] Make the appropriate choice from multiple options.

[0964] Specific examples

[0965] Example 1: A simple question

[0966] Example prompt sentence:

[0967] What is the appropriate amount of cat food?

[0968] A user posts a question: "What is the right amount of food for a cat?"

[0969] The server detects the question and analyzes the questioner's emotional information using an emotion engine.

[0970] The server sends a script to the terminal that displays a "See AI's answer" button.

[0971] When the user clicks the "See AI Answer" button, the server sends the question data to the generative AI model and receives the answer.

[0972] For example, the generative AI model generates an answer such as "2-3% of body weight is appropriate for an adult cat," which is received by the server.

[0973] The emotion engine takes into account the emotional state of the questioner and adjusts the answer accordingly.

[0974] The server displays the adjusted answer on the question page.

[0975] The user selects "Mark as Best Answer" and the server saves the answer as the best answer in the database.

[0976] Example 2: Complex Question

[0977] Example prompt sentence:

[0978] "What is the procedure for normalizing wave functions in quantum mechanics?"

[0979] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?"

[0980] The server detects the question and analyzes the questioner's emotional information using an emotion engine.

[0981] The server sends a script to the terminal that displays a "See AI's answer" button.

[0982] When the user clicks the "See AI Answer" button, the server sends the question data to the generative AI model and receives the answer.

[0983] For example, the generative AI model generates an answer such as, "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1," and the server receives this.

[0984] The emotion engine analyzes the questioner's emotions and adjusts the response appropriately.

[0985] The server displays the adjusted answer on the question page.

[0986] The user selects "Add to page without marking as best answer" and the server adds the answer to the page.

[0987] In this way, the system of the present invention can use generative AI models and emotion engines to provide users with fast and relevant answers.

[0988] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0989] Specific explanation of program processing

[0990] Step 1:

[0991] The user enters a question into the Q&A service's input form and clicks the "Post" button.

[0992] Input: The question text entered by the user.

[0993] Output: The query data is sent to the server.

[0994] Specific behavior:

[0995] The terminal obtains the contents of the user input form.

[0996] The terminal generates a request to transmit the question data to the server.

[0997] The device sends a request to the server.

[0998] Step 2:

[0999] The server stores the received question data in a database and detects when a new question has been posted.

[1000] The server sends the question to the emotion engine and analyzes the questioner's emotional information.

[1001] Input: Question data sent from the terminal.

[1002] Output: The question data is stored in a database and the sentiment analysis results are obtained.

[1003] Specific behavior:

[1004] The server writes the received question data to the database.

[1005] The server monitors the database to detect new question submissions.

[1006] The server sends the question to the emotion engine and receives the analysis results.

[1007] Step 3:

[1008] The server generates a script that adds a "See AI's Answer" button to the question page and sends it to the terminal.

[1009] Input: The event that a new question is posted.

[1010] Output: The script is sent to the terminal and the button is displayed.

[1011] Specific behavior:

[1012] The server generates script code to add a "See AI's Answer" button.

[1013] The server sends the generated script to the terminal.

[1014] Step 4:

[1015] The device receives the script, redraws the question page, and displays the "See AI's Answer" button.

[1016] Input: The script sent by the server.

[1017] Output: A "See AI's answer" button will be displayed on the question page.

[1018] Specific behavior:

[1019] The terminal parses the script received from the server.

[1020] The terminal executes the received script and redraws the question page.

[1021] The device will display a "See AI's Answer" button.

[1022] Step 5:

[1023] The user clicks the "See AI's answer" button.

[1024] Input: User click event.

[1025] Output: A click event is sent to the server.

[1026] Specific behavior:

[1027] The device detects a click event.

[1028] The device sends the click event information to the server.

[1029] Step 6:

[1030] The server sends the question data to the generative AI model, which then generates an answer.

[1031] Input: Click event information and original question data.

[1032] Output: Answer data from the generative AI model.

[1033] Specific behavior:

[1034] The server creates a request to send the original question data to the generative AI model.

[1035] The server calls the API of the generated AI model and sends the question data.

[1036] The generative AI model analyzes the question and generates an appropriate answer.

[1037] The answer data generated by the generative AI model is sent back to the server.

[1038] Step 7:

[1039] The emotion engine takes into account the emotional state of the questioner and adjusts the generated answer content.

[1040] Input: Response data from the generative AI model and pre-obtained sentiment analysis results.

[1041] Output: Adjusted response data.

[1042] Specific behavior:

[1043] The server sends the response data from the generative AI model and the emotion analysis results to the emotion engine.

[1044] The emotion engine adjusts the response content.

[1045] The emotion engine sends the adjusted response data back to the server.

[1046] Step 8:

[1047] The server receives the adjusted answer and displays it on the question page.

[1048] Input: Calibrated response data from the sentiment engine.

[1049] Output: The answer displayed on the question page.

[1050] Specific behavior:

[1051] The server receives the adjusted response data.

[1052] The server generates data for updating the question page and sends it to the terminal.

[1053] Step 9:

[1054] The device will redraw the question page and display the adjusted answer.

[1055] Input: Data to update from the server.

[1056] Output: Redrawn question page and displayed answer.

[1057] Specific behavior:

[1058] The terminal analyzes the data received from the server and redraws the question page.

[1059] The device will display the adjusted answer.

[1060] Step 10:

[1061] The server generates a UI containing three options that the questioner can choose from in response to the generated AI model's answer: "Make this the best answer," "Add it to the page without making it the best answer," or "Do not add the AI ​​answer," and sends this UI to the device.

[1062] Input: Adjusted response data.

[1063] Output: The UI shown to the questioner.

[1064] Specific behavior:

[1065] Create a script that the server uses to generate a UI with the options.

[1066] The server sends the generated script to the terminal.

[1067] Step 11:

[1068] The terminal will redraw and show the options to the questioner.

[1069] Input: Script from server.

[1070] Output: The choices presented to the questioner.

[1071] Specific behavior:

[1072] The terminal parses the script received from the server.

[1073] The device will redraw the question page and display the options.

[1074] Step 12:

[1075] The user clicks on one of the displayed options.

[1076] Input: User click event.

[1077] Output: Click event information is sent to the server.

[1078] Specific behavior:

[1079] The device detects a click event.

[1080] The device sends the click event information to the server.

[1081] Step 13:

[1082] The server processes the generative AI model's answer based on the selection results.

[1083] Input: Click event information.

[1084] Output: Processed response data.

[1085] Specific behavior:

[1086] The server appropriately processes the answer of the generative AI model based on the click event information.

[1087] If it is selected as the best answer, the server saves the generated AI model's answer as the best answer in the database and displays it on the page.

[1088] If you add the answer to a page without marking it as the best answer, the server adds the answer from the generated AI model to the page, making it available for other users to view.

[1089] If you do not add an AI answer, the server will not add the generated AI model's answer to the page or database, but will leave it as is.

[1090] Step 14:

[1091] The server collects feedback from the questioner and uses it as learning data for the emotion engine.

[1092] Input: Feedback from the questioner.

[1093] Output: Improved recognition accuracy through emotion engine training.

[1094] Specific behavior:

[1095] The server provides an interface for collecting feedback data from questioners.

[1096] The server sends the collected feedback data to the emotion engine.

[1097] The emotion engine learns from feedback data to improve emotion recognition accuracy.

[1098] (Application example 2)

[1099] 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."

[1100] Conventional Q&A services respond slowly to questions, and the answers are often inadequate. Furthermore, answers are provided without considering the feelings of the questioner, resulting in low user satisfaction. Furthermore, answers are provided one-way, making it difficult to improve the service based on user feedback. A system that can solve these problems and provide fast and appropriate answers is needed.

[1101] 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.

[1102] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, means for processing the answer of the generative AI model based on the choice of the questioner, means for analyzing emotional information of the questioner using emotion recognition means and adjusting the answer of the generative AI model based on that information, means for collecting feedback from the questioner on whether the answer to the question was appropriate, and means for training the emotion recognition means to improve emotion recognition accuracy. This makes it possible to provide quick and appropriate answers to questions, as well as adaptive answers that take the questioner's emotions into consideration, thereby improving user satisfaction.

[1103] The "means for detecting that a question has been posted" is a function that, when a user enters a question into the input form and clicks the submit button, sends the question data to the server and recognizes that a new question has been posted.

[1104] The "means for sending a request to a generative AI model" is a function for sending a detected question to a generative AI model and sending a request to obtain an appropriate answer to the question.

[1105] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to a function that receives answers generated by the generative AI model and displays them on the question page in a format that is easy for users to view.

[1106] "Means for displaying multiple options to the questioner" refers to a function that displays a user interface that provides multiple options for the questioner to select in response to the answer of the generative AI model.

[1107] "Means for processing the generative AI model's answer based on the questioner's selection" refers to a function that optimally processes the generative AI model's answer based on the option selected by the questioner.

[1108] "Means for analyzing the emotional information of the questioner using emotion recognition means and adjusting the response of the generative AI model based on that information" refers to a function that analyzes the emotional state of the questioner and appropriately adjusts the content of the response in accordance with that emotion.

[1109] "Means for collecting feedback from the questioner on whether the answer to the question was appropriate" is a function that collects direct feedback from the questioner on whether the answer provided by the generative AI model was appropriate.

[1110] "Means for training the emotion recognition means to improve the emotion recognition accuracy" is a function that improves and trains the accuracy of the emotion recognition engine based on feedback from the questioner.

[1111] This invention is a system that provides prompt and appropriate answers in a Q&A service by combining a generative AI model and emotion recognition means. The configuration of this system and the program processing flow are described in detail below.

[1112] System configuration

[1113] The system consists of the following components:

[1114] 1. How to detect question posts

[1115] When a user enters a question into the input form and clicks the send button, the terminal sends the entered question data to the server.

[1116] 2. How to send requests to the generative AI model

[1117] The server sends the detected question to a generative AI model, which then generates an answer to that question.

[1118] 3. How to receive and display answers from the generative AI model

[1119] This is a function where the server receives the answer from the generative AI model and displays it on the question page.

[1120] 4. Emotion recognition means

[1121] This function analyzes the emotions of the questioner and adjusts the response content of the generative AI model based on that. This method analyzes the emotional state of the questioner and adjusts the response content to be appropriate for the questioner.

[1122] 5. A way to present multiple options to the questioner

[1123] This function displays a user interface to the questioner in response to the answer generated by the server using the AI ​​model, with options such as "I'm satisfied with this answer," "Please tell me more," or "The answer is inappropriate."

[1124] 6. Feedback Collection Methods

[1125] This is a method of collecting data to improve emotion recognition accuracy based on feedback from questioners, which enables the emotion recognition method to perform more accurate emotion analysis.

[1126] Program processing description

[1127] Question Post Detection

[1128] When a user enters a question into the Q&A service's input form and clicks the "Send" button, the device sends the entered question data to the server. The server saves the question in a database and detects that the question has been posted. At this time, the question data is sent to the emotion recognition means, which analyzes the questioner's emotions.

[1129] Sending requests to generative AI models

[1130] The server sends the detected question to the generative AI model, which generates an answer to the question and returns the answer to the server.

[1131] Receiving and adjusting answers from generative AI models

[1132] The server receives the answer from the generative AI model and adjusts it using emotion recognition to take into account the emotional state of the questioner. The adjusted answer is then displayed on the question page.

[1133] Displaying options to the questioner

[1134] The server displays the adjusted answer to the questioner along with the options of "I'm satisfied with this answer," "Please tell me more," or "The answer is not appropriate." When the questioner selects an option, the generative AI model's answer is processed accordingly.

[1135] Gathering feedback and learning

[1136] The feedback provided by the questioner is collected by the server, and this data is used as training data to improve the accuracy of the emotion recognition means.

[1137] Specific examples

[1138] 1. Simple Questions

[1139] A user posts a question such as "What is this product made of?" The server detects the question and sends it to the generative AI model. The answer from the generative AI model, "This product is made of cotton," is adjusted by the emotion recognition means and displayed on the question page. When the user selects "I'm satisfied with this answer," feedback is collected that the answer was appropriate.

[1140] 2. Complex Questions

[1141] A user posts a question such as "Please tell me the detailed manufacturing process of this product." The server detects the question and sends it to the generative AI model. The answer from the generative AI model, "This product is manufactured through multiple processes," is adjusted using emotion recognition and displayed on the question page. The user can select "Tell me more" to provide further details.

[1142] Prompt Sentence Examples

[1143] The user enters a question such as:

[1144] "What is this item made of?"

[1145] Example prompts to send to the generative AI model:

[1146] "The user is asking about the materials used in this product. The materials used in this product are as follows:"

[1147] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1148] Step 1:

[1149] A user enters a question into the Q&A service's input form and clicks the "Submit" button. The input is the question text entered by the user. The terminal sends this question data to the server. The output is the question data sent to the server.

[1150] Step 2:

[1151] The server receives new question data and stores it in a database. The input is the question data sent from the device. The server detects that a question has been posted and analyzes the questioner's emotional information using emotion recognition means. The output is the question data and the analyzed emotional information.

