System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving an idea input by a user; means for analyzing the input idea and automatically generating a plurality of characters; means for the generated characters to discuss; means for analyzing the discussion content and deriving a conclusion; and means for presenting the conclusion to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] When it comes to the idea generation and decision-making processes that unleash creativity, individual thinking has its limitations, and biased perspectives and emotional judgments are common challenges. This can hinder the creation of high-quality ideas and objective decision-making. Furthermore, in team discussions, the opinions of specific members can be swayed, making it difficult to reach a consensus. Therefore, there is a need for a system that can solve these problems, compensate for users' biased perspectives and emotions, and support deep idea generation and objective decision-making. [Means for solving the problem]

[0005] To solve this problem, the present invention provides a system including: a means for receiving ideas input by a user; a means for automatically generating multiple characters by analyzing the input ideas; a means for the generated characters to hold discussions; a means for analyzing the content of the discussion and deriving a conclusion; and a means for presenting the conclusion to the user. In particular, the characters have roles such as moderator, bold, cautious, and balancer, thereby enabling discussions from different perspectives. Furthermore, the roles of the characters can be customized by the user, allowing for flexible discussions according to the situation. This allows users to gain diverse perspectives, generate deeper ideas, and make objective decisions.

[0006] A "user" is an entity that uses the system to input ideas and direct the progress of discussions.

[0007] An "idea" is a concept or proposal for solving a problem that a user submits to the system.

[0008] "Analysis" is the process by which the system analyzes the input ideas to understand them and generate discussions from diverse perspectives.

[0009] "Characters" are virtual entities that are automatically generated by the system and have specific roles to play in discussions.

[0010] "Automatic generation" is the process by which a system generates a character based on user input using algorithms or AI.

[0011] The "moderator" is a character whose role is to moderate the discussion and keep it running smoothly.

[0012] "Bold" characters are those who come up with new ideas and challenging proposals.

[0013] The "cautious" character is responsible for assessing risks and suggesting a cautious approach.

[0014] A "Balancer" is a character whose role is to harmonize different opinions and perspectives during a discussion and maintain overall balance.

[0015] A "discussion" is a dialogue in which generated characters exchange opinions with each other and move toward problem-solving and decision-making.

[0016] A "conclusion" is the optimal solution or decision reached as a result of the discussion.

[0017] "Presentation" is the process by which the system displays the generated conclusions to the user visually or audibly.

[0018] "Customizable" refers to the ability for users to set and change the role and characteristics of their characters on the system to suit their needs. [Brief explanation of the drawings]

[0019] [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

[0020] 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.

[0021] First, the terms used in the following description will be explained.

[0022] 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).

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 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.

[0030] 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).

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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."

[0040] The present invention is a system in which multiple generated characters develop a discussion based on an idea input by a user and reach a conclusion. The specific processing of the program of this system is explained in natural language below. The embodiment will also be explained in detail with specific examples.

[0041] Program processing

[0042] 1. User Idea Submissions

[0043] The user inputs the idea into the input form on the terminal.

[0044] The device receives the input ideas, converts them into a specific format (e.g., JSON format), and sends them to the server.

[0045] 2. Character Generation

[0046] The server analyzes the received ideas using natural language processing technology.

[0047] Based on the content of the idea, the server automatically generates the characters needed to smoothly advance the discussion (e.g., moderator, bold, cautious, balancer).

[0048] The profile and role of each character is determined and sent to the terminal.

[0049] 3. Starting a discussion

[0050] The terminal visually displays the generated characters and their roles to the user.

[0051] When a user clicks a button to start a discussion, the terminal sends a request to start a discussion to the server.

[0052] 4. Discussion Progress

[0053] The server generates utterances based on each character's role, including using AI models (e.g., GPT-3).

[0054] The server sequentially transmits the generated comments to the terminal, which then displays them to the user.

[0055] Users can join the discussion and add questions or new ideas, which are also sent to the server and reflected in the discussion.

[0056] 5. Drawing conclusions

[0057] The server analyzes the content of the discussion in real time and draws the optimal conclusion.

[0058] The server sends the generated conclusion to the terminal, which displays the final conclusion to the user.

[0059] Specific examples

[0060] Below are some specific examples on the theme of "market introduction strategies for new products."

[0061] 1. User Idea Submissions

[0062] The user enters "I would like to think of a market introduction strategy for a new health food" into the input form on the terminal.

[0063] The device converts this idea into JSON format and sends it to the server.

[0064] 2. Character Generation

[0065] The server analyzes the ideas received and extracts keywords such as "healthy food," "market introduction," and "strategy."

[0066] Based on this, the server will automatically generate characters for the moderator, bolder, cautious, and balancer.

[0067] The server determines the role and profile of each character and sends this to the device.

[0068] 3. Starting a discussion

[0069] The terminal displays information about the generated character to the user.

[0070] A discussion begins when the user clicks the "Start Discussion" button.

[0071] 4. Discussion Progress

[0072] The moderator says, "Now let's hear from the bold ones."

[0073] The bold suggestion is to launch a new social media campaign and collaborate with influencers.

[0074] Cautious voices say, "The idea is good, but we need to carefully evaluate the cost-benefit and risks."

[0075] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[0076] A user joins the discussion and types, "Should we do additional cost estimation?"

[0077] The device sends this new idea to the server and the discussion is updated.

[0078] 5. Drawing conclusions

[0079] The server analyzes the content of the discussion and generates a conclusion that "we will conduct a trial SNS campaign and gradually expand it based on the results."

[0080] The server sends this conclusion to the terminal, which displays it to the user.

[0081] This specific embodiment allows users to generate deeper ideas and make objective decisions through discussions from a variety of perspectives.

[0082] The processing flow will be explained below.

[0083] Step 1:

[0084] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[0085] Step 2:

[0086] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[0087] Step 3:

[0088] The idea data stored on the server is analyzed using natural language processing (NLP). As a result of the analysis, key topics and keywords of the ideas are extracted.

[0089] Step 4:

[0090] Based on the analysis results, the server automatically generates the characters necessary for the discussion (moderator, bold, cautious, balancer, etc.), and retrieves the character profiles and roles from a database.

[0091] Step 5:

[0092] The character information generated by the server is formatted in JSON format and sent to the terminal.

[0093] Step 6:

[0094] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[0095] Step 7:

[0096] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[0097] Step 8:

[0098] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[0099] Step 9:

[0100] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[0101] Step 10:

[0102] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[0103] Step 11:

[0104] The server updates the discussion based on the new input received, and new comments are generated by the characters based on the updated content.

[0105] Step 12:

[0106] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[0107] Step 13:

[0108] The server formats the generated conclusion in JSON format and sends it to the device, which visually displays the received conclusion to the user.

[0109] Step 14:

[0110] The user confirms the displayed conclusion, which allows the user to reach the optimal solution or decision that was reached through discussion of different perspectives.

[0111] Example 1

[0112] 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."

[0113] In the traditional idea generation and discussion process, users are responsible for all thinking alone, leading to problems of biased perspectives and reduced efficiency. Furthermore, incorporating multiple different opinions takes a lot of time and effort, making it difficult to quickly reach an optimal conclusion. Furthermore, there is a lack of support for users to smoothly facilitate discussions, which can lead to confusion.

[0114] 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.

[0115] In this invention, the server includes means for receiving information input by a user, means for converting the input information into a specific format and transmitting it to the server, means for analyzing the information received by the server using natural language processing technology, means for automatically generating multiple profiles based on the analysis, means for visually displaying the information in the generated profiles, means for the profiles to start a discussion, means for the server to analyze the content of the discussion and draw a conclusion, and means for presenting the conclusion to the user. This allows users to more effectively and quickly reach optimal conclusions through discussions from diverse perspectives. Furthermore, the discussion progresses smoothly, reducing bias in perspectives and reduced efficiency.

[0116] "User" refers to a person who uses this system to input and operate information.

[0117] "Information" refers to the ideas and data that users input into the system.

[0118] "Means" refers to methods or apparatuses for implementing the functions or processes described in the present invention.

[0119] A "profile" is information about the attributes and roles of a virtual character generated by the server for discussion purposes.

[0120] A "server" is a central computer system that receives and analyzes information from users, generates profiles, and manages discussions.

[0121] "Natural language processing technology" is a technology that allows computers to understand and analyze human language.

[0122] "Analysis" is the process of understanding, classifying, and evaluating input information.

[0123] "Discussion" is the process by which generated profiles exchange ideas and opinions and reach a conclusion.

[0124] A "conclusion" is a final judgment or recommendation that is reached as a result of a discussion.

[0125] "Visually displaying" means graphically presenting profile information and discussion content in a format that is easy for users to view.

[0126] The present invention is a system in which multiple profiles are generated based on information entered by a user, and discussions are held to reach a conclusion. A specific embodiment of the program for this system is described below. In particular, the hardware and software used, as well as specific data processing and calculations, are described in detail.

[0127] Hardware and Software Configuration

[0128] This system mainly uses the following hardware and software:

[0129] Cloud computing platforms as servers (e.g., Amazon Web Services, Google Cloud Platform)

[0130] User devices as terminals (e.g., PCs, tablets, smartphones)

[0131] NLP engines as natural language processing technologies (e.g., spaCy)

[0132] Language models as generative AI models (e.g., GPT-3)

[0133] Secure protocol used for sending and receiving data (e.g. HTTPS)

[0134] JSON format as data format

[0135] Program processing

[0136] 1. Enter your information:

[0137] The user enters information into the input form on the terminal.

[0138] The terminal receives the input information, converts it into JSON format, and sends it to the server.

[0139] 2. Analysis of Information and Profile Creation:

[0140] The information received by the server is analyzed using natural language processing technology (e.g., spaCy).

[0141] Based on the analysis results, the server automatically generates the profiles (e.g., moderator, bold, cautious, balancer) necessary to smoothly advance the discussion using a generative AI model (e.g., GPT-3).

[0142] The server sends the generated profile information to the terminal in JSON format.

[0143] 3. Visual representation of the profile:

[0144] The device visually displays the received profile information to the user using HTML, CSS, and Javascript.

[0145] 4. Starting a discussion:

[0146] When a user clicks the discussion start button, the terminal sends a request to start a discussion to the server.

[0147] The server generates messages for each profile and transmits the messages to the terminals one by one.

[0148] The terminal displays the statement to the user.

[0149] 5. Developing a discussion and drawing conclusions:

[0150] Users can join the discussion and enter new questions or ideas, which are also sent to the server and reflected in the discussion.

[0151] The server analyzes the discussion content in real time and generates the optimal conclusion.

[0152] The server sends the generated conclusion to the terminal, which displays the conclusion to the user.

[0153] Specific examples

[0154] Below are some specific examples on the theme of "market introduction strategies for new products."

[0155] 1. User Idea Submissions:

[0156] The user enters "I would like to think of a market introduction strategy for a new health food" into the input form on the terminal.

[0157] The device converts this information into JSON format and sends it to the server.

[0158] 2. Generate and view the profile:

[0159] The server analyzes the information it receives and extracts keywords such as "health food," "market introduction," and "strategy."

[0160] Based on this, the server generates profiles for Moderator, Bold, Cautious, and Balancer.

[0161] The server transmits the generated profile information to the terminal.

[0162] The terminal displays the profile information to the user.

[0163] 3. Initiating and Conducting a Discussion:

[0164] The terminal displays a "Start Discussion" button, and when the user clicks on it, the discussion begins.

[0165] The moderator says, "Now let's hear from the bold ones."

[0166] The bold suggestion is to launch a new social media campaign and collaborate with influencers.

[0167] Cautious voices say, "The idea is good, but we need to carefully evaluate the cost-benefit and risks."

[0168] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[0169] The user joins the discussion and types, "Should we do additional cost estimation?" The device sends this information to the server.

[0170] 4. Draw conclusions:

[0171] The server analyzes the content of the discussion and generates a conclusion that "we will conduct a trial SNS campaign and gradually expand it based on the results."

[0172] The server sends this conclusion to the terminal, which displays it to the user.

[0173] Examples of prompt statements

[0174] Here are some examples of prompts to input to a generative AI model:

[0175] User: I would like to think about a market introduction strategy for a new health food.

[0176] Server: The moderator (you) is responsible for keeping the discussion running smoothly.

[0177] Moderator: "Now let's hear from the bold ones."

[0178] Bold: "We should launch a new social media campaign and collaborate with influencers."

[0179] Cautious: "The idea is good, but the costs and risks need to be carefully assessed."

[0180] Balancer: "Both opinions are important. Let's compare the specific costs and expected benefits."

[0181] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0182] Step 1: Enter your information

[0183] The user inputs information (e.g., "I would like to consider a market introduction strategy for a new health food product") into an input form on the terminal.

[0184] Input: Information entered by the user into the device.

[0185] The device processes this input information using UTF-8 character encoding and converts it into a JSON-formatted object.

[0186] Data processing: Character encoding of information and conversion to JSON format.

[0187] The device sends the generated JSON object to the server using the secure HTTPS protocol.

[0188] Output: Reformatted JSON information.

[0189] Step 2: Analyze the information and create a profile

[0190] The server parses the received JSON data and uses natural language processing techniques (e.g., spaCy) to extract important keywords and phrases.

[0191] Input: JSON formatted information sent from the device.

[0192] Data processing: Keyword extraction and phrase analysis using a natural language processing engine.

[0193] Based on the extracted keywords, the server automatically generates profiles (e.g., moderator, bold, cautious, balancer) necessary to smoothly advance the discussion using a generative AI model (e.g., GPT-3).

[0194] Data calculation: Profile generation using AI models.

[0195] The server determines the attributes of each profile (e.g., name, personality, expertise, etc.) and packages them into a JSON format.

[0196] Output: JSON format data of generated profile information.

[0197] The server sends the generated profile information to the device using the secure HTTPS protocol.

[0198] Step 3: Visual representation of the profile

[0199] Renders a user interface (UI) that displays the profile information received by the device.

[0200] Input: JSON data of profile information received from the server.

[0201] The device uses HTML, CSS, and JavaScript to visually display the profile's name, role, profile picture, etc.

[0202] Behavior: Rendering and displaying the UI.

[0203] Output: Profile information visually displayed to the user.

[0204] Step 4: Start a discussion

[0205] When a user clicks the Start Discussion button, a JavaScript event handler on the device detects this.

[0206] Input: The action of the user clicking the Start Discussion button.

[0207] The terminal composes a request to start a discussion in JSON format and sends it to the server.

[0208] Data processing: Composing a discussion start request.

[0209] Output: Discussion start request sent to the server.

[0210] Step 5: Facilitate the discussion

[0211] The server receives a request to start a discussion and starts the process of generating comments for each profile.

[0212] Input: A request to start a discussion from a terminal.

[0213] The server uses a generative AI model (e.g., GPT-3) to generate utterances based on the role of each profile.

[0214] Data computation: utterance generation using AI models.

[0215] The server sequentially sends the generated comments to the terminal in JSON format.

[0216] Output: JSON data of the generated utterance.

[0217] The device displays the comments received in real time, allowing users to visually check the progress of the discussion.

[0218] What it does: Real-time display of what's being said.

[0219] Users can join the discussion and enter new questions or ideas, which are also sent back to the server and reflected in the discussion.

[0220] Input: New questions and ideas.

[0221] Data arithmetic: Processing new input information and reflecting it in discussions.

[0222] Step 6: Draw conclusions

[0223] The server analyzes all discussion content in real time and applies algorithms to derive optimal conclusions based on the data.

[0224] Input: All data from the discussion.

[0225] Data arithmetic: Applying conclusion-drawing algorithms.

[0226] The server generates the final result and sends it to the device in JSON format.

[0227] Output: JSON data of the generated conclusions.

[0228] The terminal displays the received conclusion on a user interface to provide the user with a final conclusion.

[0229] Action: Display of final conclusion.

[0230] (Application example 1)

[0231] 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."

[0232] In the development and operation of autonomous vehicles, there is a need to efficiently hold discussions from a variety of expert perspectives and derive objective, optimal conclusions. Conventional methods often make it difficult to reconcile opinions between experts and advance discussions, requiring time and effort. Furthermore, there is a problem in that it is difficult for users to actively participate in discussions, resulting in only a portion of opinions being reflected. In response to these issues, a system that supports efficient and fair discussions and allows users to actively participate is desired.

[0233] 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.

[0234] In this invention, the server includes means for receiving a task input by a user, means for analyzing the task content and automatically generating a plurality of discussion characters, means for the generated discussion characters to engage in a discussion, means for analyzing the discussion content and deriving a conclusion, means for presenting the conclusion to the user, and means for the user to input additional questions or suggestions while the discussion is in progress. This enables efficient and fair discussions from various perspectives and provides an environment in which users can actively participate in the discussion.

[0235] "User" refers to a person who inputs data into or operates the system.

[0236] "Issues" refer to problems that users want to solve or themes that they want to consider.

[0237] "Means of receiving" refers to the interface or function that allows the system to acquire tasks from users.

[0238] "Analyzing" refers to the process of understanding the content of the input task and extracting the necessary information.

[0239] "Discussion characters" refer to virtual roles automatically generated for the purpose of conducting discussions.

[0240] "Means of automatic generation" refers to the process by which the system uses AI technology to automatically create discussion characters.

[0241] "Means for holding discussions" refers to a function that allows the generated discussion characters to converse and exchange opinions with each other.

[0242] "Means of deriving a conclusion" refers to the process for generating the most appropriate conclusion based on the discussion that has taken place.

[0243] "Presentation means" refers to the interface or functionality for informing the user of the generated conclusions.

[0244] "Means for inputting additional questions and suggestions" refers to the function that allows users to input new questions or suggestions to the system during the discussion.

[0245] This invention is a system for efficiently conducting discussions from various expert viewpoints and deriving objective and optimal conclusions in the development and operation of autonomous vehicles. This system uses automatically generated discussion characters to hold discussions based on issues entered by the user, and analyzes the results to reach a conclusion.

[0246] System configuration

[0247] 1. User Device

[0248] Hardware: User interfaces such as smartphones and head-mounted displays

[0249] Software: input forms, display interfaces, communication modules

[0250] 2. Server

[0251] Hardware: Servers with high-performance computing capabilities (including cloud servers)

[0252] Software: Natural language processing technology (e.g., GPT-3), data analysis module, character generation module, discussion progression module, conclusion drawing module

[0253] System Operation Overview

[0254] 1. User Device

[0255] The user inputs a problem, such as "I want to think of a market launch strategy for a new autonomous driving algorithm," into an input form.

[0256] This input task is converted into JSON format and sent to the server.

[0257] 2. Server

[0258] The server analyzes the received problem and extracts keywords, such as "autonomous driving," "algorithm," and "market launch."

[0259] Based on the extracted keywords, the necessary discussion characters are automatically generated, including "technical experts," "safety managers," "management managers," and "user representatives."

[0260] The role and profile of each character are determined and transmitted to the user terminal.

[0261] 3. Discussion Progress

[0262] The user terminal displays the generated character information, and the user starts a discussion.

[0263] The server generates utterances according to the character's role and sends them to the user's terminal in sequence. During the discussion, the user can input additional questions or suggestions.

[0264] 4. Drawing conclusions

[0265] The server analyzes the content of the discussion and derives the optimal conclusion, which is then sent to the user's terminal and presented to the user.

[0266] Specific examples

[0267] Below is a concrete example of a discussion regarding a market launch strategy for a new autonomous driving algorithm.

[0268] Example prompt sentence:

[0269] As a technology expert, please discuss "Go-to-market strategies for new autonomous driving algorithms."

[0270] As a safety manager, please discuss the "market launch strategy for new autonomous driving algorithms."

[0271] As a manager, please discuss the "market launch strategy for new autonomous driving algorithms."

[0272] As a user representative, please discuss the "market launch strategy for new autonomous driving algorithms."

[0273] This system utilizes generative AI models (e.g., GPT-3) to enable efficient and fair discussions from diverse perspectives, allowing users to actively participate in the discussion and derive deeper ideas and objective conclusions.

[0274] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0275] Step 1:

[0276] The user inputs a task into the device's input form. This task may include a specific problem such as "I want to think about a market launch strategy for a new autonomous driving algorithm." The device converts this input into JSON format and sends it to the server.

[0277] Input: The assignment entered by the user

[0278] Output: Issue data in JSON format

[0279] Step 2:

[0280] The server analyzes the received JSON-formatted problem data and extracts keywords, such as "autonomous driving," "algorithm," and "market launch." This analysis uses natural language processing technology (e.g., GPT-3).

[0281] Input: Issue data in JSON format

[0282] Output: Extracted keyword list

[0283] Step 3:

[0284] The server generates the necessary discussion characters based on the extracted keywords. The generated characters include "technical expert," "safety manager," "management manager," and "user representative." It generates a prompt for each character and determines the character's profile and role using an AI model (e.g., GPT-3).

[0285] Input: Keyword list

[0286] Output: Character profile and role

[0287] Step 4:

[0288] The server sends the generated character's profile and role to the user's device, which receives it and displays it visually to the user. The user can start a discussion by clicking the "Start Discussion" button on their device.

[0289] Input: Character profile and role

[0290] Output: Display character information to the user

[0291] Step 5:

[0292] The server generates utterances based on the character's role. This utterance generation uses an AI model (e.g., GPT-3). The generated utterances are sequentially sent to the user's device and displayed to the user.

