System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating interview questions using a predictive model; means for connecting experts to be interviewed with a AI interviewer to conduct an interview; means for analyzing collected interview information to generate expert profiles and articles; and means for recommending content to appropriate users, wherein the content is generated based on user interests and past browsing history.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] In modern society, there are many specialists with specialized knowledge and experience in a variety of fields, but their specialized knowledge and valuable experience are not being shared sufficiently. As a result, it is difficult for people who want to find a new hobby or deepen their knowledge in a specific field to obtain appropriate information. Furthermore, there are limited ways for specialists to organize and disseminate their knowledge and experience, which makes it difficult for their passion to be widely disseminated. This invention aims to solve these problems and promote the sharing of specialized knowledge. [Means for solving the problem]

[0005] This invention provides a system that solves the above problems by using the following means: a means for generating interview questions for specialists using a generative model, a means for connecting the interviewed specialist with an AI interviewer to conduct the interview, a means for analyzing collected interview data to generate specialist profiles and articles, and a means for recommending the generated content to appropriate users based on the user's interests and past browsing history. This system allows specialists' expertise and experience to be organized, shared, and disseminated, encouraging the discovery of new hobbies and interests and building the specialists' own confidence.

[0006] A "generative model" is a data generation algorithm that uses artificial intelligence, a system for generating new information or data based on specific input data.

[0007] "Interview questions" refer to questions generated to draw out the knowledge and experience of specialists, and are automatically created by a generative model.

[0008] A "specialist" is an expert with deep knowledge and extensive experience in a particular field.

[0009] "AI Interviewer" is a program that uses artificial intelligence technology to automatically conduct interviews and extract information from specialists.

[0010] "Means for conducting the interview" refers to the technological means for conducting the dialogue and collecting information, and refers to the system for connecting with the interviewee using an AI interviewer.

[0011] "Interview data" refers to information such as specialists' responses and anecdotes collected through interviews.

[0012] A "profile" is information that systematically organizes a specialist's knowledge and experience, generated based on collected interview data.

[0013] "Articles" refer to content in the form of reading material generated from interview data, intended to help users deepen their understanding of a particular field.

[0014] "Recommendation" is a method of delivering appropriate information by suggesting highly relevant content based on a user's interests and past browsing history.

[0015] A "user profile DB" is a database that stores information about users' interests and past browsing history. [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 a system for sharing and spreading the knowledge and experience of experts using generative models. The purpose of this system is to support the discovery of new hobbies and the deepening of expertise, thereby enabling their knowledge to be widely disseminated. Specific embodiments are described below.

[0038] Server Processing

[0039] 1. Generate interview questions:

[0040] The server uses the generative model to automatically generate interview questions for specialists, for example, using machine learning algorithms to create optimal questions from a training database.

[0041] 2. Interview scheduling and management:

[0042] The server coordinates the interview date and time with the specialist to be interviewed and stores the schedule information in a database, thereby managing the specialist to participate in the interview at an appropriate time.

[0043] 3. Conducting the interview:

[0044] When the interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer in real time, with the generative model automatically asking questions and collecting the specialist's answers.

[0045] 4. Generate profiles and articles:

[0046] Based on the collected interview data, the server uses a generative model to generate specialist profiles and articles, which systematically organize the specialist's knowledge and experience.

[0047] 5. User Recommendations:

[0048] The server references the user profile database and recommends content to appropriate users based on the collected user interests and past browsing history. For example, a newly generated article about a pottery expert will be recommended to users who are interested in pottery.

[0049] Terminal handling

[0050] 1. Notification of Interview Participation:

[0051] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, and the specialist will be able to view detailed interview information.

[0052] 2. Interview phase begins:

[0053] When a specialist joins an interview, the device initiates the interview session and connects them in real time with an AI interviewer, who then communicates using voice and text.

[0054] 3. Viewing and Feedback:

[0055] Users can view the generated profiles and articles through their devices and can provide feedback on their impressions while viewing, which is then sent to the server via their devices.

[0056] User Action

[0057] 1. Interview Participation:

[0058] If the user participates in the interview as a specialist, they will check the interview details based on the notification from the device, and at the specified date and time, they will participate in the interview phase through the device.

[0059] 2. Viewing profiles and articles:

[0060] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0061] 3. Feedback and sharing:

[0062] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0063] Specific examples

[0064] For example, if film director A is the subject of an interview, the process will proceed as follows:

[0065] 1. The server generates a question for the film director: "Which directors influenced you?"

[0066] 2. Arrange an interview date and time and send a notification to Director A.

[0067] 3. On the specified date and time, Director A will participate in the interview session via the terminal.

[0068] 4. The server runs the interview in real time and collects the response data.

[0069] 5. Generate a profile and article about film director A based on the collected data.

[0070] 6. Recommend generated articles to users who are interested in movies.

[0071] 7. Movie fan users can view the generated article on their devices and provide feedback on their impressions.

[0072] In this way, a system is created in which the knowledge and experience of specialists is systematically organized and shared with many users.

[0073] The processing flow will be explained below.

[0074] Server Processing Steps

[0075] Step 1:

[0076] The server uses the generative model to generate interview questions for the specialist in question, such as "Which directors influenced you?"

[0077] Step 2:

[0078] The server coordinates interview dates and times with the interviewed specialists and stores the schedule information in a database, where it is maintained for future reference.

[0079] Step 3:

[0080] When the scheduled interview date and time arrives, the server starts the interview phase, connecting the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time.

[0081] Step 4:

[0082] The server analyzes the collected interview data and uses generative models to generate specialist profiles and articles, systematically organizing the collected data into knowledge and experience.

[0083] Step 5:

[0084] The server references the user profile database and recommends content to appropriate users based on their interests and past browsing history. For example, a newly generated article about a ceramics expert will be recommended to users who are interested in ceramics.

[0085] Terminal processing steps

[0086] Step 1:

[0087] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, and the specialist will be able to view detailed interview information.

[0088] Step 2:

[0089] At the appointed date and time, when the specialist joins the interview, the device will initiate the interview session and connect with an AI interviewer in real time, engaging in a dialogue using voice and text.

[0090] Step 3:

[0091] Users can view the generated profiles and articles through their devices and provide feedback based on their impressions of the views, which is then sent to the server via their devices.

[0092] User processing steps

[0093] Step 1:

[0094] If the user participates in the interview as a specialist, they will check the interview details based on the notification from the device, and at the specified date and time, they will participate in the interview phase through the device.

[0095] Step 2:

[0096] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0097] Step 3:

[0098] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0099] Specific examples

[0100] The processing flow when film director A is the subject of an interview is as follows:

[0101] Server Processing Steps

[0102] Step 1:

[0103] Generate a question for film director A: "Which directors influenced you?"

[0104] Step 2:

[0105] Arrange an interview date and time and send a notification to Director A.

[0106] Step 3:

[0107] At the specified date and time, film director A will participate in the interview session via the terminal.

[0108] Step 4:

[0109] Conduct interviews and collect response data in real time.

[0110] Step 5:

[0111] Based on the collected data, a profile and article about film director A is generated.

[0112] Step 6:

[0113] Recommend generated articles to users who are interested in movies.

[0114] Terminal processing steps

[0115] Step 1:

[0116] Notify Director A of the interview date and time and how to participate.

[0117] Step 2:

[0118] Film director A starts an interview session using a terminal to participate in the interview.

[0119] Step 3:

[0120] Users can view the generated articles through their devices and provide feedback.

[0121] User processing steps

[0122] Step 1:

[0123] Director A will confirm the interview details based on the notification.

[0124] Step 2:

[0125] The user views an article about recommended film director A.

[0126] Step 3:

[0127] Users provide and share feedback on articles.

[0128] Example 1

[0129] 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."

[0130] There is a growing demand for systems that can efficiently collect the knowledge and experience of experts and provide it to a wide range of users. However, current systems face challenges in effectively managing and conducting interviews with experts and recommending appropriate content based on each user's interests and browsing history.

[0131] 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.

[0132] In this invention, the server includes a means for generating interview questions using a generative model, a means for connecting an AI interviewer to the expert being interviewed using real-time communication means to conduct the interview, a means for analyzing the collected interview data to generate expert profiles and articles, and a means for recommending the generated content to appropriate users based on the user's interests and past browsing history. This makes it possible to effectively collect and organize the knowledge and experience of experts and provide personalized information to users.

[0133] A "generative model" is a method or system that uses machine learning algorithms to generate new data based on existing data.

[0134] "Interview questions" are questions created using generative models to elicit information or knowledge from experts or subjects.

[0135] An "expert" is a person who has advanced knowledge and experience in a particular field and has deep understanding and skills in that field.

[0136] "AI Interviewer" is a system that uses artificial intelligence to automatically conduct interviews and collect necessary information.

[0137] "Real-time communications means" are technologies and systems that allow information to be exchanged over the Internet or other communications networks with little or no time delay.

[0138] A "profile" is a document or data that systematically organizes an expert's knowledge and experience and shows the individual's characteristics and activities.

[0139] An "article" is a piece of writing that is created based on interview results and expert knowledge and is intended to provide information to a large number of users.

[0140] "User" means a general user who uses the system to view content and provide feedback.

[0141] "Interests" are the interest or curiosity a user shows in a particular field or topic.

[0142] "Browsing history" is a record of the content a user has viewed within the system.

[0143] "Recommendation" is the act of presenting optimal content based on a user's interests and browsing history.

[0144] This invention is a system for sharing and spreading expert knowledge and experience using generative AI models, with the aim of supporting the discovery of new hobbies, deepening of expertise, and spreading expert knowledge widely.

[0145] Server Processing

[0146] Generate interview questions

[0147] The server generates interview questions using a generative AI model (e.g., GPT-4). The server retrieves relevant keywords and past questions from a database of past interviews, and inputs prompt sentences based on these into the generative AI model. This automatically generates optimal questions for the expert. The following is a specific example.

[0148] Example prompt sentence:

[0149] "Generate interview questions for film director A. I'd like to know which directors influenced him."

[0150] Scheduling and managing interviews

[0151] The server retrieves the specialist's schedule information from the API and adjusts the interview date and time. The adjusted schedule information is saved in the database and an email or in-app notification is sent to the specialist. This series of operations ensures that the specialist can attend the interview at the appropriate time.

[0152] Running the Interview

[0153] When the interview date and time arrives, the server will start the interview session using a real-time communication method such as Zoom or Teams. The server will connect the AI ​​interviewer with the expert and conduct the interview in real time by presenting generated questions. Dialogue during the interview can include both voice and text.

[0154] Profile and Article Generation

[0155] Once the interview is complete, the server inputs the collected data into a generative AI model to generate a profile and article about the expert. These profiles and articles systematically describe the expert's knowledge and experience. The server then stores this data in a database.

[0156] User recommendations

[0157] The server references a user profile database and recommends the most appropriate content based on the user's interests and past browsing history. For example, a user interested in pottery might be recommended a newly generated interview article with a pottery expert.

[0158] Terminal handling

[0159] Notification of interview participation

[0160] Experts who are selected for an interview will be notified of the interview date and time and how to participate via email and in-app notification, and they can view detailed interview information.

[0161] Beginning of the interview phase

[0162] When the interview date and time arrives, the expert participates in the interview phase via a device, which connects to an AI interviewer in real time and begins the interview using voice and text.

[0163] Views and Feedback

[0164] Users can view the generated profiles and articles through their devices, and can also provide feedback on their impressions while viewing, which is then sent to the server via their devices.

[0165] User Action

[0166] Participating in an interview

[0167] If the user participates in the interview as an expert, he / she will check the interview details based on the notification from the device, and at the specified date and time, he / she will participate in the interview phase through the device.

[0168] View Profiles and Articles

[0169] Users can browse interesting profiles and articles recommended by the server through their devices. For example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0170] Feedback and sharing

[0171] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0172] As a result, this system can effectively collect and organize the knowledge and experience of experts, and provide personalized information to users.

[0173] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0174] Step 1:

[0175] The server generates interview questions using a generative AI model. In this step, it takes keywords obtained from past interview records and related databases as input and generates a prompt based on them. The prompt is sent to a generative AI model (e.g., GPT-4) to generate a new interview question. This question is output in the form of, for example, "What kind of movie would you like to make next?" Specifically, the server accesses the database, extracts relevant data, converts it into a prompt, sends it to the generative AI model, and stores the generated question in the database.

[0176] Step 2:

[0177] The server schedules an interview with a specialist. First, it accesses the specialist's calendar API to obtain available time slots. Based on the obtained schedule data, it confirms the interview date and time and saves it in the database. It also sends the confirmed date and time and participation instructions to the specialist via email and in-app notification. The input is the specialist's available schedule, and the output is the confirmed interview date. Specific operations include checking the schedule via the API, algorithmic processing to adjust the date and time, updating the database, and sending notifications.

[0178] Step 3:

[0179] When the interview date and time arrives, the server starts the interview phase using a real-time communication tool (e.g., Zoom, Teams). The server provides pre-generated interview questions to the AI ​​interviewer, which then presents the questions to the expert sequentially. The input is the interview date and time and the generated questions, and the output is the expert's response data. Specific operations include creating a meeting by calling Zoom's API, providing the generated questions sequentially, and collecting the expert's responses.

[0180] Step 4:

[0181] The server generates profiles and articles based on the collected interview data. The collected response data is input into a generative AI model to generate expert profiles and articles. The input for this process is the collected response data, and the output is the generated profile and article. Specific operations include text analysis of the collected data, generation of profiles and articles by the generative AI model, and storage in a database.

[0182] Step 5:

[0183] The server recommends the generated profiles and articles to users. First, it references the user profile database to obtain the user's interests and past browsing history. Based on this data, it selects highly relevant content and recommends it to the target user. The input is the user's interests and browsing history, and the output is the recommended content. Specific operations include database query, content selection using a recommendation algorithm, and notification and email sending.

[0184] Step 6:

[0185] The device notifies the selected expert of the interview date and time and how to participate. The device provides detailed interview information to the expert based on the schedule information obtained from the server. The input is the schedule information from the server, and the output is a notification to the expert. Specific operations include generating notifications and sending notifications via email or within the app.

[0186] Step 7:

[0187] When the interview date arrives, the terminal helps the expert join the interview session. The expert operates the terminal to connect to the real-time communication tool and begins a conversation with the AI ​​interviewer. The input is the meeting link provided by the server, and the output is the start of the interview session. Specific operations include providing the meeting link and guiding the connection procedure.

[0188] Step 8:

[0189] Users can view the generated profiles and articles through their devices and provide their thoughts and feedback. When users enter their feedback, the data is sent to the server via their devices. The input is the user's feedback, and the output is sending the feedback to the server. Specific actions include entering feedback, pressing the send button, and sending data to the server.

[0190] Step 9:

[0191] When a user joins an interview, the device provides the user with detailed interview information and invites them to the interview session at the specified date and time. The input is the notified interview details, and the output is participation in the interview phase. Specific actions include operating the start button for the interview phase and connecting to a real-time communication tool.

[0192] Step 10:

[0193] Users browse recommended profiles and articles and share the content they find particularly interesting with other users or on social media. The input is the recommended content, and the output is the shared content. Specific actions include viewing the content, clicking the share button, and posting to social media.

[0194] (Application example 1)

[0195] 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."

[0196] This invention relates to a system for widely sharing and disseminating the knowledge and experience of experts, aiming to effectively deliver knowledge from a variety of fields, not just specific fields, to a wide range of users. Another objective is to create more interactive and fulfilling learning opportunities by distributing expert interviews in real time and providing an environment in which users can provide immediate feedback on the content. Conventional systems have limited the dissemination of knowledge due to the difficulty of efficiently collecting and organizing expert knowledge and delivering it to the appropriate users.

[0197] 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.

[0198] In this invention, the server includes a means for generating interview questions using a generative model, a means for connecting the interviewed expert with an AI interviewer to conduct the interview, a means for analyzing collected interview data to generate expert profiles and articles, a means for recommending generated content to appropriate users based on users' interests and past browsing history, a means for live and recorded broadcasts of interviews, and a means for collecting and managing feedback from users. This allows experts' knowledge and experience to be shared effectively in real time, and users can provide instant feedback. Therefore, the diffusion of knowledge and feedback to experts occurs in both directions, creating a more comprehensive knowledge sharing platform.

[0199] A "generative model" is an algorithm or method that can automatically create new information or patterns based on data.

[0200] "Means for generating interview questions" refers to a function that uses a generative model to automatically create appropriate questions for interviewees.

[0201] An "artificial intelligence interviewer" is a virtual interviewer that uses artificial intelligence technology to conduct interviews.

[0202] "Means for conducting interviews" refers to the function of connecting the interviewee with an artificial intelligence interviewer and conducting the interview.

[0203] "Means for analyzing collected interview data" refers to the function of analyzing data collected during the interview process and extracting useful information.

[0204] "Means for generating expert profiles and articles" refers to the function of creating expert profiles and related articles based on the analyzed interview data.

[0205] "Recommendation methods" refers to a function that recommends appropriate content to a specific user based on the user's interests and past browsing history.

[0206] "Means for live streaming of interviews" refers to a function for streaming the contents of interviews in real time.

[0207] "Means for distributing recorded interviews" refers to the function of distributing recorded interviews so that users can watch them later.

[0208] "Means for collecting and managing feedback" refers to the function of collecting opinions and feedback from users and managing them systematically.

[0209] In this invention, a specific embodiment of a system for sharing and disseminating expert knowledge and experience will be described.

[0210] System program generation

[0211] Server Processing

[0212] 1. Generate interview questions:

[0213] The server uses a generative AI model to automatically generate interview questions for experts. This generative AI model uses a language model such as GPT-3. The generated questions are stored in a database.

[0214] 2. Interview scheduling and management:

[0215] The server coordinates interview dates and times with experts and stores the information in a database using a database management system such as SQLite.

[0216] 3. Conducting the interview:

[0217] When the interview date and time arrives, the server starts the interview phase in real time, connecting the expert with the AI ​​interviewer using technologies such as WebRTC.

[0218] 4. Generate profiles and articles:

[0219] The collected interview data is analyzed and a generative AI model is used to generate expert profiles and articles, which are then stored in a database.

[0220] 5. Content Recommendations:

[0221] The server uses a user profile database to recommend appropriate content based on the user's interests and past browsing history. The recommendation engine uses collaborative filtering and content-based filtering algorithms.

[0222] Terminal handling

[0223] 1. Notification of Interview Participation:

[0224] The server notifies the interviewed expert of the date and time of the interview and how to participate via email or push notification.

[0225] 2. Interview phase begins:

[0226] At the time of the interview, the device initiates the interview session, connecting the AI ​​interviewer with the expert in real time, enabling an interactive dialogue using voice and text.

[0227] 3. Watch the live and recorded interviews:

[0228] Users can watch the live interview on their device, and even after the live broadcast has ended, they can watch the recorded footage later.

[0229] 4. Providing Feedback:

[0230] Users can provide feedback on interviews and articles they have watched from their devices, and this feedback is immediately sent to the server and managed.

[0231] Specific examples

[0232] For example, if an expert film director is the subject of an interview, the process will proceed as follows:

[0233] 1. The server generates a question for the film director: "Which directors influenced you?"

[0234] 2. Arrange an interview date and time and send a notification to the film director.

[0235] 3. At the appointed date and time, the film director will participate in a real-time interview via the terminal.

[0236] 4. The server runs the interview in real time and collects the response data.

[0237] 5. Based on the collected data, generative AI models are used to generate profiles and articles about film directors.

[0238] 6. Recommend the generated articles and profiles to movie fan users.

[0239] 7. Movie fans can view the article on their devices and provide feedback on their impressions.

[0240] Example prompt sentence:

[0241] Generate a professional profile of the film director. Based on interviews with him, include information about his biggest influences and the inspiration for his latest work.

[0242] In this way, a series of processes on the server and terminals will realize a system in which the knowledge and experience of experts can be efficiently collected, organized, shared, and delivered to many users.

[0243] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0244] Step 1:

[0245] Generate interview questions

[0246] The server uses a generative AI model (e.g., GPT-3) to automatically generate questions for the expert being interviewed based on the prompt. The input is the "expert's field" and the "question prompt." The generative AI model outputs an appropriate question (e.g., "Which director influenced you?") based on these inputs and stores it in a database. Specifically, the server sends the prompt to the generative AI model, formats the returned answer as a question, and stores it in the database.

[0247] Step 2:

[0248] Scheduling and managing interviews

[0249] The server contacts the expert and arranges the interview date and time. The inputs are the expert's contact information and available time slots. Based on this, the server determines the interview date and time and saves it in the database. Specifically, the server contacts the expert via email or in-app notification and records the agreed interview date and time in the database.

[0250] Step 3:

[0251] Running the Interview

[0252] When the time for the interview arrives, the server starts the interview phase in real time. The inputs are the question and expert information. The server connects the expert and the AI ​​interviewer using technologies such as WebRTC and collects the results. Specifically, it initiates a real-time voice or text dialogue and stores the interview content in a database.

[0253] Step 4:

[0254] Profile and Article Generation

[0255] The server analyzes the collected interview data and uses a generative AI model to generate expert profiles and articles. The inputs are "interview data" and "generative AI model." The interview data is preprocessed using a data analysis tool and then input into the generative AI model to obtain profiles and articles. This output data is stored in a database. Specifically, it organizes the interview data through natural language analysis and automatically generates profiles and articles based on it.

[0256] Step 5:

[0257] User recommendations

[0258] Based on the user's interests and past browsing history, the server recommends appropriate content to the user. The inputs are a user profile database and generated profiles and articles. A recommender system is built using collaborative filtering and content-based filtering algorithms to recommend appropriate content to the user. Specifically, it analyzes the user profile, identifies users with similar interests and related content, and presents it to the user.

