system

Abstract

Provide a system. A method for receiving a request from a user includes: a means for analyzing the received request and identifying the appropriate area of ​​expertise based on keywords; a means for forwarding the request to a specialized AI having a database of expertise corresponding to the identified area of ​​expertise; A means for returning the answers obtained from the specialized AI to the user; A system including:

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] Current generative AI systems have difficulty returning accurate answers to local or specialized topics. This leaves users unable to obtain appropriate information for questions that require specialized knowledge, resulting in a lack of accuracy and reliability of the information. Furthermore, it is not realistic for a single generative AI to be able to handle all fields, limiting the depth and breadth of its knowledge. [Means for solving the problem]

[0005] The present invention solves the above problems by providing a system that integrates specialized AIs specialized in multiple fields and queries the appropriate specialized AI in response to a user request. Specifically, the system includes a means for receiving a request from a user, a means for analyzing the received request and identifying the appropriate specialized field based on keywords, a means for forwarding the request to a specialized AI that has an expert knowledge database corresponding to the identified specialized field, and a means for returning an answer obtained from the specialized AI to the user. This allows users to obtain accurate and reliable answers to even specialized questions.

[0006] A "user" is a person who makes an inquiry or request to the system, or an entity that operates a terminal.

[0007] A "request" is a question or request that a user sends to a system to obtain knowledge or information.

[0008] A "means" is a method or process used to solve a problem, or a functional element of a device or system.

[0009] "Analysis" refers to the process of breaking down the request content and extracting its meaning and important keywords.

[0010] "Keywords" are words or phrases that are particularly important in the request and are used to identify a particular area of ​​expertise.

[0011] A "specialty field" is a field in which specific knowledge or skills are systematized, and there are multiple specialties in this system.

[0012] "Identification" refers to the process of selecting the appropriate area of ​​expertise based on the analyzed request content.

[0013] A "specialized knowledge database" is a database that stores information and data related to a specialized field.

[0014] "Specialized AI" is artificial intelligence developed specifically for a specific field of expertise, with knowledge of that field and the ability to generate appropriate responses to requests.

[0015] "Forwarding" is the process of sending a request to a specified specialized AI.

[0016] "Answer" refers to the information or knowledge generated by specialized AI based on a request and provided to the user.

[0017] "Return" refers to the process of sending the answer received from the specialized AI to the user's device.

[0018] A "system" refers to a mechanism with overall functionality that is composed by integrating the above means and specialized AI. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8]FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

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

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] The present invention provides a system that integrates multiple specialized AIs and queries an AI with an appropriate specialty in response to a user request. This system includes the following means.

[0041] Program processing

[0042] 1. Request receiving method:

[0043] The terminal receives a request from the user as input. For example, the user inputs a question into the terminal such as "Tell me about medicine."

[0044] The terminal processes this user request and prepares it for transmission to the server.

[0045] 2. Request Analysis and Expertise Identification Methods:

[0046] The server analyzes the request received from the device. The server analyzes the request content and extracts important keywords from it. For example, it checks whether the keyword "medicine" is included.

[0047] The server then identifies the appropriate specialty based on the extracted keywords, in this case the "medical" specialty because "medicine" is included.

[0048] 3. Expert knowledge database and expert AI:

[0049] The server queries a database with specialized AIs corresponding to the identified specialty, in this case, a specialized AI specialized in the medical field.

[0050] Specialized AI refers to a database containing a large amount of medical-related data and information and generates appropriate answers to user questions, such as "Medicine is the scientific study of human health and disease."

[0051] 4. Response return method:

[0052] The server sends the answer received from the specialized AI back to the user's device.

[0053] The terminal displays the answer received from the server to the user.

[0054] Specific examples

[0055] Example 1: Medical enquiry

[0056] The user types "Teach me about medicine" into the terminal.

[0057] The terminal sends this request to the server.

[0058] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0059] The server forwards the request to a specialized AI in the medical field.

[0060] Based on the request received, the specialized AI generates an answer from its knowledge base: "Medicine is the scientific study of human health and disease."

[0061] The server then sends the answer received from the specialized AI back to the device.

[0062] The terminal displays the answer to the user.

[0063] Example 2: Legal enquiry

[0064] A user types into a terminal, "Tell me about the law."

[0065] The terminal sends this request to the server.

[0066] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0067] The server forwards the request to a specialized AI in the legal field.

[0068] Based on the request received, the specialized AI generates an answer from its own knowledge base: "Law is a system for establishing social rules and norms."

[0069] The server then sends the answer received from the specialized AI back to the device.

[0070] The terminal displays the answer to the user.

[0071] In this way, the present invention provides accurate information based on knowledge of each specialized field in response to user requests.

[0072] The processing flow will be explained below.

[0073] Step 1:

[0074] The user types "Teach me about medicine" into the terminal. When the user presses the send button, the terminal receives this request.

[0075] Step 2:

[0076] The terminal converts the user's request into an HTTP request and prepares it for transmission to the server.

[0077] Step 3:

[0078] The terminal sends an HTTP request to the server (for example, sends the request using the POST method).

[0079] Step 4:

[0080] The server receives the HTTP request, analyzes the request body, and extracts the user's question.

[0081] Step 5:

[0082] The server analyzes the question and extracts important keywords, for example, checking whether the keyword "medicine" is included.

[0083] Step 6:

[0084] The server identifies the appropriate specialty based on the extracted keywords. In this case, since "medicine" is included, it identifies the "medical" specialty.

[0085] Step 7:

[0086] The server selects a specialized AI corresponding to the specified specialty field. In this case, it selects a specialized AI specialized in the medical field.