[0293] Input: Character profile and role

[0294] Output: What the character says

[0295] Step 6:

[0296] Users can enter additional questions or suggestions into the discussion on their terminals. These additional inputs are also converted into JSON format and sent to the server, which receives them and updates the discussion.

[0297] Input: User's additional questions or suggestions

[0298] Output: Updated discussion

[0299] Step 7:

[0300] The server analyzes the entire discussion and derives the optimal conclusion. This analysis and conclusion generation uses an AI model (e.g., GPT-3). The server then sends the final conclusion to the user's device.

[0301] Input: Updated discussion

[0302] Output: Conclusion

[0303] Step 8:

[0304] The user terminal displays the received conclusion to the user, who can then confirm the conclusion and use it to decide what action to take next.

[0305] Input: Conclusion

[0306] Output: Display conclusion to user

[0307] 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.

[0308] The present invention is a system in which multiple generated characters develop a discussion based on an idea input by a user, and a conclusion is reached by combining an emotion engine while recognizing the user's emotional state. The specific processing of the program of this system is explained in natural language below. Also, specific examples are provided to explain the embodiment in detail.

[0309] Program processing

[0310] 1. User Idea Submissions

[0311] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[0312] 2. Character Generation

[0313] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[0314] The server analyzes the stored idea data using natural language processing (NLP), and extracts key topics and keywords from the analysis results.

[0315] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer), and retrieves the character profiles and roles from the database.

[0316] The character information generated by the server is formatted in JSON format and sent to the terminal.

[0317] 3. Starting a discussion

[0318] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[0319] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[0320] 4. Operation of the Emotion Engine

[0321] The device sends the user's input and actions (e.g., keystroke speed, mouse movements, facial recognition, etc.) to the emotion engine in real time.

[0322] The emotion engine analyzes this data to detect the user's emotional state (e.g., stress, excitement, concentration, etc.).

[0323] The device transmits the detected emotional state to the server.

[0324] 5. Discussion Progress

[0325] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[0326] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[0327] The server receives data from the emotion engine and adjusts the character's speech and tone based on the user's emotional state.

[0328] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[0329] 6. Drawing conclusions

[0330] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[0331] The server formats the generated conclusion in JSON format and sends it to the device, which visually displays the received conclusion to the user.

[0332] Specific examples

[0333] Below are some specific examples on the theme of "market introduction strategies for new products."

[0334] 1. User Idea Submissions

[0335] The user enters the idea into the input form on the device, "I want to think of a market introduction strategy for a new health food." The device converts this idea into JSON format and sends it to the server.

[0336] 2. Character Generation

[0337] The server analyzes the ideas received and extracts keywords such as "healthy food," "market introduction," and "strategy."

[0338] Based on this, the server automatically generates characters for the moderator, bold, cautious, and balancer. The server determines the role and profile of each character and sends this to the device.

[0339] 3. Starting a discussion

[0340] The terminal displays information about the generated character to the user. The user can start a discussion by clicking the "Start Discussion" button.

[0341] 4. Operation of the Emotion Engine

[0342] The device sends the user's input and behavioral data (e.g., keystroke speed and facial expression) to the emotion engine, which then detects the user's emotional state. The detected emotional state is then sent from the device to the server.

[0343] 5. Discussion Progress

[0344] The moderator says, "Now let's hear from the bold ones."

[0345] The bold suggest, "You should launch a social media campaign and collaborate with influencers."

[0346] Cautious voices say, "The idea is good, but we need to evaluate the cost-benefit and risks."

[0347] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[0348] The server adjusts the bolder's tone of speech, for example softening it, depending on the user's emotional state.

[0349] A user joins the discussion and types, "Should we do additional cost estimation?" The device sends this idea to the server and the discussion is updated.

[0350] 6. Drawing conclusions

[0351] The server analyzes the content of the discussion and generates a conclusion: "We will conduct a trial SNS campaign and gradually expand it based on the results." The conclusion is sent to the device and displayed to the user.

[0352] This specific embodiment allows users to discuss with characters with diverse perspectives, encouraging flexible responses that reflect their emotional state, enabling them to generate deeper ideas and make more objective decisions.

[0353] The processing flow will be explained below.

[0354] Step 1:

[0355] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[0356] Step 2:

[0357] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[0358] Step 3:

[0359] The idea data stored on the server is analyzed using natural language processing (NLP), which extracts key topics and keywords for the ideas.

[0360] Step 4:

[0361] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer), and retrieves each character's profile and role from the database.

[0362] Step 5:

[0363] The character information generated by the server is formatted in JSON format and sent to the terminal.

[0364] Step 6:

[0365] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[0366] Step 7:

[0367] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[0368] Step 8:

[0369] The device sends the user's input and actions (e.g., keystroke speed, mouse movement, facial recognition, etc.) to the emotion engine in real time. The emotion engine analyzes this data and detects the user's emotional state.

[0370] Step 9:

[0371] The device sends the emotional state detected by the emotion engine to the server, which then adjusts the content and tone of the character's speech based on the emotional state received.

[0372] Step 10:

[0373] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[0374] Step 11:

[0375] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[0376] Step 12:

[0377] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[0378] Step 13:

[0379] The server updates the discussion based on the new input received, and new comments are generated by the characters based on the updated content.

[0380] Step 14:

[0381] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[0382] Step 15:

[0383] The server formats the generated conclusion in JSON format and sends it to the device.

[0384] Step 16:

[0385] The terminal visually displays the received conclusion to the user, who then confirms the displayed conclusion. This allows the user to obtain the optimal solution or decision-making result obtained through discussion from different perspectives.

[0386] Example 2

[0387] 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."

[0388] In conventional idea generation support systems, it was difficult to discuss ideas entered by users from various perspectives and to respond flexibly while taking into account their emotional state. Furthermore, the generated characters' comments were not adjusted according to the user's emotions, making it difficult to advance effective discussions. The present invention aims to solve these problems and provide a system that effectively supports the user's idea generation process from multiple angles.

[0389] 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.

[0390] In this invention, the server includes means for receiving ideas input by a user, means for converting the input ideas into JSON format, and means for transmitting the converted JSON-formatted idea data to the server. This allows users to easily input ideas and process them efficiently. The server also includes means for analyzing the saved idea data using natural language processing to extract key topics and keywords, means for automatically generating multiple characters based on the analysis results, and means for formatting the generated character information in JSON format and transmitting it to a terminal. This allows users to generate characters with diverse perspectives and engage in appropriate discussions. The server also includes means for an emotion engine to detect the user's emotional state, means for adjusting the content and tone of the character's comments based on the user's emotional state, and means for utilizing a generative AI model to generate comments appropriate to each character's role. This enables flexible responses that reflect the user's emotional state, improving the quality of discussions.

[0391] A "user" is a person who operates the system and inputs ideas, and is the primary user of the system.

[0392] A "terminal" is an input device operated by a user, and is a device for inputting ideas and displaying information.

[0393] "Server" refers to a central management system that stores, analyzes, and processes data sent by users.

[0394] "JSON format" is an abbreviation for JavaScript Object Notation, and is a lightweight data exchange format that makes it easy to exchange data.

[0395] "Conversion" refers to the process of changing input data into another format.

[0396] A "POST request" is a method of sending data to a server using the HyperText Transfer Protocol (HTTP).

[0397] "Natural Language Processing (NLP)" is a technology that allows computers to understand, analyze, and generate human language.

[0398] A "character" is a virtual person automatically generated for the purpose of discussion, with a specific role and profile.

[0399] The "Emotion Engine" is a machine learning model for analyzing and detecting a user's emotional state.

[0400] A "generative AI model" refers to an artificial intelligence model used to perform tasks such as natural language generation and dialogue generation.

[0401] A "prompt" is an instruction given to a generative AI model that the model uses to generate an appropriate response or generation.

[0402] "Discussion" refers to the process in which multiple characters exchange opinions on a user's idea.

[0403] A "conclusion" is the final result or proposal reached through a discussion.

[0404] This invention is a system that supports users in discussing ideas from various perspectives and responding flexibly. This system is realized using users, terminals, a server, a generative AI model, and an emotion engine.

[0405] First, the user inputs an idea into the input form on the device. For example, this input might be in the form of "I want to think of a market introduction strategy for a new health food." This idea is converted into JSON format by the device. The device then sends this JSON-formatted idea data to the server via a POST request.

[0406] Next, the server stores the JSON-formatted idea data received from the device. A NoSQL database (e.g., MongoDB) is used for this storage. The server then analyzes the stored idea data using natural language processing (NLP). An NLP library (e.g., spaCy or NLTK) is used for this analysis, and key topics and keywords for the ideas (e.g., "healthy food," "market introduction," and "strategy") are extracted. Based on this, characters are automatically generated to participate in the discussion. Character profile and role information is retrieved from a database (e.g., PostgreSQL).

[0407] The generated character information is formatted in JSON format by the server and sent to the device. The device parses the received character information and visually displays it to the user. For example, the character's name, role, and profile information are displayed. At this point, the user clicks the "Start Discussion" button to begin the discussion.

[0408] When a discussion begins, the device collects user input and behavioral data (e.g., keystroke speed, mouse movement, and facial recognition data) in real time and sends them to the emotion engine. The emotion engine then analyzes this data using a machine learning model (e.g., a TensorFlow model) to detect the user's emotional state. The device then transmits the detected emotional state to the server.

[0409] The server uses a generative AI model (e.g., GPT-3) to generate utterances appropriate for each character's role. A prompt is used to generate the utterances. An example of a prompt is: "As a bold person, please give your opinion on the market launch strategy for health foods."

[0410] The comments generated by the server are sent to the device in sequence, and the comments received by the device are displayed to the user in chat format. This uses a conversational UI, making it appear to the user that the characters are engaged in a dialogue. The server receives data from the emotion engine and adjusts the content and tone of the characters' comments based on the user's emotional state. For example, if the user is relaxed, it softens the tone of the bolder comments. Users participate in the discussion, inputting unclear points and new ideas. Input such as "Should we do additional cost estimates?" is sent from the device to the server, and the discussion is updated.

[0411] Once the discussion has ended, the server analyzes the discussion log and stores each character's comments in a database. The server generates an optimal conclusion based on the content of the discussion, formats the conclusion into JSON format, and sends it to the device. The device then visually displays this conclusion to the user. A specific conclusion might be something like, "We will conduct a trial SNS campaign and gradually expand it based on the results."

[0412] In this way, users can interact with characters with diverse perspectives, responding flexibly to their emotional state, leading to deeper idea generation and objective decision-making.

[0413] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0414] Step 1:

[0415] The user inputs their idea into the input form on the device. For example, they might input, "I want to think of a market introduction strategy for a new health food." The device receives the user's input and converts it into JSON format. Input data: User's idea (text format) → Output data: Idea (JSON format).

[0416] Step 2:

[0417] The terminal sends the converted JSON-formatted idea data to the server via a POST request. Input data: JSON-formatted idea data → Output data: Send request to the server.

[0418] Step 3:

[0419] The server stores the JSON format idea data received from the device. A NoSQL database (e.g., MongoDB) is used. Input data: JSON format idea data → Output data: idea data stored in the database.

[0420] Step 4:

[0421] The server analyzes the stored idea data using natural language processing (NLP). It uses an NLP library (e.g., spaCy or NLTK) to extract key topics and keywords for ideas. Input data: idea data (JSON format) → Processing: NLP analysis → Output data: keywords and topics.

[0422] Step 5:

[0423] The server automatically generates characters for discussions based on the analysis results. Character profiles and role information are obtained from a database (e.g., PostgreSQL). Input data: Keywords and topics → Processing content: Automatic character generation, profile acquisition → Output data: Character information (JSON format).

[0424] Step 6:

[0425] The server formats the generated character information in JSON format and sends it to the terminal. Input data: Character information → Processing content: Format into JSON format → Output data: Send request to the terminal.

[0426] Step 7:

[0427] The device analyzes the character information received and displays it visually to the user. Using HTML and CSS, a UI is generated to display the character's name, role, and profile information. Input data: character information (JSON format) → Processing: Parse, UI generation → Output data: user interface.

[0428] Step 8:

[0429] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start a discussion to the server. Input data: User's click operation → Output data: Request to start a discussion.

[0430] Step 9:

[0431] The device collects user input and behavioral data (e.g., keystroke speed, mouse movement, facial recognition data) in real time and sends it to the emotion engine. Input data: behavioral data → Output data: transmission request to the emotion engine.

[0432] Step 10:

[0433] The emotion engine analyzes the data using a machine learning model (e.g., TensorFlow model) and detects the user's emotional state. Input data: behavioral data → Processing content: detection of emotional state → Output data: user's emotional state.

[0434] Step 11:

[0435] The server receives data from the emotion engine and adjusts the content and tone of the character's speech based on the user's emotional state. Input data: User's emotional state → Processing content: Adjustment of speech content and tone → Output data: Adjusted speech.

[0436] Step 12:

[0437] The server uses a generative AI model (e.g., GPT-3) to generate utterances appropriate for each character's role. A sample prompt might be, "As a bold person, please share your opinion on the market launch strategy for health foods."

[0438] Input data: Character role, prompt sentence → Processing content: Speech generation by AI model → Output data: Generated speech.

[0439] Step 13:

[0440] The server sends generated comments to the device in sequence. The device displays the received comments to the user in chat format. Displayed in an interactive UI. Input data: Generated comments → Processing content: Displayed in chat → Output data: Displayed to the user.

[0441] Step 14:

[0442] The user joins the discussion and inputs any unclear points or new ideas. For example, "Should we do additional cost estimates?" The device sends this content to the server, and the discussion is updated. Input data: User's new idea → Output data: Send request to the server.

[0443] Step 15:

[0444] The server analyzes the discussion log and stores each character's comments in a database. The optimal conclusion is generated based on the content of the discussion. Input data: discussion log → Processing content: NLP analysis, conclusion generation → Output data: generated conclusion.

[0445] Step 16:

[0446] The server formats the generated conclusion into JSON format and sends it to the device. The device visually displays the conclusion to the user. A specific conclusion might be something like "We will conduct a trial SNS campaign and gradually expand it based on the results." Input data: Generated conclusion → Processing content: Format into JSON format, display to user → Output data: Conclusion presented to the user.

[0447] (Application example 2)

[0448] 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."

[0449] In conventional systems, multiple characters could discuss and reach a conclusion based on ideas input by the user, but they were unable to reflect the user's emotional state in real time. This made it difficult to respond appropriately to the user's emotions or to flexibly advance the discussion, ultimately causing a decline in user satisfaction and the quality of the discussion. In customer interactions and customer service in physical stores, there was no system that could take customer emotions into account, so improvements in customer satisfaction could not be expected.

[0450] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0451] In this invention, the server includes means for receiving ideas input by users, means for analyzing the input ideas and automatically generating multiple characters, means for the generated characters to hold discussions, means for analyzing the content of the discussion and deriving a conclusion, means for presenting the conclusion to users, means for emotionally analyzing customer input and behavioral data in real time, and means for adjusting the characters' comments based on the emotional state. This enables flexible discussions that reflect the emotional state of users in real time, which is expected to improve customer satisfaction in physical stores.

[0452] "User" refers to an individual or group that uses the system to input ideas and participate in discussions.

[0453] "Ideas" refer to the ideas and thoughts that users input into the system, and are the information that forms the basis of discussion.

[0454] "Characters" refer to virtual characters who are automatically generated within the system based on the user's ideas and who play various roles in discussions.

[0455] A "discussion" refers to a conversation or debate process in which multiple characters express their opinions based on their respective roles and reach a conclusion.

[0456] "Conclusion" refers to the result or proposal that is ultimately reached after analyzing the content of the discussion.

[0457] "Emotion analysis" refers to the process of analyzing user input and behavioral data to detect the user's emotional state in real time.

[0458] "Adjusting speech" refers to changing the content and tone of a character's speech based on emotion analysis to generate appropriate speech that corresponds to the user's emotional state.

[0459] A "physical store" is a facility located in a physical location where customers can visit in person to provide goods or services.

[0460] "Customer" refers to a visitor who visits a physical store and intends to purchase or use a product or service.

[0461] "Customer service" refers to activities that involve interacting with and providing support to customers in physical stores, with the aim of improving customer satisfaction.

[0462] "Emotion engine" refers to an analysis device or program that detects the user's emotional state and reflects it in processes within the system.

[0463] The present invention is a system in which multiple characters discuss ideas input by a user and use an emotion engine to reach a conclusion while reflecting the user's emotional state in real time. This system aims to improve customer service in brick-and-mortar stores. Specific embodiments for realizing this system are described below.

[0464] 1. System Overview

[0465] Hardware used

[0466] Device: Smartphone, tablet or PC. Device used by staff or customers in a physical store.

[0467] Server: A remotely located computing device that analyzes data and runs generative AI models.

[0468] Software used

[0469] Natural Language Processing (NLP): A technique used to analyze user-entered ideas.

[0470] Generative AI model (e.g. GPT-3): A model used to generate character utterances.

[0471] Emotion engine: Software for analyzing a user's emotional state in real time.

[0472] 2. Program processing explanation

[0473] 1. Idea input and analysis

[0474] The user inputs an idea on the terminal, which is converted into JSON format and sent to the server.

[0475] The server analyzes the received ideas using natural language processing (NLP) to extract key topics and keywords.

[0476] 2. Character Generation

[0477] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer).

[0478] The role and profile of each character is retrieved from the database, and the generated character information is sent to the terminal.

[0479] 3. Starting a discussion

[0480] The terminal displays the received character information to the user, and the discussion begins when the user clicks the "Start discussion" button.

[0481] This discussion is initiated by sending a request to the server to start the discussion.

[0482] 4. Operation of the Emotion Engine

[0483] The device transmits user input and actions (e.g., keystroke speed, mouse movement, facial recognition) to the emotion engine in real time.

[0484] The emotion engine analyzes these data and detects the user's emotional state (e.g., stress, excitement, concentration, etc.), which is then sent to the server.

[0485] 5. Discussion Progress

[0486] The server uses a generative AI model to generate each character's utterances. For example, the "cautious" character might say, "A risk assessment is required."

[0487] The server adjusts the content and tone of the character's speech depending on the user's emotional state.

[0488] 6. Drawing conclusions

[0489] The server analyzes the discussion logs and generates an optimal conclusion, which is formatted in JSON and sent to the device.

[0490] The terminal visually displays the received conclusions to the user.

[0491] 3. Specific Examples

[0492] Example: If a customer enters "I'm looking for a light summer jacket"

[0493] Input prompt statement:

[0494] User Wants: A lightweight summer jacket

[0495] Bolder opinion: Experiment with different colors and designs!

[0496] Cautious observers: It's also a good idea to check the quality and durability of the materials.

[0497] Balancer's opinion: Why not take a look at some mid-priced items with good design and quality?

[0498] In this example, users can receive advice from multiple perspectives and receive real-time responses based on emotion analysis, resulting in a more satisfying purchasing experience. This system can improve the quality of customer service in physical stores.

[0499] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0500] Step 1:

[0501] Entering and receiving ideas

[0502] The user inputs ideas using a terminal. This input is often in text format and is specific, such as "I would like to come up with a market introduction strategy for a new health food."

[0503] The device receives this idea and converts it into JSON format, for example, {"idea": "I want to think about a market launch strategy for a new health food product"}.

[0504] The terminal sends the converted JSON data to the server, which receives it and stores it in a database.

[0505] Step 2:

[0506] Idea analysis and character generation

[0507] The server analyzes the received JSON-formatted idea data using natural language processing (NLP) technology. The purpose of the analysis is to extract keywords and topics for the ideas. The analysis results include keywords such as "healthy food," "market introduction," and "strategy."

[0508] Based on the analysis results, the server automatically generates characters to participate in the discussion. These characters include the moderator, the bold, the cautious, and the balancer. Each character's profile and role are retrieved from a database.

[0509] The server formats the generated character information in JSON format and sends it to the device. An example of formatted data is as follows: {"characters": [{"role": "Moderator", "profile": "Neutral moderator"}, {"role": "Bold", "profile": "Provides challenging opinions"}, {"role": "Cautious", "profile": "Provides risk-oriented opinions"}, {"role": "Balancer", "profile": "Reconcile the opinions of both parties"}]}

[0510] Step 3:

[0511] View character information and start discussions

[0512] The device parses the received character information and displays it for the user to visually confirm, with each character's role and profile displayed on the screen.

[0513] The user clicks the "Start Discussion" button. The device captures this click event and sends a request to start a discussion to the server.

[0514] Step 4:

[0515] Emotion analysis using an emotion engine

[0516] The device monitors user input and actions in real time, including keystroke speed, mouse movements, and facial recognition.

[0517] For example, if the user is typing quickly, the emotion engine detects an "excited" state.

[0518] The detected emotional state is sent to the server in JSON format, e.g. {"emotion": "excited"}

[0519] Step 5:

[0520] Discussion management and coordination

[0521] The server uses a generative AI model (e.g., GPT-3) to generate statements for each character. For example, the Cautious character might say, "That's a good idea, but we need to evaluate the cost-benefit and risks."

[0522] The server adjusts the content and tone of the characters' speech depending on the user's emotional state. For example, if the user is excited, the "Bold" character might suggest in a calmer tone, "You should launch a social media campaign and collaborate with influencers."

[0523] The server sends each character's comments in JSON format to the device, which then displays them to the user in chat format. The discussion is updated as users enter new ideas or questions.

