System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting expertise from a knowledge provider; means for storing the collected expertise in a database; means for training a generative model based on the expertise in the database; means for receiving a question from a user; means for generating an answer based on the question from the generative model; means for providing the generated answer to the user; and means for returning a portion of a consulting fee to the knowledge provider.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] While generative AI excels at summarizing conversations and information, it lacks specialized knowledge. This makes it difficult to provide appropriate advice and solutions for specific issues faced by companies. Furthermore, due to the retirement of experts and the obsolescence of knowledge, it is difficult to maintain up-to-date knowledge. [Means for solving the problem]

[0005] The present invention comprises a system that collects specialized knowledge from knowledge providers and stores it in a database. A generative model is trained based on the specialized knowledge in this database, and when a question from a user is received, an answer is generated from the generative model based on the question. The system also has a function of providing the generated answer to the user and returning a portion of the consulting fee to the knowledge provider. Such a system makes it possible to utilize specialized knowledge and provide highly accurate consulting. Furthermore, by returning rewards to the knowledge provider, it is possible to continuously provide knowledge and maintain the latest knowledge.

[0006] A "knowledge provider" is an individual or organization that has specialized knowledge and experience in a particular field and provides that knowledge to the system.

[0007] "Expertise" refers to in-depth knowledge and skills related to a particular field, and information that is useful for problem-solving and decision-making.

[0008] A "database" is a system for efficiently managing collected information and data, and making it easy to search and retrieve.

[0009] A "generative model" is an artificial intelligence algorithm or software used to learn from collected data and generate answers to questions.

[0010] A "user" is an individual or business that utilizes the system to enter questions and receive generated answers.

[0011] A "question" is a question or challenge that a user inputs into the system, and is the input data that the generative model uses to generate an answer.

[0012] An "answer" is information or advice generated by a generative model based on a user's question.

[0013] "Consulting fee" means the fee paid by a user for consulting services obtained by using the system.

[0014] "Return" refers to a system in which a portion of the consulting fee is returned to the knowledge provider to compensate them for their provision.

[0015] "Preprocessing" refers to the process of preparing data in a form that is easy for the generative model to understand, and includes methods such as tokenization and normalization.

[0016] "Extraction" is the operation of retrieving necessary data from a database based on specific conditions. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to companies and users. To this end, this invention has the following main functions: database construction, AI learning, query processing, and reward distribution.

[0039] Database construction

[0040] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, they enter "latest findings on marketing strategies." The server receives this entered data and performs an initial format check. Knowledge that passes the format check is registered in a database based on specific categories or tags (e.g., marketing, financial management, legal, etc.). The database is structured to prevent anomalies and duplicate data.

[0041] AI Learning

[0042] The server operates to train a generative model (such as GPT-3 (registered trademark)) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. Then, it inputs the preprocessed data into the generative model and trains it to answer questions appropriately.

[0043] Query Processing

[0044] A user (e.g., a company employee) uses their own device to input a question or problem into the system. For example, they might input a question like, "What is the optimal marketing strategy in the current market situation?" The server receives the user's question, analyzes it, and passes it to the generative model. The generative model then utilizes the knowledge accumulated through learning to generate an appropriate answer. The generated answer is then sent by the server to the user's device and provided to the user.

[0045] reward distribution

[0046] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0047] Specific examples

[0048] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0049] 2. The user (a corporate marketing person) inputs a question from their device: "How can we effectively promote a new product?" The server receives this question and passes it to the generative model. The generative model uses its accumulated knowledge of "effective methods for running social media ads" to propose a specific strategy, including "using social media ads to reach the target demographic." The server then provides this answer to the user.

[0050] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0051] This allows users to receive prompt and highly accurate consulting, and knowledge providers can also receive rewards.

[0052] The processing flow will be explained below.

[0053] Step 1:

[0054] Users (knowledge providers) use a dedicated web form or application to input their expertise, for example, "the latest knowledge on marketing strategies."

[0055] Step 2:

[0056] The server receives the input expert knowledge data and performs a format check, which includes checking the format of the data and basic error checks.

[0057] Step 3:

[0058] The server registers the knowledge data that has passed the format check into a database based on specific categories or tags (e.g., marketing, legal, financial management, etc.).

[0059] Step 4:

[0060] The server extracts purpose-specific knowledge data from the database for training the generative model.

[0061] Step 5:

[0062] The server performs preprocessing on the extracted knowledge data, including tokenization, text normalization, and feature engineering.

[0063] Step 6:

[0064] The server trains a generative model on the preprocessed knowledge data, which the generative model uses to improve its ability to generate appropriate answers to questions.

[0065] Step 7:

[0066] A user (a company employee) uses a terminal to input a question or problem into the system. For example, the user inputs a question such as, "What is the optimal marketing strategy in the current market situation?"

[0067] Step 8:

[0068] The server receives the question data sent by the user and analyzes the question, using natural language processing techniques to understand the intent of the question.

[0069] Step 9:

[0070] Based on the analysis results, the server passes the question data to the generative model, which then uses its learned knowledge to generate an answer to the question.

[0071] Step 10:

[0072] The server receives the answers returned by the generative model and reviews them, optionally checking their clarity and accuracy.

[0073] Step 11:

[0074] The device displays the answers sent from the server to the user, who can then make decisions and take action based on the generated answers.

[0075] Step 12:

[0076] The server aggregates transaction data for the consulting session, including tracking what knowledge was used.

[0077] Step 13:

[0078] The server calculates the compensation for the knowledge provider based on the collected data, which is calculated as part of the consulting fee.

[0079] Step 14:

[0080] The server returns the calculated reward to the knowledge provider, who can receive the reward in return for utilizing their knowledge.

[0081] Example 1

[0082] 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."

[0083] Conventional consulting systems have insufficient management of collected expertise and data organization, making it difficult to generate highly accurate answers. Furthermore, the distribution of rewards to knowledge providers is often vague and unfair. This has led to issues such as a decline in user satisfaction and the motivation of knowledge providers.

[0084] 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.

[0085] In this invention, the server includes means for collecting expertise from knowledge providers, means for storing the collected expertise in a database, means for performing preprocessing including tokenization and text normalization on the expertise in the database, means for training a generative model based on the preprocessed expertise, means for receiving questions from users, means for generating answers from the generative model based on the questions, means for providing the generated answers to the users, and means for returning a portion of the consulting fee to the knowledge providers. This improves the accuracy and management of the collected expertise, making it possible to provide accurate answers, and also realizes fair and transparent reward distribution, thereby improving the satisfaction of both users and knowledge providers.

[0086] A "knowledge provider" is an individual or organization that has specialized knowledge and is responsible for providing that knowledge to the system.

[0087] "Specialized knowledge" is information that requires advanced knowledge and skills in a specific field.

[0088] A "database" is a digital information storage system for systematically storing and managing collected specialized knowledge based on categories and tags.

[0089] "Tokenization" is the process of dividing text data into units of words or sentences.

[0090] "Text normalization" is the process of removing noise and unnecessary information from text data and converting the data into a unified format.

[0091] A "generative model" is an artificial intelligence model that learns from collected and preprocessed expert knowledge and generates answers to user questions.

[0092] A "user" is an individual or entity that uses the system to enter questions or challenges and receive generated answers.

[0093] A "consulting fee" is a fee paid by a user to obtain answers to their questions or problems.

[0094] "Reward sharing" is the process of returning a portion of the consulting fee to the knowledge provider as compensation for the expertise provided.

[0095] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to users. The system has the following main functions: database construction, AI learning, query processing, and reward distribution. Each function will be explained in detail below.

[0096] Database construction

[0097] Knowledge providers use their own devices to access a dedicated web form or application and enter their expertise. For example, they enter "latest knowledge on marketing strategies." When the server receives this input data, it first performs a format check to ensure that the data conforms to the expected format. Data that passes the format check is assigned a specific category (e.g., marketing, financial management, legal, etc.) or tag and stored in a database.

[0098] AI Learning

[0099] The server performs preprocessing, including tokenization and text normalization, on the expert knowledge stored in the database. The preprocessed data is used to train a generative AI model (e.g., GPT-3). The server feeds the data to the generative model in batches, and improves the model's accuracy by tuning hyperparameters and evaluating them every epoch.

[0100] Query Processing

[0101] Users use their own devices to access the system's interface and input questions or issues. For example, they might input a question like, "What is the optimal marketing strategy given the current market situation?" The server receives this question, performs preprocessing such as grammatical analysis and keyword extraction, and then passes the question to the generative model. The generative model generates an appropriate answer based on the learned knowledge. The generated answer is then sent to the user's device via the server.

[0102] reward distribution

[0103] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates a reward based on that data. The reward is calculated taking into account factors such as frequency of use and the weight of the data. Finally, the calculated reward is paid to the knowledge provider.

[0104] Specific examples

[0105] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the marketing category.

[0106] 2. A user (e.g., a corporate marketing manager) inputs a question from their device, such as "How can we effectively promote a new product?" The server receives this question, performs the necessary preprocessing, and passes the question to the generative model. The generative model utilizes its accumulated knowledge of "effective methods for running social media advertising" to propose a specific strategy, including "using social media advertising to reach the target demographic." The server then provides this answer to the user.

[0107] 3. The server aggregates the data from this consulting session and calculates the reward for the knowledge provider used. The calculated reward is paid to the knowledge provider.

[0108] Prompt Sentence Examples

[0109] How to effectively promote new products?

[0110] This allows users to receive prompt and highly accurate consulting, and knowledge providers can also receive rewards.

[0111] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0112] Step 1:

[0113] A knowledge provider uses their own device to input their expertise. For example, they enter "latest findings on marketing strategies" into a web form. The server receives this input data and performs a format check. This format check verifies whether the input data conforms to the expected format. Data that passes the format check is assigned specific categories and tags and registered in the database. The input is the knowledge provider's expertise data, and the output is the format-checked data registered in the database.

[0114] Step 2:

[0115] The server periodically scans the expertise data registered in the database and performs preprocessing for model training. Specifically, it divides the text data into words and sentences through tokenization, and performs text normalization to remove noise and unnecessary information. It also detects and removes duplicate data. The input is the expertise data in the database, and the output is the preprocessed data.

[0116] Step 3:

[0117] The server trains a generative AI model (e.g., GPT-3) using the preprocessed data. Data is fed to the model in batches. The server adjusts hyperparameters and evaluates each epoch to improve the model's accuracy. The input is the preprocessed data, and the output is a trained generative AI model.

[0118] Step 4:

[0119] A user uses their own terminal to access the system's interface and input a question or problem. For example, they input a question such as, "What is the optimal marketing strategy in the current market situation?" The server receives this query and performs preprocessing such as grammar analysis and keyword extraction. The input is the user's question data, and the output is the preprocessed question data.

[0120] Step 5:

[0121] The server passes the preprocessed question data to the generative AI model, which generates an appropriate answer based on the knowledge it has learned so far. The input is the preprocessed question data, and the output is the generated answer.

[0122] Step 6:

[0123] The server receives the answer returned by the generative model and sends it to the user's terminal. The user can receive the answer on the terminal. The input is the generated answer, and the output is the delivery of the answer to the user terminal.

[0124] Step 7:

[0125] The server tallies the consulting fees paid by users for each transaction and identifies which knowledge provider's data was used. Based on that data, it calculates the reward for the knowledge provider. The reward is calculated taking into account factors such as frequency of use and the weight of the data. The final calculated reward is paid to the knowledge provider. The input is the consulting fee and the knowledge data used, and the output is the reward for the knowledge provider.

[0126] (Application example 1)

[0127] 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."

[0128] Conventional navigation systems for autonomous vehicles do not adequately propose optimal routes that take real-time traffic and weather information into account, preventing improvements in operational efficiency. Furthermore, they do not provide accurate route proposals that utilize specialized knowledge, leaving issues with operational accuracy and safety. This creates challenges for the efficient and safe operation of autonomous vehicles.

[0129] 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.

[0130] In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for receiving questions from users through a user interface, means for generating answers from the generative model based on the questions, means for acquiring additional data including real-time traffic and weather information when generating answers and adding the additional data to prompt sentences, means for providing the generated answers to users, and means for returning a portion of the consulting fee to the knowledge provider. This enables optimal route proposals based on real-time information for autonomous vehicles, thereby improving operational efficiency and safety.

[0131] A "knowledge provider" is a person or organization that provides specialized knowledge and know-how to the system.

[0132] "Expertise" is deep understanding and technical information in a particular field.

[0133] A "database" is a storage device or data set for efficiently storing and managing collected specialized knowledge.

[0134] A "generative model" is a machine learning algorithm that learns from collected data and generates appropriate answers to user questions.

[0135] A "user interface" is an interface or screen through which a user exchanges information with a system.

[0136] A "prompt sentence" is input data or an instruction sentence provided to a generative model, and is a sentence necessary for the model to generate an appropriate answer.

[0137] "Traffic information" refers to data such as road congestion and accident information collected in real time.

[0138] "Weather information" is data about weather and climate collected in real time.

[0139] An "answer" is a solution or suggestion provided by a generative model in response to a user's question.

[0140] "Consulting fee" is the fee paid by the user for use of the system.

[0141] "Return" means distributing a portion of the revenue gained from the use of the system to the knowledge provider.

[0142] A system for carrying out the present invention includes the following means.

[0143] Knowledge data collection and database construction

[0144] The server first collects expertise from knowledge providers. Expertise is entered through dedicated web forms or applications. This collected data undergoes format checks and is then registered in a database. The database is optimized to prevent anomalies and duplicate data.

[0145] Training a generative model

[0146] The server trains a generative AI model based on the specialized knowledge stored in the database, including preprocessing the data (such as tokenization and text normalization). After training, the generative model can use this accumulated knowledge to generate appropriate answers to questions.

[0147] Query Processing

[0148] The user sends a question to the system through their device. For example, a question might be, "What is the best route to take, taking into account the current traffic and weather conditions?" The server receives the question and retrieves real-time traffic and weather information. Using this additional data, the server generates a prompt sentence and inputs it into the generative model.

[0149] Generate and provide answers

[0150] The generative model utilizes real-time information and specialized knowledge stored in a database to generate optimal answers to user questions, which are then provided to the user via a server.

[0151] reward distribution

[0152] The server calculates the amount of the consulting fee to be returned to the knowledge provider. The server manages the distribution of the reward appropriately based on the knowledge data used.

[0153] Hardware and software used

[0154] Server: Collects and manages expertise, trains generative models, processes queries, generates and serves answers, and manages reward distribution.

[0155] Databases: Used to efficiently store and manage specialized knowledge.

[0156] Generative AI model: A machine learning algorithm for generating appropriate answers to questions, specifically GPT-3 and higher models (e.g., GPT-3.5-turbo).

[0157] External APIs: Used to obtain real-time traffic and weather information. Examples include "Traffic Information API" and "Weather Information API."

[0158] User Interface: An interface for receiving expertise input and user questions.

[0159] Specific examples

[0160] For example, the following prompt statement is generated:

[0161] Available knowledge:

[0162] Best practices for efficient route planning

[0163] Traffic Information:

[0164] Current traffic conditions: Traffic jams, accidents Road closures: None

[0165] Weather information:

[0166] Local weather: Sunny Forecast: Light rain overnight

[0167] Use this information to suggest the best route.

[0168] Using this prompt, the generative model suggests an appropriate route, which is then provided to the user, thereby improving the operational efficiency and safety of autonomous vehicles.

[0169] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0170] Step 1:

[0171] The server collects specialized knowledge from knowledge providers. Knowledge providers enter their specialized knowledge and know-how using dedicated web forms or applications. The input data is checked for format by the server and registered in a database. The input is via a dedicated form, and the output is format-checked specialized knowledge data.

[0172] Step 2:

[0173] The server trains a generative AI model based on the expertise registered in the database. It extracts the necessary expertise data from the database and performs preprocessing such as tokenization and text normalization. It then inputs the preprocessed data into a generative model (e.g., GPT-3.5-turbo) and trains the model. The input is data from the expertise database, and the output is a trained generative model.

[0174] Step 3:

[0175] A user inputs a question into the system through a terminal. For example, a question might be, "What is the best route to take, taking into account the current traffic conditions and weather?" The question is sent to the server via the user interface. The input is the question from the user, and the output is the received question data.

[0176] Step 4:

[0177] After receiving the query, the server retrieves real-time traffic and weather information. It uses external APIs to retrieve the latest data from the traffic information API and weather information API. The input is the user's query, and the output is traffic and weather information data.

[0178] Step 5:

[0179] The server generates prompt sentences based on the user's question and the acquired real-time information. The generated prompt sentences are in the format required for input to the generative model. The inputs are the user's question, traffic information, and weather information, and the output is the generated prompt sentence.

[0180] Step 6:

[0181] The server inputs the generated prompt sentence into the generative model to generate an answer. The generative model utilizes a database of expert knowledge and real-time information to generate the optimal answer. The input is the prompt sentence, and the output is the generated answer.