[1152] Step 3:

[1153] The server sends question data to the generative AI model. The input is the question and emotion information detected and analyzed by the server. The server sends a request to the generative AI model, which then generates an answer to the question. The output is the answer from the generative AI model.

[1154] Step 4:

[1155] The server receives the answer from the generative AI model and adjusts the answer using the emotion recognition means. The input is the answer from the generative AI model and pre-analyzed emotional information. The emotion recognition means adjusts the answer according to the questioner's emotions, and the server obtains the final adjusted answer. The output is the adjusted answer.

[1156] Step 5:

[1157] The server displays the adjusted answer on the question page. The input is the adjusted answer. The server displays this answer on the question page and generates a user interface with the options "I'm happy with this answer," "Please tell me more," and "This answer is not appropriate." The output is the answer and options that the user can view.

[1158] Step 6:

[1159] The user clicks on one of the options. The input is the option selected by the user. The terminal sends the selection to the server. The output is the selection sent to the server.

[1160] Step 7:

[1161] The server processes the generative AI model's answer based on the user's selection and stores it in a database if necessary. The inputs are the user's selection and the generative AI model's answer. Based on the selection, the server saves the answer as the "best answer," adds it to the page, or does nothing and exits. The output is the result saved in the database or an update to the page.

[1162] Step 8:

[1163] The server starts training the emotion recognition means based on feedback from the user. The input is the user's feedback data. The server uses this to train the emotion recognition means to improve its accuracy. The output is an emotion recognition means with improved accuracy.

[1164] 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.

[1165] 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.

[1166] 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.

[1167] [Third embodiment]

[1168] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1169] 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.

[1170] 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).

[1171] 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.

[1172] 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.

[1173] 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).

[1174] 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.

[1175] 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.

[1176] 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.

[1177] 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.

[1178] 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.

[1179] 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."

[1180] This invention is a system that uses a generative AI model to quickly provide answers to questions in a Q&A service. To implement this system, the following program processes and operations are required.

[1181] 1. Posting and discovering questions

[1182] A user enters a question on the Q&A service platform and clicks the "Post" button.

[1183] The terminal transmits the input question data to the server.

[1184] The server detects when a question is posted, analyzes the question data, and classifies it into the appropriate category.

[1185] 2. Display of "See AI's Answer" button

[1186] The server dynamically adds a "See AI's Answer" button to the question page.

[1187] The terminal redraws the updated page to the user, displaying the new button.

[1188] 3. Sending a request to the generative AI model

[1189] The user clicks the "See AI's answer" button.

[1190] The terminal detects this click event and sends the request to the server.

[1191] The server sends the question data as a request to the API of the generative AI model.

[1192] 4. Receiving and displaying answers from the generative AI model

[1193] The generative AI model analyzes the received question data and generates an appropriate answer.

[1194] The generative AI model returns the generated answer to the server.

[1195] The server formats the received answer data and adds it to the question page.

[1196] The device re-renders the updated page to the user, displaying the generative AI model's answer.

[1197] 5. Displaying options to the questioner

[1198] The server generates a UI that displays multiple options that the questioner can choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Do not add the AI ​​answer").

[1199] The device redraws its UI and displays it to the user.

[1200] 6. Processing Choice-Based Generative AI Model Answers

[1201] The user clicks on the appropriate option.

[1202] The terminal transmits the selection result to the server.

[1203] The server processes the generative AI model's answer based on the questioner's selection.

[1204] "Mark as Best Answer": Set the generative AI model's answer as the best answer in the database and display it on the page.

[1205] "Add to page without setting as best answer": The answer from the generative AI model is added to the database as an answer, and is displayed on the page without being set as the best answer.

[1206] "Don't add AI answer": The generated AI model's answer will not be added to the page or database.

[1207] Specific examples

[1208] Example 1: A simple question

[1209] A user posts a question to a Q&A service: "What is the appropriate amount of food for a cat?"

[1210] The server detects the question and displays a "See AI's Answer" button.

[1211] When the user clicks the button, the generative AI model generates and displays the answer: "For an adult cat, 2-3% of body weight is appropriate."

[1212] The user selects "make best answer" and the server sets the answer as the best answer.

[1213] Example 2: Advanced Questions

[1214] A user posts the question "What is the procedure for normalizing a wave function in quantum mechanics?"

[1215] The server detects the question and similarly displays a "See AI Answer" button.

[1216] When the user clicks the button, the generative AI model generates and displays the answer: "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1."

[1217] The user decides that the answer is insufficient and selects "Add to page without selecting as best answer." The server adds the generated AI model's answer to the page and expects other users to provide more detailed answers.

[1218] In this way, by utilizing a generative AI model, the system of the present invention allows questioners to receive quick and appropriate answers, allowing answerers to focus on asking more advanced questions.

[1219] The processing flow will be explained below.

[1220] Step 1:

[1221] The user enters a question into the input form on the Q&A service platform and clicks the "Post" button.

[1222] Step 2:

[1223] The terminal transmits the posted question data to the server.

[1224] Step 3:

[1225] The server stores the received question data in a database and detects that a question has been posted.

[1226] Step 4:

[1227] The server generates a script that dynamically adds a "View AI Answer" button to the question page.

[1228] Step 5:

[1229] The device receives the script sent from the server, redraws the question page, and displays the "See AI's Answer" button.

[1230] Step 6:

[1231] The user clicks the "See AI's answer" button.

[1232] Step 7:

[1233] The terminal detects the button click event and sends the request to the server.

[1234] Step 8:

[1235] The server sends the question data as a request to the API of the generative AI model.

[1236] Step 9:

[1237] A generative AI model analyzes the question data and generates appropriate answers.

[1238] Step 10:

[1239] The generative AI model returns the generated answer to the server.

[1240] Step 11:

[1241] The server receives the generated response data and formats it appropriately.

[1242] Step 12:

[1243] The server adds HTML containing the formatted answer to the question page.

[1244] Step 13:

[1245] The device renders the HTML received from the server and redraws the question page to display the answer from the generative AI model.

[1246] Step 14:

[1247] The server generates a UI to display multiple options for the answer of the generated AI model ("Make it the best answer," "Don't make it the best answer but add it to the page," "Don't add the AI ​​answer").

[1248] Step 15:

[1249] The device renders the UI with the options and redraws the question page to display to the user.

[1250] Step 16:

[1251] The user clicks on one of the choices.

[1252] Step 17:

[1253] The terminal detects the selection click event and sends the result to the server.

[1254] Step 18:

[1255] The server processes the answer of the generative AI model based on the selection results.

[1256] Step 19:

[1257] If you select "Set as best answer," the server will set the generated AI model's answer as the best answer in the database and display it on the question page.

[1258] Step 20:

[1259] If you select "Add to page without setting as best answer," the server will add the generated AI model's answer to the database and display it on the question page, but will not set it as the best answer.

[1260] Step 21:

[1261] If you select "Do not add AI answer," the server will not add the generated AI model's answer to the database or the question page, and will return the result to the device.

[1262] Step 22:

[1263] The terminal redraws the question page based on the results received from the server and displays the final state to the user.

[1264] The above is the specific program processing flow of this system.

[1265] Example 1

[1266] 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."

[1267] Conventional Q&A services have the problem that it takes a long time for questioners to receive an appropriate answer. Furthermore, because the quality and speed of the answer depends on the answerer, it is difficult to obtain an appropriate answer immediately to the question. Furthermore, the system's efficiency is low because question categorization and responses are done manually. Furthermore, conventional systems have limited options for automated answers using AI, leaving users with little flexibility in selecting AI answers.

[1268] 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.

[1269] In this invention, the server includes means for detecting that a question has been posted, means for analyzing the question data and classifying it into an appropriate category, means for dynamically adding a "See AI's Answer" button to the question page, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, and means for processing the answer of the generative AI model based on the choice made by the questioner. This allows the questioner to quickly obtain an appropriate answer, while allowing the answerer to focus on more advanced questions, improving the efficiency of the entire system.

[1270] The "means for detecting that a question has been posted" refers to a method or device for detecting that a user has entered a question on the Q&A service platform and clicked the post button.

[1271] "Means for analyzing question data and classifying it into appropriate categories" refers to a method or device that analyzes input question data using natural language processing (NLP) or the like, and classifies it into relevant categories based on its content.

[1272] The "means for dynamically adding a 'See AI's Answer' button to a question page" refers to a method or device for adding and displaying a 'See AI's Answer' button on a question page in real time after a question is posted.

[1273] "Means for sending a request to the generative AI model" refers to a method or device that sends the question data as a request to the API of the generative AI model when a user clicks the "View AI's answer" button.

[1274] "Means for receiving an answer from a generative AI model and displaying it on a question page" refers to a method or device for receiving an answer returned from a generative AI model and displaying it on a question page in a form that can be viewed by a user.

[1275] "Means for displaying multiple options to the questioner" refers to a method or device that displays options for the questioner in response to the answer of the generative AI model, such as "Make this the best answer," "Add this to the page without making it the best answer," or "Do not add this to the AI ​​answer."

[1276] "Means for processing the answer of a generative AI model based on the questioner's selection" refers to a method or device that processes the answer of a generative AI model in accordance with the questioner's selection from the options presented, such as saving it in a database, displaying it, or discarding it.

[1277] This invention is a system that uses a generative AI model to provide quick answers to questions in a Q&A service. This system is realized by combining various means that operate mainly on a server, a terminal, and a user.

[1278] First, a user enters a question on the Q&A service platform and clicks the "Post" button. The device detects this action and sends the entered question data to the server. The server analyzes the received question data and detects that it is a question. The server then analyzes the question data using a natural language processing (NLP) library (e.g., spaCy) and classifies it into the appropriate category.

[1279] Next, the server generates HTML code to dynamically add a "See AI's Answer" button to the question page and sends it to the terminal. The terminal receives this code and redraws the user's browser screen to display the "See AI's Answer" button. When the user clicks this button, the terminal detects the click event and sends the request to the server.

[1280] Based on the received request, the server sends question data as a request to the generative AI model's API (e.g., OpenAI). The generative AI model analyzes the sent question data and generates an answer. The generative AI model returns the generated answer to the server. The server formats the received answer data, generates HTML code to add to the question page, and sends it to the device. The device receives this updated data and displays the generative AI model's answer by redrawing the user's browser screen.

[1281] Furthermore, the server generates a UI to display multiple options that the user can select from for the generated AI model's answer (e.g., "Make this the best answer," "Add to the page without making this the best answer," and "Don't add the AI ​​answer"). The device redraws this UI and displays it on the page. When the user clicks on an appropriate option, the device sends this selection to the server. The server processes the generated AI model's answer based on the user's selection, for example, saving it as the best answer in a database or updating the page.

[1282] Specific examples

[1283] Prompt statement example 1:

[1284] A user posts a question to a Q&A service: "What is the appropriate amount of food for a cat?" The server detects the question and displays a "See AI's Answer" button. When the user clicks the button, the generative AI model generates and displays the answer: "For an adult cat, 2-3% of its body weight is appropriate." The user selects "Make this the best answer," and the server adds the answer to the database as the best answer.

[1285] Prompt statement example 2:

[1286] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?" The server detects the question and similarly displays a "See AI's answer" button. When the user clicks the button, the generative AI model generates an answer: "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1," and displays it. The user decides that the answer is insufficient and selects "Add to page without marking as best answer." The server adds the generative AI model's answer to the page, and expects other users to provide more detailed answers.

[1287] In this way, the system utilizes generative AI models to enable users to get fast and relevant answers, allowing other users to focus on more advanced questions, improving the overall efficiency of the system and providing a better user experience by providing options for questioners.

[1288] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1289] Step 1: Post a question

[1290] Users enter their questions on the Q&A service platform and click the "Post" button.

[1291] Input: Text data of the user's question (e.g., "What is the appropriate amount of food for a cat?")

[1292] The terminal receives user input, converts the question data into JSON format, and sends it to the server.

[1293] Output: Question data in JSON format

[1294] Step 2: Question detection and parsing

[1295] The server receives the JSON data and detects that a question has been posted.

[1296] Input: JSON format question data sent from the terminal

[1297] The server uses an NLP library (e.g., spaCy) to parse the question data and classify it into appropriate categories.

[1298] Output: Question analysis results and category information (e.g., "Pet category")

[1299] Step 3: Display of the "See AI's Answer" button

[1300] The server generates HTML code to dynamically add a "View AI Answer" button to the question page.

[1301] Input: User question, analysis results, category information

[1302] The server sends a response containing the generated HTML code to the terminal.

[1303] The device uses the updated data received from the server to redraw the user's browser screen, which displays a new "See AI's Answer" button.

[1304] Output: Updated HTML code

[1305] Step 4: Sending a request to the generative AI model

[1306] The user clicks the "See AI's answer" button.

[1307] Input: User click event

[1308] The terminal detects the click event and sends a request along with the question data to the server.

[1309] Output: Request data to the server

[1310] The server sends the question data as a request to the API of the generative AI model (e.g., OpenAI).

