System

Abstract

To provide a system for utilizing expert knowledge and opinions of a user as learning data.SOLUTION: Means for transmitting question data input by a user to a server, means for storing the received question data in a database, means for providing the question data stored in the database to an artificial intelligence model, means for transmitting answer data input by another user to the server, means for storing the received answer data in the database, means for providing the answer data stored in the database to the artificial intelligence model, means for browsing the question and the answer by the user, means for transmitting evaluation data input by the user to the server, means for storing the received evaluation data in the database, and means for providing the evaluation data to the artificial intelligence model and performing relearning, and means for adjusting learning content of the artificial intelligence model based on the feedback data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Conventional artificial intelligence systems could not directly incorporate users' expertise or opinions, and only provided information unilaterally. As a result, the high level of expertise possessed by users could not be efficiently utilized, limiting the accuracy and usefulness of the AI. There was also a need for a system that would allow users to directly participate in the AI's development process, enabling the provision of more accurate and useful information. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system that allows users to directly provide their expertise and opinions to an AI and utilizes them as training data. Specifically, a user interface is provided that allows users to input questions, and the input question data is received and stored on a server. The saved question data is then provided to an AI model for training. Other users are then allowed to input answers to the questions, and the answer data is also received and stored on the server and provided to the AI ​​model for training. An interface is provided that displays questions and answers, allowing users to view them, and also accepts evaluations of the answers. The evaluation data is received and stored on the server and periodically provided to the AI ​​model for re-training. This allows the AI ​​model's learning content to be adjusted based on user feedback, improving accuracy and usefulness.

[0006] "User interface" refers to the input and display screens that allow users to input and view questions, answers, and ratings.

[0007] The term "terminal" refers to a device that provides a user interface and transmits input data to a server.

[0008] "Server" refers to a computer system that receives, stores, and provides user-submitted questions, answers, and evaluation data to an artificial intelligence model.

[0009] "Database" refers to a storage system for systematically storing and managing question data, response data, and evaluation data.

[0010] An "artificial intelligence model" refers to an AI system that learns from question and answer data provided by users, accumulates knowledge, and grows and improves.

[0011] "Question data" refers to data including the question entered by the user and the metadata associated with the question.

[0012] "Answer data" refers to data including the answer entered by the user to the question and the metadata associated with the answer.

[0013] "Evaluation data" refers to data including evaluations such as "helpful" or "not helpful" entered by users in response to answers provided by them, as well as metadata associated with such evaluations.

[0014] "Retraining" refers to the process of retraining an AI model that has already completed training by incorporating newly added questions, answers, and evaluation data.

[0015] "Feedback" refers to the evaluations and opinions that users give to questions and answers, and includes data used to adjust the learning of AI models. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

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

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] This invention is an interactive system that allows users to develop an artificial intelligence model through questions, answers, and ratings. Its details include a process for effectively collecting and storing information entered by users and providing it to the AI ​​model for learning. The specific form and processing flow of the system are described below.

[0038] Enter and save your question

[0039] Enter your question

[0040] Users can input their interests and doubts as questions through the user interface of their terminal.

[0041] Example: User A types, "What's the latest anime-related news?"

[0042] Submit a Question

[0043] The terminal transmits the question data entered by the user to the server.

[0044] This data is sent in a format such as JSON and includes metadata such as the user ID, question, and timestamp.

[0045] Saving Question Data

[0046] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified.

[0047] Providing questions to the AI ​​model

[0048] The server provides the stored question data to the AI ​​model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[0049] Enter and save your answers

[0050] Enter your answer

[0051] A user inputs an answer to a question posted by another user through a user interface.

[0052] Example: User B replies, "It has been announced that 'Work X' will be made into a movie."

[0053] Submit your answer

[0054] The device sends the answer data entered by the user to the server. The answer data is also sent in JSON format and includes the user ID, question ID, answer content, timestamp, etc.

[0055] Saving response data

[0056] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[0057] Providing answers to AI models

[0058] The server provides the stored response data to the AI ​​model, which then learns from it, allowing the AI ​​to accumulate more knowledge and improve accuracy.

[0059] View questions and answers

[0060] Ask questions and get answers

[0061] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[0062] View questions and answers

[0063] The device displays the received data on the user interface. It is also possible to display newer or more popular answers preferentially.

[0064] Enter and save your rating

[0065] Enter your rating

[0066] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[0067] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[0068] Submit your rating

[0069] The terminal sends the entered evaluation data to the server. The evaluation data includes the user ID, answer ID, evaluation content, timestamp, etc.

[0070] Saving evaluation data

[0071] The server stores the received evaluation data in a database, and this evaluation data is also assigned a unique ID.

[0072] AI learning and accuracy improvement

[0073] Periodic Relearning

[0074] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for re-training, thereby improving the accuracy of the AI ​​model.

[0075] Reflecting feedback

[0076] The server automatically adjusts the AI ​​model's learning based on user evaluation data, thereby correcting incorrect answers and strengthening correct answers.

[0077] Specific examples

[0078] For example, if user A posts a question such as "What's the latest anime-related news?", the server saves this question in a database and provides it to the AI ​​model for learning. Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also saved in the database and provided to the AI ​​model. If user C views this question and answer and rates it as "helpful," this rating data is also saved and used to retrain the AI ​​model.

[0079] In this way, users can actively participate in the growth of the AI, and the system provides highly accurate information that reflects their own knowledge.

[0080] The processing flow will be explained below.

[0081] Post a Question

[0082] Step 1:

[0083] User enters a question

[0084] Users enter their questions using a web browser or mobile app user interface.

[0085] Example: User A types, "What's the latest anime-related news?"

[0086] Step 2:

[0087] Submit your question

[0088] The terminal transmits the question entered by the user to the server.

[0089] The transmitted data includes metadata such as the user ID, question content, and timestamp.

[0090] Example: {"User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}

[0091] Step 3:

[0092] Save questions

[0093] The server stores the received question data in a database.

[0094] Give the question a unique ID (e.g. Q789) and store it in the database.

[0095] Example: Save in the database as {"Question ID": "Q789", "User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}.

[0096] Step 4:

[0097] Providing questions to the AI ​​model

[0098] The server provides the saved question data to the AI ​​model and uses it as learning data.

[0099] Example: Providing {"question": "What's the latest anime news?"} to an AI model.

[0100] Providing answers

[0101] Step 5:

[0102] User enters answer

[0103] A user inputs an answer to a question from another user.

[0104] Example: User B types, "It has been announced that 'Work X' will be made into a movie."

[0105] Step 6:

[0106] Sending answers

[0107] The terminal transmits the answer entered by the user to the server.

[0108] The transmitted data includes the user ID, question ID, answer content, timestamp, etc.

[0109] Example: {"User ID": "B456", "Question ID": "Q789", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}

[0110] Step 7:

[0111] Save answers

[0112] The server stores the received response data in a database.

[0113] Give the answer a unique ID (e.g. A987) and store it along with the question ID.

[0114] Example: Save in the database as {"Answer ID": "A987", "Question ID": "Q789", "User ID": "B456", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}.

[0115] Step 8:

[0116] Providing answers to the AI ​​model

[0117] The server provides the saved response data to the AI ​​model and uses it as learning data.

[0118] Example: Providing {"answer": "It has been announced that 'Book X' will be made into a movie."} to an AI model.

[0119] View questions and answers

[0120] Step 9:

[0121] Ask questions and get answers

[0122] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[0123] Example: Request data for {"Question ID": "Q789"} and receive question and answer data from the server.

[0124] Step 10:

[0125] View questions and answers

[0126] The terminal displays the received question and answer data on a user interface.

[0127] Example: When a user accesses the site, the question "What's the latest anime-related news?" and the answer "It has been announced that 'Work X' will be made into a movie" are displayed on the screen.

[0128] Evaluating answers

[0129] Step 11:

[0130] User-initiated rating input

[0131] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[0132] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[0133] Step 12:

[0134] Submit your evaluation

[0135] The terminal transmits the input evaluation data to the server.

[0136] The transmitted data includes the user ID, answer ID, evaluation details, timestamp, etc.

[0137] Example: {"User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}

[0138] Step 13:

[0139] Save evaluation details

[0140] The server stores the received evaluation data in a database.

[0141] Evaluation data will also be assigned a unique ID (e.g., E123).

[0142] Example: Store in database as {"Rating ID": "E123", "User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}.

[0143] AI learning and accuracy improvement

[0144] Step 14:

[0145] Periodic retraining of AI models

[0146] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[0147] Example: Retraining an AI model using all question and answer and rating data from the past month.

[0148] Step 15:

[0149] Reflecting feedback

[0150] The server automatically adjusts the learning content of the AI ​​model based on user evaluation data.

[0151] Example: Revisiting learning based on answers that were rated "not helpful" by many users.

[0152] Example 1

[0153] 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."

[0154] Conventional AI model learning systems have faced issues with inefficient management of user-entered questions, answers, and evaluation data, resulting in delayed improvements in the accuracy of AI models. Furthermore, there was a lack of a mechanism for immediately reflecting user feedback and using it to improve the model, making it difficult to improve the user experience. Furthermore, when displaying questions and answers, there was a lack of a function to prioritize the latest information and popular answers that users wanted, reducing the efficiency of information acquisition.

[0155] 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.

[0156] In this invention, the server includes a means for providing a user interface, a means for transmitting question data, answer data, and evaluation data to the server, a means for storing the received data in a database, a means for providing the data to the AI ​​model as learning data, a means for acquiring the question and answer data and displaying them on the user interface, and a means for preferentially displaying new or popular answers. This enables efficient collection, management, and provision of various data entered by users, resulting in rapid improvement in the accuracy of the AI ​​model and improved functionality of the user interface.

[0157] "User interface" refers to the screen or operating means through which a user can access a system and input and view information such as questions and answers.

[0158] "Question data" refers to data that includes information about the question entered by the user, including metadata such as the user ID, question content, and timestamp.

[0159] "Server" refers to a central computing device that receives, processes, stores, and provides data sent by users to AI models.

[0160] "Database" refers to a system in which a server systematically stores question data, answer data, evaluation data, etc. and manages them so that they can be retrieved as needed.

[0161] "Artificial intelligence model" refers to a machine learning algorithm or neural network that learns from user-provided data and answers questions, provides information, or generates responses.

[0162] "Answer data" refers to data that includes answers to questions posted by other users, and includes metadata such as user ID, question ID, answer content, and timestamp.

[0163] "Rating data" refers to data that includes information about the ratings that users have given to questions and answers, and includes metadata such as user ID, rating content, answer ID, and timestamp.

[0164] "Retraining" refers to the learning process in which questions, answers, and evaluation data stored in a database are periodically provided to an artificial intelligence model to improve the model's accuracy.

[0165] "Feedback data" refers to data used to adjust the model's learning content based on user evaluation data and improve the model's output results.

[0166] MODE FOR CARRYING OUT THE INVENTION

[0167] This invention is an interactive system in which users can develop an artificial intelligence model through questions, answers, and evaluations. The specific form and processing flow of this system will be described in detail below.

[0168] Enter and save your question

[0169] When a user types a question

[0170] A user accesses a dedicated user interface using their own terminal (for example, a personal computer or a smartphone) and inputs a question. An example of a question is "What is the latest anime-related news?"

[0171] Sending and storing question data

[0172] The device sends the question entered by the user to the server in JSON format, including the user ID, question content, timestamp, etc.

[0173] The server stores the received question data in a database (e.g., MySQL or PostgreSQL). A unique ID is assigned to each question, allowing it to be uniquely identified.

[0174] Providing question data to the AI ​​model

[0175] The server can provide the stored question data to the AI ​​model, which can then use the data as training data. This process allows the AI ​​model to learn new information from user questions and improve the accuracy of its responses to future questions.

[0176] Enter and save your answers

[0177] When the user enters an answer

[0178] Users can enter answers to questions posted by other users. An example of an answer might be, "It has been announced that 'Work X' will be made into a movie."

[0179] Sending and storing response data

[0180] The device sends the answers entered by the user to the server in JSON format, including the user ID, question ID, answer content, and timestamp.

[0181] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[0182] Providing response data to the AI ​​model

[0183] The server can provide the saved response data to the AI ​​model, which can then use the data as training data. This process allows the AI ​​model to learn new information from the user's responses and further improve the accuracy of future responses.

[0184] View questions and answers

[0185] Retrieving and Displaying Data

[0186] The device requests a list of questions and their answers from the server and receives the necessary data. The received data is displayed on the user interface. It is also possible to display new or popular answers preferentially, allowing users to quickly obtain the latest information and useful information.

[0187] Enter and save your rating

[0188] When a user enters a rating

[0189] Users can enter a rating for the displayed answer, such as "helpful" or "not helpful."

[0190] Sending and storing evaluation data

[0191] The device sends the entered evaluation data in JSON format to the server, which includes the user ID, answer ID, evaluation content, and timestamp.

[0192] The server stores the received rating data in a database, and each rating is assigned a unique ID.

[0193] AI learning and accuracy improvement

[0194] Periodic Relearning

[0195] The server periodically provides the question data, answer data, and evaluation data in the database to the AI ​​model for re-learning. This process improves the accuracy of the AI ​​model, enabling it to provide more accurate information to users.

[0196] Reflecting feedback

[0197] The server automatically adjusts the learning content of the AI ​​model based on the user's evaluation data, thereby correcting incorrect answers and strengthening correct answers.

[0198] Prompt Sentence Examples

[0199] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model. If user C views this question and answer and rates it as "helpful," this rating data is also saved and used to retrain the AI ​​model. In this way, the system allows users to actively participate in the growth and improvement of the AI's accuracy.

[0200] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0201] Step 1: User enters question

[0202] A user uses a terminal to access the user interface and inputs a question by typing the question in a text box and clicking a submit button.

[0203] Input: User ID, question, timestamp

[0204] Output: Question data (JSON format)

[0205] Step 2: The device sends the query data

[0206] The device converts the question data entered by the user into JSON format and sends it to the server. This data includes the user ID, question content, and timestamp.

[0207] Input: Question data (JSON format)

[0208] Output: Data packet to be sent

[0209] Step 3: The server receives the query data

[0210] The server receives the submitted question data and stores it in a database. As an additional process, it assigns a unique ID to each question.

[0211] Input: Data packet to send

[0212] Output: Question data stored in the database (with unique ID)

[0213] Step 4: The server provides the question data to the AI ​​model

[0214] The server extracts question data stored in the database and provides it to the AI ​​model for training. The AI ​​model uses this information to add new knowledge and improve the accuracy of future question answers.

[0215] Input: Question data stored in the database

[0216] Output: Provided as training data to the AI ​​model

[0217] Step 5: User Enters Answer

[0218] A user can enter an answer to a question posted by another user by selecting the question in the user interface, entering the answer in the text box that appears, and clicking the submit button.

[0219] Input: User ID, Question ID, Answer, Time Stamp

[0220] Output: Response data (JSON format)

[0221] Step 6: The device sends the response data

[0222] The device converts the answer data entered by the user into JSON format and sends it to the server. This data includes the user ID, question ID, answer content, and timestamp.

[0223] Input: Response data (JSON format)

[0224] Output: Data packet to be sent

[0225] Step 7: The server receives the response data

[0226] The server receives the submitted answer data and stores it in a database. As an additional process, it assigns a unique ID to each answer and associates it with the question ID.

[0227] Input: Data packet to send

[0228] Output: Answer data saved in the database (with unique ID)

[0229] Step 8: The server provides the answer data to the AI ​​model

[0230] The server extracts the response data stored in the database and provides it to an artificial intelligence model for learning. The AI ​​model uses this information to add new knowledge and improve the accuracy of future responses.

[0231] Input: Response data stored in the database

[0232] Output: Provided as training data to the AI ​​model

[0233] Step 9: The device requests data

[0234] The terminal requests question and answer data from the server in response to a user request, which is sent by clicking the "Show Questions and Answers" button on the user interface.

[0235] Input: Request data

[0236] Output: None (request sent)

[0237] Step 10: Server sends data

[0238] The server extracts the requested question and answer data from the database and sends it to the device in JSON format.

[0239] Input: Request data

[0240] Output: Question and answer data in JSON format

[0241] Step 11: The terminal displays the data

[0242] The terminal displays the question and answer data received from the server on a user interface, and can also display new or popular answers with priority.

[0243] Input: Question and answer data in JSON format

[0244] Output: Questions and answers displayed in the user interface

[0245] Step 12: User Enters Rating

[0246] The user inputs a rating for the displayed answer, such as "helpful" or "not helpful," and selects one and clicks the submit button.

[0247] Input: User ID, Answer ID, Evaluation details, Time stamp

[0248] Output: Evaluation data (JSON format)

[0249] Step 13: The device sends the evaluation data

[0250] The device converts the evaluation data entered by the user into JSON format and sends it to the server. This data includes the user ID, answer ID, evaluation content, and timestamp.

[0251] Input: Evaluation data (JSON format)

[0252] Output: Data packet to be sent

[0253] Step 14: Server receives reputation data

[0254] The server receives the submitted rating data and stores it in a database. A unique ID is assigned to each rating.

[0255] Input: Data packet to send

[0256] Output: Evaluation data stored in the database (with unique ID)

[0257] Step 15: Server periodically retrains

[0258] The server periodically provides the question data, answer data, and evaluation data in the database to the AI ​​model for re-learning, a process that improves the accuracy of the AI ​​model.

[0259] Input: Question data, response data, and evaluation data stored in the database

[0260] Output: Retrained AI model

[0261] Step 16: The server processes the feedback

[0262] The server automatically adjusts the AI ​​model's learning based on user evaluation data, improving the accuracy of answers and correcting incorrect answers.

[0263] Input: Evaluation data

[0264] Output: The tuned AI model

[0265] (Application example 1)

[0266] 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."

[0267] In recent years, as demand for interactive content distribution services has increased, AI models are being developed based on information provided by users, and there is a demand for more accurate information provision. However, current systems lack efficient collection and learning of user questions, answers, and evaluation data, resulting in slow AI model development. Furthermore, the user interface is difficult to use, making it difficult to improve user satisfaction.

[0268] 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.

[0269] In this invention, the server includes: means for providing a user interface through which a user can input a question; means for transmitting the input question data to the server; means for storing the received question data in a database; means for providing the question data stored in the database to the generative AI model; means for providing a user interface through which other users can input answers to questions; means for transmitting the input answer data to the server; means for storing the received answer data in the database; means for providing the answer data stored in the database to the generative AI model; means for providing a user interface through which users can view questions and answers; means for providing a user interface through which users can input ratings for answers; means for transmitting the input rating data to the server; means for storing the received rating data in the database; means for providing the questions, answers, and rating data in the database to the generative AI model and performing re-training; means for adjusting the learning content of the generative AI model based on feedback data; means for causing the generative AI model to generate prompt sentences based on the questions, answers, and rating data; and means for providing the generated prompt sentences to the generative AI model as further training data. This enables efficient growth of the AI ​​model and provision of highly accurate information through user interaction.

[0270] A "user interface" is an interface that provides a screen and input means for a user to interact with an application or system.

[0271] "Question data" refers to data including the content of a question that is input by a user through a user interface and sent to a server.

[0272] "Answer data" refers to data including the answer content that is input by other users to a question and sent to the server.

[0273] "Evaluation data" refers to data including evaluations such as "useful" or "not useful" that are entered by users in response to the answers and sent to the server.

[0274] A "generative AI model" is an artificial intelligence model that continuously learns based on questions, answers, and evaluation data provided by users, and provides highly accurate information.

[0275] "Server" refers to a computer system that receives question data, answer data, and evaluation data from users and stores and manages them in a database.

[0276] A "database" is a storage device for accumulating and managing question data, answer data, and evaluation data stored on a server.

[0277] "Feedback data" is data used to adjust the learning content of the generative AI model based on evaluations made by users.

[0278] A "prompt sentence" is training data generated by a generative AI model based on questions, answers, and evaluation data, and is a sentence used for further AI training.

[0279] "Interaction data" is a general term for data related to operations such as questions, answers, and ratings that users perform within the system.

[0280] The present invention provides an interactive system in which users can develop a generative AI model through questions, answers, and ratings. This system has a mechanism for effectively collecting and storing information provided by users and providing it to the generative AI model for learning. Detailed modes for implementing the invention are described below.

[0281] Generating a Program

[0282] First, a user interface is provided that allows users to input questions. This interface runs on devices such as smartphones and provides a screen for users to input their interests and questions.

[0283] The server receives the input question data and stores it in a database. The hardware used here can be a general web server, and the database can be an RDBMS such as PostgreSQL.

[0284] Generative AI models are provided with question data entered by users, and the AI ​​models learn from this data. These AI models are typically implemented using Python libraries (e.g., TensorFlow and PyTorch).

[0285] Next, a user interface is provided that allows other users to enter answers to questions. This is also an interface on a smartphone, and users can enter and submit their answers.

[0286] The received response data is sent to the server and stored in a database. Similarly, this response data is also provided to the generative AI model for further learning.

[0287] Furthermore, a user interface is provided where users can view questions and answers. In this interface, questions and their answers are displayed and users can rate them. The rating data is sent to a server and stored in a database.

[0288] The saved evaluation data is provided to the generative AI model, which retrains the AI ​​model based on the evaluation results. The learning content of the generative AI model is also adjusted based on the feedback data.

[0289] A natural language description of the process

[0290] When a user inputs a question via a device, the device sends the question data to a server. The server stores the received question data in a database and provides the data to a generative AI model. The generative AI model learns from this question data and improves its ability to respond to users' interests and questions.

[0291] When other users enter their answers via their devices, the data is also sent to the server and stored in a database, which is then provided to the generative AI model, allowing it to further learn.

[0292] When a user rates an answer, the rating data is sent to the server and stored in a database. The generative AI model then retrains based on this rating data. Furthermore, by adjusting the AI ​​model's learning content based on feedback data, the accuracy of future answers is improved.

[0293] Below are some examples of prompt sentences.

[0294] Examples of prompt statements

[0295] "A user typed in the question: 'What's the latest anime-related news?'

[0296] User B then enters the following answer:

[0297] It has been announced that "Work X" will be made into a movie.

[0298] Other users found this answer helpful.

[0299] This information is used to grow the generative AI model, enabling it to provide more accurate information to users. In this way, the AI ​​model grows efficiently through user interaction, realizing the provision of more accurate information.