[0259] Step 6:

[0260] Gathering feedback

[0261] Users can provide feedback on the content they have viewed. Inputs include "user feedback content" and "user ID." The device sends this feedback to the server, which stores it in a database. Specifically, the system receives feedback through the user interface and transfers it to the server as structured data.

[0262] This series of steps will create a system that efficiently collects, organizes, and shares expert knowledge and experience, and provides appropriate content based on user interests.

[0263] 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.

[0264] This invention is a system for sharing and spreading expert knowledge and experience using a generative model and an emotion engine. The system aims to promote the sharing of expert knowledge by recommending content generated based on the user's interests and emotions to appropriate users. Specific embodiments are described below.

[0265] Server Processing

[0266] 1. Generate interview questions:

[0267] The server uses the generative model to generate interview questions for the specialist in question, such as "Which film directors have influenced you?"

[0268] 2. Interview scheduling and management:

[0269] The server coordinates the interview date and time with the interviewee specialist and stores the schedule information in a database, which can be referenced later at the appropriate time.

[0270] 3. Conducting the interview:

[0271] When the interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time. At this time, the emotion engine also works to analyze the user's emotions.

[0272] 4. Emotional data collection and analysis:

[0273] The server uses an emotion engine to analyze user data collected during the interview in real time, assessing emotions based on data such as facial expressions, tone of voice, and survey responses selected.

[0274] 5. Profile and article generation:

[0275] Based on the interview and sentiment analysis data, the server uses a generative model to generate specialist profiles and articles that include a multifaceted organization of expertise and experience, as well as sentiment data.

[0276] 6. User Recommendations:

[0277] The server refers to the user profile database and recommends the generated content to the appropriate user based on the user's interests, past browsing history, and emotional feedback. For example, a newly generated article about a ceramics expert may be provided to a user who is interested in ceramics and has previously viewed related content and felt emotionally satisfied.

[0278] Terminal handling

[0279] 1. Notification of Interview Participation:

[0280] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and / or in-app notification, and the specialist will be able to view detailed interview information.

[0281] 2. Interview phase begins:

[0282] At the appointed date and time, the device initiates a session for the specialist to participate in the interview, connecting the device to the AI ​​interviewer in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[0283] 3. Viewing and Feedback:

[0284] Users can view the profiles and articles generated through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[0285] User Action

[0286] 1. Interview Participation:

[0287] If the user participates in the interview as a specialist, they will check the interview details based on the notification on their device, and at the designated date and time, they will participate in the interview phase through their device.

[0288] 2. Viewing profiles and articles:

[0289] Users can browse recommended profiles and articles on their devices. For example, an amateur pottery enthusiast can read an interview with a pottery expert. As the user reads the article, the emotion engine analyzes their emotions in real time and sends emotional data as feedback.

[0290] 3. Feedback and sharing:

[0291] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0292] Specific examples

[0293] For example, if film director B is the subject of an interview, the following process will be carried out.

[0294] 1. The server generates a question for the film director: "Which film directors have influenced you?"

[0295] 2. Arrange an interview date and time and send a notification to Director B.

[0296] 3. On the specified date and time, Director B will join the interview session via the terminal.

[0297] 4. The server conducts the interview in real time and collects the response data. At the same time, the emotion engine analyzes Director B's emotions.

[0298] 5. Generate a profile and article about film director B based on the collected data.

[0299] 6. Recommend generated articles to users who are interested in movies and who have had a positive emotional reaction to past related articles.

[0300] 7. Movie fan users view the generated articles on their devices and provide their impressions as feedback that is analyzed by the emotion engine.

[0301] In this way, a system that integrates an emotion engine allows experts' knowledge and experience to be shared more deeply and made available to a wider audience.

[0302] The processing flow will be explained below.

[0303] Server Processing Steps

[0304] Step 1:

[0305] The server uses the generative model to generate interview questions for the specialist in question, such as "Which film directors have influenced you?"

[0306] Step 2:

[0307] The server coordinates the interview date and time with the specialist to be interviewed and stores the schedule information in a database, thereby managing the specialist to participate in the interview at an appropriate time.

[0308] Step 3:

[0309] When the scheduled interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time. At the same time, the emotion engine also runs to collect the user's emotional data.

[0310] Step 4:

[0311] The server uses an emotion engine to analyze the user's emotional data collected during the interview in real time, assessing their emotions based on, for example, their facial expressions, tone of voice, and selected survey answers.

[0312] Step 5:

[0313] The server uses a generative model to generate specialist profiles and articles based on interview response data and sentiment data. These profiles and articles systematically organize knowledge and experience.

[0314] Step 6:

[0315] The server refers to a user profile database and recommends generated content to appropriate users based on their interests, past browsing history, and emotional feedback. For example, a newly generated article about a pottery expert will be provided to users who are interested in pottery and have previously viewed related content and had a positive emotional response.

[0316] Terminal processing steps

[0317] Step 1:

[0318] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, where the specialist can view detailed interview information.

[0319] Step 2:

[0320] At the designated date and time, the device will initiate the interview session and connect with the AI ​​interviewer in real time, while the emotion engine will simultaneously collect and analyze the user's emotional data.

[0321] Step 3:

[0322] Users can view the generated profiles and articles through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[0323] User processing steps

[0324] Step 1:

[0325] If the user participates in the interview as a specialist, they will check the interview details based on the notification on their device, and at the designated date and time, they will participate in the interview phase through their device.

[0326] Step 2:

[0327] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0328] Step 3:

[0329] Users can provide feedback on the content and share articles they find particularly interesting with other users or on social media, further spreading their expertise and interests. This feedback is analyzed by the sentiment engine and sent to the server.

[0330] Specific examples

[0331] For example, if film director B is the subject of an interview, the processing flow is as follows:

[0332] Server Processing Steps

[0333] Step 1:

[0334] Generate a question for Director B: "Which directors influenced you?"

[0335] Step 2:

[0336] The interview date and time is arranged and a notification is sent to Director B.

[0337] Step 3:

[0338] At the specified date and time, film director B will participate in the interview session via the terminal.

[0339] Step 4:

[0340] Conduct interviews in real time and collect response and sentiment data.

[0341] Step 5:

[0342] Based on the collected data, a profile and article about film director B is generated.

[0343] Step 6:

[0344] The generated articles are recommended to users who are interested in movies and who have a positive reaction based on emotional feedback.

[0345] Terminal processing steps

[0346] Step 1:

[0347] Notify Director B of the interview date and time and how to participate.

[0348] Step 2:

[0349] Director B begins the interview session using a device to participate in the interview. At this time, the emotion engine also starts to collect and analyze emotion data.

[0350] Step 3:

[0351] Users can view the generated articles on their devices and provide feedback, which is also analyzed by the emotion engine and sent to the server.

[0352] User processing steps

[0353] Step 1:

[0354] Film director B checks the interview details based on the notification.

[0355] Step 2:

[0356] The user views an article about recommended film director B.

[0357] Step 3:

[0358] Users provide feedback on articles and share it on social media, etc. This feedback is analyzed by the emotion engine and sent to the server.

[0359] Example 2

[0360] 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."

[0361] There is a need for a method to effectively share the knowledge and experience of experts and provide appropriate content based on users' interests and emotions. However, conventional methods have been unable to fully utilize the quality of interview content or users' emotional feedback, making it difficult to recommend content that is highly relevant to users. This has resulted in insufficient dissemination of expert knowledge and improvement of user satisfaction.

[0362] 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.

[0363] In this invention, the server includes means for generating interview questions using a generative model, means for connecting the interviewed expert with an AI interviewer to conduct the interview, means for collecting and analyzing user data during the interview using an emotion engine, means for analyzing the collected interview data and emotion data to generate expert profiles and articles, and means for recommending the generated content to appropriate users based on the user's interests, past browsing history, and emotion feedback. This makes it possible to effectively share and spread the knowledge and experience of experts and provide users with content that is highly relevant and satisfying.

[0364] A "generative model" is a machine learning-based model that generates new text or information based on given prompts and data.

[0365] The "means for generating interview questions" is a mechanism that uses a generative model to automatically create specific questions for the expert being interviewed.

[0366] The "means of conducting the interview" is a mechanism that connects the subject expert with an AI interviewer and conducts the actual interview.

[0367] The "emotion engine" is a system that analyzes data such as the user's facial expressions, tone of voice, and selected survey responses to evaluate the user's emotions.

[0368] "Means for collecting and analyzing user data" refers to a system for collecting data obtained from users during interviews and analyzing their emotions and interests based on that data.

[0369] "Interview data" refers collectively to expert responses and other related information collected through interviews.

[0370] The "means for generating profiles and articles" is a mechanism for generating expert profiles and related articles based on the collected interview data and sentiment data.

[0371] The "user profile database" is a database for storing and managing information such as user interests, past browsing history, and emotional feedback.

[0372] "Means for recommending content" refers to a system that recommends appropriate content to users based on their interests, past browsing history, and emotional feedback.

[0373] This invention is a system for sharing and spreading expert knowledge and experience by utilizing a generative model and an emotion engine. The system aims to promote the sharing of expert knowledge by recommending content generated based on the user's interests and emotions to appropriate users. A specific embodiment of this system is described below.

[0374] Server Processing

[0375] 1. Generate interview questions:

[0376] The server generates interview questions using a generative AI model (e.g., GPT-4). At this time, instructions such as "Create interview questions for a film director" are input as prompts. An example of a question generated by the generative model is "Which film directors have influenced you?"

[0377] 2. Interview scheduling and management:

[0378] The server searches for relevant experts based on the generated interview questions and schedules the interview. This information is stored in a database and is referenced when the interview is conducted.

[0379] 3. Conducting the interview:

[0380] When the time for the interview arrives, the server connects the expert to the AI ​​interviewer and conducts the interview. The generative model automatically asks questions and collects the expert's answers in real time. At the same time, the emotion engine runs to analyze the user's emotional data.

[0381] 4. Emotional data collection and analysis:

[0382] The server uses an emotion engine to collect and analyze data such as the user's facial expressions, tone of voice, and questionnaire responses during the interview to evaluate the user's emotions.

[0383] 5. Profile and article generation:

[0384] Based on the interview and sentiment analysis data, the server uses a generative AI model to generate expert profiles and articles that include a multifaceted breakdown of expertise and experience, as well as sentiment data.

[0385] 6. User Recommendations:

[0386] The server references a user profile database and recommends content based on the user's interests, past browsing history, and emotional feedback. For example, a user interested in pottery can be recommended a newly generated article by a pottery expert.

[0387] Terminal handling

[0388] 1. Notification of Interview Participation:

[0389] The device will notify the interviewed expert of the date and time of the interview and how to participate. This notification will be provided via email and in-app notification, and the expert will be able to view detailed interview information.

[0390] 2. Interview phase begins:

[0391] At the specified date and time, the device will automatically start the interview session and connect the AI ​​interviewer with the expert in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[0392] 3. Viewing and Feedback:

[0393] Users can view the generated profiles and articles through their devices and can also provide feedback based on their impressions while browsing, which is analyzed by the emotion engine and sent to the server.

[0394] User Action

[0395] 1. Interview Participation:

[0396] When participating in an interview as an expert, the user will check the interview details based on notifications from the device, and at the designated date and time, they will participate in the interview phase through the device.

[0397] 2. Viewing profiles and articles:

[0398] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0399] 3. Feedback and sharing:

[0400] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0401] Specific examples

[0402] For example, if Director B is chosen for an interview, the following process will occur:

[0403] 1. The server uses the generative model to generate a question for the film director: "Which film directors have influenced you?"

[0404] 2. Arrange an interview date and time and send a notification to Director B.

[0405] 3. On the specified date and time, Director B will join the interview session via the terminal.

[0406] 4. The server conducts the interview in real time and collects the response data. At the same time, the emotion engine analyzes Director B's emotions.

[0407] 5. Generate a profile and article about film director B based on the collected data.

[0408] 6. Recommend generated articles to users who are interested in movies and have had a positive emotional reaction to past related articles.

[0409] 7. Movie fan users view the generated articles on their devices and provide their impressions as feedback that is analyzed by the emotion engine.

[0410] This system integrates an emotion engine to share the knowledge and experience of experts widely and deeply, enabling it to provide highly relevant content to many users.

[0411] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0412] Step 1: Generate interview questions

[0413] Input: Interview request from user, prompt

[0414] Specific behavior:

[0415] The server receives a request from the user to "interview a film director." The server inputs the prompt "Create interview questions for the film director" into a generative AI model (e.g., GPT-4).

[0416] Data processing / calculation:

[0417] The generative AI model analyzes the prompt and generates new interview questions, referencing question generation patterns from an existing database.

[0418] Output: Generated interview questions (e.g., "Which film directors influenced you?")

[0419] Step 2: Schedule and manage interviews

[0420] Input: Interview questions, expert schedule data

[0421] Specific behavior:

[0422] The server searches the database for relevant experts (e.g., film directors) based on the generated question, checks the experts' schedules, and arranges the date and time of the interview.

[0423] Data processing / calculation:

[0424] It looks at the expert's calendar, automatically identifies available time slots, and selects the appropriate time.

[0425] Output: List of experts with interview date and time (e.g., Film Director B, October 1, 2023, 2:00 PM)

[0426] Step 3: Conducting the interview

[0427] Input: Interview questions, expert connections

[0428] Specific behavior:

[0429] When the time for the interview arrives, the server sends a notification to the expert (e.g., film director B) to start the interview. The server then starts the AI ​​interviewer and delivers the generated questions one by one.

[0430] Data processing / calculation:

[0431] The AI ​​interviewer collects the expert's answers in real time and stores them in a database, while the emotion engine collects facial and vocal data and analyzes the emotional data.

[0432] Output: Expert response data, sentiment data collected in real time

[0433] Step 4: Collect and analyze emotion data

[0434] Input: Expert response data, user sentiment data

[0435] Specific behavior:

[0436] During the interview, the server uses an emotion engine to collect and analyze emotional data from the user in real time, such as facial expressions, tone of voice, and questionnaire responses.

[0437] Data processing / calculation:

[0438] The collected data is evaluated using facial expression analysis software and tone analysis software to determine and quantify positive, negative, and neutral emotions.

[0439] Output: Analyzed sentiment data (e.g., 80% positive, 10% negative, 10% neutral)

[0440] Step 5: Generate profiles and articles

[0441] Input: Expert response data, analyzed sentiment data

[0442] Specific behavior:

[0443] The server instructs the generative AI model to generate expert profiles and articles based on the collected response data and sentiment data.

[0444] Data processing / calculation:

[0445] The generative model analyzes this data, organizes the knowledge and experience of experts from various perspectives, and converts them into article format, while also appropriately reflecting sentiment data.

[0446] Output: Expert profile and article (e.g., article about film director B)

[0447] Step 6: Recommend to users

[0448] Input: User interest data, past browsing history, emotional feedback

[0449] Specific behavior:

[0450] The server references a user profile database and selects content based on interests, past browsing history, and emotional feedback, and then notifies the selected content to the target user.

[0451] Data processing / calculation:

[0452] It uses recommendation algorithms to surface content that matches your interests, and also uses your browsing history and emotional feedback to select the most relevant articles.

[0453] Output: Content notification to user (e.g. "A new article about pottery has been published")

[0454] (Application example 2)

[0455] 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."

[0456] The lack of a system that widely shares the knowledge and experience of experts and appropriately recommends them based on user interests and emotions is another issue. Another issue is the lack of technology that effectively utilizes user feedback and emotional data to organize and provide expert content from multiple perspectives.

[0457] 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 generating interview questions using a generative model, means for connecting the interviewed expert with an AI interviewer and conducting the interview, means for analyzing the collected interview data to generate expert profiles and articles, means for recommending content generated based on the user's interests, past browsing history, and emotional feedback to appropriate users, and means for analyzing the user's feedback with an emotion engine. This makes it possible to effectively share and recommend experts' knowledge and experience based on the user's emotions and interests.

[0458] A "generative model" refers to an algorithm that automatically generates new content, questions, etc. based on data.

[0459] "Interview questions" are questions asked to experts, and these questions are generated using a generative model.

[0460] An "expert" is an individual who has knowledge or experience in a particular field.

[0461] "AI interviewer" refers to a system or software that uses artificial intelligence to ask interview questions and collect answers.

[0462] "Interview Data" refers to the experts' responses and other relevant information collected during the interview.

[0463] A "profile" is information that outlines an expert's knowledge and experience, created based on collected data.

[0464] "Articles" refer to written information created based on interview data, and are detailed descriptions of the experts' knowledge and experiences.

[0465] "Recommendation" refers to recommending appropriate content based on a user's interests, past browsing history, and emotional feedback.

[0466] "Emotional feedback" refers to the analysis of users' emotional reactions when viewing content and the data and comments based on that.

[0467] An "emotion engine" refers to software or algorithms that analyze a user's emotions from facial expressions, tone of voice, text feedback, etc.

[0468] "Content" refers to the collection of information, entertainment, etc. provided to users through communications means.

[0469] "System" refers to an integrated mechanical or software collection of related means and devices.

[0470] This invention is a system that shares the knowledge and experience of experts and recommends appropriate content based on the user's interests and emotions. The detailed configuration and operation procedures for implementing this invention are described below.

[0471] Server Processing

[0472] 1. Generate interview questions:

[0473] The server uses a generative model to generate interview questions for domain experts, such as a natural language processing model like BERT (Bidirectional Encoder Representations from Transformers).

[0474] Example: Generating questions for a film director such as "Which film directors have influenced you?"

[0475] 2. Interview scheduling and management:

[0476] The server coordinates the interview date and time with the interview subject and stores the schedule information in a database, using the Python datetime module for schedule management.

[0477] Example: An interview with a film director is scheduled for "2023-10-01 15:00:00".

[0478] 3. Conducting the interview:

[0479] When the interview date and time arrives, the server starts the interview phase and connects the expert with the AI ​​interviewer. The generative model automatically asks questions and collects the expert's answers in real time. At the same time, the emotion engine runs and analyzes the user's emotional data.

[0480] Hardware and software used: GPU-based servers, generative AI models (e.g., GPT-3), emotion analysis software (e.g., facial emotion recognition API)

[0481] 4. Emotional data collection and analysis:

[0482] The server analyzes user data collected during the interview in real time using an emotion engine, which evaluates emotions based on data such as the user's facial expressions, tone of voice, and selected survey answers.

[0483] Example: Determining user satisfaction and interest in an expert's answer in real time.

[0484] 5. Profile and article generation:

[0485] Based on the interview and sentiment analysis data, the server uses a generative model to generate expert profiles and articles that include a multifaceted organization of expertise and experience, as well as sentiment data.

[0486] Hardware and software used: Database management system (e.g., PostgreSQL), natural language generation model (e.g., OpenAI's GPT-3)

[0487] 6. User Recommendations:

[0488] The server references a user profile database and recommends content to appropriate users based on their interests, past browsing history, and emotional feedback.

[0489] Example: A newly generated article by a ceramics expert is recommended to users who are interested in ceramics.

[0490] Terminal handling

[0491] 1. Notification of Interview Participation:

[0492] Once selected, the device will notify experts of the interview date and time and how to participate, via email and in-app notification.

[0493] Example: "Your interview is scheduled for 2023-10-01 15:00:00"

[0494] 2. Interview phase begins:

[0495] At the appointed date and time, the expert will initiate a session on the device and connect with the AI ​​interviewer in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[0496] Hardware and software used: Webcam, microphone, emotion analysis software

[0497] 3. Viewing and Feedback:

[0498] Users can view the profiles and articles generated through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[0499] Example: "This article was very helpful."

[0500] User Action

[0501] 1. Interview Participation:

[0502] If the user participates in the interview as an expert, they will check the interview details based on the notification from the device, and at the designated date and time, they will participate in the interview phase through the device.

[0503] Example: A user completes an interview at a specified date and time.

[0504] 2. Viewing profiles and articles:

[0505] Users browse recommended profiles and articles that interest them through their devices, and the emotion engine analyzes their emotions in real time and sends emotional data as feedback.

[0506] Example: "I was reading an article by a ceramics expert and it was very interesting."

[0507] 3. Feedback and sharing:

[0508] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0509] Example prompts: "What would a movie-loving user like to read next?", "Recommendations for movie-related interviews."

[0510] The above is a specific embodiment for carrying out the invention. This invention makes it possible to widely share the knowledge and experience of experts and to effectively recommend content based on the interests and emotions of users.

[0511] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0512] Step 1:

[0513] The server generates the interview questions

[0514] The server uses a generative model (e.g., a natural language processing model such as BERT) to generate appropriate questions for the expert being interviewed. In this process, the generative model receives a prompt (e.g., "Which other film directors have influenced this film director?") as input and generates a question as output. Specifically, the generative model analyzes the prompt, extracts relevant information from a database, and generates a new question.

[0515] Step 2:

[0516] The server schedules the interviews

[0517] The server arranges the interview date and time with the expert and saves the schedule information in a database. The expert's free time and the server's free time are referenced as input, and the optimal date and time is automatically selected. The schedule information is then saved in the database and is also output as notification data. Specifically, the date and time are set using the Python datetime module, and the process is performed to save it in a database such as PostgreSQL.

[0518] Step 3:

[0519] The server begins the interview phase

[0520] When the interview date and time arrives, the server starts the interview phase and connects the expert with the AI ​​interviewer. At this time, the generative model automatically asks questions and collects the expert's answers in real time. The emotion engine also runs and analyzes the user's emotional data. The input is the interview questions and the expert's answers, and the output is the generated answer data. Specifically, the AI ​​interviewer conducts the interview, and the exchange is collected as text data.