[0524] Step 6:

[0525] Drawing and presenting conclusions

[0526] The server analyzes the discussion log and generates an optimal conclusion based on the discussion, such as "run a trial SNS campaign and gradually expand it based on the results."

[0527] This conclusion is formatted as JSON and sent to the device. Example: {"conclusion": "Test social media campaigns and gradually expand based on the results"}

[0528] The terminal visually displays the received conclusions to the user.

[0529] 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.

[0530] 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.

[0531] 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.

[0532] [Second embodiment]

[0533] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0534] 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.

[0535] 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).

[0536] 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.

[0537] 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.

[0538] 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).

[0539] 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.

[0540] 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.

[0541] 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.

[0542] 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.

[0543] 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.

[0544] 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."

[0545] The present invention is a system in which multiple generated characters develop a discussion based on an idea input by a user and reach a conclusion. The specific processing of the program of this system is explained in natural language below. The embodiment will also be explained in detail with specific examples.

[0546] Program processing

[0547] 1. User Idea Submissions

[0548] The user inputs the idea into the input form on the terminal.

[0549] The device receives the input ideas, converts them into a specific format (e.g., JSON format), and sends them to the server.

[0550] 2. Character Generation

[0551] The server analyzes the received ideas using natural language processing technology.

[0552] Based on the content of the idea, the server automatically generates the characters needed to smoothly advance the discussion (e.g., moderator, bold, cautious, balancer).

[0553] The profile and role of each character is determined and sent to the terminal.

[0554] 3. Starting a discussion

[0555] The terminal visually displays the generated characters and their roles to the user.

[0556] When a user clicks a button to start a discussion, the terminal sends a request to start a discussion to the server.

[0557] 4. Discussion Progress

[0558] The server generates utterances based on each character's role, including using AI models (e.g., GPT-3).

[0559] The server sequentially transmits the generated comments to the terminal, which then displays them to the user.

[0560] Users can join the discussion and add questions or new ideas, which are also sent to the server and reflected in the discussion.

[0561] 5. Drawing conclusions

[0562] The server analyzes the content of the discussion in real time and draws the optimal conclusion.

[0563] The server sends the generated conclusion to the terminal, which displays the final conclusion to the user.

[0564] Specific examples

[0565] Below are some specific examples on the theme of "market introduction strategies for new products."

[0566] 1. User Idea Submissions

[0567] The user enters "I would like to think of a market introduction strategy for a new health food" into the input form on the terminal.

[0568] The device converts this idea into JSON format and sends it to the server.

[0569] 2. Character Generation

[0570] The server analyzes the ideas received and extracts keywords such as "healthy food," "market introduction," and "strategy."

[0571] Based on this, the server will automatically generate characters for the moderator, bolder, cautious, and balancer.

[0572] The server determines the role and profile of each character and sends this to the device.

[0573] 3. Starting a discussion

[0574] The terminal displays information about the generated character to the user.

[0575] A discussion begins when the user clicks the "Start Discussion" button.

[0576] 4. Discussion Progress

[0577] The moderator says, "Now let's hear from the bold ones."

[0578] The bold suggestion is to launch a new social media campaign and collaborate with influencers.

[0579] Cautious voices say, "The idea is good, but we need to carefully evaluate the cost-benefit and risks."

[0580] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[0581] A user joins the discussion and types, "Should we do additional cost estimation?"

[0582] The device sends this new idea to the server and the discussion is updated.

[0583] 5. Drawing conclusions

[0584] The server analyzes the content of the discussion and generates a conclusion that "we will conduct a trial SNS campaign and gradually expand it based on the results."

[0585] The server sends this conclusion to the terminal, which displays it to the user.

[0586] This specific embodiment allows users to generate deeper ideas and make objective decisions through discussions from a variety of perspectives.

[0587] The processing flow will be explained below.

[0588] Step 1:

[0589] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[0590] Step 2:

[0591] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[0592] Step 3:

[0593] The idea data stored on the server is analyzed using natural language processing (NLP). As a result of the analysis, key topics and keywords of the ideas are extracted.

[0594] Step 4:

[0595] Based on the analysis results, the server automatically generates the characters necessary for the discussion (moderator, bold, cautious, balancer, etc.), and retrieves the character profiles and roles from a database.

[0596] Step 5:

[0597] The character information generated by the server is formatted in JSON format and sent to the terminal.

[0598] Step 6:

[0599] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[0600] Step 7:

[0601] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[0602] Step 8:

[0603] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[0604] Step 9:

[0605] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[0606] Step 10:

[0607] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[0608] Step 11:

[0609] The server updates the discussion based on the new input received, and new comments are generated by the characters based on the updated content.

[0610] Step 12:

[0611] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[0612] Step 13:

[0613] The server formats the generated conclusion in JSON format and sends it to the device, which visually displays the received conclusion to the user.

[0614] Step 14:

[0615] The user confirms the displayed conclusion, which allows the user to reach the optimal solution or decision that was reached through discussion of different perspectives.

[0616] Example 1

[0617] 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."

[0618] In the traditional idea generation and discussion process, users are responsible for all thinking alone, leading to problems of biased perspectives and reduced efficiency. Furthermore, incorporating multiple different opinions takes a lot of time and effort, making it difficult to quickly reach an optimal conclusion. Furthermore, there is a lack of support for users to smoothly facilitate discussions, which can lead to confusion.

[0619] 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.

[0620] In this invention, the server includes means for receiving information input by a user, means for converting the input information into a specific format and transmitting it to the server, means for analyzing the information received by the server using natural language processing technology, means for automatically generating multiple profiles based on the analysis, means for visually displaying the information in the generated profiles, means for the profiles to start a discussion, means for the server to analyze the content of the discussion and draw a conclusion, and means for presenting the conclusion to the user. This allows users to more effectively and quickly reach optimal conclusions through discussions from diverse perspectives. Furthermore, the discussion progresses smoothly, reducing bias in perspectives and reduced efficiency.

[0621] "User" refers to a person who uses this system to input and operate information.

[0622] "Information" refers to the ideas and data that users input into the system.

[0623] "Means" refers to methods or apparatuses for implementing the functions or processes described in the present invention.

[0624] A "profile" is information about the attributes and roles of a virtual character generated by the server for discussion purposes.

[0625] A "server" is a central computer system that receives and analyzes information from users, generates profiles, and manages discussions.

[0626] "Natural language processing technology" is a technology that allows computers to understand and analyze human language.

[0627] "Analysis" is the process of understanding, classifying, and evaluating input information.

[0628] "Discussion" is the process by which generated profiles exchange ideas and opinions and reach a conclusion.

[0629] A "conclusion" is a final judgment or recommendation that is reached as a result of a discussion.

[0630] "Visually displaying" means graphically presenting profile information and discussion content in a format that is easy for users to view.

[0631] The present invention is a system in which multiple profiles are generated based on information entered by a user, and discussions are held to reach a conclusion. A specific embodiment of the program for this system is described below. In particular, the hardware and software used, as well as specific data processing and calculations, are described in detail.

[0632] Hardware and Software Configuration

[0633] This system mainly uses the following hardware and software:

[0634] Cloud computing platforms as servers (e.g., Amazon Web Services, Google Cloud Platform)

[0635] User devices as terminals (e.g., PCs, tablets, smartphones)

[0636] NLP engines as natural language processing technologies (e.g., spaCy)

[0637] Language models as generative AI models (e.g., GPT-3)

[0638] Secure protocol used for sending and receiving data (e.g. HTTPS)

[0639] JSON format as data format

[0640] Program processing

[0641] 1. Enter your information:

[0642] The user enters information into the input form on the terminal.

[0643] The terminal receives the input information, converts it into JSON format, and sends it to the server.

[0644] 2. Analysis of Information and Profile Creation:

[0645] The information received by the server is analyzed using natural language processing technology (e.g., spaCy).

[0646] Based on the analysis results, the server automatically generates the profiles (e.g., moderator, bold, cautious, balancer) necessary to smoothly advance the discussion using a generative AI model (e.g., GPT-3).

[0647] The server sends the generated profile information to the terminal in JSON format.

[0648] 3. Visual representation of the profile:

[0649] The device visually displays the received profile information to the user using HTML, CSS, and Javascript.

[0650] 4. Starting a discussion:

[0651] When a user clicks the discussion start button, the terminal sends a request to start a discussion to the server.

[0652] The server generates messages for each profile and transmits the messages to the terminals one by one.

[0653] The terminal displays the statement to the user.

[0654] 5. Developing a discussion and drawing conclusions:

[0655] Users can join the discussion and enter new questions or ideas, which are also sent to the server and reflected in the discussion.

[0656] The server analyzes the discussion content in real time and generates the optimal conclusion.

[0657] The server sends the generated conclusion to the terminal, which displays the conclusion to the user.

[0658] Specific examples

[0659] Below are some specific examples on the theme of "market introduction strategies for new products."

[0660] 1. User Idea Submissions:

[0661] The user enters "I would like to think of a market introduction strategy for a new health food" into the input form on the terminal.

[0662] The device converts this information into JSON format and sends it to the server.

[0663] 2. Generate and view the profile:

[0664] The server analyzes the information it receives and extracts keywords such as "health food," "market introduction," and "strategy."

[0665] Based on this, the server generates profiles for Moderator, Bold, Cautious, and Balancer.

[0666] The server transmits the generated profile information to the terminal.

[0667] The terminal displays the profile information to the user.

[0668] 3. Initiating and Conducting a Discussion:

[0669] The terminal displays a "Start Discussion" button, and when the user clicks on it, the discussion begins.

[0670] The moderator says, "Now let's hear from the bold ones."

[0671] The bold suggestion is to launch a new social media campaign and collaborate with influencers.

[0672] Cautious voices say, "The idea is good, but we need to carefully evaluate the cost-benefit and risks."

[0673] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[0674] The user joins the discussion and types, "Should we do additional cost estimation?" The device sends this information to the server.

[0675] 4. Draw conclusions:

[0676] The server analyzes the content of the discussion and generates a conclusion that "we will conduct a trial SNS campaign and gradually expand it based on the results."

[0677] The server sends this conclusion to the terminal, which displays it to the user.

[0678] Examples of prompt statements

[0679] Here are some examples of prompts to input to a generative AI model:

[0680] User: I would like to think about a market introduction strategy for a new health food.

[0681] Server: The moderator (you) is responsible for keeping the discussion running smoothly.

[0682] Moderator: "Now let's hear from the bold ones."

[0683] Bold: "We should launch a new social media campaign and collaborate with influencers."

[0684] Cautious: "The idea is good, but the costs and risks need to be carefully assessed."

[0685] Balancer: "Both opinions are important. Let's compare the specific costs and expected benefits."

[0686] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0687] Step 1: Enter your information

[0688] The user inputs information (e.g., "I would like to consider a market introduction strategy for a new health food product") into an input form on the terminal.

[0689] Input: Information entered by the user into the device.

[0690] The device processes this input information using UTF-8 character encoding and converts it into a JSON-formatted object.

[0691] Data processing: Character encoding of information and conversion to JSON format.

[0692] The device sends the generated JSON object to the server using the secure HTTPS protocol.

[0693] Output: Reformatted JSON information.

[0694] Step 2: Analyze the information and create a profile

[0695] The server parses the received JSON data and uses natural language processing techniques (e.g., spaCy) to extract important keywords and phrases.

[0696] Input: JSON formatted information sent from the device.

[0697] Data processing: Keyword extraction and phrase analysis using a natural language processing engine.

[0698] Based on the extracted keywords, the server automatically generates profiles (e.g., moderator, bold, cautious, balancer) necessary to smoothly advance the discussion using a generative AI model (e.g., GPT-3).

[0699] Data calculation: Profile generation using AI models.

[0700] The server determines the attributes of each profile (e.g., name, personality, expertise, etc.) and packages them into a JSON format.

[0701] Output: JSON format data of generated profile information.

[0702] The server sends the generated profile information to the device using the secure HTTPS protocol.

[0703] Step 3: Visual representation of the profile

[0704] Renders a user interface (UI) that displays the profile information received by the device.

[0705] Input: JSON data of profile information received from the server.

[0706] The device uses HTML, CSS, and JavaScript to visually display the profile's name, role, profile picture, etc.

[0707] Behavior: Rendering and displaying the UI.

[0708] Output: Profile information visually displayed to the user.

[0709] Step 4: Start a discussion

[0710] When a user clicks the Start Discussion button, a JavaScript event handler on the device detects this.

[0711] Input: The action of the user clicking the Start Discussion button.

[0712] The terminal composes a request to start a discussion in JSON format and sends it to the server.

[0713] Data processing: Composing a discussion start request.

[0714] Output: Discussion start request sent to the server.

[0715] Step 5: Facilitate the discussion

[0716] The server receives a request to start a discussion and starts the process of generating comments for each profile.

[0717] Input: A request to start a discussion from a terminal.

[0718] The server uses a generative AI model (e.g., GPT-3) to generate utterances based on the role of each profile.

[0719] Data computation: utterance generation using AI models.

[0720] The server sequentially sends the generated comments to the terminal in JSON format.

[0721] Output: JSON data of the generated utterance.

[0722] The device displays the comments received in real time, allowing users to visually check the progress of the discussion.

[0723] What it does: Real-time display of what's being said.

[0724] Users can join the discussion and enter new questions or ideas, which are also sent back to the server and reflected in the discussion.

[0725] Input: New questions and ideas.

[0726] Data arithmetic: Processing new input information and reflecting it in discussions.

[0727] Step 6: Draw conclusions

[0728] The server analyzes all discussion content in real time and applies algorithms to derive optimal conclusions based on the data.

[0729] Input: All data from the discussion.

[0730] Data arithmetic: Applying conclusion-drawing algorithms.

[0731] The server generates the final result and sends it to the device in JSON format.

[0732] Output: JSON data of the generated conclusions.

[0733] The terminal displays the received conclusion on a user interface to provide the user with a final conclusion.

[0734] Action: Display of final conclusion.

[0735] (Application example 1)

[0736] 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."

[0737] In the development and operation of autonomous vehicles, there is a need to efficiently hold discussions from a variety of expert perspectives and derive objective, optimal conclusions. Conventional methods often make it difficult to reconcile opinions between experts and advance discussions, requiring time and effort. Furthermore, there is a problem in that it is difficult for users to actively participate in discussions, resulting in only a portion of opinions being reflected. In response to these issues, a system that supports efficient and fair discussions and allows users to actively participate is desired.

[0738] 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.

[0739] In this invention, the server includes means for receiving a task input by a user, means for analyzing the task content and automatically generating a plurality of discussion characters, means for the generated discussion characters to engage in a discussion, means for analyzing the discussion content and deriving a conclusion, means for presenting the conclusion to the user, and means for the user to input additional questions or suggestions while the discussion is in progress. This enables efficient and fair discussions from various perspectives and provides an environment in which users can actively participate in the discussion.

[0740] "User" refers to a person who inputs data into or operates the system.

[0741] "Issues" refer to problems that users want to solve or themes that they want to consider.

[0742] "Means of receiving" refers to the interface or function that allows the system to acquire tasks from users.

[0743] "Analyzing" refers to the process of understanding the content of the input task and extracting the necessary information.

[0744] "Discussion characters" refer to virtual roles automatically generated for the purpose of conducting discussions.

[0745] "Means of automatic generation" refers to the process by which the system uses AI technology to automatically create discussion characters.

[0746] "Means for holding discussions" refers to a function that allows the generated discussion characters to converse and exchange opinions with each other.

[0747] "Means of deriving a conclusion" refers to the process for generating the most appropriate conclusion based on the discussion that has taken place.

[0748] "Presentation means" refers to the interface or functionality for informing the user of the generated conclusions.

[0749] "Means for inputting additional questions and suggestions" refers to the function that allows users to input new questions or suggestions to the system during the discussion.

[0750] This invention is a system for efficiently conducting discussions from various expert viewpoints and deriving objective and optimal conclusions in the development and operation of autonomous vehicles. This system uses automatically generated discussion characters to hold discussions based on issues entered by the user, and analyzes the results to reach a conclusion.

[0751] System configuration

[0752] 1. User Device

[0753] Hardware: User interfaces such as smartphones and head-mounted displays

[0754] Software: input forms, display interfaces, communication modules

[0755] 2. Server

[0756] Hardware: Servers with high-performance computing capabilities (including cloud servers)

[0757] Software: Natural language processing technology (e.g., GPT-3), data analysis module, character generation module, discussion progression module, conclusion drawing module

[0758] System Operation Overview

[0759] 1. User Device

[0760] The user inputs a problem, such as "I want to think of a market launch strategy for a new autonomous driving algorithm," into an input form.

[0761] This input task is converted into JSON format and sent to the server.

[0762] 2. Server

[0763] The server analyzes the received problem and extracts keywords, such as "autonomous driving," "algorithm," and "market launch."

[0764] Based on the extracted keywords, the necessary discussion characters are automatically generated, including "technical experts," "safety managers," "management managers," and "user representatives."

[0765] The role and profile of each character are determined and transmitted to the user terminal.

[0766] 3. Discussion Progress

[0767] The user terminal displays the generated character information, and the user starts a discussion.

[0768] The server generates utterances according to the character's role and sends them to the user's terminal in sequence. During the discussion, the user can input additional questions or suggestions.

[0769] 4. Drawing conclusions

[0770] The server analyzes the content of the discussion and derives the optimal conclusion, which is then sent to the user's terminal and presented to the user.

[0771] Specific examples

[0772] Below is a concrete example of a discussion regarding a market launch strategy for a new autonomous driving algorithm.

[0773] Example prompt sentence:

[0774] As a technology expert, please discuss "Go-to-market strategies for new autonomous driving algorithms."

[0775] As a safety manager, please discuss the "market launch strategy for new autonomous driving algorithms."

[0776] As a manager, please discuss the "market launch strategy for new autonomous driving algorithms."

[0777] As a user representative, please discuss the "market launch strategy for new autonomous driving algorithms."

[0778] This system utilizes generative AI models (e.g., GPT-3) to enable efficient and fair discussions from diverse perspectives, allowing users to actively participate in the discussion and derive deeper ideas and objective conclusions.

[0779] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0780] Step 1:

[0781] The user inputs a task into the device's input form. This task may include a specific problem such as "I want to think about a market launch strategy for a new autonomous driving algorithm." The device converts this input into JSON format and sends it to the server.

[0782] Input: The assignment entered by the user

[0783] Output: Issue data in JSON format

[0784] Step 2:

[0785] The server analyzes the received JSON-formatted problem data and extracts keywords, such as "autonomous driving," "algorithm," and "market launch." This analysis uses natural language processing technology (e.g., GPT-3).

[0786] Input: Issue data in JSON format

[0787] Output: Extracted keyword list

[0788] Step 3:

[0789] The server generates the necessary discussion characters based on the extracted keywords. The generated characters include "technical expert," "safety manager," "management manager," and "user representative." It generates a prompt for each character and determines the character's profile and role using an AI model (e.g., GPT-3).

[0790] Input: Keyword list

[0791] Output: Character profile and role

[0792] Step 4:

[0793] The server sends the generated character's profile and role to the user's device, which receives it and displays it visually to the user. The user can start a discussion by clicking the "Start Discussion" button on their device.

[0794] Input: Character profile and role

[0795] Output: Display character information to the user

[0796] Step 5:

[0797] The server generates utterances based on the character's role. This utterance generation uses an AI model (e.g., GPT-3). The generated utterances are sequentially sent to the user's device and displayed to the user.

[0798] Input: Character profile and role

[0799] Output: What the character says

[0800] Step 6:

[0801] Users can enter additional questions or suggestions into the discussion on their terminals. These additional inputs are also converted into JSON format and sent to the server, which receives them and updates the discussion.

[0802] Input: User's additional questions or suggestions

[0803] Output: Updated discussion

[0804] Step 7:

[0805] The server analyzes the entire discussion and derives the optimal conclusion. This analysis and conclusion generation uses an AI model (e.g., GPT-3). The server then sends the final conclusion to the user's device.

[0806] Input: Updated discussion

[0807] Output: Conclusion

[0808] Step 8:

[0809] The user terminal displays the received conclusion to the user, who can then confirm the conclusion and use it to decide what action to take next.

[0810] Input: Conclusion

[0811] Output: Display conclusion to user

[0812] 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.

[0813] The present invention is a system in which multiple generated characters develop a discussion based on an idea input by a user, and a conclusion is reached by combining an emotion engine while recognizing the user's emotional state. The specific processing of the program of this system is explained in natural language below. Also, specific examples are provided to explain the embodiment in detail.

[0814] Program processing

[0815] 1. User Idea Submissions

[0816] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[0817] 2. Character Generation

[0818] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[0819] The server analyzes the stored idea data using natural language processing (NLP), and extracts key topics and keywords from the analysis results.

[0820] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer), and retrieves the character profiles and roles from the database.

[0821] The character information generated by the server is formatted in JSON format and sent to the terminal.

[0822] 3. Starting a discussion

[0823] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[0824] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[0825] 4. Operation of the Emotion Engine

[0826] The device sends the user's input and actions (e.g., keystroke speed, mouse movements, facial recognition, etc.) to the emotion engine in real time.

[0827] The emotion engine analyzes this data to detect the user's emotional state (e.g., stress, excitement, concentration, etc.).

[0828] The device transmits the detected emotional state to the server.

[0829] 5. Discussion Progress

[0830] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[0831] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[0832] The server receives data from the emotion engine and adjusts the character's speech and tone based on the user's emotional state.