[0300] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0301] Step 1:

[0302] The user inputs a question via the terminal. The user interface has a text input field where the user inputs the question in text format. The input question is converted into JSON format data on the terminal and sent to the server. At this time, the input data includes the user ID, question content, and timestamp.

[0303] Step 2:

[0304] The server stores the received question data in a database. A unique ID is assigned to the question data and stored in the database. At the same time, the server provides this question data to the generative AI model, which then learns from the data. At this time, the server transfers the data to the generative AI model's learning engine, which processes and calculates the data using the question data as input.

[0305] Step 3:

[0306] Other users input answers to questions. The device's user interface provides a text input field for entering answers to questions displayed on the screen. The user enters the answer there, and the answer data is sent to the server in JSON format. The input data includes the user ID, question ID, answer content, and timestamp.

[0307] Step 4:

[0308] The server stores the received answer data in a database. The received data is stored along with a unique ID associated with the answer and associated with the question ID. At the same time, the server provides this answer data to a generative AI model, which then learns from the data.

[0309] Step 5:

[0310] The user browses questions and answers. The device's user interface sends a request to the server to retrieve a list of saved questions and answers. The server retrieves the relevant data from the database and sends it to the device. The device displays the received questions and answers in the user interface. This display includes a mechanism to prioritize new and popular answers.

[0311] Step 6:

[0312] The user enters a rating for the answer. The user interface provides buttons for entering ratings such as "helpful" or "not helpful." When the user presses the rating button, the rating data is sent to the server in JSON format. The rating data includes the user ID, answer ID, rating content, and timestamp.

[0313] Step 7:

[0314] The server stores the received evaluation data in a database. The evaluation data is stored with its associated unique ID and associated with the answer ID. At the same time, the server provides this evaluation data to the generative AI model, which uses the data for retraining. As a result of the retraining, the generative AI model grows to provide more accurate and useful answers.

[0315] Step 8:

[0316] The server provides the questions, answers, and evaluation data in the database to the generative AI model, and periodically retrains it. This allows the generative AI model to constantly learn the latest information. The learning content of the generative AI model is adjusted based on the feedback data. The performance of the AI ​​model improves as a result of data processing and data calculation.

[0317] Step 9:

[0318] The generative AI model generates prompts based on the questions, answers, and evaluation data. The generated prompts are used as further training data for the AI ​​model. The generated prompts are as follows:

[0319] "A user typed in the question: 'What's the latest anime-related news?'

[0320] User B then enters the following answer:

[0321] It has been announced that "Work X" will be made into a movie.

[0322] Other users found this answer helpful.

[0323] Based on this prompt, the generative AI model re-learns and improves.

[0324] 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.

[0325] The present invention combines an emotional engine with an interactive system in which users develop an artificial intelligence model through questions, answers, and ratings. This system recognizes the user's emotional state and can adjust the display order of questions and answers and the generation of feedback based on that state. The specific form and processing flow of this system are described below.

[0326] Enter and save your question

[0327] Enter your question

[0328] Users can input their interests and doubts as questions through the user interface of their terminal.

[0329] Example: User A types, "What's the latest anime-related news?"

[0330] Submit a Question

[0331] The terminal transmits the question data entered by the user to the server.

[0332] This data is sent in a format such as JSON and includes metadata such as the user ID, question, and timestamp.

[0333] Saving Question Data

[0334] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified.

[0335] Providing questions to the AI ​​model

[0336] The server provides the stored question data to the AI ​​model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[0337] Sentiment analysis of question data

[0338] The server provides the question data to the emotion engine to analyze the user's emotions.

[0339] Example: In response to User A's question, "What's the latest anime-related news?", the emotion engine recognizes emotions such as "interesting" and "expecting."

[0340] Enter and save your answers

[0341] Enter your answer

[0342] A user inputs an answer to a question posted by another user through a user interface.

[0343] Example: User B replies, "It has been announced that 'Work X' will be made into a movie."

[0344] Submit your answer

[0345] The device sends the answer data entered by the user to the server. The answer data is also sent in JSON format and includes the user ID, question ID, answer content, timestamp, etc.

[0346] Saving response data

[0347] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[0348] Providing answers to AI models

[0349] The server provides the stored response data to the AI ​​model, which then learns from it, allowing the AI ​​to accumulate more knowledge and improve accuracy.

[0350] Sentiment analysis of response data

[0351] The server provides the response data to an emotion engine to analyze the user's emotions.

[0352] Example: In response to User B's response, "It has been announced that 'Work X' will be made into a movie," the emotion engine recognizes emotions such as "joy" and "satisfaction."

[0353] View questions and answers

[0354] Ask questions and get answers

[0355] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[0356] View questions and answers

[0357] The device displays the received data on the user interface. It can prioritize new or popular answers and adjust the display order based on the user's emotions as recognized by the emotion engine.

[0358] For example, if the emotion engine indicates a high level of interest in response to the question "What is the latest anime-related news?", the question and related answers will be displayed preferentially.

[0359] Enter and save your rating

[0360] Enter your rating

[0361] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[0362] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[0363] Submit your rating

[0364] The terminal sends the entered evaluation data to the server. The evaluation data includes the user ID, answer ID, evaluation content, timestamp, etc.

[0365] Saving evaluation data

[0366] The server stores the received evaluation data in a database, and this evaluation data is also assigned a unique ID.

[0367] Sentiment analysis of rating data

[0368] The server provides the rating data to the emotion engine, which analyzes the user's emotions regarding the ratings.

[0369] Example: Recognizing emotions such as "satisfied" or "dissatisfied" from the content of the evaluation.

[0370] AI learning and accuracy improvement

[0371] Periodic Relearning

[0372] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining, which improves the accuracy of the AI ​​model.

[0373] Reflecting feedback

[0374] The server automatically adjusts the AI ​​model's learning based on user ratings and sentiment data, correcting incorrect answers and reinforcing correct answers.

[0375] Specific examples

[0376] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. The emotion engine also recognizes this question as "interest." Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model, with the emotion engine recognizing it as "joy." If user C views this question and answer and rates it as "helpful," the evaluation data is also saved, and the emotion engine recognizes it as "satisfaction." In this way, the system promotes the growth of AI that also reflects the user's emotions.

[0377] The processing flow will be explained below.

[0378] Post a Question

[0379] Step 1:

[0380] User enters a question

[0381] Users enter their questions using a web browser or mobile app user interface.

[0382] Example: User A types, "What's the latest anime-related news?"

[0383] Step 2:

[0384] Submit your question

[0385] The terminal transmits the question entered by the user to the server.

[0386] The transmitted data includes metadata such as the user ID, question content, and timestamp.

[0387] Example: {"User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}

[0388] Step 3:

[0389] Save questions

[0390] The server stores the received question data in a database.

[0391] Give the question a unique ID (e.g. Q789) and store it in the database.

[0392] Example: Save in the database as {"Question ID": "Q789", "User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}.

[0393] Step 4:

[0394] Providing questions to the AI ​​model

[0395] The server provides the saved question data to the AI ​​model and uses it as learning data.

[0396] Example: Providing {"question": "What's the latest anime news?"} to an AI model.

[0397] Step 5:

[0398] Sentiment analysis of question data

[0399] The server sends the received question data to the emotion engine to analyze the user's emotions.

[0400] Example: The emotion engine recognizes "interest" for the question "What's the latest anime-related news?"

[0401] Providing answers

[0402] Step 6:

[0403] User enters answer

[0404] A user inputs an answer to a question from another user.

[0405] Example: User B types, "It has been announced that 'Work X' will be made into a movie."

[0406] Step 7:

[0407] Sending answers

[0408] The terminal transmits the answer entered by the user to the server.

[0409] The transmitted data includes the user ID, question ID, answer content, timestamp, etc.

[0410] Example: {"User ID": "B456", "Question ID": "Q789", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}

[0411] Step 8:

[0412] Save answers

[0413] The server stores the received response data in a database.

[0414] The answer is given a unique ID (e.g. A987) and stored in association with the question ID.

[0415] Example: Save in the database as {"Answer ID": "A987", "Question ID": "Q789", "User ID": "B456", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}.

[0416] Step 9:

[0417] Providing answers to the AI ​​model

[0418] The server provides the saved response data to the AI ​​model and uses it as learning data.

[0419] Example: Providing {"answer": "It has been announced that 'Book X' will be made into a movie."} to an AI model.

[0420] Step 10:

[0421] Sentiment analysis of response data

[0422] The server sends the received response data to an emotion engine to analyze the user's emotions.

[0423] Example: The emotion engine recognizes "joy" in response to the answer "It has been announced that 'Work X' will be made into a movie."

[0424] View questions and answers

[0425] Step 11:

[0426] Ask questions and get answers

[0427] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[0428] Example: Request {"Question ID": "Q789"} and the server will send the related questions and answers.

[0429] Step 12:

[0430] View questions and answers

[0431] The terminal displays the received question and answer data on a user interface.

[0432] Example: When a user accesses the site, the question "What's the latest anime-related news?" and the answer "It has been announced that 'Work X' will be made into a movie" are displayed on the screen.

[0433] Step 13:

[0434] Adjusting display order based on emotions

[0435] The server adjusts the display order of questions and answers based on the emotion data recognized by the emotion engine.

[0436] For example, questions that you have shown interest in will be displayed before other questions.

[0437] Evaluating answers

[0438] Step 14:

[0439] User-initiated rating input

[0440] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[0441] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[0442] Step 15:

[0443] Submit your evaluation

[0444] The terminal transmits the input evaluation data to the server.

[0445] The transmitted data includes the user ID, answer ID, evaluation details, timestamp, etc.

[0446] Example: {"User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}

[0447] Step 16:

[0448] Save evaluation details

[0449] The server stores the received evaluation data in a database.

[0450] Evaluation data will also be assigned a unique ID (e.g., E123).

[0451] Example: Store in database as {"Rating ID": "E123", "User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}.

[0452] Step 17:

[0453] Sentiment analysis of rating data

[0454] The server transmits the received evaluation data to an emotion engine to analyze the user's emotions.

[0455] For example, the sentiment engine recognizes the rating "helpful" as "satisfied."

[0456] AI learning and accuracy improvement

[0457] Step 18:

[0458] Periodic retraining of AI models

[0459] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[0460] Example: Retraining an AI model using all question and answer and rating data from the past month.

[0461] Step 19:

[0462] Reflecting feedback

[0463] The server automatically adjusts the learning content of the AI ​​model based on user evaluation data and emotion data from the emotion engine.

[0464] Example: Learning from an answer that was rated "not helpful" by many users can be corrected to provide more accurate answers.

[0465] Through the above processing steps, the present invention realizes an interactive system that allows users to actively participate in the growth of AI, and provides advanced information by combining an emotion engine.

[0466] Example 2

[0467] 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."

[0468] Modern information systems are required to not only provide information based on user input, but also to understand the user's emotions and intentions and provide feedback accordingly. However, conventional systems have had difficulty in realizing advanced interactions such as analyzing the user's emotions and adjusting the order in which information is displayed based on those emotions.

[0469] 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.

[0470] In this invention, the server includes means for providing a user interface through which a user can input a question, means for transmitting input question data to the server, means for saving received question data in a database, means for providing the question data saved in the database to the artificial intelligence model, means for providing the question data to an emotion engine and analyzing the user's emotions, means for providing a user interface through which other users can input answers to questions, means for transmitting input answer data to the server, means for saving received answer data in a database, means for providing the answer data saved in the database to the artificial intelligence model, means for providing the answer data to the emotion engine and analyzing the user's emotions, means for providing a user interface through which users can view questions and answers, means for adjusting the display order of questions and answers based on the results of the user emotion analysis, means for providing a user interface through which users can input ratings for answers, means for transmitting input rating data to the server, means for saving received rating data in a database, means for providing the rating data to the emotion engine and analyzing the user's emotions, means for providing the questions, answers, and rating data in the database to the artificial intelligence model and performing relearning, and means for adjusting the learning content of the artificial intelligence model based on the feedback data. This makes it possible to provide information that reflects the user's emotions and to create an interactive system that adjusts the display based on that information.

[0471] "User interface" refers to the screens and operating means that allow users to interact with the system to exchange information.

[0472] "Question data" is information including the question entered by the user, and includes metadata such as the user ID, question content, and timestamp.

[0473] A "server" refers to a computer system that receives requests over a network and is responsible for processing and storing data.

[0474] A "database" refers to a data management system that allows a system to centrally store data and make it searchable and retrievalable.

[0475] An "artificial intelligence model" refers to an algorithm that learns patterns and rules from large amounts of data and makes inferences and predictions based on input.

[0476] An "emotion engine" is an algorithm or system that analyzes user input data and identifies the emotion or intent behind it.

[0477] "Answer data" is information including the answer entered by the user to the question, and includes the user ID, question ID, answer, timestamp, and the like.

[0478] "Evaluation data" is information including the content of the user's evaluation of an answer, and includes the user ID, answer ID, evaluation content, timestamp, and the like.

[0479] "Retraining" refers to the learning process of using new data in a database to improve the accuracy of an artificial intelligence model.

[0480] "Feedback data" refers to information used to adjust system processing based on user evaluations and the results of sentiment analysis.

[0481] The present invention is an interactive system that combines an emotion engine and aims to develop an artificial intelligence model through user questions, answers, and ratings. This system recognizes the user's emotional state and can adjust the display order of questions and answers and the generation of feedback based on that state. The specific form and processing flow of the system are described below.

[0482] Enter and save your question

[0483] Users input their interests and concerns as questions through the user interface of their device. This user interface provides a text input area, allowing users to enter the content of their questions as text. For example, User A enters "What's the latest anime-related news?" This question data is sent from the device to the server in a format such as JSON, and includes metadata such as the user ID, question content, and timestamp.

[0484] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified. The server then provides the stored question data to an artificial intelligence model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[0485] The question data is also provided to the emotion engine, which analyzes the user's emotions. For example, in response to User A's question, "What's the latest anime-related news?", the emotion engine recognizes emotions such as "interesting" and "expecting." The results are saved as metadata.

[0486] Enter and save your answers

[0487] Users input answers to questions posted by other users through the user interface. For example, User B answers, "It has been announced that 'Work X' will be made into a movie." This answer data is sent from the device to the server in JSON format or similar, and includes the user ID, question ID, answer content, and timestamp.

[0488] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID. The server then provides the stored answer data to an artificial intelligence model, which then learns from the data. This allows the AI ​​to accumulate more knowledge and improve accuracy.

[0489] The response data is also provided to the emotion engine, which analyzes the user's emotions. For example, in response to User B's response, "It has been announced that 'Work X' will be made into a movie," the emotion engine recognizes emotions such as "joy" and "satisfaction." These results are also saved as metadata.

[0490] View questions and answers

[0491] The device requests a list of questions and their answers from the server, and receives the data from the server. This request includes the user ID and the specific question ID, and the server searches for the corresponding data and returns it to the device.

[0492] The device displays the received data on the user interface. It is possible to display newer or more popular answers with priority. The display order is also adjusted based on the user's emotions as recognized by the emotion engine. For example, if the emotion engine indicates a high level of "interest" in response to the question "What's the latest anime-related news?", that question and related answers will be displayed with priority.

[0493] Enter and save your rating

[0494] The user inputs a rating such as "helpful" or "not helpful" for the displayed answer. For example, a user may rate the answer "It has been announced that 'Work X' will be made into a movie." as "helpful." This rating data is sent from the device to the server and includes the user ID, answer ID, rating content, and timestamp.

[0495] The server stores the received evaluation data in a database. This evaluation data is also assigned a unique ID. The evaluation data is also provided to the emotion engine, which analyzes the user's emotions. For example, emotions such as "satisfied" or "dissatisfied" are recognized from the content of the evaluation. This result is also stored as metadata.

[0496] AI learning and accuracy improvement

[0497] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for re-learning. For example, once a week, the server collects all new data and re-trains the AI ​​model. This improves the accuracy of the AI ​​model.

[0498] Furthermore, the server automatically adjusts the learning content of the AI ​​model based on user evaluation and emotion data. If there are many incorrect answers, the data is corrected and correct answers are strengthened, improving the model's performance.

[0499] Specific examples

[0500] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. The emotion engine also recognizes this question as "interest." Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model, with the emotion engine recognizing it as "joy." If user C views this question and answer and rates it as "helpful," the evaluation data is also saved, and the emotion engine recognizes it as "satisfaction." In this way, the system promotes the growth of AI that also reflects user emotions.

[0501] Prompt Sentence Examples

[0502] "What's the latest anime-related news?"

[0503] "It has been announced that 'Work X' will be made into a movie."

[0504] "It was helpful"

[0505] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0506] Step 1:

[0507] Enter your question

[0508] The user enters a question through the user interface of their terminal.

[0509] Specific actions: Enter a question in the text input area and click the submit button.

[0510] Input: The question the user types into the text entry area.

[0511] Output: JSON formatted question data sent by the device.

[0512] Step 2:

[0513] Submit a Question

[0514] The terminal transmits the question data entered by the user to the server.

[0515] Specific operation: The device sends question data including metadata such as the user ID, question content, and timestamp to the server.

[0516] Input: The question data entered by the user.

[0517] Output: The question data sent to the server.

[0518] Step 3:

[0519] Saving Question Data

[0520] The server stores the received question data in a database, and each question is assigned a unique ID.

[0521] Specific operation: The server assigns a unique ID to the question data and saves it in the database.

[0522] Input: Question data sent from the terminal.

[0523] Output: Question data stored in a database.

[0524] Step 4:

[0525] Providing questions to the AI ​​model

[0526] The server provides the stored question data to the AI ​​model, which then learns from the data.

[0527] Specific operation: The server reads question data from the database and sends it to the AI ​​model for learning.

[0528] Input: Question data stored in the database.

[0529] Output: The input data for the AI ​​model to learn from.

[0530] Step 5:

[0531] Sentiment analysis of question data

[0532] The server provides the question data to the emotion engine to analyze the user's emotions.

[0533] Specific operation: The server sends the question data to the emotion engine, which analyzes the emotion.

[0534] Input: Question data stored in the database.

[0535] Output: Sentiment data parsed by the sentiment engine.

[0536] Step 6:

[0537] Enter your answer

[0538] A user inputs an answer to a question posted by another user through a user interface.

[0539] Specific actions: Enter your answer in the text input area and click the submit button.

[0540] Input: The answer the user entered into the text entry area.

[0541] Output: JSON formatted response data sent by the device.

[0542] Step 7:

[0543] Submit your answer

[0544] The terminal transmits the answer data entered by the user to the server.

[0545] Specific operation: The device sends the answer data, including metadata such as the user ID, question ID, answer content, and timestamp, to the server.

[0546] Input: The answer data entered by the user.

[0547] Output: The response data sent to the server.

[0548] Step 8:

[0549] Saving response data

[0550] The server stores the received response data in a database, and each response is assigned a unique ID.

[0551] Specific operation: The server assigns a unique ID to the response data and saves it in the database.

[0552] Input: Response data sent from the device.

[0553] Output: Response data stored in a database.

[0554] Step 9:

[0555] Providing answers to AI models

[0556] The server provides the saved response data to the AI ​​model, which then learns from the data.

[0557] Specific operation: The server reads the response data from the database and sends it to the AI ​​model for learning.

[0558] Input: Response data stored in the database.

[0559] Output: The input data for the AI ​​model to learn from.

[0560] Step 10:

[0561] Sentiment analysis of response data

[0562] The server provides the response data to an emotion engine to analyze the user's emotions.

[0563] Specific operation: The server sends the response data to the emotion engine, which analyzes the emotion.

[0564] Input: Response data stored in the database.

[0565] Output: Sentiment data parsed by the sentiment engine.

[0566] Step 11:

[0567] Ask questions and get answers

[0568] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[0569] Specific operation: The device sends a request to the server, and the server returns the relevant question and answer data.

[0570] Input: The request from the device (including the user ID and the specific question ID).

[0571] Output: Question and answer data received from the server.

[0572] Step 12:

[0573] View questions and answers

[0574] The terminal displays the received data on a user interface.

[0575] Specific operation: The device converts the data into a display format and displays it on the user interface. The display order is adjusted based on the user's emotions recognized by the emotion engine.

[0576] Input: Question and answer data received from the server.

[0577] Output: The question and answer data that is displayed in the user interface.

[0578] Step 13:

[0579] Enter your rating

[0580] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[0581] Specific actions: Click the rating button, select the rating content, and click the submit button.

[0582] Input: The rating selected by the user.

[0583] Output: JSON formatted rating data sent by the device.

[0584] Step 14:

[0585] Submit your rating

[0586] The terminal transmits the input evaluation data to the server.

[0587] Specific operation: The device sends evaluation data including metadata such as user ID, answer ID, evaluation content, and timestamp to the server.

[0588] Input: The rating data entered by the user.

[0589] Output: The rating data sent to the server.

[0590] Step 15:

[0591] Saving evaluation data

[0592] The server stores the received evaluation data in a database. The evaluation data is also assigned a unique ID.

[0593] Specific operation: The server assigns a unique ID to the evaluation data and saves it in the database.

[0594] Input: Rating data sent from the device.

[0595] Output: Evaluation data stored in a database.

[0596] Step 16:

[0597] Sentiment analysis of rating data

[0598] The server provides the rating data to the emotion engine to analyze the user's emotions.

[0599] Specific operation: The server sends the evaluation data to the emotion engine, which analyzes the emotions.

[0600] Input: Evaluation data stored in a database.

[0601] Output: Sentiment data parsed by the sentiment engine.

[0602] Step 17:

[0603] Retraining the AI

[0604] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[0605] What it does: The server reads all new data from the database and periodically retrains the AI ​​model.

[0606] Input: Question, answer, and assessment data stored in a database.

[0607] Output: The retrained AI model.

[0608] Step 18:

[0609] Reflecting feedback

[0610] The server automatically adjusts the learning content of the AI ​​model based on user evaluations and emotional data.

[0611] Specific operation: The server adjusts the learning weights of the AI ​​model based on the evaluation and emotion data and corrects incorrect answers.

[0612] Input: Rating data, emotion data.

[0613] Output: The tuned AI model.

[0614] (Application example 2)

[0615] 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."