[0521] Step 4:

[0522] The server collects and analyzes emotion data

[0523] During the interview phase, the server uses an emotion engine to analyze emotional data such as the user's facial expressions, tone of voice, and selected survey answers. The input is the user's real-time reaction data, and the output is analyzed emotional evaluation data. Specifically, emotion analysis software (e.g., facial expression analysis API or voice emotion analysis API) is used to analyze the emotional data in real time, and the results are stored on the server.

[0524] Step 5:

[0525] Server generates profiles and articles

[0526] Based on the interview and sentiment analysis data, the server uses a generative model to generate expert profiles and articles. These profiles and articles organize expertise and experience from multiple angles and also include sentiment data. The input is the interview response data and sentiment analysis data, and the output is the generated profile and article. Specifically, a natural language generation model (e.g., GPT-3) is used to convert the data into text and create expert profiles and articles.

[0527] Step 6:

[0528] The server recommends content to the user

[0529] The server references a user profile database and recommends content to appropriate users based on their interests, past browsing history, and emotional feedback. The input is profile information from the user database, past browsing history, and emotional feedback, and the output is recommended content. Specifically, a machine learning model is used to analyze the user's past behavior and feedback and recommend the next content to be viewed.

[0530] 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.

[0531] 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.

[0532] 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.

[0533] [Second embodiment]

[0534] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0535] 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.

[0536] 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).

[0537] 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.

[0538] 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.

[0539] 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).

[0540] 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.

[0541] 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.

[0542] 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.

[0543] 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.

[0544] 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.

[0545] 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."

[0546] This invention is a system for sharing and spreading the knowledge and experience of experts using generative models. The purpose of this system is to support the discovery of new hobbies and the deepening of expertise, thereby enabling their knowledge to be widely disseminated. Specific embodiments are described below.

[0547] Server Processing

[0548] 1. Generate interview questions:

[0549] The server uses the generative model to automatically generate interview questions for specialists, for example, using machine learning algorithms to create optimal questions from a training database.

[0550] 2. Interview scheduling and management:

[0551] The server coordinates the interview date and time with the specialist to be interviewed and stores the schedule information in a database, thereby managing the specialist to participate in the interview at an appropriate time.

[0552] 3. Conducting the interview:

[0553] When the interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer in real time, with the generative model automatically asking questions and collecting the specialist's answers.

[0554] 4. Generate profiles and articles:

[0555] Based on the collected interview data, the server uses a generative model to generate specialist profiles and articles, which systematically organize the specialist's knowledge and experience.

[0556] 5. User Recommendations:

[0557] The server references the user profile database and recommends content to appropriate users based on the collected user interests and past browsing history. For example, a newly generated article about a pottery expert will be recommended to users who are interested in pottery.

[0558] Terminal handling

[0559] 1. Notification of Interview Participation:

[0560] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, and the specialist will be able to view detailed interview information.

[0561] 2. Interview phase begins:

[0562] When a specialist joins an interview, the device initiates the interview session and connects them in real time with an AI interviewer, who then communicates using voice and text.

[0563] 3. Viewing and Feedback:

[0564] Users can view the generated profiles and articles through their devices and can provide feedback on their impressions while viewing, which is then sent to the server via their devices.

[0565] User Action

[0566] 1. Interview Participation:

[0567] If the user participates in the interview as a specialist, they will check the interview details based on the notification from the device, and at the specified date and time, they will participate in the interview phase through the device.

[0568] 2. Viewing profiles and articles:

[0569] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0570] 3. Feedback and sharing:

[0571] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0572] Specific examples

[0573] For example, if film director A is the subject of an interview, the process will proceed as follows:

[0574] 1. The server generates a question for the film director: "Which directors influenced you?"

[0575] 2. Arrange an interview date and time and send a notification to Director A.

[0576] 3. On the specified date and time, Director A will participate in the interview session via the terminal.

[0577] 4. The server runs the interview in real time and collects the response data.

[0578] 5. Generate a profile and article about film director A based on the collected data.

[0579] 6. Recommend generated articles to users who are interested in movies.

[0580] 7. Movie fan users can view the generated article on their devices and provide feedback on their impressions.

[0581] In this way, a system is created in which the knowledge and experience of specialists is systematically organized and shared with many users.

[0582] The processing flow will be explained below.

[0583] Server Processing Steps

[0584] Step 1:

[0585] The server uses the generative model to generate interview questions for the specialist in question, such as "Which directors influenced you?"

[0586] Step 2:

[0587] The server coordinates interview dates and times with the interviewed specialists and stores the schedule information in a database, where it is maintained for future reference.

[0588] Step 3:

[0589] When the scheduled interview date and time arrives, the server starts the interview phase, connecting the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time.

[0590] Step 4:

[0591] The server analyzes the collected interview data and uses generative models to generate specialist profiles and articles, systematically organizing the collected data into knowledge and experience.

[0592] Step 5:

[0593] The server references the user profile database and recommends content to appropriate users based on their interests and past browsing history. For example, a newly generated article about a ceramics expert will be recommended to users who are interested in ceramics.

[0594] Terminal processing steps

[0595] Step 1:

[0596] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, and the specialist will be able to view detailed interview information.

[0597] Step 2:

[0598] At the appointed date and time, when the specialist joins the interview, the device will initiate the interview session and connect with an AI interviewer in real time, engaging in a dialogue using voice and text.

[0599] Step 3:

[0600] Users can view the generated profiles and articles through their devices and provide feedback based on their impressions of the views, which is then sent to the server via their devices.

[0601] User processing steps

[0602] Step 1:

[0603] If the user participates in the interview as a specialist, they will check the interview details based on the notification from the device, and at the specified date and time, they will participate in the interview phase through the device.

[0604] Step 2:

[0605] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0606] Step 3:

[0607] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0608] Specific examples

[0609] The processing flow when film director A is the subject of an interview is as follows:

[0610] Server Processing Steps

[0611] Step 1:

[0612] Generate a question for film director A: "Which directors influenced you?"

[0613] Step 2:

[0614] Arrange an interview date and time and send a notification to Director A.

[0615] Step 3:

[0616] At the specified date and time, film director A will participate in the interview session via the terminal.

[0617] Step 4:

[0618] Conduct interviews and collect response data in real time.

[0619] Step 5:

[0620] Based on the collected data, a profile and article about film director A is generated.

[0621] Step 6:

[0622] Recommend generated articles to users who are interested in movies.

[0623] Terminal processing steps

[0624] Step 1:

[0625] Notify Director A of the interview date and time and how to participate.

[0626] Step 2:

[0627] Film director A starts an interview session using a terminal to participate in the interview.

[0628] Step 3:

[0629] Users can view the generated articles through their devices and provide feedback.

[0630] User processing steps

[0631] Step 1:

[0632] Director A will confirm the interview details based on the notification.

[0633] Step 2:

[0634] The user views an article about recommended film director A.

[0635] Step 3:

[0636] Users provide and share feedback on articles.

[0637] Example 1

[0638] 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."

[0639] There is a growing demand for systems that can efficiently collect the knowledge and experience of experts and provide it to a wide range of users. However, current systems face challenges in effectively managing and conducting interviews with experts and recommending appropriate content based on each user's interests and browsing history.

[0640] 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.

[0641] In this invention, the server includes a means for generating interview questions using a generative model, a means for connecting an AI interviewer to the expert being interviewed using real-time communication means to conduct the interview, a means for analyzing the collected interview data to generate expert profiles and articles, and a means for recommending the generated content to appropriate users based on the user's interests and past browsing history. This makes it possible to effectively collect and organize the knowledge and experience of experts and provide personalized information to users.

[0642] A "generative model" is a method or system that uses machine learning algorithms to generate new data based on existing data.

[0643] "Interview questions" are questions created using generative models to elicit information or knowledge from experts or subjects.

[0644] An "expert" is a person who has advanced knowledge and experience in a particular field and has deep understanding and skills in that field.

[0645] "AI Interviewer" is a system that uses artificial intelligence to automatically conduct interviews and collect necessary information.

[0646] "Real-time communications means" are technologies and systems that allow information to be exchanged over the Internet or other communications networks with little or no time delay.

[0647] A "profile" is a document or data that systematically organizes an expert's knowledge and experience and shows the individual's characteristics and activities.

[0648] An "article" is a piece of writing that is created based on interview results and expert knowledge and is intended to provide information to a large number of users.

[0649] "User" means a general user who uses the system to view content and provide feedback.

[0650] "Interests" are the interest or curiosity a user shows in a particular field or topic.

[0651] "Browsing history" is a record of the content a user has viewed within the system.

[0652] "Recommendation" is the act of presenting optimal content based on a user's interests and browsing history.

[0653] This invention is a system for sharing and spreading expert knowledge and experience using generative AI models, with the aim of supporting the discovery of new hobbies, deepening of expertise, and spreading expert knowledge widely.

[0654] Server Processing

[0655] Generate interview questions

[0656] The server generates interview questions using a generative AI model (e.g., GPT-4). The server retrieves relevant keywords and past questions from a database of past interviews, and inputs prompt sentences based on these into the generative AI model. This automatically generates optimal questions for the expert. The following is a specific example.

[0657] Example prompt sentence:

[0658] "Generate interview questions for film director A. I'd like to know which directors influenced him."

[0659] Scheduling and managing interviews

[0660] The server retrieves the specialist's schedule information from the API and adjusts the interview date and time. The adjusted schedule information is saved in the database and an email or in-app notification is sent to the specialist. This series of operations ensures that the specialist can attend the interview at the appropriate time.

[0661] Running the Interview

[0662] When the interview date and time arrives, the server will start the interview session using a real-time communication method such as Zoom or Teams. The server will connect the AI ​​interviewer with the expert and conduct the interview in real time by presenting generated questions. Dialogue during the interview can include both voice and text.

[0663] Profile and Article Generation

[0664] Once the interview is complete, the server inputs the collected data into a generative AI model to generate a profile and article about the expert. These profiles and articles systematically describe the expert's knowledge and experience. The server then stores this data in a database.

[0665] User recommendations

[0666] The server references a user profile database and recommends the most appropriate content based on the user's interests and past browsing history. For example, a user interested in pottery might be recommended a newly generated interview article with a pottery expert.

[0667] Terminal handling

[0668] Notification of interview participation

[0669] Experts who are selected for an interview will be notified of the interview date and time and how to participate via email and in-app notification, and they can view detailed interview information.

[0670] Beginning of the interview phase

[0671] When the interview date and time arrives, the expert participates in the interview phase via a device, which connects to an AI interviewer in real time and begins the interview using voice and text.

[0672] Views and Feedback

[0673] Users can view the generated profiles and articles through their devices, and can also provide feedback on their impressions while viewing, which is then sent to the server via their devices.

[0674] User Action

[0675] Participating in an interview

[0676] If the user participates in the interview as an expert, he / she will check the interview details based on the notification from the device, and at the specified date and time, he / she will participate in the interview phase through the device.

[0677] View Profiles and Articles

[0678] Users can browse interesting profiles and articles recommended by the server through their devices. For example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0679] Feedback and sharing

[0680] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0681] As a result, this system can effectively collect and organize the knowledge and experience of experts, and provide personalized information to users.

[0682] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0683] Step 1:

[0684] The server generates interview questions using a generative AI model. In this step, it takes keywords obtained from past interview records and related databases as input and generates a prompt based on them. The prompt is sent to a generative AI model (e.g., GPT-4) to generate a new interview question. This question is output in the form of, for example, "What kind of movie would you like to make next?" Specifically, the server accesses the database, extracts relevant data, converts it into a prompt, sends it to the generative AI model, and stores the generated question in the database.

[0685] Step 2:

[0686] The server schedules an interview with a specialist. First, it accesses the specialist's calendar API to obtain available time slots. Based on the obtained schedule data, it confirms the interview date and time and saves it in the database. It also sends the confirmed date and time and participation instructions to the specialist via email and in-app notification. The input is the specialist's available schedule, and the output is the confirmed interview date. Specific operations include checking the schedule via the API, algorithmic processing to adjust the date and time, updating the database, and sending notifications.

[0687] Step 3:

[0688] When the interview date and time arrives, the server starts the interview phase using a real-time communication tool (e.g., Zoom, Teams). The server provides pre-generated interview questions to the AI ​​interviewer, which then presents the questions to the expert sequentially. The input is the interview date and time and the generated questions, and the output is the expert's response data. Specific operations include creating a meeting by calling Zoom's API, providing the generated questions sequentially, and collecting the expert's responses.

[0689] Step 4:

[0690] The server generates profiles and articles based on the collected interview data. The collected response data is input into a generative AI model to generate expert profiles and articles. The input for this process is the collected response data, and the output is the generated profile and article. Specific operations include text analysis of the collected data, generation of profiles and articles by the generative AI model, and storage in a database.

[0691] Step 5:

[0692] The server recommends the generated profiles and articles to users. First, it references the user profile database to obtain the user's interests and past browsing history. Based on this data, it selects highly relevant content and recommends it to the target user. The input is the user's interests and browsing history, and the output is the recommended content. Specific operations include database query, content selection using a recommendation algorithm, and notification and email sending.

[0693] Step 6:

[0694] The device notifies the selected expert of the interview date and time and how to participate. The device provides detailed interview information to the expert based on the schedule information obtained from the server. The input is the schedule information from the server, and the output is a notification to the expert. Specific operations include generating notifications and sending notifications via email or within the app.

[0695] Step 7:

[0696] When the interview date arrives, the terminal helps the expert join the interview session. The expert operates the terminal to connect to the real-time communication tool and begins a conversation with the AI ​​interviewer. The input is the meeting link provided by the server, and the output is the start of the interview session. Specific operations include providing the meeting link and guiding the connection procedure.

[0697] Step 8:

[0698] Users can view the generated profiles and articles through their devices and provide their thoughts and feedback. When users enter their feedback, the data is sent to the server via their devices. The input is the user's feedback, and the output is sending the feedback to the server. Specific actions include entering feedback, pressing the send button, and sending data to the server.

[0699] Step 9:

[0700] When a user joins an interview, the device provides the user with detailed interview information and invites them to the interview session at the specified date and time. The input is the notified interview details, and the output is participation in the interview phase. Specific actions include operating the start button for the interview phase and connecting to a real-time communication tool.

[0701] Step 10:

[0702] Users browse recommended profiles and articles and share the content they find particularly interesting with other users or on social media. The input is the recommended content, and the output is the shared content. Specific actions include viewing the content, clicking the share button, and posting to social media.

[0703] (Application example 1)

[0704] 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."

[0705] This invention relates to a system for widely sharing and disseminating the knowledge and experience of experts, aiming to effectively deliver knowledge from a variety of fields, not just specific fields, to a wide range of users. Another objective is to create more interactive and fulfilling learning opportunities by distributing expert interviews in real time and providing an environment in which users can provide immediate feedback on the content. Conventional systems have limited the dissemination of knowledge due to the difficulty of efficiently collecting and organizing expert knowledge and delivering it to the appropriate users.

[0706] 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.

[0707] In this invention, the server includes a means for generating interview questions using a generative model, a means for connecting the interviewed expert with an AI interviewer to conduct the interview, a means for analyzing collected interview data to generate expert profiles and articles, a means for recommending generated content to appropriate users based on users' interests and past browsing history, a means for live and recorded broadcasts of interviews, and a means for collecting and managing feedback from users. This allows experts' knowledge and experience to be shared effectively in real time, and users can provide instant feedback. Therefore, the diffusion of knowledge and feedback to experts occurs in both directions, creating a more comprehensive knowledge sharing platform.

[0708] A "generative model" is an algorithm or method that can automatically create new information or patterns based on data.

[0709] "Means for generating interview questions" refers to a function that uses a generative model to automatically create appropriate questions for interviewees.

[0710] An "artificial intelligence interviewer" is a virtual interviewer that uses artificial intelligence technology to conduct interviews.

[0711] "Means for conducting interviews" refers to the function of connecting the interviewee with an artificial intelligence interviewer and conducting the interview.

[0712] "Means for analyzing collected interview data" refers to the function of analyzing data collected during the interview process and extracting useful information.

[0713] "Means for generating expert profiles and articles" refers to the function of creating expert profiles and related articles based on the analyzed interview data.

[0714] "Recommendation methods" refers to a function that recommends appropriate content to a specific user based on the user's interests and past browsing history.

[0715] "Means for live streaming of interviews" refers to a function for streaming the contents of interviews in real time.

[0716] "Means for distributing recorded interviews" refers to the function of distributing recorded interviews so that users can watch them later.

[0717] "Means for collecting and managing feedback" refers to the function of collecting opinions and feedback from users and managing them systematically.

[0718] In this invention, a specific embodiment of a system for sharing and disseminating expert knowledge and experience will be described.

[0719] System program generation

[0720] Server Processing

[0721] 1. Generate interview questions:

[0722] The server uses a generative AI model to automatically generate interview questions for experts. This generative AI model uses a language model such as GPT-3. The generated questions are stored in a database.

[0723] 2. Interview scheduling and management:

[0724] The server coordinates interview dates and times with experts and stores the information in a database using a database management system such as SQLite.

[0725] 3. Conducting the interview:

[0726] When the interview date and time arrives, the server starts the interview phase in real time, connecting the expert with the AI ​​interviewer using technologies such as WebRTC.

[0727] 4. Generate profiles and articles:

[0728] The collected interview data is analyzed and a generative AI model is used to generate expert profiles and articles, which are then stored in a database.

[0729] 5. Content Recommendations:

[0730] The server uses a user profile database to recommend appropriate content based on the user's interests and past browsing history. The recommendation engine uses collaborative filtering and content-based filtering algorithms.

[0731] Terminal handling

[0732] 1. Notification of Interview Participation:

[0733] The server notifies the interviewed expert of the date and time of the interview and how to participate via email or push notification.

[0734] 2. Interview phase begins:

[0735] At the time of the interview, the device initiates the interview session, connecting the AI ​​interviewer with the expert in real time, enabling an interactive dialogue using voice and text.

[0736] 3. Watch the live and recorded interviews:

[0737] Users can watch the live interview on their device, and even after the live broadcast has ended, they can watch the recorded footage later.

[0738] 4. Providing Feedback:

[0739] Users can provide feedback on interviews and articles they have watched from their devices, and this feedback is immediately sent to the server and managed.

[0740] Specific examples

[0741] For example, if an expert film director is the subject of an interview, the process will proceed as follows:

[0742] 1. The server generates a question for the film director: "Which directors influenced you?"

[0743] 2. Arrange an interview date and time and send a notification to the film director.

[0744] 3. At the appointed date and time, the film director will participate in a real-time interview via the terminal.

[0745] 4. The server runs the interview in real time and collects the response data.

[0746] 5. Based on the collected data, generative AI models are used to generate profiles and articles about film directors.

[0747] 6. Recommend the generated articles and profiles to movie fan users.

[0748] 7. Movie fans can view the article on their devices and provide feedback on their impressions.

[0749] Example prompt sentence:

[0750] Generate a professional profile of the film director. Based on interviews with him, include information about his biggest influences and the inspiration for his latest work.

[0751] In this way, a series of processes on the server and terminals will realize a system in which the knowledge and experience of experts can be efficiently collected, organized, shared, and delivered to many users.

[0752] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0753] Step 1:

[0754] Generate interview questions

[0755] The server uses a generative AI model (e.g., GPT-3) to automatically generate questions for the expert being interviewed based on the prompt. The input is the "expert's field" and the "question prompt." The generative AI model outputs an appropriate question (e.g., "Which director influenced you?") based on these inputs and stores it in a database. Specifically, the server sends the prompt to the generative AI model, formats the returned answer as a question, and stores it in the database.

[0756] Step 2:

[0757] Scheduling and managing interviews

[0758] The server contacts the expert and arranges the interview date and time. The inputs are the expert's contact information and available time slots. Based on this, the server determines the interview date and time and saves it in the database. Specifically, the server contacts the expert via email or in-app notification and records the agreed interview date and time in the database.

[0759] Step 3:

[0760] Running the Interview

[0761] When the time for the interview arrives, the server starts the interview phase in real time. The inputs are the question and expert information. The server connects the expert and the AI ​​interviewer using technologies such as WebRTC and collects the results. Specifically, it initiates a real-time voice or text dialogue and stores the interview content in a database.

[0762] Step 4:

[0763] Profile and Article Generation

[0764] The server analyzes the collected interview data and uses a generative AI model to generate expert profiles and articles. The inputs are "interview data" and "generative AI model." The interview data is preprocessed using a data analysis tool and then input into the generative AI model to obtain profiles and articles. This output data is stored in a database. Specifically, it organizes the interview data through natural language analysis and automatically generates profiles and articles based on it.

[0765] Step 5:

[0766] User recommendations

[0767] Based on the user's interests and past browsing history, the server recommends appropriate content to the user. The inputs are a user profile database and generated profiles and articles. A recommender system is built using collaborative filtering and content-based filtering algorithms to recommend appropriate content to the user. Specifically, it analyzes the user profile, identifies users with similar interests and related content, and presents it to the user.

[0768] Step 6:

[0769] Gathering feedback

[0770] Users can provide feedback on the content they have viewed. Inputs include "user feedback content" and "user ID." The device sends this feedback to the server, which stores it in a database. Specifically, the system receives feedback through the user interface and transfers it to the server as structured data.

[0771] This series of steps will create a system that efficiently collects, organizes, and shares expert knowledge and experience, and provides appropriate content based on user interests.

[0772] 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.

[0773] This invention is a system for sharing and spreading expert knowledge and experience using a generative model and an emotion engine. The system aims to promote the sharing of expert knowledge by recommending content generated based on the user's interests and emotions to appropriate users. Specific embodiments are described below.