[0182] Step 7:

[0183] The server provides the generated answer to the user, and sends the answer to the user's terminal for display. The input is the generated answer, and the output is the display of the answer on the user's terminal.

[0184] Step 8:

[0185] The server performs calculations to return a portion of the consulting fee to the knowledge provider. It distributes the appropriate reward based on the knowledge data used. The input is the system usage data and the consulting fee, and the output is the reward return to the knowledge provider.

[0186] The above steps will enable the system to propose optimal routes for autonomous vehicles based on real-time information, improving operational efficiency and safety.

[0187] 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.

[0188] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to companies and users. Furthermore, this invention also includes a function to improve the user experience by combining it with an emotion engine that recognizes user emotions. The main functions of this system are database construction, AI learning, emotion analysis, query processing, and reward distribution.

[0189] Database construction

[0190] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, they enter "latest findings on marketing strategies." The server receives this data and performs an initial format check. Knowledge that passes the format check is registered in a database based on specific categories and tags (e.g., marketing, financial management, law, etc.).

[0191] AI Learning

[0192] The server operates to train a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. Then it inputs the preprocessed data into the generative model, and trains the model to answer questions appropriately.

[0193] Emotion analysis

[0194] When a user (company representative) uses a device to input a question or issue, the emotion engine captures and analyzes the user's emotional state. For example, it analyzes emotions (joy, anger, sadness, surprise, etc.) from the user's input text. Based on this analysis, the server deeply understands the context of the question and passes it to the appropriate generative model.

[0195] Query Processing

[0196] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The generated answer is sent by the server to the user's device and provided to the user.

[0197] reward distribution

[0198] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0199] Specific examples

[0200] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0201] 2. A user (a corporate marketing person) inputs the question "How can we effectively promote a new product?" into a device. The emotion engine analyzes the emotion from the user's input and detects that the user is "impatient." The server passes this information to a generative model, which generates an answer that takes into account the user's emotional state: "Use social media advertising to reach the target demographic." The device then provides this answer to the user.

[0202] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0203] This allows users to receive prompt and accurate consultation, and knowledge providers can earn rewards. Furthermore, by providing answers that take into account the user's emotional state, the user experience is improved.

[0204] The processing flow will be explained below.

[0205] Step 1:

[0206] Users (knowledge providers) use a dedicated web form or application to input their own expertise and know-how, for example, "the latest knowledge on marketing strategies."

[0207] Step 2:

[0208] The server receives the input expert knowledge data and performs a format check, which includes checking the format of the data and basic error checks.

[0209] Step 3:

[0210] The server registers the knowledge data that has passed the format check into a database based on specific categories or tags (e.g., marketing, legal, financial management, etc.).

[0211] Step 4:

[0212] The server extracts purpose-specific knowledge data from the database for training the generative model.

[0213] Step 5:

[0214] The server performs preprocessing on the extracted knowledge data, including tokenization, text normalization, and feature engineering.

[0215] Step 6:

[0216] The server trains a generative model on the preprocessed knowledge data, which the generative model uses to improve its ability to generate appropriate answers to questions.

[0217] Step 7:

[0218] A user (a company employee) uses a terminal to input a question or problem into the system. For example, the user inputs a question such as, "What is the optimal marketing strategy in the current market situation?"

[0219] Step 8:

[0220] The emotion engine captures and analyzes emotions from the user's input sentences, detecting, for example, the user's emotional state, such as "anxious" or "worried."

[0221] Step 9:

[0222] The server receives the question data sent by the user and analyzes the question, taking into account the results of the emotion engine.

[0223] Step 10:

[0224] Based on the analysis results, the server passes the question data to the generative model, which then uses its learned knowledge to generate an appropriate answer in a tone that reflects the user's emotional state.

[0225] Step 11:

[0226] The server reviews the generated answers to ensure clarity and accuracy of the content.

[0227] Step 12:

[0228] The device displays the answers sent from the server to the user, who can then make decisions and take action based on the generated answers.

[0229] Step 13:

[0230] The server aggregates transaction data for the consulting session, including tracking what knowledge was used.

[0231] Step 14:

[0232] The server calculates the compensation for the knowledge provider based on the collected data, which is calculated as part of the consulting fee.

[0233] Step 15:

[0234] The server returns the calculated reward to the knowledge provider, who can receive the reward in return for utilizing their knowledge.

[0235] Example 2

[0236] 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."

[0237] In conventional consulting systems, it has been difficult to efficiently manage the expertise collected from knowledge providers and create generative models based on that data. It has also been difficult to generate appropriate answers that take the user's emotional state into account, resulting in insufficient improvement in the user experience. Furthermore, the distribution of rewards to knowledge providers has also lacked transparency and appropriateness. The present invention aims to solve these problems and provide a consulting system with higher accuracy and a superior user experience.

[0238] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for analyzing the emotional state of a user, means for generating an answer from the generative model based on a question from the user, means for providing the generated answer to the user, and means for returning a portion of the consulting fee to the knowledge provider. This makes it possible to generate highly accurate answers that take the emotional state of the user into consideration, improve the user experience, and appropriately distribute rewards to knowledge providers.

[0239] A "knowledge provider" is an individual or organization that has specialized knowledge or know-how and provides that information to the system.

[0240] "Specialized knowledge" is information related to advanced knowledge and skills in a particular field.

[0241] A "database" is an information system for systematically storing and managing collected specialized knowledge and information.

[0242] A "generative model" is an algorithm or system that learns from collected data and generates new information or answers.

[0243] "Preprocessing" refers to processes such as text normalization and tokenization that are performed on data before it is trained by a generative model.

[0244] "Means for analyzing emotional state" refers to a technology or mechanism for analyzing emotions from user input data and identifying that state.

[0245] "Means for receiving questions" refers to the method or device by which users input questions or issues to the consulting system.

[0246] A "means for generating an answer" is a method or system that uses a generative model to create an appropriate answer to a question from a user.

[0247] "Consulting Fee" means the fee paid by the User when using the Consulting Service.

[0248] "Means of return" refers to a method or system for distributing a portion of the consulting fee to the knowledge provider.

[0249] This invention is a system that collects specialized knowledge from knowledge providers, stores it in a database, trains a generative model, and provides highly accurate consulting to users. Furthermore, it aims to improve the user experience by analyzing the user's emotional state and providing optimal answers. The main functions of this invention are data collection from knowledge providers, database construction, AI learning, sentiment analysis, query processing, and reward distribution.

[0250] 1. Database construction

[0251] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, a marketing expert might enter "effective methods for running social media ads." The server receives this data and performs a format check. Knowledge that passes the format check is registered in a database based on specific categories and tags (e.g., marketing, financial management, law, etc.).

[0252] 2. AI Learning

[0253] The server trains a generative model (such as GPT-3) based on the collected knowledge data. It extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. It then inputs the preprocessed data into the generative model, training it to answer questions appropriately.

[0254] 3. Emotion analysis

[0255] When a user uses a device to input a question or problem, the emotion engine captures and analyzes the user's emotional state. For example, it analyzes emotions (joy, anger, sadness, surprise, etc.) from the user's input text. Based on this analysis, the server deeply understands the context of the question and passes it to the appropriate generative model.

[0256] 4. Query Processing

[0257] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The generated answer is sent by the server to the user's device and provided to the user.

[0258] 5. Reward distribution

[0259] The server tallies the consulting fees paid by users for each transaction. It identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0260] Specific examples

[0261] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0262] 2. A user (a corporate marketing person) inputs a question into a device, such as "How can we effectively promote a new product?" The emotion engine analyzes the emotion in the user's input and detects that the user is "impatient." The server passes this information to a generative model, which generates an answer that takes into account the user's emotional state (e.g., "Use social media advertising to reach the target demographic"). The device then provides this answer to the user.

[0263] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0264] This allows users to receive prompt and accurate consultation, and knowledge providers can earn rewards. It also significantly improves the user experience by providing answers that take into account the user's emotional state. This system offers a new way to quickly respond to the diverse needs of modern businesses and utilize expertise more efficiently.

[0265] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0266] System program processing flow

[0267] Step 1: Gather knowledge

[0268] Input: Expertise and know-how from knowledge providers

[0269] Output: Format-checked data transferred to the server

[0270] Specific behavior:

[0271] 1. Knowledge providers use a dedicated web form or application to enter their expertise and know-how (e.g., how to effectively run social media ads).

[0272] 2. The server receives this input data.

[0273] 3. The server performs a format check on the data, which includes checking for required fields and validating the data format.

[0274] 4. The server sends the data, which has passed the format check, to the database.

[0275] Step 2: Registering with the database

[0276] Input: Format-checked expert knowledge data

[0277] Output: Categorized and tagged database entries

[0278] Specific behavior:

[0279] 1. The server parses the received, format-checked data and assigns it appropriate categories and tags (e.g., marketing, financial management, legal, etc.).

[0280] 2. The server stores the categories and tagged data in a database.

[0281] Step 3: Data extraction and preprocessing

[0282] Input: Expert knowledge data in the database

[0283] Output: Preprocessed data

[0284] Specific behavior:

[0285] 1. The server extracts data from the database to use in the generative model (e.g., GPT-3).

[0286] 2. The server performs preprocessing on the extracted data (tokenization, text normalization, etc.) to make the data suitable for the generative model.

[0287] Step 4: Training the generative model

[0288] Input: Preprocessed data

[0289] Output: A trained generative model

[0290] Specific behavior:

[0291] 1. The server feeds the preprocessed data into the generative model.

[0292] 2. The server trains the generative model, improving the model's ability to generate appropriate answers to user questions.

[0293] Step 5: User sentiment analysis

[0294] Input: User questions and challenges

[0295] Output: Emotion analysis results

[0296] Specific behavior:

[0297] 1. A user uses a device to enter a question or challenge (e.g., "How can we effectively promote a new product?").

[0298] 2. The device sends the input data to the server.

[0299] 3. The server uses an emotion engine to analyze emotions from the user's input data (e.g., to detect joy, anger, sadness, surprise, etc.).

[0300] Step 6: Parsing the question and generating an answer

[0301] Input: User question and sentiment analysis results

[0302] Output: Answers that take into account the user's emotional state

[0303] Specific behavior:

[0304] 1. The server analyzes the user's question and takes into account the results of sentiment analysis.

[0305] 2. The server passes the question to the generative model in the appropriate context.

[0306] 3. The generative model leverages its trained knowledge to generate appropriate answers in a tone that reflects the user's emotional state.

[0307] 4. The server sends the generated answer to the user's device.

[0308] Step 7: Provide your answers

[0309] Input: Answer generation result

[0310] Output: The answer provided to the user

[0311] Specific behavior:

[0312] 1. The server generates a response and sends it to the user's device.

[0313] 2. The device displays and provides the answer to the user.

[0314] Step 8: Aggregating and distributing rewards

[0315] Input: User consulting fees and knowledge data used

[0316] Output: Reward returned to knowledge provider

[0317] Specific behavior:

[0318] 1. The server tallies the consulting fees paid by the user for each transaction.

[0319] 2. The server identifies which knowledge provider's data was used.

[0320] 3. The server calculates and returns the reward to the knowledge provider based on this.

[0321] The above is a specific processing flow and operation of the system according to the present invention.

[0322] (Application example 2)

[0323] 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."

[0324] In customer service operations in brick-and-mortar stores, there is a demand for methods to quickly and accurately answer customer questions. Furthermore, it is expected that customer satisfaction will be improved by providing responses in an appropriate tone based on the customer's emotional state. Current customer service systems have difficulty in providing such advanced interactions, and there are many situations where they are forced to respond with limited information and knowledge. As a result, customer dissatisfaction is likely to occur, affecting the quality of customer service.

[0325] 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.

[0326] In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for receiving questions from users, means for generating answers from the generative model based on the questions, means for providing the generated answers to the users, means for analyzing the emotional state of the user at the time of asking the question, means for generating answers in a tone corresponding to the emotional state, means for displaying the generated answer information on a visual display device, and means for returning a portion of the consulting fee to the knowledge provider. This makes it possible to quickly provide appropriate answers to customer questions according to the emotional state when serving customers in a physical store, thereby improving customer satisfaction.

[0327] A "knowledge provider" is someone who provides specialized knowledge and know-how.

[0328] "Expertise" refers to advanced knowledge and experience in a particular field.

[0329] A "database" is a system that systematically stores information and allows it to be easily searched and updated.

[0330] A "generative model" is an artificial intelligence model that learns from collected data and generates new information and answers.

[0331] "User" means a person who uses the system to ask questions or seek consultation.

[0332] A "question" is a question that a user asks the system for information or advice.

[0333] An "answer" is information or advice that the system provides in response to a question.

[0334] "Emotional state" refers to the emotions such as joy, anger, sadness, surprise, etc. that the user is feeling at the time of asking the question.

[0335] A "visual display device" is a device for visually presenting information to a user.

[0336] "Consulting Fee" means the fee paid by User for Consulting Services.

[0337] "Return" refers to returning a portion of the consulting fee to the knowledge provider as compensation.

[0338] The system to realize this application example mainly requires a server, a user terminal, and a visual display device (smart glasses). The specific system configuration and processing method are described below.

[0339] System configuration and processing method

[0340] 1. Gathering expertise from knowledge providers

[0341] Knowledge providers provide their expertise and know-how using a dedicated web form or application. For example, they enter "latest insights on marketing strategies." The server receives this data, performs a format check, and then registers it in a database. This database is organized according to specific categories or tags (e.g., marketing, financial management, legal, etc.).

[0342] 2. Training the generative model

[0343] The server trains a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. The preprocessed data is then input into the generative model, and the model is trained to be able to answer various questions appropriately.

[0344] 3. Receiving user questions and analyzing emotional states

[0345] When a user (company representative) uses a device to input a question or issue, the emotion engine captures and analyzes the user's emotional state. For example, if a user asks, "I'd like to know more about this product," the emotion (joy, anger, sadness, surprise, etc.) is analyzed from the input sentence. Based on the results of this analysis, the server gains a deep understanding of the context of the question and passes it to the generative model.

[0346] 4. Generating answers based on questions

[0347] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. For example, if the user is anxious, an answer in a "relaxing tone" is generated. The generated answer is sent by the server to the user's smart glasses and displayed visually.

[0348] 5. Consulting fee aggregation and reward returns

[0349] The server aggregates the data from the consulting session and returns a certain amount of compensation to the knowledge provider. It identifies which knowledge provider's data was used, calculates the compensation based on that usage, and pays it to the knowledge provider.

[0350] Hardware and software used

[0351] Hardware:

[0352] Smart glasses (e.g., Google® Glass®, Vuzix)

[0353] software:

[0354] OpenAI(R) API: Use of generative AI model (GPT-3)

[0355] TextBlob: A Python library for sentiment analysis

[0356] Specific processing examples

[0357] For example, if a user asks about the features of a new smartphone:

[0358] User asks: "What are the features of this new smartphone?"

[0359] Example prompt sentence:

[0360] Customer asks: "What are the features of this new smartphone?" Provide detailed information and suggestions if needed. Product info: "Product Info for new smartphone: High quality, cost-effective, and trendy."

[0361] The server analyzes emotions and provides appropriate answers based on the user's questions and knowledge database, and displays the information visually through smart glasses to support customer service in brick-and-mortar stores, thereby improving customer satisfaction.

[0362] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0363] Step 1:

[0364] The server collects expertise from knowledge providers. Knowledge providers use a dedicated web form or application to input their expertise and know-how. For example, they input "latest findings on marketing strategies." The server receives this data and performs a format check. Data that passes the format check is registered in the database. The input is the knowledge provider's expertise, and the output is important information stored in the database.

[0365] Step 2:

[0366] The server trains a generative model based on the specialized knowledge in the database. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. The preprocessed data is input into the generative model, and the model is trained to be able to answer various questions appropriately. The input is the knowledge data in the database, and the output is the trained generative model.

[0367] Step 3:

[0368] The user inputs a question or problem using a terminal. For example, the user might ask, "I'd like to know more about this product." The server receives this question and simultaneously analyzes the user's emotional state using an emotion engine. The input is the user's question and emotional state, and the output is the analysis results.

[0369] Step 4:

[0370] The server analyzes the question sent by the user and passes the analysis results from the emotion engine to the generative model. The generative model utilizes knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The input is the user's question and emotional state, and the output is the generated answer.

[0371] Step 5:

[0372] The server transmits the generated answer to the user's device. In particular, when smart glasses are used, the answer information is visually displayed. For example, when the user is wearing smart glasses, the generated answer is displayed on the display of the glasses. The input is the generated answer, and the output is the answer displayed on the user's visual display device.

[0373] Step 6:

[0374] The server returns a portion of the consulting fee to the knowledge provider. The server aggregates the data from the consulting session and identifies which knowledge provider's data was used. It then calculates a reward based on the knowledge used and pays it to the knowledge provider. The input is the consulting session data, and the output is the reward payment to the knowledge provider.

[0375] 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.

[0376] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0377] 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.

[0378] [Second embodiment]

[0379] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0380] 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.

[0381] 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).

[0382] 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.

[0383] 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.