[0616] Conventional customer support systems have difficulty in responding to users' emotional states, making it difficult to improve user experience and service quality. In addition, there has been a lack of methods to provide artificial intelligence models that can learn interactively by reflecting user feedback and emotions.

[0617] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for providing a user interface through which a user can input a question; means for transmitting the input question data to the server; means for saving the received question data in a database; means for providing the question data saved in the database to the AI ​​model; means for providing a user interface through which other users can input answers to questions; means for transmitting the input answer data to the server; means for saving the received answer data in the database; means for providing the answer data saved in the database to the AI ​​model; means for providing a user interface through which users can view questions and answers; means for providing a user interface through which users can input ratings for answers; means for transmitting the input rating data to the server; means for saving the received rating data in the database; means for providing the questions, answers, and rating data in the database to the AI ​​model and performing re-learning; means for adjusting the learning content of the AI ​​model based on feedback data; means for analyzing the emotional state of the input question and answer and means for adjusting the display order and feedback based on the emotional state. This makes it possible to realize an interactive customer support system that reflects the user's emotions.

[0618] A "user interface" is an interface through which a user interacts directly with a system and inputs questions and answers.

[0619] A "server" is a computer that receives question data and answer data sent by users and stores them in a database.

[0620] A "database" is a storage system that centrally manages and stores question data, response data, and evaluation data.

[0621] An "artificial intelligence model" is an algorithm that learns from questions and answers entered by users and improves the quality of subsequent questions and answers.

[0622] An "emotional state" is a psychological state recognized as a result of the emotion engine's analysis of the questions and answers entered by the user.

[0623] "Display order" refers to the order in which questions and answers are displayed when a user browses.

[0624] "Feedback" refers to the evaluations and comments that users make on answers, and is data used to train the system.

[0625] "Relearning" is a process in which an artificial intelligence model periodically learns based on data accumulated in a database to improve its accuracy.

[0626] "Question data" is data including the content of questions posed to the system by users.

[0627] "Answer data" is data including answers provided by other users to questions.

[0628] "Evaluation data" is data including the evaluation content given by the user to the answer.

[0629] The "emotion engine" is a system component that analyzes the emotional state of input questions and answers.

[0630] The present invention combines an emotion engine with an interactive system in which users can develop an artificial intelligence model through questions, answers, and ratings. Specific embodiments of this system are described below.

[0631] Enter and save your question

[0632] User Interface:

[0633] It provides a user interface for users to input questions. Users can input their interests and questions and send the question data to the server via their terminal. This data includes metadata such as the user ID, question content, and timestamp.

[0634] server:

[0635] The server stores the received question data in a database and assigns a unique ID to uniquely identify it.The server then provides the stored question data to an artificial intelligence model for learning.Furthermore, the server passes the question data to an emotion engine to analyze the user's emotional state.

[0636] Enter and save your answers

[0637] User Interface:

[0638] It provides a user interface that allows other users to enter answers to questions. The entered answer data includes information such as the answer content, user ID, question ID, and timestamp.

[0639] server:

[0640] The server receives the response data and stores it in a database. Here, a unique ID is assigned to each response, making it uniquely identified. The stored response data is then provided to an artificial intelligence model, which further trains the AI. The response data is also passed to an emotion engine, which analyzes the user's emotional state.

[0641] View questions and answers

[0642] User Interface:

[0643] It provides an interface where users can browse questions and their answers. The interface is adjusted to prioritize new and popular answers. The display order is also dynamically adjusted based on the user's emotions as recognized by the emotion engine.

[0644] Enter and save your rating

[0645] User Interface:

[0646] Users can rate the displayed answers. The rating data includes the user ID, answer ID, rating content, timestamp, etc.

[0647] server:

[0648] The evaluation data is received and stored in a database. The evaluation data is analyzed by an emotion engine to recognize the user's emotional state. The acquired evaluation data is used to retrain the AI ​​model.

[0649] AI model training and accuracy improvement

[0650] server:

[0651] The server provides the AI ​​model with the questions, answers, and evaluation data from the database and periodically retrains it, improving its accuracy. It also automatically adjusts its learning based on user feedback and sentiment data.

[0652] Hardware and software used

[0653] Hardware: Smartphones, head-mounted displays, servers

[0654] Software: Emotion engine library, AI model library, database connection library

[0655] Specific example explanation

[0656] For example, User A opens the app to ask a question about a product in a store and types the question, "Does this product run on batteries?" This question is sent to the server and stored in the database. The AI ​​model learns this question, and the emotion engine recognizes it as "curiosity." As a result, the store staff is notified, "We have received your question. Please check the details." An example of this prompt is as follows:

[0657] Prompt Sentence Examples

[0658] User Question:

[0659] Question: 'Does this item run on batteries?'

[0660] User ID: 'user123'

[0661] Timestamp: '2023-10-10T10:00:00Z'

[0662] Emotion Engine Analysis Results:

[0663] Emotional state: 'Curiosity'

[0664] Action: 'Increase display priority'

[0665] In this way, the present invention provides an interactive customer support system that reflects the user's feelings.

[0666] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0667] Step 1:

[0668] The user inputs a question using a smartphone or head-mounted display. The question data input from the user interface includes the user ID, question content, and timestamp. For example, the user might input, "Does this product run on batteries?"

[0669] Step 2:

[0670] The device sends the entered question data in JSON format to the server, which includes the user ID, question content, and timestamp.

[0671] Step 3:

[0672] The server stores the received question data in a database. A unique ID is assigned when the data is saved to uniquely identify it. The input data includes the user ID, question content, timestamp, and unique ID.

[0673] Step 4:

[0674] The server provides the stored question data to the AI ​​model for learning. Based on the provided data, the AI ​​model analyzes the question content and improves the accuracy of answers to future questions.

[0675] Step 5:

[0676] The server passes the question data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the question and outputs an emotion such as "curiosity."

[0677] Step 6:

[0678] Another user inputs an answer to the question using a smartphone or head-mounted display. The answer data input from the user interface includes the answer content, user ID, question ID, and timestamp. For example, the user might input, "Yes, this product runs on batteries."

[0679] Step 7:

[0680] The device sends the entered answer data in JSON format to the server. The sent data includes the user ID, question ID, answer content, and timestamp.

[0681] Step 8:

[0682] The server stores the received answer data in a database. A unique ID is assigned when saving the data to uniquely identify it. The input data includes the user ID, question ID, answer content, timestamp, and unique ID.

[0683] Step 9:

[0684] The server provides the stored response data to the AI ​​model for learning. Based on the provided data, the AI ​​model analyzes the content of the responses and improves the accuracy of future responses.

[0685] Step 10:

[0686] The server passes the answer data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the answer and outputs an emotion such as "a sense of relief."

[0687] Step 11:

[0688] The user interface displays questions and answers to the user. The display data includes the question content and multiple answers. The display order is dynamically adjusted based on the analysis results of the emotion engine so that newer and more popular answers are displayed preferentially.

[0689] Step 12:

[0690] Users rate the displayed answers. The rating data includes the user ID, answer ID, rating content, and timestamp. For example, a user may rate the answer as "helpful."

[0691] Step 13:

[0692] The device sends the entered evaluation data in JSON format to the server. The sent data includes the user ID, answer ID, evaluation content, and timestamp.

[0693] Step 14:

[0694] The server stores the received evaluation data in a database. A unique ID is assigned when the data is saved to uniquely identify it. The input data includes the user ID, answer ID, evaluation content, timestamp, and unique ID.

[0695] Step 15:

[0696] The server passes the evaluation data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the evaluation content and outputs an emotion such as "satisfaction."

[0697] Step 16:

[0698] The server provides the AI ​​model with the questions, answers, and evaluation data stored in the database and periodically retrains it, improving its accuracy. Furthermore, by automatically adjusting the learning content based on emotion data, the AI ​​model can provide more appropriate responses to users.

[0699] 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.

[0700] 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.

[0701] 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.

[0702] [Second embodiment]

[0703] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0704] 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.

[0705] 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).

[0706] 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.

[0707] 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.

[0708] 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).

[0709] 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.

[0710] 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.

[0711] 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.

[0712] 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.

[0713] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0714] 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."

[0715] This invention is an interactive system that allows users to develop an artificial intelligence model through questions, answers, and ratings. Its details include a process for effectively collecting and storing information entered by users and providing it to the AI ​​model for learning. The specific form and processing flow of the system are described below.

[0716] Enter and save your question

[0717] Enter your question

[0718] Users can input their interests and doubts as questions through the user interface of their terminal.

[0719] Example: User A types, "What's the latest anime-related news?"

[0720] Submit a Question

[0721] The terminal transmits the question data entered by the user to the server.

[0722] This data is sent in a format such as JSON and includes metadata such as the user ID, question, and timestamp.

[0723] Saving Question Data

[0724] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified.

[0725] Providing questions to the AI ​​model

[0726] The server provides the stored question data to the AI ​​model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[0727] Enter and save your answers

[0728] Enter your answer

[0729] A user inputs an answer to a question posted by another user through a user interface.

[0730] Example: User B replies, "It has been announced that 'Work X' will be made into a movie."

[0731] Submit your answer

[0732] The device sends the answer data entered by the user to the server. The answer data is also sent in JSON format and includes the user ID, question ID, answer content, timestamp, etc.

[0733] Saving response data

[0734] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[0735] Providing answers to AI models

[0736] The server provides the stored response data to the AI ​​model, which then learns from it, allowing the AI ​​to accumulate more knowledge and improve accuracy.

[0737] View questions and answers

[0738] Ask questions and get answers

[0739] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[0740] View questions and answers

[0741] The device displays the received data on the user interface. It is also possible to display newer or more popular answers preferentially.

[0742] Enter and save your rating

[0743] Enter your rating

[0744] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[0745] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[0746] Submit your rating

[0747] The terminal sends the entered evaluation data to the server. The evaluation data includes the user ID, answer ID, evaluation content, timestamp, etc.

[0748] Saving evaluation data

[0749] The server stores the received evaluation data in a database, and this evaluation data is also assigned a unique ID.

[0750] AI learning and accuracy improvement

[0751] Periodic Relearning

[0752] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for re-training, thereby improving the accuracy of the AI ​​model.

[0753] Reflecting feedback

[0754] The server automatically adjusts the AI ​​model's learning based on user evaluation data, thereby correcting incorrect answers and strengthening correct answers.

[0755] Specific examples

[0756] For example, if user A posts a question such as "What's the latest anime-related news?", the server saves this question in a database and provides it to the AI ​​model for learning. Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also saved in the database and provided to the AI ​​model. If user C views this question and answer and rates it as "helpful," this rating data is also saved and used to retrain the AI ​​model.

[0757] In this way, users can actively participate in the growth of the AI, and the system provides highly accurate information that reflects their own knowledge.

[0758] The processing flow will be explained below.

[0759] Post a Question

[0760] Step 1:

[0761] User enters a question

[0762] Users enter their questions using a web browser or mobile app user interface.

[0763] Example: User A types, "What's the latest anime-related news?"

[0764] Step 2:

[0765] Submit your question

[0766] The terminal transmits the question entered by the user to the server.

[0767] The transmitted data includes metadata such as the user ID, question content, and timestamp.

[0768] Example: {"User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}

[0769] Step 3:

[0770] Save questions

[0771] The server stores the received question data in a database.

[0772] Give the question a unique ID (e.g. Q789) and store it in the database.

[0773] Example: Save in the database as {"Question ID": "Q789", "User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}.

[0774] Step 4:

[0775] Providing questions to the AI ​​model

[0776] The server provides the saved question data to the AI ​​model and uses it as learning data.

[0777] Example: Providing {"question": "What's the latest anime news?"} to an AI model.

[0778] Providing answers

[0779] Step 5:

[0780] User enters answer

[0781] A user inputs an answer to a question from another user.

[0782] Example: User B types, "It has been announced that 'Work X' will be made into a movie."

[0783] Step 6:

[0784] Sending answers

[0785] The terminal transmits the answer entered by the user to the server.

[0786] The transmitted data includes the user ID, question ID, answer content, timestamp, etc.

[0787] Example: {"User ID": "B456", "Question ID": "Q789", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}

[0788] Step 7:

[0789] Save answers

[0790] The server stores the received response data in a database.

[0791] Give the answer a unique ID (e.g. A987) and store it along with the question ID.

[0792] Example: Save in the database as {"Answer ID": "A987", "Question ID": "Q789", "User ID": "B456", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}.

[0793] Step 8:

[0794] Providing answers to the AI ​​model

[0795] The server provides the saved response data to the AI ​​model and uses it as learning data.

[0796] Example: Providing {"answer": "It has been announced that 'Book X' will be made into a movie."} to an AI model.

[0797] View questions and answers

[0798] Step 9:

[0799] Ask questions and get answers

[0800] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[0801] Example: Request data for {"Question ID": "Q789"} and receive question and answer data from the server.

[0802] Step 10:

[0803] View questions and answers

[0804] The terminal displays the received question and answer data on a user interface.

[0805] Example: When a user accesses the site, the question "What's the latest anime-related news?" and the answer "It has been announced that 'Work X' will be made into a movie" are displayed on the screen.

[0806] Evaluating answers

[0807] Step 11:

[0808] User-initiated rating input

[0809] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[0810] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[0811] Step 12:

[0812] Submit your evaluation

[0813] The terminal transmits the input evaluation data to the server.

[0814] The transmitted data includes the user ID, answer ID, evaluation details, timestamp, etc.

[0815] Example: {"User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}

[0816] Step 13:

[0817] Save evaluation details

[0818] The server stores the received evaluation data in a database.

[0819] Evaluation data will also be assigned a unique ID (e.g., E123).

[0820] Example: Store in database as {"Rating ID": "E123", "User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}.

[0821] AI learning and accuracy improvement

[0822] Step 14:

[0823] Periodic retraining of AI models

[0824] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[0825] Example: Retraining an AI model using all question and answer and rating data from the past month.

[0826] Step 15:

[0827] Reflecting feedback

[0828] The server automatically adjusts the learning content of the AI ​​model based on user evaluation data.

[0829] Example: Revisiting learning based on answers that were rated "not helpful" by many users.

[0830] Example 1

[0831] 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."

[0832] Conventional AI model learning systems have faced issues with inefficient management of user-entered questions, answers, and evaluation data, resulting in delayed improvements in the accuracy of AI models. Furthermore, there was a lack of a mechanism for immediately reflecting user feedback and using it to improve the model, making it difficult to improve the user experience. Furthermore, when displaying questions and answers, there was a lack of a function to prioritize the latest information and popular answers that users wanted, reducing the efficiency of information acquisition.

[0833] 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.

[0834] In this invention, the server includes a means for providing a user interface, a means for transmitting question data, answer data, and evaluation data to the server, a means for storing the received data in a database, a means for providing the data to the AI ​​model as learning data, a means for acquiring the question and answer data and displaying them on the user interface, and a means for preferentially displaying new or popular answers. This enables efficient collection, management, and provision of various data entered by users, resulting in rapid improvement in the accuracy of the AI ​​model and improved functionality of the user interface.

[0835] "User interface" refers to the screen or operating means through which a user can access a system and input and view information such as questions and answers.

[0836] "Question data" refers to data that includes information about the question entered by the user, including metadata such as the user ID, question content, and timestamp.

[0837] "Server" refers to a central computing device that receives, processes, stores, and provides data sent by users to AI models.

[0838] "Database" refers to a system in which a server systematically stores question data, answer data, evaluation data, etc. and manages them so that they can be retrieved as needed.

[0839] "Artificial intelligence model" refers to a machine learning algorithm or neural network that learns from user-provided data and answers questions, provides information, or generates responses.

[0840] "Answer data" refers to data that includes answers to questions posted by other users, and includes metadata such as user ID, question ID, answer content, and timestamp.

[0841] "Rating data" refers to data that includes information about the ratings that users have given to questions and answers, and includes metadata such as user ID, rating content, answer ID, and timestamp.

[0842] "Retraining" refers to the learning process in which questions, answers, and evaluation data stored in a database are periodically provided to an artificial intelligence model to improve the model's accuracy.

[0843] "Feedback data" refers to data used to adjust the model's learning content based on user evaluation data and improve the model's output results.

[0844] MODE FOR CARRYING OUT THE INVENTION

[0845] This invention is an interactive system in which users can develop an artificial intelligence model through questions, answers, and evaluations. The specific form and processing flow of this system will be described in detail below.

[0846] Enter and save your question

[0847] When a user types a question

[0848] A user accesses a dedicated user interface using their own terminal (for example, a personal computer or a smartphone) and inputs a question. An example of a question is "What is the latest anime-related news?"

[0849] Sending and storing question data

[0850] The device sends the question entered by the user to the server in JSON format, including the user ID, question content, timestamp, etc.

[0851] The server stores the received question data in a database (e.g., MySQL or PostgreSQL). A unique ID is assigned to each question, allowing it to be uniquely identified.

[0852] Providing question data to the AI ​​model

[0853] The server can provide the stored question data to the AI ​​model, which can then use the data as training data. This process allows the AI ​​model to learn new information from user questions and improve the accuracy of its responses to future questions.

[0854] Enter and save your answers

[0855] When the user enters an answer

[0856] Users can enter answers to questions posted by other users. An example of an answer might be, "It has been announced that 'Work X' will be made into a movie."

[0857] Sending and storing response data

[0858] The device sends the answers entered by the user to the server in JSON format, including the user ID, question ID, answer content, and timestamp.

[0859] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[0860] Providing response data to the AI ​​model

[0861] The server can provide the saved response data to the AI ​​model, which can then use the data as training data. This process allows the AI ​​model to learn new information from the user's responses and further improve the accuracy of future responses.

[0862] View questions and answers

[0863] Retrieving and Displaying Data

[0864] The device requests a list of questions and their answers from the server and receives the necessary data. The received data is displayed on the user interface. It is also possible to display new or popular answers preferentially, allowing users to quickly obtain the latest information and useful information.

[0865] Enter and save your rating

[0866] When a user enters a rating

[0867] Users can enter a rating for the displayed answer, such as "helpful" or "not helpful."

[0868] Sending and storing evaluation data

[0869] The device sends the entered evaluation data in JSON format to the server, which includes the user ID, answer ID, evaluation content, and timestamp.

[0870] The server stores the received rating data in a database, and each rating is assigned a unique ID.

[0871] AI learning and accuracy improvement

[0872] Periodic Relearning

[0873] The server periodically provides the question data, answer data, and evaluation data in the database to the AI ​​model for re-learning. This process improves the accuracy of the AI ​​model, enabling it to provide more accurate information to users.

[0874] Reflecting feedback

[0875] The server automatically adjusts the learning content of the AI ​​model based on the user's evaluation data, thereby correcting incorrect answers and strengthening correct answers.

[0876] Prompt Sentence Examples

[0877] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model. If user C views this question and answer and rates it as "helpful," this rating data is also saved and used to retrain the AI ​​model. In this way, the system allows users to actively participate in the growth and improvement of the AI's accuracy.

[0878] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0879] Step 1: User enters question

[0880] A user uses a terminal to access the user interface and inputs a question by typing the question in a text box and clicking a submit button.

[0881] Input: User ID, question, timestamp

[0882] Output: Question data (JSON format)

[0883] Step 2: The device sends the query data

[0884] The device converts the question data entered by the user into JSON format and sends it to the server. This data includes the user ID, question content, and timestamp.

[0885] Input: Question data (JSON format)

[0886] Output: Data packet to be sent

[0887] Step 3: The server receives the query data

[0888] The server receives the submitted question data and stores it in a database. As an additional process, it assigns a unique ID to each question.

[0889] Input: Data packet to send

[0890] Output: Question data stored in the database (with unique ID)

[0891] Step 4: The server provides the question data to the AI ​​model

[0892] The server extracts question data stored in the database and provides it to the AI ​​model for training. The AI ​​model uses this information to add new knowledge and improve the accuracy of future question answers.

[0893] Input: Question data stored in the database

[0894] Output: Provided as training data to the AI ​​model

[0895] Step 5: User Enters Answer

[0896] A user can enter an answer to a question posted by another user by selecting the question in the user interface, entering the answer in the text box that appears, and clicking the submit button.

[0897] Input: User ID, Question ID, Answer, Time Stamp

[0898] Output: Response data (JSON format)

[0899] Step 6: The device sends the response data

[0900] The device converts the answer data entered by the user into JSON format and sends it to the server. This data includes the user ID, question ID, answer content, and timestamp.

[0901] Input: Response data (JSON format)

[0902] Output: Data packet to be sent

[0903] Step 7: The server receives the response data

[0904] The server receives the submitted answer data and stores it in a database. As an additional process, it assigns a unique ID to each answer and associates it with the question ID.

[0905] Input: Data packet to send

[0906] Output: Answer data saved in the database (with unique ID)

[0907] Step 8: The server provides the answer data to the AI ​​model

[0908] The server extracts the response data stored in the database and provides it to an artificial intelligence model for learning. The AI ​​model uses this information to add new knowledge and improve the accuracy of future responses.

[0909] Input: Response data stored in the database

[0910] Output: Provided as training data to the AI ​​model

[0911] Step 9: The device requests data

[0912] The terminal requests question and answer data from the server in response to a user request, which is sent by clicking the "Show Questions and Answers" button on the user interface.

[0913] Input: Request data

[0914] Output: None (request sent)

[0915] Step 10: Server sends data

[0916] The server extracts the requested question and answer data from the database and sends it to the device in JSON format.

[0917] Input: Request data

[0918] Output: Question and answer data in JSON format

[0919] Step 11: The terminal displays the data

[0920] The terminal displays the question and answer data received from the server on a user interface, and can also display new or popular answers with priority.

[0921] Input: Question and answer data in JSON format

[0922] Output: Questions and answers displayed in the user interface

[0923] Step 12: User Enters Rating

[0924] The user inputs a rating for the displayed answer, such as "helpful" or "not helpful," and selects one and clicks the submit button.

[0925] Input: User ID, Answer ID, Evaluation details, Time stamp

[0926] Output: Evaluation data (JSON format)

[0927] Step 13: The device sends the evaluation data

[0928] The device converts the evaluation data entered by the user into JSON format and sends it to the server. This data includes the user ID, answer ID, evaluation content, and timestamp.

[0929] Input: Evaluation data (JSON format)

[0930] Output: Data packet to be sent

[0931] Step 14: Server receives reputation data

[0932] The server receives the evaluation data and stores it in a database. Each evaluation is assigned a unique ID.