[0774] Server Processing

[0775] 1. Generate interview questions:

[0776] The server uses the generative model to generate interview questions for the specialist in question, such as "Which film directors have influenced you?"

[0777] 2. Interview scheduling and management:

[0778] The server coordinates the interview date and time with the interviewee specialist and stores the schedule information in a database, which can be referenced later at the appropriate time.

[0779] 3. Conducting the interview:

[0780] When the interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time. At this time, the emotion engine also works to analyze the user's emotions.

[0781] 4. Emotional data collection and analysis:

[0782] The server uses an emotion engine to analyze user data collected during the interview in real time, assessing emotions based on data such as facial expressions, tone of voice, and survey responses selected.

[0783] 5. Profile and article generation:

[0784] Based on the interview and sentiment analysis data, the server uses a generative model to generate specialist profiles and articles that include a multifaceted organization of expertise and experience, as well as sentiment data.

[0785] 6. User Recommendations:

[0786] The server refers to the user profile database and recommends the generated content to the appropriate user based on the user's interests, past browsing history, and emotional feedback. For example, a newly generated article about a ceramics expert may be provided to a user who is interested in ceramics and has previously viewed related content and felt emotionally satisfied.

[0787] Terminal handling

[0788] 1. Notification of Interview Participation:

[0789] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and / or in-app notification, and the specialist will be able to view detailed interview information.

[0790] 2. Interview phase begins:

[0791] At the appointed date and time, the device initiates a session for the specialist to participate in the interview, connecting the device to the AI ​​interviewer in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[0792] 3. Viewing and Feedback:

[0793] Users can view the profiles and articles generated through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[0794] User Action

[0795] 1. Interview Participation:

[0796] If the user participates in the interview as a specialist, they will check the interview details based on the notification on their device, and at the designated date and time, they will participate in the interview phase through their device.

[0797] 2. Viewing profiles and articles:

[0798] Users can browse recommended profiles and articles on their devices. For example, an amateur pottery enthusiast can read an interview with a pottery expert. As the user reads the article, the emotion engine analyzes their emotions in real time and sends emotional data as feedback.

[0799] 3. Feedback and sharing:

[0800] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0801] Specific examples

[0802] For example, if film director B is the subject of an interview, the following process will be carried out.

[0803] 1. The server generates a question for the film director: "Which film directors have influenced you?"

[0804] 2. Arrange an interview date and time and send a notification to Director B.

[0805] 3. On the specified date and time, Director B will join the interview session via the terminal.

[0806] 4. The server conducts the interview in real time and collects the response data. At the same time, the emotion engine analyzes Director B's emotions.

[0807] 5. Generate a profile and article about film director B based on the collected data.

[0808] 6. Recommend generated articles to users who are interested in movies and who have had a positive emotional reaction to past related articles.

[0809] 7. Movie fan users view the generated articles on their devices and provide their impressions as feedback that is analyzed by the emotion engine.

[0810] In this way, a system that integrates an emotion engine allows experts' knowledge and experience to be shared more deeply and made available to a wider audience.

[0811] The processing flow will be explained below.

[0812] Server Processing Steps

[0813] Step 1:

[0814] The server uses the generative model to generate interview questions for the specialist in question, such as "Which film directors have influenced you?"

[0815] Step 2:

[0816] The server coordinates the interview date and time with the specialist to be interviewed and stores the schedule information in a database, thereby managing the specialist to participate in the interview at an appropriate time.

[0817] Step 3:

[0818] When the scheduled interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time. At the same time, the emotion engine also runs to collect the user's emotional data.

[0819] Step 4:

[0820] The server uses an emotion engine to analyze the user's emotional data collected during the interview in real time, assessing their emotions based on, for example, their facial expressions, tone of voice, and selected survey answers.

[0821] Step 5:

[0822] The server uses a generative model to generate specialist profiles and articles based on interview response data and sentiment data. These profiles and articles systematically organize knowledge and experience.

[0823] Step 6:

[0824] The server refers to a user profile database and recommends generated content to appropriate users based on their interests, past browsing history, and emotional feedback. For example, a newly generated article about a pottery expert will be provided to users who are interested in pottery and have previously viewed related content and had a positive emotional response.

[0825] Terminal processing steps

[0826] Step 1:

[0827] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, where the specialist can view detailed interview information.

[0828] Step 2:

[0829] At the designated date and time, the device will initiate the interview session and connect with the AI ​​interviewer in real time, while the emotion engine will simultaneously collect and analyze the user's emotional data.

[0830] Step 3:

[0831] Users can view the generated profiles and articles through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[0832] User processing steps

[0833] Step 1:

[0834] If the user participates in the interview as a specialist, they will check the interview details based on the notification on their device, and at the designated date and time, they will participate in the interview phase through their device.

[0835] Step 2:

[0836] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0837] Step 3:

[0838] Users can provide feedback on the content and share articles they find particularly interesting with other users or on social media, further spreading their expertise and interests. This feedback is analyzed by the sentiment engine and sent to the server.

[0839] Specific examples

[0840] For example, if film director B is the subject of an interview, the processing flow is as follows:

[0841] Server Processing Steps

[0842] Step 1:

[0843] Generate a question for Director B: "Which directors influenced you?"

[0844] Step 2:

[0845] The interview date and time is arranged and a notification is sent to Director B.

[0846] Step 3:

[0847] At the specified date and time, film director B will participate in the interview session via the terminal.

[0848] Step 4:

[0849] Conduct interviews in real time and collect response and sentiment data.

[0850] Step 5:

[0851] Based on the collected data, a profile and article about film director B is generated.

[0852] Step 6:

[0853] The generated articles are recommended to users who are interested in movies and who have a positive reaction based on emotional feedback.

[0854] Terminal processing steps

[0855] Step 1:

[0856] Notify Director B of the interview date and time and how to participate.

[0857] Step 2:

[0858] Director B begins the interview session using a device to participate in the interview. At this time, the emotion engine also starts to collect and analyze emotion data.

[0859] Step 3:

[0860] Users can view the generated articles on their devices and provide feedback, which is also analyzed by the emotion engine and sent to the server.

[0861] User processing steps

[0862] Step 1:

[0863] Film director B checks the interview details based on the notification.

[0864] Step 2:

[0865] The user views an article about recommended film director B.

[0866] Step 3:

[0867] Users provide feedback on articles and share it on social media, etc. This feedback is analyzed by the emotion engine and sent to the server.

[0868] Example 2

[0869] 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."

[0870] There is a need for a method to effectively share the knowledge and experience of experts and provide appropriate content based on users' interests and emotions. However, conventional methods have been unable to fully utilize the quality of interview content or users' emotional feedback, making it difficult to recommend content that is highly relevant to users. This has resulted in insufficient dissemination of expert knowledge and improvement of user satisfaction.

[0871] 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.

[0872] In this invention, the server includes means for generating interview questions using a generative model, means for connecting the interviewed expert with an AI interviewer to conduct the interview, means for collecting and analyzing user data during the interview using an emotion engine, means for analyzing the collected interview data and emotion data to generate expert profiles and articles, and means for recommending the generated content to appropriate users based on the user's interests, past browsing history, and emotion feedback. This makes it possible to effectively share and spread the knowledge and experience of experts and provide users with content that is highly relevant and satisfying.

[0873] A "generative model" is a machine learning-based model that generates new text or information based on given prompts and data.

[0874] The "means for generating interview questions" is a mechanism that uses a generative model to automatically create specific questions for the expert being interviewed.

[0875] The "means of conducting the interview" is a mechanism that connects the subject expert with an AI interviewer and conducts the actual interview.

[0876] The "emotion engine" is a system that analyzes data such as the user's facial expressions, tone of voice, and selected survey responses to evaluate the user's emotions.

[0877] "Means for collecting and analyzing user data" refers to a system for collecting data obtained from users during interviews and analyzing their emotions and interests based on that data.

[0878] "Interview data" refers collectively to expert responses and other related information collected through interviews.

[0879] The "means for generating profiles and articles" is a mechanism for generating expert profiles and related articles based on the collected interview data and sentiment data.

[0880] The "user profile database" is a database for storing and managing information such as user interests, past browsing history, and emotional feedback.

[0881] "Means for recommending content" refers to a system that recommends appropriate content to users based on their interests, past browsing history, and emotional feedback.

[0882] This invention is a system for sharing and spreading expert knowledge and experience by utilizing a generative model and an emotion engine. The system aims to promote the sharing of expert knowledge by recommending content generated based on the user's interests and emotions to appropriate users. A specific embodiment of this system is described below.

[0883] Server Processing

[0884] 1. Generate interview questions:

[0885] The server generates interview questions using a generative AI model (e.g., GPT-4). At this time, instructions such as "Create interview questions for a film director" are input as prompts. An example of a question generated by the generative model is "Which film directors have influenced you?"

[0886] 2. Interview scheduling and management:

[0887] The server searches for relevant experts based on the generated interview questions and schedules the interview. This information is stored in a database and is referenced when the interview is conducted.

[0888] 3. Conducting the interview:

[0889] When the time for the interview arrives, the server connects the expert to the AI ​​interviewer and conducts the interview. The generative model automatically asks questions and collects the expert's answers in real time. At the same time, the emotion engine runs to analyze the user's emotional data.

[0890] 4. Emotional data collection and analysis:

[0891] The server uses an emotion engine to collect and analyze data such as the user's facial expressions, tone of voice, and questionnaire responses during the interview to evaluate the user's emotions.

[0892] 5. Profile and article generation:

[0893] Based on the interview and sentiment analysis data, the server uses a generative AI model to generate expert profiles and articles that include a multifaceted breakdown of expertise and experience, as well as sentiment data.

[0894] 6. User Recommendations:

[0895] The server references a user profile database and recommends content based on the user's interests, past browsing history, and emotional feedback. For example, a user interested in pottery can be recommended a newly generated article by a pottery expert.

[0896] Terminal handling

[0897] 1. Notification of Interview Participation:

[0898] The device will notify the interviewed expert of the date and time of the interview and how to participate. This notification will be provided via email and in-app notification, and the expert will be able to view detailed interview information.

[0899] 2. Interview phase begins:

[0900] At the specified date and time, the device will automatically start the interview session and connect the AI ​​interviewer with the expert in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[0901] 3. Viewing and Feedback:

[0902] Users can view the generated profiles and articles through their devices and can also provide feedback based on their impressions while browsing, which is analyzed by the emotion engine and sent to the server.

[0903] User Action

[0904] 1. Interview Participation:

[0905] When participating in an interview as an expert, the user will check the interview details based on notifications from the device, and at the designated date and time, they will participate in the interview phase through the device.

[0906] 2. Viewing profiles and articles:

[0907] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[0908] 3. Feedback and sharing:

[0909] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[0910] Specific examples

[0911] For example, if Director B is chosen for an interview, the following process will occur:

[0912] 1. The server uses the generative model to generate a question for the film director: "Which film directors have influenced you?"

[0913] 2. Arrange an interview date and time and send a notification to Director B.

[0914] 3. On the specified date and time, Director B will join the interview session via the terminal.

[0915] 4. The server conducts the interview in real time and collects the response data. At the same time, the emotion engine analyzes Director B's emotions.

[0916] 5. Generate a profile and article about film director B based on the collected data.

[0917] 6. Recommend generated articles to users who are interested in movies and have had a positive emotional reaction to past related articles.

[0918] 7. Movie fan users view the generated articles on their devices and provide their impressions as feedback that is analyzed by the emotion engine.

[0919] This system integrates an emotion engine to share the knowledge and experience of experts widely and deeply, enabling it to provide highly relevant content to many users.

[0920] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0921] Step 1: Generate interview questions

[0922] Input: Interview request from user, prompt

[0923] Specific behavior:

[0924] The server receives a request from the user to "interview a film director." The server inputs the prompt "Create interview questions for the film director" into a generative AI model (e.g., GPT-4).

[0925] Data processing / calculation:

[0926] The generative AI model analyzes the prompt and generates new interview questions, referencing question generation patterns from an existing database.

[0927] Output: Generated interview questions (e.g., "Which film directors influenced you?")

[0928] Step 2: Schedule and manage interviews

[0929] Input: Interview questions, expert schedule data

[0930] Specific behavior:

[0931] The server searches the database for relevant experts (e.g., film directors) based on the generated question, checks the experts' schedules, and arranges the date and time of the interview.

[0932] Data processing / calculation:

[0933] It looks at the expert's calendar, automatically identifies available time slots, and selects the appropriate time.

[0934] Output: List of experts with interview date and time (e.g., Film Director B, October 1, 2023, 2:00 PM)

[0935] Step 3: Conducting the interview

[0936] Input: Interview questions, expert connections

[0937] Specific behavior:

[0938] When the time for the interview arrives, the server sends a notification to the expert (e.g., film director B) to start the interview. The server then starts the AI ​​interviewer and delivers the generated questions one by one.

[0939] Data processing / calculation:

[0940] The AI ​​interviewer collects the expert's answers in real time and stores them in a database, while the emotion engine collects facial and vocal data and analyzes the emotional data.

[0941] Output: Expert response data, sentiment data collected in real time

[0942] Step 4: Collect and analyze emotion data

[0943] Input: Expert response data, user sentiment data

[0944] Specific behavior:

[0945] During the interview, the server uses an emotion engine to collect and analyze emotional data from the user in real time, such as facial expressions, tone of voice, and questionnaire responses.

[0946] Data processing / calculation:

[0947] The collected data is evaluated using facial expression analysis software and tone analysis software to determine and quantify positive, negative, and neutral emotions.

[0948] Output: Analyzed sentiment data (e.g., 80% positive, 10% negative, 10% neutral)

[0949] Step 5: Generate profiles and articles

[0950] Input: Expert response data, analyzed sentiment data

[0951] Specific behavior:

[0952] The server instructs the generative AI model to generate expert profiles and articles based on the collected response data and sentiment data.

[0953] Data processing / calculation:

[0954] The generative model analyzes this data, organizes the knowledge and experience of experts from various perspectives, and converts them into article format, while also appropriately reflecting sentiment data.

[0955] Output: Expert profile and article (e.g., article about film director B)

[0956] Step 6: Recommend to users

[0957] Input: User interest data, past browsing history, emotional feedback

[0958] Specific behavior:

[0959] The server references a user profile database and selects content based on interests, past browsing history, and emotional feedback, and then notifies the selected content to the target user.

[0960] Data processing / calculation:

[0961] It uses recommendation algorithms to surface content that matches your interests, and also uses your browsing history and emotional feedback to select the most relevant articles.

[0962] Output: Content notification to user (e.g. "A new article about pottery has been published")

[0963] (Application example 2)

[0964] 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."

[0965] The lack of a system that widely shares the knowledge and experience of experts and appropriately recommends them based on user interests and emotions is another issue. Another issue is the lack of technology that effectively utilizes user feedback and emotional data to organize and provide expert content from multiple perspectives.

[0966] 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 generating interview questions using a generative model, means for connecting the interviewed expert with an AI interviewer and conducting the interview, means for analyzing the collected interview data to generate expert profiles and articles, means for recommending content generated based on the user's interests, past browsing history, and emotional feedback to appropriate users, and means for analyzing the user's feedback with an emotion engine. This makes it possible to effectively share and recommend experts' knowledge and experience based on the user's emotions and interests.

[0967] A "generative model" refers to an algorithm that automatically generates new content, questions, etc. based on data.

[0968] "Interview questions" are questions asked to experts, and these questions are generated using a generative model.

[0969] An "expert" is an individual who has knowledge or experience in a particular field.

[0970] "AI interviewer" refers to a system or software that uses artificial intelligence to ask interview questions and collect answers.

[0971] "Interview Data" refers to the experts' responses and other relevant information collected during the interview.

[0972] A "profile" is information that outlines an expert's knowledge and experience, created based on collected data.

[0973] "Articles" refer to written information created based on interview data, and are detailed descriptions of the experts' knowledge and experiences.

[0974] "Recommendation" refers to recommending appropriate content based on a user's interests, past browsing history, and emotional feedback.

[0975] "Emotional feedback" refers to the analysis of users' emotional reactions when viewing content and the data and comments based on that.

[0976] An "emotion engine" refers to software or algorithms that analyze a user's emotions from facial expressions, tone of voice, text feedback, etc.

[0977] "Content" refers to the collection of information, entertainment, etc. provided to users through communications means.

[0978] "System" refers to an integrated mechanical or software collection of related means and devices.

[0979] This invention is a system that shares the knowledge and experience of experts and recommends appropriate content based on the user's interests and emotions. The detailed configuration and operation procedures for implementing this invention are described below.

[0980] Server Processing

[0981] 1. Generate interview questions:

[0982] The server uses a generative model to generate interview questions for domain experts, such as a natural language processing model like BERT (Bidirectional Encoder Representations from Transformers).

[0983] Example: Generating questions for a film director such as "Which film directors have influenced you?"

[0984] 2. Interview scheduling and management:

[0985] The server coordinates the interview date and time with the interview subject and stores the schedule information in a database, using the Python datetime module for schedule management.

[0986] Example: An interview with a film director is scheduled for "2023-10-01 15:00:00".

[0987] 3. Conducting the interview:

[0988] When the interview date and time arrives, the server starts the interview phase and connects the expert with the AI ​​interviewer. The generative model automatically asks questions and collects the expert's answers in real time. At the same time, the emotion engine runs and analyzes the user's emotional data.

[0989] Hardware and software used: GPU-based servers, generative AI models (e.g., GPT-3), emotion analysis software (e.g., facial emotion recognition API)

[0990] 4. Emotional data collection and analysis:

[0991] The server analyzes user data collected during the interview in real time using an emotion engine, which evaluates emotions based on data such as the user's facial expressions, tone of voice, and selected survey answers.

[0992] Example: Determining user satisfaction and interest in an expert's answer in real time.

[0993] 5. Profile and article generation:

[0994] Based on the interview and sentiment analysis data, the server uses a generative model to generate expert profiles and articles that include a multifaceted organization of expertise and experience, as well as sentiment data.

[0995] Hardware and software used: Database management system (e.g., PostgreSQL), natural language generation model (e.g., OpenAI's GPT-3)

[0996] 6. User Recommendations:

[0997] The server references a user profile database and recommends content to appropriate users based on their interests, past browsing history, and emotional feedback.

[0998] Example: A newly generated article by a ceramics expert is recommended to users who are interested in ceramics.

[0999] Terminal handling

[1000] 1. Notification of Interview Participation:

[1001] Once selected, the device will notify experts of the interview date and time and how to participate, via email and in-app notification.

[1002] Example: "Your interview is scheduled for 2023-10-01 15:00:00"

[1003] 2. Interview phase begins:

[1004] At the appointed date and time, the expert will initiate a session on the device and connect with the AI ​​interviewer in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[1005] Hardware and software used: Webcam, microphone, emotion analysis software

[1006] 3. Viewing and Feedback:

[1007] Users can view the profiles and articles generated through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[1008] Example: "This article was very helpful."

[1009] User Action

[1010] 1. Interview Participation:

[1011] If the user participates in the interview as an expert, they will check the interview details based on the notification from the device, and at the designated date and time, they will participate in the interview phase through the device.

[1012] Example: A user completes an interview at a specified date and time.

[1013] 2. Viewing profiles and articles:

[1014] Users browse recommended profiles and articles that interest them through their devices, and the emotion engine analyzes their emotions in real time and sends emotional data as feedback.

[1015] Example: "I was reading an article by a ceramics expert and it was very interesting."

[1016] 3. Feedback and sharing:

[1017] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1018] Example prompts: "What would a movie-loving user like to read next?", "Recommendations for movie-related interviews."

[1019] The above is a specific embodiment for carrying out the invention. This invention makes it possible to widely share the knowledge and experience of experts and to effectively recommend content based on the interests and emotions of users.

[1020] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1021] Step 1:

[1022] The server generates the interview questions

[1023] The server uses a generative model (e.g., a natural language processing model such as BERT) to generate appropriate questions for the expert being interviewed. In this process, the generative model receives a prompt (e.g., "Which other film directors have influenced this film director?") as input and generates a question as output. Specifically, the generative model analyzes the prompt, extracts relevant information from a database, and generates a new question.

[1024] Step 2:

[1025] The server schedules the interviews

[1026] The server arranges the interview date and time with the expert and saves the schedule information in a database. The expert's free time and the server's free time are referenced as input, and the optimal date and time is automatically selected. The schedule information is then saved in the database and is also output as notification data. Specifically, the date and time are set using the Python datetime module, and the process is performed to save it in a database such as PostgreSQL.

[1027] Step 3:

[1028] The server begins the interview phase

[1029] When the interview date and time arrives, the server starts the interview phase and connects the expert with the AI ​​interviewer. At this time, the generative model automatically asks questions and collects the expert's answers in real time. The emotion engine also runs and analyzes the user's emotional data. The input is the interview questions and the expert's answers, and the output is the generated answer data. Specifically, the AI ​​interviewer conducts the interview, and the exchange is collected as text data.

[1030] Step 4:

[1031] The server collects and analyzes emotion data

[1032] During the interview phase, the server uses an emotion engine to analyze emotional data such as the user's facial expressions, tone of voice, and selected survey answers. The input is the user's real-time reaction data, and the output is analyzed emotional evaluation data. Specifically, emotion analysis software (e.g., facial expression analysis API or voice emotion analysis API) is used to analyze the emotional data in real time, and the results are stored on the server.

[1033] Step 5:

[1034] Server generates profiles and articles

[1035] Based on the interview and sentiment analysis data, the server uses a generative model to generate expert profiles and articles. These profiles and articles organize expertise and experience from multiple angles and also include sentiment data. The input is the interview response data and sentiment analysis data, and the output is the generated profile and article. Specifically, a natural language generation model (e.g., GPT-3) is used to convert the data into text and create expert profiles and articles.