[0833] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[0834] 6. Drawing conclusions

[0835] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[0836] The server formats the generated conclusion in JSON format and sends it to the device, which visually displays the received conclusion to the user.

[0837] Specific examples

[0838] Below are some specific examples on the theme of "market introduction strategies for new products."

[0839] 1. User Idea Submissions

[0840] The user enters the idea into the input form on the device, "I want to think of a market introduction strategy for a new health food." The device converts this idea into JSON format and sends it to the server.

[0841] 2. Character Generation

[0842] The server analyzes the ideas received and extracts keywords such as "healthy food," "market introduction," and "strategy."

[0843] Based on this, the server automatically generates characters for the moderator, bold, cautious, and balancer. The server determines the role and profile of each character and sends this to the device.

[0844] 3. Starting a discussion

[0845] The terminal displays information about the generated character to the user. The user can start a discussion by clicking the "Start Discussion" button.

[0846] 4. Operation of the Emotion Engine

[0847] The device sends the user's input and behavioral data (e.g., keystroke speed and facial expression) to the emotion engine, which then detects the user's emotional state. The detected emotional state is then sent from the device to the server.

[0848] 5. Discussion Progress

[0849] The moderator says, "Now let's hear from the bold ones."

[0850] The bold suggest, "You should launch a social media campaign and collaborate with influencers."

[0851] Cautious voices say, "The idea is good, but we need to evaluate the cost-benefit and risks."

[0852] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[0853] The server adjusts the bolder's tone of speech, for example softening it, depending on the user's emotional state.

[0854] A user joins the discussion and types, "Should we do additional cost estimation?" The device sends this idea to the server and the discussion is updated.

[0855] 6. Drawing conclusions

[0856] The server analyzes the content of the discussion and generates a conclusion: "We will conduct a trial SNS campaign and gradually expand it based on the results." The conclusion is sent to the device and displayed to the user.

[0857] This specific embodiment allows users to discuss with characters with diverse perspectives, encouraging flexible responses that reflect their emotional state, enabling them to generate deeper ideas and make more objective decisions.

[0858] The processing flow will be explained below.

[0859] Step 1:

[0860] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[0861] Step 2:

[0862] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[0863] Step 3:

[0864] The idea data stored on the server is analyzed using natural language processing (NLP), which extracts key topics and keywords for the ideas.

[0865] Step 4:

[0866] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer), and retrieves each character's profile and role from the database.

[0867] Step 5:

[0868] The character information generated by the server is formatted in JSON format and sent to the terminal.

[0869] Step 6:

[0870] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[0871] Step 7:

[0872] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[0873] Step 8:

[0874] The device sends the user's input and actions (e.g., keystroke speed, mouse movement, facial recognition, etc.) to the emotion engine in real time. The emotion engine analyzes this data and detects the user's emotional state.

[0875] Step 9:

[0876] The device sends the emotional state detected by the emotion engine to the server, which then adjusts the content and tone of the character's speech based on the emotional state received.

[0877] Step 10:

[0878] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[0879] Step 11:

[0880] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[0881] Step 12:

[0882] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[0883] Step 13:

[0884] The server updates the discussion based on the new input received, and new comments are generated by the characters based on the updated content.

[0885] Step 14:

[0886] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[0887] Step 15:

[0888] The server formats the generated conclusion in JSON format and sends it to the device.

[0889] Step 16:

[0890] The terminal visually displays the received conclusion to the user, who then confirms the displayed conclusion. This allows the user to obtain the optimal solution or decision-making result obtained through discussion from different perspectives.

[0891] Example 2

[0892] 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."

[0893] In conventional idea generation support systems, it was difficult to discuss ideas entered by users from various perspectives and to respond flexibly while taking into account their emotional state. Furthermore, the generated characters' comments were not adjusted according to the user's emotions, making it difficult to advance effective discussions. The present invention aims to solve these problems and provide a system that effectively supports the user's idea generation process from multiple angles.

[0894] 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.

[0895] In this invention, the server includes means for receiving ideas input by a user, means for converting the input ideas into JSON format, and means for transmitting the converted JSON-formatted idea data to the server. This allows users to easily input ideas and process them efficiently. The server also includes means for analyzing the saved idea data using natural language processing to extract key topics and keywords, means for automatically generating multiple characters based on the analysis results, and means for formatting the generated character information in JSON format and transmitting it to a terminal. This allows users to generate characters with diverse perspectives and engage in appropriate discussions. The server also includes means for an emotion engine to detect the user's emotional state, means for adjusting the content and tone of the character's comments based on the user's emotional state, and means for utilizing a generative AI model to generate comments appropriate to each character's role. This enables flexible responses that reflect the user's emotional state, improving the quality of discussions.

[0896] A "user" is a person who operates the system and inputs ideas, and is the primary user of the system.

[0897] A "terminal" is an input device operated by a user, and is a device for inputting ideas and displaying information.

[0898] "Server" refers to a central management system that stores, analyzes, and processes data sent by users.

[0899] "JSON format" is an abbreviation for JavaScript Object Notation, and is a lightweight data exchange format that makes it easy to exchange data.

[0900] "Conversion" refers to the process of changing input data into another format.

[0901] A "POST request" is a method of sending data to a server using the HyperText Transfer Protocol (HTTP).

[0902] "Natural Language Processing (NLP)" is a technology that allows computers to understand, analyze, and generate human language.

[0903] A "character" is a virtual person automatically generated for the purpose of discussion, with a specific role and profile.

[0904] The "Emotion Engine" is a machine learning model for analyzing and detecting a user's emotional state.

[0905] A "generative AI model" refers to an artificial intelligence model used to perform tasks such as natural language generation and dialogue generation.

[0906] A "prompt" is an instruction given to a generative AI model that the model uses to generate an appropriate response or generation.

[0907] "Discussion" refers to the process in which multiple characters exchange opinions on a user's idea.

[0908] A "conclusion" is the final result or proposal reached through a discussion.

[0909] This invention is a system that supports users in discussing ideas from various perspectives and responding flexibly. This system is realized using users, terminals, a server, a generative AI model, and an emotion engine.

[0910] First, the user inputs an idea into the input form on the device. For example, this input might be in the form of "I want to think of a market introduction strategy for a new health food." This idea is converted into JSON format by the device. The device then sends this JSON-formatted idea data to the server via a POST request.

[0911] Next, the server stores the JSON-formatted idea data received from the device. A NoSQL database (e.g., MongoDB) is used for this storage. The server then analyzes the stored idea data using natural language processing (NLP). An NLP library (e.g., spaCy or NLTK) is used for this analysis, and key topics and keywords for the ideas (e.g., "healthy food," "market introduction," and "strategy") are extracted. Based on this, characters are automatically generated to participate in the discussion. Character profile and role information is retrieved from a database (e.g., PostgreSQL).

[0912] The generated character information is formatted in JSON format by the server and sent to the device. The device parses the received character information and visually displays it to the user. For example, the character's name, role, and profile information are displayed. At this point, the user clicks the "Start Discussion" button to begin the discussion.

[0913] When a discussion begins, the device collects user input and behavioral data (e.g., keystroke speed, mouse movement, and facial recognition data) in real time and sends them to the emotion engine. The emotion engine then analyzes this data using a machine learning model (e.g., a TensorFlow model) to detect the user's emotional state. The device then transmits the detected emotional state to the server.

[0914] The server uses a generative AI model (e.g., GPT-3) to generate utterances appropriate for each character's role. A prompt is used to generate the utterances. An example of a prompt is: "As a bold person, please give your opinion on the market launch strategy for health foods."

[0915] The comments generated by the server are sent to the device in sequence, and the comments received by the device are displayed to the user in chat format. This uses a conversational UI, making it appear to the user that the characters are engaged in a dialogue. The server receives data from the emotion engine and adjusts the content and tone of the characters' comments based on the user's emotional state. For example, if the user is relaxed, it softens the tone of the bolder comments. Users participate in the discussion, inputting unclear points and new ideas. Input such as "Should we do additional cost estimates?" is sent from the device to the server, and the discussion is updated.

[0916] Once the discussion has ended, the server analyzes the discussion log and stores each character's comments in a database. The server generates an optimal conclusion based on the content of the discussion, formats the conclusion into JSON format, and sends it to the device. The device then visually displays this conclusion to the user. A specific conclusion might be something like, "We will conduct a trial SNS campaign and gradually expand it based on the results."

[0917] In this way, users can interact with characters with diverse perspectives, responding flexibly to their emotional state, leading to deeper idea generation and objective decision-making.

[0918] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0919] Step 1:

[0920] The user inputs their idea into the input form on the device. For example, they might input, "I want to think of a market introduction strategy for a new health food." The device receives the user's input and converts it into JSON format. Input data: User's idea (text format) → Output data: Idea (JSON format).

[0921] Step 2:

[0922] The terminal sends the converted JSON-formatted idea data to the server via a POST request. Input data: JSON-formatted idea data → Output data: Send request to the server.

[0923] Step 3:

[0924] The server stores the JSON format idea data received from the device. A NoSQL database (e.g., MongoDB) is used. Input data: JSON format idea data → Output data: idea data stored in the database.

[0925] Step 4:

[0926] The server analyzes the stored idea data using natural language processing (NLP). It uses an NLP library (e.g., spaCy or NLTK) to extract key topics and keywords for ideas. Input data: idea data (JSON format) → Processing: NLP analysis → Output data: keywords and topics.

[0927] Step 5:

[0928] The server automatically generates characters for discussions based on the analysis results. Character profiles and role information are obtained from a database (e.g., PostgreSQL). Input data: Keywords and topics → Processing content: Automatic character generation, profile acquisition → Output data: Character information (JSON format).

[0929] Step 6:

[0930] The server formats the generated character information in JSON format and sends it to the terminal. Input data: Character information → Processing content: Format into JSON format → Output data: Send request to the terminal.

[0931] Step 7:

[0932] The device analyzes the character information received and displays it visually to the user. Using HTML and CSS, a UI is generated to display the character's name, role, and profile information. Input data: character information (JSON format) → Processing: Parse, UI generation → Output data: user interface.

[0933] Step 8:

[0934] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start a discussion to the server. Input data: User's click operation → Output data: Request to start a discussion.

[0935] Step 9:

[0936] The device collects user input and behavioral data (e.g., keystroke speed, mouse movement, facial recognition data) in real time and sends it to the emotion engine. Input data: behavioral data → Output data: transmission request to the emotion engine.

[0937] Step 10:

[0938] The emotion engine analyzes the data using a machine learning model (e.g., TensorFlow model) and detects the user's emotional state. Input data: behavioral data → Processing content: detection of emotional state → Output data: user's emotional state.

[0939] Step 11:

[0940] The server receives data from the emotion engine and adjusts the content and tone of the character's speech based on the user's emotional state. Input data: User's emotional state → Processing content: Adjustment of speech content and tone → Output data: Adjusted speech.

[0941] Step 12:

[0942] The server uses a generative AI model (e.g., GPT-3) to generate utterances appropriate for each character's role. A sample prompt might be, "As a bold person, please share your opinion on the market launch strategy for health foods."

[0943] Input data: Character role, prompt sentence → Processing content: Speech generation by AI model → Output data: Generated speech.

[0944] Step 13:

[0945] The server sends generated comments to the device in sequence. The device displays the received comments to the user in chat format. Displayed in an interactive UI. Input data: Generated comments → Processing content: Displayed in chat → Output data: Displayed to the user.

[0946] Step 14:

[0947] The user joins the discussion and inputs any unclear points or new ideas. For example, "Should we do additional cost estimates?" The device sends this content to the server, and the discussion is updated. Input data: User's new idea → Output data: Send request to the server.

[0948] Step 15:

[0949] The server analyzes the discussion log and stores each character's comments in a database. The optimal conclusion is generated based on the content of the discussion. Input data: discussion log → Processing content: NLP analysis, conclusion generation → Output data: generated conclusion.

[0950] Step 16:

[0951] The server formats the generated conclusion into JSON format and sends it to the device. The device visually displays the conclusion to the user. A specific conclusion might be something like "We will conduct a trial SNS campaign and gradually expand it based on the results." Input data: Generated conclusion → Processing content: Format into JSON format, display to user → Output data: Conclusion presented to the user.

[0952] (Application example 2)

[0953] 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."

[0954] In conventional systems, multiple characters could discuss and reach a conclusion based on ideas input by the user, but they were unable to reflect the user's emotional state in real time. This made it difficult to respond appropriately to the user's emotions or to flexibly advance the discussion, ultimately causing a decline in user satisfaction and the quality of the discussion. In customer interactions and customer service in physical stores, there was no system that could take customer emotions into account, so improvements in customer satisfaction could not be expected.

[0955] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0956] In this invention, the server includes means for receiving ideas input by users, means for analyzing the input ideas and automatically generating multiple characters, means for the generated characters to hold discussions, means for analyzing the content of the discussion and deriving a conclusion, means for presenting the conclusion to users, means for emotionally analyzing customer input and behavioral data in real time, and means for adjusting the characters' comments based on the emotional state. This enables flexible discussions that reflect the emotional state of users in real time, which is expected to improve customer satisfaction in physical stores.

[0957] "User" refers to an individual or group that uses the system to input ideas and participate in discussions.

[0958] "Ideas" refer to the ideas and thoughts that users input into the system, and are the information that forms the basis of discussion.

[0959] "Characters" refer to virtual characters who are automatically generated within the system based on the user's ideas and who play various roles in discussions.

[0960] A "discussion" refers to a conversation or debate process in which multiple characters express their opinions based on their respective roles and reach a conclusion.

[0961] "Conclusion" refers to the result or proposal that is ultimately reached after analyzing the content of the discussion.

[0962] "Emotion analysis" refers to the process of analyzing user input and behavioral data to detect the user's emotional state in real time.

[0963] "Adjusting speech" refers to changing the content and tone of a character's speech based on emotion analysis to generate appropriate speech that corresponds to the user's emotional state.

[0964] A "physical store" is a facility located in a physical location where customers can visit in person to provide goods or services.

[0965] "Customer" refers to a visitor who visits a physical store and intends to purchase or use a product or service.

[0966] "Customer service" refers to activities that involve interacting with and providing support to customers in physical stores, with the aim of improving customer satisfaction.

[0967] "Emotion engine" refers to an analysis device or program that detects the user's emotional state and reflects it in processes within the system.

[0968] The present invention is a system in which multiple characters discuss ideas input by a user and use an emotion engine to reach a conclusion while reflecting the user's emotional state in real time. This system aims to improve customer service in brick-and-mortar stores. Specific embodiments for realizing this system are described below.

[0969] 1. System Overview

[0970] Hardware used

[0971] Device: Smartphone, tablet or PC. Device used by staff or customers in a physical store.

[0972] Server: A remotely located computing device that analyzes data and runs generative AI models.

[0973] Software used

[0974] Natural Language Processing (NLP): A technique used to analyze user-entered ideas.

[0975] Generative AI model (e.g. GPT-3): A model used to generate character utterances.

[0976] Emotion engine: Software for analyzing a user's emotional state in real time.

[0977] 2. Program processing explanation

[0978] 1. Idea input and analysis

[0979] The user inputs an idea on the terminal, which is converted into JSON format and sent to the server.

[0980] The server analyzes the received ideas using natural language processing (NLP) to extract key topics and keywords.

[0981] 2. Character Generation

[0982] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer).

[0983] The role and profile of each character is retrieved from the database, and the generated character information is sent to the terminal.

[0984] 3. Starting a discussion

[0985] The terminal displays the received character information to the user, and the discussion begins when the user clicks the "Start discussion" button.

[0986] This discussion is initiated by sending a request to the server to start the discussion.

[0987] 4. Operation of the Emotion Engine

[0988] The device transmits user input and actions (e.g., keystroke speed, mouse movement, facial recognition) to the emotion engine in real time.

[0989] The emotion engine analyzes these data and detects the user's emotional state (e.g., stress, excitement, concentration, etc.), which is then sent to the server.

[0990] 5. Discussion Progress

[0991] The server uses a generative AI model to generate each character's utterances. For example, the "cautious" character might say, "A risk assessment is required."

[0992] The server adjusts the content and tone of the character's speech depending on the user's emotional state.

[0993] 6. Drawing conclusions

[0994] The server analyzes the discussion logs and generates an optimal conclusion, which is formatted in JSON and sent to the device.

[0995] The terminal visually displays the received conclusions to the user.

[0996] 3. Specific Examples

[0997] Example: If a customer enters "I'm looking for a light summer jacket"

[0998] Input prompt statement:

[0999] User Wants: A lightweight summer jacket

[1000] Bolder opinion: Experiment with different colors and designs!

[1001] Cautious observers: It's also a good idea to check the quality and durability of the materials.

[1002] Balancer's opinion: Why not take a look at some mid-priced items with good design and quality?

[1003] In this example, users can receive advice from multiple perspectives and receive real-time responses based on emotion analysis, resulting in a more satisfying purchasing experience. This system can improve the quality of customer service in physical stores.

[1004] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1005] Step 1:

[1006] Entering and receiving ideas

[1007] The user inputs ideas using a terminal. This input is often in text format and is specific, such as "I would like to come up with a market introduction strategy for a new health food."

[1008] The device receives this idea and converts it into JSON format, for example, {"idea": "I want to think about a market launch strategy for a new health food product"}.

[1009] The terminal sends the converted JSON data to the server, which receives it and stores it in a database.

[1010] Step 2:

[1011] Idea analysis and character generation

[1012] The server analyzes the received JSON-formatted idea data using natural language processing (NLP) technology. The purpose of the analysis is to extract keywords and topics for the ideas. The analysis results include keywords such as "healthy food," "market introduction," and "strategy."

[1013] Based on the analysis results, the server automatically generates characters to participate in the discussion. These characters include the moderator, the bold, the cautious, and the balancer. Each character's profile and role are retrieved from a database.

[1014] The server formats the generated character information in JSON format and sends it to the device. An example of formatted data is as follows: {"characters": [{"role": "Moderator", "profile": "Neutral moderator"}, {"role": "Bold", "profile": "Provides challenging opinions"}, {"role": "Cautious", "profile": "Provides risk-oriented opinions"}, {"role": "Balancer", "profile": "Reconcile the opinions of both parties"}]}

[1015] Step 3:

[1016] View character information and start discussions

[1017] The device parses the received character information and displays it for the user to visually confirm, with each character's role and profile displayed on the screen.

[1018] The user clicks the "Start Discussion" button. The device captures this click event and sends a request to start a discussion to the server.

[1019] Step 4:

[1020] Emotion analysis using an emotion engine

[1021] The device monitors user input and actions in real time, including keystroke speed, mouse movements, and facial recognition.

[1022] For example, if the user is typing quickly, the emotion engine detects an "excited" state.

[1023] The detected emotional state is sent to the server in JSON format, e.g. {"emotion": "excited"}

[1024] Step 5:

[1025] Discussion management and coordination

[1026] The server uses a generative AI model (e.g., GPT-3) to generate statements for each character. For example, the Cautious character might say, "That's a good idea, but we need to evaluate the cost-benefit and risks."

[1027] The server adjusts the content and tone of the characters' speech depending on the user's emotional state. For example, if the user is excited, the "Bold" character might suggest in a calmer tone, "You should launch a social media campaign and collaborate with influencers."

[1028] The server sends each character's comments in JSON format to the device, which then displays them to the user in chat format. The discussion is updated as users enter new ideas or questions.

[1029] Step 6:

[1030] Drawing and presenting conclusions

[1031] The server analyzes the discussion log and generates an optimal conclusion based on the discussion, such as "run a trial SNS campaign and gradually expand it based on the results."

[1032] This conclusion is formatted as JSON and sent to the device. Example: {"conclusion": "Test social media campaigns and gradually expand based on the results"}

[1033] The terminal visually displays the received conclusions to the user.

[1034] 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.

[1035] 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.

[1036] 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.

[1037] [Third embodiment]

[1038] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1039] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1040] 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).

[1041] 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.

[1042] 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.

[1043] 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).

[1044] 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.

[1045] 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.

[1046] 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.

[1047] 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.

[1048] 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.

[1049] 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."

[1050] The present invention is a system in which multiple generated characters develop a discussion based on an idea input by a user and reach a conclusion. The specific processing of the program of this system is explained in natural language below. The embodiment will also be explained in detail with specific examples.

[1051] Program processing

[1052] 1. User Idea Submissions

[1053] The user inputs the idea into the input form on the terminal.

[1054] The device receives the input ideas, converts them into a specific format (e.g., JSON format), and sends them to the server.

[1055] 2. Character Generation

[1056] The server analyzes the received ideas using natural language processing technology.

[1057] Based on the content of the idea, the server automatically generates the characters needed to smoothly advance the discussion (e.g., moderator, bold, cautious, balancer).

[1058] The profile and role of each character is determined and sent to the terminal.

[1059] 3. Starting a discussion

[1060] The terminal visually displays the generated characters and their roles to the user.

[1061] When a user clicks a button to start a discussion, the terminal sends a request to start a discussion to the server.

[1062] 4. Discussion Progress

[1063] The server generates utterances based on each character's role, including using AI models (e.g., GPT-3).

[1064] The server sequentially transmits the generated comments to the terminal, which then displays them to the user.

[1065] Users can join the discussion and add questions or new ideas, which are also sent to the server and reflected in the discussion.

[1066] 5. Drawing conclusions

[1067] The server analyzes the content of the discussion in real time and draws the optimal conclusion.

[1068] The server sends the generated conclusion to the terminal, which displays the final conclusion to the user.