[0933] Input: Data packet to send

[0934] Output: Evaluation data stored in the database (with unique ID)

[0935] Step 15: Server periodically retrains

[0936] The server periodically provides the question data, answer data, and evaluation data in the database to the AI ​​model for re-learning, a process that improves the accuracy of the AI ​​model.

[0937] Input: Question data, response data, and evaluation data stored in the database

[0938] Output: Retrained AI model

[0939] Step 16: The server processes the feedback

[0940] The server automatically adjusts the AI ​​model's learning based on user evaluation data, improving the accuracy of answers and correcting incorrect answers.

[0941] Input: Evaluation data

[0942] Output: The tuned AI model

[0943] (Application example 1)

[0944] 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."

[0945] In recent years, as demand for interactive content distribution services has increased, AI models are being developed based on information provided by users, and there is a demand for more accurate information provision. However, current systems lack efficient collection and learning of user questions, answers, and evaluation data, resulting in slow AI model development. Furthermore, the user interface is difficult to use, making it difficult to improve user satisfaction.

[0946] 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.

[0947] In this invention, the server includes: means for providing a user interface through which a user can input a question; means for transmitting the input question data to the server; means for storing the received question data in a database; means for providing the question data stored in the database to the generative AI model; means for providing a user interface through which other users can input answers to questions; means for transmitting the input answer data to the server; means for storing the received answer data in the database; means for providing the answer data stored in the database to the generative AI model; means for providing a user interface through which users can view questions and answers; means for providing a user interface through which users can input ratings for answers; means for transmitting the input rating data to the server; means for storing the received rating data in the database; means for providing the questions, answers, and rating data in the database to the generative AI model and performing re-training; means for adjusting the learning content of the generative AI model based on feedback data; means for causing the generative AI model to generate prompt sentences based on the questions, answers, and rating data; and means for providing the generated prompt sentences to the generative AI model as further training data. This enables efficient growth of the AI ​​model and provision of highly accurate information through user interaction.

[0948] A "user interface" is an interface that provides a screen and input means for a user to interact with an application or system.

[0949] "Question data" refers to data including the content of a question that is input by a user through a user interface and sent to a server.

[0950] "Answer data" refers to data including the answer content that is input by other users to a question and sent to the server.

[0951] "Evaluation data" refers to data including evaluations such as "useful" or "not useful" that are entered by users in response to the answers and sent to the server.

[0952] A "generative AI model" is an artificial intelligence model that continuously learns based on questions, answers, and evaluation data provided by users, and provides highly accurate information.

[0953] "Server" refers to a computer system that receives question data, answer data, and evaluation data from users and stores and manages them in a database.

[0954] A "database" is a storage device for accumulating and managing question data, answer data, and evaluation data stored on a server.

[0955] "Feedback data" is data used to adjust the learning content of the generative AI model based on evaluations made by users.

[0956] A "prompt sentence" is training data generated by a generative AI model based on questions, answers, and evaluation data, and is a sentence used for further AI training.

[0957] "Interaction data" is a general term for data related to operations such as questions, answers, and ratings that users perform within the system.

[0958] The present invention provides an interactive system in which users can develop a generative AI model through questions, answers, and ratings. This system has a mechanism for effectively collecting and storing information provided by users and providing it to the generative AI model for learning. Detailed modes for implementing the invention are described below.

[0959] Generating a Program

[0960] First, a user interface is provided that allows users to input questions. This interface runs on devices such as smartphones and provides a screen for users to input their interests and questions.

[0961] The server receives the input question data and stores it in a database. The hardware used here can be a general web server, and the database can be an RDBMS such as PostgreSQL.

[0962] Generative AI models are provided with question data entered by users, and the AI ​​models learn from this data. These AI models are typically implemented using Python libraries (e.g., TensorFlow and PyTorch).

[0963] Next, a user interface is provided that allows other users to enter answers to questions. This is also an interface on a smartphone, and users can enter and submit their answers.

[0964] The received response data is sent to the server and stored in a database. Similarly, this response data is also provided to the generative AI model for further learning.

[0965] Furthermore, a user interface is provided where users can view questions and answers. In this interface, questions and their answers are displayed and users can rate them. The rating data is sent to a server and stored in a database.

[0966] The saved evaluation data is provided to the generative AI model, which retrains the AI ​​model based on the evaluation results. The learning content of the generative AI model is also adjusted based on the feedback data.

[0967] A natural language description of the process

[0968] When a user inputs a question via a device, the device sends the question data to a server. The server stores the received question data in a database and provides the data to a generative AI model. The generative AI model learns from this question data and improves its ability to respond to users' interests and questions.

[0969] When other users enter their answers via their devices, the data is also sent to the server and stored in a database, which is then provided to the generative AI model, allowing it to further learn.

[0970] When a user rates an answer, the rating data is sent to the server and stored in a database. The generative AI model then retrains based on this rating data. Furthermore, by adjusting the AI ​​model's learning content based on feedback data, the accuracy of future answers is improved.

[0971] Below are some examples of prompt sentences.

[0972] Examples of prompt statements

[0973] "A user typed in the question: 'What's the latest anime-related news?'

[0974] User B then enters the following answer:

[0975] It has been announced that "Work X" will be made into a movie.

[0976] Other users found this answer helpful.

[0977] This information is used to grow the generative AI model, enabling it to provide more accurate information to users. In this way, the AI ​​model grows efficiently through user interaction, realizing the provision of more accurate information.

[0978] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0979] Step 1:

[0980] The user inputs a question via the terminal. The user interface has a text input field where the user inputs the question in text format. The input question is converted into JSON format data on the terminal and sent to the server. At this time, the input data includes the user ID, question content, and timestamp.

[0981] Step 2:

[0982] The server stores the received question data in a database. A unique ID is assigned to the question data and stored in the database. At the same time, the server provides this question data to the generative AI model, which then learns from the data. At this time, the server transfers the data to the generative AI model's learning engine, which processes and calculates the data using the question data as input.

[0983] Step 3:

[0984] Other users input answers to questions. The device's user interface provides a text input field for entering answers to questions displayed on the screen. The user enters the answer there, and the answer data is sent to the server in JSON format. The input data includes the user ID, question ID, answer content, and timestamp.

[0985] Step 4:

[0986] The server stores the received answer data in a database. The received data is stored along with a unique ID associated with the answer and associated with the question ID. At the same time, the server provides this answer data to a generative AI model, which then learns from the data.

[0987] Step 5:

[0988] The user browses questions and answers. The device's user interface sends a request to the server to retrieve a list of saved questions and answers. The server retrieves the relevant data from the database and sends it to the device. The device displays the received questions and answers in the user interface. This display includes a mechanism to prioritize new and popular answers.

[0989] Step 6:

[0990] The user enters a rating for the answer. The user interface provides buttons for entering ratings such as "helpful" or "not helpful." When the user presses the rating button, the rating data is sent to the server in JSON format. The rating data includes the user ID, answer ID, rating content, and timestamp.

[0991] Step 7:

[0992] The server stores the received evaluation data in a database. The evaluation data is stored with its associated unique ID and associated with the answer ID. At the same time, the server provides this evaluation data to the generative AI model, which uses the data for retraining. As a result of the retraining, the generative AI model grows to provide more accurate and useful answers.

[0993] Step 8:

[0994] The server provides the questions, answers, and evaluation data in the database to the generative AI model, and periodically retrains it. This allows the generative AI model to constantly learn the latest information. The learning content of the generative AI model is adjusted based on the feedback data. The performance of the AI ​​model improves as a result of data processing and data calculation.

[0995] Step 9:

[0996] The generative AI model generates prompts based on the questions, answers, and evaluation data. The generated prompts are used as further training data for the AI ​​model. The generated prompts are as follows:

[0997] "A user typed in the question: 'What's the latest anime-related news?'

[0998] User B then enters the following answer:

[0999] It has been announced that "Work X" will be made into a movie.

[1000] Other users found this answer helpful.

[1001] Based on this prompt, the generative AI model re-learns and improves.

[1002] 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.

[1003] The present invention combines an emotional engine with an interactive system in which users develop an artificial intelligence model through questions, answers, and ratings. This system recognizes the user's emotional state and can adjust the display order of questions and answers and the generation of feedback based on that state. The specific form and processing flow of this system are described below.

[1004] Enter and save your question

[1005] Enter your question

[1006] Users can input their interests and doubts as questions through the user interface of their terminal.

[1007] Example: User A types, "What's the latest anime-related news?"

[1008] Submit a Question

[1009] The terminal transmits the question data entered by the user to the server.

[1010] This data is sent in a format such as JSON and includes metadata such as the user ID, question, and timestamp.

[1011] Saving Question Data

[1012] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified.

[1013] Providing questions to the AI ​​model

[1014] The server provides the stored question data to the AI ​​model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[1015] Sentiment analysis of question data

[1016] The server provides the question data to the emotion engine to analyze the user's emotions.

[1017] Example: In response to User A's question, "What's the latest anime-related news?", the emotion engine recognizes emotions such as "interesting" and "expecting."

[1018] Enter and save your answers

[1019] Enter your answer

[1020] A user inputs an answer to a question posted by another user through a user interface.

[1021] Example: User B replies, "It has been announced that 'Work X' will be made into a movie."

[1022] Submit your answer

[1023] The device sends the answer data entered by the user to the server. The answer data is also sent in JSON format and includes the user ID, question ID, answer content, timestamp, etc.

[1024] Saving response data

[1025] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[1026] Providing answers to AI models

[1027] The server provides the stored response data to the AI ​​model, which then learns from it, allowing the AI ​​to accumulate more knowledge and improve accuracy.

[1028] Sentiment analysis of response data

[1029] The server provides the response data to an emotion engine to analyze the user's emotions.

[1030] Example: In response to User B's response, "It has been announced that 'Work X' will be made into a movie," the emotion engine recognizes emotions such as "joy" and "satisfaction."

[1031] View questions and answers

[1032] Ask questions and get answers

[1033] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[1034] View questions and answers

[1035] The device displays the received data on the user interface. It can prioritize new or popular answers and adjust the display order based on the user's emotions as recognized by the emotion engine.

[1036] For example, if the emotion engine indicates a high level of interest in response to the question "What is the latest anime-related news?", the question and related answers will be displayed preferentially.

[1037] Enter and save your rating

[1038] Enter your rating

[1039] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[1040] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[1041] Submit your rating

[1042] The terminal sends the entered evaluation data to the server. The evaluation data includes the user ID, answer ID, evaluation content, timestamp, etc.

[1043] Saving evaluation data

[1044] The server stores the received evaluation data in a database, and this evaluation data is also assigned a unique ID.

[1045] Sentiment analysis of rating data

[1046] The server provides the rating data to the emotion engine, which analyzes the user's emotions regarding the ratings.

[1047] Example: Recognizing emotions such as "satisfied" or "dissatisfied" from the content of the evaluation.

[1048] AI learning and accuracy improvement

[1049] Periodic Relearning

[1050] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining, which improves the accuracy of the AI ​​model.

[1051] Reflecting feedback

[1052] The server automatically adjusts the AI ​​model's learning based on user ratings and sentiment data, correcting incorrect answers and reinforcing correct answers.

[1053] Specific examples

[1054] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. The emotion engine also recognizes this question as "interest." Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model, with the emotion engine recognizing it as "joy." If user C views this question and answer and rates it as "helpful," the evaluation data is also saved, and the emotion engine recognizes it as "satisfaction." In this way, the system promotes the growth of AI that also reflects the user's emotions.

[1055] The processing flow will be explained below.

[1056] Post a Question

[1057] Step 1:

[1058] User enters a question

[1059] Users enter their questions using a web browser or mobile app user interface.

[1060] Example: User A types, "What's the latest anime-related news?"

[1061] Step 2:

[1062] Submit your question

[1063] The terminal transmits the question entered by the user to the server.

[1064] The transmitted data includes metadata such as the user ID, question content, and timestamp.

[1065] Example: {"User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}

[1066] Step 3:

[1067] Save questions

[1068] The server stores the received question data in a database.

[1069] Give the question a unique ID (e.g. Q789) and store it in the database.

[1070] Example: Save in the database as {"Question ID": "Q789", "User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}.

[1071] Step 4:

[1072] Providing questions to the AI ​​model

[1073] The server provides the saved question data to the AI ​​model and uses it as learning data.

[1074] Example: Providing {"question": "What's the latest anime news?"} to an AI model.

[1075] Step 5:

[1076] Sentiment analysis of question data

[1077] The server sends the received question data to the emotion engine to analyze the user's emotions.

[1078] Example: The emotion engine recognizes "interest" for the question "What's the latest anime-related news?"

[1079] Providing answers

[1080] Step 6:

[1081] User enters answer

[1082] A user inputs an answer to a question from another user.

[1083] Example: User B types, "It has been announced that 'Work X' will be made into a movie."

[1084] Step 7:

[1085] Sending answers

[1086] The terminal transmits the answer entered by the user to the server.

[1087] The transmitted data includes the user ID, question ID, answer content, timestamp, etc.

[1088] Example: {"User ID": "B456", "Question ID": "Q789", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}

[1089] Step 8:

[1090] Save answers

[1091] The server stores the received response data in a database.

[1092] The answer is given a unique ID (e.g. A987) and stored in association with the question ID.

[1093] Example: Save in the database as {"Answer ID": "A987", "Question ID": "Q789", "User ID": "B456", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}.

[1094] Step 9:

[1095] Providing answers to the AI ​​model

[1096] The server provides the saved response data to the AI ​​model and uses it as learning data.

[1097] Example: Providing {"answer": "It has been announced that 'Book X' will be made into a movie."} to an AI model.

[1098] Step 10:

[1099] Sentiment analysis of response data

[1100] The server sends the received response data to an emotion engine to analyze the user's emotions.

[1101] Example: The emotion engine recognizes "joy" in response to the answer "It has been announced that 'Work X' will be made into a movie."

[1102] View questions and answers

[1103] Step 11:

[1104] Ask questions and get answers

[1105] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[1106] Example: Request {"Question ID": "Q789"} and the server will send the related questions and answers.

[1107] Step 12:

[1108] View questions and answers

[1109] The terminal displays the received question and answer data on a user interface.

[1110] Example: When a user accesses the site, the question "What's the latest anime-related news?" and the answer "It has been announced that 'Work X' will be made into a movie" are displayed on the screen.

[1111] Step 13:

[1112] Adjusting display order based on emotions

[1113] The server adjusts the display order of questions and answers based on the emotion data recognized by the emotion engine.

[1114] For example, questions that you have shown interest in will be displayed before other questions.

[1115] Evaluating answers

[1116] Step 14:

[1117] User-initiated rating input

[1118] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[1119] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[1120] Step 15:

[1121] Submit your evaluation

[1122] The terminal transmits the input evaluation data to the server.

[1123] The transmitted data includes the user ID, answer ID, evaluation details, timestamp, etc.

[1124] Example: {"User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}

[1125] Step 16:

[1126] Save evaluation details

[1127] The server stores the received evaluation data in a database.

[1128] Evaluation data will also be assigned a unique ID (e.g., E123).

[1129] Example: Store in database as {"Rating ID": "E123", "User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}.

[1130] Step 17:

[1131] Sentiment analysis of rating data

[1132] The server transmits the received evaluation data to an emotion engine to analyze the user's emotions.

[1133] For example, the sentiment engine recognizes the rating "helpful" as "satisfied."

[1134] AI learning and accuracy improvement

[1135] Step 18:

[1136] Periodic retraining of AI models

[1137] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[1138] Example: Retraining an AI model using all question and answer and rating data from the past month.

[1139] Step 19:

[1140] Reflecting feedback

[1141] The server automatically adjusts the learning content of the AI ​​model based on user evaluation data and emotion data from the emotion engine.

[1142] Example: Learning from an answer that was rated "not helpful" by many users can be corrected to provide more accurate answers.

[1143] Through the above processing steps, the present invention realizes an interactive system that allows users to actively participate in the growth of AI, and provides advanced information by combining an emotion engine.

[1144] Example 2

[1145] 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."

[1146] Modern information systems are required to not only provide information based on user input, but also to understand the user's emotions and intentions and provide feedback accordingly. However, conventional systems have had difficulty in realizing advanced interactions such as analyzing the user's emotions and adjusting the order in which information is displayed based on those emotions.

[1147] 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.

[1148] In this invention, the server includes means for providing a user interface through which a user can input a question, means for transmitting input question data to the server, means for saving received question data in a database, means for providing the question data saved in the database to the artificial intelligence model, means for providing the question data to an emotion engine and analyzing the user's emotions, means for providing a user interface through which other users can input answers to questions, means for transmitting input answer data to the server, means for saving received answer data in a database, means for providing the answer data saved in the database to the artificial intelligence model, means for providing the answer data to the emotion engine and analyzing the user's emotions, means for providing a user interface through which users can view questions and answers, means for adjusting the display order of questions and answers based on the results of the user emotion analysis, means for providing a user interface through which users can input ratings for answers, means for transmitting input rating data to the server, means for saving received rating data in a database, means for providing the rating data to the emotion engine and analyzing the user's emotions, means for providing the questions, answers, and rating data in the database to the artificial intelligence model and performing relearning, and means for adjusting the learning content of the artificial intelligence model based on the feedback data. This makes it possible to provide information that reflects the user's emotions and to create an interactive system that adjusts the display based on that information.

[1149] "User interface" refers to the screens and operating means that allow users to interact with the system to exchange information.

[1150] "Question data" is information including the question entered by the user, and includes metadata such as the user ID, question content, and timestamp.

[1151] A "server" refers to a computer system that receives requests over a network and is responsible for processing and storing data.

[1152] A "database" refers to a data management system that allows a system to centrally store data and make it searchable and retrievalable.

[1153] An "artificial intelligence model" refers to an algorithm that learns patterns and rules from large amounts of data and makes inferences and predictions based on input.

[1154] An "emotion engine" is an algorithm or system that analyzes user input data and identifies the emotion or intent behind it.

[1155] "Answer data" is information including the answer entered by the user to the question, and includes the user ID, question ID, answer, timestamp, and the like.

[1156] "Evaluation data" is information including the content of the user's evaluation of an answer, and includes the user ID, answer ID, evaluation content, timestamp, and the like.

[1157] "Retraining" refers to the learning process of using new data in a database to improve the accuracy of an artificial intelligence model.

[1158] "Feedback data" refers to information used to adjust system processing based on user evaluations and the results of sentiment analysis.

[1159] The present invention is an interactive system that combines an emotion engine and aims to develop an artificial intelligence model through user questions, answers, and ratings. This system recognizes the user's emotional state and can adjust the display order of questions and answers and the generation of feedback based on that state. The specific form and processing flow of the system are described below.

[1160] Enter and save your question

[1161] Users input their interests and concerns as questions through the user interface of their device. This user interface provides a text input area, allowing users to enter the content of their questions as text. For example, User A enters "What's the latest anime-related news?" This question data is sent from the device to the server in a format such as JSON, and includes metadata such as the user ID, question content, and timestamp.

[1162] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified. The server then provides the stored question data to an artificial intelligence model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[1163] The question data is also provided to the emotion engine, which analyzes the user's emotions. For example, in response to User A's question, "What's the latest anime-related news?", the emotion engine recognizes emotions such as "interesting" and "expecting." The results are saved as metadata.

[1164] Enter and save your answers

[1165] Users input answers to questions posted by other users through the user interface. For example, User B answers, "It has been announced that 'Work X' will be made into a movie." This answer data is sent from the device to the server in JSON format or similar, and includes the user ID, question ID, answer content, and timestamp.

[1166] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID. The server then provides the stored answer data to an artificial intelligence model, which then learns from the data. This allows the AI ​​to accumulate more knowledge and improve accuracy.

[1167] The response data is also provided to the emotion engine, which analyzes the user's emotions. For example, in response to User B's response, "It has been announced that 'Work X' will be made into a movie," the emotion engine recognizes emotions such as "joy" and "satisfaction." These results are also saved as metadata.

[1168] View questions and answers

[1169] The device requests a list of questions and their answers from the server, and receives the data from the server. This request includes the user ID and the specific question ID, and the server searches for the corresponding data and returns it to the device.

[1170] The device displays the received data on the user interface. It is possible to display newer or more popular answers with priority. The display order is also adjusted based on the user's emotions as recognized by the emotion engine. For example, if the emotion engine indicates a high level of "interest" in response to the question "What's the latest anime-related news?", that question and related answers will be displayed with priority.

[1171] Enter and save your rating

[1172] The user inputs a rating such as "helpful" or "not helpful" for the displayed answer. For example, a user may rate the answer "It has been announced that 'Work X' will be made into a movie." as "helpful." This rating data is sent from the device to the server and includes the user ID, answer ID, rating content, and timestamp.

[1173] The server stores the received evaluation data in a database. This evaluation data is also assigned a unique ID. The evaluation data is also provided to the emotion engine, which analyzes the user's emotions. For example, emotions such as "satisfied" or "dissatisfied" are recognized from the content of the evaluation. This result is also stored as metadata.

[1174] AI learning and accuracy improvement

[1175] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for re-learning. For example, once a week, the server collects all new data and re-trains the AI ​​model. This improves the accuracy of the AI ​​model.

[1176] Furthermore, the server automatically adjusts the learning content of the AI ​​model based on user evaluation and emotion data. If there are many incorrect answers, the data is corrected and correct answers are strengthened, improving the model's performance.

[1177] Specific examples

[1178] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. The emotion engine also recognizes this question as "interest." Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model, with the emotion engine recognizing it as "joy." If user C views this question and answer and rates it as "helpful," the evaluation data is also saved, and the emotion engine recognizes it as "satisfaction." In this way, the system promotes the growth of AI that also reflects user emotions.

[1179] Prompt Sentence Examples

[1180] "What's the latest anime-related news?"

[1181] "It has been announced that 'Work X' will be made into a movie."

[1182] "It was helpful"

[1183] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1184] Step 1:

[1185] Enter your question

[1186] The user enters a question through the user interface of their terminal.

[1187] Specific actions: Enter a question in the text input area and click the submit button.

[1188] Input: The question the user types into the text entry area.

[1189] Output: JSON formatted question data sent by the device.

[1190] Step 2:

[1191] Submit a Question

[1192] The terminal transmits the question data entered by the user to the server.