[1311] Step 5: Receive and display the answer from the generative AI model

[1312] The server receives the response from the generative AI model.

[1313] Input: Answer data from the generative AI model

[1314] The generative AI model analyzes the question data and generates an appropriate answer (e.g., "For adult cats, 2-3% of body weight is appropriate").

[1315] The server formats the generated answer and generates HTML code to add to the question page.

[1316] Output: Updated HTML code and generated AI answer data

[1317] The device receives the updated page data and redraws the user's browser screen, which displays the answer from the generated AI model.

[1318] Step 6: Displaying the options

[1319] The server generates a UI to display multiple options for the user to choose from in response to the generated AI model's answer.

[1320] Input: Answer data for the generative AI model

[1321] The server transmits the generated UI data to the terminal.

[1322] The device uses the UI data to display the options to the user.

[1323] Output: Updated UI with choices

[1324] Step 7: Processing based on user selection

[1325] The user clicks on the appropriate option.

[1326] Input: User click event

[1327] The terminal transmits the selection result to the server.

[1328] Output: Request data to the server

[1329] The server processes the generative AI model's answer based on the questioner's selection.

[1330] "Mark as Best Answer": Set the generative AI model's answer as the best answer in the database and display it on the page.

[1331] "Add to page without setting as best answer": Add the generative AI model's answer and display it on the page without setting it as the best answer.

[1332] "Don't add AI answer": The generated AI model's answer will not be added to the page or database.

[1333] The above are the specific processing steps and operations of the program for this system.

[1334] (Application example 1)

[1335] 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."

[1336] In recent years, with the spread of electronic payment services, the number of inquiries from users has increased dramatically. This has resulted in delayed support responses, leading to problems that lead to a decline in user satisfaction. Furthermore, support staff are required to respond to a large number of inquiries, increasing their workload. To solve this problem, a system that can provide prompt and appropriate responses is needed. However, existing systems have difficulty in providing prompt and appropriate responses to inquiries.

[1337] 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.

[1338] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, means for processing the answer from the generative AI model based on the questioner's selection, means for providing a user interface for inputting questions about electronic payments, and means for receiving the answer from the generative AI model and displaying it on the question page related to electronic payments. This allows users to receive quick and accurate answers to inquiries about electronic payments, improving the efficiency of support responses and reducing workloads.

[1339] The "means for detecting that a question has been posted" is a mechanism by which the system determines that a user has entered and posted a question on the system.

[1340] The "means for sending a request to the generative AI model" is a mechanism for sending the posted question to the generative AI model and sending a request to generate an appropriate answer.

[1341] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to a mechanism for receiving answers returned from the generative AI model and displaying them on the user interface.

[1342] "Means for displaying multiple options to the questioner" refers to a mechanism that displays multiple options so that the user can select further actions in response to the answer from the generative AI model.

[1343] The "means for processing the generative AI model's answer based on the questioner's selection" refers to a mechanism that, after the user selects an option, stores or displays the generative AI model's answer in a database based on that selection.

[1344] The "means for providing a user interface for inputting questions about electronic payments" is a mechanism for providing an interface designed to allow users to intuitively input questions about electronic payments.

[1345] "Means for receiving the answer of the generative AI model and displaying it on the question page related to electronic payments" refers to a mechanism for receiving the answer related to electronic payments returned from the generative AI model and displaying it on the relevant question page.

[1346] This invention is a system for quickly and accurately answering questions about electronic payments. The system consists of posting and detecting questions, sending requests to a generative AI model, receiving and displaying answers, presenting options to the user, and processing the answers based on the selections.

[1347] The core of the system resides in the server, which has the following main functions:

[1348] 1. Post and discover questions:

[1349] When a user enters a question about electronic payments on the smartphone application and clicks the "Post" button, the device sends the question data to the server, which detects the post, analyzes the received question data, and classifies it into the appropriate category.

[1350] 2. Display of "See AI's Answer" button:

[1351] When a question is submitted, the server dynamically adds a "See AI's Answer" button to the question page. After this button is added to the page, the device redraws the page and displays the new button to the user.

[1352] 3. Sending a request to the generative AI model:

[1353] When a user clicks the "View AI's Answer" button, the click event is detected by the device and a request is sent to the server. The server then sends the received question data as a request to the API of the generative AI model.

[1354] 4. Receive and display answers from the generative AI model:

[1355] The generative AI model analyzes the received question data and generates an appropriate answer. The generated answer is returned to the server, which formats the answer data and adds it to the question page. The device redraws the updated page and displays the generative AI model's answer to the user.

[1356] 5. Displaying options to the questioner:

[1357] The server generates a UI to display multiple options for the questioner to choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Don't add the AI ​​answer"), and the device redraws the UI to display the options to the user.

[1358] 6. Processing the generative AI model's answers based on your choices:

[1359] When the user clicks on the appropriate option, the device sends the selection to the server, which then takes appropriate action on the generative AI model's answer based on the user's selection, such as setting it as the best answer, adding it to the page, or ignoring it.

[1360] The following hardware and software are used to realize this system.

[1361] 1. Smartphone application:

[1362] Platform: iOS and Android

[1363] Development language: Swift (iOS), Kotlin (Android)

[1364] 2. Server:

[1365] Platform: Linux-based server environment

[1366] Software used: Node.js, Django, Flask

[1367] 3. Generative AI Model:

[1368] Model Name: OpenAI GPT-3 or GPT-4

[1369] API used: OpenAI API

[1370] As a concrete example, if a question is posted asking "Where can I check my credit card statement?", the system will operate as follows:

[1371] 1. The server receives the question and adds a "See AI's Answer" button to the question page.

[1372] 2. When the user clicks this button, the question data is sent to the generative AI model.

[1373] 3. The generative AI model generates the answer, "You can check your credit card statement on the 'Statement' page of the official app or website," and returns it to the server.

[1374] 4. The server displays this answer on the question page and gives the user the options of "Make it the best answer," "Add it to the page without making it the best answer," or "Don't add the AI ​​answer."

[1375] 5. If the user selects "Mark as Best Answer," the answer is saved in the database and set as the best answer on the question page.

[1376] This system will enable users to quickly obtain information about electronic payments and is expected to reduce the burden on support staff.

[1377] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1378] Step 1: Posting and discovering questions

[1379] A user enters a question about electronic payment into a smartphone application and clicks the "Post" button. At this time, the question text entered by the user is generated. The device sends this question text to the server. The server analyzes the received question data and detects that a question has been posted. Next, it classifies the question data into appropriate categories. This data processing process organizes the question content into categories.

[1380] Input: The question text entered by the user

[1381] Data processing: categorization of question data

[1382] Output: Question data with categories

[1383] Step 2: Displaying the "See AI's Answer" button

[1384] The server dynamically adds a "See AI's Answer" button to the question page based on the question data it received. It then instructs the device to redraw the updated page and display the new button to the user. This operation allows the user to expect a quick answer from the AI.

[1385] Input: Question data with categories

[1386] Data processing: Update the question page and add a button

[1387] Output: Updated question page with a "See AI's answer" button

[1388] Step 3: Sending a request to the generative AI model

[1389] When the user clicks the "View AI Answer" button, the device detects this click event. Based on this detection, the device resends the question data to the server. The server then sends the received question data as a request to the generative AI model's API. This data transmission causes the generative AI model to begin analyzing the question content.

[1390] Input: Click event of the "View AI's answer" button and question data

[1391] Data processing: Sending a request to the generative AI model API

[1392] Output: The question request received by the generative AI model

[1393] Step 4: Receive and display the answer from the generative AI model

[1394] The generative AI model analyzes the received question data and generates an appropriate answer. This answer is returned to the server. The server formats the received answer data and adds it to the question page. The device then redraws the updated page and displays the generative AI model's answer, allowing the user to see the answer provided by the AI ​​model.

[1395] Input: Answer data generated by a generative AI model

[1396] Data processing: Formatting of response data

[1397] Output: Question page with formatted answer data

[1398] Step 5: Present the options to the questioner

[1399] The server generates a UI to display multiple options for the questioner to choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Don't add the AI ​​answer") The device redraws this UI and displays it to the user.

[1400] Input: Answer data for the generative AI model

[1401] Data processing: generating options and updating the UI

[1402] Output: Question page with updated UI

[1403] Step 6: Processing the generative AI model's answer based on your selection

[1404] When the user clicks on the appropriate option, the device detects this click event and sends the selection result to the server. Based on the selection, the server saves the generative AI model's answer in a database or displays it on the page. For example, if the user selects "Set as best answer," the answer is set as the best answer in the database and displayed on the question page.

[1405] Input: User selection

[1406] Data processing: Save answers to database and update page display

[1407] Output: Updated question page and database state

[1408] 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.

[1409] This invention is a system that provides quick and appropriate answers to questions in a Q&A service using a generative AI model and an emotion engine. The configuration of this system and the program processing flow are described in detail below.

[1410] System configuration

[1411] The system mainly consists of the following components:

[1412] 1. Question Post Detection

[1413] The server has a means to detect new questions posted by users.

[1414] 2. Sending a request to the generative AI model

[1415] The server has the ability to send the detected questions to the generative AI model.

[1416] 3. Receiving and displaying answers from the generative AI model

[1417] The server has a means to receive the answer from the generative AI model and display it on the question page.

[1418] 4. Emotion Engine

[1419] The server has the means to recognize the questioner's emotions and analyze that information.

[1420] The emotion engine adjusts the generative AI model's responses based on the questioner's emotional state.

[1421] 5. Display of questioner's options

[1422] The server has a means to display a UI that offers the questioner multiple options for the answer of the generated AI model.

[1423] 6. Processing Choice-Based Answers

[1424] The server has a means to process the generative AI model's answer based on the questioner's selection.

[1425] The server has the function of storing the emotion information recognized by the emotion engine in a database.

[1426] 7. Learning the Emotion Engine

[1427] The server has a learning mechanism to improve emotion recognition accuracy based on feedback from the questioner.

[1428] Program flow

[1429] Posting and discovering questions

[1430] A user enters a question into the Q&A service's input form and clicks the "Post" button. The device sends the entered question data to the server. The server saves the question in a database, detects that the question has been posted, and analyzes the questioner's emotional information using an emotion engine.

[1431] Display of "See AI's answer" button

[1432] The server generates a script that dynamically adds a "See AI's Answer" button to the question page. The device receives this script, redraws the question page, and displays the "See AI's Answer" button.

[1433] Linking generative AI models and emotion engines

[1434] The user clicks the "See AI's Answer" button. The device detects the click event and sends the request to the server. The server sends the question data to the generative AI model, which then generates an answer.

[1435] The emotion engine adjusts the generated answer based on the emotional state of the questioner. The server receives the adjusted answer and displays it on the question page.

[1436] Displaying options to the questioner

[1437] The server generates a UI for the questioner to choose from three options for the AI ​​model's answer: "Make it the best answer," "Add it to the page without making it the best answer," or "Don't add it to the AI ​​answer." The device redraws the page and displays the options to the questioner.

[1438] Processing questioner selections and answers

[1439] The user clicks on one of the options. The device sends the selection to the server, which processes the answer of the generative AI model based on the selection.

[1440] Mark as best answer: The server saves the generated AI model's answer in the database as the best answer and displays it on the page.

[1441] Add to page without marking as best answer: The server adds the generated AI model's answer to the page, making it visible to other users.

[1442] Do not add AI answer: The server does not add the generated AI model answer to the page or database, but leaves it as is.

[1443] Emotion Engine Learning

[1444] Based on the feedback from the questioner, the server performs a learning process to improve the emotion recognition accuracy of the emotion engine.

[1445] Specific examples

[1446] Example 1: A simple question

[1447] A user posts a question: "What is the appropriate amount of food for a cat?" The server detects the question and displays a "See AI's Answer" button. When the user clicks the button, the generative AI model generates the answer: "For an adult cat, 2-3% of its body weight is appropriate," and the emotion engine appropriately adjusts the answer based on the questioner's emotional information and displays it. The user selects "Make Best Answer," and the server sets the answer as the best answer.

[1448] Example 2: A more complex question

[1449] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?" The server detects the question and displays a "See AI's answer" button. When the user clicks the button, the generative AI model generates the answer: "The procedure is to integrate the wave function over the entire space and make the square of the absolute value of that result 1." The emotion engine analyzes the questioner's emotions and adjusts the answer. The user selects "Add to page without marking as best answer," and the server adds the generative AI model's answer to the page.

[1450] In this way, the system of the present invention uses a generative AI model and an emotion engine in combination, allowing the questioner to receive a quick and appropriate answer while allowing the answerer to focus on asking more advanced questions.

[1451] The processing flow will be explained below.

[1452] Step 1:

[1453] The user enters a question into the Q&A service's input form and clicks the "Post" button.

[1454] Step 2:

[1455] The terminal transmits the posted question data to the server.

[1456] Step 3:

[1457] The server stores the received question data in a database and detects that a question has been posted.

[1458] Step 4:

[1459] The server calls the emotion engine and analyzes the questioner's emotional information based on the question data.