[0384] 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).

[0385] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0386] 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.

[0387] 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.

[0388] 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.

[0389] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0390] 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."

[0391] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to companies and users. To this end, this invention has the following main functions: database construction, AI learning, query processing, and reward distribution.

[0392] Database construction

[0393] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, they enter "latest findings on marketing strategies." The server receives this entered data and performs an initial format check. Knowledge that passes the format check is registered in a database based on specific categories or tags (e.g., marketing, financial management, legal, etc.). The database is structured to prevent anomalies and duplicate data.

[0394] AI Learning

[0395] The server operates to train a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. Then it inputs the preprocessed data into the generative model, and trains the model to answer questions appropriately.

[0396] Query Processing

[0397] A user (e.g., a company employee) uses their own device to input a question or problem into the system. For example, they might input a question like, "What is the optimal marketing strategy in the current market situation?" The server receives the user's question, analyzes it, and passes it to the generative model. The generative model then utilizes the knowledge accumulated through learning to generate an appropriate answer. The generated answer is then sent by the server to the user's device and provided to the user.

[0398] reward distribution

[0399] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0400] Specific examples

[0401] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0402] 2. The user (a corporate marketing person) inputs a question from their device: "How can we effectively promote a new product?" The server receives this question and passes it to the generative model. The generative model uses its accumulated knowledge of "effective methods for running social media ads" to propose a specific strategy, including "using social media ads to reach the target demographic." The server then provides this answer to the user.

[0403] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0404] This allows users to receive prompt and highly accurate consulting, and knowledge providers can also receive rewards.

[0405] The processing flow will be explained below.

[0406] Step 1:

[0407] Users (knowledge providers) use a dedicated web form or application to input their expertise, for example, "the latest knowledge on marketing strategies."

[0408] Step 2:

[0409] The server receives the input expert knowledge data and performs a format check, which includes checking the format of the data and basic error checks.

[0410] Step 3:

[0411] The server registers the knowledge data that has passed the format check into a database based on specific categories or tags (e.g., marketing, legal, financial management, etc.).

[0412] Step 4:

[0413] The server extracts purpose-specific knowledge data from the database for training the generative model.

[0414] Step 5:

[0415] The server performs preprocessing on the extracted knowledge data, including tokenization, text normalization, and feature engineering.

[0416] Step 6:

[0417] The server trains a generative model on the preprocessed knowledge data, which the generative model uses to improve its ability to generate appropriate answers to questions.

[0418] Step 7:

[0419] A user (a company employee) uses a terminal to input a question or problem into the system. For example, the user inputs a question such as, "What is the optimal marketing strategy in the current market situation?"

[0420] Step 8:

[0421] The server receives the question data sent by the user and analyzes the question, using natural language processing techniques to understand the intent of the question.

[0422] Step 9:

[0423] Based on the analysis results, the server passes the question data to the generative model, which then uses its learned knowledge to generate an answer to the question.

[0424] Step 10:

[0425] The server receives the answers returned by the generative model and reviews them, optionally checking their clarity and accuracy.

[0426] Step 11:

[0427] The device displays the answers sent from the server to the user, who can then make decisions and take action based on the generated answers.

[0428] Step 12:

[0429] The server aggregates transaction data for the consulting session, including tracking what knowledge was used.

[0430] Step 13:

[0431] The server calculates the compensation for the knowledge provider based on the collected data, which is calculated as part of the consulting fee.

[0432] Step 14:

[0433] The server returns the calculated reward to the knowledge provider, who can receive the reward in return for utilizing their knowledge.

[0434] Example 1

[0435] 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."

[0436] Conventional consulting systems have insufficient management of collected expertise and data organization, making it difficult to generate highly accurate answers. Furthermore, the distribution of rewards to knowledge providers is often vague and unfair. This has led to issues such as a decline in user satisfaction and the motivation of knowledge providers.

[0437] 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.

[0438] In this invention, the server includes means for collecting expertise from knowledge providers, means for storing the collected expertise in a database, means for performing preprocessing including tokenization and text normalization on the expertise in the database, means for training a generative model based on the preprocessed expertise, means for receiving questions from users, means for generating answers from the generative model based on the questions, means for providing the generated answers to the users, and means for returning a portion of the consulting fee to the knowledge providers. This improves the accuracy and management of the collected expertise, making it possible to provide accurate answers, and also realizes fair and transparent reward distribution, thereby improving the satisfaction of both users and knowledge providers.

[0439] A "knowledge provider" is an individual or organization that has specialized knowledge and is responsible for providing that knowledge to the system.

[0440] "Specialized knowledge" is information that requires advanced knowledge and skills in a specific field.

[0441] A "database" is a digital information storage system for systematically storing and managing collected specialized knowledge based on categories and tags.

[0442] "Tokenization" is the process of dividing text data into units of words or sentences.

[0443] "Text normalization" is the process of removing noise and unnecessary information from text data and converting the data into a unified format.

[0444] A "generative model" is an artificial intelligence model that learns from collected and preprocessed expert knowledge and generates answers to user questions.

[0445] A "user" is an individual or entity that uses the system to enter questions or challenges and receive generated answers.

[0446] A "consulting fee" is a fee paid by a user to obtain answers to their questions or problems.

[0447] "Reward sharing" is the process of returning a portion of the consulting fee to the knowledge provider as compensation for the expertise provided.

[0448] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to users. The system has the following main functions: database construction, AI learning, query processing, and reward distribution. Each function will be explained in detail below.

[0449] Database construction

[0450] Knowledge providers use their own devices to access a dedicated web form or application and enter their expertise. For example, they enter "latest knowledge on marketing strategies." When the server receives this input data, it first performs a format check to ensure that the data conforms to the expected format. Data that passes the format check is assigned a specific category (e.g., marketing, financial management, legal, etc.) or tag and stored in a database.

[0451] AI Learning

[0452] The server performs preprocessing, including tokenization and text normalization, on the expert knowledge stored in the database. The preprocessed data is used to train a generative AI model (e.g., GPT-3). The server feeds the data to the generative model in batches, and improves the model's accuracy by tuning hyperparameters and evaluating them every epoch.

[0453] Query Processing

[0454] Users use their own devices to access the system's interface and input questions or issues. For example, they might input a question like, "What is the optimal marketing strategy given the current market situation?" The server receives this question, performs preprocessing such as grammatical analysis and keyword extraction, and then passes the question to the generative model. The generative model generates an appropriate answer based on the learned knowledge. The generated answer is then sent to the user's device via the server.

[0455] reward distribution

[0456] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates a reward based on that data. The reward is calculated taking into account factors such as frequency of use and the weight of the data. Finally, the calculated reward is paid to the knowledge provider.

[0457] Specific examples

[0458] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the marketing category.

[0459] 2. A user (e.g., a corporate marketing manager) inputs a question from their device, such as "How can we effectively promote a new product?" The server receives this question, performs the necessary preprocessing, and passes the question to the generative model. The generative model utilizes its accumulated knowledge of "effective methods for running social media advertising" to propose a specific strategy, including "using social media advertising to reach the target demographic." The server then provides this answer to the user.

[0460] 3. The server aggregates the data from this consulting session and calculates the reward for the knowledge provider used. The calculated reward is paid to the knowledge provider.

[0461] Prompt Sentence Examples

[0462] How to effectively promote new products?

[0463] This allows users to receive prompt and highly accurate consulting, and knowledge providers can also receive rewards.

[0464] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0465] Step 1:

[0466] A knowledge provider uses their own device to input their expertise. For example, they enter "latest findings on marketing strategies" into a web form. The server receives this input data and performs a format check. This format check verifies whether the input data conforms to the expected format. Data that passes the format check is assigned specific categories and tags and registered in the database. The input is the knowledge provider's expertise data, and the output is the format-checked data registered in the database.

[0467] Step 2:

[0468] The server periodically scans the expertise data registered in the database and performs preprocessing for model training. Specifically, it divides the text data into words and sentences through tokenization, and performs text normalization to remove noise and unnecessary information. It also detects and removes duplicate data. The input is the expertise data in the database, and the output is the preprocessed data.

[0469] Step 3:

[0470] The server trains a generative AI model (e.g., GPT-3) using the preprocessed data. Data is fed to the model in batches. The server adjusts hyperparameters and evaluates each epoch to improve the model's accuracy. The input is the preprocessed data, and the output is a trained generative AI model.

[0471] Step 4:

[0472] A user uses their own terminal to access the system's interface and input a question or problem. For example, they input a question such as, "What is the optimal marketing strategy in the current market situation?" The server receives this query and performs preprocessing such as grammar analysis and keyword extraction. The input is the user's question data, and the output is the preprocessed question data.

[0473] Step 5:

[0474] The server passes the preprocessed question data to the generative AI model, which generates an appropriate answer based on the knowledge it has learned so far. The input is the preprocessed question data, and the output is the generated answer.

[0475] Step 6:

[0476] The server receives the answer returned by the generative model and sends it to the user's terminal. The user can receive the answer on the terminal. The input is the generated answer, and the output is the delivery of the answer to the user terminal.

[0477] Step 7:

[0478] The server tallies the consulting fees paid by users for each transaction and identifies which knowledge provider's data was used. Based on that data, it calculates the reward for the knowledge provider. The reward is calculated taking into account factors such as frequency of use and the weight of the data. The final calculated reward is paid to the knowledge provider. The input is the consulting fee and the knowledge data used, and the output is the reward for the knowledge provider.

[0479] (Application example 1)

[0480] 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."

[0481] Conventional navigation systems for autonomous vehicles do not adequately propose optimal routes that take real-time traffic and weather information into account, preventing improvements in operational efficiency. Furthermore, they do not provide accurate route proposals that utilize specialized knowledge, leaving issues with operational accuracy and safety. This creates challenges for the efficient and safe operation of autonomous vehicles.

[0482] 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.

[0483] In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for receiving questions from users through a user interface, means for generating answers from the generative model based on the questions, means for acquiring additional data including real-time traffic and weather information when generating answers and adding the additional data to prompt sentences, means for providing the generated answers to users, and means for returning a portion of the consulting fee to the knowledge provider. This enables optimal route proposals based on real-time information for autonomous vehicles, thereby improving operational efficiency and safety.

[0484] A "knowledge provider" is a person or organization that provides specialized knowledge and know-how to the system.

[0485] "Expertise" is deep understanding and technical information in a particular field.

[0486] A "database" is a storage device or data set for efficiently storing and managing collected specialized knowledge.

[0487] A "generative model" is a machine learning algorithm that learns from collected data and generates appropriate answers to user questions.

[0488] A "user interface" is an interface or screen through which a user exchanges information with a system.

[0489] A "prompt sentence" is input data or an instruction sentence provided to a generative model, and is a sentence necessary for the model to generate an appropriate answer.

[0490] "Traffic information" refers to data such as road congestion and accident information collected in real time.

[0491] "Weather information" is data about weather and climate collected in real time.

[0492] An "answer" is a solution or suggestion provided by a generative model in response to a user's question.

[0493] "Consulting fee" is the fee paid by the user for use of the system.

[0494] "Return" means distributing a portion of the revenue gained from the use of the system to the knowledge provider.

[0495] A system for carrying out the present invention includes the following means.

[0496] Knowledge data collection and database construction

[0497] The server first collects expertise from knowledge providers. Expertise is entered through dedicated web forms or applications. This collected data undergoes format checks and is then registered in a database. The database is optimized to prevent anomalies and duplicate data.

[0498] Training a generative model

[0499] The server trains a generative AI model based on the specialized knowledge stored in the database, including preprocessing the data (such as tokenization and text normalization). After training, the generative model can use this accumulated knowledge to generate appropriate answers to questions.

[0500] Query Processing

[0501] The user sends a question to the system through their device. For example, a question might be, "What is the best route to take, taking into account the current traffic and weather conditions?" The server receives the question and retrieves real-time traffic and weather information. Using this additional data, the server generates a prompt sentence and inputs it into the generative model.

[0502] Generate and provide answers

[0503] The generative model utilizes real-time information and specialized knowledge stored in a database to generate optimal answers to user questions, which are then provided to the user via a server.

[0504] reward distribution

[0505] The server calculates the amount of the consulting fee to be returned to the knowledge provider. The server manages the distribution of the reward appropriately based on the knowledge data used.

[0506] Hardware and software used

[0507] Server: Collects and manages expertise, trains generative models, processes queries, generates and serves answers, and manages reward distribution.

[0508] Databases: Used to efficiently store and manage specialized knowledge.

[0509] Generative AI model: A machine learning algorithm for generating appropriate answers to questions, specifically GPT-3 and higher models (e.g., GPT-3.5-turbo).

[0510] External APIs: Used to obtain real-time traffic and weather information. Examples include "Traffic Information API" and "Weather Information API."

[0511] User Interface: An interface for receiving expertise input and user questions.

[0512] Specific examples

[0513] For example, the following prompt statement is generated:

[0514] Available knowledge:

[0515] Best practices for efficient route planning

[0516] Traffic Information:

[0517] Current traffic conditions: Traffic jams, accidents Road closures: None

[0518] Weather information:

[0519] Local weather: Sunny Forecast: Light rain overnight

[0520] Use this information to suggest the best route.

[0521] Using this prompt, the generative model suggests an appropriate route, which is then provided to the user, thereby improving the operational efficiency and safety of autonomous vehicles.

[0522] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0523] Step 1:

[0524] The server collects specialized knowledge from knowledge providers. Knowledge providers enter their specialized knowledge and know-how using dedicated web forms or applications. The input data is checked for format by the server and registered in a database. The input is via a dedicated form, and the output is format-checked specialized knowledge data.

[0525] Step 2:

[0526] The server trains a generative AI model based on the expertise registered in the database. It extracts the necessary expertise data from the database and performs preprocessing such as tokenization and text normalization. It then inputs the preprocessed data into a generative model (e.g., GPT-3.5-turbo) and trains the model. The input is data from the expertise database, and the output is a trained generative model.

[0527] Step 3:

[0528] A user inputs a question into the system through a terminal. For example, a question might be, "What is the best route to take, taking into account the current traffic conditions and weather?" The question is sent to the server via the user interface. The input is the question from the user, and the output is the received question data.

[0529] Step 4:

[0530] After receiving the query, the server retrieves real-time traffic and weather information. It uses external APIs to retrieve the latest data from the traffic information API and weather information API. The input is the user's query, and the output is traffic and weather information data.

[0531] Step 5:

[0532] The server generates prompt sentences based on the user's question and the acquired real-time information. The generated prompt sentences are in the format required for input to the generative model. The inputs are the user's question, traffic information, and weather information, and the output is the generated prompt sentence.

[0533] Step 6:

[0534] The server inputs the generated prompt sentence into the generative model to generate an answer. The generative model utilizes a database of expert knowledge and real-time information to generate the optimal answer. The input is the prompt sentence, and the output is the generated answer.

[0535] Step 7:

[0536] The server provides the generated answer to the user, and sends the answer to the user's terminal for display. The input is the generated answer, and the output is the display of the answer on the user's terminal.

[0537] Step 8:

[0538] The server performs calculations to return a portion of the consulting fee to the knowledge provider. It distributes the appropriate reward based on the knowledge data used. The input is the system usage data and the consulting fee, and the output is the reward return to the knowledge provider.

[0539] The above steps will enable the system to propose optimal routes for autonomous vehicles based on real-time information, improving operational efficiency and safety.

[0540] 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.

[0541] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to companies and users. Furthermore, this invention also includes a function to improve the user experience by combining it with an emotion engine that recognizes user emotions. The main functions of this system are database construction, AI learning, emotion analysis, query processing, and reward distribution.

[0542] Database construction

[0543] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, they enter "latest findings on marketing strategies." The server receives this data and performs an initial format check. Knowledge that passes the format check is registered in a database based on specific categories and tags (e.g., marketing, financial management, law, etc.).

[0544] AI Learning

[0545] The server operates to train a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. Then it inputs the preprocessed data into the generative model, and trains the model to answer questions appropriately.

[0546] Emotion analysis

[0547] When a user (company representative) uses a device to input a question or issue, the emotion engine captures and analyzes the user's emotional state. For example, it analyzes emotions (joy, anger, sadness, surprise, etc.) from the user's input text. Based on this analysis, the server deeply understands the context of the question and passes it to the appropriate generative model.

[0548] Query Processing

[0549] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The generated answer is sent by the server to the user's device and provided to the user.

[0550] reward distribution

[0551] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0552] Specific examples

[0553] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0554] 2. A user (a corporate marketing person) inputs the question "How can we effectively promote a new product?" into a device. The emotion engine analyzes the emotion from the user's input and detects that the user is "impatient." The server passes this information to a generative model, which generates an answer that takes into account the user's emotional state: "Use social media advertising to reach the target demographic." The device then provides this answer to the user.

[0555] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0556] This allows users to receive prompt and accurate consultation, and knowledge providers can earn rewards. Furthermore, by providing answers that take into account the user's emotional state, the user experience is improved.

[0557] The processing flow will be explained below.