[1193] Specific operation: The device sends question data including metadata such as the user ID, question content, and timestamp to the server.

[1194] Input: The question data entered by the user.

[1195] Output: The question data sent to the server.

[1196] Step 3:

[1197] Saving Question Data

[1198] The server stores the received question data in a database, and each question is assigned a unique ID.

[1199] Specific operation: The server assigns a unique ID to the question data and saves it in the database.

[1200] Input: Question data sent from the terminal.

[1201] Output: Question data stored in a database.

[1202] Step 4:

[1203] Providing questions to the AI ​​model

[1204] The server provides the stored question data to the AI ​​model, which then learns from the data.

[1205] Specific operation: The server reads question data from the database and sends it to the AI ​​model for learning.

[1206] Input: Question data stored in the database.

[1207] Output: The input data for the AI ​​model to learn from.

[1208] Step 5:

[1209] Sentiment analysis of question data

[1210] The server provides the question data to the emotion engine to analyze the user's emotions.

[1211] Specific operation: The server sends the question data to the emotion engine, which analyzes the emotion.

[1212] Input: Question data stored in the database.

[1213] Output: Sentiment data parsed by the sentiment engine.

[1214] Step 6:

[1215] Enter your answer

[1216] A user inputs an answer to a question posted by another user through a user interface.

[1217] Specific actions: Enter your answer in the text input area and click the submit button.

[1218] Input: The answer the user entered into the text entry area.

[1219] Output: JSON formatted response data sent by the device.

[1220] Step 7:

[1221] Submit your answer

[1222] The terminal transmits the answer data entered by the user to the server.

[1223] Specific operation: The device sends the answer data, including metadata such as the user ID, question ID, answer content, and timestamp, to the server.

[1224] Input: The answer data entered by the user.

[1225] Output: The response data sent to the server.

[1226] Step 8:

[1227] Saving response data

[1228] The server stores the received response data in a database, and each response is assigned a unique ID.

[1229] Specific operation: The server assigns a unique ID to the response data and saves it in the database.

[1230] Input: Response data sent from the device.

[1231] Output: Response data stored in a database.

[1232] Step 9:

[1233] Providing answers to AI models

[1234] The server provides the saved response data to the AI ​​model, which then learns from the data.

[1235] Specific operation: The server reads the response data from the database and sends it to the AI ​​model for learning.

[1236] Input: Response data stored in the database.

[1237] Output: The input data for the AI ​​model to learn from.

[1238] Step 10:

[1239] Sentiment analysis of response data

[1240] The server provides the response data to an emotion engine to analyze the user's emotions.

[1241] Specific operation: The server sends the response data to the emotion engine, which analyzes the emotion.

[1242] Input: Response data stored in the database.

[1243] Output: Sentiment data parsed by the sentiment engine.

[1244] Step 11:

[1245] Ask questions and get answers

[1246] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[1247] Specific operation: The device sends a request to the server, and the server returns the relevant question and answer data.

[1248] Input: The request from the device (including the user ID and the specific question ID).

[1249] Output: Question and answer data received from the server.

[1250] Step 12:

[1251] View questions and answers

[1252] The terminal displays the received data on a user interface.

[1253] Specific operation: The device converts the data into a display format and displays it on the user interface. The display order is adjusted based on the user's emotions recognized by the emotion engine.

[1254] Input: Question and answer data received from the server.

[1255] Output: The question and answer data that is displayed in the user interface.

[1256] Step 13:

[1257] Enter your rating

[1258] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[1259] Specific actions: Click the rating button, select the rating content, and click the submit button.

[1260] Input: The rating selected by the user.

[1261] Output: JSON formatted rating data sent by the device.

[1262] Step 14:

[1263] Submit your rating

[1264] The terminal transmits the input evaluation data to the server.

[1265] Specific operation: The device sends evaluation data including metadata such as user ID, answer ID, evaluation content, and timestamp to the server.

[1266] Input: The rating data entered by the user.

[1267] Output: The rating data sent to the server.

[1268] Step 15:

[1269] Saving evaluation data

[1270] The server stores the received evaluation data in a database. The evaluation data is also assigned a unique ID.

[1271] Specific operation: The server assigns a unique ID to the evaluation data and saves it in the database.

[1272] Input: Rating data sent from the device.

[1273] Output: Evaluation data stored in a database.

[1274] Step 16:

[1275] Sentiment analysis of rating data

[1276] The server provides the rating data to the emotion engine to analyze the user's emotions.

[1277] Specific operation: The server sends the evaluation data to the emotion engine, which analyzes the emotions.

[1278] Input: Evaluation data stored in a database.

[1279] Output: Sentiment data parsed by the sentiment engine.

[1280] Step 17:

[1281] Retraining the AI

[1282] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[1283] What it does: The server reads all new data from the database and periodically retrains the AI ​​model.

[1284] Input: Question, answer, and assessment data stored in a database.

[1285] Output: The retrained AI model.

[1286] Step 18:

[1287] Reflecting feedback

[1288] The server automatically adjusts the learning content of the AI ​​model based on user evaluations and emotional data.

[1289] Specific operation: The server adjusts the learning weights of the AI ​​model based on the evaluation and emotion data and corrects incorrect answers.

[1290] Input: Rating data, emotion data.

[1291] Output: The tuned AI model.

[1292] (Application example 2)

[1293] 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."

[1294] Conventional customer support systems have difficulty in responding to users' emotional states, making it difficult to improve user experience and service quality. In addition, there has been a lack of methods to provide artificial intelligence models that can learn interactively by reflecting user feedback and emotions.

[1295] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for providing a user interface through which a user can input a question; means for transmitting the input question data to the server; means for saving the received question data in a database; means for providing the question data saved in the database to the AI ​​model; means for providing a user interface through which other users can input answers to questions; means for transmitting the input answer data to the server; means for saving the received answer data in the database; means for providing the answer data saved in the database to the AI ​​model; means for providing a user interface through which users can view questions and answers; means for providing a user interface through which users can input ratings for answers; means for transmitting the input rating data to the server; means for saving the received rating data in the database; means for providing the questions, answers, and rating data in the database to the AI ​​model and performing re-learning; means for adjusting the learning content of the AI ​​model based on feedback data; means for analyzing the emotional state of the input question and answer and means for adjusting the display order and feedback based on the emotional state. This makes it possible to realize an interactive customer support system that reflects the user's emotions.

[1296] A "user interface" is an interface through which a user interacts directly with a system and inputs questions and answers.

[1297] A "server" is a computer that receives question data and answer data sent by users and stores them in a database.

[1298] A "database" is a storage system that centrally manages and stores question data, response data, and evaluation data.

[1299] An "artificial intelligence model" is an algorithm that learns from questions and answers entered by users and improves the quality of subsequent questions and answers.

[1300] An "emotional state" is a psychological state recognized as a result of the emotion engine's analysis of the questions and answers entered by the user.

[1301] "Display order" refers to the order in which questions and answers are displayed when a user browses.

[1302] "Feedback" refers to the evaluations and comments that users make on answers, and is data used to train the system.

[1303] "Relearning" is a process in which an artificial intelligence model periodically learns based on data accumulated in a database to improve its accuracy.

[1304] "Question data" is data including the content of questions posed to the system by users.

[1305] "Answer data" is data including answers provided by other users to questions.

[1306] "Evaluation data" is data including the evaluation content given by the user to the answer.

[1307] The "emotion engine" is a system component that analyzes the emotional state of input questions and answers.

[1308] The present invention combines an emotion engine with an interactive system in which users can develop an artificial intelligence model through questions, answers, and ratings. Specific embodiments of this system are described below.

[1309] Enter and save your question

[1310] User Interface:

[1311] It provides a user interface for users to input questions. Users can input their interests and questions and send the question data to the server via their terminal. This data includes metadata such as the user ID, question content, and timestamp.

[1312] server:

[1313] The server stores the received question data in a database and assigns a unique ID to uniquely identify it.The server then provides the stored question data to an artificial intelligence model for learning.Furthermore, the server passes the question data to an emotion engine to analyze the user's emotional state.

[1314] Enter and save your answers

[1315] User Interface:

[1316] It provides a user interface that allows other users to enter answers to questions. The entered answer data includes information such as the answer content, user ID, question ID, and timestamp.

[1317] server:

[1318] The server receives the response data and stores it in a database. Here, a unique ID is assigned to each response, making it uniquely identified. The stored response data is then provided to an artificial intelligence model, which further trains the AI. The response data is also passed to an emotion engine, which analyzes the user's emotional state.

[1319] View questions and answers

[1320] User Interface:

[1321] It provides an interface where users can browse questions and their answers. The interface is adjusted to prioritize new and popular answers. The display order is also dynamically adjusted based on the user's emotions as recognized by the emotion engine.

[1322] Enter and save your rating

[1323] User Interface:

[1324] Users can rate the displayed answers. The rating data includes the user ID, answer ID, rating content, timestamp, etc.

[1325] server:

[1326] The evaluation data is received and stored in a database. The evaluation data is analyzed by an emotion engine to recognize the user's emotional state. The acquired evaluation data is used to retrain the AI ​​model.

[1327] AI model training and accuracy improvement

[1328] server:

[1329] The server provides the AI ​​model with the questions, answers, and evaluation data from the database and periodically retrains it, improving its accuracy. It also automatically adjusts its learning based on user feedback and sentiment data.

[1330] Hardware and software used

[1331] Hardware: Smartphones, head-mounted displays, servers

[1332] Software: Emotion engine library, AI model library, database connection library

[1333] Specific example explanation

[1334] For example, User A opens the app to ask a question about a product in a store and types the question, "Does this product run on batteries?" This question is sent to the server and stored in the database. The AI ​​model learns this question, and the emotion engine recognizes it as "curiosity." As a result, the store staff is notified, "We have received your question. Please check the details." An example of this prompt is as follows:

[1335] Prompt Sentence Examples

[1336] User Question:

[1337] Question: 'Does this item run on batteries?'

[1338] User ID: 'user123'

[1339] Timestamp: '2023-10-10T10:00:00Z'

[1340] Emotion Engine Analysis Results:

[1341] Emotional state: 'Curiosity'

[1342] Action: 'Increase display priority'

[1343] In this way, the present invention provides an interactive customer support system that reflects the user's feelings.

[1344] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1345] Step 1:

[1346] The user inputs a question using a smartphone or head-mounted display. The question data input from the user interface includes the user ID, question content, and timestamp. For example, the user might input, "Does this product run on batteries?"

[1347] Step 2:

[1348] The device sends the entered question data in JSON format to the server, which includes the user ID, question content, and timestamp.

[1349] Step 3:

[1350] The server stores the received question data in a database. A unique ID is assigned when the data is saved to uniquely identify it. The input data includes the user ID, question content, timestamp, and unique ID.

[1351] Step 4:

[1352] The server provides the stored question data to the AI ​​model for learning. Based on the provided data, the AI ​​model analyzes the question content and improves the accuracy of answers to future questions.

[1353] Step 5:

[1354] The server passes the question data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the question and outputs an emotion such as "curiosity."

[1355] Step 6:

[1356] Another user inputs an answer to the question using a smartphone or head-mounted display. The answer data input from the user interface includes the answer content, user ID, question ID, and timestamp. For example, the user might input, "Yes, this product runs on batteries."

[1357] Step 7:

[1358] The device sends the entered answer data in JSON format to the server. The sent data includes the user ID, question ID, answer content, and timestamp.

[1359] Step 8:

[1360] The server stores the received answer data in a database. A unique ID is assigned when saving the data to uniquely identify it. The input data includes the user ID, question ID, answer content, timestamp, and unique ID.

[1361] Step 9:

[1362] The server provides the stored response data to the AI ​​model for learning. Based on the provided data, the AI ​​model analyzes the content of the responses and improves the accuracy of future responses.

[1363] Step 10:

[1364] The server passes the answer data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the answer and outputs an emotion such as "a sense of relief."

[1365] Step 11:

[1366] The user interface displays questions and answers to the user. The display data includes the question content and multiple answers. The display order is dynamically adjusted based on the analysis results of the emotion engine so that newer and more popular answers are displayed preferentially.

[1367] Step 12:

[1368] Users rate the displayed answers. The rating data includes the user ID, answer ID, rating content, and timestamp. For example, a user may rate the answer as "helpful."

[1369] Step 13:

[1370] The device sends the entered evaluation data in JSON format to the server. The sent data includes the user ID, answer ID, evaluation content, and timestamp.

[1371] Step 14:

[1372] The server stores the received evaluation data in a database. A unique ID is assigned when the data is saved to uniquely identify it. The input data includes the user ID, answer ID, evaluation content, timestamp, and unique ID.

[1373] Step 15:

[1374] The server passes the evaluation data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the evaluation content and outputs an emotion such as "satisfaction."

[1375] Step 16:

[1376] The server provides the AI ​​model with the questions, answers, and evaluation data stored in the database and periodically retrains it, improving its accuracy. Furthermore, by automatically adjusting the learning content based on emotion data, the AI ​​model can provide more appropriate responses to users.

[1377] 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.

[1378] 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.

[1379] 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.

[1380] [Third embodiment]

[1381] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1382] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1383] 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).

[1384] 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.

[1385] 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.

[1386] 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).

[1387] 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.

[1388] 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.

[1389] 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.

[1390] 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.

[1391] 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.

[1392] 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."

[1393] This invention is an interactive system that allows users to develop an artificial intelligence model through questions, answers, and ratings. Its details include a process for effectively collecting and storing information entered by users and providing it to the AI ​​model for learning. The specific form and processing flow of the system are described below.

[1394] Enter and save your question

[1395] Enter your question

[1396] Users can input their interests and doubts as questions through the user interface of their terminal.

[1397] Example: User A types, "What's the latest anime-related news?"

[1398] Submit a Question

[1399] The terminal transmits the question data entered by the user to the server.

[1400] This data is sent in a format such as JSON and includes metadata such as the user ID, question, and timestamp.

[1401] Saving Question Data

[1402] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified.

[1403] Providing questions to the AI ​​model

[1404] The server provides the stored question data to the AI ​​model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[1405] Enter and save your answers

[1406] Enter your answer

[1407] A user inputs an answer to a question posted by another user through a user interface.

[1408] Example: User B replies, "It has been announced that 'Work X' will be made into a movie."

[1409] Submit your answer

[1410] The device sends the answer data entered by the user to the server. The answer data is also sent in JSON format and includes the user ID, question ID, answer content, timestamp, etc.

[1411] Saving response data

[1412] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[1413] Providing answers to AI models

[1414] The server provides the stored response data to the AI ​​model, which then learns from it, allowing the AI ​​to accumulate more knowledge and improve accuracy.

[1415] View questions and answers

[1416] Ask questions and get answers

[1417] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[1418] View questions and answers

[1419] The device displays the received data on the user interface. It is also possible to display newer or more popular answers preferentially.

[1420] Enter and save your rating

[1421] Enter your rating

[1422] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[1423] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[1424] Submit your rating

[1425] The terminal sends the entered evaluation data to the server. The evaluation data includes the user ID, answer ID, evaluation content, timestamp, etc.

[1426] Saving evaluation data

[1427] The server stores the received evaluation data in a database, and this evaluation data is also assigned a unique ID.

[1428] AI learning and accuracy improvement

[1429] Periodic Relearning

[1430] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for re-training, thereby improving the accuracy of the AI ​​model.

[1431] Reflecting feedback

[1432] The server automatically adjusts the AI ​​model's learning based on user evaluation data, thereby correcting incorrect answers and strengthening correct answers.

[1433] Specific examples

[1434] For example, if user A posts a question such as "What's the latest anime-related news?", the server saves this question in a database and provides it to the AI ​​model for learning. Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also saved in the database and provided to the AI ​​model. If user C views this question and answer and rates it as "helpful," this rating data is also saved and used to retrain the AI ​​model.

[1435] In this way, users can actively participate in the growth of the AI, and the system provides highly accurate information that reflects their own knowledge.

[1436] The processing flow will be explained below.

[1437] Post a Question

[1438] Step 1:

[1439] User enters a question

[1440] Users enter their questions using a web browser or mobile app user interface.

[1441] Example: User A types, "What's the latest anime-related news?"

[1442] Step 2:

[1443] Submit your question

[1444] The terminal transmits the question entered by the user to the server.

[1445] The transmitted data includes metadata such as the user ID, question content, and timestamp.

[1446] Example: {"User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}

[1447] Step 3:

[1448] Save questions

[1449] The server stores the received question data in a database.

[1450] Give the question a unique ID (e.g. Q789) and store it in the database.

[1451] Example: Save in the database as {"Question ID": "Q789", "User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}.

[1452] Step 4:

[1453] Providing questions to the AI ​​model

[1454] The server provides the saved question data to the AI ​​model and uses it as learning data.

[1455] Example: Providing {"question": "What's the latest anime news?"} to an AI model.

[1456] Providing answers

[1457] Step 5:

[1458] User enters answer

[1459] A user inputs an answer to a question from another user.

[1460] Example: User B types, "It has been announced that 'Work X' will be made into a movie."

[1461] Step 6:

[1462] Sending answers

[1463] The terminal transmits the answer entered by the user to the server.

[1464] The transmitted data includes the user ID, question ID, answer content, timestamp, etc.

[1465] Example: {"User ID": "B456", "Question ID": "Q789", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}

[1466] Step 7:

[1467] Save answers

[1468] The server stores the received response data in a database.

[1469] Give the answer a unique ID (e.g. A987) and store it with the question ID.

[1470] Example: Save in the database as {"Answer ID": "A987", "Question ID": "Q789", "User ID": "B456", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}.

[1471] Step 8:

[1472] Providing answers to the AI ​​model

[1473] The server provides the saved response data to the AI ​​model and uses it as learning data.

[1474] Example: Providing {"answer": "It has been announced that 'Book X' will be made into a movie."} to an AI model.

[1475] View questions and answers

[1476] Step 9:

[1477] Ask questions and get answers

[1478] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[1479] Example: Request data for {"Question ID": "Q789"} and receive question and answer data from the server.

[1480] Step 10:

[1481] View questions and answers

[1482] The terminal displays the received question and answer data on a user interface.

[1483] Example: When a user accesses the site, the question "What's the latest anime-related news?" and the answer "It has been announced that 'Work X' will be made into a movie" are displayed on the screen.

[1484] Evaluating answers

[1485] Step 11:

[1486] User-initiated rating input

[1487] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[1488] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[1489] Step 12:

[1490] Submit your evaluation

[1491] The terminal transmits the input evaluation data to the server.

[1492] The transmitted data includes the user ID, answer ID, evaluation details, timestamp, etc.

[1493] Example: {"User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}

[1494] Step 13:

[1495] Save evaluation details

[1496] The server stores the received evaluation data in a database.

[1497] Evaluation data will also be assigned a unique ID (e.g., E123).

[1498] Example: Store in database as {"Rating ID": "E123", "User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}.

[1499] AI learning and accuracy improvement

[1500] Step 14:

[1501] Periodic retraining of AI models

[1502] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[1503] Example: Retraining an AI model using all question and answer and rating data from the past month.

[1504] Step 15:

[1505] Reflecting feedback

[1506] The server automatically adjusts the learning content of the AI ​​model based on user evaluation data.

[1507] Example: Revisiting learning based on answers that were rated "not helpful" by many users.

[1508] Example 1

[1509] 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."

[1510] Conventional AI model learning systems have faced issues with inefficient management of user-entered questions, answers, and evaluation data, resulting in delayed improvements in the accuracy of AI models. Furthermore, there was a lack of a mechanism for immediately reflecting user feedback and using it to improve the model, making it difficult to improve the user experience. Furthermore, when displaying questions and answers, there was a lack of a function to prioritize the latest information and popular answers that users wanted, reducing the efficiency of information acquisition.

[1511] 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.

[1512] In this invention, the server includes a means for providing a user interface, a means for transmitting question data, answer data, and evaluation data to the server, a means for storing the received data in a database, a means for providing the data to the AI ​​model as learning data, a means for acquiring the question and answer data and displaying them on the user interface, and a means for preferentially displaying new or popular answers. This enables efficient collection, management, and provision of various data entered by users, resulting in rapid improvement in the accuracy of the AI ​​model and improved functionality of the user interface.

[1513] "User interface" refers to the screen or operating means through which a user can access a system and input and view information such as questions and answers.

[1514] "Question data" refers to data that includes information about the question entered by the user, including metadata such as the user ID, question content, and timestamp.

[1515] "Server" refers to a central computing device that receives, processes, stores, and provides data sent by users to AI models.

[1516] "Database" refers to a system in which a server systematically stores question data, answer data, evaluation data, etc. and manages them so that they can be retrieved as needed.

[1517] "Artificial intelligence model" refers to a machine learning algorithm or neural network that learns from user-provided data and answers questions, provides information, or generates responses.

[1518] "Answer data" refers to data that includes answers to questions posted by other users, and includes metadata such as user ID, question ID, answer content, and timestamp.

[1519] "Rating data" refers to data that includes information about the ratings that users have given to questions and answers, and includes metadata such as user ID, rating content, answer ID, and timestamp.

[1520] "Retraining" refers to the learning process in which questions, answers, and evaluation data stored in a database are periodically provided to an artificial intelligence model to improve the model's accuracy.

[1521] "Feedback data" refers to data used to adjust the model's learning content based on user evaluation data and improve the model's output results.

[1522] MODE FOR CARRYING OUT THE INVENTION

[1523] This invention is an interactive system in which users can develop an artificial intelligence model through questions, answers, and evaluations. The specific form and processing flow of this system will be described in detail below.

[1524] Enter and save your question

[1525] When a user types a question

[1526] A user accesses a dedicated user interface using their own terminal (for example, a personal computer or a smartphone) and inputs a question. An example of a question is "What is the latest anime-related news?"

[1527] Sending and storing question data

[1528] The device sends the question entered by the user to the server in JSON format, including the user ID, question content, timestamp, etc.

[1529] The server stores the received question data in a database (e.g., MySQL or PostgreSQL). A unique ID is assigned to each question, allowing it to be uniquely identified.

[1530] Providing question data to the AI ​​model

[1531] The server can provide the stored question data to the AI ​​model, which can then use the data as training data. This process allows the AI ​​model to learn new information from user questions and improve the accuracy of its responses to future questions.