[1460] Step 5:

[1461] The server stores the emotion analysis results in a database.

[1462] Step 6:

[1463] The server generates a script that dynamically adds a "View AI Answer" button to the question page.

[1464] Step 7:

[1465] The device receives the script sent from the server, redraws the question page, and displays the "See AI's Answer" button.

[1466] Step 8:

[1467] The user clicks the "See AI's answer" button.

[1468] Step 9:

[1469] The terminal detects the button click event and sends the request to the server.

[1470] Step 10:

[1471] The server sends the question data and emotion information from the emotion engine as a request to the API of the generative AI model.

[1472] Step 11:

[1473] The generative AI model analyzes question data and emotional information to generate appropriate answers.

[1474] Step 12:

[1475] The generative AI model returns the generated answer to the server.

[1476] Step 13:

[1477] The server receives the generated response data and formats it appropriately.

[1478] Step 14:

[1479] The server adds HTML containing the formatted answer to the question page.

[1480] Step 15:

[1481] The device renders the HTML received from the server and redraws the question page to display the answer from the generative AI model.

[1482] Step 16:

[1483] The server generates a UI that provides multiple options for the answer of the generated AI model.

[1484] Step 17:

[1485] The device renders the UI with the options and redraws the question page to display to the user.

[1486] Step 18:

[1487] The user can choose to either "Make this the best answer," "Add it to the page without making it the best answer," or "Do not add the AI ​​answer."

[1488] Step 19:

[1489] The terminal transmits the user's selection to the server.

[1490] Step 20:

[1491] The server processes the answer of the generative AI model based on the selection results.

[1492] Step 21:

[1493] If you select "Set as best answer," the server will set the generated AI model's answer as the best answer in the database and display it on the question page.

[1494] Step 22:

[1495] If you select "Add to page without setting as best answer," the server will add the generated AI model's answer to the database and display it on the question page, but will not set it as the best answer.

[1496] Step 23:

[1497] If you select "Do not add AI answer", the server will not add the generated AI model's answer to the database or the question page.

[1498] Step 24:

[1499] The server sends feedback from the questioner to the emotion engine, which then learns to improve emotion recognition accuracy.

[1500] Step 25:

[1501] The terminal redraws the question page based on the final result received from the server and displays it to the user.

[1502] The above is the specific program processing flow of this system.

[1503] Example 2

[1504] 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."

[1505] With conventional Q&A services, it was difficult for questioners to obtain prompt and appropriate answers. Furthermore, the answers generated by the AI ​​model were not properly adjusted according to the questioner's feelings, resulting in low satisfaction with the answer. Furthermore, there was a lack of a way to clearly indicate which option the questioner should take in response to the question posted, which did not improve the user experience.

[1506] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1507] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for analyzing the emotion of the questioner, means for adjusting the content of the answer of the generative AI model based on the analyzed emotion, means for displaying multiple options to the questioner, means for processing the answer of the generative AI model based on the selection of the questioner, and means for storing the emotion information in a database. This allows the questioner to receive a quick and appropriate answer, and further enables the answer to be adjusted according to the questioner's emotion, thereby improving the user experience.

[1508] "Question detection" is the process by which the server automatically recognizes questions posted by users.

[1509] "Sending a request to the generative AI model" is a means of sending information from the server to the generative AI model to analyze the detected question.

[1510] "Receiving and displaying answers from the generative AI model" refers to the process in which the server receives the answers provided by the generative AI model and displays them on the question page.

[1511] "Emotion analysis" is a means of determining the questioner's emotional state using an emotion engine.

[1512] "Adjusting response content based on emotions" is a method of appropriately changing the response content provided by the generative AI model based on analyzed emotional information.

[1513] "Displaying options" is the process by which the server provides a user interface with multiple options that the questioner can select from in response to the answer provided by the generative AI model.

[1514] "Choice-based answer processing" is the process by which the server appropriately processes the answer of the generative AI model depending on the option selected by the questioner.

[1515] "Storing emotional information" is a method of storing the emotional information analyzed by the emotion engine in a database.

[1516] This invention is a system that provides quick and appropriate answers to questions in a Q&A service using a generative AI model and an emotion engine. The configuration of this system and the program processing flow are described in detail below.

[1517] System configuration

[1518] The system consists of the following main components:

[1519] 1. Server

[1520] The server has a means for receiving and detecting queries from users.

[1521] It has the ability to send question data to a generative AI model.

[1522] It has a means to receive answers from the generative AI model and display them on the question page.

[1523] It integrates an emotion engine to analyze the questioner's emotions.

[1524] It has the ability to adjust the answers of the generative AI model based on the analyzed emotional information.

[1525] It has a means to display multiple options in the user interface.

[1526] It has a means to process the generative AI model's answers based on the choices.

[1527] It has the function of storing emotional information in a database.

[1528] 2. Terminal

[1529] The terminal provides an interface where the user can enter a question and click a "Post" button.

[1530] It has a means for sending question data to the server.

[1531] It has the means to receive and execute a script to dynamically display a "View AI Answer" button on the question page.

[1532] It has the function of sending the selection results to the server.

[1533] 3. Users

[1534] Users post questions using the input form of the Q&A service.

[1535] Click the "See AI's Answer" button to see the answer from the generated AI model.

[1536] Make the appropriate choice from multiple options.

[1537] Specific examples

[1538] Example 1: A simple question

[1539] Example prompt sentence:

[1540] What is the appropriate amount of cat food?

[1541] A user posts a question: "What is the right amount of food for a cat?"

[1542] The server detects the question and analyzes the questioner's emotional information using an emotion engine.

[1543] The server sends a script to the terminal that displays a "See AI's answer" button.

[1544] When the user clicks the "See AI Answer" button, the server sends the question data to the generative AI model and receives the answer.

[1545] For example, the generative AI model generates an answer such as "2-3% of body weight is appropriate for an adult cat," which is received by the server.

[1546] The emotion engine takes into account the emotional state of the questioner and adjusts the answer accordingly.

[1547] The server displays the adjusted answer on the question page.

[1548] The user selects "Mark as Best Answer" and the server saves the answer as the best answer in the database.

[1549] Example 2: Complex Question

[1550] Example prompt sentence:

[1551] "What is the procedure for normalizing wave functions in quantum mechanics?"

[1552] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?"

[1553] The server detects the question and analyzes the questioner's emotional information using an emotion engine.

[1554] The server sends a script to the terminal that displays a "See AI's answer" button.

[1555] When the user clicks the "See AI Answer" button, the server sends the question data to the generative AI model and receives the answer.

[1556] For example, the generative AI model generates an answer such as, "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1," and the server receives this.

[1557] The emotion engine analyzes the questioner's emotions and adjusts the response appropriately.

[1558] The server displays the adjusted answer on the question page.

[1559] The user selects "Add to page without marking as best answer" and the server adds the answer to the page.

[1560] In this way, the system of the present invention can use generative AI models and emotion engines to provide users with fast and relevant answers.

[1561] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1562] Specific explanation of program processing

[1563] Step 1:

[1564] The user enters a question into the Q&A service's input form and clicks the "Post" button.

[1565] Input: The question text entered by the user.

[1566] Output: The query data is sent to the server.

[1567] Specific behavior:

[1568] The terminal obtains the contents of the user input form.

[1569] The terminal generates a request to transmit the question data to the server.

[1570] The device sends a request to the server.

[1571] Step 2:

[1572] The server stores the received question data in a database and detects when a new question has been posted.

[1573] The server sends the question to the emotion engine and analyzes the questioner's emotional information.

[1574] Input: Question data sent from the terminal.

[1575] Output: The question data is stored in a database and the sentiment analysis results are obtained.

[1576] Specific behavior:

[1577] The server writes the received question data to the database.

[1578] The server monitors the database to detect new question submissions.

[1579] The server sends the question to the emotion engine and receives the analysis results.

[1580] Step 3:

[1581] The server generates a script that adds a "See AI's Answer" button to the question page and sends it to the terminal.

[1582] Input: The event that a new question is posted.

[1583] Output: The script is sent to the terminal and the button is displayed.

[1584] Specific behavior:

[1585] The server generates script code to add a "See AI's Answer" button.

[1586] The server sends the generated script to the terminal.

[1587] Step 4:

[1588] The device receives the script, redraws the question page, and displays the "See AI's Answer" button.

[1589] Input: The script sent by the server.

[1590] Output: A "See AI's answer" button will be displayed on the question page.

[1591] Specific behavior:

[1592] The terminal parses the script received from the server.

[1593] The terminal executes the received script and redraws the question page.

[1594] The device will display a "See AI's Answer" button.

[1595] Step 5:

[1596] The user clicks the "See AI's answer" button.

[1597] Input: User click event.

[1598] Output: A click event is sent to the server.

[1599] Specific behavior:

[1600] The device detects a click event.

[1601] The device sends the click event information to the server.

[1602] Step 6:

[1603] The server sends the question data to the generative AI model, which then generates an answer.

[1604] Input: Click event information and original question data.

[1605] Output: Answer data from the generative AI model.

[1606] Specific behavior:

[1607] The server creates a request to send the original question data to the generative AI model.

[1608] The server calls the API of the generated AI model and sends the question data.

[1609] The generative AI model analyzes the question and generates an appropriate answer.

[1610] The answer data generated by the generative AI model is sent back to the server.

[1611] Step 7:

[1612] The emotion engine takes into account the emotional state of the questioner and adjusts the generated answer content.

[1613] Input: Response data from the generative AI model and pre-obtained sentiment analysis results.

[1614] Output: Adjusted response data.

[1615] Specific behavior:

[1616] The server sends the response data from the generative AI model and the emotion analysis results to the emotion engine.

[1617] The emotion engine adjusts the response content.

[1618] The emotion engine sends the adjusted response data back to the server.

[1619] Step 8:

[1620] The server receives the adjusted answer and displays it on the question page.

[1621] Input: Calibrated response data from the sentiment engine.

[1622] Output: The answer displayed on the question page.

[1623] Specific behavior:

[1624] The server receives the adjusted response data.

[1625] The server generates data for updating the question page and sends it to the terminal.

[1626] Step 9:

[1627] The device will redraw the question page and display the adjusted answer.

[1628] Input: Data to update from the server.

[1629] Output: Redrawn question page and displayed answer.

[1630] Specific behavior:

[1631] The terminal analyzes the data received from the server and redraws the question page.

[1632] The device will display the adjusted answer.

[1633] Step 10:

[1634] The server generates a UI containing three options that the questioner can choose from in response to the generated AI model's answer: "Make this the best answer," "Add it to the page without making it the best answer," or "Do not add the AI ​​answer," and sends this UI to the device.

[1635] Input: Adjusted response data.

[1636] Output: The UI shown to the questioner.

[1637] Specific behavior:

[1638] Create a script that the server uses to generate a UI with the options.

[1639] The server sends the generated script to the terminal.

[1640] Step 11:

[1641] The terminal will redraw and show the options to the questioner.

[1642] Input: Script from server.

[1643] Output: The choices presented to the questioner.

[1644] Specific behavior:

[1645] The terminal parses the script received from the server.

[1646] The device will redraw the question page and display the options.

[1647] Step 12:

[1648] The user clicks on one of the displayed options.

[1649] Input: User click event.

[1650] Output: Click event information is sent to the server.

[1651] Specific behavior:

[1652] The device detects a click event.

[1653] The device sends the click event information to the server.

[1654] Step 13:

[1655] The server processes the generative AI model's answer based on the selection results.

[1656] Input: Click event information.

[1657] Output: Processed response data.

[1658] Specific behavior:

[1659] The server appropriately processes the answer of the generative AI model based on the click event information.

[1660] If it is selected as the best answer, the server saves the generated AI model's answer as the best answer in the database and displays it on the page.

[1661] If you add the answer to a page without marking it as the best answer, the server adds the answer from the generated AI model to the page, making it available for other users to view.

[1662] If you do not add an AI answer, the server will not add the generated AI model's answer to the page or database, but will leave it as is.

[1663] Step 14:

[1664] The server collects feedback from the questioner and uses it as learning data for the emotion engine.

[1665] Input: Feedback from the questioner.

[1666] Output: Improved recognition accuracy through emotion engine training.

[1667] Specific behavior:

[1668] The server provides an interface for collecting feedback data from questioners.

[1669] The server sends the collected feedback data to the emotion engine.

[1670] The emotion engine learns from feedback data to improve emotion recognition accuracy.

[1671] (Application example 2)

[1672] 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."

[1673] Conventional Q&A services respond slowly to questions, and the answers are often inadequate. Furthermore, answers are provided without considering the feelings of the questioner, resulting in low user satisfaction. Furthermore, answers are provided one-way, making it difficult to improve the service based on user feedback. A system that can solve these problems and provide fast and appropriate answers is needed.

[1674] 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.