[0558] Step 1:

[0559] Users (knowledge providers) use a dedicated web form or application to input their own expertise and know-how, for example, "the latest knowledge on marketing strategies."

[0560] Step 2:

[0561] The server receives the input expert knowledge data and performs a format check, which includes checking the format of the data and basic error checks.

[0562] Step 3:

[0563] The server registers the knowledge data that has passed the format check into a database based on specific categories or tags (e.g., marketing, legal, financial management, etc.).

[0564] Step 4:

[0565] The server extracts purpose-specific knowledge data from the database for training the generative model.

[0566] Step 5:

[0567] The server performs preprocessing on the extracted knowledge data, including tokenization, text normalization, and feature engineering.

[0568] Step 6:

[0569] The server trains a generative model on the preprocessed knowledge data, which the generative model uses to improve its ability to generate appropriate answers to questions.

[0570] Step 7:

[0571] A user (a company employee) uses a terminal to input a question or problem into the system. For example, the user inputs a question such as, "What is the optimal marketing strategy in the current market situation?"

[0572] Step 8:

[0573] The emotion engine captures and analyzes emotions from the user's input sentences, detecting, for example, the user's emotional state, such as "anxious" or "worried."

[0574] Step 9:

[0575] The server receives the question data sent by the user and analyzes the question, taking into account the results of the emotion engine.

[0576] Step 10:

[0577] Based on the analysis results, the server passes the question data to the generative model, which then uses its learned knowledge to generate an appropriate answer in a tone that reflects the user's emotional state.

[0578] Step 11:

[0579] The server reviews the generated answers to ensure clarity and accuracy of the content.

[0580] Step 12:

[0581] The device displays the answers sent from the server to the user, who can then make decisions and take action based on the generated answers.

[0582] Step 13:

[0583] The server aggregates transaction data for the consulting session, including tracking what knowledge was used.

[0584] Step 14:

[0585] The server calculates the compensation for the knowledge provider based on the collected data, which is calculated as part of the consulting fee.

[0586] Step 15:

[0587] The server returns the calculated reward to the knowledge provider, who can receive the reward in return for utilizing their knowledge.

[0588] Example 2

[0589] 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."

[0590] In conventional consulting systems, it has been difficult to efficiently manage the expertise collected from knowledge providers and create generative models based on that data. It has also been difficult to generate appropriate answers that take the user's emotional state into account, resulting in insufficient improvement in the user experience. Furthermore, the distribution of rewards to knowledge providers has also lacked transparency and appropriateness. The present invention aims to solve these problems and provide a consulting system with higher accuracy and a superior user experience.

[0591] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for analyzing the emotional state of a user, means for generating an answer from the generative model based on a question from the user, means for providing the generated answer to the user, and means for returning a portion of the consulting fee to the knowledge provider. This makes it possible to generate highly accurate answers that take the emotional state of the user into consideration, improve the user experience, and appropriately distribute rewards to knowledge providers.

[0592] A "knowledge provider" is an individual or organization that has specialized knowledge or know-how and provides that information to the system.

[0593] "Specialized knowledge" is information related to advanced knowledge and skills in a particular field.

[0594] A "database" is an information system for systematically storing and managing collected specialized knowledge and information.

[0595] A "generative model" is an algorithm or system that learns from collected data and generates new information or answers.

[0596] "Preprocessing" refers to processes such as text normalization and tokenization that are performed on data before it is trained by a generative model.

[0597] "Means for analyzing emotional state" refers to a technology or mechanism for analyzing emotions from user input data and identifying that state.

[0598] "Means for receiving questions" refers to the method or device by which users input questions or issues to the consulting system.

[0599] A "means for generating an answer" is a method or system that uses a generative model to create an appropriate answer to a question from a user.

[0600] "Consulting Fee" means the fee paid by the User when using the Consulting Service.

[0601] "Means of return" refers to a method or system for distributing a portion of the consulting fee to the knowledge provider.

[0602] This invention is a system that collects specialized knowledge from knowledge providers, stores it in a database, trains a generative model, and provides highly accurate consulting to users. Furthermore, it aims to improve the user experience by analyzing the user's emotional state and providing optimal answers. The main functions of this invention are data collection from knowledge providers, database construction, AI learning, sentiment analysis, query processing, and reward distribution.

[0603] 1. Database construction

[0604] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, a marketing expert might enter "effective methods for running social media ads." The server receives this data and performs a format check. Knowledge that passes the format check is registered in a database based on specific categories and tags (e.g., marketing, financial management, law, etc.).

[0605] 2. AI Learning

[0606] The server trains a generative model (such as GPT-3) based on the collected knowledge data. It extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. It then inputs the preprocessed data into the generative model, training it to answer questions appropriately.

[0607] 3. Emotion analysis

[0608] When a user uses a device to input a question or problem, the emotion engine captures and analyzes the user's emotional state. For example, it analyzes emotions (joy, anger, sadness, surprise, etc.) from the user's input text. Based on this analysis, the server deeply understands the context of the question and passes it to the appropriate generative model.

[0609] 4. Query Processing

[0610] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The generated answer is sent by the server to the user's device and provided to the user.

[0611] 5. Reward distribution

[0612] The server tallies the consulting fees paid by users for each transaction. It identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0613] Specific examples

[0614] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0615] 2. A user (a corporate marketing person) inputs a question into a device, such as "How can we effectively promote a new product?" The emotion engine analyzes the emotion in the user's input and detects that the user is "impatient." The server passes this information to a generative model, which generates an answer that takes into account the user's emotional state (e.g., "Use social media advertising to reach the target demographic"). The device then provides this answer to the user.

[0616] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0617] This allows users to receive prompt and accurate consultation, and knowledge providers can earn rewards. It also significantly improves the user experience by providing answers that take into account the user's emotional state. This system offers a new way to quickly respond to the diverse needs of modern businesses and utilize expertise more efficiently.

[0618] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0619] System program processing flow

[0620] Step 1: Gather knowledge

[0621] Input: Expertise and know-how from knowledge providers

[0622] Output: Format-checked data transferred to the server

[0623] Specific behavior:

[0624] 1. Knowledge providers use a dedicated web form or application to enter their expertise and know-how (e.g., how to effectively run social media ads).

[0625] 2. The server receives this input data.

[0626] 3. The server performs a format check on the data, which includes checking for required fields and validating the data format.

[0627] 4. The server sends the data, which has passed the format check, to the database.

[0628] Step 2: Registering with the database

[0629] Input: Format-checked expert knowledge data

[0630] Output: Categorized and tagged database entries

[0631] Specific behavior:

[0632] 1. The server parses the received, format-checked data and assigns it appropriate categories and tags (e.g., marketing, financial management, legal, etc.).

[0633] 2. The server stores the categories and tagged data in a database.

[0634] Step 3: Data extraction and preprocessing

[0635] Input: Expert knowledge data in the database

[0636] Output: Preprocessed data

[0637] Specific behavior:

[0638] 1. The server extracts data from the database to use in the generative model (e.g., GPT-3).

[0639] 2. The server performs preprocessing on the extracted data (tokenization, text normalization, etc.) to make the data suitable for the generative model.

[0640] Step 4: Training the generative model

[0641] Input: Preprocessed data

[0642] Output: A trained generative model

[0643] Specific behavior:

[0644] 1. The server feeds the preprocessed data into the generative model.

[0645] 2. The server trains the generative model, improving the model's ability to generate appropriate answers to user questions.

[0646] Step 5: User sentiment analysis

[0647] Input: User questions and challenges

[0648] Output: Emotion analysis results

[0649] Specific behavior:

[0650] 1. A user uses a device to enter a question or challenge (e.g., "How can we effectively promote a new product?").

[0651] 2. The device sends the input data to the server.

[0652] 3. The server uses an emotion engine to analyze emotions from the user's input data (e.g., to detect joy, anger, sadness, surprise, etc.).

[0653] Step 6: Parsing the question and generating an answer

[0654] Input: User question and sentiment analysis results

[0655] Output: Answers that take into account the user's emotional state

[0656] Specific behavior:

[0657] 1. The server analyzes the user's question and takes into account the results of sentiment analysis.

[0658] 2. The server passes the question to the generative model in the appropriate context.

[0659] 3. The generative model leverages its trained knowledge to generate appropriate answers in a tone that reflects the user's emotional state.

[0660] 4. The server sends the generated answer to the user's device.

[0661] Step 7: Provide your answers

[0662] Input: Answer generation result

[0663] Output: The answer provided to the user

[0664] Specific behavior:

[0665] 1. The server generates a response and sends it to the user's device.

[0666] 2. The device displays and provides the answer to the user.

[0667] Step 8: Aggregating and distributing rewards

[0668] Input: User consulting fees and knowledge data used

[0669] Output: Reward returned to knowledge provider

[0670] Specific behavior:

[0671] 1. The server tallies the consulting fees paid by the user for each transaction.

[0672] 2. The server identifies which knowledge provider's data was used.

[0673] 3. The server calculates and returns the reward to the knowledge provider based on this.

[0674] The above is a specific processing flow and operation of the system according to the present invention.

[0675] (Application example 2)

[0676] 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."

[0677] In customer service operations in brick-and-mortar stores, there is a demand for methods to quickly and accurately answer customer questions. Furthermore, it is expected that customer satisfaction will be improved by providing responses in an appropriate tone based on the customer's emotional state. Current customer service systems have difficulty in providing such advanced interactions, and there are many situations where they are forced to respond with limited information and knowledge. As a result, customer dissatisfaction is likely to occur, affecting the quality of customer service.

[0678] 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.

[0679] In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for receiving questions from users, means for generating answers from the generative model based on the questions, means for providing the generated answers to the users, means for analyzing the emotional state of the user at the time of asking the question, means for generating answers in a tone corresponding to the emotional state, means for displaying the generated answer information on a visual display device, and means for returning a portion of the consulting fee to the knowledge provider. This makes it possible to quickly provide appropriate answers to customer questions according to the emotional state when serving customers in a physical store, thereby improving customer satisfaction.

[0680] A "knowledge provider" is someone who provides specialized knowledge and know-how.

[0681] "Expertise" refers to advanced knowledge and experience in a particular field.

[0682] A "database" is a system that systematically stores information and allows it to be easily searched and updated.

[0683] A "generative model" is an artificial intelligence model that learns from collected data and generates new information and answers.

[0684] "User" means a person who uses the system to ask questions or seek consultation.

[0685] A "question" is a question that a user asks the system for information or advice.

[0686] An "answer" is information or advice that the system provides in response to a question.

[0687] "Emotional state" refers to the emotions such as joy, anger, sadness, surprise, etc. that the user is feeling at the time of asking the question.

[0688] A "visual display device" is a device for visually presenting information to a user.

[0689] "Consulting Fee" means the fee paid by User for Consulting Services.

[0690] "Return" refers to returning a portion of the consulting fee to the knowledge provider as compensation.

[0691] The system to realize this application example mainly requires a server, a user terminal, and a visual display device (smart glasses). The specific system configuration and processing method are described below.

[0692] System configuration and processing method

[0693] 1. Gathering expertise from knowledge providers

[0694] Knowledge providers provide their expertise and know-how using a dedicated web form or application. For example, they enter "latest insights on marketing strategies." The server receives this data, performs a format check, and then registers it in a database. This database is organized according to specific categories or tags (e.g., marketing, financial management, legal, etc.).

[0695] 2. Training the generative model

[0696] The server trains a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. The preprocessed data is then input into the generative model, and the model is trained to be able to answer various questions appropriately.

[0697] 3. Receiving user questions and analyzing emotional states

[0698] When a user (company representative) uses a device to input a question or issue, the emotion engine captures and analyzes the user's emotional state. For example, if a user asks, "I'd like to know more about this product," the emotion (joy, anger, sadness, surprise, etc.) is analyzed from the input sentence. Based on the results of this analysis, the server gains a deep understanding of the context of the question and passes it to the generative model.

[0699] 4. Generating answers based on questions

[0700] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. For example, if the user is anxious, an answer in a "relaxing tone" is generated. The generated answer is sent by the server to the user's smart glasses and displayed visually.

[0701] 5. Consulting fee aggregation and reward returns

[0702] The server aggregates the data from the consulting session and returns a certain amount of compensation to the knowledge provider. It identifies which knowledge provider's data was used, calculates the compensation based on that usage, and pays it to the knowledge provider.

[0703] Hardware and software used

[0704] Hardware:

[0705] Smart glasses (e.g. Google Glass, Vuzix)

[0706] software:

[0707] OpenAI API: Using generative AI models (GPT-3)

[0708] TextBlob: A Python library for sentiment analysis

[0709] Specific processing examples

[0710] For example, if a user asks about the features of a new smartphone:

[0711] User asks: "What are the features of this new smartphone?"

[0712] Example prompt sentence:

[0713] Customer asks: "What are the features of this new smartphone?" Provide detailed information and suggestions if needed. Product info: "Product Info for new smartphone: High quality, cost-effective, and trendy."

[0714] The server analyzes emotions and provides appropriate answers based on the user's questions and knowledge database, and displays the information visually through smart glasses to support customer service in brick-and-mortar stores, thereby improving customer satisfaction.

[0715] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0716] Step 1:

[0717] The server collects expertise from knowledge providers. Knowledge providers use a dedicated web form or application to input their expertise and know-how. For example, they input "latest findings on marketing strategies." The server receives this data and performs a format check. Data that passes the format check is registered in the database. The input is the knowledge provider's expertise, and the output is important information stored in the database.

[0718] Step 2:

[0719] The server trains a generative model based on the specialized knowledge in the database. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. The preprocessed data is input into the generative model, and the model is trained to be able to answer various questions appropriately. The input is the knowledge data in the database, and the output is the trained generative model.

[0720] Step 3:

[0721] The user inputs a question or problem using a terminal. For example, the user might ask, "I'd like to know more about this product." The server receives this question and simultaneously analyzes the user's emotional state using an emotion engine. The input is the user's question and emotional state, and the output is the analysis results.

[0722] Step 4:

[0723] The server analyzes the question sent by the user and passes the analysis results from the emotion engine to the generative model. The generative model utilizes knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The input is the user's question and emotional state, and the output is the generated answer.

[0724] Step 5:

[0725] The server transmits the generated answer to the user's device. In particular, when smart glasses are used, the answer information is visually displayed. For example, when the user is wearing smart glasses, the generated answer is displayed on the display of the glasses. The input is the generated answer, and the output is the answer displayed on the user's visual display device.

[0726] Step 6:

[0727] The server returns a portion of the consulting fee to the knowledge provider. The server aggregates the data from the consulting session and identifies which knowledge provider's data was used. It then calculates a reward based on the knowledge used and pays it to the knowledge provider. The input is the consulting session data, and the output is the reward payment to the knowledge provider.

[0728] 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.

[0729] 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.

[0730] 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.

[0731] [Third embodiment]

[0732] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0733] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0734] 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).

[0735] 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.

[0736] 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.

[0737] 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).

[0738] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0739] 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.

[0740] 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.

[0741] 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.

[0742] 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.

[0743] 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."

[0744] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to companies and users. To this end, this invention has the following main functions: database construction, AI learning, query processing, and reward distribution.

[0745] Database construction

[0746] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, they enter "latest findings on marketing strategies." The server receives this entered data and performs an initial format check. Knowledge that passes the format check is registered in a database based on specific categories or tags (e.g., marketing, financial management, legal, etc.). The database is structured to prevent anomalies and duplicate data.

[0747] AI Learning

[0748] The server operates to train a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. Then it inputs the preprocessed data into the generative model, and trains the model to answer questions appropriately.

[0749] Query Processing

[0750] A user (e.g., a company employee) uses their own device to input a question or problem into the system. For example, they might input a question like, "What is the optimal marketing strategy in the current market situation?" The server receives the user's question, analyzes it, and passes it to the generative model. The generative model then utilizes the knowledge accumulated through learning to generate an appropriate answer. The generated answer is then sent by the server to the user's device and provided to the user.

[0751] reward distribution

[0752] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0753] Specific examples

[0754] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0755] 2. The user (a corporate marketing person) inputs a question from their device: "How can we effectively promote a new product?" The server receives this question and passes it to the generative model. The generative model uses its accumulated knowledge of "effective methods for running social media ads" to propose a specific strategy, including "using social media ads to reach the target demographic." The server then provides this answer to the user.

[0756] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0757] This allows users to receive prompt and highly accurate consulting, and knowledge providers can also receive rewards.

[0758] The processing flow will be explained below.

[0759] Step 1:

[0760] Users (knowledge providers) use a dedicated web form or application to input their expertise, for example, "the latest knowledge on marketing strategies."

[0761] Step 2:

[0762] The server receives the input expert knowledge data and performs a format check, which includes checking the format of the data and basic error checks.

[0763] Step 3:

[0764] The server registers the knowledge data that has passed the format check into a database based on specific categories or tags (e.g., marketing, legal, financial management, etc.).

[0765] Step 4:

[0766] The server extracts purpose-specific knowledge data from the database for training the generative model.