[1532] Enter and save your answers

[1533] When the user enters an answer

[1534] Users can enter answers to questions posted by other users. An example of an answer might be, "It has been announced that 'Work X' will be made into a movie."

[1535] Sending and storing response data

[1536] The device sends the answers entered by the user to the server in JSON format, including the user ID, question ID, answer content, and timestamp.

[1537] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[1538] Providing response data to the AI ​​model

[1539] The server can provide the saved response data to the AI ​​model, which can then use the data as training data. This process allows the AI ​​model to learn new information from the user's responses and further improve the accuracy of future responses.

[1540] View questions and answers

[1541] Retrieving and Displaying Data

[1542] The device requests a list of questions and their answers from the server and receives the necessary data. The received data is displayed on the user interface. It is also possible to display new or popular answers preferentially, allowing users to quickly obtain the latest information and useful information.

[1543] Enter and save your rating

[1544] When a user enters a rating

[1545] Users can enter a rating for the displayed answer, such as "helpful" or "not helpful."

[1546] Sending and storing evaluation data

[1547] The device sends the entered evaluation data in JSON format to the server, which includes the user ID, answer ID, evaluation content, and timestamp.

[1548] The server stores the received rating data in a database, and each rating is assigned a unique ID.

[1549] AI learning and accuracy improvement

[1550] Periodic Relearning

[1551] The server periodically provides the question data, answer data, and evaluation data in the database to the AI ​​model for re-learning. This process improves the accuracy of the AI ​​model, enabling it to provide more accurate information to users.

[1552] Reflecting feedback

[1553] The server automatically adjusts the learning content of the AI ​​model based on the user's evaluation data, thereby correcting incorrect answers and strengthening correct answers.

[1554] Prompt Sentence Examples

[1555] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model. If user C views this question and answer and rates it as "helpful," this rating data is also saved and used to retrain the AI ​​model. In this way, the system allows users to actively participate in the growth and improvement of the AI's accuracy.

[1556] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1557] Step 1: User enters question

[1558] A user uses a terminal to access the user interface and inputs a question by typing the question in a text box and clicking a submit button.

[1559] Input: User ID, question, timestamp

[1560] Output: Question data (JSON format)

[1561] Step 2: The device sends the query data

[1562] The device converts the question data entered by the user into JSON format and sends it to the server. This data includes the user ID, question content, and timestamp.

[1563] Input: Question data (JSON format)

[1564] Output: Data packet to be sent

[1565] Step 3: The server receives the query data

[1566] The server receives the submitted question data and stores it in a database. As an additional process, it assigns a unique ID to each question.

[1567] Input: Data packet to send

[1568] Output: Question data stored in the database (with unique ID)

[1569] Step 4: The server provides the question data to the AI ​​model

[1570] The server extracts question data stored in the database and provides it to the AI ​​model for training. The AI ​​model uses this information to add new knowledge and improve the accuracy of future question answers.

[1571] Input: Question data stored in the database

[1572] Output: Provided as training data to the AI ​​model

[1573] Step 5: User Enters Answer

[1574] A user can enter an answer to a question posted by another user by selecting the question in the user interface, entering the answer in the text box that appears, and clicking the submit button.

[1575] Input: User ID, Question ID, Answer, Time Stamp

[1576] Output: Response data (JSON format)

[1577] Step 6: The device sends the response data

[1578] The device converts the answer data entered by the user into JSON format and sends it to the server. This data includes the user ID, question ID, answer content, and timestamp.

[1579] Input: Response data (JSON format)

[1580] Output: Data packet to be sent

[1581] Step 7: The server receives the response data

[1582] The server receives the submitted answer data and stores it in a database. As an additional process, it assigns a unique ID to each answer and associates it with the question ID.

[1583] Input: Data packet to send

[1584] Output: Answer data saved in the database (with unique ID)

[1585] Step 8: The server provides the answer data to the AI ​​model

[1586] The server extracts the response data stored in the database and provides it to an artificial intelligence model for learning. The AI ​​model uses this information to add new knowledge and improve the accuracy of future responses.

[1587] Input: Response data stored in the database

[1588] Output: Provided as training data to the AI ​​model

[1589] Step 9: The device requests data

[1590] The terminal requests question and answer data from the server in response to a user request, which is sent by clicking the "Show Questions and Answers" button on the user interface.

[1591] Input: Request data

[1592] Output: None (request sent)

[1593] Step 10: Server sends data

[1594] The server extracts the requested question and answer data from the database and sends it to the device in JSON format.

[1595] Input: Request data

[1596] Output: Question and answer data in JSON format

[1597] Step 11: The terminal displays the data

[1598] The terminal displays the question and answer data received from the server on a user interface, and can also display new or popular answers with priority.

[1599] Input: Question and answer data in JSON format

[1600] Output: Questions and answers displayed in the user interface

[1601] Step 12: User Enters Rating

[1602] The user inputs a rating for the displayed answer, such as "helpful" or "not helpful," and selects one and clicks the submit button.

[1603] Input: User ID, Answer ID, Evaluation details, Time stamp

[1604] Output: Evaluation data (JSON format)

[1605] Step 13: The device sends the evaluation data

[1606] The device converts the evaluation data entered by the user into JSON format and sends it to the server. This data includes the user ID, answer ID, evaluation content, and timestamp.

[1607] Input: Evaluation data (JSON format)

[1608] Output: Data packet to be sent

[1609] Step 14: Server receives reputation data

[1610] The server receives the evaluation data and stores it in a database. Each evaluation is assigned a unique ID.

[1611] Input: Data packet to send

[1612] Output: Evaluation data stored in the database (with unique ID)

[1613] Step 15: Server periodically retrains

[1614] The server periodically provides the question data, answer data, and evaluation data in the database to the AI ​​model for re-learning, a process that improves the accuracy of the AI ​​model.

[1615] Input: Question data, response data, and evaluation data stored in the database

[1616] Output: Retrained AI model

[1617] Step 16: The server processes the feedback

[1618] The server automatically adjusts the AI ​​model's learning based on user evaluation data, improving the accuracy of answers and correcting incorrect answers.

[1619] Input: Evaluation data

[1620] Output: The tuned AI model

[1621] (Application example 1)

[1622] 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."

[1623] In recent years, as demand for interactive content distribution services has increased, AI models are being developed based on information provided by users, and there is a demand for more accurate information provision. However, current systems lack efficient collection and learning of user questions, answers, and evaluation data, resulting in slow AI model development. Furthermore, the user interface is difficult to use, making it difficult to improve user satisfaction.

[1624] 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.

[1625] In this invention, the server includes: means for providing a user interface through which a user can input a question; means for transmitting the input question data to the server; means for storing the received question data in a database; means for providing the question data stored in the database to the generative AI model; means for providing a user interface through which other users can input answers to questions; means for transmitting the input answer data to the server; means for storing the received answer data in the database; means for providing the answer data stored in the database to the generative AI model; means for providing a user interface through which users can view questions and answers; means for providing a user interface through which users can input ratings for answers; means for transmitting the input rating data to the server; means for storing the received rating data in the database; means for providing the questions, answers, and rating data in the database to the generative AI model and performing re-training; means for adjusting the learning content of the generative AI model based on feedback data; means for causing the generative AI model to generate prompt sentences based on the questions, answers, and rating data; and means for providing the generated prompt sentences to the generative AI model as further training data. This enables efficient growth of the AI ​​model and provision of highly accurate information through user interaction.

[1626] A "user interface" is an interface that provides a screen and input means for a user to interact with an application or system.

[1627] "Question data" refers to data including the content of a question that is input by a user through a user interface and sent to a server.

[1628] "Answer data" refers to data including the answer content that is input by other users to a question and sent to the server.

[1629] "Evaluation data" refers to data including evaluations such as "useful" or "not useful" that are entered by users in response to the answers and sent to the server.

[1630] A "generative AI model" is an artificial intelligence model that continuously learns based on questions, answers, and evaluation data provided by users, and provides highly accurate information.

[1631] "Server" refers to a computer system that receives question data, answer data, and evaluation data from users and stores and manages them in a database.

[1632] A "database" is a storage device for accumulating and managing question data, answer data, and evaluation data stored on a server.

[1633] "Feedback data" is data used to adjust the learning content of the generative AI model based on evaluations made by users.

[1634] A "prompt sentence" is training data generated by a generative AI model based on questions, answers, and evaluation data, and is a sentence used for further AI training.

[1635] "Interaction data" is a general term for data related to operations such as questions, answers, and ratings that users perform within the system.

[1636] The present invention provides an interactive system in which users can develop a generative AI model through questions, answers, and ratings. This system has a mechanism for effectively collecting and storing information provided by users and providing it to the generative AI model for learning. Detailed modes for implementing the invention are described below.

[1637] Generating a Program

[1638] First, a user interface is provided that allows users to input questions. This interface runs on devices such as smartphones and provides a screen for users to input their interests and questions.

[1639] The server receives the input question data and stores it in a database. The hardware used here can be a general web server, and the database can be an RDBMS such as PostgreSQL.

[1640] Generative AI models are provided with question data entered by users, and the AI ​​models learn from this data. These AI models are typically implemented using Python libraries (e.g., TensorFlow and PyTorch).

[1641] Next, a user interface is provided that allows other users to enter answers to questions. This is also an interface on a smartphone, and users can enter and submit their answers.

[1642] The received response data is sent to the server and stored in a database. Similarly, this response data is also provided to the generative AI model for further learning.

[1643] Furthermore, a user interface is provided where users can view questions and answers. In this interface, questions and their answers are displayed and users can rate them. The rating data is sent to a server and stored in a database.

[1644] The saved evaluation data is provided to the generative AI model, which retrains the AI ​​model based on the evaluation results. The learning content of the generative AI model is also adjusted based on the feedback data.

[1645] A natural language description of the process

[1646] When a user inputs a question via a device, the device sends the question data to a server. The server stores the received question data in a database and provides the data to a generative AI model. The generative AI model learns from this question data and improves its ability to respond to users' interests and questions.

[1647] When other users enter their answers via their devices, the data is also sent to the server and stored in a database, which is then provided to the generative AI model, allowing it to further learn.

[1648] When a user rates an answer, the rating data is sent to the server and stored in a database. The generative AI model then retrains based on this rating data. Furthermore, by adjusting the AI ​​model's learning content based on feedback data, the accuracy of future answers is improved.

[1649] Below are some examples of prompt sentences.

[1650] Examples of prompt statements

[1651] "A user typed in the question: 'What's the latest anime-related news?'

[1652] User B then enters the following answer:

[1653] It has been announced that "Work X" will be made into a movie.

[1654] Other users found this answer helpful.

[1655] This information is used to grow the generative AI model, enabling it to provide more accurate information to users. In this way, the AI ​​model grows efficiently through user interaction, realizing the provision of more accurate information.

[1656] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1657] Step 1:

[1658] The user inputs a question via the terminal. The user interface has a text input field where the user inputs the question in text format. The input question is converted into JSON format data on the terminal and sent to the server. At this time, the input data includes the user ID, question content, and timestamp.

[1659] Step 2:

[1660] The server stores the received question data in a database. A unique ID is assigned to the question data and stored in the database. At the same time, the server provides this question data to the generative AI model, which then learns from the data. At this time, the server transfers the data to the generative AI model's learning engine, which processes and calculates the data using the question data as input.

[1661] Step 3:

[1662] Other users input answers to questions. The device's user interface provides a text input field for entering answers to questions displayed on the screen. The user enters the answer there, and the answer data is sent to the server in JSON format. The input data includes the user ID, question ID, answer content, and timestamp.

[1663] Step 4:

[1664] The server stores the received answer data in a database. The received data is stored along with a unique ID associated with the answer and associated with the question ID. At the same time, the server provides this answer data to a generative AI model, which then learns from the data.

[1665] Step 5:

[1666] The user browses questions and answers. The device's user interface sends a request to the server to retrieve a list of saved questions and answers. The server retrieves the relevant data from the database and sends it to the device. The device displays the received questions and answers in the user interface. This display includes a mechanism to prioritize new and popular answers.

[1667] Step 6:

[1668] The user enters a rating for the answer. The user interface provides buttons for entering ratings such as "helpful" or "not helpful." When the user presses the rating button, the rating data is sent to the server in JSON format. The rating data includes the user ID, answer ID, rating content, and timestamp.

[1669] Step 7:

[1670] The server stores the received evaluation data in a database. The evaluation data is stored with its associated unique ID and associated with the answer ID. At the same time, the server provides this evaluation data to the generative AI model, which uses the data for retraining. As a result of the retraining, the generative AI model grows to provide more accurate and useful answers.

[1671] Step 8:

[1672] The server provides the questions, answers, and evaluation data in the database to the generative AI model, and periodically retrains it. This allows the generative AI model to constantly learn the latest information. The learning content of the generative AI model is adjusted based on the feedback data. The performance of the AI ​​model improves as a result of data processing and data calculation.

[1673] Step 9:

[1674] The generative AI model generates prompts based on the questions, answers, and evaluation data. The generated prompts are used as further training data for the AI ​​model. The generated prompts are as follows:

[1675] "A user typed in the question: 'What's the latest anime-related news?'

[1676] User B then enters the following answer:

[1677] It has been announced that "Work X" will be made into a movie.

[1678] Other users found this answer helpful.

[1679] Based on this prompt, the generative AI model re-learns and improves.

[1680] 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.

[1681] The present invention combines an emotional engine with an interactive system in which users develop an artificial intelligence model through questions, answers, and ratings. This system recognizes the user's emotional state and can adjust the display order of questions and answers and the generation of feedback based on that state. The specific form and processing flow of this system are described below.

[1682] Enter and save your question

[1683] Enter your question

[1684] Users can input their interests and doubts as questions through the user interface of their terminal.

[1685] Example: User A types, "What's the latest anime-related news?"

[1686] Submit a Question

[1687] The terminal transmits the question data entered by the user to the server.

[1688] This data is sent in a format such as JSON and includes metadata such as the user ID, question, and timestamp.

[1689] Saving Question Data

[1690] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified.

[1691] Providing questions to the AI ​​model

[1692] The server provides the stored question data to the AI ​​model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[1693] Sentiment analysis of question data

[1694] The server provides the question data to the emotion engine to analyze the user's emotions.

[1695] Example: In response to User A's question, "What's the latest anime-related news?", the emotion engine recognizes emotions such as "interesting" and "expecting."

[1696] Enter and save your answers

[1697] Enter your answer

[1698] A user inputs an answer to a question posted by another user through a user interface.

[1699] Example: User B replies, "It has been announced that 'Work X' will be made into a movie."

[1700] Submit your answer

[1701] The device sends the answer data entered by the user to the server. The answer data is also sent in JSON format and includes the user ID, question ID, answer content, timestamp, etc.

[1702] Saving response data

[1703] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[1704] Providing answers to AI models

[1705] The server provides the stored response data to the AI ​​model, which then learns from it, allowing the AI ​​to accumulate more knowledge and improve accuracy.

[1706] Sentiment analysis of response data

[1707] The server provides the response data to an emotion engine to analyze the user's emotions.

[1708] Example: In response to User B's response, "It has been announced that 'Work X' will be made into a movie," the emotion engine recognizes emotions such as "joy" and "satisfaction."

[1709] View questions and answers

[1710] Ask questions and get answers

[1711] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[1712] View questions and answers

[1713] The device displays the received data on the user interface. It can prioritize new or popular answers and adjust the display order based on the user's emotions as recognized by the emotion engine.

[1714] For example, if the emotion engine indicates a high level of interest in response to the question "What is the latest anime-related news?", the question and related answers will be displayed preferentially.

[1715] Enter and save your rating

[1716] Enter your rating

[1717] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[1718] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[1719] Submit your rating

[1720] The terminal sends the entered evaluation data to the server. The evaluation data includes the user ID, answer ID, evaluation content, timestamp, etc.

[1721] Saving evaluation data

[1722] The server stores the received evaluation data in a database, and this evaluation data is also assigned a unique ID.

[1723] Sentiment analysis of rating data

[1724] The server provides the rating data to the emotion engine, which analyzes the user's emotions regarding the ratings.

[1725] Example: Recognizing emotions such as "satisfied" or "dissatisfied" from the content of the evaluation.

[1726] AI learning and accuracy improvement

[1727] Periodic Relearning

[1728] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining, which improves the accuracy of the AI ​​model.

[1729] Reflecting feedback

[1730] The server automatically adjusts the AI ​​model's learning based on user ratings and sentiment data, correcting incorrect answers and reinforcing correct answers.

[1731] Specific examples

[1732] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. The emotion engine also recognizes this question as "interest." Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model, with the emotion engine recognizing it as "joy." If user C views this question and answer and rates it as "helpful," the evaluation data is also saved, and the emotion engine recognizes it as "satisfaction." In this way, the system promotes the growth of AI that also reflects the user's emotions.

[1733] The processing flow will be explained below.

[1734] Post a Question

[1735] Step 1:

[1736] User enters a question

[1737] Users enter their questions using a web browser or mobile app user interface.

[1738] Example: User A types, "What's the latest anime-related news?"

[1739] Step 2:

[1740] Submit your question

[1741] The terminal transmits the question entered by the user to the server.

[1742] The transmitted data includes metadata such as the user ID, question content, and timestamp.

[1743] Example: {"User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}

[1744] Step 3:

[1745] Save questions

[1746] The server stores the received question data in a database.

[1747] Give the question a unique ID (e.g. Q789) and store it in the database.

[1748] Example: Save in the database as {"Question ID": "Q789", "User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}.

[1749] Step 4:

[1750] Providing questions to the AI ​​model

[1751] The server provides the saved question data to the AI ​​model and uses it as learning data.

[1752] Example: Providing {"question": "What's the latest anime news?"} to an AI model.

[1753] Step 5:

[1754] Sentiment analysis of question data

[1755] The server sends the received question data to the emotion engine to analyze the user's emotions.

[1756] Example: The emotion engine recognizes "interest" for the question "What's the latest anime-related news?"

[1757] Providing answers

[1758] Step 6:

[1759] User enters answer

[1760] A user inputs an answer to a question from another user.

[1761] Example: User B types, "It has been announced that 'Work X' will be made into a movie."

[1762] Step 7:

[1763] Sending answers

[1764] The terminal transmits the answer entered by the user to the server.

[1765] The transmitted data includes the user ID, question ID, answer content, timestamp, etc.

[1766] Example: {"User ID": "B456", "Question ID": "Q789", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}

[1767] Step 8:

[1768] Save answers

[1769] The server stores the received response data in a database.

[1770] The answer is given a unique ID (e.g. A987) and stored in association with the question ID.

[1771] Example: Save in the database as {"Answer ID": "A987", "Question ID": "Q789", "User ID": "B456", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}.

[1772] Step 9:

[1773] Providing answers to the AI ​​model

[1774] The server provides the saved response data to the AI ​​model and uses it as learning data.

[1775] Example: Providing {"answer": "It has been announced that 'Book X' will be made into a movie."} to an AI model.

[1776] Step 10:

[1777] Sentiment analysis of response data

[1778] The server sends the received response data to an emotion engine to analyze the user's emotions.

[1779] Example: The emotion engine recognizes "joy" in response to the answer "It has been announced that 'Work X' will be made into a movie."

[1780] View questions and answers

[1781] Step 11:

[1782] Ask questions and get answers

[1783] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[1784] Example: Request {"Question ID": "Q789"} and the server will send the related questions and answers.

[1785] Step 12:

[1786] View questions and answers

[1787] The terminal displays the received question and answer data on a user interface.

[1788] Example: When a user accesses the site, the question "What's the latest anime-related news?" and the answer "It has been announced that 'Work X' will be made into a movie" are displayed on the screen.

[1789] Step 13:

[1790] Adjusting display order based on emotions

[1791] The server adjusts the display order of questions and answers based on the emotion data recognized by the emotion engine.

[1792] For example, questions that you have shown interest in will be displayed before other questions.

[1793] Evaluating answers

[1794] Step 14:

[1795] User-initiated rating input

[1796] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[1797] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[1798] Step 15:

[1799] Submit your evaluation

[1800] The terminal transmits the input evaluation data to the server.

[1801] The transmitted data includes the user ID, answer ID, evaluation details, timestamp, etc.

[1802] Example: {"User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}

[1803] Step 16:

[1804] Save evaluation details

[1805] The server stores the received evaluation data in a database.

[1806] Evaluation data will also be assigned a unique ID (e.g., E123).

[1807] Example: Store in database as {"Rating ID": "E123", "User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}.

[1808] Step 17:

[1809] Sentiment analysis of rating data

[1810] The server transmits the received evaluation data to an emotion engine to analyze the user's emotions.

[1811] For example, the sentiment engine recognizes the rating "helpful" as "satisfied."

[1812] AI learning and accuracy improvement

[1813] Step 18:

[1814] Periodic retraining of AI models

[1815] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[1816] Example: Retraining an AI model using all question and answer and rating data from the past month.

[1817] Step 19:

[1818] Reflecting feedback

[1819] The server automatically adjusts the learning content of the AI ​​model based on user evaluation data and emotion data from the emotion engine.

[1820] Example: Learning from an answer that was rated "not helpful" by many users can be corrected to provide more accurate answers.

[1821] Through the above processing steps, the present invention realizes an interactive system that allows users to actively participate in the growth of AI, and provides advanced information by combining an emotion engine.

[1822] Example 2

[1823] 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."

[1824] Modern information systems are required to not only provide information based on user input, but also to understand the user's emotions and intentions and provide feedback accordingly. However, conventional systems have had difficulty in realizing advanced interactions such as analyzing the user's emotions and adjusting the order in which information is displayed based on those emotions.

[1825] 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.

[1826] In this invention, the server includes means for providing a user interface through which a user can input a question, means for transmitting input question data to the server, means for saving received question data in a database, means for providing the question data saved in the database to the artificial intelligence model, means for providing the question data to an emotion engine and analyzing the user's emotions, means for providing a user interface through which other users can input answers to questions, means for transmitting input answer data to the server, means for saving received answer data in a database, means for providing the answer data saved in the database to the artificial intelligence model, means for providing the answer data to the emotion engine and analyzing the user's emotions, means for providing a user interface through which users can view questions and answers, means for adjusting the display order of questions and answers based on the results of the user emotion analysis, means for providing a user interface through which users can input ratings for answers, means for transmitting input rating data to the server, means for saving received rating data in a database, means for providing the rating data to the emotion engine and analyzing the user's emotions, means for providing the questions, answers, and rating data in the database to the artificial intelligence model and performing relearning, and means for adjusting the learning content of the artificial intelligence model based on the feedback data. This makes it possible to provide information that reflects the user's emotions and to create an interactive system that adjusts the display based on that information.