[1675] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, means for processing the answer of the generative AI model based on the choice of the questioner, means for analyzing emotional information of the questioner using emotion recognition means and adjusting the answer of the generative AI model based on that information, means for collecting feedback from the questioner on whether the answer to the question was appropriate, and means for training the emotion recognition means to improve emotion recognition accuracy. This makes it possible to provide quick and appropriate answers to questions, as well as adaptive answers that take the questioner's emotions into consideration, thereby improving user satisfaction.

[1676] The "means for detecting that a question has been posted" is a function that, when a user enters a question into the input form and clicks the submit button, sends the question data to the server and recognizes that a new question has been posted.

[1677] The "means for sending a request to a generative AI model" is a function for sending a detected question to a generative AI model and sending a request to obtain an appropriate answer to the question.

[1678] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to a function that receives answers generated by the generative AI model and displays them on the question page in a format that is easy for users to view.

[1679] "Means for displaying multiple options to the questioner" refers to a function that displays a user interface that provides multiple options for the questioner to select in response to the answer of the generative AI model.

[1680] "Means for processing the generative AI model's answer based on the questioner's selection" refers to a function that optimally processes the generative AI model's answer based on the option selected by the questioner.

[1681] "Means for analyzing the emotional information of the questioner using emotion recognition means and adjusting the response of the generative AI model based on that information" refers to a function that analyzes the emotional state of the questioner and appropriately adjusts the content of the response in accordance with that emotion.

[1682] "Means for collecting feedback from the questioner on whether the answer to the question was appropriate" is a function that collects direct feedback from the questioner on whether the answer provided by the generative AI model was appropriate.

[1683] "Means for training the emotion recognition means to improve the emotion recognition accuracy" is a function that improves and trains the accuracy of the emotion recognition engine based on feedback from the questioner.

[1684] This invention is a system that provides prompt and appropriate answers in a Q&A service by combining a generative AI model and emotion recognition means. The configuration of this system and the program processing flow are described in detail below.

[1685] System configuration

[1686] The system consists of the following components:

[1687] 1. How to detect question posts

[1688] When a user enters a question into the input form and clicks the send button, the terminal sends the entered question data to the server.

[1689] 2. How to send requests to the generative AI model

[1690] The server sends the detected question to a generative AI model, which then generates an answer to that question.

[1691] 3. How to receive and display answers from the generative AI model

[1692] This is a function where the server receives the answer from the generative AI model and displays it on the question page.

[1693] 4. Emotion recognition means

[1694] This function analyzes the emotions of the questioner and adjusts the response content of the generative AI model based on that. This method analyzes the emotional state of the questioner and adjusts the response content to be appropriate for the questioner.

[1695] 5. A way to present multiple options to the questioner

[1696] This function displays a user interface to the questioner in response to the answer generated by the server using the AI ​​model, with options such as "I'm satisfied with this answer," "Please tell me more," or "The answer is inappropriate."

[1697] 6. Feedback Collection Methods

[1698] This is a method of collecting data to improve emotion recognition accuracy based on feedback from questioners, which enables the emotion recognition method to perform more accurate emotion analysis.

[1699] Program processing description

[1700] Question Post Detection

[1701] When a user enters a question into the Q&A service's input form and clicks the "Send" button, the device sends the entered question data to the server. The server saves the question in a database and detects that the question has been posted. At this time, the question data is sent to the emotion recognition means, which analyzes the questioner's emotions.

[1702] Sending requests to generative AI models

[1703] The server sends the detected question to the generative AI model, which generates an answer to the question and returns the answer to the server.

[1704] Receiving and adjusting answers from generative AI models

[1705] The server receives the answer from the generative AI model and adjusts it using emotion recognition to take into account the emotional state of the questioner. The adjusted answer is then displayed on the question page.

[1706] Displaying options to the questioner

[1707] The server displays the adjusted answer to the questioner along with the options of "I'm satisfied with this answer," "Please tell me more," or "The answer is not appropriate." When the questioner selects an option, the generative AI model's answer is processed accordingly.

[1708] Gathering feedback and learning

[1709] The feedback provided by the questioner is collected by the server, and this data is used as training data to improve the accuracy of the emotion recognition means.

[1710] Specific examples

[1711] 1. Simple Questions

[1712] A user posts a question such as "What is this product made of?" The server detects the question and sends it to the generative AI model. The answer from the generative AI model, "This product is made of cotton," is adjusted by the emotion recognition means and displayed on the question page. When the user selects "I'm satisfied with this answer," feedback is collected that the answer was appropriate.

[1713] 2. Complex Questions

[1714] A user posts a question such as "Please tell me the detailed manufacturing process of this product." The server detects the question and sends it to the generative AI model. The answer from the generative AI model, "This product is manufactured through multiple processes," is adjusted using emotion recognition and displayed on the question page. The user can select "Tell me more" to provide further details.

[1715] Prompt Sentence Examples

[1716] The user enters a question such as:

[1717] "What is this item made of?"

[1718] Example prompts to send to the generative AI model:

[1719] "The user is asking about the materials used in this product. The materials used in this product are as follows:"

[1720] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1721] Step 1:

[1722] A user enters a question into the Q&A service's input form and clicks the "Submit" button. The input is the question text entered by the user. The terminal sends this question data to the server. The output is the question data sent to the server.

[1723] Step 2:

[1724] The server receives new question data and stores it in a database. The input is the question data sent from the device. The server detects that a question has been posted and analyzes the questioner's emotional information using emotion recognition means. The output is the question data and the analyzed emotional information.

[1725] Step 3:

[1726] The server sends question data to the generative AI model. The input is the question and emotion information detected and analyzed by the server. The server sends a request to the generative AI model, which then generates an answer to the question. The output is the answer from the generative AI model.

[1727] Step 4:

[1728] The server receives the answer from the generative AI model and adjusts the answer using the emotion recognition means. The input is the answer from the generative AI model and pre-analyzed emotional information. The emotion recognition means adjusts the answer according to the questioner's emotions, and the server obtains the final adjusted answer. The output is the adjusted answer.

[1729] Step 5:

[1730] The server displays the adjusted answer on the question page. The input is the adjusted answer. The server displays this answer on the question page and generates a user interface with the options "I'm happy with this answer," "Please tell me more," and "This answer is not appropriate." The output is the answer and options that the user can view.

[1731] Step 6:

[1732] The user clicks on one of the options. The input is the option selected by the user. The terminal sends the selection to the server. The output is the selection sent to the server.

[1733] Step 7:

[1734] The server processes the generative AI model's answer based on the user's selection and stores it in a database if necessary. The inputs are the user's selection and the generative AI model's answer. Based on the selection, the server saves the answer as the "best answer," adds it to the page, or does nothing and exits. The output is the result saved in the database or an update to the page.

[1735] Step 8:

[1736] The server starts training the emotion recognition means based on feedback from the user. The input is the user's feedback data. The server uses this to train the emotion recognition means to improve its accuracy. The output is an emotion recognition means with improved accuracy.

[1737] 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.

[1738] 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.

[1739] 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.

[1740] [Fourth embodiment]

[1741] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1742] 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.

[1743] 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).

[1744] 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.

[1745] 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.

[1746] 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).

[1747] 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.

[1748] 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.

[1749] 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.

[1750] 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.

[1751] 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.

[1752] 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.

[1753] 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."

[1754] This invention is a system that uses a generative AI model to quickly provide answers to questions in a Q&A service. To implement this system, the following program processes and operations are required.

[1755] 1. Posting and discovering questions

[1756] A user enters a question on the Q&A service platform and clicks the "Post" button.

[1757] The terminal transmits the input question data to the server.

[1758] The server detects when a question is posted, analyzes the question data, and classifies it into the appropriate category.

[1759] 2. Display of "See AI's Answer" button

[1760] The server dynamically adds a "See AI's Answer" button to the question page.

[1761] The terminal redraws the updated page to the user, displaying the new button.

[1762] 3. Sending a request to the generative AI model

[1763] The user clicks the "See AI's answer" button.

[1764] The terminal detects this click event and sends the request to the server.

[1765] The server sends the question data as a request to the API of the generative AI model.

[1766] 4. Receiving and displaying answers from the generative AI model

[1767] The generative AI model analyzes the received question data and generates an appropriate answer.

[1768] The generative AI model returns the generated answer to the server.

[1769] The server formats the received answer data and adds it to the question page.

[1770] The device re-renders the updated page to the user, displaying the generative AI model's answer.

[1771] 5. Displaying options to the questioner

[1772] The server generates a UI that displays multiple options that the questioner can choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Do not add the AI ​​answer").

[1773] The device redraws its UI and displays it to the user.

[1774] 6. Processing Choice-Based Generative AI Model Answers

[1775] The user clicks on the appropriate option.

[1776] The terminal transmits the selection result to the server.

[1777] The server processes the generative AI model's answer based on the questioner's selection.

[1778] "Mark as Best Answer": Set the generative AI model's answer as the best answer in the database and display it on the page.

[1779] "Add to page without setting as best answer": The answer from the generative AI model is added to the database as an answer, and is displayed on the page without being set as the best answer.

[1780] "Don't add AI answer": The generated AI model's answer will not be added to the page or database.

[1781] Specific examples

[1782] Example 1: A simple question

[1783] A user posts a question to a Q&A service: "What is the appropriate amount of food for a cat?"

[1784] The server detects the question and displays a "See AI's Answer" button.

[1785] When the user clicks the button, the generative AI model generates and displays the answer: "For an adult cat, 2-3% of body weight is appropriate."

[1786] The user selects "make best answer" and the server sets the answer as the best answer.

[1787] Example 2: Advanced Questions

[1788] A user posts the question "What is the procedure for normalizing a wave function in quantum mechanics?"

[1789] The server detects the question and similarly displays a "See AI Answer" button.

[1790] When the user clicks the button, the generative AI model generates and displays the answer: "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1."

[1791] The user decides that the answer is insufficient and selects "Add to page without selecting as best answer." The server adds the generated AI model's answer to the page and expects other users to provide more detailed answers.

[1792] In this way, by utilizing a generative AI model, the system of the present invention allows questioners to receive quick and appropriate answers, allowing answerers to focus on asking more advanced questions.

[1793] The processing flow will be explained below.

[1794] Step 1:

[1795] The user enters a question into the input form on the Q&A service platform and clicks the "Post" button.

[1796] Step 2:

[1797] The terminal transmits the posted question data to the server.

[1798] Step 3:

[1799] The server stores the received question data in a database and detects that a question has been posted.

[1800] Step 4:

[1801] The server generates a script that dynamically adds a "View AI Answer" button to the question page.

[1802] Step 5:

[1803] The device receives the script sent from the server, redraws the question page, and displays the "See AI's Answer" button.

[1804] Step 6:

[1805] The user clicks the "See AI's answer" button.

[1806] Step 7:

[1807] The terminal detects the button click event and sends the request to the server.

[1808] Step 8:

[1809] The server sends the question data as a request to the API of the generative AI model.

[1810] Step 9:

[1811] A generative AI model analyzes the question data and generates appropriate answers.

[1812] Step 10:

[1813] The generative AI model returns the generated answer to the server.

[1814] Step 11:

[1815] The server receives the generated response data and formats it appropriately.

[1816] Step 12:

[1817] The server adds HTML containing the formatted answer to the question page.

[1818] Step 13:

[1819] The device renders the HTML received from the server and redraws the question page to display the answer from the generative AI model.

[1820] Step 14:

[1821] The server generates a UI to display multiple options for the answer of the generated AI model ("Make it the best answer," "Don't make it the best answer but add it to the page," "Don't add the AI ​​answer").

[1822] Step 15:

[1823] The device renders the UI with the options and redraws the question page to display to the user.

[1824] Step 16:

[1825] The user clicks on one of the choices.

[1826] Step 17:

[1827] The terminal detects the selection click event and sends the result to the server.

[1828] Step 18:

[1829] The server processes the answer of the generative AI model based on the selection results.

[1830] Step 19:

[1831] If you select "Set as best answer," the server will set the generated AI model's answer as the best answer in the database and display it on the question page.

[1832] Step 20:

[1833] If you select "Add to page without setting as best answer," the server will add the generated AI model's answer to the database and display it on the question page, but will not set it as the best answer.

[1834] Step 21:

[1835] If you select "Do not add AI answer," the server will not add the generated AI model's answer to the database or the question page, and will return the result to the device.

[1836] Step 22:

[1837] The terminal redraws the question page based on the results received from the server and displays the final state to the user.

[1838] The above is the specific program processing flow of this system.

[1839] Example 1

[1840] 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."

[1841] Conventional Q&A services have the problem that it takes a long time for questioners to receive an appropriate answer. Furthermore, because the quality and speed of the answer depends on the answerer, it is difficult to obtain an appropriate answer immediately to the question. Furthermore, the system's efficiency is low because question categorization and responses are done manually. Furthermore, conventional systems have limited options for automated answers using AI, leaving users with little flexibility in selecting AI answers.

[1842] 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.

[1843] In this invention, the server includes means for detecting that a question has been posted, means for analyzing the question data and classifying it into an appropriate category, means for dynamically adding a "See AI's Answer" button to the question page, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, and means for processing the answer of the generative AI model based on the choice made by the questioner. This allows the questioner to quickly obtain an appropriate answer, while allowing the answerer to focus on more advanced questions, improving the efficiency of the entire system.