[0767] Step 5:

[0768] The server performs preprocessing on the extracted knowledge data, including tokenization, text normalization, and feature engineering.

[0769] Step 6:

[0770] The server trains a generative model on the preprocessed knowledge data, which the generative model uses to improve its ability to generate appropriate answers to questions.

[0771] Step 7:

[0772] A user (a company employee) uses a terminal to input a question or problem into the system. For example, the user inputs a question such as, "What is the optimal marketing strategy in the current market situation?"

[0773] Step 8:

[0774] The server receives the question data sent by the user and analyzes the question, using natural language processing techniques to understand the intent of the question.

[0775] Step 9:

[0776] Based on the analysis results, the server passes the question data to the generative model, which then uses its learned knowledge to generate an answer to the question.

[0777] Step 10:

[0778] The server receives the answers returned by the generative model and reviews them, optionally checking their clarity and accuracy.

[0779] Step 11:

[0780] The device displays the answers sent from the server to the user, who can then make decisions and take action based on the generated answers.

[0781] Step 12:

[0782] The server aggregates transaction data for the consulting session, including tracking what knowledge was used.

[0783] Step 13:

[0784] The server calculates the compensation for the knowledge provider based on the collected data, which is calculated as part of the consulting fee.

[0785] Step 14:

[0786] The server returns the calculated reward to the knowledge provider, who can receive the reward in return for utilizing their knowledge.

[0787] Example 1

[0788] 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."

[0789] Conventional consulting systems have insufficient management of collected expertise and data organization, making it difficult to generate highly accurate answers. Furthermore, the distribution of rewards to knowledge providers is often vague and unfair. This has led to issues such as a decline in user satisfaction and the motivation of knowledge providers.

[0790] 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.

[0791] In this invention, the server includes means for collecting expertise from knowledge providers, means for storing the collected expertise in a database, means for performing preprocessing including tokenization and text normalization on the expertise in the database, means for training a generative model based on the preprocessed expertise, means for receiving questions from users, means for generating answers from the generative model based on the questions, means for providing the generated answers to the users, and means for returning a portion of the consulting fee to the knowledge providers. This improves the accuracy and management of the collected expertise, making it possible to provide accurate answers, and also realizes fair and transparent reward distribution, thereby improving the satisfaction of both users and knowledge providers.

[0792] A "knowledge provider" is an individual or organization that has specialized knowledge and is responsible for providing that knowledge to the system.

[0793] "Specialized knowledge" is information that requires advanced knowledge and skills in a specific field.

[0794] A "database" is a digital information storage system for systematically storing and managing collected specialized knowledge based on categories and tags.

[0795] "Tokenization" is the process of dividing text data into units of words or sentences.

[0796] "Text normalization" is the process of removing noise and unnecessary information from text data and converting the data into a unified format.

[0797] A "generative model" is an artificial intelligence model that learns from collected and preprocessed expert knowledge and generates answers to user questions.

[0798] A "user" is an individual or entity that uses the system to enter questions or challenges and receive generated answers.

[0799] A "consulting fee" is a fee paid by a user to obtain answers to their questions or problems.

[0800] "Reward sharing" is the process of returning a portion of the consulting fee to the knowledge provider as compensation for the expertise provided.

[0801] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to users. The system has the following main functions: database construction, AI learning, query processing, and reward distribution. Each function will be explained in detail below.

[0802] Database construction

[0803] Knowledge providers use their own devices to access a dedicated web form or application and enter their expertise. For example, they enter "latest knowledge on marketing strategies." When the server receives this input data, it first performs a format check to ensure that the data conforms to the expected format. Data that passes the format check is assigned a specific category (e.g., marketing, financial management, legal, etc.) or tag and stored in a database.

[0804] AI Learning

[0805] The server performs preprocessing, including tokenization and text normalization, on the expert knowledge stored in the database. The preprocessed data is used to train a generative AI model (e.g., GPT-3). The server feeds the data to the generative model in batches, and improves the model's accuracy by tuning hyperparameters and evaluating them every epoch.

[0806] Query Processing

[0807] Users use their own devices to access the system's interface and input questions or issues. For example, they might input a question like, "What is the optimal marketing strategy given the current market situation?" The server receives this question, performs preprocessing such as grammatical analysis and keyword extraction, and then passes the question to the generative model. The generative model generates an appropriate answer based on the learned knowledge. The generated answer is then sent to the user's device via the server.

[0808] reward distribution

[0809] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates a reward based on that data. The reward is calculated taking into account factors such as frequency of use and the weight of the data. Finally, the calculated reward is paid to the knowledge provider.

[0810] Specific examples

[0811] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the marketing category.

[0812] 2. A user (e.g., a corporate marketing manager) inputs a question from their device, such as "How can we effectively promote a new product?" The server receives this question, performs the necessary preprocessing, and passes the question to the generative model. The generative model utilizes its accumulated knowledge of "effective methods for running social media advertising" to propose a specific strategy, including "using social media advertising to reach the target demographic." The server then provides this answer to the user.

[0813] 3. The server aggregates the data from this consulting session and calculates the reward for the knowledge provider used. The calculated reward is paid to the knowledge provider.

[0814] Prompt Sentence Examples

[0815] How to effectively promote new products?

[0816] This allows users to receive prompt and highly accurate consulting, and knowledge providers can also receive rewards.

[0817] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0818] Step 1:

[0819] A knowledge provider uses their own device to input their expertise. For example, they enter "latest findings on marketing strategies" into a web form. The server receives this input data and performs a format check. This format check verifies whether the input data conforms to the expected format. Data that passes the format check is assigned specific categories and tags and registered in the database. The input is the knowledge provider's expertise data, and the output is the format-checked data registered in the database.

[0820] Step 2:

[0821] The server periodically scans the expertise data registered in the database and performs preprocessing for model training. Specifically, it divides the text data into words and sentences through tokenization, and performs text normalization to remove noise and unnecessary information. It also detects and removes duplicate data. The input is the expertise data in the database, and the output is the preprocessed data.

[0822] Step 3:

[0823] The server trains a generative AI model (e.g., GPT-3) using the preprocessed data. Data is fed to the model in batches. The server adjusts hyperparameters and evaluates each epoch to improve the model's accuracy. The input is the preprocessed data, and the output is a trained generative AI model.

[0824] Step 4:

[0825] A user uses their own terminal to access the system's interface and input a question or problem. For example, they input a question such as, "What is the optimal marketing strategy in the current market situation?" The server receives this query and performs preprocessing such as grammar analysis and keyword extraction. The input is the user's question data, and the output is the preprocessed question data.

[0826] Step 5:

[0827] The server passes the preprocessed question data to the generative AI model, which generates an appropriate answer based on the knowledge it has learned so far. The input is the preprocessed question data, and the output is the generated answer.

[0828] Step 6:

[0829] The server receives the answer returned by the generative model and sends it to the user's terminal. The user can receive the answer on the terminal. The input is the generated answer, and the output is the delivery of the answer to the user terminal.

[0830] Step 7:

[0831] The server tallies the consulting fees paid by users for each transaction and identifies which knowledge provider's data was used. Based on that data, it calculates the reward for the knowledge provider. The reward is calculated taking into account factors such as frequency of use and the weight of the data. The final calculated reward is paid to the knowledge provider. The input is the consulting fee and the knowledge data used, and the output is the reward for the knowledge provider.

[0832] (Application example 1)

[0833] 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."

[0834] Conventional navigation systems for autonomous vehicles do not adequately propose optimal routes that take real-time traffic and weather information into account, preventing improvements in operational efficiency. Furthermore, they do not provide accurate route proposals that utilize specialized knowledge, leaving issues with operational accuracy and safety. This creates challenges for the efficient and safe operation of autonomous vehicles.

[0835] 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.

[0836] In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for receiving questions from users through a user interface, means for generating answers from the generative model based on the questions, means for acquiring additional data including real-time traffic and weather information when generating answers and adding the additional data to prompt sentences, means for providing the generated answers to users, and means for returning a portion of the consulting fee to the knowledge provider. This enables optimal route proposals based on real-time information for autonomous vehicles, thereby improving operational efficiency and safety.

[0837] A "knowledge provider" is a person or organization that provides specialized knowledge and know-how to the system.

[0838] "Expertise" is deep understanding and technical information in a particular field.

[0839] A "database" is a storage device or data set for efficiently storing and managing collected specialized knowledge.

[0840] A "generative model" is a machine learning algorithm that learns from collected data and generates appropriate answers to user questions.

[0841] A "user interface" is an interface or screen through which a user exchanges information with a system.

[0842] A "prompt sentence" is input data or an instruction sentence provided to a generative model, and is a sentence necessary for the model to generate an appropriate answer.

[0843] "Traffic information" refers to data such as road congestion and accident information collected in real time.

[0844] "Weather information" is data about weather and climate collected in real time.

[0845] An "answer" is a solution or suggestion provided by a generative model in response to a user's question.

[0846] "Consulting fee" is the fee paid by the user for use of the system.

[0847] "Return" means distributing a portion of the revenue gained from the use of the system to the knowledge provider.

[0848] A system for carrying out the present invention includes the following means.

[0849] Knowledge data collection and database construction

[0850] The server first collects expertise from knowledge providers. Expertise is entered through dedicated web forms or applications. This collected data undergoes format checks and is then registered in a database. The database is optimized to prevent anomalies and duplicate data.

[0851] Training a generative model

[0852] The server trains a generative AI model based on the specialized knowledge stored in the database, including preprocessing the data (such as tokenization and text normalization). After training, the generative model can use this accumulated knowledge to generate appropriate answers to questions.

[0853] Query Processing

[0854] The user sends a question to the system through their device. For example, a question might be, "What is the best route to take, taking into account the current traffic and weather conditions?" The server receives the question and retrieves real-time traffic and weather information. Using this additional data, the server generates a prompt sentence and inputs it into the generative model.

[0855] Generate and provide answers

[0856] The generative model utilizes real-time information and specialized knowledge stored in a database to generate optimal answers to user questions, which are then provided to the user via a server.

[0857] reward distribution

[0858] The server calculates the amount of the consulting fee to be returned to the knowledge provider. The server manages the distribution of the reward appropriately based on the knowledge data used.

[0859] Hardware and software used

[0860] Server: Collects and manages expertise, trains generative models, processes queries, generates and serves answers, and manages reward distribution.

[0861] Databases: Used to efficiently store and manage specialized knowledge.

[0862] Generative AI model: A machine learning algorithm for generating appropriate answers to questions, specifically GPT-3 and higher models (e.g., GPT-3.5-turbo).

[0863] External APIs: Used to obtain real-time traffic and weather information. Examples include "Traffic Information API" and "Weather Information API."

[0864] User Interface: An interface for receiving expertise input and user questions.

[0865] Specific examples

[0866] For example, the following prompt statement is generated:

[0867] Available knowledge:

[0868] Best practices for efficient route planning

[0869] Traffic Information:

[0870] Current traffic conditions: Traffic jams, accidents Road closures: None

[0871] Weather information:

[0872] Local weather: Sunny Forecast: Light rain overnight

[0873] Use this information to suggest the best route.

[0874] Using this prompt, the generative model suggests an appropriate route, which is then provided to the user, thereby improving the operational efficiency and safety of autonomous vehicles.

[0875] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0876] Step 1:

[0877] The server collects specialized knowledge from knowledge providers. Knowledge providers enter their specialized knowledge and know-how using dedicated web forms or applications. The input data is checked for format by the server and registered in a database. The input is via a dedicated form, and the output is format-checked specialized knowledge data.

[0878] Step 2:

[0879] The server trains a generative AI model based on the expertise registered in the database. It extracts the necessary expertise data from the database and performs preprocessing such as tokenization and text normalization. It then inputs the preprocessed data into a generative model (e.g., GPT-3.5-turbo) and trains the model. The input is data from the expertise database, and the output is a trained generative model.

[0880] Step 3:

[0881] A user inputs a question into the system through a terminal. For example, a question might be, "What is the best route to take, taking into account the current traffic conditions and weather?" The question is sent to the server via the user interface. The input is the question from the user, and the output is the received question data.

[0882] Step 4:

[0883] After receiving the query, the server retrieves real-time traffic and weather information. It uses external APIs to retrieve the latest data from the traffic information API and weather information API. The input is the user's query, and the output is traffic and weather information data.

[0884] Step 5:

[0885] The server generates prompt sentences based on the user's question and the acquired real-time information. The generated prompt sentences are in the format required for input to the generative model. The inputs are the user's question, traffic information, and weather information, and the output is the generated prompt sentence.

[0886] Step 6:

[0887] The server inputs the generated prompt sentence into the generative model to generate an answer. The generative model utilizes a database of expert knowledge and real-time information to generate the optimal answer. The input is the prompt sentence, and the output is the generated answer.

[0888] Step 7:

[0889] The server provides the generated answer to the user, and sends the answer to the user's terminal for display. The input is the generated answer, and the output is the display of the answer on the user's terminal.

[0890] Step 8:

[0891] The server performs calculations to return a portion of the consulting fee to the knowledge provider. It distributes the appropriate reward based on the knowledge data used. The input is the system usage data and the consulting fee, and the output is the reward return to the knowledge provider.

[0892] The above steps will enable the system to propose optimal routes for autonomous vehicles based on real-time information, improving operational efficiency and safety.

[0893] 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.

[0894] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to companies and users. Furthermore, this invention also includes a function to improve the user experience by combining it with an emotion engine that recognizes user emotions. The main functions of this system are database construction, AI learning, emotion analysis, query processing, and reward distribution.

[0895] Database construction

[0896] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, they enter "latest findings on marketing strategies." The server receives this data and performs an initial format check. Knowledge that passes the format check is registered in a database based on specific categories and tags (e.g., marketing, financial management, law, etc.).

[0897] AI Learning

[0898] The server operates to train a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. Then it inputs the preprocessed data into the generative model, and trains the model to answer questions appropriately.

[0899] Emotion analysis

[0900] When a user (company representative) uses a device to input a question or issue, the emotion engine captures and analyzes the user's emotional state. For example, it analyzes emotions (joy, anger, sadness, surprise, etc.) from the user's input text. Based on this analysis, the server deeply understands the context of the question and passes it to the appropriate generative model.

[0901] Query Processing

[0902] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The generated answer is sent by the server to the user's device and provided to the user.

[0903] reward distribution

[0904] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0905] Specific examples

[0906] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0907] 2. A user (a corporate marketing person) inputs the question "How can we effectively promote a new product?" into a device. The emotion engine analyzes the emotion from the user's input and detects that the user is "impatient." The server passes this information to a generative model, which generates an answer that takes into account the user's emotional state: "Use social media advertising to reach the target demographic." The device then provides this answer to the user.

[0908] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0909] This allows users to receive prompt and accurate consultation, and knowledge providers can earn rewards. Furthermore, by providing answers that take into account the user's emotional state, the user experience is improved.

[0910] The processing flow will be explained below.

[0911] Step 1:

[0912] Users (knowledge providers) use a dedicated web form or application to input their own expertise and know-how, for example, "the latest knowledge on marketing strategies."

[0913] Step 2:

[0914] The server receives the input expert knowledge data and performs a format check, which includes checking the format of the data and basic error checks.

[0915] Step 3:

[0916] The server registers the knowledge data that has passed the format check into a database based on specific categories or tags (e.g., marketing, legal, financial management, etc.).

[0917] Step 4:

[0918] The server extracts purpose-specific knowledge data from the database for training the generative model.

[0919] Step 5:

[0920] The server performs preprocessing on the extracted knowledge data, including tokenization, text normalization, and feature engineering.

[0921] Step 6:

[0922] The server trains a generative model on the preprocessed knowledge data, which the generative model uses to improve its ability to generate appropriate answers to questions.

[0923] Step 7:

[0924] A user (a company employee) uses a terminal to input a question or problem into the system. For example, the user inputs a question such as, "What is the optimal marketing strategy in the current market situation?"

[0925] Step 8:

[0926] The emotion engine captures and analyzes emotions from the user's input sentences, detecting, for example, the user's emotional state, such as "anxious" or "worried."

[0927] Step 9:

[0928] The server receives the question data sent by the user and analyzes the question, taking into account the results of the emotion engine.

[0929] Step 10:

[0930] Based on the analysis results, the server passes the question data to the generative model, which then uses its learned knowledge to generate an appropriate answer in a tone that reflects the user's emotional state.

[0931] Step 11:

[0932] The server reviews the generated answers to ensure clarity and accuracy of the content.

[0933] Step 12:

[0934] The device displays the answers sent from the server to the user, who can then make decisions and take action based on the generated answers.

[0935] Step 13:

[0936] The server aggregates transaction data for the consulting session, including tracking what knowledge was used.

[0937] Step 14:

[0938] The server calculates the compensation for the knowledge provider based on the collected data, which is calculated as part of the consulting fee.

[0939] Step 15:

[0940] The server returns the calculated reward to the knowledge provider, who can receive the reward in return for utilizing their knowledge.

[0941] Example 2

[0942] 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."