[1036] Step 6:

[1037] The server recommends content to the user

[1038] The server references a user profile database and recommends content to appropriate users based on their interests, past browsing history, and emotional feedback. The input is profile information from the user database, past browsing history, and emotional feedback, and the output is recommended content. Specifically, a machine learning model is used to analyze the user's past behavior and feedback and recommend the next content to be viewed.

[1039] 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.

[1040] 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.

[1041] 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.

[1042] [Third embodiment]

[1043] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1044] 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.

[1045] 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).

[1046] 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.

[1047] 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.

[1048] 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).

[1049] 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.

[1050] 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.

[1051] 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.

[1052] 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.

[1053] 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.

[1054] 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."

[1055] This invention is a system for sharing and spreading the knowledge and experience of experts using generative models. The purpose of this system is to support the discovery of new hobbies and the deepening of expertise, thereby enabling their knowledge to be widely disseminated. Specific embodiments are described below.

[1056] Server Processing

[1057] 1. Generate interview questions:

[1058] The server uses the generative model to automatically generate interview questions for specialists, for example, using machine learning algorithms to create optimal questions from a training database.

[1059] 2. Interview scheduling and management:

[1060] The server coordinates the interview date and time with the specialist to be interviewed and stores the schedule information in a database, thereby managing the specialist to participate in the interview at an appropriate time.

[1061] 3. Conducting the interview:

[1062] When the interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer in real time, with the generative model automatically asking questions and collecting the specialist's answers.

[1063] 4. Generate profiles and articles:

[1064] Based on the collected interview data, the server uses a generative model to generate specialist profiles and articles, which systematically organize the specialist's knowledge and experience.

[1065] 5. User Recommendations:

[1066] The server references the user profile database and recommends content to appropriate users based on the collected user interests and past browsing history. For example, a newly generated article about a pottery expert will be recommended to users who are interested in pottery.

[1067] Terminal handling

[1068] 1. Notification of Interview Participation:

[1069] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, and the specialist will be able to view detailed interview information.

[1070] 2. Interview phase begins:

[1071] When a specialist joins an interview, the device initiates the interview session and connects them in real time with an AI interviewer, who then communicates using voice and text.

[1072] 3. Viewing and Feedback:

[1073] Users can view the generated profiles and articles through their devices and can provide feedback on their impressions while viewing, which is then sent to the server via their devices.

[1074] User Action

[1075] 1. Interview Participation:

[1076] If the user participates in the interview as a specialist, they will check the interview details based on the notification from the device, and at the specified date and time, they will participate in the interview phase through the device.

[1077] 2. Viewing profiles and articles:

[1078] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1079] 3. Feedback and sharing:

[1080] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1081] Specific examples

[1082] For example, if film director A is the subject of an interview, the process will proceed as follows:

[1083] 1. The server generates a question for the film director: "Which directors influenced you?"

[1084] 2. Arrange an interview date and time and send a notification to Director A.

[1085] 3. On the specified date and time, Director A will participate in the interview session via the terminal.

[1086] 4. The server runs the interview in real time and collects the response data.

[1087] 5. Generate a profile and article about film director A based on the collected data.

[1088] 6. Recommend generated articles to users who are interested in movies.

[1089] 7. Movie fan users can view the generated article on their devices and provide feedback on their impressions.

[1090] In this way, a system is created in which the knowledge and experience of specialists is systematically organized and shared with many users.

[1091] The processing flow will be explained below.

[1092] Server Processing Steps

[1093] Step 1:

[1094] The server uses the generative model to generate interview questions for the specialist in question, such as "Which directors influenced you?"

[1095] Step 2:

[1096] The server coordinates interview dates and times with the interviewed specialists and stores the schedule information in a database, where it is maintained for future reference.

[1097] Step 3:

[1098] When the scheduled interview date and time arrives, the server starts the interview phase, connecting the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time.

[1099] Step 4:

[1100] The server analyzes the collected interview data and uses generative models to generate specialist profiles and articles, systematically organizing the collected data into knowledge and experience.

[1101] Step 5:

[1102] The server references the user profile database and recommends content to appropriate users based on their interests and past browsing history. For example, a newly generated article about a ceramics expert will be recommended to users who are interested in ceramics.

[1103] Terminal processing steps

[1104] Step 1:

[1105] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, and the specialist will be able to view detailed interview information.

[1106] Step 2:

[1107] At the appointed date and time, when the specialist joins the interview, the device will initiate the interview session and connect with an AI interviewer in real time, engaging in a dialogue using voice and text.

[1108] Step 3:

[1109] Users can view the generated profiles and articles through their devices and provide feedback based on their impressions of the views, which is then sent to the server via their devices.

[1110] User processing steps

[1111] Step 1:

[1112] If the user participates in the interview as a specialist, they will check the interview details based on the notification from the device, and at the specified date and time, they will participate in the interview phase through the device.

[1113] Step 2:

[1114] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1115] Step 3:

[1116] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1117] Specific examples

[1118] The processing flow when film director A is the subject of an interview is as follows:

[1119] Server Processing Steps

[1120] Step 1:

[1121] Generate a question for film director A: "Which directors influenced you?"

[1122] Step 2:

[1123] Arrange an interview date and time and send a notification to Director A.

[1124] Step 3:

[1125] At the specified date and time, film director A will participate in the interview session via the terminal.

[1126] Step 4:

[1127] Conduct interviews and collect response data in real time.

[1128] Step 5:

[1129] Based on the collected data, a profile and article about film director A is generated.

[1130] Step 6:

[1131] Recommend generated articles to users who are interested in movies.

[1132] Terminal processing steps

[1133] Step 1:

[1134] Notify Director A of the interview date and time and how to participate.

[1135] Step 2:

[1136] Film director A starts an interview session using a terminal to participate in the interview.

[1137] Step 3:

[1138] Users can view the generated articles through their devices and provide feedback.

[1139] User processing steps

[1140] Step 1:

[1141] Director A will confirm the interview details based on the notification.

[1142] Step 2:

[1143] The user views an article about recommended film director A.

[1144] Step 3:

[1145] Users provide and share feedback on articles.

[1146] Example 1

[1147] 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."

[1148] There is a growing demand for systems that can efficiently collect the knowledge and experience of experts and provide it to a wide range of users. However, current systems face challenges in effectively managing and conducting interviews with experts and recommending appropriate content based on each user's interests and browsing history.

[1149] 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.

[1150] In this invention, the server includes a means for generating interview questions using a generative model, a means for connecting an AI interviewer to the expert being interviewed using real-time communication means to conduct the interview, a means for analyzing the collected interview data to generate expert profiles and articles, and a means for recommending the generated content to appropriate users based on the user's interests and past browsing history. This makes it possible to effectively collect and organize the knowledge and experience of experts and provide personalized information to users.

[1151] A "generative model" is a method or system that uses machine learning algorithms to generate new data based on existing data.

[1152] "Interview questions" are questions created using generative models to elicit information or knowledge from experts or subjects.

[1153] An "expert" is a person who has advanced knowledge and experience in a particular field and has deep understanding and skills in that field.

[1154] "AI Interviewer" is a system that uses artificial intelligence to automatically conduct interviews and collect necessary information.

[1155] "Real-time communications means" are technologies and systems that allow information to be exchanged over the Internet or other communications networks with little or no time delay.

[1156] A "profile" is a document or data that systematically organizes an expert's knowledge and experience and shows the individual's characteristics and activities.

[1157] An "article" is a piece of writing that is created based on interview results and expert knowledge and is intended to provide information to a large number of users.

[1158] "User" means a general user who uses the system to view content and provide feedback.

[1159] "Interests" are the interest or curiosity a user shows in a particular field or topic.

[1160] "Browsing history" is a record of the content a user has viewed within the system.

[1161] "Recommendation" is the act of presenting optimal content based on a user's interests and browsing history.

[1162] This invention is a system for sharing and spreading expert knowledge and experience using generative AI models, with the aim of supporting the discovery of new hobbies, deepening of expertise, and spreading expert knowledge widely.

[1163] Server Processing

[1164] Generate interview questions

[1165] The server generates interview questions using a generative AI model (e.g., GPT-4). The server retrieves relevant keywords and past questions from a database of past interviews, and inputs prompt sentences based on these into the generative AI model. This automatically generates optimal questions for the expert. The following is a specific example.

[1166] Example prompt sentence:

[1167] "Generate interview questions for film director A. I'd like to know which directors influenced him."

[1168] Scheduling and managing interviews

[1169] The server retrieves the specialist's schedule information from the API and adjusts the interview date and time. The adjusted schedule information is saved in the database and an email or in-app notification is sent to the specialist. This series of operations ensures that the specialist can attend the interview at the appropriate time.

[1170] Running the Interview

[1171] When the interview date and time arrives, the server will start the interview session using a real-time communication method such as Zoom or Teams. The server will connect the AI ​​interviewer with the expert and conduct the interview in real time by presenting generated questions. Dialogue during the interview can include both voice and text.

[1172] Profile and Article Generation

[1173] Once the interview is complete, the server inputs the collected data into a generative AI model to generate a profile and article about the expert. These profiles and articles systematically describe the expert's knowledge and experience. The server then stores this data in a database.

[1174] User recommendations

[1175] The server references a user profile database and recommends the most appropriate content based on the user's interests and past browsing history. For example, a user interested in pottery might be recommended a newly generated interview article with a pottery expert.

[1176] Terminal handling

[1177] Notification of interview participation

[1178] Experts who are selected for an interview will be notified of the interview date and time and how to participate via email and in-app notification, and they can view detailed interview information.

[1179] Beginning of the interview phase

[1180] When the interview date and time arrives, the expert participates in the interview phase via a device, which connects to an AI interviewer in real time and begins the interview using voice and text.

[1181] Views and Feedback

[1182] Users can view the generated profiles and articles through their devices, and can also provide feedback on their impressions while viewing, which is then sent to the server via their devices.

[1183] User Action

[1184] Participating in an interview

[1185] If the user participates in the interview as an expert, he / she will check the interview details based on the notification from the device, and at the specified date and time, he / she will participate in the interview phase through the device.

[1186] View Profiles and Articles

[1187] Users can browse interesting profiles and articles recommended by the server through their devices. For example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1188] Feedback and sharing

[1189] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1190] As a result, this system can effectively collect and organize the knowledge and experience of experts, and provide personalized information to users.

[1191] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1192] Step 1:

[1193] The server generates interview questions using a generative AI model. In this step, it takes keywords obtained from past interview records and related databases as input and generates a prompt based on them. The prompt is sent to a generative AI model (e.g., GPT-4) to generate a new interview question. This question is output in the form of, for example, "What kind of movie would you like to make next?" Specifically, the server accesses the database, extracts relevant data, converts it into a prompt, sends it to the generative AI model, and stores the generated question in the database.

[1194] Step 2:

[1195] The server schedules an interview with a specialist. First, it accesses the specialist's calendar API to obtain available time slots. Based on the obtained schedule data, it confirms the interview date and time and saves it in the database. It also sends the confirmed date and time and participation instructions to the specialist via email and in-app notification. The input is the specialist's available schedule, and the output is the confirmed interview date. Specific operations include checking the schedule via the API, algorithmic processing to adjust the date and time, updating the database, and sending notifications.

[1196] Step 3:

[1197] When the interview date and time arrives, the server starts the interview phase using a real-time communication tool (e.g., Zoom, Teams). The server provides pre-generated interview questions to the AI ​​interviewer, which then presents the questions to the expert sequentially. The input is the interview date and time and the generated questions, and the output is the expert's response data. Specific operations include creating a meeting by calling Zoom's API, providing the generated questions sequentially, and collecting the expert's responses.

[1198] Step 4:

[1199] The server generates profiles and articles based on the collected interview data. The collected response data is input into a generative AI model to generate expert profiles and articles. The input for this process is the collected response data, and the output is the generated profile and article. Specific operations include text analysis of the collected data, generation of profiles and articles by the generative AI model, and storage in a database.

[1200] Step 5:

[1201] The server recommends the generated profiles and articles to users. First, it references the user profile database to obtain the user's interests and past browsing history. Based on this data, it selects highly relevant content and recommends it to the target user. The input is the user's interests and browsing history, and the output is the recommended content. Specific operations include database query, content selection using a recommendation algorithm, and notification and email sending.

[1202] Step 6:

[1203] The device notifies the selected expert of the interview date and time and how to participate. The device provides detailed interview information to the expert based on the schedule information obtained from the server. The input is the schedule information from the server, and the output is a notification to the expert. Specific operations include generating notifications and sending notifications via email or within the app.

[1204] Step 7:

[1205] When the interview date arrives, the terminal helps the expert join the interview session. The expert operates the terminal to connect to the real-time communication tool and begins a conversation with the AI ​​interviewer. The input is the meeting link provided by the server, and the output is the start of the interview session. Specific operations include providing the meeting link and guiding the connection procedure.

[1206] Step 8:

[1207] Users can view the generated profiles and articles through their devices and provide their thoughts and feedback. When users enter their feedback, the data is sent to the server via their devices. The input is the user's feedback, and the output is sending the feedback to the server. Specific actions include entering feedback, pressing the send button, and sending data to the server.

[1208] Step 9:

[1209] When a user joins an interview, the device provides the user with detailed interview information and invites them to the interview session at the specified date and time. The input is the notified interview details, and the output is participation in the interview phase. Specific actions include operating the start button for the interview phase and connecting to a real-time communication tool.

[1210] Step 10:

[1211] Users browse recommended profiles and articles and share the content they find particularly interesting with other users or on social media. The input is the recommended content, and the output is the shared content. Specific actions include viewing the content, clicking the share button, and posting to social media.

[1212] (Application example 1)

[1213] 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."

[1214] This invention relates to a system for widely sharing and disseminating the knowledge and experience of experts, aiming to effectively deliver knowledge from a variety of fields, not just specific fields, to a wide range of users. Another objective is to create more interactive and fulfilling learning opportunities by distributing expert interviews in real time and providing an environment in which users can provide immediate feedback on the content. Conventional systems have limited the dissemination of knowledge due to the difficulty of efficiently collecting and organizing expert knowledge and delivering it to the appropriate users.

[1215] 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.

[1216] In this invention, the server includes a means for generating interview questions using a generative model, a means for connecting the interviewed expert with an AI interviewer to conduct the interview, a means for analyzing collected interview data to generate expert profiles and articles, a means for recommending generated content to appropriate users based on users' interests and past browsing history, a means for live and recorded broadcasts of interviews, and a means for collecting and managing feedback from users. This allows experts' knowledge and experience to be shared effectively in real time, and users can provide instant feedback. Therefore, the diffusion of knowledge and feedback to experts occurs in both directions, creating a more comprehensive knowledge sharing platform.

[1217] A "generative model" is an algorithm or method that can automatically create new information or patterns based on data.

[1218] "Means for generating interview questions" refers to a function that uses a generative model to automatically create appropriate questions for interviewees.

[1219] An "artificial intelligence interviewer" is a virtual interviewer that uses artificial intelligence technology to conduct interviews.

[1220] "Means for conducting interviews" refers to the function of connecting the interviewee with an artificial intelligence interviewer and conducting the interview.

[1221] "Means for analyzing collected interview data" refers to the function of analyzing data collected during the interview process and extracting useful information.

[1222] "Means for generating expert profiles and articles" refers to the function of creating expert profiles and related articles based on the analyzed interview data.

[1223] "Recommendation methods" refers to a function that recommends appropriate content to a specific user based on the user's interests and past browsing history.

[1224] "Means for live streaming of interviews" refers to a function for streaming the contents of interviews in real time.

[1225] "Means for distributing recorded interviews" refers to the function of distributing recorded interviews so that users can watch them later.

[1226] "Means for collecting and managing feedback" refers to the function of collecting opinions and feedback from users and managing them systematically.

[1227] In this invention, a specific embodiment of a system for sharing and disseminating expert knowledge and experience will be described.

[1228] System program generation

[1229] Server Processing

[1230] 1. Generate interview questions:

[1231] The server uses a generative AI model to automatically generate interview questions for experts. This generative AI model uses a language model such as GPT-3. The generated questions are stored in a database.

[1232] 2. Interview scheduling and management:

[1233] The server coordinates interview dates and times with experts and stores the information in a database using a database management system such as SQLite.

[1234] 3. Conducting the interview:

[1235] When the interview date and time arrives, the server starts the interview phase in real time, connecting the expert with the AI ​​interviewer using technologies such as WebRTC.

[1236] 4. Generate profiles and articles:

[1237] The collected interview data is analyzed and a generative AI model is used to generate expert profiles and articles, which are then stored in a database.

[1238] 5. Content Recommendations:

[1239] The server uses a user profile database to recommend appropriate content based on the user's interests and past browsing history. The recommendation engine uses collaborative filtering and content-based filtering algorithms.

[1240] Terminal handling

[1241] 1. Notification of Interview Participation:

[1242] The server notifies the interviewed expert of the date and time of the interview and how to participate via email or push notification.

[1243] 2. Interview phase begins:

[1244] At the time of the interview, the device initiates the interview session, connecting the AI ​​interviewer with the expert in real time, enabling an interactive dialogue using voice and text.

[1245] 3. Watch the live and recorded interviews:

[1246] Users can watch the live interview on their device, and even after the live broadcast has ended, they can watch the recorded footage later.

[1247] 4. Providing Feedback:

[1248] Users can provide feedback on interviews and articles they have watched from their devices, and this feedback is immediately sent to the server and managed.

[1249] Specific examples

[1250] For example, if an expert film director is the subject of an interview, the process will proceed as follows:

[1251] 1. The server generates a question for the film director: "Which directors influenced you?"

[1252] 2. Arrange an interview date and time and send a notification to the film director.

[1253] 3. At the appointed date and time, the film director will participate in a real-time interview via the terminal.

[1254] 4. The server runs the interview in real time and collects the response data.

[1255] 5. Based on the collected data, generative AI models are used to generate profiles and articles about film directors.

[1256] 6. Recommend the generated articles and profiles to movie fan users.

[1257] 7. Movie fans can view the article on their devices and provide feedback on their impressions.

[1258] Example prompt sentence:

[1259] Generate a professional profile of the film director. Based on interviews with him, include information about his biggest influences and the inspiration for his latest work.

[1260] In this way, a series of processes on the server and terminals will realize a system in which the knowledge and experience of experts can be efficiently collected, organized, shared, and delivered to many users.

[1261] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1262] Step 1:

[1263] Generate interview questions

[1264] The server uses a generative AI model (e.g., GPT-3) to automatically generate questions for the expert being interviewed based on the prompt. The input is the "expert's field" and the "question prompt." The generative AI model outputs an appropriate question (e.g., "Which director influenced you?") based on these inputs and stores it in a database. Specifically, the server sends the prompt to the generative AI model, formats the returned answer as a question, and stores it in the database.

[1265] Step 2:

[1266] Scheduling and managing interviews

[1267] The server contacts the expert and arranges the interview date and time. The inputs are the expert's contact information and available time slots. Based on this, the server determines the interview date and time and saves it in the database. Specifically, the server contacts the expert via email or in-app notification and records the agreed interview date and time in the database.

[1268] Step 3:

[1269] Running the Interview

[1270] When the time for the interview arrives, the server starts the interview phase in real time. The inputs are the question and expert information. The server connects the expert and the AI ​​interviewer using technologies such as WebRTC and collects the results. Specifically, it initiates a real-time voice or text dialogue and stores the interview content in a database.

[1271] Step 4:

[1272] Profile and Article Generation

[1273] The server analyzes the collected interview data and uses a generative AI model to generate expert profiles and articles. The inputs are "interview data" and "generative AI model." The interview data is preprocessed using a data analysis tool and then input into the generative AI model to obtain profiles and articles. This output data is stored in a database. Specifically, it organizes the interview data through natural language analysis and automatically generates profiles and articles based on it.

[1274] Step 5:

[1275] User recommendations

[1276] Based on the user's interests and past browsing history, the server recommends appropriate content to the user. The inputs are a user profile database and generated profiles and articles. A recommender system is built using collaborative filtering and content-based filtering algorithms to recommend appropriate content to the user. Specifically, it analyzes the user profile, identifies users with similar interests and related content, and presents it to the user.

[1277] Step 6:

[1278] Gathering feedback

[1279] Users can provide feedback on the content they have viewed. Inputs include "user feedback content" and "user ID." The device sends this feedback to the server, which stores it in a database. Specifically, the system receives feedback through the user interface and transfers it to the server as structured data.

[1280] This series of steps will create a system that efficiently collects, organizes, and shares expert knowledge and experience, and provides appropriate content based on user interests.

[1281] 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.

[1282] This invention is a system for sharing and spreading expert knowledge and experience using a generative model and an emotion engine. The system aims to promote the sharing of expert knowledge by recommending content generated based on the user's interests and emotions to appropriate users. Specific embodiments are described below.

[1283] Server Processing

[1284] 1. Generate interview questions:

[1285] The server uses the generative model to generate interview questions for the specialist in question, such as "Which film directors have influenced you?"

[1286] 2. Interview scheduling and management:

[1287] The server coordinates the interview date and time with the interviewee specialist and stores the schedule information in a database, which can be referenced later at the appropriate time.

[1288] 3. Conducting the interview:

[1289] When the interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time. At this time, the emotion engine also works to analyze the user's emotions.

[1290] 4. Emotional data collection and analysis:

[1291] The server uses an emotion engine to analyze user data collected during the interview in real time, assessing emotions based on data such as facial expressions, tone of voice, and survey responses selected.

[1292] 5. Profile and article generation:

[1293] Based on the interview and sentiment analysis data, the server uses a generative model to generate specialist profiles and articles that include a multifaceted organization of expertise and experience, as well as sentiment data.