[1827] "User interface" refers to the screens and operating means that allow users to interact with the system to exchange information.

[1828] "Question data" is information including the question entered by the user, and includes metadata such as the user ID, question content, and timestamp.

[1829] A "server" refers to a computer system that receives requests over a network and is responsible for processing and storing data.

[1830] A "database" refers to a data management system that allows a system to centrally store data and make it searchable and retrievalable.

[1831] An "artificial intelligence model" refers to an algorithm that learns patterns and rules from large amounts of data and makes inferences and predictions based on input.

[1832] An "emotion engine" is an algorithm or system that analyzes user input data and identifies the emotion or intent behind it.

[1833] "Answer data" is information including the answer entered by the user to the question, and includes the user ID, question ID, answer, timestamp, and the like.

[1834] "Evaluation data" is information including the content of the user's evaluation of an answer, and includes the user ID, answer ID, evaluation content, timestamp, and the like.

[1835] "Retraining" refers to the learning process of using new data in a database to improve the accuracy of an artificial intelligence model.

[1836] "Feedback data" refers to information used to adjust system processing based on user evaluations and the results of sentiment analysis.

[1837] The present invention is an interactive system that combines an emotion engine and aims to develop an artificial intelligence model through user questions, answers, and ratings. This system recognizes the user's emotional state and can adjust the display order of questions and answers and the generation of feedback based on that state. The specific form and processing flow of the system are described below.

[1838] Enter and save your question

[1839] Users input their interests and concerns as questions through the user interface of their device. This user interface provides a text input area, allowing users to enter the content of their questions as text. For example, User A enters "What's the latest anime-related news?" This question data is sent from the device to the server in a format such as JSON, and includes metadata such as the user ID, question content, and timestamp.

[1840] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified. The server then provides the stored question data to an artificial intelligence model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[1841] The question data is also provided to the emotion engine, which analyzes the user's emotions. For example, in response to User A's question, "What's the latest anime-related news?", the emotion engine recognizes emotions such as "interesting" and "expecting." The results are saved as metadata.

[1842] Enter and save your answers

[1843] Users input answers to questions posted by other users through the user interface. For example, User B answers, "It has been announced that 'Work X' will be made into a movie." This answer data is sent from the device to the server in JSON format or similar, and includes the user ID, question ID, answer content, and timestamp.

[1844] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID. The server then provides the stored answer data to an artificial intelligence model, which then learns from the data. This allows the AI ​​to accumulate more knowledge and improve accuracy.

[1845] The response data is also provided to the emotion engine, which analyzes the user's emotions. For example, in response to User B's response, "It has been announced that 'Work X' will be made into a movie," the emotion engine recognizes emotions such as "joy" and "satisfaction." These results are also saved as metadata.

[1846] View questions and answers

[1847] The device requests a list of questions and their answers from the server, and receives the data from the server. This request includes the user ID and the specific question ID, and the server searches for the corresponding data and returns it to the device.

[1848] The device displays the received data on the user interface. It is possible to prioritize displaying newer or more popular answers. The display order is also adjusted based on the user's emotions as recognized by the emotion engine. For example, if the emotion engine indicates a high level of "interest" in response to the question "What's the latest anime-related news?", that question and related answers will be displayed with priority.

[1849] Enter and save your rating

[1850] The user inputs a rating such as "helpful" or "not helpful" for the displayed answer. For example, a user may rate the answer "It has been announced that 'Work X' will be made into a movie." as "helpful." This rating data is sent from the device to the server and includes the user ID, answer ID, rating content, and timestamp.

[1851] The server stores the received evaluation data in a database. This evaluation data is also assigned a unique ID. The evaluation data is also provided to the emotion engine, which analyzes the user's emotions. For example, emotions such as "satisfied" or "dissatisfied" are recognized from the content of the evaluation. This result is also stored as metadata.

[1852] AI learning and accuracy improvement

[1853] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for re-learning. For example, once a week, the server collects all new data and re-trains the AI ​​model. This improves the accuracy of the AI ​​model.

[1854] Furthermore, the server automatically adjusts the learning content of the AI ​​model based on user evaluation and emotion data. If there are many incorrect answers, the data is corrected and correct answers are strengthened, improving the model's performance.

[1855] Specific examples

[1856] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. The emotion engine also recognizes this question as "interest." Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model, with the emotion engine recognizing it as "joy." If user C views this question and answer and rates it as "helpful," the evaluation data is also saved, and the emotion engine recognizes it as "satisfaction." In this way, the system promotes the growth of AI that also reflects user emotions.

[1857] Prompt Sentence Examples

[1858] "What's the latest anime-related news?"

[1859] "It has been announced that 'Work X' will be made into a movie."

[1860] "It was helpful"

[1861] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1862] Step 1:

[1863] Enter your question

[1864] The user enters a question through the user interface of their terminal.

[1865] Specific actions: Enter a question in the text input area and click the submit button.

[1866] Input: The question the user types into the text entry area.

[1867] Output: JSON formatted question data sent by the device.

[1868] Step 2:

[1869] Submit a Question

[1870] The terminal transmits the question data entered by the user to the server.

[1871] Specific operation: The device sends question data including metadata such as the user ID, question content, and timestamp to the server.

[1872] Input: The question data entered by the user.

[1873] Output: The question data sent to the server.

[1874] Step 3:

[1875] Saving Question Data

[1876] The server stores the received question data in a database, and each question is assigned a unique ID.

[1877] Specific operation: The server assigns a unique ID to the question data and saves it in the database.

[1878] Input: Question data sent from the terminal.

[1879] Output: Question data stored in a database.

[1880] Step 4:

[1881] Providing questions to the AI ​​model

[1882] The server provides the stored question data to the AI ​​model, which then learns from the data.

[1883] Specific operation: The server reads question data from the database and sends it to the AI ​​model for learning.

[1884] Input: Question data stored in the database.

[1885] Output: The input data for the AI ​​model to learn from.

[1886] Step 5:

[1887] Sentiment analysis of question data

[1888] The server provides the question data to the emotion engine to analyze the user's emotions.

[1889] Specific operation: The server sends the question data to the emotion engine, which analyzes the emotion.

[1890] Input: Question data stored in the database.

[1891] Output: Sentiment data parsed by the sentiment engine.

[1892] Step 6:

[1893] Enter your answer

[1894] A user inputs an answer to a question posted by another user through a user interface.

[1895] Specific actions: Enter your answer in the text input area and click the submit button.

[1896] Input: The answer the user entered into the text entry area.

[1897] Output: JSON formatted response data sent by the device.

[1898] Step 7:

[1899] Submit your answer

[1900] The terminal transmits the answer data entered by the user to the server.

[1901] Specific operation: The device sends the answer data, including metadata such as the user ID, question ID, answer content, and timestamp, to the server.

[1902] Input: The answer data entered by the user.

[1903] Output: The response data sent to the server.

[1904] Step 8:

[1905] Saving response data

[1906] The server stores the received response data in a database, and each response is assigned a unique ID.

[1907] Specific operation: The server assigns a unique ID to the response data and saves it in the database.

[1908] Input: Response data sent from the device.

[1909] Output: Response data stored in a database.

[1910] Step 9:

[1911] Providing answers to AI models

[1912] The server provides the saved response data to the AI ​​model, which then learns from the data.

[1913] Specific operation: The server reads the response data from the database and sends it to the AI ​​model for learning.

[1914] Input: Response data stored in the database.

[1915] Output: The input data for the AI ​​model to learn from.

[1916] Step 10:

[1917] Sentiment analysis of response data

[1918] The server provides the response data to an emotion engine to analyze the user's emotions.

[1919] Specific operation: The server sends the response data to the emotion engine, which analyzes the emotion.

[1920] Input: Response data stored in the database.

[1921] Output: Sentiment data parsed by the sentiment engine.

[1922] Step 11:

[1923] Ask questions and get answers

[1924] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[1925] Specific operation: The device sends a request to the server, and the server returns the relevant question and answer data.

[1926] Input: The request from the device (including the user ID and the specific question ID).

[1927] Output: Question and answer data received from the server.

[1928] Step 12:

[1929] View questions and answers

[1930] The terminal displays the received data on a user interface.

[1931] Specific operation: The device converts the data into a display format and displays it on the user interface. The display order is adjusted based on the user's emotions recognized by the emotion engine.

[1932] Input: Question and answer data received from the server.

[1933] Output: The question and answer data that is displayed in the user interface.

[1934] Step 13:

[1935] Enter your rating

[1936] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[1937] Specific actions: Click the rating button, select the rating content, and click the submit button.

[1938] Input: The rating selected by the user.

[1939] Output: JSON formatted rating data sent by the device.

[1940] Step 14:

[1941] Submit your rating

[1942] The terminal transmits the input evaluation data to the server.

[1943] Specific operation: The device sends evaluation data including metadata such as user ID, answer ID, evaluation content, and timestamp to the server.

[1944] Input: The rating data entered by the user.

[1945] Output: The rating data sent to the server.

[1946] Step 15:

[1947] Saving evaluation data

[1948] The server stores the received evaluation data in a database. The evaluation data is also assigned a unique ID.

[1949] Specific operation: The server assigns a unique ID to the evaluation data and saves it in the database.

[1950] Input: Rating data sent from the device.

[1951] Output: Evaluation data stored in a database.

[1952] Step 16:

[1953] Sentiment analysis of rating data

[1954] The server provides the rating data to the emotion engine to analyze the user's emotions.

[1955] Specific operation: The server sends the evaluation data to the emotion engine, which analyzes the emotions.

[1956] Input: Evaluation data stored in a database.

[1957] Output: Sentiment data parsed by the sentiment engine.

[1958] Step 17:

[1959] Retraining the AI

[1960] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[1961] What it does: The server reads all new data from the database and periodically retrains the AI ​​model.

[1962] Input: Question, answer, and assessment data stored in a database.

[1963] Output: The retrained AI model.

[1964] Step 18:

[1965] Reflecting feedback

[1966] The server automatically adjusts the learning content of the AI ​​model based on user evaluations and emotional data.

[1967] Specific operation: The server adjusts the learning weights of the AI ​​model based on the evaluation and emotion data and corrects incorrect answers.

[1968] Input: Rating data, emotion data.

[1969] Output: The tuned AI model.

[1970] (Application example 2)

[1971] 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."

[1972] Conventional customer support systems have difficulty in responding to users' emotional states, making it difficult to improve user experience and service quality. In addition, there has been a lack of methods to provide artificial intelligence models that can learn interactively by reflecting user feedback and emotions.

[1973] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for providing a user interface through which a user can input a question; means for transmitting the input question data to the server; means for saving the received question data in a database; means for providing the question data saved in the database to the AI ​​model; means for providing a user interface through which other users can input answers to questions; means for transmitting the input answer data to the server; means for saving the received answer data in the database; means for providing the answer data saved in the database to the AI ​​model; means for providing a user interface through which users can view questions and answers; means for providing a user interface through which users can input ratings for answers; means for transmitting the input rating data to the server; means for saving the received rating data in the database; means for providing the questions, answers, and rating data in the database to the AI ​​model and performing re-learning; means for adjusting the learning content of the AI ​​model based on feedback data; means for analyzing the emotional state of the input question and answer and means for adjusting the display order and feedback based on the emotional state. This makes it possible to realize an interactive customer support system that reflects the user's emotions.

[1974] A "user interface" is an interface through which a user interacts directly with a system and inputs questions and answers.

[1975] A "server" is a computer that receives question data and answer data sent by users and stores them in a database.

[1976] A "database" is a storage system that centrally manages and stores question data, response data, and evaluation data.

[1977] An "artificial intelligence model" is an algorithm that learns from questions and answers entered by users and improves the quality of subsequent questions and answers.

[1978] An "emotional state" is a psychological state recognized as a result of the emotion engine's analysis of the questions and answers entered by the user.

[1979] "Display order" refers to the order in which questions and answers are displayed when a user browses.

[1980] "Feedback" refers to the evaluations and comments that users make on answers, and is data used to train the system.

[1981] "Relearning" is a process in which an artificial intelligence model periodically learns based on data accumulated in a database to improve its accuracy.

[1982] "Question data" is data including the content of questions posed to the system by users.

[1983] "Answer data" is data including answers provided by other users to questions.

[1984] "Evaluation data" is data including the evaluation content given by the user to the answer.

[1985] The "emotion engine" is a system component that analyzes the emotional state of input questions and answers.

[1986] The present invention combines an emotion engine with an interactive system in which users can develop an artificial intelligence model through questions, answers, and ratings. Specific embodiments of this system are described below.

[1987] Enter and save your question

[1988] User Interface:

[1989] It provides a user interface for users to input questions. Users can input their interests and questions and send the question data to the server via their terminal. This data includes metadata such as the user ID, question content, and timestamp.

[1990] server:

[1991] The server stores the received question data in a database and assigns a unique ID to uniquely identify it.The server then provides the stored question data to an artificial intelligence model for learning.Furthermore, the server passes the question data to an emotion engine to analyze the user's emotional state.

[1992] Enter and save your answers

[1993] User Interface:

[1994] It provides a user interface that allows other users to enter answers to questions. The entered answer data includes information such as the answer content, user ID, question ID, and timestamp.

[1995] server:

[1996] The server receives the response data and stores it in a database. Here, a unique ID is assigned to each response, making it uniquely identified. The stored response data is then provided to an artificial intelligence model, which further trains the AI. The response data is also passed to an emotion engine, which analyzes the user's emotional state.

[1997] View questions and answers

[1998] User Interface:

[1999] It provides an interface where users can browse questions and their answers. The interface is adjusted to prioritize new and popular answers. The display order is also dynamically adjusted based on the user's emotions as recognized by the emotion engine.

[2000] Enter and save your rating

[2001] User Interface:

[2002] Users can rate the displayed answers. The rating data includes the user ID, answer ID, rating content, timestamp, etc.

[2003] server:

[2004] The evaluation data is received and stored in a database. The evaluation data is analyzed by an emotion engine to recognize the user's emotional state. The acquired evaluation data is used to retrain the AI ​​model.

[2005] AI model training and accuracy improvement

[2006] server:

[2007] The server provides the AI ​​model with the questions, answers, and evaluation data from the database and periodically retrains it, improving its accuracy. It also automatically adjusts its learning based on user feedback and sentiment data.

[2008] Hardware and software used

[2009] Hardware: Smartphones, head-mounted displays, servers

[2010] Software: Emotion engine library, AI model library, database connection library

[2011] Specific example explanation

[2012] For example, User A opens the app to ask a question about a product in a store and types the question, "Does this product run on batteries?" This question is sent to the server and stored in the database. The AI ​​model learns this question, and the emotion engine recognizes it as "curiosity." As a result, the store staff is notified, "We have received your question. Please check the details." An example of this prompt is as follows:

[2013] Prompt Sentence Examples

[2014] User Question:

[2015] Question: 'Does this item run on batteries?'

[2016] User ID: 'user123'

[2017] Timestamp: '2023-10-10T10:00:00Z'

[2018] Emotion Engine Analysis Results:

[2019] Emotional state: 'Curiosity'

[2020] Action: 'Increase display priority'

[2021] In this way, the present invention provides an interactive customer support system that reflects the user's feelings.

[2022] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2023] Step 1:

[2024] The user inputs a question using a smartphone or head-mounted display. The question data input from the user interface includes the user ID, question content, and timestamp. For example, the user might input, "Does this product run on batteries?"

[2025] Step 2:

[2026] The device sends the entered question data in JSON format to the server, which includes the user ID, question content, and timestamp.

[2027] Step 3:

[2028] The server stores the received question data in a database. A unique ID is assigned when the data is saved to uniquely identify it. The input data includes the user ID, question content, timestamp, and unique ID.

[2029] Step 4:

[2030] The server provides the stored question data to the AI ​​model for learning. Based on the provided data, the AI ​​model analyzes the question content and improves the accuracy of answers to future questions.

[2031] Step 5:

[2032] The server passes the question data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the question and outputs an emotion such as "curiosity."

[2033] Step 6:

[2034] Another user inputs an answer to the question using a smartphone or head-mounted display. The answer data input from the user interface includes the answer content, user ID, question ID, and timestamp. For example, the user might input, "Yes, this product runs on batteries."

[2035] Step 7:

[2036] The device sends the entered answer data in JSON format to the server. The sent data includes the user ID, question ID, answer content, and timestamp.

[2037] Step 8:

[2038] The server stores the received answer data in a database. A unique ID is assigned when saving the data to uniquely identify it. The input data includes the user ID, question ID, answer content, timestamp, and unique ID.

[2039] Step 9:

[2040] The server provides the stored response data to the AI ​​model for learning. Based on the provided data, the AI ​​model analyzes the content of the responses and improves the accuracy of future responses.

[2041] Step 10:

[2042] The server passes the answer data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the answer and outputs an emotion such as "a sense of relief."

[2043] Step 11:

[2044] The user interface displays questions and answers to the user. The display data includes the question content and multiple answers. The display order is dynamically adjusted based on the analysis results of the emotion engine so that newer and more popular answers are displayed preferentially.

[2045] Step 12:

[2046] Users rate the displayed answers. The rating data includes the user ID, answer ID, rating content, and timestamp. For example, a user may rate the answer as "helpful."

[2047] Step 13:

[2048] The device sends the entered evaluation data in JSON format to the server. The sent data includes the user ID, answer ID, evaluation content, and timestamp.

[2049] Step 14:

[2050] The server stores the received evaluation data in a database. A unique ID is assigned when the data is saved to uniquely identify it. The input data includes the user ID, answer ID, evaluation content, timestamp, and unique ID.

[2051] Step 15:

[2052] The server passes the evaluation data to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the evaluation content and outputs an emotion such as "satisfaction."

[2053] Step 16:

[2054] The server provides the AI ​​model with the questions, answers, and evaluation data stored in the database and periodically retrains it, improving its accuracy. Furthermore, by automatically adjusting the learning content based on emotion data, the AI ​​model can provide more appropriate responses to users.

[2055] 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.

[2056] 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.

[2057] 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.

[2058] [Fourth embodiment]

[2059] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[2060] 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.

[2061] 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).

[2062] 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.

[2063] 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.

[2064] 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).

[2065] 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.

[2066] 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.

[2067] 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.

[2068] 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.

[2069] 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.

[2070] 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.

[2071] 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."

[2072] This invention is an interactive system that allows users to develop an artificial intelligence model through questions, answers, and ratings. Its details include a process for effectively collecting and storing information entered by users and providing it to the AI ​​model for learning. The specific form and processing flow of the system are described below.

[2073] Enter and save your question

[2074] Enter your question

[2075] Users can input their interests and doubts as questions through the user interface of their terminal.

[2076] Example: User A types, "What's the latest anime-related news?"

[2077] Submit a Question

[2078] The terminal transmits the question data entered by the user to the server.

[2079] This data is sent in a format such as JSON and includes metadata such as the user ID, question, and timestamp.

[2080] Saving Question Data

[2081] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified.

[2082] Providing questions to the AI ​​model

[2083] The server provides the stored question data to the AI ​​model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[2084] Enter and save your answers

[2085] Enter your answer

[2086] A user inputs an answer to a question posted by another user through a user interface.

[2087] Example: User B replies, "It has been announced that 'Work X' will be made into a movie."

[2088] Submit your answer

[2089] The device sends the answer data entered by the user to the server. The answer data is also sent in JSON format and includes the user ID, question ID, answer content, timestamp, etc.

[2090] Saving response data

[2091] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[2092] Providing answers to AI models

[2093] The server provides the stored response data to the AI ​​model, which then learns from it, allowing the AI ​​to accumulate more knowledge and improve accuracy.

[2094] View questions and answers

[2095] Ask questions and get answers

[2096] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[2097] View questions and answers

[2098] The device displays the received data on the user interface. It is also possible to display newer or more popular answers preferentially.

[2099] Enter and save your rating

[2100] Enter your rating

[2101] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[2102] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[2103] Submit your rating

[2104] The terminal sends the entered evaluation data to the server. The evaluation data includes the user ID, answer ID, evaluation content, timestamp, etc.

[2105] Saving evaluation data

[2106] The server stores the received evaluation data in a database, and this evaluation data is also assigned a unique ID.

[2107] AI learning and accuracy improvement

[2108] Periodic Relearning

[2109] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for re-training, thereby improving the accuracy of the AI ​​model.

[2110] Reflecting feedback

[2111] The server automatically adjusts the AI ​​model's learning based on user evaluation data, thereby correcting incorrect answers and strengthening correct answers.

[2112] Specific examples

[2113] For example, if user A posts a question such as "What's the latest anime-related news?", the server saves this question in a database and provides it to the AI ​​model for learning. Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also saved in the database and provided to the AI ​​model. If user C views this question and answer and rates it as "helpful," this rating data is also saved and used to retrain the AI ​​model.

[2114] In this way, users can actively participate in the growth of the AI, and the system provides highly accurate information that reflects their own knowledge.

[2115] The processing flow will be explained below.

[2116] Post a Question

[2117] Step 1:

[2118] User enters a question

[2119] Users enter their questions using a web browser or mobile app user interface.

[2120] Example: User A types, "What's the latest anime-related news?"

[2121] Step 2:

[2122] Submit your question

[2123] The terminal transmits the question entered by the user to the server.

[2124] The transmitted data includes metadata such as the user ID, question content, and timestamp.

[2125] Example: {"User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}

[2126] Step 3:

[2127] Save questions

[2128] The server stores the received question data in a database.

[2129] Give the question a unique ID (e.g. Q789) and store it in the database.

[2130] Example: Save in the database as {"Question ID": "Q789", "User ID": "A123", "Question": "What's the latest anime-related news?", "Timestamp": "2023-10-10T15:00:00Z"}.

[2131] Step 4:

[2132] Providing questions to the AI ​​model

[2133] The server provides the saved question data to the AI ​​model and uses it as learning data.

[2134] Example: Providing {"question": "What's the latest anime news?"} to an AI model.

[2135] Providing answers

[2136] Step 5:

[2137] User enters answer

[2138] A user inputs an answer to a question from another user.