[0943] In conventional consulting systems, it has been difficult to efficiently manage the expertise collected from knowledge providers and create generative models based on that data. It has also been difficult to generate appropriate answers that take the user's emotional state into account, resulting in insufficient improvement in the user experience. Furthermore, the distribution of rewards to knowledge providers has also lacked transparency and appropriateness. The present invention aims to solve these problems and provide a consulting system with higher accuracy and a superior user experience.

[0944] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for analyzing the emotional state of a user, means for generating an answer from the generative model based on a question from the user, means for providing the generated answer to the user, and means for returning a portion of the consulting fee to the knowledge provider. This makes it possible to generate highly accurate answers that take the emotional state of the user into consideration, improve the user experience, and appropriately distribute rewards to knowledge providers.

[0945] A "knowledge provider" is an individual or organization that has specialized knowledge or know-how and provides that information to the system.

[0946] "Specialized knowledge" is information related to advanced knowledge and skills in a particular field.

[0947] A "database" is an information system for systematically storing and managing collected specialized knowledge and information.

[0948] A "generative model" is an algorithm or system that learns from collected data and generates new information or answers.

[0949] "Preprocessing" refers to processes such as text normalization and tokenization that are performed on data before it is trained by a generative model.

[0950] "Means for analyzing emotional state" refers to a technology or mechanism for analyzing emotions from user input data and identifying that state.

[0951] "Means for receiving questions" refers to the method or device by which users input questions or issues to the consulting system.

[0952] A "means for generating an answer" is a method or system that uses a generative model to create an appropriate answer to a question from a user.

[0953] "Consulting Fee" means the fee paid by the User when using the Consulting Service.

[0954] "Means of return" refers to a method or system for distributing a portion of the consulting fee to the knowledge provider.

[0955] This invention is a system that collects specialized knowledge from knowledge providers, stores it in a database, trains a generative model, and provides highly accurate consulting to users. Furthermore, it aims to improve the user experience by analyzing the user's emotional state and providing optimal answers. The main functions of this invention are data collection from knowledge providers, database construction, AI learning, sentiment analysis, query processing, and reward distribution.

[0956] 1. Database construction

[0957] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, a marketing expert might enter "effective methods for running social media ads." The server receives this data and performs a format check. Knowledge that passes the format check is registered in a database based on specific categories and tags (e.g., marketing, financial management, law, etc.).

[0958] 2. AI Learning

[0959] The server trains a generative model (such as GPT-3) based on the collected knowledge data. It extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. It then inputs the preprocessed data into the generative model, training it to answer questions appropriately.

[0960] 3. Emotion analysis

[0961] When a user uses a device to input a question or problem, the emotion engine captures and analyzes the user's emotional state. For example, it analyzes emotions (joy, anger, sadness, surprise, etc.) from the user's input text. Based on this analysis, the server deeply understands the context of the question and passes it to the appropriate generative model.

[0962] 4. Query Processing

[0963] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The generated answer is sent by the server to the user's device and provided to the user.

[0964] 5. Reward distribution

[0965] The server tallies the consulting fees paid by users for each transaction. It identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[0966] Specific examples

[0967] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[0968] 2. A user (a corporate marketing person) inputs a question into a device, such as "How can we effectively promote a new product?" The emotion engine analyzes the emotion in the user's input and detects that the user is "impatient." The server passes this information to a generative model, which generates an answer that takes into account the user's emotional state (e.g., "Use social media advertising to reach the target demographic"). The device then provides this answer to the user.

[0969] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[0970] This allows users to receive prompt and accurate consultation, and knowledge providers can earn rewards. It also significantly improves the user experience by providing answers that take into account the user's emotional state. This system offers a new way to quickly respond to the diverse needs of modern businesses and utilize expertise more efficiently.

[0971] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0972] System program processing flow

[0973] Step 1: Gather knowledge

[0974] Input: Expertise and know-how from knowledge providers

[0975] Output: Format-checked data transferred to the server

[0976] Specific behavior:

[0977] 1. Knowledge providers use a dedicated web form or application to enter their expertise and know-how (e.g., how to effectively run social media ads).

[0978] 2. The server receives this input data.

[0979] 3. The server performs a format check on the data, which includes checking for required fields and validating the data format.

[0980] 4. The server sends the data, which has passed the format check, to the database.

[0981] Step 2: Registering with the database

[0982] Input: Format-checked expert knowledge data

[0983] Output: Categorized and tagged database entries

[0984] Specific behavior:

[0985] 1. The server parses the received, format-checked data and assigns it appropriate categories and tags (e.g., marketing, financial management, legal, etc.).

[0986] 2. The server stores the categories and tagged data in a database.

[0987] Step 3: Data extraction and preprocessing

[0988] Input: Expert knowledge data in the database

[0989] Output: Preprocessed data

[0990] Specific behavior:

[0991] 1. The server extracts data from the database to use in the generative model (e.g., GPT-3).

[0992] 2. The server performs preprocessing on the extracted data (tokenization, text normalization, etc.) to make the data suitable for the generative model.

[0993] Step 4: Training the generative model

[0994] Input: Preprocessed data

[0995] Output: A trained generative model

[0996] Specific behavior:

[0997] 1. The server feeds the preprocessed data into the generative model.

[0998] 2. The server trains the generative model, improving the model's ability to generate appropriate answers to user questions.

[0999] Step 5: User sentiment analysis

[1000] Input: User questions and challenges

[1001] Output: Emotion analysis results

[1002] Specific behavior:

[1003] 1. A user uses a device to enter a question or challenge (e.g., "How can we effectively promote a new product?").

[1004] 2. The device sends the input data to the server.

[1005] 3. The server uses an emotion engine to analyze emotions from the user's input data (e.g., to detect joy, anger, sadness, surprise, etc.).

[1006] Step 6: Parsing the question and generating an answer

[1007] Input: User question and sentiment analysis results

[1008] Output: Answers that take into account the user's emotional state

[1009] Specific behavior:

[1010] 1. The server analyzes the user's question and takes into account the results of sentiment analysis.

[1011] 2. The server passes the question to the generative model in the appropriate context.

[1012] 3. The generative model leverages its trained knowledge to generate appropriate answers in a tone that reflects the user's emotional state.

[1013] 4. The server sends the generated answer to the user's device.

[1014] Step 7: Provide your answers

[1015] Input: Answer generation result

[1016] Output: The answer provided to the user

[1017] Specific behavior:

[1018] 1. The server generates a response and sends it to the user's device.

[1019] 2. The device displays and provides the answer to the user.

[1020] Step 8: Aggregating and distributing rewards

[1021] Input: User consulting fees and knowledge data used

[1022] Output: Reward returned to knowledge provider

[1023] Specific behavior:

[1024] 1. The server tallies the consulting fees paid by the user for each transaction.

[1025] 2. The server identifies which knowledge provider's data was used.

[1026] 3. The server calculates and returns the reward to the knowledge provider based on this.

[1027] The above is a specific processing flow and operation of the system according to the present invention.

[1028] (Application example 2)

[1029] 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."

[1030] In customer service operations in brick-and-mortar stores, there is a demand for methods to quickly and accurately answer customer questions. Furthermore, it is expected that customer satisfaction will be improved by providing responses in an appropriate tone based on the customer's emotional state. Current customer service systems have difficulty in providing such advanced interactions, and there are many situations where they are forced to respond with limited information and knowledge. As a result, customer dissatisfaction is likely to occur, affecting the quality of customer service.

[1031] 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.

[1032] In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for receiving questions from users, means for generating answers from the generative model based on the questions, means for providing the generated answers to the users, means for analyzing the emotional state of the user at the time of asking the question, means for generating answers in a tone corresponding to the emotional state, means for displaying the generated answer information on a visual display device, and means for returning a portion of the consulting fee to the knowledge provider. This makes it possible to quickly provide appropriate answers to customer questions according to the emotional state when serving customers in a physical store, thereby improving customer satisfaction.

[1033] A "knowledge provider" is someone who provides specialized knowledge and know-how.

[1034] "Expertise" refers to advanced knowledge and experience in a particular field.

[1035] A "database" is a system that systematically stores information and allows it to be easily searched and updated.

[1036] A "generative model" is an artificial intelligence model that learns from collected data and generates new information and answers.

[1037] "User" means a person who uses the system to ask questions or seek consultation.

[1038] A "question" is a question that a user asks the system for information or advice.

[1039] An "answer" is information or advice that the system provides in response to a question.

[1040] "Emotional state" refers to the emotions such as joy, anger, sadness, surprise, etc. that the user is feeling at the time of asking the question.

[1041] A "visual display device" is a device for visually presenting information to a user.

[1042] "Consulting Fee" means the fee paid by User for Consulting Services.

[1043] "Return" refers to returning a portion of the consulting fee to the knowledge provider as compensation.

[1044] The system to realize this application example mainly requires a server, a user terminal, and a visual display device (smart glasses). The specific system configuration and processing method are described below.

[1045] System configuration and processing method

[1046] 1. Gathering expertise from knowledge providers

[1047] Knowledge providers provide their expertise and know-how using a dedicated web form or application. For example, they enter "latest insights on marketing strategies." The server receives this data, performs a format check, and then registers it in a database. This database is organized according to specific categories or tags (e.g., marketing, financial management, legal, etc.).

[1048] 2. Training the generative model

[1049] The server trains a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. The preprocessed data is then input into the generative model, and the model is trained to be able to answer various questions appropriately.

[1050] 3. Receiving user questions and analyzing emotional states

[1051] When a user (company representative) uses a device to input a question or issue, the emotion engine captures and analyzes the user's emotional state. For example, if a user asks, "I'd like to know more about this product," the emotion (joy, anger, sadness, surprise, etc.) is analyzed from the input sentence. Based on the results of this analysis, the server gains a deep understanding of the context of the question and passes it to the generative model.

[1052] 4. Generating answers based on questions

[1053] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. For example, if the user is anxious, an answer in a "relaxing tone" is generated. The generated answer is sent by the server to the user's smart glasses and displayed visually.

[1054] 5. Consulting fee aggregation and reward returns

[1055] The server aggregates the data from the consulting session and returns a certain amount of compensation to the knowledge provider. It identifies which knowledge provider's data was used, calculates the compensation based on that usage, and pays it to the knowledge provider.

[1056] Hardware and software used

[1057] Hardware:

[1058] Smart glasses (e.g. Google Glass, Vuzix)

[1059] software:

[1060] OpenAI API: Using generative AI models (GPT-3)

[1061] TextBlob: A Python library for sentiment analysis

[1062] Specific processing examples

[1063] For example, if a user asks about the features of a new smartphone:

[1064] User asks: "What are the features of this new smartphone?"

[1065] Example prompt sentence:

[1066] Customer asks: "What are the features of this new smartphone?" Provide detailed information and suggestions if needed. Product info: "Product Info for new smartphone: High quality, cost-effective, and trendy."

[1067] The server analyzes emotions and provides appropriate answers based on the user's questions and knowledge database, and displays the information visually through smart glasses to support customer service in brick-and-mortar stores, thereby improving customer satisfaction.

[1068] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1069] Step 1:

[1070] The server collects expertise from knowledge providers. Knowledge providers use a dedicated web form or application to input their expertise and know-how. For example, they input "latest findings on marketing strategies." The server receives this data and performs a format check. Data that passes the format check is registered in the database. The input is the knowledge provider's expertise, and the output is important information stored in the database.

[1071] Step 2:

[1072] The server trains a generative model based on the specialized knowledge in the database. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. The preprocessed data is input into the generative model, and the model is trained to be able to answer various questions appropriately. The input is the knowledge data in the database, and the output is the trained generative model.

[1073] Step 3:

[1074] The user inputs a question or problem using a terminal. For example, the user might ask, "I'd like to know more about this product." The server receives this question and simultaneously analyzes the user's emotional state using an emotion engine. The input is the user's question and emotional state, and the output is the analysis results.

[1075] Step 4:

[1076] The server analyzes the question sent by the user and passes the analysis results from the emotion engine to the generative model. The generative model utilizes knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The input is the user's question and emotional state, and the output is the generated answer.

[1077] Step 5:

[1078] The server transmits the generated answer to the user's device. In particular, when smart glasses are used, the answer information is visually displayed. For example, when the user is wearing smart glasses, the generated answer is displayed on the display of the glasses. The input is the generated answer, and the output is the answer displayed on the user's visual display device.

[1079] Step 6:

[1080] The server returns a portion of the consulting fee to the knowledge provider. The server aggregates the data from the consulting session and identifies which knowledge provider's data was used. It then calculates a reward based on the knowledge used and pays it to the knowledge provider. The input is the consulting session data, and the output is the reward payment to the knowledge provider.

[1081] 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.

[1082] 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.

[1083] 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.

[1084] [Fourth embodiment]

[1085] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1086] 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.

[1087] 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).

[1088] 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.

[1089] 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.

[1090] 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).

[1091] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1092] 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.

[1093] 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.

[1094] 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.

[1095] 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.

[1096] 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.

[1097] 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."

[1098] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to companies and users. To this end, this invention has the following main functions: database construction, AI learning, query processing, and reward distribution.

[1099] Database construction

[1100] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, they enter "latest findings on marketing strategies." The server receives this entered data and performs an initial format check. Knowledge that passes the format check is registered in a database based on specific categories or tags (e.g., marketing, financial management, legal, etc.). The database is structured to prevent anomalies and duplicate data.

[1101] AI Learning

[1102] The server operates to train a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. Then it inputs the preprocessed data into the generative model, and trains the model to answer questions appropriately.

[1103] Query Processing

[1104] A user (e.g., a company employee) uses their own device to input a question or problem into the system. For example, they might input a question like, "What is the optimal marketing strategy in the current market situation?" The server receives the user's question, analyzes it, and passes it to the generative model. The generative model then utilizes the knowledge accumulated through learning to generate an appropriate answer. The generated answer is then sent by the server to the user's device and provided to the user.

[1105] reward distribution

[1106] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[1107] Specific examples

[1108] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[1109] 2. The user (a corporate marketing person) inputs a question from their device: "How can we effectively promote a new product?" The server receives this question and passes it to the generative model. The generative model uses its accumulated knowledge of "effective methods for running social media ads" to propose a specific strategy, including "using social media ads to reach the target demographic." The server then provides this answer to the user.

[1110] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[1111] This allows users to receive prompt and highly accurate consulting, and knowledge providers can also receive rewards.

[1112] The processing flow will be explained below.

[1113] Step 1:

[1114] Users (knowledge providers) use a dedicated web form or application to input their expertise, for example, "the latest knowledge on marketing strategies."

[1115] Step 2:

[1116] The server receives the input expert knowledge data and performs a format check, which includes checking the format of the data and basic error checks.

[1117] Step 3:

[1118] The server registers the knowledge data that has passed the format check into a database based on specific categories or tags (e.g., marketing, legal, financial management, etc.).

[1119] Step 4:

[1120] The server extracts purpose-specific knowledge data from the database for training the generative model.

[1121] Step 5:

[1122] The server performs preprocessing on the extracted knowledge data, including tokenization, text normalization, and feature engineering.

[1123] Step 6:

[1124] The server trains a generative model on the preprocessed knowledge data, which the generative model uses to improve its ability to generate appropriate answers to questions.

[1125] Step 7:

[1126] A user (a company employee) uses a terminal to input a question or problem into the system. For example, the user inputs a question such as, "What is the optimal marketing strategy in the current market situation?"

[1127] Step 8:

[1128] The server receives the question data sent by the user and analyzes the question, using natural language processing techniques to understand the intent of the question.

[1129] Step 9:

[1130] Based on the analysis results, the server passes the question data to the generative model, which then uses its learned knowledge to generate an answer to the question.

[1131] Step 10:

[1132] The server receives the answers returned by the generative model and reviews them, optionally checking their clarity and accuracy.

[1133] Step 11:

[1134] The device displays the answers sent from the server to the user, who can then make decisions and take action based on the generated answers.

[1135] Step 12:

[1136] The server aggregates transaction data for the consulting session, including tracking what knowledge was used.

[1137] Step 13:

[1138] The server calculates the compensation for the knowledge provider based on the collected data, which is calculated as part of the consulting fee.

[1139] Step 14:

[1140] The server returns the calculated reward to the knowledge provider, who can receive the reward in return for utilizing their knowledge.

[1141] Example 1

[1142] 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."

[1143] Conventional consulting systems have insufficient management of collected expertise and data organization, making it difficult to generate highly accurate answers. Furthermore, the distribution of rewards to knowledge providers is often vague and unfair. This has led to issues such as a decline in user satisfaction and the motivation of knowledge providers.

[1144] 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.

[1145] In this invention, the server includes means for collecting expertise from knowledge providers, means for storing the collected expertise in a database, means for performing preprocessing including tokenization and text normalization on the expertise in the database, means for training a generative model based on the preprocessed expertise, means for receiving questions from users, means for generating answers from the generative model based on the questions, means for providing the generated answers to the users, and means for returning a portion of the consulting fee to the knowledge providers. This improves the accuracy and management of the collected expertise, making it possible to provide accurate answers, and also realizes fair and transparent reward distribution, thereby improving the satisfaction of both users and knowledge providers.