[1294] 6. User Recommendations:

[1295] The server refers to the user profile database and recommends the generated content to the appropriate user based on the user's interests, past browsing history, and emotional feedback. For example, a newly generated article about a ceramics expert may be provided to a user who is interested in ceramics and has previously viewed related content and felt emotionally satisfied.

[1296] Terminal handling

[1297] 1. Notification of Interview Participation:

[1298] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and / or in-app notification, and the specialist will be able to view detailed interview information.

[1299] 2. Interview phase begins:

[1300] At the appointed date and time, the device initiates a session for the specialist to participate in the interview, connecting the device to the AI ​​interviewer in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[1301] 3. Viewing and Feedback:

[1302] Users can view the profiles and articles generated through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[1303] User Action

[1304] 1. Interview Participation:

[1305] If the user participates in the interview as a specialist, they will check the interview details based on the notification on their device, and at the designated date and time, they will participate in the interview phase through their device.

[1306] 2. Viewing profiles and articles:

[1307] Users can browse recommended profiles and articles on their devices. For example, an amateur pottery enthusiast can read an interview with a pottery expert. As the user reads the article, the emotion engine analyzes their emotions in real time and sends emotional data as feedback.

[1308] 3. Feedback and sharing:

[1309] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1310] Specific examples

[1311] For example, if film director B is the subject of an interview, the following process will be carried out.

[1312] 1. The server generates a question for the film director: "Which film directors have influenced you?"

[1313] 2. Arrange an interview date and time and send a notification to Director B.

[1314] 3. On the specified date and time, Director B will join the interview session via the terminal.

[1315] 4. The server conducts the interview in real time and collects the response data. At the same time, the emotion engine analyzes Director B's emotions.

[1316] 5. Generate a profile and article about film director B based on the collected data.

[1317] 6. Recommend generated articles to users who are interested in movies and who have had a positive emotional reaction to past related articles.

[1318] 7. Movie fan users view the generated articles on their devices and provide their impressions as feedback that is analyzed by the emotion engine.

[1319] In this way, a system that integrates an emotion engine allows experts' knowledge and experience to be shared more deeply and made available to a wider audience.

[1320] The processing flow will be explained below.

[1321] Server Processing Steps

[1322] Step 1:

[1323] The server uses the generative model to generate interview questions for the specialist in question, such as "Which film directors have influenced you?"

[1324] Step 2:

[1325] The server coordinates the interview date and time with the specialist to be interviewed and stores the schedule information in a database, thereby managing the specialist to participate in the interview at an appropriate time.

[1326] Step 3:

[1327] When the scheduled interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time. At the same time, the emotion engine also runs to collect the user's emotional data.

[1328] Step 4:

[1329] The server uses an emotion engine to analyze the user's emotional data collected during the interview in real time, assessing their emotions based on, for example, their facial expressions, tone of voice, and selected survey answers.

[1330] Step 5:

[1331] The server uses a generative model to generate specialist profiles and articles based on interview response data and sentiment data. These profiles and articles systematically organize knowledge and experience.

[1332] Step 6:

[1333] The server refers to a user profile database and recommends generated content to appropriate users based on their interests, past browsing history, and emotional feedback. For example, a newly generated article about a pottery expert will be provided to users who are interested in pottery and have previously viewed related content and had a positive emotional response.

[1334] Terminal processing steps

[1335] Step 1:

[1336] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, where the specialist can view detailed interview information.

[1337] Step 2:

[1338] At the designated date and time, the device will initiate the interview session and connect with the AI ​​interviewer in real time, while the emotion engine will simultaneously collect and analyze the user's emotional data.

[1339] Step 3:

[1340] Users can view the generated profiles and articles through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[1341] User processing steps

[1342] Step 1:

[1343] If the user participates in the interview as a specialist, they will check the interview details based on the notification on their device, and at the designated date and time, they will participate in the interview phase through their device.

[1344] Step 2:

[1345] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1346] Step 3:

[1347] Users can provide feedback on the content and share articles they find particularly interesting with other users or on social media, further spreading their expertise and interests. This feedback is analyzed by the sentiment engine and sent to the server.

[1348] Specific examples

[1349] For example, if film director B is the subject of an interview, the processing flow is as follows:

[1350] Server Processing Steps

[1351] Step 1:

[1352] Generate a question for Director B: "Which directors influenced you?"

[1353] Step 2:

[1354] The interview date and time is arranged and a notification is sent to Director B.

[1355] Step 3:

[1356] At the specified date and time, film director B will participate in the interview session via the terminal.

[1357] Step 4:

[1358] Conduct interviews in real time and collect response and sentiment data.

[1359] Step 5:

[1360] Based on the collected data, a profile and article about film director B is generated.

[1361] Step 6:

[1362] The generated articles are recommended to users who are interested in movies and who have a positive reaction based on emotional feedback.

[1363] Terminal processing steps

[1364] Step 1:

[1365] Notify Director B of the interview date and time and how to participate.

[1366] Step 2:

[1367] Director B begins the interview session using a device to participate in the interview. At this time, the emotion engine also starts to collect and analyze emotion data.

[1368] Step 3:

[1369] Users can view the generated articles on their devices and provide feedback, which is also analyzed by the emotion engine and sent to the server.

[1370] User processing steps

[1371] Step 1:

[1372] Film director B checks the interview details based on the notification.

[1373] Step 2:

[1374] The user views an article about recommended film director B.

[1375] Step 3:

[1376] Users provide feedback on articles and share it on social media, etc. This feedback is analyzed by the emotion engine and sent to the server.

[1377] Example 2

[1378] 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."

[1379] There is a need for a method to effectively share the knowledge and experience of experts and provide appropriate content based on users' interests and emotions. However, conventional methods have been unable to fully utilize the quality of interview content or users' emotional feedback, making it difficult to recommend content that is highly relevant to users. This has resulted in insufficient dissemination of expert knowledge and improvement of user satisfaction.

[1380] 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.

[1381] In this invention, the server includes means for generating interview questions using a generative model, means for connecting the interviewed expert with an AI interviewer to conduct the interview, means for collecting and analyzing user data during the interview using an emotion engine, means for analyzing the collected interview data and emotion data to generate expert profiles and articles, and means for recommending the generated content to appropriate users based on the user's interests, past browsing history, and emotion feedback. This makes it possible to effectively share and spread the knowledge and experience of experts and provide users with content that is highly relevant and satisfying.

[1382] A "generative model" is a machine learning-based model that generates new text or information based on given prompts and data.

[1383] The "means for generating interview questions" is a mechanism that uses a generative model to automatically create specific questions for the expert being interviewed.

[1384] The "means of conducting the interview" is a mechanism that connects the subject expert with an AI interviewer and conducts the actual interview.

[1385] The "emotion engine" is a system that analyzes data such as the user's facial expressions, tone of voice, and selected survey responses to evaluate the user's emotions.

[1386] "Means for collecting and analyzing user data" refers to a system for collecting data obtained from users during interviews and analyzing their emotions and interests based on that data.

[1387] "Interview data" refers collectively to expert responses and other related information collected through interviews.

[1388] The "means for generating profiles and articles" is a mechanism for generating expert profiles and related articles based on the collected interview data and sentiment data.

[1389] The "user profile database" is a database for storing and managing information such as user interests, past browsing history, and emotional feedback.

[1390] "Means for recommending content" refers to a system that recommends appropriate content to users based on their interests, past browsing history, and emotional feedback.

[1391] This invention is a system for sharing and spreading expert knowledge and experience by utilizing a generative model and an emotion engine. The system aims to promote the sharing of expert knowledge by recommending content generated based on the user's interests and emotions to appropriate users. A specific embodiment of this system is described below.

[1392] Server Processing

[1393] 1. Generate interview questions:

[1394] The server generates interview questions using a generative AI model (e.g., GPT-4). At this time, instructions such as "Create interview questions for a film director" are input as prompts. An example of a question generated by the generative model is "Which film directors have influenced you?"

[1395] 2. Interview scheduling and management:

[1396] The server searches for relevant experts based on the generated interview questions and schedules the interview. This information is stored in a database and is referenced when the interview is conducted.

[1397] 3. Conducting the interview:

[1398] When the time for the interview arrives, the server connects the expert to the AI ​​interviewer and conducts the interview. The generative model automatically asks questions and collects the expert's answers in real time. At the same time, the emotion engine runs to analyze the user's emotional data.

[1399] 4. Emotional data collection and analysis:

[1400] The server uses an emotion engine to collect and analyze data such as the user's facial expressions, tone of voice, and questionnaire responses during the interview to evaluate the user's emotions.

[1401] 5. Profile and article generation:

[1402] Based on the interview and sentiment analysis data, the server uses a generative AI model to generate expert profiles and articles that include a multifaceted breakdown of expertise and experience, as well as sentiment data.

[1403] 6. User Recommendations:

[1404] The server references a user profile database and recommends content based on the user's interests, past browsing history, and emotional feedback. For example, a user interested in pottery can be recommended a newly generated article by a pottery expert.

[1405] Terminal handling

[1406] 1. Notification of Interview Participation:

[1407] The device will notify the interviewed expert of the date and time of the interview and how to participate. This notification will be provided via email and in-app notification, and the expert will be able to view detailed interview information.

[1408] 2. Interview phase begins:

[1409] At the specified date and time, the device will automatically start the interview session and connect the AI ​​interviewer with the expert in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[1410] 3. Viewing and Feedback:

[1411] Users can view the generated profiles and articles through their devices and can also provide feedback based on their impressions while browsing, which is analyzed by the emotion engine and sent to the server.

[1412] User Action

[1413] 1. Interview Participation:

[1414] When participating in an interview as an expert, the user will check the interview details based on notifications from the device, and at the designated date and time, they will participate in the interview phase through the device.

[1415] 2. Viewing profiles and articles:

[1416] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1417] 3. Feedback and sharing:

[1418] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1419] Specific examples

[1420] For example, if Director B is chosen for an interview, the following process will occur:

[1421] 1. The server uses the generative model to generate a question for the film director: "Which film directors have influenced you?"

[1422] 2. Arrange an interview date and time and send a notification to Director B.

[1423] 3. On the specified date and time, Director B will join the interview session via the terminal.

[1424] 4. The server conducts the interview in real time and collects the response data. At the same time, the emotion engine analyzes Director B's emotions.

[1425] 5. Generate a profile and article about film director B based on the collected data.

[1426] 6. Recommend generated articles to users who are interested in movies and have had a positive emotional reaction to past related articles.

[1427] 7. Movie fan users view the generated articles on their devices and provide their impressions as feedback that is analyzed by the emotion engine.

[1428] This system integrates an emotion engine to share the knowledge and experience of experts widely and deeply, enabling it to provide highly relevant content to many users.

[1429] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1430] Step 1: Generate interview questions

[1431] Input: Interview request from user, prompt

[1432] Specific behavior:

[1433] The server receives a request from the user to "interview a film director." The server inputs the prompt "Create interview questions for the film director" into a generative AI model (e.g., GPT-4).

[1434] Data processing / calculation:

[1435] The generative AI model analyzes the prompt and generates new interview questions, referencing question generation patterns from an existing database.

[1436] Output: Generated interview questions (e.g., "Which film directors influenced you?")

[1437] Step 2: Schedule and manage interviews

[1438] Input: Interview questions, expert schedule data

[1439] Specific behavior:

[1440] The server searches the database for relevant experts (e.g., film directors) based on the generated question, checks the experts' schedules, and arranges the date and time of the interview.

[1441] Data processing / calculation:

[1442] It looks at the expert's calendar, automatically identifies available time slots, and selects the appropriate time.

[1443] Output: List of experts with interview date and time (e.g., Film Director B, October 1, 2023, 2:00 PM)

[1444] Step 3: Conducting the interview

[1445] Input: Interview questions, expert connections

[1446] Specific behavior:

[1447] When the time for the interview arrives, the server sends a notification to the expert (e.g., film director B) to start the interview. The server then starts the AI ​​interviewer and delivers the generated questions one by one.

[1448] Data processing / calculation:

[1449] The AI ​​interviewer collects the expert's answers in real time and stores them in a database, while the emotion engine collects facial and vocal data and analyzes the emotional data.

[1450] Output: Expert response data, sentiment data collected in real time

[1451] Step 4: Collect and analyze emotion data

[1452] Input: Expert response data, user sentiment data

[1453] Specific behavior:

[1454] During the interview, the server uses an emotion engine to collect and analyze emotional data from the user in real time, such as facial expressions, tone of voice, and questionnaire responses.

[1455] Data processing / calculation:

[1456] The collected data is evaluated using facial expression analysis software and tone analysis software to determine and quantify positive, negative, and neutral emotions.

[1457] Output: Analyzed sentiment data (e.g., 80% positive, 10% negative, 10% neutral)

[1458] Step 5: Generate profiles and articles

[1459] Input: Expert response data, analyzed sentiment data

[1460] Specific behavior:

[1461] The server instructs the generative AI model to generate expert profiles and articles based on the collected response data and sentiment data.

[1462] Data processing / calculation:

[1463] The generative model analyzes this data, organizes the knowledge and experience of experts from various perspectives, and converts them into article format, while also appropriately reflecting sentiment data.

[1464] Output: Expert profile and article (e.g., article about film director B)

[1465] Step 6: Recommend to users

[1466] Input: User interest data, past browsing history, emotional feedback

[1467] Specific behavior:

[1468] The server references a user profile database and selects content based on interests, past browsing history, and emotional feedback, and then notifies the selected content to the target user.

[1469] Data processing / calculation:

[1470] It uses recommendation algorithms to surface content that matches your interests, and also uses your browsing history and emotional feedback to select the most relevant articles.

[1471] Output: Content notification to user (e.g. "A new article about pottery has been published")

[1472] (Application example 2)

[1473] 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."

[1474] The lack of a system that widely shares the knowledge and experience of experts and appropriately recommends them based on user interests and emotions is another issue. Another issue is the lack of technology that effectively utilizes user feedback and emotional data to organize and provide expert content from multiple perspectives.

[1475] 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 generating interview questions using a generative model, means for connecting the interviewed expert with an AI interviewer and conducting the interview, means for analyzing the collected interview data to generate expert profiles and articles, means for recommending content generated based on the user's interests, past browsing history, and emotional feedback to appropriate users, and means for analyzing the user's feedback with an emotion engine. This makes it possible to effectively share and recommend experts' knowledge and experience based on the user's emotions and interests.

[1476] A "generative model" refers to an algorithm that automatically generates new content, questions, etc. based on data.

[1477] "Interview questions" are questions asked to experts, and these questions are generated using a generative model.

[1478] An "expert" is an individual who has knowledge or experience in a particular field.

[1479] "AI interviewer" refers to a system or software that uses artificial intelligence to ask interview questions and collect answers.

[1480] "Interview Data" refers to the experts' responses and other relevant information collected during the interview.

[1481] A "profile" is information that outlines an expert's knowledge and experience, created based on collected data.

[1482] "Articles" refer to written information created based on interview data, and are detailed descriptions of the experts' knowledge and experiences.

[1483] "Recommendation" refers to recommending appropriate content based on a user's interests, past browsing history, and emotional feedback.

[1484] "Emotional feedback" refers to the analysis of users' emotional reactions when viewing content and the data and comments based on that.

[1485] An "emotion engine" refers to software or algorithms that analyze a user's emotions from facial expressions, tone of voice, text feedback, etc.

[1486] "Content" refers to the collection of information, entertainment, etc. provided to users through communications means.

[1487] "System" refers to an integrated mechanical or software collection of related means and devices.

[1488] This invention is a system that shares the knowledge and experience of experts and recommends appropriate content based on the user's interests and emotions. The detailed configuration and operation procedures for implementing this invention are described below.

[1489] Server Processing

[1490] 1. Generate interview questions:

[1491] The server uses a generative model to generate interview questions for domain experts, such as a natural language processing model like BERT (Bidirectional Encoder Representations from Transformers).

[1492] Example: Generating questions for a film director such as "Which film directors have influenced you?"

[1493] 2. Interview scheduling and management:

[1494] The server coordinates the interview date and time with the interview subject and stores the schedule information in a database, using the Python datetime module for schedule management.

[1495] Example: An interview with a film director is scheduled for "2023-10-01 15:00:00".

[1496] 3. Conducting the interview:

[1497] When the interview date and time arrives, the server starts the interview phase and connects the expert with the AI ​​interviewer. The generative model automatically asks questions and collects the expert's answers in real time. At the same time, the emotion engine runs and analyzes the user's emotional data.

[1498] Hardware and software used: GPU-based servers, generative AI models (e.g., GPT-3), emotion analysis software (e.g., facial emotion recognition API)

[1499] 4. Emotional data collection and analysis:

[1500] The server analyzes user data collected during the interview in real time using an emotion engine, which evaluates emotions based on data such as the user's facial expressions, tone of voice, and selected survey answers.

[1501] Example: Determining user satisfaction and interest in an expert's answer in real time.

[1502] 5. Profile and article generation:

[1503] Based on the interview and sentiment analysis data, the server uses a generative model to generate expert profiles and articles that include a multifaceted organization of expertise and experience, as well as sentiment data.

[1504] Hardware and software used: Database management system (e.g., PostgreSQL), natural language generation model (e.g., OpenAI's GPT-3)

[1505] 6. User Recommendations:

[1506] The server references a user profile database and recommends content to appropriate users based on their interests, past browsing history, and emotional feedback.

[1507] Example: A newly generated article by a ceramics expert is recommended to users who are interested in ceramics.

[1508] Terminal handling

[1509] 1. Notification of Interview Participation:

[1510] Once selected, the device will notify experts of the interview date and time and how to participate, via email and in-app notification.

[1511] Example: "Your interview is scheduled for 2023-10-01 15:00:00"

[1512] 2. Interview phase begins:

[1513] At the appointed date and time, the expert will initiate a session on the device and connect with the AI ​​interviewer in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[1514] Hardware and software used: Webcam, microphone, emotion analysis software

[1515] 3. Viewing and Feedback:

[1516] Users can view the profiles and articles generated through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[1517] Example: "This article was very helpful."

[1518] User Action

[1519] 1. Interview Participation:

[1520] If the user participates in the interview as an expert, they will check the interview details based on the notification from the device, and at the designated date and time, they will participate in the interview phase through the device.

[1521] Example: A user completes an interview at a specified date and time.

[1522] 2. Viewing profiles and articles:

[1523] Users browse recommended profiles and articles that interest them through their devices, and the emotion engine analyzes their emotions in real time and sends emotional data as feedback.

[1524] Example: "I was reading an article by a ceramics expert and it was very interesting."

[1525] 3. Feedback and sharing:

[1526] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1527] Example prompts: "What would a movie-loving user like to read next?", "Recommendations for movie-related interviews."

[1528] The above is a specific embodiment for carrying out the invention. This invention makes it possible to widely share the knowledge and experience of experts and to effectively recommend content based on the interests and emotions of users.

[1529] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1530] Step 1:

[1531] The server generates the interview questions

[1532] The server uses a generative model (e.g., a natural language processing model such as BERT) to generate appropriate questions for the expert being interviewed. In this process, the generative model receives a prompt (e.g., "Which other film directors have influenced this film director?") as input and generates a question as output. Specifically, the generative model analyzes the prompt, extracts relevant information from a database, and generates a new question.

[1533] Step 2:

[1534] The server schedules the interviews

[1535] The server arranges the interview date and time with the expert and saves the schedule information in a database. The expert's free time and the server's free time are referenced as input, and the optimal date and time is automatically selected. The schedule information is then saved in the database and is also output as notification data. Specifically, the date and time are set using the Python datetime module, and the process is performed to save it in a database such as PostgreSQL.

[1536] Step 3:

[1537] The server begins the interview phase

[1538] When the interview date and time arrives, the server starts the interview phase and connects the expert with the AI ​​interviewer. At this time, the generative model automatically asks questions and collects the expert's answers in real time. The emotion engine also runs and analyzes the user's emotional data. The input is the interview questions and the expert's answers, and the output is the generated answer data. Specifically, the AI ​​interviewer conducts the interview, and the exchange is collected as text data.

[1539] Step 4:

[1540] The server collects and analyzes emotion data

[1541] During the interview phase, the server uses an emotion engine to analyze emotional data such as the user's facial expressions, tone of voice, and selected survey answers. The input is the user's real-time reaction data, and the output is analyzed emotional evaluation data. Specifically, emotion analysis software (e.g., facial expression analysis API or voice emotion analysis API) is used to analyze the emotional data in real time, and the results are stored on the server.

[1542] Step 5:

[1543] Server generates profiles and articles

[1544] Based on the interview and sentiment analysis data, the server uses a generative model to generate expert profiles and articles. These profiles and articles organize expertise and experience from multiple angles and also include sentiment data. The input is the interview response data and sentiment analysis data, and the output is the generated profile and article. Specifically, a natural language generation model (e.g., GPT-3) is used to convert the data into text and create expert profiles and articles.

[1545] Step 6:

[1546] The server recommends content to the user

[1547] The server references a user profile database and recommends content to appropriate users based on their interests, past browsing history, and emotional feedback. The input is profile information from the user database, past browsing history, and emotional feedback, and the output is recommended content. Specifically, a machine learning model is used to analyze the user's past behavior and feedback and recommend the next content to be viewed.

[1548] 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.

[1549] 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.

[1550] 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.

[1551] [Fourth embodiment]

[1552] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1553] 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.

[1554] 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).

[1555] 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.

[1556] 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.

[1557] 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).

[1558] 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.

[1559] 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.

[1560] 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.

[1561] 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.

[1562] 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.

[1563] 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.

[1564] 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."