[2139] Example: User B types, "It has been announced that 'Work X' will be made into a movie."

[2140] Step 6:

[2141] Sending answers

[2142] The terminal transmits the answer entered by the user to the server.

[2143] The transmitted data includes the user ID, question ID, answer content, timestamp, etc.

[2144] Example: {"User ID": "B456", "Question ID": "Q789", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}

[2145] Step 7:

[2146] Save answers

[2147] The server stores the received response data in a database.

[2148] Give the answer a unique ID (e.g. A987) and store it along with the question ID.

[2149] Example: Save in the database as {"Answer ID": "A987", "Question ID": "Q789", "User ID": "B456", "Answer": "It has been announced that 'Work X' will be made into a movie.", "Timestamp": "2023-10-10T16:00:00Z"}.

[2150] Step 8:

[2151] Providing answers to the AI ​​model

[2152] The server provides the saved response data to the AI ​​model and uses it as learning data.

[2153] Example: Providing {"answer": "It has been announced that 'Book X' will be made into a movie."} to an AI model.

[2154] View questions and answers

[2155] Step 9:

[2156] Ask questions and get answers

[2157] The terminal requests a list of questions and their answers from the server and receives the data from the server.

[2158] Example: Request data for {"Question ID": "Q789"} and receive question and answer data from the server.

[2159] Step 10:

[2160] View questions and answers

[2161] The terminal displays the received question and answer data on a user interface.

[2162] Example: When a user accesses the site, the question "What's the latest anime-related news?" and the answer "It has been announced that 'Work X' will be made into a movie" are displayed on the screen.

[2163] Evaluating answers

[2164] Step 11:

[2165] User-initiated rating input

[2166] The user inputs an evaluation such as "helpful" or "not helpful" for the displayed answer.

[2167] Example: A user rates the answer "It has been announced that 'Work X' will be made into a movie." as "helpful."

[2168] Step 12:

[2169] Submit your evaluation

[2170] The terminal transmits the input evaluation data to the server.

[2171] The transmitted data includes the user ID, answer ID, evaluation details, timestamp, etc.

[2172] Example: {"User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}

[2173] Step 13:

[2174] Save evaluation details

[2175] The server stores the received evaluation data in a database.

[2176] Evaluation data will also be assigned a unique ID (e.g., E123).

[2177] Example: Store in database as {"Rating ID": "E123", "User ID": "C789", "Answer ID": "A987", "Rating": "Helpful", "Timestamp": "2023-10-10T17:00:00Z"}.

[2178] AI learning and accuracy improvement

[2179] Step 14:

[2180] Periodic retraining of AI models

[2181] The server periodically provides the questions, answers, and evaluation data in the database to the AI ​​model for retraining.

[2182] Example: Retraining an AI model using all question and answer and rating data from the past month.

[2183] Step 15:

[2184] Reflecting feedback

[2185] The server automatically adjusts the learning content of the AI ​​model based on user evaluation data.

[2186] Example: Revisiting learning based on answers that were rated "not helpful" by many users.

[2187] Example 1

[2188] 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."

[2189] Conventional AI model learning systems have faced issues with inefficient management of user-entered questions, answers, and evaluation data, resulting in delayed improvements in the accuracy of AI models. Furthermore, there was a lack of a mechanism for immediately reflecting user feedback and using it to improve the model, making it difficult to improve the user experience. Furthermore, when displaying questions and answers, there was a lack of a function to prioritize the latest information and popular answers that users wanted, reducing the efficiency of information acquisition.

[2190] 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.

[2191] In this invention, the server includes a means for providing a user interface, a means for transmitting question data, answer data, and evaluation data to the server, a means for storing the received data in a database, a means for providing the data to the AI ​​model as learning data, a means for acquiring the question and answer data and displaying them on the user interface, and a means for preferentially displaying new or popular answers. This enables efficient collection, management, and provision of various data entered by users, resulting in rapid improvement in the accuracy of the AI ​​model and improved functionality of the user interface.

[2192] "User interface" refers to the screen or operating means through which a user can access a system and input and view information such as questions and answers.

[2193] "Question data" refers to data that includes information about the question entered by the user, including metadata such as the user ID, question content, and timestamp.

[2194] "Server" refers to a central computing device that receives, processes, stores, and provides data sent by users to AI models.

[2195] "Database" refers to a system in which a server systematically stores question data, answer data, evaluation data, etc. and manages them so that they can be retrieved as needed.

[2196] "Artificial intelligence model" refers to a machine learning algorithm or neural network that learns from user-provided data and answers questions, provides information, or generates responses.

[2197] "Answer data" refers to data that includes answers to questions posted by other users, and includes metadata such as user ID, question ID, answer content, and timestamp.

[2198] "Rating data" refers to data that includes information about the ratings that users have given to questions and answers, and includes metadata such as user ID, rating content, answer ID, and timestamp.

[2199] "Retraining" refers to the learning process in which questions, answers, and evaluation data stored in a database are periodically provided to an artificial intelligence model to improve the model's accuracy.

[2200] "Feedback data" refers to data used to adjust the model's learning content based on user evaluation data and improve the model's output results.

[2201] MODE FOR CARRYING OUT THE INVENTION

[2202] This invention is an interactive system in which users can develop an artificial intelligence model through questions, answers, and evaluations. The specific form and processing flow of this system will be described in detail below.

[2203] Enter and save your question

[2204] When a user types a question

[2205] A user accesses a dedicated user interface using their own terminal (for example, a personal computer or a smartphone) and inputs a question. An example of a question is "What is the latest anime-related news?"

[2206] Sending and storing question data

[2207] The device sends the question entered by the user to the server in JSON format, including the user ID, question content, timestamp, etc.

[2208] The server stores the received question data in a database (e.g., MySQL or PostgreSQL). A unique ID is assigned to each question, allowing it to be uniquely identified.

[2209] Providing question data to the AI ​​model

[2210] The server can provide the stored question data to the AI ​​model, which can then use the data as training data. This process allows the AI ​​model to learn new information from user questions and improve the accuracy of its responses to future questions.

[2211] Enter and save your answers

[2212] When the user enters an answer

[2213] Users can enter answers to questions posted by other users. An example of an answer might be, "It has been announced that 'Work X' will be made into a movie."

[2214] Sending and storing response data

[2215] The device sends the answers entered by the user to the server in JSON format, including the user ID, question ID, answer content, and timestamp.

[2216] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID.

[2217] Providing response data to the AI ​​model

[2218] The server can provide the saved response data to the AI ​​model, which can then use the data as training data. This process allows the AI ​​model to learn new information from the user's responses and further improve the accuracy of future responses.

[2219] View questions and answers

[2220] Retrieving and Displaying Data

[2221] The device requests a list of questions and their answers from the server and receives the necessary data. The received data is displayed on the user interface. It is also possible to display new or popular answers preferentially, allowing users to quickly obtain the latest information and useful information.

[2222] Enter and save your rating

[2223] When a user enters a rating

[2224] Users can enter a rating for the displayed answer, such as "helpful" or "not helpful."

[2225] Sending and storing evaluation data

[2226] The device sends the entered evaluation data in JSON format to the server, which includes the user ID, answer ID, evaluation content, and timestamp.

[2227] The server stores the received rating data in a database, and each rating is assigned a unique ID.

[2228] AI learning and accuracy improvement

[2229] Periodic Relearning

[2230] The server periodically provides the question data, answer data, and evaluation data in the database to the AI ​​model for re-learning. This process improves the accuracy of the AI ​​model, enabling it to provide more accurate information to users.

[2231] Reflecting feedback

[2232] The server automatically adjusts the learning content of the AI ​​model based on the user's evaluation data, thereby correcting incorrect answers and strengthening correct answers.

[2233] Prompt Sentence Examples

[2234] For example, if user A posts a question such as "What's the latest anime-related news?", the server stores this question in a database and provides it to the AI ​​model for learning. Next, if user B posts an answer such as "It's been announced that 'Work X' will be made into a movie," this answer is also stored in the database and provided to the AI ​​model. If user C views this question and answer and rates it as "helpful," this rating data is also saved and used to retrain the AI ​​model. In this way, the system allows users to actively participate in the growth and improvement of the AI's accuracy.

[2235] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2236] Step 1: User enters question

[2237] A user uses a terminal to access the user interface and inputs a question by typing the question in a text box and clicking a submit button.

[2238] Input: User ID, question, timestamp

[2239] Output: Question data (JSON format)

[2240] Step 2: The device sends the query data

[2241] The device converts the question data entered by the user into JSON format and sends it to the server. This data includes the user ID, question content, and timestamp.

[2242] Input: Question data (JSON format)

[2243] Output: Data packet to be sent

[2244] Step 3: The server receives the query data

[2245] The server receives the submitted question data and stores it in a database. As an additional process, it assigns a unique ID to each question.

[2246] Input: Data packet to send

[2247] Output: Question data stored in the database (with unique ID)

[2248] Step 4: The server provides the question data to the AI ​​model

[2249] The server extracts question data stored in the database and provides it to the AI ​​model for training. The AI ​​model uses this information to add new knowledge and improve the accuracy of future question answers.

[2250] Input: Question data stored in the database

[2251] Output: Provided as training data to the AI ​​model

[2252] Step 5: User Enters Answer

[2253] A user can enter an answer to a question posted by another user by selecting the question in the user interface, entering the answer in the text box that appears, and clicking the submit button.

[2254] Input: User ID, Question ID, Answer, Time Stamp

[2255] Output: Response data (JSON format)

[2256] Step 6: The device sends the response data

[2257] The device converts the answer data entered by the user into JSON format and sends it to the server. This data includes the user ID, question ID, answer content, and timestamp.

[2258] Input: Response data (JSON format)

[2259] Output: Data packet to be sent

[2260] Step 7: The server receives the response data

[2261] The server receives the submitted answer data and stores it in a database. As an additional process, it assigns a unique ID to each answer and associates it with the question ID.

[2262] Input: Data packet to send

[2263] Output: Answer data saved in the database (with unique ID)

[2264] Step 8: The server provides the answer data to the AI ​​model

[2265] The server extracts the response data stored in the database and provides it to an artificial intelligence model for learning. The AI ​​model uses this information to add new knowledge and improve the accuracy of future responses.

[2266] Input: Response data stored in the database

[2267] Output: Provided as training data to the AI ​​model

[2268] Step 9: The device requests data

[2269] The terminal requests question and answer data from the server in response to a user request, which is sent by clicking the "Show Questions and Answers" button on the user interface.

[2270] Input: Request data

[2271] Output: None (request sent)

[2272] Step 10: Server sends data

[2273] The server extracts the requested question and answer data from the database and sends it to the device in JSON format.

[2274] Input: Request data

[2275] Output: Question and answer data in JSON format

[2276] Step 11: The terminal displays the data

[2277] The terminal displays the question and answer data received from the server on a user interface, and can also display new or popular answers with priority.

[2278] Input: Question and answer data in JSON format

[2279] Output: Questions and answers displayed in the user interface

[2280] Step 12: User Enters Rating

[2281] The user inputs a rating for the displayed answer, such as "helpful" or "not helpful," and selects one and clicks the submit button.

[2282] Input: User ID, Answer ID, Evaluation details, Time stamp

[2283] Output: Evaluation data (JSON format)

[2284] Step 13: The device sends the evaluation data

[2285] The device converts the evaluation data entered by the user into JSON format and sends it to the server. This data includes the user ID, answer ID, evaluation content, and timestamp.

[2286] Input: Evaluation data (JSON format)

[2287] Output: Data packet to be sent

[2288] Step 14: Server receives reputation data

[2289] The server receives the evaluation data and stores it in a database. Each evaluation is assigned a unique ID.

[2290] Input: Data packet to send

[2291] Output: Evaluation data stored in the database (with unique ID)

[2292] Step 15: Server periodically retrains

[2293] The server periodically provides the question data, answer data, and evaluation data in the database to the AI ​​model for re-learning, a process that improves the accuracy of the AI ​​model.

[2294] Input: Question data, response data, and evaluation data stored in the database

[2295] Output: Retrained AI model

[2296] Step 16: The server processes the feedback

[2297] The server automatically adjusts the AI ​​model's learning based on user evaluation data, improving the accuracy of answers and correcting incorrect answers.

[2298] Input: Evaluation data

[2299] Output: The tuned AI model

[2300] (Application example 1)

[2301] 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."

[2302] In recent years, as demand for interactive content distribution services has increased, AI models are being developed based on information provided by users, and there is a demand for more accurate information provision. However, current systems lack efficient collection and learning of user questions, answers, and evaluation data, resulting in slow AI model development. Furthermore, the user interface is difficult to use, making it difficult to improve user satisfaction.

[2303] 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.

[2304] In this invention, the server includes: means for providing a user interface through which a user can input a question; means for transmitting the input question data to the server; means for storing the received question data in a database; means for providing the question data stored in the database to the generative AI model; means for providing a user interface through which other users can input answers to questions; means for transmitting the input answer data to the server; means for storing the received answer data in the database; means for providing the answer data stored in the database to the generative AI model; means for providing a user interface through which users can view questions and answers; means for providing a user interface through which users can input ratings for answers; means for transmitting the input rating data to the server; means for storing the received rating data in the database; means for providing the questions, answers, and rating data in the database to the generative AI model and performing re-training; means for adjusting the learning content of the generative AI model based on feedback data; means for causing the generative AI model to generate prompt sentences based on the questions, answers, and rating data; and means for providing the generated prompt sentences to the generative AI model as further training data. This enables efficient growth of the AI ​​model and provision of highly accurate information through user interaction.

[2305] A "user interface" is an interface that provides a screen and input means for a user to interact with an application or system.

[2306] "Question data" refers to data including the content of a question that is input by a user through a user interface and sent to a server.

[2307] "Answer data" refers to data including the answer content that is input by other users to a question and sent to the server.

[2308] "Evaluation data" refers to data including evaluations such as "useful" or "not useful" that are entered by users in response to the answers and sent to the server.

[2309] A "generative AI model" is an artificial intelligence model that continuously learns based on questions, answers, and evaluation data provided by users, and provides highly accurate information.

[2310] "Server" refers to a computer system that receives question data, answer data, and evaluation data from users and stores and manages them in a database.

[2311] A "database" is a storage device for accumulating and managing question data, answer data, and evaluation data stored on a server.

[2312] "Feedback data" is data used to adjust the learning content of the generative AI model based on evaluations made by users.

[2313] A "prompt sentence" is training data generated by a generative AI model based on questions, answers, and evaluation data, and is a sentence used for further AI training.

[2314] "Interaction data" is a general term for data related to operations such as questions, answers, and ratings that users perform within the system.

[2315] The present invention provides an interactive system in which users can develop a generative AI model through questions, answers, and ratings. This system has a mechanism for effectively collecting and storing information provided by users and providing it to the generative AI model for learning. Detailed modes for implementing the invention are described below.

[2316] Generating a Program

[2317] First, a user interface is provided that allows users to input questions. This interface runs on devices such as smartphones and provides a screen for users to input their interests and questions.

[2318] The server receives the input question data and stores it in a database. The hardware used here can be a general web server, and the database can be an RDBMS such as PostgreSQL.

[2319] Generative AI models are provided with question data entered by users, and the AI ​​models learn from this data. These AI models are typically implemented using Python libraries (e.g., TensorFlow and PyTorch).

[2320] Next, a user interface is provided that allows other users to enter answers to questions. This is also an interface on a smartphone, and users can enter and submit their answers.

[2321] The received response data is sent to the server and stored in a database. Similarly, this response data is also provided to the generative AI model for further learning.

[2322] Furthermore, a user interface is provided where users can view questions and answers. In this interface, questions and their answers are displayed and users can rate them. The rating data is sent to a server and stored in a database.

[2323] The saved evaluation data is provided to the generative AI model, which retrains the AI ​​model based on the evaluation results. The learning content of the generative AI model is also adjusted based on the feedback data.

[2324] A natural language description of the process

[2325] When a user inputs a question via a device, the device sends the question data to a server. The server stores the received question data in a database and provides the data to a generative AI model. The generative AI model learns from this question data and improves its ability to respond to users' interests and questions.

[2326] When other users enter their answers via their devices, the data is also sent to the server and stored in a database, which is then provided to the generative AI model, allowing it to further learn.

[2327] When a user rates an answer, the rating data is sent to the server and stored in a database. The generative AI model then retrains based on this rating data. Furthermore, by adjusting the AI ​​model's learning content based on feedback data, the accuracy of future answers is improved.

[2328] Below are some examples of prompt sentences.

[2329] Examples of prompt statements

[2330] "A user typed in the question: 'What's the latest anime-related news?'

[2331] User B then enters the following answer:

[2332] It has been announced that "Work X" will be made into a movie.

[2333] Other users found this answer helpful.

[2334] This information is used to grow the generative AI model, enabling it to provide more accurate information to users. In this way, the AI ​​model grows efficiently through user interaction, realizing the provision of more accurate information.

[2335] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2336] Step 1:

[2337] The user inputs a question via the terminal. The user interface has a text input field where the user inputs the question in text format. The input question is converted into JSON format data on the terminal and sent to the server. At this time, the input data includes the user ID, question content, and timestamp.

[2338] Step 2:

[2339] The server stores the received question data in a database. A unique ID is assigned to the question data and stored in the database. At the same time, the server provides this question data to the generative AI model, which then learns from the data. At this time, the server transfers the data to the generative AI model's learning engine, which processes and calculates the data using the question data as input.

[2340] Step 3:

[2341] Other users input answers to questions. The device's user interface provides a text input field for entering answers to questions displayed on the screen. The user enters the answer there, and the answer data is sent to the server in JSON format. The input data includes the user ID, question ID, answer content, and timestamp.

[2342] Step 4:

[2343] The server stores the received answer data in a database. The received data is stored along with a unique ID associated with the answer and associated with the question ID. At the same time, the server provides this answer data to a generative AI model, which then learns from the data.

[2344] Step 5:

[2345] The user browses questions and answers. The device's user interface sends a request to the server to retrieve a list of saved questions and answers. The server retrieves the relevant data from the database and sends it to the device. The device displays the received questions and answers in the user interface. This display includes a mechanism to prioritize new and popular answers.

[2346] Step 6:

[2347] The user enters a rating for the answer. The user interface provides buttons for entering ratings such as "helpful" or "not helpful." When the user presses the rating button, the rating data is sent to the server in JSON format. The rating data includes the user ID, answer ID, rating content, and timestamp.

[2348] Step 7:

[2349] The server stores the received evaluation data in a database. The evaluation data is stored with its associated unique ID and associated with the answer ID. At the same time, the server provides this evaluation data to the generative AI model, which uses the data for retraining. As a result of the retraining, the generative AI model grows to provide more accurate and useful answers.

[2350] Step 8:

[2351] The server provides the questions, answers, and evaluation data in the database to the generative AI model, and periodically retrains it. This allows the generative AI model to constantly learn the latest information. The learning content of the generative AI model is adjusted based on the feedback data. The performance of the AI ​​model improves as a result of data processing and data calculation.

[2352] Step 9:

[2353] The generative AI model generates prompts based on the questions, answers, and evaluation data. The generated prompts are used as further training data for the AI ​​model. The generated prompts are as follows:

[2354] "A user typed in the question: 'What's the latest anime-related news?'

[2355] User B then enters the following answer:

[2356] It has been announced that "Work X" will be made into a movie.

[2357] Other users found this answer helpful.

[2358] Based on this prompt, the generative AI model re-learns and improves.

[2359] 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.

[2360] The present invention combines an emotional engine with an interactive system in which users develop an artificial intelligence model through questions, answers, and ratings. This system recognizes the user's emotional state and can adjust the display order of questions and answers and the generation of feedback based on that state. The specific form and processing flow of this system are described below.

[2361] Enter and save your question

[2362] Enter your question

[2363] Users can input their interests and doubts as questions through the user interface of their terminal.

[2364] Example: User A types, "What's the latest anime-related news?"

[2365] Submit a Question

[2366] The terminal transmits the question data entered by the user to the server.

[2367] This data is sent in a format such as JSON and includes metadata such as the user ID, question, and timestamp.

[2368] Saving Question Data

[2369] The server stores the received question data in a database. A unique ID is assigned to each question, allowing the question content to be uniquely identified.

[2370] Providing questions to the AI ​​model

[2371] The server provides the stored question data to the AI ​​model, which then learns from the data. Through this learning, the AI ​​accumulates information about the user's interests and interests.

[2372] Sentiment analysis of question data

[2373] The server provides the question data to the emotion engine to analyze the user's emotions.

[2374] Example: In response to User A's question, "What's the latest anime-related news?", the emotion engine recognizes emotions such as "interesting" and "expecting."

[2375] Enter and save your answers

[2376] Enter your answer

[2377] A user inputs an answer to a question posted by another user through a user interface.

[2378] Example: User B replies, "It has been announced that 'Work X' will be made into a movie."

[2379] Submit your answer

[2380] The device sends the answer data entered by the user to the server. The answer data is also sent in JSON format and includes the user ID, question ID, answer content, timestamp, etc.

[2381] Saving response data

[2382] The server stores the received answer data in a database. Each answer is assigned a unique ID and associated with the question ID. 【...

Claims

1. means for providing a user interface through which a user can input a question; means for transmitting the input question data to a server; a means for storing the received question data in a database; a means for providing the question data stored in the database to the artificial intelligence model; means for providing a user interface that allows other users to input answers to questions; means for transmitting the inputted answer data to a server; A means for storing the received response data in a database; a means for providing the response data stored in the database to an artificial intelligence model; means for providing a user interface through which a user can view questions and answers; means for providing a user interface that allows a user to input an evaluation of an answer; means for transmitting the input evaluation data to a server; means for storing the received evaluation data in a database; a means for providing the questions, answers, and evaluation data in the database to the artificial intelligence model for retraining; A system that includes a means for adjusting the learning content of an artificial intelligence model based on feedback data.

2. A means for providing the question data stored in the database to an artificial intelligence model and using it as learning data; A means for providing the response data stored in the database to an artificial intelligence model and using it as learning data; 10. The system of claim 1, further comprising means for modifying the artificial intelligence model based on the evaluation data.

3. 2. The system of claim 1, wherein the user interface through which the user can view questions and answers includes means for preferentially displaying popular answers.

Images