[1844] The "means for detecting that a question has been posted" refers to a method or device for detecting that a user has entered a question on the Q&A service platform and clicked the post button.

[1845] "Means for analyzing question data and classifying it into appropriate categories" refers to a method or device that analyzes input question data using natural language processing (NLP) or the like, and classifies it into relevant categories based on its content.

[1846] The "means for dynamically adding a 'See AI's Answer' button to a question page" refers to a method or device for adding and displaying a 'See AI's Answer' button on a question page in real time after a question is posted.

[1847] "Means for sending a request to the generative AI model" refers to a method or device that sends the question data as a request to the API of the generative AI model when a user clicks the "View AI's answer" button.

[1848] "Means for receiving an answer from a generative AI model and displaying it on a question page" refers to a method or device for receiving an answer returned from a generative AI model and displaying it on a question page in a form that can be viewed by a user.

[1849] "Means for displaying multiple options to the questioner" refers to a method or device that displays options for the questioner in response to the answer of the generative AI model, such as "Make this the best answer," "Add this to the page without making it the best answer," or "Do not add this to the AI ​​answer."

[1850] "Means for processing the answer of a generative AI model based on the questioner's selection" refers to a method or device that processes the answer of a generative AI model in accordance with the questioner's selection from the options presented, such as saving it in a database, displaying it, or discarding it.

[1851] This invention is a system that uses a generative AI model to provide quick answers to questions in a Q&A service. This system is realized by combining various means that operate mainly on a server, a terminal, and a user.

[1852] First, a user enters a question on the Q&A service platform and clicks the "Post" button. The device detects this action and sends the entered question data to the server. The server analyzes the received question data and detects that it is a question. The server then analyzes the question data using a natural language processing (NLP) library (e.g., spaCy) and classifies it into the appropriate category.

[1853] Next, the server generates HTML code to dynamically add a "See AI's Answer" button to the question page and sends it to the terminal. The terminal receives this code and redraws the user's browser screen to display the "See AI's Answer" button. When the user clicks this button, the terminal detects the click event and sends the request to the server.

[1854] Based on the received request, the server sends question data as a request to the generative AI model's API (e.g., OpenAI). The generative AI model analyzes the sent question data and generates an answer. The generative AI model returns the generated answer to the server. The server formats the received answer data, generates HTML code to add to the question page, and sends it to the device. The device receives this updated data and displays the generative AI model's answer by redrawing the user's browser screen.

[1855] Furthermore, the server generates a UI to display multiple options that the user can select from for the generated AI model's answer (e.g., "Make this the best answer," "Add to the page without making this the best answer," and "Don't add the AI ​​answer"). The device redraws this UI and displays it on the page. When the user clicks on an appropriate option, the device sends this selection to the server. The server processes the generated AI model's answer based on the user's selection, for example, saving it as the best answer in a database or updating the page.

[1856] Specific examples

[1857] Prompt statement example 1:

[1858] A user posts a question to a Q&A service: "What is the appropriate amount of food for a cat?" The server detects the question and displays a "See AI's Answer" button. When the user clicks the button, the generative AI model generates and displays the answer: "For an adult cat, 2-3% of its body weight is appropriate." The user selects "Make this the best answer," and the server adds the answer to the database as the best answer.

[1859] Prompt statement example 2:

[1860] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?" The server detects the question and similarly displays a "See AI's answer" button. When the user clicks the button, the generative AI model generates an answer: "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1," and displays it. The user decides that the answer is insufficient and selects "Add to page without marking as best answer." The server adds the generative AI model's answer to the page, and expects other users to provide more detailed answers.

[1861] In this way, the system utilizes generative AI models to enable users to get fast and relevant answers, allowing other users to focus on more advanced questions, improving the overall efficiency of the system and providing a better user experience by providing options for questioners.

[1862] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1863] Step 1: Post a question

[1864] Users enter their questions on the Q&A service platform and click the "Post" button.

[1865] Input: Text data of the user's question (e.g., "What is the appropriate amount of food for a cat?")

[1866] The terminal receives user input, converts the question data into JSON format, and sends it to the server.

[1867] Output: Question data in JSON format

[1868] Step 2: Question detection and parsing

[1869] The server receives the JSON data and detects that a question has been posted.

[1870] Input: JSON format question data sent from the terminal

[1871] The server uses an NLP library (e.g., spaCy) to parse the question data and classify it into appropriate categories.

[1872] Output: Question analysis results and category information (e.g., "Pet category")

[1873] Step 3: Display of the "See AI's Answer" button

[1874] The server generates HTML code to dynamically add a "View AI Answer" button to the question page.

[1875] Input: User question, analysis results, category information

[1876] The server sends a response containing the generated HTML code to the terminal.

[1877] The device uses the updated data received from the server to redraw the user's browser screen, which displays a new "See AI's Answer" button.

[1878] Output: Updated HTML code

[1879] Step 4: Sending a request to the generative AI model

[1880] The user clicks the "See AI's answer" button.

[1881] Input: User click event

[1882] The terminal detects the click event and sends a request along with the question data to the server.

[1883] Output: Request data to the server

[1884] The server sends the question data as a request to the API of the generative AI model (e.g., OpenAI).

[1885] Step 5: Receive and display the answer from the generative AI model

[1886] The server receives the response from the generative AI model.

[1887] Input: Answer data from the generative AI model

[1888] The generative AI model analyzes the question data and generates an appropriate answer (e.g., "For adult cats, 2-3% of body weight is appropriate").

[1889] The server formats the generated answer and generates HTML code to add to the question page.

[1890] Output: Updated HTML code and generated AI answer data

[1891] The device receives the updated page data and redraws the user's browser screen, which displays the answer from the generated AI model.

[1892] Step 6: Displaying the options

[1893] The server generates a UI to display multiple options for the user to choose from in response to the generated AI model's answer.

[1894] Input: Answer data for the generative AI model

[1895] The server transmits the generated UI data to the terminal.

[1896] The device uses the UI data to display the options to the user.

[1897] Output: Updated UI with choices

[1898] Step 7: Processing based on user selection

[1899] The user clicks on the appropriate option.

[1900] Input: User click event

[1901] The terminal transmits the selection result to the server.

[1902] Output: Request data to the server

[1903] The server processes the generative AI model's answer based on the questioner's selection.

[1904] "Mark as Best Answer": Set the generative AI model's answer as the best answer in the database and display it on the page.

[1905] "Add to page without setting as best answer": Add the generative AI model's answer and display it on the page without setting it as the best answer.

[1906] "Don't add AI answer": The generated AI model's answer will not be added to the page or database.

[1907] The above are the specific processing steps and operations of the program for this system.

[1908] (Application example 1)

[1909] 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."

[1910] In recent years, with the spread of electronic payment services, the number of inquiries from users has increased dramatically. This has resulted in delayed support responses, leading to problems that lead to a decline in user satisfaction. Furthermore, support staff are required to respond to a large number of inquiries, increasing their workload. To solve this problem, a system that can provide prompt and appropriate responses is needed. However, existing systems have difficulty in providing prompt and appropriate responses to inquiries.

[1911] 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.

[1912] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, means for processing the answer from the generative AI model based on the questioner's selection, means for providing a user interface for inputting questions about electronic payments, and means for receiving the answer from the generative AI model and displaying it on the question page related to electronic payments. This allows users to receive quick and accurate answers to inquiries about electronic payments, improving the efficiency of support responses and reducing workloads.

[1913] The "means for detecting that a question has been posted" is a mechanism by which the system determines that a user has entered and posted a question on the system.

[1914] The "means for sending a request to the generative AI model" is a mechanism for sending the posted question to the generative AI model and sending a request to generate an appropriate answer.

[1915] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to a mechanism for receiving answers returned from the generative AI model and displaying them on the user interface.

[1916] "Means for displaying multiple options to the questioner" refers to a mechanism that displays multiple options so that the user can select further actions in response to the answer from the generative AI model.

[1917] The "means for processing the generative AI model's answer based on the questioner's selection" refers to a mechanism that, after the user selects an option, stores or displays the generative AI model's answer in a database based on that selection.

[1918] The "means for providing a user interface for inputting questions about electronic payments" is a mechanism for providing an interface designed to allow users to intuitively input questions about electronic payments.

[1919] "Means for receiving the answer of the generative AI model and displaying it on the question page related to electronic payments" refers to a mechanism for receiving the answer related to electronic payments returned from the generative AI model and displaying it on the relevant question page.

[1920] This invention is a system for quickly and accurately answering questions about electronic payments. The system consists of posting and detecting questions, sending requests to a generative AI model, receiving and displaying answers, presenting options to the user, and processing the answers based on the selections.

[1921] The core of the system resides in the server, which has the following main functions:

[1922] 1. Post and discover questions:

[1923] When a user enters a question about electronic payments on the smartphone application and clicks the "Post" button, the device sends the question data to the server, which detects the post, analyzes the received question data, and classifies it into the appropriate category.

[1924] 2. Display of "See AI's Answer" button:

[1925] When a question is submitted, the server dynamically adds a "See AI's Answer" button to the question page. After this button is added to the page, the device redraws the page and displays the new button to the user.

[1926] 3. Sending a request to the generative AI model:

[1927] When a user clicks the "View AI's Answer" button, the click event is detected by the device and a request is sent to the server. The server then sends the received question data as a request to the API of the generative AI model.

[1928] 4. Receive and display answers from the generative AI model:

[1929] The generative AI model analyzes the received question data and generates an appropriate answer. The generated answer is returned to the server, which formats the answer data and adds it to the question page. The device redraws the updated page and displays the generative AI model's answer to the user.

[1930] 5. Displaying options to the questioner:

[1931] The server generates a UI to display multiple options for the questioner to choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Don't add the AI ​​answer"), and the device redraws the UI to display the options to the user.

[1932] 6. Processing the generative AI model's answers based on your choices:

[1933] When the user clicks on the appropriate option, the device sends the selection to the server, which then takes appropriate action on the generative AI model's answer based on the user's selection, such as setting it as the best answer, adding it to the page, or ignoring it.

[1934] The following hardware and software are used to realize this system.

[1935] 1. Smartphone application:

[1936] Platform: iOS and Android

[1937] Development language: Swift (iOS), Kotlin (Android)

[1938] 2. Server:

[1939] Platform: Linux-based server environment

[1940] Software used: Node.js, Django, Flask

[1941] 3. Generative AI Model:

[1942] Model Name: OpenAI GPT-3 or GPT-4

[1943] API used: OpenAI API

[1944] As a concrete example, if a question is posted asking "Where can I check my credit card statement?", the system will operate as follows:

[1945] 1. The server receives the question and adds a "See AI's Answer" button to the question page.

[1946] 2. When the user clicks this button, the question data is sent to the generative AI model.

[1947] 3. The generative AI model generates the answer, "You can check your credit card statement on the 'Statement' page of the official app or website," and returns it to the server.

[1948] 4. The server displays this answer on the question page and gives the user the options of "Make it the best answer," "Add it to the page without making it the best answer," or "Don't add the AI ​​answer."

[1949] 5. If the user selects "Mark as Best Answer," the answer is saved in the database and set as the best answer on the question page.

[1950] This system will enable users to quickly obtain information about electronic payments and is expected to reduce the burden on support staff.

[1951] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1952] Step 1: Posting and discovering questions

[1953] A user enters a question about electronic payment into a smartphone application and clicks the "Post" button. At this time, the question text entered by the user is generated. The device sends this question text to the server. The server analyzes the received question data and detects that a question has been posted. Next, it classifies the question data into appropriate categories. This data processing process organizes the question content into categories.

[1954] Input: The question text entered by the user

[1955] Data processing: categorization of question data

[1956] Output: Question data with categories

[1957] Step 2: Displaying the "See AI's Answer" button

[1958] The server dynamically adds a "See AI's Answer" button to the question page based on the question data it received. It then instructs the device to redraw the updated page and display the new button to the user. This operation allows the user to expect a quick answer from the AI.

[1959] Input: Question data with categories

[1960] Data processing: Update the question page and add a button

[1961] Output: Updated question page with a "See AI's answer" button

[1962] Step 3: Sending a request to the generative AI model

[1963] When the user clicks the "View AI Answer" button, the device detects this click event. Based on this detection, the device resends the question data to the server. The server then sends the received question data as a request to the generative AI model's API. This data transmission causes the generative AI model to begin analyzing the question content.

[1964] Input: Click event of the "View AI's answer" button and question data

[1965] Data processing: Sending a request to the generative AI model API

[1966] Output: The question request received by the generative AI model

[1967] Step 4: Receive and display the answer from the generative AI model

[1968] The generative AI model analyzes the received question data and generates an appropriate answer. This answer is returned to the server. The server formats the received answer data and adds it to the question page. The device then redraws the updated page and displays the generative AI model's answer, allowing the user to see the answer provided by the AI ​​model.