[1146] A "knowledge provider" is an individual or organization that has specialized knowledge and is responsible for providing that knowledge to the system.

[1147] "Specialized knowledge" is information that requires advanced knowledge and skills in a specific field.

[1148] A "database" is a digital information storage system for systematically storing and managing collected specialized knowledge based on categories and tags.

[1149] "Tokenization" is the process of dividing text data into units of words or sentences.

[1150] "Text normalization" is the process of removing noise and unnecessary information from text data and converting the data into a unified format.

[1151] A "generative model" is an artificial intelligence model that learns from collected and preprocessed expert knowledge and generates answers to user questions.

[1152] A "user" is an individual or entity that uses the system to enter questions or challenges and receive generated answers.

[1153] A "consulting fee" is a fee paid by a user to obtain answers to their questions or problems.

[1154] "Reward sharing" is the process of returning a portion of the consulting fee to the knowledge provider as compensation for the expertise provided.

[1155] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to users. The system has the following main functions: database construction, AI learning, query processing, and reward distribution. Each function will be explained in detail below.

[1156] Database construction

[1157] Knowledge providers use their own devices to access a dedicated web form or application and enter their expertise. For example, they enter "latest knowledge on marketing strategies." When the server receives this input data, it first performs a format check to ensure that the data conforms to the expected format. Data that passes the format check is assigned a specific category (e.g., marketing, financial management, legal, etc.) or tag and stored in a database.

[1158] AI Learning

[1159] The server performs preprocessing, including tokenization and text normalization, on the expert knowledge stored in the database. The preprocessed data is used to train a generative AI model (e.g., GPT-3). The server feeds the data to the generative model in batches, and improves the model's accuracy by tuning hyperparameters and evaluating them every epoch.

[1160] Query Processing

[1161] Users use their own devices to access the system's interface and input questions or issues. For example, they might input a question like, "What is the optimal marketing strategy given the current market situation?" The server receives this question, performs preprocessing such as grammatical analysis and keyword extraction, and then passes the question to the generative model. The generative model generates an appropriate answer based on the learned knowledge. The generated answer is then sent to the user's device via the server.

[1162] reward distribution

[1163] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates a reward based on that data. The reward is calculated taking into account factors such as frequency of use and the weight of the data. Finally, the calculated reward is paid to the knowledge provider.

[1164] Specific examples

[1165] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the marketing category.

[1166] 2. A user (e.g., a corporate marketing manager) inputs a question from their device, such as "How can we effectively promote a new product?" The server receives this question, performs the necessary preprocessing, and passes the question to the generative model. The generative model utilizes its accumulated knowledge of "effective methods for running social media advertising" to propose a specific strategy, including "using social media advertising to reach the target demographic." The server then provides this answer to the user.

[1167] 3. The server aggregates the data from this consulting session and calculates the reward for the knowledge provider used. The calculated reward is paid to the knowledge provider.

[1168] Prompt Sentence Examples

[1169] How to effectively promote new products?

[1170] This allows users to receive prompt and highly accurate consulting, and knowledge providers can also receive rewards.

[1171] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1172] Step 1:

[1173] A knowledge provider uses their own device to input their expertise. For example, they enter "latest findings on marketing strategies" into a web form. The server receives this input data and performs a format check. This format check verifies whether the input data conforms to the expected format. Data that passes the format check is assigned specific categories and tags and registered in the database. The input is the knowledge provider's expertise data, and the output is the format-checked data registered in the database.

[1174] Step 2:

[1175] The server periodically scans the expertise data registered in the database and performs preprocessing for model training. Specifically, it divides the text data into words and sentences through tokenization, and performs text normalization to remove noise and unnecessary information. It also detects and removes duplicate data. The input is the expertise data in the database, and the output is the preprocessed data.

[1176] Step 3:

[1177] The server trains a generative AI model (e.g., GPT-3) using the preprocessed data. Data is fed to the model in batches. The server adjusts hyperparameters and evaluates each epoch to improve the model's accuracy. The input is the preprocessed data, and the output is a trained generative AI model.

[1178] Step 4:

[1179] A user uses their own terminal to access the system's interface and input a question or problem. For example, they input a question such as, "What is the optimal marketing strategy in the current market situation?" The server receives this query and performs preprocessing such as grammar analysis and keyword extraction. The input is the user's question data, and the output is the preprocessed question data.

[1180] Step 5:

[1181] The server passes the preprocessed question data to the generative AI model, which generates an appropriate answer based on the knowledge it has learned so far. The input is the preprocessed question data, and the output is the generated answer.

[1182] Step 6:

[1183] The server receives the answer returned by the generative model and sends it to the user's terminal. The user can receive the answer on the terminal. The input is the generated answer, and the output is the delivery of the answer to the user terminal.

[1184] Step 7:

[1185] The server tallies the consulting fees paid by users for each transaction and identifies which knowledge provider's data was used. Based on that data, it calculates the reward for the knowledge provider. The reward is calculated taking into account factors such as frequency of use and the weight of the data. The final calculated reward is paid to the knowledge provider. The input is the consulting fee and the knowledge data used, and the output is the reward for the knowledge provider.

[1186] (Application example 1)

[1187] 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."

[1188] Conventional navigation systems for autonomous vehicles do not adequately propose optimal routes that take real-time traffic and weather information into account, preventing improvements in operational efficiency. Furthermore, they do not provide accurate route proposals that utilize specialized knowledge, leaving issues with operational accuracy and safety. This creates challenges for the efficient and safe operation of autonomous vehicles.

[1189] 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.

[1190] In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for receiving questions from users through a user interface, means for generating answers from the generative model based on the questions, means for acquiring additional data including real-time traffic and weather information when generating answers and adding the additional data to prompt sentences, means for providing the generated answers to users, and means for returning a portion of the consulting fee to the knowledge provider. This enables optimal route proposals based on real-time information for autonomous vehicles, thereby improving operational efficiency and safety.

[1191] A "knowledge provider" is a person or organization that provides specialized knowledge and know-how to the system.

[1192] "Expertise" is deep understanding and technical information in a particular field.

[1193] A "database" is a storage device or data set for efficiently storing and managing collected specialized knowledge.

[1194] A "generative model" is a machine learning algorithm that learns from collected data and generates appropriate answers to user questions.

[1195] A "user interface" is an interface or screen through which a user exchanges information with a system.

[1196] A "prompt sentence" is input data or an instruction sentence provided to a generative model, and is a sentence necessary for the model to generate an appropriate answer.

[1197] "Traffic information" refers to data such as road congestion and accident information collected in real time.

[1198] "Weather information" is data about weather and climate collected in real time.

[1199] An "answer" is a solution or suggestion provided by a generative model in response to a user's question.

[1200] "Consulting fee" is the fee paid by the user for use of the system.

[1201] "Return" means distributing a portion of the revenue gained from the use of the system to the knowledge provider.

[1202] A system for carrying out the present invention includes the following means.

[1203] Knowledge data collection and database construction

[1204] The server first collects expertise from knowledge providers. Expertise is entered through dedicated web forms or applications. This collected data undergoes format checks and is then registered in a database. The database is optimized to prevent anomalies and duplicate data.

[1205] Training a generative model

[1206] The server trains a generative AI model based on the specialized knowledge stored in the database, including preprocessing the data (such as tokenization and text normalization). After training, the generative model can use this accumulated knowledge to generate appropriate answers to questions.

[1207] Query Processing

[1208] The user sends a question to the system through their device. For example, a question might be, "What is the best route to take, taking into account the current traffic and weather conditions?" The server receives the question and retrieves real-time traffic and weather information. Using this additional data, the server generates a prompt sentence and inputs it into the generative model.

[1209] Generate and provide answers

[1210] The generative model utilizes real-time information and specialized knowledge stored in a database to generate optimal answers to user questions, which are then provided to the user via a server.

[1211] reward distribution

[1212] The server calculates the amount of the consulting fee to be returned to the knowledge provider. The server manages the distribution of the reward appropriately based on the knowledge data used.

[1213] Hardware and software used

[1214] Server: Collects and manages expertise, trains generative models, processes queries, generates and serves answers, and manages reward distribution.

[1215] Databases: Used to efficiently store and manage specialized knowledge.

[1216] Generative AI model: A machine learning algorithm for generating appropriate answers to questions, specifically GPT-3 and higher models (e.g., GPT-3.5-turbo).

[1217] External APIs: Used to obtain real-time traffic and weather information. Examples include "Traffic Information API" and "Weather Information API."

[1218] User Interface: An interface for receiving expertise input and user questions.

[1219] Specific examples

[1220] For example, the following prompt statement is generated:

[1221] Available knowledge:

[1222] Best practices for efficient route planning

[1223] Traffic Information:

[1224] Current traffic conditions: Traffic jams, accidents Road closures: None

[1225] Weather information:

[1226] Local weather: Sunny Forecast: Light rain overnight

[1227] Use this information to suggest the best route.

[1228] Using this prompt, the generative model suggests an appropriate route, which is then provided to the user, thereby improving the operational efficiency and safety of autonomous vehicles.

[1229] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1230] Step 1:

[1231] The server collects specialized knowledge from knowledge providers. Knowledge providers enter their specialized knowledge and know-how using dedicated web forms or applications. The input data is checked for format by the server and registered in a database. The input is via a dedicated form, and the output is format-checked specialized knowledge data.

[1232] Step 2:

[1233] The server trains a generative AI model based on the expertise registered in the database. It extracts the necessary expertise data from the database and performs preprocessing such as tokenization and text normalization. It then inputs the preprocessed data into a generative model (e.g., GPT-3.5-turbo) and trains the model. The input is data from the expertise database, and the output is a trained generative model.

[1234] Step 3:

[1235] A user inputs a question into the system through a terminal. For example, a question might be, "What is the best route to take, taking into account the current traffic conditions and weather?" The question is sent to the server via the user interface. The input is the question from the user, and the output is the received question data.

[1236] Step 4:

[1237] After receiving the query, the server retrieves real-time traffic and weather information. It uses external APIs to retrieve the latest data from the traffic information API and weather information API. The input is the user's query, and the output is traffic and weather information data.

[1238] Step 5:

[1239] The server generates prompt sentences based on the user's question and the acquired real-time information. The generated prompt sentences are in the format required for input to the generative model. The inputs are the user's question, traffic information, and weather information, and the output is the generated prompt sentence.

[1240] Step 6:

[1241] The server inputs the generated prompt sentence into the generative model to generate an answer. The generative model utilizes a database of expert knowledge and real-time information to generate the optimal answer. The input is the prompt sentence, and the output is the generated answer.

[1242] Step 7:

[1243] The server provides the generated answer to the user, and sends the answer to the user's terminal for display. The input is the generated answer, and the output is the display of the answer on the user's terminal.

[1244] Step 8:

[1245] The server performs calculations to return a portion of the consulting fee to the knowledge provider. It distributes the appropriate reward based on the knowledge data used. The input is the system usage data and the consulting fee, and the output is the reward return to the knowledge provider.

[1246] The above steps will enable the system to propose optimal routes for autonomous vehicles based on real-time information, improving operational efficiency and safety.

[1247] 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.

[1248] This invention is a system that collects specialized knowledge from knowledge providers, trains a generative model based on that knowledge, and provides highly accurate consulting services to companies and users. Furthermore, this invention also includes a function to improve the user experience by combining it with an emotion engine that recognizes user emotions. The main functions of this system are database construction, AI learning, emotion analysis, query processing, and reward distribution.

[1249] Database construction

[1250] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, they enter "latest findings on marketing strategies." The server receives this data and performs an initial format check. Knowledge that passes the format check is registered in a database based on specific categories and tags (e.g., marketing, financial management, law, etc.).

[1251] AI Learning

[1252] The server operates to train a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. Then it inputs the preprocessed data into the generative model, and trains the model to answer questions appropriately.

[1253] Emotion analysis

[1254] When a user (company representative) uses a device to input a question or issue, the emotion engine captures and analyzes the user's emotional state. For example, it analyzes emotions (joy, anger, sadness, surprise, etc.) from the user's input text. Based on this analysis, the server deeply understands the context of the question and passes it to the appropriate generative model.

[1255] Query Processing

[1256] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The generated answer is sent by the server to the user's device and provided to the user.

[1257] reward distribution

[1258] The server tallies the consulting fees paid by users for each transaction. It then identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[1259] Specific examples

[1260] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[1261] 2. A user (a corporate marketing person) inputs the question "How can we effectively promote a new product?" into a device. The emotion engine analyzes the emotion from the user's input and detects that the user is "impatient." The server passes this information to a generative model, which generates an answer that takes into account the user's emotional state: "Use social media advertising to reach the target demographic." The device then provides this answer to the user.

[1262] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[1263] This allows users to receive prompt and accurate consultation, and knowledge providers can earn rewards. Furthermore, by providing answers that take into account the user's emotional state, the user experience is improved.

[1264] The processing flow will be explained below.

[1265] Step 1:

[1266] Users (knowledge providers) use a dedicated web form or application to input their own expertise and know-how, for example, "the latest knowledge on marketing strategies."

[1267] Step 2:

[1268] The server receives the input expert knowledge data and performs a format check, which includes checking the format of the data and basic error checks.

[1269] Step 3:

[1270] The server registers the knowledge data that has passed the format check into a database based on specific categories or tags (e.g., marketing, legal, financial management, etc.).

[1271] Step 4:

[1272] The server extracts purpose-specific knowledge data from the database for training the generative model.

[1273] Step 5:

[1274] The server performs preprocessing on the extracted knowledge data, including tokenization, text normalization, and feature engineering.

[1275] Step 6:

[1276] The server trains a generative model on the preprocessed knowledge data, which the generative model uses to improve its ability to generate appropriate answers to questions.

[1277] Step 7:

[1278] A user (a company employee) uses a terminal to input a question or problem into the system. For example, the user inputs a question such as, "What is the optimal marketing strategy in the current market situation?"

[1279] Step 8:

[1280] The emotion engine captures and analyzes emotions from the user's input sentences, detecting, for example, the user's emotional state, such as "anxious" or "worried."

[1281] Step 9:

[1282] The server receives the question data sent by the user and analyzes the question, taking into account the results of the emotion engine.

[1283] Step 10:

[1284] Based on the analysis results, the server passes the question data to the generative model, which then uses its learned knowledge to generate an appropriate answer in a tone that reflects the user's emotional state.

[1285] Step 11:

[1286] The server reviews the generated answers to ensure clarity and accuracy of the content.

[1287] Step 12:

[1288] The device displays the answers sent from the server to the user, who can then make decisions and take action based on the generated answers.

[1289] Step 13:

[1290] The server aggregates transaction data for the consulting session, including tracking what knowledge was used.

[1291] Step 14:

[1292] The server calculates the compensation for the knowledge provider based on the collected data, which is calculated as part of the consulting fee.

[1293] Step 15:

[1294] The server returns the calculated reward to the knowledge provider, who can receive the reward in return for utilizing their knowledge.

[1295] Example 2

[1296] 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."

[1297] In conventional consulting systems, it has been difficult to efficiently manage the expertise collected from knowledge providers and create generative models based on that data. It has also been difficult to generate appropriate answers that take the user's emotional state into account, resulting in insufficient improvement in the user experience. Furthermore, the distribution of rewards to knowledge providers has also lacked transparency and appropriateness. The present invention aims to solve these problems and provide a consulting system with higher accuracy and a superior user experience.

[1298] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for analyzing the emotional state of a user, means for generating an answer from the generative model based on a question from the user, means for providing the generated answer to the user, and means for returning a portion of the consulting fee to the knowledge provider. This makes it possible to generate highly accurate answers that take the emotional state of the user into consideration, improve the user experience, and appropriately distribute rewards to knowledge providers.

[1299] A "knowledge provider" is an individual or organization that has specialized knowledge or know-how and provides that information to the system.

[1300] "Specialized knowledge" is information related to advanced knowledge and skills in a particular field.

[1301] A "database" is an information system for systematically storing and managing collected specialized knowledge and information.

[1302] A "generative model" is an algorithm or system that learns from collected data and generates new information or answers.

[1303] "Preprocessing" refers to processes such as text normalization and tokenization that are performed on data before it is trained by a generative model.

[1304] "Means for analyzing emotional state" refers to a technology or mechanism for analyzing emotions from user input data and identifying that state.

[1305] "Means for receiving questions" refers to the method or device by which users input questions or issues to the consulting system.

[1306] A "means for generating an answer" is a method or system that uses a generative model to create an appropriate answer to a question from a user.

[1307] "Consulting Fee" means the fee paid by the User when using the Consulting Service.

[1308] "Means of return" refers to a method or system for distributing a portion of the consulting fee to the knowledge provider.

[1309] This invention is a system that collects specialized knowledge from knowledge providers, stores it in a database, trains a generative model, and provides highly accurate consulting to users. Furthermore, it aims to improve the user experience by analyzing the user's emotional state and providing optimal answers. The main functions of this invention are data collection from knowledge providers, database construction, AI learning, sentiment analysis, query processing, and reward distribution.