[1069] Specific examples

[1070] Below are some specific examples on the theme of "market introduction strategies for new products."

[1071] 1. User Idea Submissions

[1072] The user enters "I would like to think of a market introduction strategy for a new health food" into the input form on the terminal.

[1073] The device converts this idea into JSON format and sends it to the server.

[1074] 2. Character Generation

[1075] The server analyzes the ideas received and extracts keywords such as "healthy food," "market introduction," and "strategy."

[1076] Based on this, the server will automatically generate characters for the moderator, bolder, cautious, and balancer.

[1077] The server determines the role and profile of each character and sends this to the device.

[1078] 3. Starting a discussion

[1079] The terminal displays information about the generated character to the user.

[1080] A discussion begins when the user clicks the "Start Discussion" button.

[1081] 4. Discussion Progress

[1082] The moderator says, "Now let's hear from the bold ones."

[1083] The bold suggestion is to launch a new social media campaign and collaborate with influencers.

[1084] Cautious voices say, "The idea is good, but we need to carefully evaluate the cost-benefit and risks."

[1085] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[1086] A user joins the discussion and types, "Should we do additional cost estimation?"

[1087] The device sends this new idea to the server and the discussion is updated.

[1088] 5. Drawing conclusions

[1089] The server analyzes the content of the discussion and generates a conclusion that "we will conduct a trial SNS campaign and gradually expand it based on the results."

[1090] The server sends this conclusion to the terminal, which displays it to the user.

[1091] This specific embodiment allows users to generate deeper ideas and make objective decisions through discussions from a variety of perspectives.

[1092] The processing flow will be explained below.

[1093] Step 1:

[1094] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[1095] Step 2:

[1096] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[1097] Step 3:

[1098] The idea data stored on the server is analyzed using natural language processing (NLP). As a result of the analysis, key topics and keywords of the ideas are extracted.

[1099] Step 4:

[1100] Based on the analysis results, the server automatically generates the characters necessary for the discussion (moderator, bold, cautious, balancer, etc.), and retrieves the character profiles and roles from a database.

[1101] Step 5:

[1102] The character information generated by the server is formatted in JSON format and sent to the terminal.

[1103] Step 6:

[1104] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[1105] Step 7:

[1106] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[1107] Step 8:

[1108] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[1109] Step 9:

[1110] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[1111] Step 10:

[1112] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[1113] Step 11:

[1114] The server updates the discussion based on the new input received, and new comments are generated by the characters based on the updated content.

[1115] Step 12:

[1116] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[1117] Step 13:

[1118] The server formats the generated conclusion in JSON format and sends it to the device, which visually displays the received conclusion to the user.

[1119] Step 14:

[1120] The user confirms the displayed conclusion, which allows the user to reach the optimal solution or decision that was reached through discussion of different perspectives.

[1121] Example 1

[1122] 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."

[1123] In the traditional idea generation and discussion process, users are responsible for all thinking alone, leading to problems of biased perspectives and reduced efficiency. Furthermore, incorporating multiple different opinions takes a lot of time and effort, making it difficult to quickly reach an optimal conclusion. Furthermore, there is a lack of support for users to smoothly facilitate discussions, which can lead to confusion.

[1124] 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.

[1125] In this invention, the server includes means for receiving information input by a user, means for converting the input information into a specific format and transmitting it to the server, means for analyzing the information received by the server using natural language processing technology, means for automatically generating multiple profiles based on the analysis, means for visually displaying the information in the generated profiles, means for the profiles to start a discussion, means for the server to analyze the content of the discussion and draw a conclusion, and means for presenting the conclusion to the user. This allows users to more effectively and quickly reach optimal conclusions through discussions from diverse perspectives. Furthermore, the discussion progresses smoothly, reducing bias in perspectives and reduced efficiency.

[1126] "User" refers to a person who uses this system to input and operate information.

[1127] "Information" refers to the ideas and data that users input into the system.

[1128] "Means" refers to methods or apparatuses for implementing the functions or processes described in the present invention.

[1129] A "profile" is information about the attributes and roles of a virtual character generated by the server for discussion purposes.

[1130] A "server" is a central computer system that receives and analyzes information from users, generates profiles, and manages discussions.

[1131] "Natural language processing technology" is a technology that allows computers to understand and analyze human language.

[1132] "Analysis" is the process of understanding, classifying, and evaluating input information.

[1133] "Discussion" is the process by which generated profiles exchange ideas and opinions and reach a conclusion.

[1134] A "conclusion" is a final judgment or recommendation that is reached as a result of a discussion.

[1135] "Visually displaying" means graphically presenting profile information and discussion content in a format that is easy for users to view.

[1136] The present invention is a system in which multiple profiles are generated based on information entered by a user, and discussions are held to reach a conclusion. A specific embodiment of the program for this system is described below. In particular, the hardware and software used, as well as specific data processing and calculations, are described in detail.

[1137] Hardware and Software Configuration

[1138] This system mainly uses the following hardware and software:

[1139] Cloud computing platforms as servers (e.g., Amazon Web Services, Google Cloud Platform)

[1140] User devices as terminals (e.g., PCs, tablets, smartphones)

[1141] NLP engines as natural language processing technologies (e.g., spaCy)

[1142] Language models as generative AI models (e.g., GPT-3)

[1143] Secure protocol used for sending and receiving data (e.g. HTTPS)

[1144] JSON format as data format

[1145] Program processing

[1146] 1. Enter your information:

[1147] The user enters information into the input form on the terminal.

[1148] The terminal receives the input information, converts it into JSON format, and sends it to the server.

[1149] 2. Analysis of Information and Profile Creation:

[1150] The information received by the server is analyzed using natural language processing technology (e.g., spaCy).

[1151] Based on the analysis results, the server automatically generates the profiles (e.g., moderator, bold, cautious, balancer) necessary to smoothly advance the discussion using a generative AI model (e.g., GPT-3).

[1152] The server sends the generated profile information to the terminal in JSON format.

[1153] 3. Visual representation of the profile:

[1154] The device visually displays the received profile information to the user using HTML, CSS, and Javascript.

[1155] 4. Starting a discussion:

[1156] When a user clicks the discussion start button, the terminal sends a request to start a discussion to the server.

[1157] The server generates messages for each profile and transmits the messages to the terminals one by one.

[1158] The terminal displays the statement to the user.

[1159] 5. Developing a discussion and drawing conclusions:

[1160] Users can join the discussion and enter new questions or ideas, which are also sent to the server and reflected in the discussion.

[1161] The server analyzes the discussion content in real time and generates the optimal conclusion.

[1162] The server sends the generated conclusion to the terminal, which displays the conclusion to the user.

[1163] Specific examples

[1164] Below are some specific examples on the theme of "market introduction strategies for new products."

[1165] 1. User Idea Submissions:

[1166] The user enters "I would like to think of a market introduction strategy for a new health food" into the input form on the terminal.

[1167] The device converts this information into JSON format and sends it to the server.

[1168] 2. Generate and view the profile:

[1169] The server analyzes the information it receives and extracts keywords such as "health food," "market introduction," and "strategy."

[1170] Based on this, the server generates profiles for Moderator, Bold, Cautious, and Balancer.

[1171] The server transmits the generated profile information to the terminal.

[1172] The terminal displays the profile information to the user.

[1173] 3. Initiating and Conducting a Discussion:

[1174] The terminal displays a "Start Discussion" button, and when the user clicks on it, the discussion begins.

[1175] The moderator says, "Now let's hear from the bold ones."

[1176] The bold suggestion is to launch a new social media campaign and collaborate with influencers.

[1177] Cautious voices say, "The idea is good, but we need to carefully evaluate the cost-benefit and risks."

[1178] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[1179] The user joins the discussion and types, "Should we do additional cost estimation?" The device sends this information to the server.

[1180] 4. Draw conclusions:

[1181] The server analyzes the content of the discussion and generates a conclusion that "we will conduct a trial SNS campaign and gradually expand it based on the results."

[1182] The server sends this conclusion to the terminal, which displays it to the user.

[1183] Examples of prompt statements

[1184] Here are some examples of prompts to input to a generative AI model:

[1185] User: I would like to think about a market introduction strategy for a new health food.

[1186] Server: The moderator (you) is responsible for keeping the discussion running smoothly.

[1187] Moderator: "Now let's hear from the bold ones."

[1188] Bold: "We should launch a new social media campaign and collaborate with influencers."

[1189] Cautious: "The idea is good, but the costs and risks need to be carefully assessed."

[1190] Balancer: "Both opinions are important. Let's compare the specific costs and expected benefits."

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

[1192] Step 1: Enter your information

[1193] The user inputs information (e.g., "I would like to consider a market introduction strategy for a new health food product") into an input form on the terminal.

[1194] Input: Information entered by the user into the device.

[1195] The device processes this input information using UTF-8 character encoding and converts it into a JSON-formatted object.

[1196] Data processing: Character encoding of information and conversion to JSON format.

[1197] The device sends the generated JSON object to the server using the secure HTTPS protocol.

[1198] Output: Reformatted JSON information.

[1199] Step 2: Analyze the information and create a profile

[1200] The server parses the received JSON data and uses natural language processing techniques (e.g., spaCy) to extract important keywords and phrases.

[1201] Input: JSON formatted information sent from the device.

[1202] Data processing: Keyword extraction and phrase analysis using a natural language processing engine.

[1203] Based on the extracted keywords, the server automatically generates profiles (e.g., moderator, bold, cautious, balancer) necessary to smoothly advance the discussion using a generative AI model (e.g., GPT-3).

[1204] Data calculation: Profile generation using AI models.

[1205] The server determines the attributes of each profile (e.g., name, personality, expertise, etc.) and packages them into a JSON format.

[1206] Output: JSON format data of generated profile information.

[1207] The server sends the generated profile information to the device using the secure HTTPS protocol.

[1208] Step 3: Visual representation of the profile

[1209] Renders a user interface (UI) that displays the profile information received by the device.

[1210] Input: JSON data of profile information received from the server.

[1211] The device uses HTML, CSS, and JavaScript to visually display the profile's name, role, profile picture, etc.

[1212] Behavior: Rendering and displaying the UI.

[1213] Output: Profile information visually displayed to the user.

[1214] Step 4: Start a discussion

[1215] When a user clicks the Start Discussion button, a JavaScript event handler on the device detects this.

[1216] Input: The action of the user clicking the Start Discussion button.

[1217] The terminal composes a request to start a discussion in JSON format and sends it to the server.

[1218] Data processing: Composing a discussion start request.

[1219] Output: Discussion start request sent to the server.

[1220] Step 5: Facilitate the discussion

[1221] The server receives a request to start a discussion and starts the process of generating comments for each profile.

[1222] Input: A request to start a discussion from a terminal.

[1223] The server uses a generative AI model (e.g., GPT-3) to generate utterances based on the role of each profile.

[1224] Data computation: utterance generation using AI models.

[1225] The server sequentially sends the generated comments to the terminal in JSON format.

[1226] Output: JSON data of the generated utterance.

[1227] The device displays the comments received in real time, allowing users to visually check the progress of the discussion.

[1228] What it does: Real-time display of what's being said.

[1229] Users can join the discussion and enter new questions or ideas, which are also sent back to the server and reflected in the discussion.

[1230] Input: New questions and ideas.

[1231] Data arithmetic: Processing new input information and reflecting it in discussions.

[1232] Step 6: Draw conclusions

[1233] The server analyzes all discussion content in real time and applies algorithms to derive optimal conclusions based on the data.

[1234] Input: All data from the discussion.

[1235] Data arithmetic: Applying conclusion-drawing algorithms.

[1236] The server generates the final result and sends it to the device in JSON format.

[1237] Output: JSON data of the generated conclusions.

[1238] The terminal displays the received conclusion on a user interface to provide the user with a final conclusion.

[1239] Action: Display of final conclusion.

[1240] (Application example 1)

[1241] 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."

[1242] In the development and operation of autonomous vehicles, there is a need to efficiently hold discussions from a variety of expert perspectives and derive objective, optimal conclusions. Conventional methods often make it difficult to reconcile opinions between experts and advance discussions, requiring time and effort. Furthermore, there is a problem in that it is difficult for users to actively participate in discussions, resulting in only a portion of opinions being reflected. In response to these issues, a system that supports efficient and fair discussions and allows users to actively participate is desired.

[1243] 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.

[1244] In this invention, the server includes means for receiving a task input by a user, means for analyzing the task content and automatically generating a plurality of discussion characters, means for the generated discussion characters to engage in a discussion, means for analyzing the discussion content and deriving a conclusion, means for presenting the conclusion to the user, and means for the user to input additional questions or suggestions while the discussion is in progress. This enables efficient and fair discussions from various perspectives and provides an environment in which users can actively participate in the discussion.

[1245] "User" refers to a person who inputs data into or operates the system.

[1246] "Issues" refer to problems that users want to solve or themes that they want to consider.

[1247] "Means of receiving" refers to the interface or function that allows the system to acquire tasks from users.

[1248] "Analyzing" refers to the process of understanding the content of the input task and extracting the necessary information.

[1249] "Discussion characters" refer to virtual roles automatically generated for the purpose of conducting discussions.

[1250] "Means of automatic generation" refers to the process by which the system uses AI technology to automatically create discussion characters.

[1251] "Means for holding discussions" refers to a function that allows the generated discussion characters to converse and exchange opinions with each other.

[1252] "Means of deriving a conclusion" refers to the process for generating the most appropriate conclusion based on the discussion that has taken place.

[1253] "Presentation means" refers to the interface or functionality for informing the user of the generated conclusions.

[1254] "Means for inputting additional questions and suggestions" refers to the function that allows users to input new questions or suggestions to the system during the discussion.

[1255] This invention is a system for efficiently conducting discussions from various expert viewpoints and deriving objective and optimal conclusions in the development and operation of autonomous vehicles. This system uses automatically generated discussion characters to hold discussions based on issues entered by the user, and analyzes the results to reach a conclusion.

[1256] System configuration

[1257] 1. User Device

[1258] Hardware: User interfaces such as smartphones and head-mounted displays

[1259] Software: input forms, display interfaces, communication modules

[1260] 2. Server

[1261] Hardware: Servers with high-performance computing capabilities (including cloud servers)

[1262] Software: Natural language processing technology (e.g., GPT-3), data analysis module, character generation module, discussion progression module, conclusion drawing module

[1263] System Operation Overview

[1264] 1. User Device

[1265] The user inputs a problem, such as "I want to think of a market launch strategy for a new autonomous driving algorithm," into an input form.

[1266] This input task is converted into JSON format and sent to the server.

[1267] 2. Server

[1268] The server analyzes the received problem and extracts keywords, such as "autonomous driving," "algorithm," and "market launch."

[1269] Based on the extracted keywords, the necessary discussion characters are automatically generated, including "technical experts," "safety managers," "management managers," and "user representatives."

[1270] The role and profile of each character are determined and transmitted to the user terminal.

[1271] 3. Discussion Progress

[1272] The user terminal displays the generated character information, and the user starts a discussion.

[1273] The server generates utterances according to the character's role and sends them to the user's terminal in sequence. During the discussion, the user can input additional questions or suggestions.

[1274] 4. Drawing conclusions

[1275] The server analyzes the content of the discussion and derives the optimal conclusion, which is then sent to the user's terminal and presented to the user.

[1276] Specific examples

[1277] Below is a concrete example of a discussion regarding a market launch strategy for a new autonomous driving algorithm.

[1278] Example prompt sentence:

[1279] As a technology expert, please discuss "Go-to-market strategies for new autonomous driving algorithms."

[1280] As a safety manager, please discuss the "market launch strategy for new autonomous driving algorithms."

[1281] As a manager, please discuss the "market launch strategy for new autonomous driving algorithms."

[1282] As a user representative, please discuss the "market launch strategy for new autonomous driving algorithms."

[1283] This system utilizes generative AI models (e.g., GPT-3) to enable efficient and fair discussions from diverse perspectives, allowing users to actively participate in the discussion and derive deeper ideas and objective conclusions.

[1284] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1285] Step 1:

[1286] The user inputs a task into the device's input form. This task may include a specific problem such as "I want to think about a market launch strategy for a new autonomous driving algorithm." The device converts this input into JSON format and sends it to the server.

[1287] Input: The assignment entered by the user

[1288] Output: Issue data in JSON format

[1289] Step 2:

[1290] The server analyzes the received JSON-formatted problem data and extracts keywords, such as "autonomous driving," "algorithm," and "market launch." This analysis uses natural language processing technology (e.g., GPT-3).

[1291] Input: Issue data in JSON format

[1292] Output: Extracted keyword list

[1293] Step 3:

[1294] The server generates the necessary discussion characters based on the extracted keywords. The generated characters include "technical expert," "safety manager," "management manager," and "user representative." It generates a prompt for each character and determines the character's profile and role using an AI model (e.g., GPT-3).

[1295] Input: Keyword list

[1296] Output: Character profile and role

[1297] Step 4:

[1298] The server sends the generated character's profile and role to the user's device, which receives it and displays it visually to the user. The user can start a discussion by clicking the "Start Discussion" button on their device.

[1299] Input: Character profile and role

[1300] Output: Display character information to the user

[1301] Step 5:

[1302] The server generates utterances based on the character's role. This utterance generation uses an AI model (e.g., GPT-3). The generated utterances are sequentially sent to the user's device and displayed to the user.

[1303] Input: Character profile and role

[1304] Output: What the character says

[1305] Step 6:

[1306] Users can enter additional questions or suggestions into the discussion on their terminals. These additional inputs are also converted into JSON format and sent to the server, which receives them and updates the discussion.

[1307] Input: User's additional questions or suggestions

[1308] Output: Updated discussion

[1309] Step 7:

[1310] The server analyzes the entire discussion and derives the optimal conclusion. This analysis and conclusion generation uses an AI model (e.g., GPT-3). The server then sends the final conclusion to the user's device.

[1311] Input: Updated discussion

[1312] Output: Conclusion

[1313] Step 8:

[1314] The user terminal displays the received conclusion to the user, who can then confirm the conclusion and use it to decide what action to take next.

[1315] Input: Conclusion

[1316] Output: Display conclusion to user

[1317] 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.

[1318] The present invention is a system in which multiple generated characters develop a discussion based on an idea input by a user, and a conclusion is reached by combining an emotion engine while recognizing the user's emotional state. The specific processing of the program of this system is explained in natural language below. Also, specific examples are provided to explain the embodiment in detail.

[1319] Program processing

[1320] 1. User Idea Submissions

[1321] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[1322] 2. Character Generation

[1323] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[1324] The server analyzes the stored idea data using natural language processing (NLP), and extracts key topics and keywords from the analysis results.

[1325] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer), and retrieves the character profiles and roles from the database.

[1326] The character information generated by the server is formatted in JSON format and sent to the terminal.

[1327] 3. Starting a discussion

[1328] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[1329] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[1330] 4. Operation of the Emotion Engine

[1331] The device sends the user's input and actions (e.g., keystroke speed, mouse movements, facial recognition, etc.) to the emotion engine in real time.

[1332] The emotion engine analyzes this data to detect the user's emotional state (e.g., stress, excitement, concentration, etc.).

[1333] The device transmits the detected emotional state to the server.

[1334] 5. Discussion Progress

[1335] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[1336] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[1337] The server receives data from the emotion engine and adjusts the character's speech and tone based on the user's emotional state.

[1338] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[1339] 6. Drawing conclusions

[1340] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[1341] The server formats the generated conclusion in JSON format and sends it to the device, which visually displays the received conclusion to the user.

[1342] Specific examples

[1343] Below are some specific examples on the theme of "market introduction strategies for new products."

[1344] 1. User Idea Submissions

[1345] The user enters the idea into the input form on the device, "I want to think of a market introduction strategy for a new health food." The device converts this idea into JSON format and sends it to the server.

[1346] 2. Character Generation

[1347] The server analyzes the ideas received and extracts keywords such as "healthy food," "market introduction," and "strategy."

[1348] Based on this, the server automatically generates characters for the moderator, bold, cautious, and balancer. The server determines the role and profile of each character and sends this to the device.

[1349] 3. Starting a discussion

[1350] The terminal displays information about the generated character to the user. The user can start a discussion by clicking the "Start Discussion" button.

[1351] 4. Operation of the Emotion Engine

[1352] The device sends the user's input and behavioral data (e.g., keystroke speed and facial expression) to the emotion engine, which then detects the user's emotional state. The detected emotional state is then sent from the device to the server.

[1353] 5. Discussion Progress

[1354] The moderator says, "Now let's hear from the bold ones."

[1355] The bold suggest, "You should launch a social media campaign and collaborate with influencers."

[1356] Cautious voices say, "The idea is good, but we need to evaluate the cost-benefit and risks."

[1357] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[1358] The server adjusts the bolder's tone of speech, for example softening it, depending on the user's emotional state.

[1359] A user joins the discussion and types, "Should we do additional cost estimation?" The device sends this idea to the server and the discussion is updated.

[1360] 6. Drawing conclusions

[1361] The server analyzes the content of the discussion and generates a conclusion: "We will conduct a trial SNS campaign and gradually expand it based on the results." The conclusion is sent to the device and displayed to the user.

[1362] This specific embodiment allows users to discuss with characters with diverse perspectives, encouraging flexible responses that reflect their emotional state, enabling them to generate deeper ideas and make more objective decisions.

[1363] The processing flow will be explained below.