[1969] Input: Answer data generated by a generative AI model

[1970] Data processing: Formatting of response data

[1971] Output: Question page with formatted answer data

[1972] Step 5: Present the options to the questioner

[1973] The server generates a UI to display multiple options for the questioner to choose from in response to the generated AI model's answer ("Make this the best answer," "Add to the page without making this the best answer," "Don't add the AI ​​answer") The device redraws this UI and displays it to the user.

[1974] Input: Answer data for the generative AI model

[1975] Data processing: generating options and updating the UI

[1976] Output: Question page with updated UI

[1977] Step 6: Processing the generative AI model's answer based on your selection

[1978] When the user clicks on the appropriate option, the device detects this click event and sends the selection result to the server. Based on the selection, the server saves the generative AI model's answer in a database or displays it on the page. For example, if the user selects "Set as best answer," the answer is set as the best answer in the database and displayed on the question page.

[1979] Input: User selection

[1980] Data processing: Save answers to database and update page display

[1981] Output: Updated question page and database state

[1982] 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.

[1983] This invention is a system that provides quick and appropriate answers to questions in a Q&A service using a generative AI model and an emotion engine. The configuration of this system and the program processing flow are described in detail below.

[1984] System configuration

[1985] The system mainly consists of the following components:

[1986] 1. Question Post Detection

[1987] The server has a means to detect new questions posted by users.

[1988] 2. Sending a request to the generative AI model

[1989] The server has the ability to send the detected questions to the generative AI model.

[1990] 3. Receiving and displaying answers from the generative AI model

[1991] The server has a means to receive the answer from the generative AI model and display it on the question page.

[1992] 4. Emotion Engine

[1993] The server has the means to recognize the questioner's emotions and analyze that information.

[1994] The emotion engine adjusts the generative AI model's responses based on the questioner's emotional state.

[1995] 5. Display of questioner's options

[1996] The server has a means to display a UI that offers the questioner multiple options for the answer of the generated AI model.

[1997] 6. Processing Choice-Based Answers

[1998] The server has a means to process the generative AI model's answer based on the questioner's selection.

[1999] The server has the function of storing the emotion information recognized by the emotion engine in a database.

[2000] 7. Learning the Emotion Engine

[2001] The server has a learning mechanism to improve emotion recognition accuracy based on feedback from the questioner.

[2002] Program flow

[2003] Posting and discovering questions

[2004] A user enters a question into the Q&A service's input form and clicks the "Post" button. The device sends the entered question data to the server. The server saves the question in a database, detects that the question has been posted, and analyzes the questioner's emotional information using an emotion engine.

[2005] Display of "See AI's answer" button

[2006] The server generates a script that dynamically adds a "See AI's Answer" button to the question page. The device receives this script, redraws the question page, and displays the "See AI's Answer" button.

[2007] Linking generative AI models and emotion engines

[2008] The user clicks the "See AI's Answer" button. The device detects the click event and sends the request to the server. The server sends the question data to the generative AI model, which then generates an answer.

[2009] The emotion engine adjusts the generated answer based on the emotional state of the questioner. The server receives the adjusted answer and displays it on the question page.

[2010] Displaying options to the questioner

[2011] The server generates a UI for the questioner to choose from three options for the AI ​​model's answer: "Make it the best answer," "Add it to the page without making it the best answer," or "Don't add it to the AI ​​answer." The device redraws the page and displays the options to the questioner.

[2012] Processing questioner selections and answers

[2013] The user clicks on one of the options. The device sends the selection to the server, which processes the answer of the generative AI model based on the selection.

[2014] Mark as best answer: The server saves the generated AI model's answer in the database as the best answer and displays it on the page.

[2015] Add to page without marking as best answer: The server adds the generated AI model's answer to the page, making it visible to other users.

[2016] Do not add AI answer: The server does not add the generated AI model answer to the page or database, but leaves it as is.

[2017] Emotion Engine Learning

[2018] Based on the feedback from the questioner, the server performs a learning process to improve the emotion recognition accuracy of the emotion engine.

[2019] Specific examples

[2020] Example 1: A simple question

[2021] A user posts a question: "What is the appropriate amount of food for a cat?" The server detects the question and displays a "See AI's Answer" button. When the user clicks the button, the generative AI model generates the answer: "For an adult cat, 2-3% of its body weight is appropriate," and the emotion engine appropriately adjusts the answer based on the questioner's emotional information and displays it. The user selects "Make Best Answer," and the server sets the answer as the best answer.

[2022] Example 2: A more complex question

[2023] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?" The server detects the question and displays a "See AI's answer" button. When the user clicks the button, the generative AI model generates the answer: "The procedure is to integrate the wave function over the entire space and make the square of the absolute value of that result 1." The emotion engine analyzes the questioner's emotions and adjusts the answer. The user selects "Add to page without marking as best answer," and the server adds the generative AI model's answer to the page.

[2024] In this way, the system of the present invention uses a generative AI model and an emotion engine in combination, allowing the questioner to receive a quick and appropriate answer while allowing the answerer to focus on asking more advanced questions.

[2025] The processing flow will be explained below.

[2026] Step 1:

[2027] The user enters a question into the Q&A service's input form and clicks the "Post" button.

[2028] Step 2:

[2029] The terminal transmits the posted question data to the server.

[2030] Step 3:

[2031] The server stores the received question data in a database and detects that a question has been posted.

[2032] Step 4:

[2033] The server calls the emotion engine and analyzes the questioner's emotional information based on the question data.

[2034] Step 5:

[2035] The server stores the emotion analysis results in a database.

[2036] Step 6:

[2037] The server generates a script that dynamically adds a "View AI Answer" button to the question page.

[2038] Step 7:

[2039] The device receives the script sent from the server, redraws the question page, and displays the "See AI's Answer" button.

[2040] Step 8:

[2041] The user clicks the "See AI's answer" button.

[2042] Step 9:

[2043] The terminal detects the button click event and sends the request to the server.

[2044] Step 10:

[2045] The server sends the question data and emotion information from the emotion engine as a request to the API of the generative AI model.

[2046] Step 11:

[2047] The generative AI model analyzes question data and emotional information to generate appropriate answers.

[2048] Step 12:

[2049] The generative AI model returns the generated answer to the server.

[2050] Step 13:

[2051] The server receives the generated response data and formats it appropriately.

[2052] Step 14:

[2053] The server adds HTML containing the formatted answer to the question page.

[2054] Step 15:

[2055] The device renders the HTML received from the server and redraws the question page to display the answer from the generative AI model.

[2056] Step 16:

[2057] The server generates a UI that provides multiple options for the answer of the generated AI model.

[2058] Step 17:

[2059] The device renders the UI with the options and redraws the question page to display to the user.

[2060] Step 18:

[2061] The user can choose to either "Make this the best answer," "Add it to the page without making it the best answer," or "Do not add the AI ​​answer."

[2062] Step 19:

[2063] The terminal transmits the user's selection to the server.

[2064] Step 20:

[2065] The server processes the answer of the generative AI model based on the selection results.

[2066] Step 21:

[2067] If you select "Set as best answer," the server will set the generated AI model's answer as the best answer in the database and display it on the question page.

[2068] Step 22:

[2069] If you select "Add to page without setting as best answer," the server will add the generated AI model's answer to the database and display it on the question page, but will not set it as the best answer.

[2070] Step 23:

[2071] If you select "Do not add AI answer", the server will not add the generated AI model's answer to the database or the question page.

[2072] Step 24:

[2073] The server sends feedback from the questioner to the emotion engine, which then learns to improve emotion recognition accuracy.

[2074] Step 25:

[2075] The terminal redraws the question page based on the final result received from the server and displays it to the user.

[2076] The above is the specific program processing flow of this system.

[2077] Example 2

[2078] 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."

[2079] With conventional Q&A services, it was difficult for questioners to obtain prompt and appropriate answers. Furthermore, the answers generated by the AI ​​model were not properly adjusted according to the questioner's feelings, resulting in low satisfaction with the answer. Furthermore, there was a lack of a way to clearly indicate which option the questioner should take in response to the question posted, which did not improve the user experience.

[2080] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[2081] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for analyzing the emotion of the questioner, means for adjusting the content of the answer of the generative AI model based on the analyzed emotion, means for displaying multiple options to the questioner, means for processing the answer of the generative AI model based on the selection of the questioner, and means for storing the emotion information in a database. This allows the questioner to receive a quick and appropriate answer, and further enables the answer to be adjusted according to the questioner's emotion, thereby improving the user experience.

[2082] "Question detection" is the process by which the server automatically recognizes questions posted by users.

[2083] "Sending a request to the generative AI model" is a means of sending information from the server to the generative AI model to analyze the detected question.

[2084] "Receiving and displaying answers from the generative AI model" refers to the process in which the server receives the answers provided by the generative AI model and displays them on the question page.

[2085] "Emotion analysis" is a means of determining the questioner's emotional state using an emotion engine.

[2086] "Adjusting response content based on emotions" is a method of appropriately changing the response content provided by the generative AI model based on analyzed emotional information.

[2087] "Displaying options" is the process by which the server provides a user interface with multiple options that the questioner can select from in response to the answer provided by the generative AI model.

[2088] "Choice-based answer processing" is the process by which the server appropriately processes the answer of the generative AI model depending on the option selected by the questioner.

[2089] "Storing emotional information" is a method of storing the emotional information analyzed by the emotion engine in a database.

[2090] This invention is a system that provides quick and appropriate answers to questions in a Q&A service using a generative AI model and an emotion engine. The configuration of this system and the program processing flow are described in detail below.

[2091] System configuration

[2092] The system consists of the following main components:

[2093] 1. Server

[2094] The server has a means for receiving and detecting queries from users.

[2095] It has the ability to send question data to a generative AI model.

[2096] It has a means to receive answers from the generative AI model and display them on the question page.

[2097] It integrates an emotion engine to analyze the questioner's emotions.

[2098] It has the ability to adjust the answers of the generative AI model based on the analyzed emotional information.

[2099] It has a means to display multiple options in the user interface.

[2100] It has a means to process the generative AI model's answers based on the choices.

[2101] It has the function of storing emotional information in a database.

[2102] 2. Terminal

[2103] The terminal provides an interface where the user can enter a question and click a "Post" button.

[2104] It has a means for sending question data to the server.

[2105] It has the means to receive and execute a script to dynamically display a "View AI Answer" button on the question page.

[2106] It has the function of sending the selection results to the server.

[2107] 3. Users

[2108] Users post questions using the input form of the Q&A service.

[2109] Click the "See AI's Answer" button to see the answer from the generated AI model.

[2110] Make the appropriate choice from multiple options.

[2111] Specific examples

[2112] Example 1: A simple question

[2113] Example prompt sentence:

[2114] What is the appropriate amount of cat food?

[2115] A user posts a question: "What is the right amount of food for a cat?"

[2116] The server detects the question and analyzes the questioner's emotional information using an emotion engine.

[2117] The server sends a script to the terminal that displays a "See AI's answer" button.

[2118] When the user clicks the "See AI Answer" button, the server sends the question data to the generative AI model and receives the answer.

[2119] For example, the generative AI model generates an answer such as "2-3% of body weight is appropriate for an adult cat," which is received by the server.

[2120] The emotion engine takes into account the emotional state of the questioner and adjusts the answer accordingly.

[2121] The server displays the adjusted answer on the question page.

[2122] The user selects "Mark as Best Answer" and the server saves the answer as the best answer in the database.

[2123] Example 2: Complex Question

[2124] Example prompt sentence:

[2125] "What is the procedure for normalizing wave functions in quantum mechanics?"

[2126] A user posts a question: "What is the procedure for normalizing a wave function in quantum mechanics?"

[2127] The server detects the question and analyzes the questioner's emotional information using an emotion engine.

[2128] The server sends a script to the terminal that displays a "See AI's answer" button.

[2129] When the user clicks the "See AI Answer" button, the server sends the question data to the generative AI model and receives the answer.

[2130] For example, the generative AI model generates an answer such as, "The procedure is to integrate the wave function over the entire space and make the square of its absolute value 1," and the server receives this.

[2131] The emotion engine analyzes the questioner's emotions and adjusts the response appropriately.

[2132] The server displays the adjusted answer on the question page.

[2133] The user selects "Add to page without marking as best answer" and the server adds the answer to the page.

[2134] In this way, the system of the present invention can use generative AI models and emotion engines to provide users with fast and relevant answers.

[2135] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2136] Specific explanation of program processing

[2137] Step 1:

[2138] The user enters a question into the Q&A service's input form and clicks the "Post" button.

[2139] Input: The question text entered by the user.

[2140] Output: The query data is sent to the server.

[2141] Specific behavior:

[2142] The terminal obtains the contents of the user input form.

[2143] The terminal generates a request to transmit the question data to the server.

[2144] The device sends a request to the server.

[2145] Step 2:

[2146] The server stores the received question data in a database and detects when a new question has been posted.

[2147] The server sends the question to the emotion engine and analyzes the questioner's emotional information.

[2148] Input: Question data sent from the terminal.