[1310] 1. Database construction

[1311] Knowledge providers enter their expertise and know-how using a dedicated web form or application. For example, a marketing expert might enter "effective methods for running social media ads." The server receives this data and performs a format check. Knowledge that passes the format check is registered in a database based on specific categories and tags (e.g., marketing, financial management, law, etc.).

[1312] 2. AI Learning

[1313] The server trains a generative model (such as GPT-3) based on the collected knowledge data. It extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. It then inputs the preprocessed data into the generative model, training it to answer questions appropriately.

[1314] 3. Emotion analysis

[1315] When a user uses a device to input a question or problem, the emotion engine captures and analyzes the user's emotional state. For example, it analyzes emotions (joy, anger, sadness, surprise, etc.) from the user's input text. Based on this analysis, the server deeply understands the context of the question and passes it to the appropriate generative model.

[1316] 4. Query Processing

[1317] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The generated answer is sent by the server to the user's device and provided to the user.

[1318] 5. Reward distribution

[1319] The server tallies the consulting fees paid by users for each transaction. It identifies which knowledge provider's data was used and calculates the reward to be returned to the knowledge provider for the knowledge used. The calculated reward is paid to the knowledge provider by the server.

[1320] Specific examples

[1321] 1. The knowledge provider enters "effective methods for running social media ads" into a web form as the "latest marketing technology." The server checks the format of this data and registers it in the database.

[1322] 2. A user (a corporate marketing person) inputs a question into a device, such as "How can we effectively promote a new product?" The emotion engine analyzes the emotion in the user's input and detects that the user is "impatient." The server passes this information to a generative model, which generates an answer that takes into account the user's emotional state (e.g., "Use social media advertising to reach the target demographic"). The device then provides this answer to the user.

[1323] 3. The server aggregates the data from this consulting session and returns a certain amount of reward to the knowledge provider.

[1324] This allows users to receive prompt and accurate consultation, and knowledge providers can earn rewards. It also significantly improves the user experience by providing answers that take into account the user's emotional state. This system offers a new way to quickly respond to the diverse needs of modern businesses and utilize expertise more efficiently.

[1325] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1326] System program processing flow

[1327] Step 1: Gather knowledge

[1328] Input: Expertise and know-how from knowledge providers

[1329] Output: Format-checked data transferred to the server

[1330] Specific behavior:

[1331] 1. Knowledge providers use a dedicated web form or application to enter their expertise and know-how (e.g., how to effectively run social media ads).

[1332] 2. The server receives this input data.

[1333] 3. The server performs a format check on the data, which includes checking for required fields and validating the data format.

[1334] 4. The server sends the data, which has passed the format check, to the database.

[1335] Step 2: Registering with the database

[1336] Input: Format-checked expert knowledge data

[1337] Output: Categorized and tagged database entries

[1338] Specific behavior:

[1339] 1. The server parses the received, format-checked data and assigns it appropriate categories and tags (e.g., marketing, financial management, legal, etc.).

[1340] 2. The server stores the categories and tagged data in a database.

[1341] Step 3: Data extraction and preprocessing

[1342] Input: Expert knowledge data in the database

[1343] Output: Preprocessed data

[1344] Specific behavior:

[1345] 1. The server extracts data from the database to use in the generative model (e.g., GPT-3).

[1346] 2. The server performs preprocessing on the extracted data (tokenization, text normalization, etc.) to make the data suitable for the generative model.

[1347] Step 4: Training the generative model

[1348] Input: Preprocessed data

[1349] Output: A trained generative model

[1350] Specific behavior:

[1351] 1. The server feeds the preprocessed data into the generative model.

[1352] 2. The server trains the generative model, improving the model's ability to generate appropriate answers to user questions.

[1353] Step 5: User sentiment analysis

[1354] Input: User questions and challenges

[1355] Output: Emotion analysis results

[1356] Specific behavior:

[1357] 1. A user uses a device to enter a question or challenge (e.g., "How can we effectively promote a new product?").

[1358] 2. The device sends the input data to the server.

[1359] 3. The server uses an emotion engine to analyze emotions from the user's input data (e.g., to detect joy, anger, sadness, surprise, etc.).

[1360] Step 6: Parsing the question and generating an answer

[1361] Input: User question and sentiment analysis results

[1362] Output: Answers that take into account the user's emotional state

[1363] Specific behavior:

[1364] 1. The server analyzes the user's question and takes into account the results of sentiment analysis.

[1365] 2. The server passes the question to the generative model in the appropriate context.

[1366] 3. The generative model leverages its trained knowledge to generate appropriate answers in a tone that reflects the user's emotional state.

[1367] 4. The server sends the generated answer to the user's device.

[1368] Step 7: Provide your answers

[1369] Input: Answer generation result

[1370] Output: The answer provided to the user

[1371] Specific behavior:

[1372] 1. The server generates a response and sends it to the user's device.

[1373] 2. The device displays and provides the answer to the user.

[1374] Step 8: Aggregating and distributing rewards

[1375] Input: User consulting fees and knowledge data used

[1376] Output: Reward returned to knowledge provider

[1377] Specific behavior:

[1378] 1. The server tallies the consulting fees paid by the user for each transaction.

[1379] 2. The server identifies which knowledge provider's data was used.

[1380] 3. The server calculates and returns the reward to the knowledge provider based on this.

[1381] The above is a specific processing flow and operation of the system according to the present invention.

[1382] (Application example 2)

[1383] 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."

[1384] In customer service operations in brick-and-mortar stores, there is a demand for methods to quickly and accurately answer customer questions. Furthermore, it is expected that customer satisfaction will be improved by providing responses in an appropriate tone based on the customer's emotional state. Current customer service systems have difficulty in providing such advanced interactions, and there are many situations where they are forced to respond with limited information and knowledge. As a result, customer dissatisfaction is likely to occur, affecting the quality of customer service.

[1385] 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.

[1386] In this invention, the server includes means for collecting specialized knowledge from knowledge providers, means for storing the collected specialized knowledge in a database, means for training a generative model based on the specialized knowledge in the database, means for receiving questions from users, means for generating answers from the generative model based on the questions, means for providing the generated answers to the users, means for analyzing the emotional state of the user at the time of asking the question, means for generating answers in a tone corresponding to the emotional state, means for displaying the generated answer information on a visual display device, and means for returning a portion of the consulting fee to the knowledge provider. This makes it possible to quickly provide appropriate answers to customer questions according to the emotional state when serving customers in a physical store, thereby improving customer satisfaction.

[1387] A "knowledge provider" is someone who provides specialized knowledge and know-how.

[1388] "Expertise" refers to advanced knowledge and experience in a particular field.

[1389] A "database" is a system that systematically stores information and allows it to be easily searched and updated.

[1390] A "generative model" is an artificial intelligence model that learns from collected data and generates new information and answers.

[1391] "User" means a person who uses the system to ask questions or seek consultation.

[1392] A "question" is a question that a user asks the system for information or advice.

[1393] An "answer" is information or advice that the system provides in response to a question.

[1394] "Emotional state" refers to the emotions such as joy, anger, sadness, surprise, etc. that the user is feeling at the time of asking the question.

[1395] A "visual display device" is a device for visually presenting information to a user.

[1396] "Consulting Fee" means the fee paid by User for Consulting Services.

[1397] "Return" refers to returning a portion of the consulting fee to the knowledge provider as compensation.

[1398] The system to realize this application example mainly requires a server, a user terminal, and a visual display device (smart glasses). The specific system configuration and processing method are described below.

[1399] System configuration and processing method

[1400] 1. Gathering expertise from knowledge providers

[1401] Knowledge providers provide their expertise and know-how using a dedicated web form or application. For example, they enter "latest insights on marketing strategies." The server receives this data, performs a format check, and then registers it in a database. This database is organized according to specific categories or tags (e.g., marketing, financial management, legal, etc.).

[1402] 2. Training the generative model

[1403] The server trains a generative model (such as GPT-3) based on the collected knowledge data. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. The preprocessed data is then input into the generative model, and the model is trained to be able to answer various questions appropriately.

[1404] 3. Receiving user questions and analyzing emotional states

[1405] When a user (company representative) uses a device to input a question or issue, the emotion engine captures and analyzes the user's emotional state. For example, if a user asks, "I'd like to know more about this product," the emotion (joy, anger, sadness, surprise, etc.) is analyzed from the input sentence. Based on the results of this analysis, the server gains a deep understanding of the context of the question and passes it to the generative model.

[1406] 4. Generating answers based on questions

[1407] The server analyzes the question sent by the user and passes it to the generative model, taking into account the results of the emotion engine. The generative model utilizes the knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. For example, if the user is anxious, an answer in a "relaxing tone" is generated. The generated answer is sent by the server to the user's smart glasses and displayed visually.

[1408] 5. Consulting fee aggregation and reward returns

[1409] The server aggregates the data from the consulting session and returns a certain amount of compensation to the knowledge provider. It identifies which knowledge provider's data was used, calculates the compensation based on that usage, and pays it to the knowledge provider.

[1410] Hardware and software used

[1411] Hardware:

[1412] Smart glasses (e.g. Google Glass, Vuzix)

[1413] software:

[1414] OpenAI API: Using generative AI models (GPT-3)

[1415] TextBlob: A Python library for sentiment analysis

[1416] Specific processing examples

[1417] For example, if a user asks about the features of a new smartphone:

[1418] User asks: "What are the features of this new smartphone?"

[1419] Example prompt sentence:

[1420] Customer asks: "What are the features of this new smartphone?" Provide detailed information and suggestions if needed. Product info: "Product Info for new smartphone: High quality, cost-effective, and trendy."

[1421] The server analyzes emotions and provides appropriate answers based on the user's questions and knowledge database, and displays the information visually through smart glasses to support customer service in brick-and-mortar stores, thereby improving customer satisfaction.

[1422] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1423] Step 1:

[1424] The server collects expertise from knowledge providers. Knowledge providers use a dedicated web form or application to input their expertise and know-how. For example, they input "latest findings on marketing strategies." The server receives this data and performs a format check. Data that passes the format check is registered in the database. The input is the knowledge provider's expertise, and the output is important information stored in the database.

[1425] Step 2:

[1426] The server trains a generative model based on the specialized knowledge in the database. First, it extracts the necessary knowledge from the database and performs preprocessing (tokenization, text normalization, etc.) on that knowledge. The preprocessed data is input into the generative model, and the model is trained to be able to answer various questions appropriately. The input is the knowledge data in the database, and the output is the trained generative model.

[1427] Step 3:

[1428] The user inputs a question or problem using a terminal. For example, the user might ask, "I'd like to know more about this product." The server receives this question and simultaneously analyzes the user's emotional state using an emotion engine. The input is the user's question and emotional state, and the output is the analysis results.

[1429] Step 4:

[1430] The server analyzes the question sent by the user and passes the analysis results from the emotion engine to the generative model. The generative model utilizes knowledge accumulated through learning to generate an appropriate answer in a tone that corresponds to the user's emotional state. The input is the user's question and emotional state, and the output is the generated answer.

[1431] Step 5:

[1432] The server transmits the generated answer to the user's device. In particular, when smart glasses are used, the answer information is visually displayed. For example, when the user is wearing smart glasses, the generated answer is displayed on the display of the glasses. The input is the generated answer, and the output is the answer displayed on the user's visual display device.

[1433] Step 6:

[1434] The server returns a portion of the consulting fee to the knowledge provider. The server aggregates the data from the consulting session and identifies which knowledge provider's data was used. It then calculates a reward based on the knowledge used and pays it to the knowledge provider. The input is the consulting session data, and the output is the reward payment to the knowledge provider.

[1435] 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.

[1436] 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.

[1437] 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.

[1438] 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.

[1439] 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.

[1440] 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.

[1441] 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).

[1442] 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.

[1443] 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."

[1444] 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.

[1445] 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).

[1446] 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.

[1447] 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.

[1448] 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.

[1449] 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.

[1450] 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.

[1451] 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.

[1452] 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.

[1453] 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.

[1454] 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.

[1455] 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.

[1456] The following is further disclosed regarding the above embodiment.

[1457] (Claim 1)

[1458] a means of gathering expertise from knowledge providers;

[1459] means for storing said collected expertise in a database;

[1460] a means of training a generative model based on the expertise in the database; and

[1461] means for receiving questions from users;

[1462] means for generating an answer from the generative model based on the question;

[1463] means for providing the generated answer to a user;

[1464] A means to return a portion of the consulting fee to the knowledge provider,

[1465] A system including:

[1466] (Claim 2)

[1467] 2. The system of claim 1, wherein the database registration of the specialized knowledge is based on specific categories or tags.

[1468] (Claim 3)

[1469] The system of claim 1, wherein the specialized knowledge extracted from the database is preprocessed and used to train a generative model.

[1470] "Example 1"

[1471] (Claim 1)

[1472] a means of gathering expertise from knowledge providers;

[1473] means for storing said collected expertise in a database;

[1474] means for performing preprocessing, including tokenization and text normalization, on the expertise in the database;

[1475] A means of training a generative model based on preprocessed expert knowledge; and

[1476] means for receiving questions from users;

[1477] means for generating an answer from the generative model based on the question;

[1478] means for providing the generated answer to a user;

[1479] A means to return a portion of the consulting fee to the knowledge provider,

[1480] A system including:

[1481] (Claim 2)

[1482] 2. The system of claim 1, wherein the database registration of the specialized knowledge is based on specific categories or tags.

[1483] (Claim 3)

[1484] 2. The system according to claim 1, wherein the question from the user is analyzed, and after performing necessary preprocessing, the question is passed to the generative model.

[1485] "Application Example 1"

[1486] (Claim 1)

[1487] a means of gathering expertise from knowledge providers;

[1488] means for storing said collected expertise in a database;

[1489] a means of training a generative model based on the expertise in the database; and

[1490] means for receiving a query from a user through a user interface;

[1491] means for generating an answer from the generative model based on the question;

[1492] means for obtaining additional data, including real-time traffic and weather information, during answer generation and adding it to the prompt;

[1493] means for providing the generated answer to a user;

[1494] A means to return a portion of the consulting fee to the knowledge provider,

[1495] A system including:

[1496] (Claim 2)

[1497] 2. The system of claim 1, wherein the database registration of the specialized knowledge is based on specific categories or tags.

[1498] (Claim 3)

[1499] The system of claim 1, wherein the specialized knowledge extracted from the database is preprocessed and used to train a generative model.

[1500] "Example 2: Combining Emotion Engines"

[1501] (Claim 1)

[1502] a means of gathering expertise from knowledge providers;

[1503] means for storing said collected expertise in a database;

[1504] a means of training a generative model based on the expertise in the database; and

[1505] means for receiving questions from users;

[1506] means for generating an answer from the generative model based on the question;

[1507] means for analyzing the emotional state of a user;

[1508] means for providing the generated answer to a user;

[1509] A means to return a portion of the consulting fee to the knowledge provider,

[1510] A system including:

[1511] (Claim 2)

[1512] 2. The system of claim 1, wherein the database registration of the specialized knowledge is based on specific categories or tags.

[1513] (Claim 3)

[1514] The system of claim 1, wherein the specialized knowledge extracted from the database is preprocessed and used to train a generative model.

[1515] (Claim 4)

[1516] 10. The system of claim 1, further comprising: analyzing the user's emotional state; and adjusting responses based on the analysis.

[1517] "Application example 2 when combining emotion engines"

[1518] (Claim 1)

[1519] a means of gathering expertise from knowledge providers;

[1520] means for storing said collected expertise in a database;

[1521] a means of training a generative model based on the expertise in the database; and

[1522] means for receiving questions from users;

[1523] means for generating an answer from the generative model based on the question;

[1524] means for providing the generated answer to a user;

[1525] means for analyzing the emotional state of a user when asking a question;

[1526] means for generating a response in a tone according to said emotional state;

[1527] means for displaying the generated answer information on a visual display device;

[1528] A means to return a portion of the consulting fee to the knowledge provider,

[1529] A system including:

[1530] (Claim 2)

[1531] 2. The system of claim 1, wherein the database registration of the specialized knowledge is based on specific categories or tags.

[1532] (Claim 3)

[1533] The system of claim 1, wherein the specialized knowledge extracted from the database is preprocessed and used to train a generative model.

[1534] (Claim 4)

[1535] 10. The system of claim 1, wherein the visual display device is smart glasses. [Explanation of symbols]

[1536] 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 of gathering expertise from knowledge providers; means for storing said collected expertise in a database; a means of training a generative model based on the expertise in the database; and means for receiving questions from users; means for generating an answer from the generative model based on the question; means for providing the generated answer to a user; A means to return a portion of the consulting fee to the knowledge provider, A system including:

2. 2. The system of claim 1, wherein the database of specialized knowledge is based on specific categories or tags.

3. The system of claim 1 , wherein the specialized knowledge extracted from the database is preprocessed and used to train a generative model.

Images