[1565] This invention is a system for sharing and spreading the knowledge and experience of experts using generative models. The purpose of this system is to support the discovery of new hobbies and the deepening of expertise, thereby enabling their knowledge to be widely disseminated. Specific embodiments are described below.

[1566] Server Processing

[1567] 1. Generate interview questions:

[1568] The server uses the generative model to automatically generate interview questions for specialists, for example, using machine learning algorithms to create optimal questions from a training database.

[1569] 2. Interview scheduling and management:

[1570] The server coordinates the interview date and time with the specialist to be interviewed and stores the schedule information in a database, thereby managing the specialist to participate in the interview at an appropriate time.

[1571] 3. Conducting the interview:

[1572] When the interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer in real time, with the generative model automatically asking questions and collecting the specialist's answers.

[1573] 4. Generate profiles and articles:

[1574] Based on the collected interview data, the server uses a generative model to generate specialist profiles and articles, which systematically organize the specialist's knowledge and experience.

[1575] 5. User Recommendations:

[1576] The server references the user profile database and recommends content to appropriate users based on the collected user interests and past browsing history. For example, a newly generated article about a pottery expert will be recommended to users who are interested in pottery.

[1577] Terminal handling

[1578] 1. Notification of Interview Participation:

[1579] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, and the specialist will be able to view detailed interview information.

[1580] 2. Interview phase begins:

[1581] When a specialist joins an interview, the device initiates the interview session and connects them in real time with an AI interviewer, who then communicates using voice and text.

[1582] 3. Viewing and Feedback:

[1583] Users can view the generated profiles and articles through their devices and can provide feedback on their impressions while viewing, which is then sent to the server via their devices.

[1584] User Action

[1585] 1. Interview Participation:

[1586] If the user participates in the interview as a specialist, they will check the interview details based on the notification from the device, and at the specified date and time, they will participate in the interview phase through the device.

[1587] 2. Viewing profiles and articles:

[1588] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1589] 3. Feedback and sharing:

[1590] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1591] Specific examples

[1592] For example, if film director A is the subject of an interview, the process will proceed as follows:

[1593] 1. The server generates a question for the film director: "Which directors influenced you?"

[1594] 2. Arrange an interview date and time and send a notification to Director A.

[1595] 3. On the specified date and time, Director A will participate in the interview session via the terminal.

[1596] 4. The server runs the interview in real time and collects the response data.

[1597] 5. Generate a profile and article about film director A based on the collected data.

[1598] 6. Recommend generated articles to users who are interested in movies.

[1599] 7. Movie fan users can view the generated article on their devices and provide feedback on their impressions.

[1600] In this way, a system is created in which the knowledge and experience of specialists is systematically organized and shared with many users.

[1601] The processing flow will be explained below.

[1602] Server Processing Steps

[1603] Step 1:

[1604] The server uses the generative model to generate interview questions for the specialist in question, such as "Which directors influenced you?"

[1605] Step 2:

[1606] The server coordinates interview dates and times with the interviewed specialists and stores the schedule information in a database, where it is maintained for future reference.

[1607] Step 3:

[1608] When the scheduled interview date and time arrives, the server starts the interview phase, connecting the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time.

[1609] Step 4:

[1610] The server analyzes the collected interview data and uses generative models to generate specialist profiles and articles, systematically organizing the collected data into knowledge and experience.

[1611] Step 5:

[1612] The server references the user profile database and recommends content to appropriate users based on their interests and past browsing history. For example, a newly generated article about a ceramics expert will be recommended to users who are interested in ceramics.

[1613] Terminal processing steps

[1614] Step 1:

[1615] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, and the specialist will be able to view detailed interview information.

[1616] Step 2:

[1617] At the appointed date and time, when the specialist joins the interview, the device will initiate the interview session and connect with an AI interviewer in real time, engaging in a dialogue using voice and text.

[1618] Step 3:

[1619] Users can view the generated profiles and articles through their devices and provide feedback based on their impressions of the views, which is then sent to the server via their devices.

[1620] User processing steps

[1621] Step 1:

[1622] If the user participates in the interview as a specialist, they will check the interview details based on the notification from the device, and at the specified date and time, they will participate in the interview phase through the device.

[1623] Step 2:

[1624] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1625] Step 3:

[1626] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1627] Specific examples

[1628] The processing flow when film director A is the subject of an interview is as follows:

[1629] Server Processing Steps

[1630] Step 1:

[1631] Generate a question for film director A: "Which directors influenced you?"

[1632] Step 2:

[1633] Arrange an interview date and time and send a notification to Director A.

[1634] Step 3:

[1635] At the specified date and time, film director A will participate in the interview session via the terminal.

[1636] Step 4:

[1637] Conduct interviews and collect response data in real time.

[1638] Step 5:

[1639] Based on the collected data, a profile and article about film director A is generated.

[1640] Step 6:

[1641] Recommend generated articles to users who are interested in movies.

[1642] Terminal processing steps

[1643] Step 1:

[1644] Notify Director A of the interview date and time and how to participate.

[1645] Step 2:

[1646] Film director A starts an interview session using a terminal to participate in the interview.

[1647] Step 3:

[1648] Users can view the generated articles through their devices and provide feedback.

[1649] User processing steps

[1650] Step 1:

[1651] Director A will confirm the interview details based on the notification.

[1652] Step 2:

[1653] The user views an article about recommended film director A.

[1654] Step 3:

[1655] Users provide and share feedback on articles.

[1656] Example 1

[1657] 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."

[1658] There is a growing demand for systems that can efficiently collect the knowledge and experience of experts and provide it to a wide range of users. However, current systems face challenges in effectively managing and conducting interviews with experts and recommending appropriate content based on each user's interests and browsing history.

[1659] 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.

[1660] In this invention, the server includes a means for generating interview questions using a generative model, a means for connecting an AI interviewer to the expert being interviewed using real-time communication means to conduct the interview, a means for analyzing the collected interview data to generate expert profiles and articles, and a means for recommending the generated content to appropriate users based on the user's interests and past browsing history. This makes it possible to effectively collect and organize the knowledge and experience of experts and provide personalized information to users.

[1661] A "generative model" is a method or system that uses machine learning algorithms to generate new data based on existing data.

[1662] "Interview questions" are questions created using generative models to elicit information or knowledge from experts or subjects.

[1663] An "expert" is a person who has advanced knowledge and experience in a particular field and has deep understanding and skills in that field.

[1664] "AI Interviewer" is a system that uses artificial intelligence to automatically conduct interviews and collect necessary information.

[1665] "Real-time communications means" are technologies and systems that allow information to be exchanged over the Internet or other communications networks with little or no time delay.

[1666] A "profile" is a document or data that systematically organizes an expert's knowledge and experience and shows the individual's characteristics and activities.

[1667] An "article" is a piece of writing that is created based on interview results and expert knowledge and is intended to provide information to a large number of users.

[1668] "User" means a general user who uses the system to view content and provide feedback.

[1669] "Interests" are the interest or curiosity a user shows in a particular field or topic.

[1670] "Browsing history" is a record of the content a user has viewed within the system.

[1671] "Recommendation" is the act of presenting optimal content based on a user's interests and browsing history.

[1672] This invention is a system for sharing and spreading expert knowledge and experience using generative AI models, with the aim of supporting the discovery of new hobbies, deepening of expertise, and spreading expert knowledge widely.

[1673] Server Processing

[1674] Generate interview questions

[1675] The server generates interview questions using a generative AI model (e.g., GPT-4). The server retrieves relevant keywords and past questions from a database of past interviews, and inputs prompt sentences based on these into the generative AI model. This automatically generates optimal questions for the expert. The following is a specific example.

[1676] Example prompt sentence:

[1677] "Generate interview questions for film director A. I'd like to know which directors influenced him."

[1678] Scheduling and managing interviews

[1679] The server retrieves the specialist's schedule information from the API and adjusts the interview date and time. The adjusted schedule information is saved in the database and an email or in-app notification is sent to the specialist. This series of operations ensures that the specialist can attend the interview at the appropriate time.

[1680] Running the Interview

[1681] When the interview date and time arrives, the server will start the interview session using a real-time communication method such as Zoom or Teams. The server will connect the AI ​​interviewer with the expert and conduct the interview in real time by presenting generated questions. Dialogue during the interview can include both voice and text.

[1682] Profile and Article Generation

[1683] Once the interview is complete, the server inputs the collected data into a generative AI model to generate a profile and article about the expert. These profiles and articles systematically describe the expert's knowledge and experience. The server then stores this data in a database.

[1684] User recommendations

[1685] The server references a user profile database and recommends the most appropriate content based on the user's interests and past browsing history. For example, a user interested in pottery might be recommended a newly generated interview article with a pottery expert.

[1686] Terminal handling

[1687] Notification of interview participation

[1688] Experts who are selected for an interview will be notified of the interview date and time and how to participate via email and in-app notification, and they can view detailed interview information.

[1689] Beginning of the interview phase

[1690] When the interview date and time arrives, the expert participates in the interview phase via a device, which connects to an AI interviewer in real time and begins the interview using voice and text.

[1691] Views and Feedback

[1692] Users can view the generated profiles and articles through their devices, and can also provide feedback on their impressions while viewing, which is then sent to the server via their devices.

[1693] User Action

[1694] Participating in an interview

[1695] If the user participates in the interview as an expert, he / she will check the interview details based on the notification from the device, and at the specified date and time, he / she will participate in the interview phase through the device.

[1696] View Profiles and Articles

[1697] Users can browse interesting profiles and articles recommended by the server through their devices. For example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1698] Feedback and sharing

[1699] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1700] As a result, this system can effectively collect and organize the knowledge and experience of experts, and provide personalized information to users.

[1701] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1702] Step 1:

[1703] The server generates interview questions using a generative AI model. In this step, it takes keywords obtained from past interview records and related databases as input and generates a prompt based on them. The prompt is sent to a generative AI model (e.g., GPT-4) to generate a new interview question. This question is output in the form of, for example, "What kind of movie would you like to make next?" Specifically, the server accesses the database, extracts relevant data, converts it into a prompt, sends it to the generative AI model, and stores the generated question in the database.

[1704] Step 2:

[1705] The server schedules an interview with a specialist. First, it accesses the specialist's calendar API to obtain available time slots. Based on the obtained schedule data, it confirms the interview date and time and saves it in the database. It also sends the confirmed date and time and participation instructions to the specialist via email and in-app notification. The input is the specialist's available schedule, and the output is the confirmed interview date. Specific operations include checking the schedule via the API, algorithmic processing to adjust the date and time, updating the database, and sending notifications.

[1706] Step 3:

[1707] When the interview date and time arrives, the server starts the interview phase using a real-time communication tool (e.g., Zoom, Teams). The server provides pre-generated interview questions to the AI ​​interviewer, which then presents the questions to the expert sequentially. The input is the interview date and time and the generated questions, and the output is the expert's response data. Specific operations include creating a meeting by calling Zoom's API, providing the generated questions sequentially, and collecting the expert's responses.

[1708] Step 4:

[1709] The server generates profiles and articles based on the collected interview data. The collected response data is input into a generative AI model to generate expert profiles and articles. The input for this process is the collected response data, and the output is the generated profile and article. Specific operations include text analysis of the collected data, generation of profiles and articles by the generative AI model, and storage in a database.

[1710] Step 5:

[1711] The server recommends the generated profiles and articles to users. First, it references the user profile database to obtain the user's interests and past browsing history. Based on this data, it selects highly relevant content and recommends it to the target user. The input is the user's interests and browsing history, and the output is the recommended content. Specific operations include database query, content selection using a recommendation algorithm, and notification and email sending.

[1712] Step 6:

[1713] The device notifies the selected expert of the interview date and time and how to participate. The device provides detailed interview information to the expert based on the schedule information obtained from the server. The input is the schedule information from the server, and the output is a notification to the expert. Specific operations include generating notifications and sending notifications via email or within the app.

[1714] Step 7:

[1715] When the interview date arrives, the terminal helps the expert join the interview session. The expert operates the terminal to connect to the real-time communication tool and begins a conversation with the AI ​​interviewer. The input is the meeting link provided by the server, and the output is the start of the interview session. Specific operations include providing the meeting link and guiding the connection procedure.

[1716] Step 8:

[1717] Users can view the generated profiles and articles through their devices and provide their thoughts and feedback. When users enter their feedback, the data is sent to the server via their devices. The input is the user's feedback, and the output is sending the feedback to the server. Specific actions include entering feedback, pressing the send button, and sending data to the server.

[1718] Step 9:

[1719] When a user joins an interview, the device provides the user with detailed interview information and invites them to the interview session at the specified date and time. The input is the notified interview details, and the output is participation in the interview phase. Specific actions include operating the start button for the interview phase and connecting to a real-time communication tool.

[1720] Step 10:

[1721] Users browse recommended profiles and articles and share the content they find particularly interesting with other users or on social media. The input is the recommended content, and the output is the shared content. Specific actions include viewing the content, clicking the share button, and posting to social media.

[1722] (Application example 1)

[1723] 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."

[1724] This invention relates to a system for widely sharing and disseminating the knowledge and experience of experts, aiming to effectively deliver knowledge from a variety of fields, not just specific fields, to a wide range of users. Another objective is to create more interactive and fulfilling learning opportunities by distributing expert interviews in real time and providing an environment in which users can provide immediate feedback on the content. Conventional systems have limited the dissemination of knowledge due to the difficulty of efficiently collecting and organizing expert knowledge and delivering it to the appropriate users.

[1725] 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.

[1726] In this invention, the server includes a means for generating interview questions using a generative model, a means for connecting the interviewed expert with an AI interviewer to conduct the interview, a means for analyzing collected interview data to generate expert profiles and articles, a means for recommending generated content to appropriate users based on users' interests and past browsing history, a means for live and recorded broadcasts of interviews, and a means for collecting and managing feedback from users. This allows experts' knowledge and experience to be shared effectively in real time, and users can provide instant feedback. Therefore, the diffusion of knowledge and feedback to experts occurs in both directions, creating a more comprehensive knowledge sharing platform.

[1727] A "generative model" is an algorithm or method that can automatically create new information or patterns based on data.

[1728] "Means for generating interview questions" refers to a function that uses a generative model to automatically create appropriate questions for interviewees.

[1729] An "artificial intelligence interviewer" is a virtual interviewer that uses artificial intelligence technology to conduct interviews.

[1730] "Means for conducting interviews" refers to the function of connecting the interviewee with an artificial intelligence interviewer and conducting the interview.

[1731] "Means for analyzing collected interview data" refers to the function of analyzing data collected during the interview process and extracting useful information.

[1732] "Means for generating expert profiles and articles" refers to the function of creating expert profiles and related articles based on the analyzed interview data.

[1733] "Recommendation methods" refers to a function that recommends appropriate content to a specific user based on the user's interests and past browsing history.

[1734] "Means for live streaming of interviews" refers to a function for streaming the contents of interviews in real time.

[1735] "Means for distributing recorded interviews" refers to the function of distributing recorded interviews so that users can watch them later.

[1736] "Means for collecting and managing feedback" refers to the function of collecting opinions and feedback from users and managing them systematically.

[1737] In this invention, a specific embodiment of a system for sharing and disseminating expert knowledge and experience will be described.

[1738] System program generation

[1739] Server Processing

[1740] 1. Generate interview questions:

[1741] The server uses a generative AI model to automatically generate interview questions for experts. This generative AI model uses a language model such as GPT-3. The generated questions are stored in a database.

[1742] 2. Interview scheduling and management:

[1743] The server coordinates interview dates and times with experts and stores the information in a database using a database management system such as SQLite.

[1744] 3. Conducting the interview:

[1745] When the interview date and time arrives, the server starts the interview phase in real time, connecting the expert with the AI ​​interviewer using technologies such as WebRTC.

[1746] 4. Generate profiles and articles:

[1747] The collected interview data is analyzed and a generative AI model is used to generate expert profiles and articles, which are then stored in a database.

[1748] 5. Content Recommendations:

[1749] The server uses a user profile database to recommend appropriate content based on the user's interests and past browsing history. The recommendation engine uses collaborative filtering and content-based filtering algorithms.

[1750] Terminal handling

[1751] 1. Notification of Interview Participation:

[1752] The server notifies the interviewed expert of the date and time of the interview and how to participate via email or push notification.

[1753] 2. Interview phase begins:

[1754] At the time of the interview, the device initiates the interview session, connecting the AI ​​interviewer with the expert in real time, enabling an interactive dialogue using voice and text.

[1755] 3. Watch the live and recorded interviews:

[1756] Users can watch the live interview on their device, and even after the live broadcast has ended, they can watch the recorded footage later.

[1757] 4. Providing Feedback:

[1758] Users can provide feedback on interviews and articles they have watched from their devices, and this feedback is immediately sent to the server and managed.

[1759] Specific examples

[1760] For example, if an expert film director is the subject of an interview, the process will proceed as follows:

[1761] 1. The server generates a question for the film director: "Which directors influenced you?"

[1762] 2. Arrange an interview date and time and send a notification to the film director.

[1763] 3. At the appointed date and time, the film director will participate in a real-time interview via the terminal.

[1764] 4. The server runs the interview in real time and collects the response data.

[1765] 5. Based on the collected data, generative AI models are used to generate profiles and articles about film directors.

[1766] 6. Recommend the generated articles and profiles to movie fan users.

[1767] 7. Movie fans can view the article on their devices and provide feedback on their impressions.

[1768] Example prompt sentence:

[1769] Generate a professional profile of the film director. Based on interviews with him, include information about his biggest influences and the inspiration for his latest work.

[1770] In this way, a series of processes on the server and terminals will realize a system in which the knowledge and experience of experts can be efficiently collected, organized, shared, and delivered to many users.

[1771] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1772] Step 1:

[1773] Generate interview questions

[1774] The server uses a generative AI model (e.g., GPT-3) to automatically generate questions for the expert being interviewed based on the prompt. The input is the "expert's field" and the "question prompt." The generative AI model outputs an appropriate question (e.g., "Which director influenced you?") based on these inputs and stores it in a database. Specifically, the server sends the prompt to the generative AI model, formats the returned answer as a question, and stores it in the database.

[1775] Step 2:

[1776] Scheduling and managing interviews

[1777] The server contacts the expert and arranges the interview date and time. The inputs are the expert's contact information and available time slots. Based on this, the server determines the interview date and time and saves it in the database. Specifically, the server contacts the expert via email or in-app notification and records the agreed interview date and time in the database.

[1778] Step 3:

[1779] Running the Interview

[1780] When the time for the interview arrives, the server starts the interview phase in real time. The inputs are the question and expert information. The server connects the expert and the AI ​​interviewer using technologies such as WebRTC and collects the results. Specifically, it initiates a real-time voice or text dialogue and stores the interview content in a database.

[1781] Step 4:

[1782] Profile and Article Generation

[1783] The server analyzes the collected interview data and uses a generative AI model to generate expert profiles and articles. The inputs are "interview data" and "generative AI model." The interview data is preprocessed using a data analysis tool and then input into the generative AI model to obtain profiles and articles. This output data is stored in a database. Specifically, it organizes the interview data through natural language analysis and automatically generates profiles and articles based on it.

[1784] Step 5:

[1785] User recommendations

[1786] Based on the user's interests and past browsing history, the server recommends appropriate content to the user. The inputs are a user profile database and generated profiles and articles. A recommender system is built using collaborative filtering and content-based filtering algorithms to recommend appropriate content to the user. Specifically, it analyzes the user profile, identifies users with similar interests and related content, and presents it to the user.

[1787] Step 6:

[1788] Gathering feedback

[1789] Users can provide feedback on the content they have viewed. Inputs include "user feedback content" and "user ID." The device sends this feedback to the server, which stores it in a database. Specifically, the system receives feedback through the user interface and transfers it to the server as structured data.

[1790] This series of steps will create a system that efficiently collects, organizes, and shares expert knowledge and experience, and provides appropriate content based on user interests.

[1791] 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.

[1792] This invention is a system for sharing and spreading expert knowledge and experience using a generative model and an emotion engine. The system aims to promote the sharing of expert knowledge by recommending content generated based on the user's interests and emotions to appropriate users. Specific embodiments are described below.

[1793] Server Processing

[1794] 1. Generate interview questions:

[1795] The server uses the generative model to generate interview questions for the specialist in question, such as "Which film directors have influenced you?"

[1796] 2. Interview scheduling and management:

[1797] The server coordinates the interview date and time with the interviewee specialist and stores the schedule information in a database, which can be referenced later at the appropriate time.

[1798] 3. Conducting the interview:

[1799] When the interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time. At this time, the emotion engine also works to analyze the user's emotions.

[1800] 4. Emotional data collection and analysis:

[1801] The server uses an emotion engine to analyze user data collected during the interview in real time, assessing emotions based on data such as facial expressions, tone of voice, and survey responses selected.

[1802] 5. Profile and article generation:

[1803] Based on the interview and sentiment analysis data, the server uses a generative model to generate specialist profiles and articles that include a multifaceted organization of expertise and experience, as well as sentiment data.

[1804] 6. User Recommendations:

[1805] The server refers to the user profile database and recommends the generated content to the appropriate user based on the user's interests, past browsing history, and emotional feedback. For example, a newly generated article about a ceramics expert may be provided to a user who is interested in ceramics and has previously viewed related content and felt emotionally satisfied.

[1806] Terminal handling

[1807] 1. Notification of Interview Participation:

[1808] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and / or in-app notification, and the specialist will be able to view detailed interview information.

[1809] 2. Interview phase begins:

[1810] At the appointed date and time, the device initiates a session for the specialist to participate in the interview, connecting the device to the AI ​​interviewer in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[1811] 3. Viewing and Feedback:

[1812] Users can view the profiles and articles generated through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[1813] User Action

[1814] 1. Interview Participation:

[1815] If the user participates in the interview as a specialist, they will check the interview details based on the notification on their device, and at the designated date and time, they will participate in the interview phase through their device.