[1364] Step 1:

[1365] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[1366] Step 2:

[1367] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[1368] Step 3:

[1369] The idea data stored on the server is analyzed using natural language processing (NLP), which extracts key topics and keywords for the ideas.

[1370] Step 4:

[1371] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer), and retrieves each character's profile and role from the database.

[1372] Step 5:

[1373] The character information generated by the server is formatted in JSON format and sent to the terminal.

[1374] Step 6:

[1375] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[1376] Step 7:

[1377] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[1378] Step 8:

[1379] The device sends the user's input and actions (e.g., keystroke speed, mouse movement, facial recognition, etc.) to the emotion engine in real time. The emotion engine analyzes this data and detects the user's emotional state.

[1380] Step 9:

[1381] The device sends the emotional state detected by the emotion engine to the server, which then adjusts the content and tone of the character's speech based on the emotional state received.

[1382] Step 10:

[1383] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[1384] Step 11:

[1385] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[1386] Step 12:

[1387] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[1388] Step 13:

[1389] The server updates the discussion based on the new input received, and new comments are generated by the characters based on the updated content.

[1390] Step 14:

[1391] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[1392] Step 15:

[1393] The server formats the generated conclusion in JSON format and sends it to the device.

[1394] Step 16:

[1395] The terminal visually displays the received conclusion to the user, who then confirms the displayed conclusion. This allows the user to obtain the optimal solution or decision-making result obtained through discussion from different perspectives.

[1396] Example 2

[1397] 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."

[1398] In conventional idea generation support systems, it was difficult to discuss ideas entered by users from various perspectives and to respond flexibly while taking into account their emotional state. Furthermore, the generated characters' comments were not adjusted according to the user's emotions, making it difficult to advance effective discussions. The present invention aims to solve these problems and provide a system that effectively supports the user's idea generation process from multiple angles.

[1399] 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.

[1400] In this invention, the server includes means for receiving ideas input by a user, means for converting the input ideas into JSON format, and means for transmitting the converted JSON-formatted idea data to the server. This allows users to easily input ideas and process them efficiently. The server also includes means for analyzing the saved idea data using natural language processing to extract key topics and keywords, means for automatically generating multiple characters based on the analysis results, and means for formatting the generated character information in JSON format and transmitting it to a terminal. This allows users to generate characters with diverse perspectives and engage in appropriate discussions. The server also includes means for an emotion engine to detect the user's emotional state, means for adjusting the content and tone of the character's comments based on the user's emotional state, and means for utilizing a generative AI model to generate comments appropriate to each character's role. This enables flexible responses that reflect the user's emotional state, improving the quality of discussions.

[1401] A "user" is a person who operates the system and inputs ideas, and is the primary user of the system.

[1402] A "terminal" is an input device operated by a user, and is a device for inputting ideas and displaying information.

[1403] "Server" refers to a central management system that stores, analyzes, and processes data sent by users.

[1404] "JSON format" is an abbreviation for JavaScript Object Notation, and is a lightweight data exchange format that makes it easy to exchange data.

[1405] "Conversion" refers to the process of changing input data into another format.

[1406] A "POST request" is a method of sending data to a server using the HyperText Transfer Protocol (HTTP).

[1407] "Natural Language Processing (NLP)" is a technology that allows computers to understand, analyze, and generate human language.

[1408] A "character" is a virtual person automatically generated for the purpose of discussion, with a specific role and profile.

[1409] The "Emotion Engine" is a machine learning model for analyzing and detecting a user's emotional state.

[1410] A "generative AI model" refers to an artificial intelligence model used to perform tasks such as natural language generation and dialogue generation.

[1411] A "prompt" is an instruction given to a generative AI model that the model uses to generate an appropriate response or generation.

[1412] "Discussion" refers to the process in which multiple characters exchange opinions on a user's idea.

[1413] A "conclusion" is the final result or proposal reached through a discussion.

[1414] This invention is a system that supports users in discussing ideas from various perspectives and responding flexibly. This system is realized using users, terminals, a server, a generative AI model, and an emotion engine.

[1415] First, the user inputs an idea into the input form on the device. For example, this input might be in the form of "I want to think of a market introduction strategy for a new health food." This idea is converted into JSON format by the device. The device then sends this JSON-formatted idea data to the server via a POST request.

[1416] Next, the server stores the JSON-formatted idea data received from the device. A NoSQL database (e.g., MongoDB) is used for this storage. The server then analyzes the stored idea data using natural language processing (NLP). An NLP library (e.g., spaCy or NLTK) is used for this analysis, and key topics and keywords for the ideas (e.g., "healthy food," "market introduction," and "strategy") are extracted. Based on this, characters are automatically generated to participate in the discussion. Character profile and role information is retrieved from a database (e.g., PostgreSQL).

[1417] The generated character information is formatted in JSON format by the server and sent to the device. The device parses the received character information and visually displays it to the user. For example, the character's name, role, and profile information are displayed. At this point, the user clicks the "Start Discussion" button to begin the discussion.

[1418] When a discussion begins, the device collects user input and behavioral data (e.g., keystroke speed, mouse movement, and facial recognition data) in real time and sends them to the emotion engine. The emotion engine then analyzes this data using a machine learning model (e.g., a TensorFlow model) to detect the user's emotional state. The device then transmits the detected emotional state to the server.

[1419] The server uses a generative AI model (e.g., GPT-3) to generate utterances appropriate for each character's role. A prompt is used to generate the utterances. An example of a prompt is: "As a bold person, please give your opinion on the market launch strategy for health foods."

[1420] The comments generated by the server are sent to the device in sequence, and the comments received by the device are displayed to the user in chat format. This uses a conversational UI, making it appear to the user that the characters are engaged in a dialogue. The server receives data from the emotion engine and adjusts the content and tone of the characters' comments based on the user's emotional state. For example, if the user is relaxed, it softens the tone of the bolder comments. Users participate in the discussion, inputting unclear points and new ideas. Input such as "Should we do additional cost estimates?" is sent from the device to the server, and the discussion is updated.

[1421] Once the discussion has ended, the server analyzes the discussion log and stores each character's comments in a database. The server generates an optimal conclusion based on the content of the discussion, formats the conclusion into JSON format, and sends it to the device. The device then visually displays this conclusion to the user. A specific conclusion might be something like, "We will conduct a trial SNS campaign and gradually expand it based on the results."

[1422] In this way, users can interact with characters with diverse perspectives, responding flexibly to their emotional state, leading to deeper idea generation and objective decision-making.

[1423] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1424] Step 1:

[1425] The user inputs their idea into the input form on the device. For example, they might input, "I want to think of a market introduction strategy for a new health food." The device receives the user's input and converts it into JSON format. Input data: User's idea (text format) → Output data: Idea (JSON format).

[1426] Step 2:

[1427] The terminal sends the converted JSON-formatted idea data to the server via a POST request. Input data: JSON-formatted idea data → Output data: Send request to the server.

[1428] Step 3:

[1429] The server stores the JSON format idea data received from the device. A NoSQL database (e.g., MongoDB) is used. Input data: JSON format idea data → Output data: idea data stored in the database.

[1430] Step 4:

[1431] The server analyzes the stored idea data using natural language processing (NLP). It uses an NLP library (e.g., spaCy or NLTK) to extract key topics and keywords for ideas. Input data: idea data (JSON format) → Processing: NLP analysis → Output data: keywords and topics.

[1432] Step 5:

[1433] The server automatically generates characters for discussions based on the analysis results. Character profiles and role information are obtained from a database (e.g., PostgreSQL). Input data: Keywords and topics → Processing content: Automatic character generation, profile acquisition → Output data: Character information (JSON format).

[1434] Step 6:

[1435] The server formats the generated character information in JSON format and sends it to the terminal. Input data: Character information → Processing content: Format into JSON format → Output data: Send request to the terminal.

[1436] Step 7:

[1437] The device analyzes the character information received and displays it visually to the user. Using HTML and CSS, a UI is generated to display the character's name, role, and profile information. Input data: character information (JSON format) → Processing: Parse, UI generation → Output data: user interface.

[1438] Step 8:

[1439] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start a discussion to the server. Input data: User's click operation → Output data: Request to start a discussion.

[1440] Step 9:

[1441] The device collects user input and behavioral data (e.g., keystroke speed, mouse movement, facial recognition data) in real time and sends it to the emotion engine. Input data: behavioral data → Output data: transmission request to the emotion engine.

[1442] Step 10:

[1443] The emotion engine analyzes the data using a machine learning model (e.g., TensorFlow model) and detects the user's emotional state. Input data: behavioral data → Processing content: detection of emotional state → Output data: user's emotional state.

[1444] Step 11:

[1445] The server receives data from the emotion engine and adjusts the content and tone of the character's speech based on the user's emotional state. Input data: User's emotional state → Processing content: Adjustment of speech content and tone → Output data: Adjusted speech.

[1446] Step 12:

[1447] The server uses a generative AI model (e.g., GPT-3) to generate utterances appropriate for each character's role. A sample prompt might be, "As a bold person, please share your opinion on the market launch strategy for health foods."

[1448] Input data: Character role, prompt sentence → Processing content: Speech generation by AI model → Output data: Generated speech.

[1449] Step 13:

[1450] The server sends generated comments to the device in sequence. The device displays the received comments to the user in chat format. Displayed in an interactive UI. Input data: Generated comments → Processing content: Displayed in chat → Output data: Displayed to the user.

[1451] Step 14:

[1452] The user joins the discussion and inputs any unclear points or new ideas. For example, "Should we do additional cost estimates?" The device sends this content to the server, and the discussion is updated. Input data: User's new idea → Output data: Send request to the server.

[1453] Step 15:

[1454] The server analyzes the discussion log and stores each character's comments in a database. The optimal conclusion is generated based on the content of the discussion. Input data: discussion log → Processing content: NLP analysis, conclusion generation → Output data: generated conclusion.

[1455] Step 16:

[1456] The server formats the generated conclusion into JSON format and sends it to the device. The device visually displays the conclusion to the user. A specific conclusion might be something like "We will conduct a trial SNS campaign and gradually expand it based on the results." Input data: Generated conclusion → Processing content: Format into JSON format, display to user → Output data: Conclusion presented to the user.

[1457] (Application example 2)

[1458] 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."

[1459] In conventional systems, multiple characters could discuss and reach a conclusion based on ideas input by the user, but they were unable to reflect the user's emotional state in real time. This made it difficult to respond appropriately to the user's emotions or to flexibly advance the discussion, ultimately causing a decline in user satisfaction and the quality of the discussion. In customer interactions and customer service in physical stores, there was no system that could take customer emotions into account, so improvements in customer satisfaction could not be expected.

[1460] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1461] In this invention, the server includes means for receiving ideas input by users, means for analyzing the input ideas and automatically generating multiple characters, means for the generated characters to hold discussions, means for analyzing the content of the discussion and deriving a conclusion, means for presenting the conclusion to users, means for emotionally analyzing customer input and behavioral data in real time, and means for adjusting the characters' comments based on the emotional state. This enables flexible discussions that reflect the emotional state of users in real time, which is expected to improve customer satisfaction in physical stores.

[1462] "User" refers to an individual or group that uses the system to input ideas and participate in discussions.

[1463] "Ideas" refer to the ideas and thoughts that users input into the system, and are the information that forms the basis of discussion.

[1464] "Characters" refer to virtual characters who are automatically generated within the system based on the user's ideas and who play various roles in discussions.

[1465] A "discussion" refers to a conversation or debate process in which multiple characters express their opinions based on their respective roles and reach a conclusion.

[1466] "Conclusion" refers to the result or proposal that is ultimately reached after analyzing the content of the discussion.

[1467] "Emotion analysis" refers to the process of analyzing user input and behavioral data to detect the user's emotional state in real time.

[1468] "Adjusting speech" refers to changing the content and tone of a character's speech based on emotion analysis to generate appropriate speech that corresponds to the user's emotional state.

[1469] A "physical store" is a facility located in a physical location where customers can visit in person to provide goods or services.

[1470] "Customer" refers to a visitor who visits a physical store and intends to purchase or use a product or service.

[1471] "Customer service" refers to activities that involve interacting with and providing support to customers in physical stores, with the aim of improving customer satisfaction.

[1472] "Emotion engine" refers to an analysis device or program that detects the user's emotional state and reflects it in processes within the system.

[1473] The present invention is a system in which multiple characters discuss ideas input by a user and use an emotion engine to reach a conclusion while reflecting the user's emotional state in real time. This system aims to improve customer service in brick-and-mortar stores. Specific embodiments for realizing this system are described below.

[1474] 1. System Overview

[1475] Hardware used

[1476] Device: Smartphone, tablet or PC. Device used by staff or customers in a physical store.

[1477] Server: A remotely located computing device that analyzes data and runs generative AI models.

[1478] Software used

[1479] Natural Language Processing (NLP): A technique used to analyze user-entered ideas.

[1480] Generative AI model (e.g. GPT-3): A model used to generate character utterances.

[1481] Emotion engine: Software for analyzing a user's emotional state in real time.

[1482] 2. Program processing explanation

[1483] 1. Idea input and analysis

[1484] The user inputs an idea on the terminal, which is converted into JSON format and sent to the server.

[1485] The server analyzes the received ideas using natural language processing (NLP) to extract key topics and keywords.

[1486] 2. Character Generation

[1487] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer).

[1488] The role and profile of each character is retrieved from the database, and the generated character information is sent to the terminal.

[1489] 3. Starting a discussion

[1490] The terminal displays the received character information to the user, and the discussion begins when the user clicks the "Start discussion" button.

[1491] This discussion is initiated by sending a request to the server to start the discussion.

[1492] 4. Operation of the Emotion Engine

[1493] The device transmits user input and actions (e.g., keystroke speed, mouse movement, facial recognition) to the emotion engine in real time.

[1494] The emotion engine analyzes these data and detects the user's emotional state (e.g., stress, excitement, concentration, etc.), which is then sent to the server.

[1495] 5. Discussion Progress

[1496] The server uses a generative AI model to generate each character's utterances. For example, the "cautious" character might say, "A risk assessment is required."

[1497] The server adjusts the content and tone of the character's speech depending on the user's emotional state.

[1498] 6. Drawing conclusions

[1499] The server analyzes the discussion logs and generates an optimal conclusion, which is formatted in JSON and sent to the device.

[1500] The terminal visually displays the received conclusions to the user.

[1501] 3. Specific Examples

[1502] Example: If a customer enters "I'm looking for a light summer jacket"

[1503] Input prompt statement:

[1504] User Wants: A lightweight summer jacket

[1505] Bolder opinion: Experiment with different colors and designs!

[1506] Cautious observers: It's also a good idea to check the quality and durability of the materials.

[1507] Balancer's opinion: Why not take a look at some mid-priced items with good design and quality?

[1508] In this example, users can receive advice from multiple perspectives and receive real-time responses based on emotion analysis, resulting in a more satisfying purchasing experience. This system can improve the quality of customer service in physical stores.

[1509] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1510] Step 1:

[1511] Entering and receiving ideas

[1512] The user inputs ideas using a terminal. This input is often in text format and is specific, such as "I would like to come up with a market introduction strategy for a new health food."

[1513] The device receives this idea and converts it into JSON format, for example, {"idea": "I want to think about a market launch strategy for a new health food product"}.

[1514] The terminal sends the converted JSON data to the server, which receives it and stores it in a database.

[1515] Step 2:

[1516] Idea analysis and character generation

[1517] The server analyzes the received JSON-formatted idea data using natural language processing (NLP) technology. The purpose of the analysis is to extract keywords and topics for the ideas. The analysis results include keywords such as "healthy food," "market introduction," and "strategy."

[1518] Based on the analysis results, the server automatically generates characters to participate in the discussion. These characters include the moderator, the bold, the cautious, and the balancer. Each character's profile and role are retrieved from a database.

[1519] The server formats the generated character information in JSON format and sends it to the device. An example of formatted data is as follows: {"characters": [{"role": "Moderator", "profile": "Neutral moderator"}, {"role": "Bold", "profile": "Provides challenging opinions"}, {"role": "Cautious", "profile": "Provides risk-oriented opinions"}, {"role": "Balancer", "profile": "Reconcile the opinions of both parties"}]}

[1520] Step 3:

[1521] View character information and start discussions

[1522] The device parses the received character information and displays it for the user to visually confirm, with each character's role and profile displayed on the screen.

[1523] The user clicks the "Start Discussion" button. The device captures this click event and sends a request to start a discussion to the server.

[1524] Step 4:

[1525] Emotion analysis using an emotion engine

[1526] The device monitors user input and actions in real time, including keystroke speed, mouse movements, and facial recognition.

[1527] For example, if the user is typing quickly, the emotion engine detects an "excited" state.

[1528] The detected emotional state is sent to the server in JSON format, e.g. {"emotion": "excited"}

[1529] Step 5:

[1530] Discussion management and coordination

[1531] The server uses a generative AI model (e.g., GPT-3) to generate statements for each character. For example, the Cautious character might say, "That's a good idea, but we need to evaluate the cost-benefit and risks."

[1532] The server adjusts the content and tone of the characters' speech depending on the user's emotional state. For example, if the user is excited, the "Bold" character might suggest in a calmer tone, "You should launch a social media campaign and collaborate with influencers."

[1533] The server sends each character's comments in JSON format to the device, which then displays them to the user in chat format. The discussion is updated as users enter new ideas or questions.

[1534] Step 6:

[1535] Drawing and presenting conclusions

[1536] The server analyzes the discussion log and generates an optimal conclusion based on the discussion, such as "run a trial SNS campaign and gradually expand it based on the results."

[1537] This conclusion is formatted as JSON and sent to the device. Example: {"conclusion": "Test social media campaigns and gradually expand based on the results"}

[1538] The terminal visually displays the received conclusions to the user.

[1539] 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.

[1540] 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.

[1541] 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.

[1542] [Fourth embodiment]

[1543] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1544] 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.

[1545] 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).

[1546] 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.

[1547] 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.

[1548] 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).

[1549] 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.

[1550] 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.

[1551] 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.

[1552] 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.

[1553] 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.

[1554] 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.

[1555] 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."

[1556] The present invention is a system in which multiple generated characters develop a discussion based on an idea input by a user and reach a conclusion. The specific processing of the program of this system is explained in natural language below. The embodiment will also be explained in detail with specific examples.

[1557] Program processing

[1558] 1. User Idea Submissions

[1559] The user inputs the idea into the input form on the terminal.

[1560] The device receives the input ideas, converts them into a specific format (e.g., JSON format), and sends them to the server.

[1561] 2. Character Generation

[1562] The server analyzes the received ideas using natural language processing technology.

[1563] Based on the content of the idea, the server automatically generates the characters needed to smoothly advance the discussion (e.g., moderator, bold, cautious, balancer).

[1564] The profile and role of each character is determined and sent to the terminal.

[1565] 3. Starting a discussion

[1566] The terminal visually displays the generated characters and their roles to the user.

[1567] When a user clicks a button to start a discussion, the terminal sends a request to start a discussion to the server.

[1568] 4. Discussion Progress

[1569] The server generates utterances based on each character's role, including using AI models (e.g., GPT-3).

[1570] The server sequentially transmits the generated comments to the terminal, which then displays them to the user.

[1571] Users can join the discussion and add questions or new ideas, which are also sent to the server and reflected in the discussion.

[1572] 5. Drawing conclusions

[1573] The server analyzes the content of the discussion in real time and draws the optimal conclusion.

[1574] The server sends the generated conclusion to the terminal, which displays the final conclusion to the user.

[1575] Specific examples

[1576] Below are some specific examples on the theme of "market introduction strategies for new products."

[1577] 1. User Idea Submissions

[1578] The user enters "I would like to think of a market introduction strategy for a new health food" into the input form on the terminal.

[1579] The device converts this idea into JSON format and sends it to the server.

[1580] 2. Character Generation

[1581] The server analyzes the ideas received and extracts keywords such as "healthy food," "market introduction," and "strategy."

[1582] Based on this, the server will automatically generate characters for the moderator, bolder, cautious, and balancer.

[1583] The server determines the role and profile of each character and sends this to the device.

[1584] 3. Starting a discussion

[1585] The terminal displays information about the generated character to the user.

[1586] A discussion begins when the user clicks the "Start Discussion" button.

[1587] 4. Discussion Progress

[1588] The moderator says, "Now let's hear from the bold ones."

[1589] The bold suggestion is to launch a new social media campaign and collaborate with influencers.

[1590] Cautious voices say, "The idea is good, but we need to carefully evaluate the cost-benefit and risks."

[1591] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[1592] A user joins the discussion and types, "Should we do additional cost estimation?"

[1593] The device sends this new idea to the server and the discussion is updated.

[1594] 5. Drawing conclusions

[1595] The server analyzes the content of the discussion and generates a conclusion that "we will conduct a trial SNS campaign and gradually expand it based on the results."

[1596] The server sends this conclusion to the terminal, which displays it to the user.

[1597] This specific embodiment allows users to generate deeper ideas and make objective decisions through discussions from a variety of perspectives.

[1598] The processing flow will be explained below.

[1599] Step 1:

[1600] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[1601] Step 2:

[1602] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[1603] Step 3:

[1604] The idea data stored on the server is analyzed using natural language processing (NLP). As a result of the analysis, key topics and keywords of the ideas are extracted.

[1605] Step 4:

[1606] Based on the analysis results, the server automatically generates the characters necessary for the discussion (moderator, bold, cautious, balancer, etc.), and retrieves the character profiles and roles from a database.