[2149] Output: The question data is stored in a database and the sentiment analysis results are obtained.

[2150] Specific behavior:

[2151] The server writes the received question data to the database.

[2152] The server monitors the database to detect new question submissions.

[2153] The server sends the question to the emotion engine and receives the analysis results.

[2154] Step 3:

[2155] The server generates a script that adds a "See AI's Answer" button to the question page and sends it to the terminal.

[2156] Input: The event that a new question is posted.

[2157] Output: The script is sent to the terminal and the button is displayed.

[2158] Specific behavior:

[2159] The server generates script code to add a "See AI's Answer" button.

[2160] The server sends the generated script to the terminal.

[2161] Step 4:

[2162] The device receives the script, redraws the question page, and displays the "See AI's Answer" button.

[2163] Input: The script sent by the server.

[2164] Output: A "See AI's answer" button will be displayed on the question page.

[2165] Specific behavior:

[2166] The terminal parses the script received from the server.

[2167] The terminal executes the received script and redraws the question page.

[2168] The device will display a "See AI's Answer" button.

[2169] Step 5:

[2170] The user clicks the "See AI's answer" button.

[2171] Input: User click event.

[2172] Output: A click event is sent to the server.

[2173] Specific behavior:

[2174] The device detects a click event.

[2175] The device sends the click event information to the server.

[2176] Step 6:

[2177] The server sends the question data to the generative AI model, which then generates an answer.

[2178] Input: Click event information and original question data.

[2179] Output: Answer data from the generative AI model.

[2180] Specific behavior:

[2181] The server creates a request to send the original question data to the generative AI model.

[2182] The server calls the API of the generated AI model and sends the question data.

[2183] The generative AI model analyzes the question and generates an appropriate answer.

[2184] The answer data generated by the generative AI model is sent back to the server.

[2185] Step 7:

[2186] The emotion engine takes into account the emotional state of the questioner and adjusts the generated answer content.

[2187] Input: Response data from the generative AI model and pre-obtained sentiment analysis results.

[2188] Output: Adjusted response data.

[2189] Specific behavior:

[2190] The server sends the response data from the generative AI model and the emotion analysis results to the emotion engine.

[2191] The emotion engine adjusts the response content.

[2192] The emotion engine sends the adjusted response data back to the server.

[2193] Step 8:

[2194] The server receives the adjusted answer and displays it on the question page.

[2195] Input: Calibrated response data from the sentiment engine.

[2196] Output: The answer displayed on the question page.

[2197] Specific behavior:

[2198] The server receives the adjusted response data.

[2199] The server generates data for updating the question page and sends it to the terminal.

[2200] Step 9:

[2201] The device will redraw the question page and display the adjusted answer.

[2202] Input: Data to update from the server.

[2203] Output: Redrawn question page and displayed answer.

[2204] Specific behavior:

[2205] The terminal analyzes the data received from the server and redraws the question page.

[2206] The device will display the adjusted answer.

[2207] Step 10:

[2208] The server generates a UI containing three options that the questioner can choose from in response to the generated AI model's answer: "Make this the best answer," "Add it to the page without making it the best answer," or "Do not add the AI ​​answer," and sends this UI to the device.

[2209] Input: Adjusted response data.

[2210] Output: The UI shown to the questioner.

[2211] Specific behavior:

[2212] Create a script that the server uses to generate a UI with the options.

[2213] The server sends the generated script to the terminal.

[2214] Step 11:

[2215] The terminal will redraw and show the options to the questioner.

[2216] Input: Script from server.

[2217] Output: The choices presented to the questioner.

[2218] Specific behavior:

[2219] The terminal parses the script received from the server.

[2220] The device will redraw the question page and display the options.

[2221] Step 12:

[2222] The user clicks on one of the displayed options.

[2223] Input: User click event.

[2224] Output: Click event information is sent to the server.

[2225] Specific behavior:

[2226] The device detects a click event.

[2227] The device sends the click event information to the server.

[2228] Step 13:

[2229] The server processes the generative AI model's answer based on the selection results.

[2230] Input: Click event information.

[2231] Output: Processed response data.

[2232] Specific behavior:

[2233] The server appropriately processes the answer of the generative AI model based on the click event information.

[2234] If it is selected as the best answer, the server saves the generated AI model's answer as the best answer in the database and displays it on the page.

[2235] If you add the answer to a page without marking it as the best answer, the server adds the answer from the generated AI model to the page, making it available for other users to view.

[2236] If you do not add an AI answer, the server will not add the generated AI model's answer to the page or database, but will leave it as is.

[2237] Step 14:

[2238] The server collects feedback from the questioner and uses it as learning data for the emotion engine.

[2239] Input: Feedback from the questioner.

[2240] Output: Improved recognition accuracy through emotion engine training.

[2241] Specific behavior:

[2242] The server provides an interface for collecting feedback data from questioners.

[2243] The server sends the collected feedback data to the emotion engine.

[2244] The emotion engine learns from feedback data to improve emotion recognition accuracy.

[2245] (Application example 2)

[2246] 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."

[2247] Conventional Q&A services respond slowly to questions, and the answers are often inadequate. Furthermore, answers are provided without considering the feelings of the questioner, resulting in low user satisfaction. Furthermore, answers are provided one-way, making it difficult to improve the service based on user feedback. A system that can solve these problems and provide fast and appropriate answers is needed.

[2248] 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.

[2249] In this invention, the server includes means for detecting that a question has been posted, means for sending a request to the generative AI model, means for receiving an answer from the generative AI model and displaying it on the question page, means for displaying multiple options to the questioner, means for processing the answer of the generative AI model based on the choice of the questioner, means for analyzing emotional information of the questioner using emotion recognition means and adjusting the answer of the generative AI model based on that information, means for collecting feedback from the questioner on whether the answer to the question was appropriate, and means for training the emotion recognition means to improve emotion recognition accuracy. This makes it possible to provide quick and appropriate answers to questions, as well as adaptive answers that take the questioner's emotions into consideration, thereby improving user satisfaction.

[2250] The "means for detecting that a question has been posted" is a function that, when a user enters a question into the input form and clicks the submit button, sends the question data to the server and recognizes that a new question has been posted.

[2251] The "means for sending a request to a generative AI model" is a function for sending a detected question to a generative AI model and sending a request to obtain an appropriate answer to the question.

[2252] "Means for receiving answers from the generative AI model and displaying them on the question page" refers to a function that receives answers generated by the generative AI model and displays them on the question page in a format that is easy for users to view.

[2253] "Means for displaying multiple options to the questioner" refers to a function that displays a user interface that provides multiple options for the questioner to select in response to the answer of the generative AI model.

[2254] "Means for processing the generative AI model's answer based on the questioner's selection" refers to a function that optimally processes the generative AI model's answer based on the option selected by the questioner.

[2255] "Means for analyzing the emotional information of the questioner using emotion recognition means and adjusting the response of the generative AI model based on that information" refers to a function that analyzes the emotional state of the questioner and appropriately adjusts the content of the response in accordance with that emotion.

[2256] "Means for collecting feedback from the questioner on whether the answer to the question was appropriate" is a function that collects direct feedback from the questioner on whether the answer provided by the generative AI model was appropriate.

[2257] "Means for training the emotion recognition means to improve the emotion recognition accuracy" is a function that improves and trains the accuracy of the emotion recognition engine based on feedback from the questioner.

[2258] This invention is a system that provides prompt and appropriate answers in a Q&A service by combining a generative AI model and emotion recognition means. The configuration of this system and the program processing flow are described in detail below.

[2259] System configuration

[2260] The system consists of the following components:

[2261] 1. How to detect question posts

[2262] When a user enters a question into the input form and clicks the send button, the terminal sends the entered question data to the server.

[2263] 2. How to send requests to the generative AI model

[2264] The server sends the detected question to a generative AI model, which then generates an answer to that question.

[2265] 3. How to receive and display answers from the generative AI model

[2266] This is a function where the server receives the answer from the generative AI model and displays it on the question page.

[2267] 4. Emotion recognition means

[2268] This function analyzes the emotions of the questioner and adjusts the response content of the generative AI model based on that. This method analyzes the emotional state of the questioner and adjusts the response content to be appropriate for the questioner.

[2269] 5. A way to present multiple options to the questioner

[2270] This function displays a user interface to the questioner in response to the answer generated by the server using the AI ​​model, with options such as "I'm satisfied with this answer," "Please tell me more," or "The answer is inappropriate."

[2271] 6. Feedback Collection Methods

[2272] This is a method of collecting data to improve emotion recognition accuracy based on feedback from questioners, which enables the emotion recognition method to perform more accurate emotion analysis.

[2273] Program processing description

[2274] Question Post Detection

[2275] When a user enters a question into the Q&A service's input form and clicks the "Send" button, the device sends the entered question data to the server. The server saves the question in a database and detects that the question has been posted. At this time, the question data is sent to the emotion recognition means, which analyzes the questioner's emotions.

[2276] Sending requests to generative AI models

[2277] The server sends the detected question to the generative AI model, which generates an answer to the question and returns the answer to the server.

[2278] Receiving and adjusting answers from generative AI models

[2279] The server receives the answer from the generative AI model and adjusts it using emotion recognition to take into account the emotional state of the questioner. The adjusted answer is then displayed on the question page.

[2280] Displaying options to the questioner

[2281] The server displays the adjusted answer to the questioner along with the options of "I'm satisfied with this answer," "Please tell me more," or "The answer is not appropriate." When the questioner selects an option, the generative AI model's answer is processed accordingly.

[2282] Gathering feedback and learning

[2283] The feedback provided by the questioner is collected by the server, and this data is used as training data to improve the accuracy of the emotion recognition means.

[2284] Specific examples

[2285] 1. Simple Questions

[2286] A user posts a question such as "What is this product made of?" The server detects the question and sends it to the generative AI model. The answer from the generative AI model, "This product is made of cotton," is adjusted by the emotion recognition means and displayed on the question page. When the user selects "I'm satisfied with this answer," feedback is collected that the answer was appropriate.

[2287] 2. Complex Questions

[2288] A user posts a question such as "Please tell me the detailed manufacturing process of this product." The server detects the question and sends it to the generative AI model. The answer from the generative AI model, "This product is manufactured through multiple processes," is adjusted using emotion recognition and displayed on the question page. The user can select "Tell me more" to provide further details.

[2289] Prompt Sentence Examples

[2290] The user enters a question such as:

[2291] "What is this item made of?"

[2292] Example prompts to send to the generative AI model:

[2293] "The user is asking about the materials used in this product. The materials used in this product are as follows:"

[2294] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2295] Step 1:

[2296] A user enters a question into the Q&A service's input form and clicks the "Submit" button. The input is the question text entered by the user. The terminal sends this question data to the server. The output is the question data sent to the server.

[2297] Step 2:

[2298] The server receives new question data and stores it in a database. The input is the question data sent from the device. The server detects that a question has been posted and analyzes the questioner's emotional information using emotion recognition means. The output is the question data and the analyzed emotional information.

[2299] Step 3:

[2300] The server sends question data to the generative AI model. The input is the question and emotion information detected and analyzed by the server. The server sends a request to the generative AI model, which then generates an answer to the question. The output is the answer from the generative AI model.

[2301] Step 4:

[2302] The server receives the answer from the generative AI model and adjusts the answer using the emotion recognition means. The input is the answer from the generative AI model and pre-analyzed emotional information. The emotion recognition means adjusts the answer according to the questioner's emotions, and the server obtains the final adjusted answer. The output is the adjusted answer.

[2303] Step 5:

[2304] The server displays the adjusted answer on the question page. The input is the adjusted answer. The server displays this answer on the question page and generates a user interface with the options "I'm happy with this answer," "Please tell me more," and "This answer is not appropriate." The output is the answer and options that the user can view.

[2305] Step 6:

[2306] The user clicks on one of the options. The input is the option selected by the user. The terminal sends the selection to the server. The output is the selection sent to the server.

[2307] Step 7:

[2308] The server processes the generative AI model's answer based on the user's selection and stores it in a database if necessary. The inputs are the user's selection and the generative AI model's answer. Based on the selection, the server saves the answer as the "best answer," adds it to the page, or does nothing and exits. The output is the result saved in the database or an update to the page.

[2309] Step 8:

[2310] The server starts training the emotion recognition means based on feedback from the user. The input is the user's feedback data. The server uses this to train the emotion recognition means to improve its accuracy. The output is an emotion recognition means with improved accuracy.

[2311] 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.

[2312] 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.

[2313] 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.

[2314] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion i...

Claims

1. a means for detecting when a question has been posted; a means for sending requests to the generative AI model; A means to receive the answer from the generative AI model and display it on the question page; and a means for displaying multiple options to a questioner; a means for processing the generative AI model's answer based on the questioner's selection; and A system including:

2. The system of claim 1 , further comprising: means for displaying a “See AI Answer” button on the page when a question is posted.

3. The system of claim 1 , further comprising means for storing the answer of the generative AI model in a database and setting the answer as a best answer based on a selection by the questioner.

Images