[1816] 2. Viewing profiles and articles:

[1817] Users can browse recommended profiles and articles on their devices. For example, an amateur pottery enthusiast can read an interview with a pottery expert. As the user reads the article, the emotion engine analyzes their emotions in real time and sends emotional data as feedback.

[1818] 3. Feedback and sharing:

[1819] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1820] Specific examples

[1821] For example, if film director B is the subject of an interview, the following process will be carried out.

[1822] 1. The server generates a question for the film director: "Which film directors have influenced you?"

[1823] 2. Arrange an interview date and time and send a notification to Director B.

[1824] 3. On the specified date and time, Director B will join the interview session via the terminal.

[1825] 4. The server conducts the interview in real time and collects the response data. At the same time, the emotion engine analyzes Director B's emotions.

[1826] 5. Generate a profile and article about film director B based on the collected data.

[1827] 6. Recommend generated articles to users who are interested in movies and who have had a positive emotional reaction to past related articles.

[1828] 7. Movie fan users view the generated articles on their devices and provide their impressions as feedback that is analyzed by the emotion engine.

[1829] In this way, a system that integrates an emotion engine allows experts' knowledge and experience to be shared more deeply and made available to a wider audience.

[1830] The processing flow will be explained below.

[1831] Server Processing Steps

[1832] Step 1:

[1833] The server uses the generative model to generate interview questions for the specialist in question, such as "Which film directors have influenced you?"

[1834] Step 2:

[1835] The server coordinates the interview date and time with the specialist to be interviewed and stores the schedule information in a database, thereby managing the specialist to participate in the interview at an appropriate time.

[1836] Step 3:

[1837] When the scheduled interview date and time arrives, the server starts the interview phase and connects the specialist with the AI ​​interviewer. The generative model automatically asks questions and collects the specialist's answers in real time. At the same time, the emotion engine also runs to collect the user's emotional data.

[1838] Step 4:

[1839] The server uses an emotion engine to analyze the user's emotional data collected during the interview in real time, assessing their emotions based on, for example, their facial expressions, tone of voice, and selected survey answers.

[1840] Step 5:

[1841] The server uses a generative model to generate specialist profiles and articles based on interview response data and sentiment data. These profiles and articles systematically organize knowledge and experience.

[1842] Step 6:

[1843] The server refers to a user profile database and recommends generated content to appropriate users based on their interests, past browsing history, and emotional feedback. For example, a newly generated article about a pottery expert will be provided to users who are interested in pottery and have previously viewed related content and had a positive emotional response.

[1844] Terminal processing steps

[1845] Step 1:

[1846] If a specialist is selected for an interview, the device will notify them of the interview date and time and how to participate. This notification will be provided via email and in-app notification, where the specialist can view detailed interview information.

[1847] Step 2:

[1848] At the designated date and time, the device will initiate the interview session and connect with the AI ​​interviewer in real time, while the emotion engine will simultaneously collect and analyze the user's emotional data.

[1849] Step 3:

[1850] Users can view the generated profiles and articles through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[1851] User processing steps

[1852] Step 1:

[1853] If the user participates in the interview as a specialist, they will check the interview details based on the notification on their device, and at the designated date and time, they will participate in the interview phase through their device.

[1854] Step 2:

[1855] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1856] Step 3:

[1857] Users can provide feedback on the content and share articles they find particularly interesting with other users or on social media, further spreading their expertise and interests. This feedback is analyzed by the sentiment engine and sent to the server.

[1858] Specific examples

[1859] For example, if film director B is the subject of an interview, the processing flow is as follows:

[1860] Server Processing Steps

[1861] Step 1:

[1862] Generate a question for Director B: "Which directors influenced you?"

[1863] Step 2:

[1864] The interview date and time is arranged and a notification is sent to Director B.

[1865] Step 3:

[1866] At the specified date and time, film director B will participate in the interview session via the terminal.

[1867] Step 4:

[1868] Conduct interviews in real time and collect response and sentiment data.

[1869] Step 5:

[1870] Based on the collected data, a profile and article about film director B is generated.

[1871] Step 6:

[1872] The generated articles are recommended to users who are interested in movies and who have a positive reaction based on emotional feedback.

[1873] Terminal processing steps

[1874] Step 1:

[1875] Notify Director B of the interview date and time and how to participate.

[1876] Step 2:

[1877] Director B begins the interview session using a device to participate in the interview. At this time, the emotion engine also starts to collect and analyze emotion data.

[1878] Step 3:

[1879] Users can view the generated articles on their devices and provide feedback, which is also analyzed by the emotion engine and sent to the server.

[1880] User processing steps

[1881] Step 1:

[1882] Film director B checks the interview details based on the notification.

[1883] Step 2:

[1884] The user views an article about recommended film director B.

[1885] Step 3:

[1886] Users provide feedback on articles and share it on social media, etc. This feedback is analyzed by the emotion engine and sent to the server.

[1887] Example 2

[1888] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1889] There is a need for a method to effectively share the knowledge and experience of experts and provide appropriate content based on users' interests and emotions. However, conventional methods have been unable to fully utilize the quality of interview content or users' emotional feedback, making it difficult to recommend content that is highly relevant to users. This has resulted in insufficient dissemination of expert knowledge and improvement of user satisfaction.

[1890] 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.

[1891] In this invention, the server includes means for generating interview questions using a generative model, means for connecting the interviewed expert with an AI interviewer to conduct the interview, means for collecting and analyzing user data during the interview using an emotion engine, means for analyzing the collected interview data and emotion data to generate expert profiles and articles, and means for recommending the generated content to appropriate users based on the user's interests, past browsing history, and emotion feedback. This makes it possible to effectively share and spread the knowledge and experience of experts and provide users with content that is highly relevant and satisfying.

[1892] A "generative model" is a machine learning-based model that generates new text or information based on given prompts and data.

[1893] The "means for generating interview questions" is a mechanism that uses a generative model to automatically create specific questions for the expert being interviewed.

[1894] The "means of conducting the interview" is a mechanism that connects the subject expert with an AI interviewer and conducts the actual interview.

[1895] The "emotion engine" is a system that analyzes data such as the user's facial expressions, tone of voice, and selected survey responses to evaluate the user's emotions.

[1896] "Means for collecting and analyzing user data" refers to a system for collecting data obtained from users during interviews and analyzing their emotions and interests based on that data.

[1897] "Interview data" refers collectively to expert responses and other related information collected through interviews.

[1898] The "means for generating profiles and articles" is a mechanism for generating expert profiles and related articles based on the collected interview data and sentiment data.

[1899] The "user profile database" is a database for storing and managing information such as user interests, past browsing history, and emotional feedback.

[1900] "Means for recommending content" refers to a system that recommends appropriate content to users based on their interests, past browsing history, and emotional feedback.

[1901] This invention is a system for sharing and spreading expert knowledge and experience by utilizing a generative model and an emotion engine. The system aims to promote the sharing of expert knowledge by recommending content generated based on the user's interests and emotions to appropriate users. A specific embodiment of this system is described below.

[1902] Server Processing

[1903] 1. Generate interview questions:

[1904] The server generates interview questions using a generative AI model (e.g., GPT-4). At this time, instructions such as "Create interview questions for a film director" are input as prompts. An example of a question generated by the generative model is "Which film directors have influenced you?"

[1905] 2. Interview scheduling and management:

[1906] The server searches for relevant experts based on the generated interview questions and schedules the interview. This information is stored in a database and is referenced when the interview is conducted.

[1907] 3. Conducting the interview:

[1908] When the time for the interview arrives, the server connects the expert to the AI ​​interviewer and conducts the interview. The generative model automatically asks questions and collects the expert's answers in real time. At the same time, the emotion engine runs to analyze the user's emotional data.

[1909] 4. Emotional data collection and analysis:

[1910] The server uses an emotion engine to collect and analyze data such as the user's facial expressions, tone of voice, and questionnaire responses during the interview to evaluate the user's emotions.

[1911] 5. Profile and article generation:

[1912] Based on the interview and sentiment analysis data, the server uses a generative AI model to generate expert profiles and articles that include a multifaceted breakdown of expertise and experience, as well as sentiment data.

[1913] 6. User Recommendations:

[1914] The server references a user profile database and recommends content based on the user's interests, past browsing history, and emotional feedback. For example, a user interested in pottery can be recommended a newly generated article by a pottery expert.

[1915] Terminal handling

[1916] 1. Notification of Interview Participation:

[1917] The device will notify the interviewed expert of the date and time of the interview and how to participate. This notification will be provided via email and in-app notification, and the expert will be able to view detailed interview information.

[1918] 2. Interview phase begins:

[1919] At the specified date and time, the device will automatically start the interview session and connect the AI ​​interviewer with the expert in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[1920] 3. Viewing and Feedback:

[1921] Users can view the generated profiles and articles through their devices and can also provide feedback based on their impressions while browsing, which is analyzed by the emotion engine and sent to the server.

[1922] User Action

[1923] 1. Interview Participation:

[1924] When participating in an interview as an expert, the user will check the interview details based on notifications from the device, and at the designated date and time, they will participate in the interview phase through the device.

[1925] 2. Viewing profiles and articles:

[1926] Users can browse recommended profiles and articles of interest through their devices, for example, an amateur pottery enthusiast can read an interview article with a pottery expert.

[1927] 3. Feedback and sharing:

[1928] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[1929] Specific examples

[1930] For example, if Director B is chosen for an interview, the following process will occur:

[1931] 1. The server uses the generative model to generate a question for the film director: "Which film directors have influenced you?"

[1932] 2. Arrange an interview date and time and send a notification to Director B.

[1933] 3. On the specified date and time, Director B will join the interview session via the terminal.

[1934] 4. The server conducts the interview in real time and collects the response data. At the same time, the emotion engine analyzes Director B's emotions.

[1935] 5. Generate a profile and article about film director B based on the collected data.

[1936] 6. Recommend generated articles to users who are interested in movies and have had a positive emotional reaction to past related articles.

[1937] 7. Movie fan users view the generated articles on their devices and provide their impressions as feedback that is analyzed by the emotion engine.

[1938] This system integrates an emotion engine to share the knowledge and experience of experts widely and deeply, enabling it to provide highly relevant content to many users.

[1939] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1940] Step 1: Generate interview questions

[1941] Input: Interview request from user, prompt

[1942] Specific behavior:

[1943] The server receives a request from the user to "interview a film director." The server inputs the prompt "Create interview questions for the film director" into a generative AI model (e.g., GPT-4).

[1944] Data processing / calculation:

[1945] The generative AI model analyzes the prompt and generates new interview questions, referencing question generation patterns from an existing database.

[1946] Output: Generated interview questions (e.g., "Which film directors influenced you?")

[1947] Step 2: Schedule and manage interviews

[1948] Input: Interview questions, expert schedule data

[1949] Specific behavior:

[1950] The server searches the database for relevant experts (e.g., film directors) based on the generated question, checks the experts' schedules, and arranges the date and time of the interview.

[1951] Data processing / calculation:

[1952] It looks at the expert's calendar, automatically identifies available time slots, and selects the appropriate time.

[1953] Output: List of experts with interview date and time (e.g., Film Director B, October 1, 2023, 2:00 PM)

[1954] Step 3: Conducting the interview

[1955] Input: Interview questions, expert connections

[1956] Specific behavior:

[1957] When the time for the interview arrives, the server sends a notification to the expert (e.g., film director B) to start the interview. The server then starts the AI ​​interviewer and delivers the generated questions one by one.

[1958] Data processing / calculation:

[1959] The AI ​​interviewer collects the expert's answers in real time and stores them in a database, while the emotion engine collects facial and vocal data and analyzes the emotional data.

[1960] Output: Expert response data, sentiment data collected in real time

[1961] Step 4: Collect and analyze emotion data

[1962] Input: Expert response data, user sentiment data

[1963] Specific behavior:

[1964] During the interview, the server uses an emotion engine to collect and analyze emotional data from the user in real time, such as facial expressions, tone of voice, and questionnaire responses.

[1965] Data processing / calculation:

[1966] The collected data is evaluated using facial expression analysis software and tone analysis software to determine and quantify positive, negative, and neutral emotions.

[1967] Output: Analyzed sentiment data (e.g., 80% positive, 10% negative, 10% neutral)

[1968] Step 5: Generate profiles and articles

[1969] Input: Expert response data, analyzed sentiment data

[1970] Specific behavior:

[1971] The server instructs the generative AI model to generate expert profiles and articles based on the collected response data and sentiment data.

[1972] Data processing / calculation:

[1973] The generative model analyzes this data, organizes the knowledge and experience of experts from various perspectives, and converts them into article format, while also appropriately reflecting sentiment data.

[1974] Output: Expert profile and article (e.g., article about film director B)

[1975] Step 6: Recommend to users

[1976] Input: User interest data, past browsing history, emotional feedback

[1977] Specific behavior:

[1978] The server references a user profile database and selects content based on interests, past browsing history, and emotional feedback, and then notifies the selected content to the target user.

[1979] Data processing / calculation:

[1980] It uses recommendation algorithms to surface content that matches your interests, and also uses your browsing history and emotional feedback to select the most relevant articles.

[1981] Output: Content notification to user (e.g. "A new article about pottery has been published")

[1982] (Application example 2)

[1983] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1984] The lack of a system that widely shares the knowledge and experience of experts and appropriately recommends them based on user interests and emotions is another issue. Another issue is the lack of technology that effectively utilizes user feedback and emotional data to organize and provide expert content from multiple perspectives.

[1985] 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 generating interview questions using a generative model, means for connecting the interviewed expert with an AI interviewer and conducting the interview, means for analyzing the collected interview data to generate expert profiles and articles, means for recommending content generated based on the user's interests, past browsing history, and emotional feedback to appropriate users, and means for analyzing the user's feedback with an emotion engine. This makes it possible to effectively share and recommend experts' knowledge and experience based on the user's emotions and interests.

[1986] A "generative model" refers to an algorithm that automatically generates new content, questions, etc. based on data.

[1987] "Interview questions" are questions asked to experts, and these questions are generated using a generative model.

[1988] An "expert" is an individual who has knowledge or experience in a particular field.

[1989] "AI interviewer" refers to a system or software that uses artificial intelligence to ask interview questions and collect answers.

[1990] "Interview Data" refers to the experts' responses and other relevant information collected during the interview.

[1991] A "profile" is information that outlines an expert's knowledge and experience, created based on collected data.

[1992] "Articles" refer to written information created based on interview data, and are detailed descriptions of the experts' knowledge and experiences.

[1993] "Recommendation" refers to recommending appropriate content based on a user's interests, past browsing history, and emotional feedback.

[1994] "Emotional feedback" refers to the analysis of users' emotional reactions when viewing content and the data and comments based on that.

[1995] An "emotion engine" refers to software or algorithms that analyze a user's emotions from facial expressions, tone of voice, text feedback, etc.

[1996] "Content" refers to the collection of information, entertainment, etc. provided to users through communications means.

[1997] "System" refers to an integrated mechanical or software collection of related means and devices.

[1998] This invention is a system that shares the knowledge and experience of experts and recommends appropriate content based on the user's interests and emotions. The detailed configuration and operation procedures for implementing this invention are described below.

[1999] Server Processing

[2000] 1. Generate interview questions:

[2001] The server uses a generative model to generate interview questions for domain experts, such as a natural language processing model like BERT (Bidirectional Encoder Representations from Transformers).

[2002] Example: Generating questions for a film director such as "Which film directors have influenced you?"

[2003] 2. Interview scheduling and management:

[2004] The server coordinates the interview date and time with the interview subject and stores the schedule information in a database, using the Python datetime module for schedule management.

[2005] Example: An interview with a film director is scheduled for "2023-10-01 15:00:00".

[2006] 3. Conducting the interview:

[2007] When the interview date and time arrives, the server starts the interview phase and connects the expert with the AI ​​interviewer. The generative model automatically asks questions and collects the expert's answers in real time. At the same time, the emotion engine runs and analyzes the user's emotional data.

[2008] Hardware and software used: GPU-based servers, generative AI models (e.g., GPT-3), emotion analysis software (e.g., facial emotion recognition API)

[2009] 4. Emotional data collection and analysis:

[2010] The server analyzes user data collected during the interview in real time using an emotion engine, which evaluates emotions based on data such as the user's facial expressions, tone of voice, and selected survey answers.

[2011] Example: Determining user satisfaction and interest in an expert's answer in real time.

[2012] 5. Profile and article generation:

[2013] Based on the interview and sentiment analysis data, the server uses a generative model to generate expert profiles and articles that include a multifaceted organization of expertise and experience, as well as sentiment data.

[2014] Hardware and software used: Database management system (e.g., PostgreSQL), natural language generation model (e.g., OpenAI's GPT-3)

[2015] 6. User Recommendations:

[2016] The server references a user profile database and recommends content to appropriate users based on their interests, past browsing history, and emotional feedback.

[2017] Example: A newly generated article by a ceramics expert is recommended to users who are interested in ceramics.

[2018] Terminal handling

[2019] 1. Notification of Interview Participation:

[2020] Once selected, the device will notify experts of the interview date and time and how to participate, via email and in-app notification.

[2021] Example: "Your interview is scheduled for 2023-10-01 15:00:00"

[2022] 2. Interview phase begins:

[2023] At the appointed date and time, the expert will initiate a session on the device and connect with the AI ​​interviewer in real time. This connection also runs an emotion engine, which collects and analyzes the user's emotional data.

[2024] Hardware and software used: Webcam, microphone, emotion analysis software

[2025] 3. Viewing and Feedback:

[2026] Users can view the profiles and articles generated through their devices and provide feedback based on their impressions while browsing. This feedback is also analyzed by the emotion engine and sent to the server.

[2027] Example: "This article was very helpful."

[2028] User Action

[2029] 1. Interview Participation:

[2030] If the user participates in the interview as an expert, they will check the interview details based on the notification from the device, and at the designated date and time, they will participate in the interview phase through the device.

[2031] Example: A user completes an interview at a specified date and time.

[2032] 2. Viewing profiles and articles:

[2033] Users browse recommended profiles and articles that interest them through their devices, and the emotion engine analyzes their emotions in real time and sends emotional data as feedback.

[2034] Example: "I was reading an article by a ceramics expert and it was very interesting."

[2035] 3. Feedback and sharing:

[2036] Users can provide feedback on content and share articles they find particularly interesting with other users and on social media, further spreading expertise and interest.

[2037] Example prompts: "What would a movie-loving user like to read next?", "Recommendations for movie-related interviews."

[2038] The above is a specific embodiment for carrying out the invention. This invention makes it possible to widely share the knowledge and experience of experts and to effectively recommend content based on the interests and emotions of users.

[2039] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2040] Step 1:

[2041] The server generates the interview questions

[2042] The server uses a generative model (e.g., a natural language processing model such as BERT) to generate appropriate questions for the expert being interviewed. In this process, the generative model receives a prompt (e.g., "Which other film directors have influenced this film director?") as input and generates a question as output. Specifically, the generative model analyzes the prompt, extracts relevant information from a database, and generates a new question.

[2043] Step 2:

[2044] The server schedules the interviews

[2045] The server arranges the interview date and time with the expert and saves the schedule information in a database. The expert's free time and the server's free time are referenced as input, and the optimal date and time is automatically selected. The schedule information is then saved in the database and is also output as notification data. Specifically, the date and time are set using the Python datetime module, and the process is performed to save it in a database such as PostgreSQL.

[2046] Step 3:

[2047] The server begins the interview phase

[2048] When the interview date and time arrives, the server starts the interview phase and connects the expert with the AI ​​interviewer. At this time, the generative model automatically asks questions and collects the expert's answers in real time. The emotion engine also runs and analyzes the user's emotional data. The input is the interview questions and the expert's answers, and the output is the generated answer data. Specifically, the AI ​​interviewer conducts the interview, and the exchange is collected as text data.

[2049] Step 4:

[2050] The server collects and analyzes emotion data

[2051] During the interview phase, the server uses an emotion engine to analyze emotional data such as the user's facial expressions, tone of voice, and selected survey answers. The input is the user's real-time reaction data, and the output is analyzed emotional evaluation data. Specifically, emotion analysis software (e.g., facial expression analysis API or voice emotion analysis API) is used to analyze the emotional data in real time, and the results are stored on the server.

[2052] Step 5:

[2053] Server generates profiles and articles

[2054] Based on the interview and sentiment analysis data, the server uses a generative model to generate expert profiles and articles. These profiles and articles organize expertise and experience from multiple angles and also include sentiment data. The input is the interview response data and sentiment analysis data, and the output is the generated profile and article. Specifically, a natural language generation model (e.g., GPT-3) is used to convert the data into text and create expert profiles and articles.

[2055] Step 6:

[2056] The server recommends content to the user

[2057] The server references a user profile database and recommends content to appropriate users based on their interests, past browsing history, and emotional feedback. The input is profile information from the user database, past browsing history, and emotional feedback, and the output is recommended content. Specifically, a machine learning model is used to analyze the user's past behavior and feedback and recommend the next content to be viewed.

[2058] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2059] 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: / / o...

Claims

1. means for generating interview questions using the generative model; A means of connecting the interviewee expert with an AI interviewer to conduct the interview; a means of analysing the collected interview data to generate expert profiles and articles; A means of recommending content to appropriate users based on their interests and past browsing history, and A system including:

2. 2. The system according to claim 1, wherein the interview execution means comprises means for conducting interviews in real time and collecting response data.

3. 2. The system according to claim 1, wherein the means for generating profiles and articles comprises means for organizing the knowledge and experience of experts from various angles using the collected data.

Images