[1607] Step 5:

[1608] The character information generated by the server is formatted in JSON format and sent to the terminal.

[1609] Step 6:

[1610] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[1611] Step 7:

[1612] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[1613] Step 8:

[1614] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[1615] Step 9:

[1616] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[1617] Step 10:

[1618] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[1619] Step 11:

[1620] The server updates the discussion based on the new input received, and new comments are generated by the characters based on the updated content.

[1621] Step 12:

[1622] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[1623] Step 13:

[1624] The server formats the generated conclusion in JSON format and sends it to the device, which visually displays the received conclusion to the user.

[1625] Step 14:

[1626] The user confirms the displayed conclusion, which allows the user to reach the optimal solution or decision that was reached through discussion of different perspectives.

[1627] Example 1

[1628] 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."

[1629] In the traditional idea generation and discussion process, users are responsible for all thinking alone, leading to problems of biased perspectives and reduced efficiency. Furthermore, incorporating multiple different opinions takes a lot of time and effort, making it difficult to quickly reach an optimal conclusion. Furthermore, there is a lack of support for users to smoothly facilitate discussions, which can lead to confusion.

[1630] 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.

[1631] In this invention, the server includes means for receiving information input by a user, means for converting the input information into a specific format and transmitting it to the server, means for analyzing the information received by the server using natural language processing technology, means for automatically generating multiple profiles based on the analysis, means for visually displaying the information in the generated profiles, means for the profiles to start a discussion, means for the server to analyze the content of the discussion and draw a conclusion, and means for presenting the conclusion to the user. This allows users to more effectively and quickly reach optimal conclusions through discussions from diverse perspectives. Furthermore, the discussion progresses smoothly, reducing bias in perspectives and reduced efficiency.

[1632] "User" refers to a person who uses this system to input and operate information.

[1633] "Information" refers to the ideas and data that users input into the system.

[1634] "Means" refers to methods or apparatuses for implementing the functions or processes described in the present invention.

[1635] A "profile" is information about the attributes and roles of a virtual character generated by the server for discussion purposes.

[1636] A "server" is a central computer system that receives and analyzes information from users, generates profiles, and manages discussions.

[1637] "Natural language processing technology" is a technology that allows computers to understand and analyze human language.

[1638] "Analysis" is the process of understanding, classifying, and evaluating input information.

[1639] "Discussion" is the process by which generated profiles exchange ideas and opinions and reach a conclusion.

[1640] A "conclusion" is a final judgment or recommendation that is reached as a result of a discussion.

[1641] "Visually displaying" means graphically presenting profile information and discussion content in a format that is easy for users to view.

[1642] The present invention is a system in which multiple profiles are generated based on information entered by a user, and discussions are held to reach a conclusion. A specific embodiment of the program for this system is described below. In particular, the hardware and software used, as well as specific data processing and calculations, are described in detail.

[1643] Hardware and Software Configuration

[1644] This system mainly uses the following hardware and software:

[1645] Cloud computing platforms as servers (e.g., Amazon Web Services, Google Cloud Platform)

[1646] User devices as terminals (e.g., PCs, tablets, smartphones)

[1647] NLP engines as natural language processing technologies (e.g., spaCy)

[1648] Language models as generative AI models (e.g., GPT-3)

[1649] Secure protocol used for sending and receiving data (e.g. HTTPS)

[1650] JSON format as data format

[1651] Program processing

[1652] 1. Enter your information:

[1653] The user enters information into the input form on the terminal.

[1654] The terminal receives the input information, converts it into JSON format, and sends it to the server.

[1655] 2. Analysis of Information and Profile Creation:

[1656] The information received by the server is analyzed using natural language processing technology (e.g., spaCy).

[1657] Based on the analysis results, the server automatically generates the profiles (e.g., moderator, bold, cautious, balancer) necessary to smoothly advance the discussion using a generative AI model (e.g., GPT-3).

[1658] The server sends the generated profile information to the terminal in JSON format.

[1659] 3. Visual representation of the profile:

[1660] The device visually displays the received profile information to the user using HTML, CSS, and Javascript.

[1661] 4. Starting a discussion:

[1662] When a user clicks the discussion start button, the terminal sends a request to start a discussion to the server.

[1663] The server generates messages for each profile and transmits the messages to the terminals one by one.

[1664] The terminal displays the statement to the user.

[1665] 5. Developing a discussion and drawing conclusions:

[1666] Users can join the discussion and enter new questions or ideas, which are also sent to the server and reflected in the discussion.

[1667] The server analyzes the discussion content in real time and generates the optimal conclusion.

[1668] The server sends the generated conclusion to the terminal, which displays the conclusion to the user.

[1669] Specific examples

[1670] Below are some specific examples on the theme of "market introduction strategies for new products."

[1671] 1. User Idea Submissions:

[1672] The user enters "I would like to think of a market introduction strategy for a new health food" into the input form on the terminal.

[1673] The device converts this information into JSON format and sends it to the server.

[1674] 2. Generate and view the profile:

[1675] The server analyzes the information it receives and extracts keywords such as "health food," "market introduction," and "strategy."

[1676] Based on this, the server generates profiles for Moderator, Bold, Cautious, and Balancer.

[1677] The server transmits the generated profile information to the terminal.

[1678] The terminal displays the profile information to the user.

[1679] 3. Initiating and Conducting a Discussion:

[1680] The terminal displays a "Start Discussion" button, and when the user clicks on it, the discussion begins.

[1681] The moderator says, "Now let's hear from the bold ones."

[1682] The bold suggestion is to launch a new social media campaign and collaborate with influencers.

[1683] Cautious voices say, "The idea is good, but we need to carefully evaluate the cost-benefit and risks."

[1684] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[1685] The user joins the discussion and types, "Should we do additional cost estimation?" The device sends this information to the server.

[1686] 4. Draw conclusions:

[1687] The server analyzes the content of the discussion and generates a conclusion that "we will conduct a trial SNS campaign and gradually expand it based on the results."

[1688] The server sends this conclusion to the terminal, which displays it to the user.

[1689] Examples of prompt statements

[1690] Here are some examples of prompts to input to a generative AI model:

[1691] User: I would like to think about a market introduction strategy for a new health food.

[1692] Server: The moderator (you) is responsible for keeping the discussion running smoothly.

[1693] Moderator: "Now let's hear from the bold ones."

[1694] Bold: "We should launch a new social media campaign and collaborate with influencers."

[1695] Cautious: "The idea is good, but the costs and risks need to be carefully assessed."

[1696] Balancer: "Both opinions are important. Let's compare the specific costs and expected benefits."

[1697] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1698] Step 1: Enter your information

[1699] The user inputs information (e.g., "I would like to consider a market introduction strategy for a new health food product") into an input form on the terminal.

[1700] Input: Information entered by the user into the device.

[1701] The device processes this input information using UTF-8 character encoding and converts it into a JSON-formatted object.

[1702] Data processing: Character encoding of information and conversion to JSON format.

[1703] The device sends the generated JSON object to the server using the secure HTTPS protocol.

[1704] Output: Reformatted JSON information.

[1705] Step 2: Analyze the information and create a profile

[1706] The server parses the received JSON data and uses natural language processing techniques (e.g., spaCy) to extract important keywords and phrases.

[1707] Input: JSON formatted information sent from the device.

[1708] Data processing: Keyword extraction and phrase analysis using a natural language processing engine.

[1709] Based on the extracted keywords, the server automatically generates profiles (e.g., moderator, bold, cautious, balancer) necessary to smoothly advance the discussion using a generative AI model (e.g., GPT-3).

[1710] Data calculation: Profile generation using AI models.

[1711] The server determines the attributes of each profile (e.g., name, personality, expertise, etc.) and packages them into a JSON format.

[1712] Output: JSON format data of generated profile information.

[1713] The server sends the generated profile information to the device using the secure HTTPS protocol.

[1714] Step 3: Visual representation of the profile

[1715] Renders a user interface (UI) that displays the profile information received by the device.

[1716] Input: JSON data of profile information received from the server.

[1717] The device uses HTML, CSS, and JavaScript to visually display the profile's name, role, profile picture, etc.

[1718] Behavior: Rendering and displaying the UI.

[1719] Output: Profile information visually displayed to the user.

[1720] Step 4: Start a discussion

[1721] When a user clicks the Start Discussion button, a JavaScript event handler on the device detects this.

[1722] Input: The action of the user clicking the Start Discussion button.

[1723] The terminal composes a request to start a discussion in JSON format and sends it to the server.

[1724] Data processing: Composing a discussion start request.

[1725] Output: Discussion start request sent to the server.

[1726] Step 5: Facilitate the discussion

[1727] The server receives a request to start a discussion and starts the process of generating comments for each profile.

[1728] Input: A request to start a discussion from a terminal.

[1729] The server uses a generative AI model (e.g., GPT-3) to generate utterances based on the role of each profile.

[1730] Data computation: utterance generation using AI models.

[1731] The server sequentially sends the generated comments to the terminal in JSON format.

[1732] Output: JSON data of the generated utterance.

[1733] The device displays the comments received in real time, allowing users to visually check the progress of the discussion.

[1734] What it does: Real-time display of what's being said.

[1735] Users can join the discussion and enter new questions or ideas, which are also sent back to the server and reflected in the discussion.

[1736] Input: New questions and ideas.

[1737] Data arithmetic: Processing new input information and reflecting it in discussions.

[1738] Step 6: Draw conclusions

[1739] The server analyzes all discussion content in real time and applies algorithms to derive optimal conclusions based on the data.

[1740] Input: All data from the discussion.

[1741] Data arithmetic: Applying conclusion-drawing algorithms.

[1742] The server generates the final result and sends it to the device in JSON format.

[1743] Output: JSON data of the generated conclusions.

[1744] The terminal displays the received conclusion on a user interface to provide the user with a final conclusion.

[1745] Action: Display of final conclusion.

[1746] (Application example 1)

[1747] 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."

[1748] In the development and operation of autonomous vehicles, there is a need to efficiently hold discussions from a variety of expert perspectives and derive objective, optimal conclusions. Conventional methods often make it difficult to reconcile opinions between experts and advance discussions, requiring time and effort. Furthermore, there is a problem in that it is difficult for users to actively participate in discussions, resulting in only a portion of opinions being reflected. In response to these issues, a system that supports efficient and fair discussions and allows users to actively participate is desired.

[1749] 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.

[1750] In this invention, the server includes means for receiving a task input by a user, means for analyzing the task content and automatically generating a plurality of discussion characters, means for the generated discussion characters to engage in a discussion, means for analyzing the discussion content and deriving a conclusion, means for presenting the conclusion to the user, and means for the user to input additional questions or suggestions while the discussion is in progress. This enables efficient and fair discussions from various perspectives and provides an environment in which users can actively participate in the discussion.

[1751] "User" refers to a person who inputs data into or operates the system.

[1752] "Issues" refer to problems that users want to solve or themes that they want to consider.

[1753] "Means of receiving" refers to the interface or function that allows the system to acquire tasks from users.

[1754] "Analyzing" refers to the process of understanding the content of the input task and extracting the necessary information.

[1755] "Discussion characters" refer to virtual roles automatically generated for the purpose of conducting discussions.

[1756] "Means of automatic generation" refers to the process by which the system uses AI technology to automatically create discussion characters.

[1757] "Means for holding discussions" refers to a function that allows the generated discussion characters to converse and exchange opinions with each other.

[1758] "Means of deriving a conclusion" refers to the process for generating the most appropriate conclusion based on the discussion that has taken place.

[1759] "Presentation means" refers to the interface or functionality for informing the user of the generated conclusions.

[1760] "Means for inputting additional questions and suggestions" refers to the function that allows users to input new questions or suggestions to the system during the discussion.

[1761] This invention is a system for efficiently conducting discussions from various expert viewpoints and deriving objective and optimal conclusions in the development and operation of autonomous vehicles. This system uses automatically generated discussion characters to hold discussions based on issues entered by the user, and analyzes the results to reach a conclusion.

[1762] System configuration

[1763] 1. User Device

[1764] Hardware: User interfaces such as smartphones and head-mounted displays

[1765] Software: input forms, display interfaces, communication modules

[1766] 2. Server

[1767] Hardware: Servers with high-performance computing capabilities (including cloud servers)

[1768] Software: Natural language processing technology (e.g., GPT-3), data analysis module, character generation module, discussion progression module, conclusion drawing module

[1769] System Operation Overview

[1770] 1. User Device

[1771] The user inputs a problem, such as "I want to think of a market launch strategy for a new autonomous driving algorithm," into an input form.

[1772] This input task is converted into JSON format and sent to the server.

[1773] 2. Server

[1774] The server analyzes the received problem and extracts keywords, such as "autonomous driving," "algorithm," and "market launch."

[1775] Based on the extracted keywords, the necessary discussion characters are automatically generated, including "technical experts," "safety managers," "management managers," and "user representatives."

[1776] The role and profile of each character are determined and transmitted to the user terminal.

[1777] 3. Discussion Progress

[1778] The user terminal displays the generated character information, and the user starts a discussion.

[1779] The server generates utterances according to the character's role and sends them to the user's terminal in sequence. During the discussion, the user can input additional questions or suggestions.

[1780] 4. Drawing conclusions

[1781] The server analyzes the content of the discussion and derives the optimal conclusion, which is then sent to the user's terminal and presented to the user.

[1782] Specific examples

[1783] Below is a concrete example of a discussion regarding a market launch strategy for a new autonomous driving algorithm.

[1784] Example prompt sentence:

[1785] As a technology expert, please discuss "Go-to-market strategies for new autonomous driving algorithms."

[1786] As a safety manager, please discuss the "market launch strategy for new autonomous driving algorithms."

[1787] As a manager, please discuss the "market launch strategy for new autonomous driving algorithms."

[1788] As a user representative, please discuss the "market launch strategy for new autonomous driving algorithms."

[1789] This system utilizes generative AI models (e.g., GPT-3) to enable efficient and fair discussions from diverse perspectives, allowing users to actively participate in the discussion and derive deeper ideas and objective conclusions.

[1790] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1791] Step 1:

[1792] The user inputs a task into the device's input form. This task may include a specific problem such as "I want to think about a market launch strategy for a new autonomous driving algorithm." The device converts this input into JSON format and sends it to the server.

[1793] Input: The assignment entered by the user

[1794] Output: Issue data in JSON format

[1795] Step 2:

[1796] The server analyzes the received JSON-formatted problem data and extracts keywords, such as "autonomous driving," "algorithm," and "market launch." This analysis uses natural language processing technology (e.g., GPT-3).

[1797] Input: Issue data in JSON format

[1798] Output: Extracted keyword list

[1799] Step 3:

[1800] The server generates the necessary discussion characters based on the extracted keywords. The generated characters include "technical expert," "safety manager," "management manager," and "user representative." It generates a prompt for each character and determines the character's profile and role using an AI model (e.g., GPT-3).

[1801] Input: Keyword list

[1802] Output: Character profile and role

[1803] Step 4:

[1804] The server sends the generated character's profile and role to the user's device, which receives it and displays it visually to the user. The user can start a discussion by clicking the "Start Discussion" button on their device.

[1805] Input: Character profile and role

[1806] Output: Display character information to the user

[1807] Step 5:

[1808] The server generates utterances based on the character's role. This utterance generation uses an AI model (e.g., GPT-3). The generated utterances are sequentially sent to the user's device and displayed to the user.

[1809] Input: Character profile and role

[1810] Output: What the character says

[1811] Step 6:

[1812] Users can enter additional questions or suggestions into the discussion on their terminals. These additional inputs are also converted into JSON format and sent to the server, which receives them and updates the discussion.

[1813] Input: User's additional questions or suggestions

[1814] Output: Updated discussion

[1815] Step 7:

[1816] The server analyzes the entire discussion and derives the optimal conclusion. This analysis and conclusion generation uses an AI model (e.g., GPT-3). The server then sends the final conclusion to the user's device.

[1817] Input: Updated discussion

[1818] Output: Conclusion

[1819] Step 8:

[1820] The user terminal displays the received conclusion to the user, who can then confirm the conclusion and use it to decide what action to take next.

[1821] Input: Conclusion

[1822] Output: Display conclusion to user

[1823] 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.

[1824] The present invention is a system in which multiple generated characters develop a discussion based on an idea input by a user, and a conclusion is reached by combining an emotion engine while recognizing the user's emotional state. The specific processing of the program of this system is explained in natural language below. Also, specific examples are provided to explain the embodiment in detail.

[1825] Program processing

[1826] 1. User Idea Submissions

[1827] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[1828] 2. Character Generation

[1829] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[1830] The server analyzes the stored idea data using natural language processing (NLP), and extracts key topics and keywords from the analysis results.

[1831] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer), and retrieves the character profiles and roles from the database.

[1832] The character information generated by the server is formatted in JSON format and sent to the terminal.

[1833] 3. Starting a discussion

[1834] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[1835] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[1836] 4. Operation of the Emotion Engine

[1837] The device sends the user's input and actions (e.g., keystroke speed, mouse movements, facial recognition, etc.) to the emotion engine in real time.

[1838] The emotion engine analyzes this data to detect the user's emotional state (e.g., stress, excitement, concentration, etc.).

[1839] The device transmits the detected emotional state to the server.

[1840] 5. Discussion Progress

[1841] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[1842] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[1843] The server receives data from the emotion engine and adjusts the character's speech and tone based on the user's emotional state.

[1844] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[1845] 6. Drawing conclusions

[1846] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[1847] The server formats the generated conclusion in JSON format and sends it to the device, which visually displays the received conclusion to the user.

[1848] Specific examples

[1849] Below are some specific examples on the theme of "market introduction strategies for new products."

[1850] 1. User Idea Submissions

[1851] The user enters the idea into the input form on the device, "I want to think of a market introduction strategy for a new health food." The device converts this idea into JSON format and sends it to the server.

[1852] 2. Character Generation

[1853] The server analyzes the ideas received and extracts keywords such as "healthy food," "market introduction," and "strategy."

[1854] Based on this, the server automatically generates characters for the moderator, bold, cautious, and balancer. The server determines the role and profile of each character and sends this to the device.

[1855] 3. Starting a discussion

[1856] The terminal displays information about the generated character to the user. The user can start a discussion by clicking the "Start Discussion" button.

[1857] 4. Operation of the Emotion Engine

[1858] The device sends the user's input and behavioral data (e.g., keystroke speed and facial expression) to the emotion engine, which then detects the user's emotional state. The detected emotional state is then sent from the device to the server.

[1859] 5. Discussion Progress

[1860] The moderator says, "Now let's hear from the bold ones."

[1861] The bold suggest, "You should launch a social media campaign and collaborate with influencers."

[1862] Cautious voices say, "The idea is good, but we need to evaluate the cost-benefit and risks."

[1863] The balancer makes the adjustment by saying, "Both opinions are important. Let's compare the specific costs and expected effects."

[1864] The server adjusts the bolder's tone of speech, for example softening it, depending on the user's emotional state.

[1865] A user joins the discussion and types, "Should we do additional cost estimation?" The device sends this idea to the server and the discussion is updated.

[1866] 6. Drawing conclusions

[1867] The server analyzes the content of the discussion and generates a conclusion: "We will conduct a trial SNS campaign and gradually expand it based on the results." The conclusion is sent to the device and displayed to the user.

[1868] This specific embodiment allows users to discuss with characters with diverse perspectives, encouraging flexible responses that reflect their emotional state, enabling them to generate deeper ideas and make more objective decisions.

[1869] The processing flow will be explained below.

[1870] Step 1:

[1871] The user inputs their idea into the input form on the device. The device receives the input idea and converts it into JSON format.

[1872] Step 2:

[1873] The device sends a POST request to the server with the converted JSON format idea data. The server saves the received idea data.

[1874] Step 3:

[1875] The idea data stored on the server is analyzed using natural language processing (NLP), which extracts key topics and keywords for the ideas.

[1876] Step 4:

[1877] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer), and retrieves each character's profile and role from the database.

[1878] Step 5:

[1879] The character information generated by the server is formatted in JSON format and sent to the terminal.

[1880] Step 6:

[1881] The device parses the received character information and visually displays it to the user, presenting the character's role and profile in an easy-to-understand format.

[1882] Step 7:

[1883] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start the discussion to the server.

[1884] Step 8:

[1885] The device sends the user's input and actions (e.g., keystroke speed, mouse movement, facial recognition, etc.) to the emotion engine in real time. The emotion engine analyzes this data and detects the user's emotional state.

[1886] Step 9:

[1887] The device sends the emotional state detected by the emotion engine to the server, which then adjusts the content and tone of the character's speech based on the emotional state received.

[1888] Step 10:

[1889] The server generates utterances based on each character's role. An AI model (e.g., GPT-3) is used to generate utterances. Different utterances are generated for each character.

[1890] Step 11:

[1891] The server sends generated messages to the terminal in sequence, and the received messages are displayed to the user in chat format.

[1892] Step 12:

[1893] Users can join the discussion and input any questions or new ideas they have. The device sends the input information to the server.

[1894] Step 13:

[1895] The server updates the discussion based on the new input received, and new comments are generated by the characters based on the updated content.

[1896] Step 14:

[1897] The server analyzes the discussion log and stores each character's comments in a database. It then generates the optimal conclusion based on the content of the discussion.

[1898] Step 15:

[1899] The server formats the generated conclusion in JSON format and sends it to the device.

[1900] Step 16:

[1901] The terminal visually displays the received conclusion to the user, who then confirms the displayed conclusion. This allows the user to obtain the optimal solution or decision-making result obtained through discussion from different perspectives.

[1902] Example 2

[1903] 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."

[1904] In conventional idea generation support systems, it was difficult to discuss ideas entered by users from various perspectives and to respond flexibly while taking into account their emotional state. Furthermore, the generated characters' comments were not adjusted according to the user's emotions, making it difficult to advance effective discussions. The present invention aims to solve these problems and provide a system that effectively supports the user's idea generation process from multiple angles.

[1905] 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.

[1906] In this invention, the server includes means for receiving ideas input by a user, means for converting the input ideas into JSON format, and means for transmitting the converted JSON-formatted idea data to the server. This allows users to easily input ideas and process them efficiently. The server also includes means for analyzing the saved idea data using natural language processing to extract key topics and keywords, means for automatically generating multiple characters based on the analysis results, and means for formatting the generated character information in JSON format and transmitting it to a terminal. This allows users to generate characters with diverse perspectives and engage in appropriate discussions. The server also includes means for an emotion engine to detect the user's emotional state, means for adjusting the content and tone of the character's comments based on the user's emotional state, and means for utilizing a generative AI model to generate comments appropriate to each character's role. This enables flexible responses that reflect the user's emotional state, improving the quality of discussions.

[1907] A "user" is a person who operates the system and inputs ideas, and is the primary user of the system.

[1908] A "terminal" is an input device operated by a user, and is a device for inputting ideas and displaying information.

[1909] "Server" refers to a central management system that stores, analyzes, and processes data sent by users.

[1910] "JSON format" is an abbreviation for JavaScript Object Notation, and is a lightweight data exchange format that makes it easy to exchange data.

[1911] "Conversion" refers to the process of changing input data into another format.

[1912] A "POST request" is a method of sending data to a server using the HyperText Transfer Protocol (HTTP).

[1913] "Natural Language Processing (NLP)" is a technology that allows computers to understand, analyze, and generate human language.

[1914] A "character" is a virtual person automatically generated for the purpose of discussion, with a specific role and profile.

[1915] The "Emotion Engine" is a machine learning model for analyzing and detecting a user's emotional state.

[1916] A "generative AI model" refers to an artificial intelligence model used to perform tasks such as natural language generation and dialogue generation.

[1917] A "prompt" is an instruction given to a generative AI model that the model uses to generate an appropriate response or generation.

[1918] "Discussion" refers to the process in which multiple characters exchange opinions on a user's idea.

[1919] A "conclusion" is the final result or proposal reached through a discussion.

[1920] This invention is a system that supports users in discussing ideas from various perspectives and responding flexibly. This system is realized using users, terminals, a server, a generative AI model, and an emotion engine.

[1921] First, the user inputs an idea into the input form on the device. For example, this input might be in the form of "I want to think of a market introduction strategy for a new health food." This idea is converted into JSON format by the device. The device then sends this JSON-formatted idea data to the server via a POST request.

[1922] Next, the server stores the JSON-formatted idea data received from the device. A NoSQL database (e.g., MongoDB) is used for this storage. The server then analyzes the stored idea data using natural language processing (NLP). An NLP library (e.g., spaCy or NLTK) is used for this analysis, and key topics and keywords for the ideas (e.g., "healthy food," "market introduction," and "strategy") are extracted. Based on this, characters are automatically generated to participate in the discussion. Character profile and role information is retrieved from a database (e.g., PostgreSQL).

[1923] The generated character information is formatted in JSON format by the server and sent to the device. The device parses the received character information and visually displays it to the user. For example, the character's name, role, and profile information are displayed. At this point, the user clicks the "Start Discussion" button to begin the discussion.

[1924] When a discussion begins, the device collects user input and behavioral data (e.g., keystroke speed, mouse movement, and facial recognition data) in real time and sends them to the emotion engine. The emotion engine then analyzes this data using a machine learning model (e.g., a TensorFlow model) to detect the user's emotional state. The device then transmits the detected emotional state to the server.

[1925] The server uses a generative AI model (e.g., GPT-3) to generate utterances appropriate for each character's role. A prompt is used to generate the utterances. An example of a prompt is: "As a bold person, please give your opinion on the market launch strategy for health foods."

[1926] The comments generated by the server are sent to the device in sequence, and the comments received by the device are displayed to the user in chat format. This uses a conversational UI, making it appear to the user that the characters are engaged in a dialogue. The server receives data from the emotion engine and adjusts the content and tone of the characters' comments based on the user's emotional state. For example, if the user is relaxed, it softens the tone of the bolder comments. Users participate in the discussion, inputting unclear points and new ideas. Input such as "Should we do additional cost estimates?" is sent from the device to the server, and the discussion is updated.

[1927] Once the discussion has ended, the server analyzes the discussion log and stores each character's comments in a database. The server generates an optimal conclusion based on the content of the discussion, formats the conclusion into JSON format, and sends it to the device. The device then visually displays this conclusion to the user. A specific conclusion might be something like, "We will conduct a trial SNS campaign and gradually expand it based on the results."

[1928] In this way, users can interact with characters with diverse perspectives, responding flexibly to their emotional state, leading to deeper idea generation and objective decision-making.

[1929] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1930] Step 1:

[1931] The user inputs their idea into the input form on the device. For example, they might input, "I want to think of a market introduction strategy for a new health food." The device receives the user's input and converts it into JSON format. Input data: User's idea (text format) → Output data: Idea (JSON format).

[1932] Step 2:

[1933] The terminal sends the converted JSON-formatted idea data to the server via a POST request. Input data: JSON-formatted idea data → Output data: Send request to the server.

[1934] Step 3:

[1935] The server stores the JSON format idea data received from the device. A NoSQL database (e.g., MongoDB) is used. Input data: JSON format idea data → Output data: idea data stored in the database.

[1936] Step 4:

[1937] The server analyzes the stored idea data using natural language processing (NLP). It uses an NLP library (e.g., spaCy or NLTK) to extract key topics and keywords for ideas. Input data: idea data (JSON format) → Processing: NLP analysis → Output data: keywords and topics.

[1938] Step 5:

[1939] The server automatically generates characters for discussions based on the analysis results. Character profiles and role information are obtained from a database (e.g., PostgreSQL). Input data: Keywords and topics → Processing content: Automatic character generation, profile acquisition → Output data: Character information (JSON format).

[1940] Step 6:

[1941] The server formats the generated character information in JSON format and sends it to the terminal. Input data: Character information → Processing content: Format into JSON format → Output data: Send request to the terminal.

[1942] Step 7:

[1943] The device analyzes the character information received and displays it visually to the user. Using HTML and CSS, a UI is generated to display the character's name, role, and profile information. Input data: character information (JSON format) → Processing: Parse, UI generation → Output data: user interface.

[1944] Step 8:

[1945] The user clicks the "Start discussion" button. The device receives the user's instruction and sends a request to start a discussion to the server. Input data: User's click operation → Output data: Request to start a discussion.

[1946] Step 9:

[1947] The device collects user input and behavioral data (e.g., keystroke speed, mouse movement, facial recognition data) in real time and sends it to the emotion engine. Input data: behavioral data → Output data: transmission request to the emotion engine.

[1948] Step 10:

[1949] The emotion engine analyzes the data using a machine learning model (e.g., TensorFlow model) and detects the user's emotional state. Input data: behavioral data → Processing content: detection of emotional state → Output data: user's emotional state.

[1950] Step 11:

[1951] The server receives data from the emotion engine and adjusts the content and tone of the character's speech based on the user's emotional state. Input data: User's emotional state → Processing content: Adjustment of speech content and tone → Output data: Adjusted speech.

[1952] Step 12:

[1953] The server uses a generative AI model (e.g., GPT-3) to generate utterances appropriate for each character's role. A sample prompt might be, "As a bold person, please share your opinion on the market launch strategy for health foods."

[1954] Input data: Character role, prompt sentence → Processing content: Speech generation by AI model → Output data: Generated speech.

[1955] Step 13:

[1956] The server sends generated comments to the device in sequence. The device displays the received comments to the user in chat format. Displayed in an interactive UI. Input data: Generated comments → Processing content: Displayed in chat → Output data: Displayed to the user.

[1957] Step 14:

[1958] The user joins the discussion and inputs any unclear points or new ideas. For example, "Should we do additional cost estimates?" The device sends this content to the server, and the discussion is updated. Input data: User's new idea → Output data: Send request to the server.

[1959] Step 15:

[1960] The server analyzes the discussion log and stores each character's comments in a database. The optimal conclusion is generated based on the content of the discussion. Input data: discussion log → Processing content: NLP analysis, conclusion generation → Output data: generated conclusion.

[1961] Step 16:

[1962] The server formats the generated conclusion into JSON format and sends it to the device. The device visually displays the conclusion to the user. A specific conclusion might be something like "We will conduct a trial SNS campaign and gradually expand it based on the results." Input data: Generated conclusion → Processing content: Format into JSON format, display to user → Output data: Conclusion presented to the user.

[1963] (Application example 2)

[1964] 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."

[1965] In conventional systems, multiple characters could discuss and reach a conclusion based on ideas input by the user, but they were unable to reflect the user's emotional state in real time. This made it difficult to respond appropriately to the user's emotions or to flexibly advance the discussion, ultimately causing a decline in user satisfaction and the quality of the discussion. In customer interactions and customer service in physical stores, there was no system that could take customer emotions into account, so improvements in customer satisfaction could not be expected.

[1966] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1967] In this invention, the server includes means for receiving ideas input by users, means for analyzing the input ideas and automatically generating multiple characters, means for the generated characters to hold discussions, means for analyzing the content of the discussion and deriving a conclusion, means for presenting the conclusion to users, means for emotionally analyzing customer input and behavioral data in real time, and means for adjusting the characters' comments based on the emotional state. This enables flexible discussions that reflect the emotional state of users in real time, which is expected to improve customer satisfaction in physical stores.

[1968] "User" refers to an individual or group that uses the system to input ideas and participate in discussions.

[1969] "Ideas" refer to the ideas and thoughts that users input into the system, and are the information that forms the basis of discussion.

[1970] "Characters" refer to virtual characters who are automatically generated within the system based on the user's ideas and who play various roles in discussions.

[1971] A "discussion" refers to a conversation or debate process in which multiple characters express their opinions based on their respective roles and reach a conclusion.

[1972] "Conclusion" refers to the result or proposal that is ultimately reached after analyzing the content of the discussion.

[1973] "Emotion analysis" refers to the process of analyzing user input and behavioral data to detect the user's emotional state in real time.

[1974] "Adjusting speech" refers to changing the content and tone of a character's speech based on emotion analysis to generate appropriate speech that corresponds to the user's emotional state.

[1975] A "physical store" is a facility located in a physical location where customers can visit in person to provide goods or services.

[1976] "Customer" refers to a visitor who visits a physical store and intends to purchase or use a product or service.

[1977] "Customer service" refers to activities that involve interacting with and providing support to customers in physical stores, with the aim of improving customer satisfaction.

[1978] "Emotion engine" refers to an analysis device or program that detects the user's emotional state and reflects it in processes within the system.

[1979] The present invention is a system in which multiple characters discuss ideas input by a user and use an emotion engine to reach a conclusion while reflecting the user's emotional state in real time. This system aims to improve customer service in brick-and-mortar stores. Specific embodiments for realizing this system are described below.

[1980] 1. System Overview

[1981] Hardware used

[1982] Device: Smartphone, tablet or PC. Device used by staff or customers in a physical store.

[1983] Server: A remotely located computing device that analyzes data and runs generative AI models.

[1984] Software used

[1985] Natural Language Processing (NLP): A technique used to analyze user-entered ideas.

[1986] Generative AI model (e.g. GPT-3): A model used to generate character utterances.

[1987] Emotion engine: Software for analyzing a user's emotional state in real time.

[1988] 2. Program processing explanation

[1989] 1. Idea input and analysis

[1990] The user inputs an idea on the terminal, which is converted into JSON format and sent to the server.

[1991] The server analyzes the received ideas using natural language processing (NLP) to extract key topics and keywords.

[1992] 2. Character Generation

[1993] Based on the analysis results, the server automatically generates the characters necessary for the discussion (e.g., moderator, bold, cautious, balancer).

[1994] The role and profile of each character is retrieved from the database, and the generated character information is sent to the terminal.

[1995] 3. Starting a discussion

[1996] The terminal displays the received character information to the user, and the discussion begins when the user clicks the "Start discussion" button.

[1997] This discussion is initiated by sending a request to the server to start the discussion.

[1998] 4. Operation of the Emotion Engine

[1999] The device transmits user input and actions (e.g., keystroke speed, mouse movement, facial recognition) to the emotion engine in real time.

[2000] The emotion engine analyzes these data and detects the user's emotional state (e.g., stress, excitement, concentration, etc.), which is then sent to the server.

[2001] 5. Discussion Progress

[2002] The server uses a generative AI model to generate each character's utterances. For example, the "cautious" character might say, "A risk assessment is required."

[2003] The server adjusts the content and tone of the character's speech depending on the user's emotional state.

[2004] 6. Drawing conclusions

[2005] The server analyzes the discussion logs and generates an optimal conclusion, which is formatted in JSON and sent to the device.

[2006] The terminal visually displays the received conclusions to the user.

[2007] 3. Specific Examples

[2008] Example: If a customer enters "I'm looking for a light summer jacket"

[2009] Input prompt statement:

[2010] User Wants: A lightweight summer jacket

[2011] Bolder opinion: Experiment with different colors and designs!

[2012] Cautious observers: It's also a good idea to check the quality and durability of the materials.

[2013] Balancer's opinion: Why not take a look at some mid-priced items with good design and quality?

[2014] In this example, users can receive advice from multiple perspectives and receive real-time responses based on emotion analysis, resulting in a more satisfying purchasing experience. This system can improve the quality of customer service in physical stores.

[2015] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2016] Step 1:

[2017] Entering and receiving ideas

[2018] The user inputs ideas using a terminal. This input is often in text format and is specific, such as "I would like to come up with a market introduction strategy for a new health food."

[2019] The device receives this idea and converts it into JSON format, for example, {"idea": "I want to think about a market launch strategy for a new health food product"}.

[2020] The terminal sends the converted JSON data to the server, which receives it and stores it in a database.

[2021] Step 2:

[2022] Idea analysis and character generation

[2023] The server analyzes the received JSON-formatted idea data using natural language processing (NLP) technology. The purpose of the analysis is to extract keywords and topics for the ideas. The analysis results include keywords such as "healthy food," "market introduction," and "strategy."

[2024] Based on the analysis results, the server automatically generates characters to participate in the discussion. These characters include the moderator, the bold, the cautious, and the balancer. Each character's profile and role are retrieved from a database.

[2025] The server formats the generated character information in JSON format and sends it to the device. An example of formatted data is as follows: {"characters": [{"role": "Moderator", "profile": "Neutral moderator"}, {"role": "Bold", "profile": "Provides challenging opinions"}, {"role": "Cautious", "profile": "Provides risk-oriented opinions"}, {"role": "Balancer", "profile": "Reconcile the opinions of both parties"}]}

[2026] Step 3:

[2027] View character information and start discussions

[2028] The device parses the received character information and displays it for the user to visually confirm, with each character's role and profile displayed on the screen.

[2029] The user clicks the "Start Discussion" button. The device captures this click event and sends a request to start a discussion to the server.

[2030] Step 4:

[2031] Emotion analysis using an emotion engine

[2032] The device monitors user input and actions in real time, including keystroke speed, mouse movements, and facial recognition.

[2033] For example, if the user is typing quickly, the emotion engine detects an "excited" state.

[2034] The detected emotional state is sent to the server in JSON format, e.g. {"emotion": "excited"}

[2035] Step 5:

[2036] Discussion management and coordination

[2037] The server uses a generative AI model (e.g., GPT-3) to generate statements for each character. For example, the Cautious character might say, "That's a good idea, but we need to evaluate the cost-benefit and risks."

[2038] The server adjusts the content and tone of the characters' speech depending on the user's emotional state. For example, if the user is excited, the "Bold" character might suggest in a calmer tone, "You should launch a social media campaign and collaborate with influencers."

[2039] The server sends each character's comments in JSON format to the device, which then displays them to the user in chat format. The discussion is updated as users enter new ideas or questions.

[2040] Step 6:

[2041] Drawing and presenting conclusions

[2042] The server analyzes the discussion log and generates an optimal conclusion based on the discussion, such as "run a trial SNS campaign and gradually expand it based on the results."

[2043] This conclusion is formatted as JSON and sent to the device. Example: {"conclusion": "Test social media campaigns and gradually expand based on the results"}

[2044] The terminal visually displays the received conclusions to the user.

[2045] 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.

[2046] 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.

[2047] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2048] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2049] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2050] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2051] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2052] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2053] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2054] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2055] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2056] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2057] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2058] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2059] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2060] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2061] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2062] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2063] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2064] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2065] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2066] The following is further disclosed regarding the above embodiment.

[2067] (Claim 1)

[2068] a means for receiving user-entered ideas;

[2069] means for analyzing the input idea and automatically generating a plurality of characters;

[2070] A means for the generated characters to hold discussions;

[2071] means for analyzing the content of the discussion and drawing conclusions;

[2072] The system includes a means for presenting said conclusion to a user.

[2073] (Claim 2)

[2074] The system of claim 1 , wherein the characters have roles of moderator, bold, cautious, and balancer.

[2075] (Claim 3)

[2076] The system of claim 1 , wherein the character's role is user customizable.

[2077] "Example 1"

[2078] (Claim 1)

[2079] means for receiving user-entered information;

[2080] means for converting the input information into a specific format and transmitting the converted information to a server;

[2081] means for analyzing the information received by the server using natural language processing technology;

[2082] means for automatically generating a plurality of profiles based on said analysis;

[2083] a means for visually displaying the generated profile information;

[2084] a means for said profile to initiate a discussion;

[2085] A means for the server to analyze the content of the discussion and draw a conclusion;

[2086] The system includes a means for presenting said conclusion to a user.

[2087] (Claim 2)

[2088] 2. The system of claim 1, wherein the profiles have roles of moderator, bold, cautious, and balancer.

[2089] (Claim 3)

[2090] 10. The system of claim 1, wherein the profile roles are user customizable.

[2091] "Application Example 1"

[2092] (Claim 1)

[2093] means for receiving user-entered assignments;

[2094] means for analyzing the input topic and automatically generating a plurality of discussion characters;

[2095] A means for the generated discussion character to hold a discussion;

[2096] means for analyzing the content of the discussion and drawing conclusions;

[2097] means for presenting said conclusion to a user;

[2098] A system that includes a means for users to enter additional questions or suggestions while a discussion is in progress.

[2099] (Claim 2)

[2100] The system of claim 1 , wherein the discussion characters have roles of technical expert, safety manager, management, and user representative.

[2101] (Claim 3)

[2102] The system of claim 1 , wherein the discussion character roles are user customizable.

[2103] "Example 2: Combining Emotion Engines"

[2104] (Claim 1)

[2105] a means for receiving user-entered ideas;

[2106] means for converting the input ideas into a JSON format;

[2107] A means for transmitting the converted JSON format idea data to a server;

[2108] a means for storing the idea data received by the server;

[2109] A method for analyzing saved idea data using natural language processing to extract key topics and keywords;

[2110] A means of automatically generating multiple characters based on the analysis results,

[2111] A means to format the generated character information in JSON format and send it to the device,

[2112] A means for visually displaying character information to a user and receiving an instruction to start a discussion;

[2113] a means of transmitting user input and behavioral data to the emotion engine in real time;

[2114] means for the emotion engine to detect the user's emotional state;

[2115] a means for adjusting the content and tone of the character's speech based on the user's emotional state;

[2116] A means of utilizing a generative AI model to generate utterances appropriate for each character's role, and

[2117] The means by which the characters conduct their arguments;

[2118] means for analyzing the content of the discussion and drawing conclusions;

[2119] The system includes a means for presenting said conclusion to a user.

[2120] (Claim 2)

[2121] The system of claim 1 , wherein the characters have roles of moderator, bold, cautious, and balancer.

[2122] (Claim 3)

[2123] The system of claim 1 , wherein the character's role is user customizable.

[2124] "Application example 2 when combining emotion engines"

[2125] (Claim 1)

[2126] a means for receiving user-entered ideas;

[2127] means for analyzing the input idea and automatically generating a plurality of characters;

[2128] A means for the generated characters to hold discussions;

[2129] means for analyzing the content of the discussion and drawing conclusions;

[2130] means for presenting said conclusion to a user;

[2131] A means of analyzing customer input and behavioral data in real time,

[2132] The system includes means for adjusting the character's speech based on the emotional state.

[2133] (Claim 2)

[2134] The system of claim 1 , wherein the characters have roles of moderator, bold, cautious, and balancer.

[2135] (Claim 3)

[2136] The system of claim 1 , wherein the character's role is user customizable. [Explanation of symbols]

[2137] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for receiving user-entered ideas; means for analyzing the input idea and automatically generating a plurality of characters; A means for the generated characters to hold discussions; means for analyzing the content of the discussion and drawing conclusions; The system includes a means for presenting said conclusion to a user.

2. The system of claim 1 , wherein the characters have roles of a moderator, a bold person, a cautious person, and a balancer.

3. The system of claim 1 , wherein the character's role is user customizable.

Images