[0087] Step 8:

[0088] The server forwards the request to the selected specialized AI. The server calls the medical_ai's query method and passes the request content.

[0089] Step 9:

[0090] The specialized AI receives the request and uses its knowledge base to generate an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[0091] Step 10:

[0092] The specialized AI returns the generated answer to the server, which then prepares the answer received from the specialized AI for sending back to the user's device.

[0093] Step 11:

[0094] The server generates an HTTP response for the user's device, includes the answer, and sends this HTTP response to the device.

[0095] Step 12:

[0096] The terminal analyzes the HTTP response received from the server, extracts the answer, and displays the extracted answer to the user.

[0097] Step 13:

[0098] The user can read the answers displayed on the terminal and obtain the desired information.

[0099] Example 1

[0100] 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."

[0101] In today's information society, users need to quickly and accurately obtain detailed information in various specialized fields. However, in current systems, AIs that provide specialized information in specific fields are distributed across each field, making it difficult for users to obtain appropriate information tailored to their respective fields. Furthermore, due to low accuracy in analyzing request content, it may not be possible to identify the exact specialized field the user is looking for. Therefore, there is a need for a system that can quickly provide appropriate answers to user requests.

[0102] 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.

[0103] In this invention, the server includes means for receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty based on keywords, means for transferring the request to a generative AI model having an expert knowledge database corresponding to the identified specialty, and means for returning an answer obtained from the generative AI model to the user. This makes it possible to integrate AI models specialized in different specialty areas and provide information on the appropriate specialty in response to a user request quickly and accurately.

[0104] A "user" is a person who requests information from the system.

[0105] A "request" is an inquiry that a user makes to the system requesting information or services.

[0106] The "means for receiving" is a mechanism or method for capturing requests from users and incorporating them into the system.

[0107] The "analyzing means" refers to a technique or method for analyzing a received request and extracting keywords contained therein.

[0108] "Keywords" are important words or phrases in a request that identify an area of ​​expertise.

[0109] A "discipline" is an area in which information or knowledge falls into a particular category or field.

[0110] A "generative AI model" is an artificial intelligence model that generates and analyzes information based on large amounts of data.

[0111] A "transfer means" is a method or technique for sending a request to a generative AI model in a specified area of ​​expertise.

[0112] A "means for returning" is a mechanism or method for returning answers received from a generative AI model to a user.

[0113] "Network communications" refers to communications technologies that connect computers and devices to send and receive data.

[0114] The present invention is a system that queries AI with an appropriate field of expertise in response to a user request. The system receives the user's request as input, analyzes its content, and identifies the appropriate field of expertise based on keywords. The system then forwards the request to a generative AI model corresponding to the identified field of expertise and returns the generated answer to the user. Specific embodiments of the system are described below.

[0115] Hardware and software used

[0116] 1. Terminal: A device through which a user inputs a request. Examples include smartphones, tablets, and computers.

[0117] 2. Server: A central processing unit responsible for analyzing requests, identifying areas of expertise, forwarding requests to generative AI models, and returning answers. The server has a high-bandwidth network connection and high-performance computing capabilities. For example, a virtual server provided by a cloud service provider (e.g., AWS (registered trademark) or Azure (registered trademark)) can be used.

[0118] 3. Generative AI model: An artificial intelligence model that generates information according to a specific field of expertise. For example, large-scale language models such as GPT-3 (registered trademark) and BERT can be used.

[0119] 4. Natural language processing software: Used to analyze requests and extract keywords. Examples include Python's NLTK and spaCy.

[0120] System Operation Overview

[0121] 1. Request received:

[0122] The user types a request into the terminal, for example, "Tell me about the law."

[0123] The device receives the request and converts it to a format for sending to the server. Specifically, it converts the text to JSON format and sends it to the server as an HTTP request.

[0124] 2. Request analysis and expertise identification:

[0125] The server analyzes the incoming request, using natural language processing software (e.g., spaCy) to tokenize the request, tag it with parts of speech, and analyze dependencies.

[0126] The server extracts important keywords using a keyword extraction algorithm (e.g., TF-IDF). For example, it detects the keyword "law."

[0127] The server identifies the appropriate field of expertise (in this case, "law") based on the extracted keywords, and classifies it using a machine learning model (e.g., sklearn classifier).

[0128] 3. Querying the database of expert knowledge and expert AI:

[0129] The server forwards the request to the generative AI model in the specified domain of expertise. Specifically, it sends the request as an API request to the generative AI model.

[0130] The generative AI model refers to a knowledge database in that field and generates an appropriate answer, such as "Law is a system for defining social rules and norms."

[0131] The generative AI model sends the generated answer back to the server.

[0132] 4. Return of Response:

[0133] The server formats the answer received from the generative AI model and sends it to the user's device, for example by converting it into HTML format.

[0134] The device displays the response received from the server to the user, for example, on the browser or app screen.

[0135] Specific examples

[0136] Example 1: Medical enquiry

[0137] The user types "Teach me about medicine" into the terminal.

[0138] The terminal sends this request to the server.

[0139] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0140] The server forwards the request to a generative AI model in the medical field.

[0141] The generative AI model generates the answer "Medicine is the scientific study of human health and disease" and sends it back to the server.

[0142] The server then sends the answer received from the generative AI model back to the device.

[0143] The terminal displays the answer to the user.

[0144] Example 2: Legal enquiry

[0145] A user types into a terminal, "Tell me about the law."

[0146] The terminal sends this request to the server.

[0147] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0148] The server forwards the request to a generative AI model in the legal domain.

[0149] The generative AI model generates the answer, "Law is a system for defining social rules and norms," ​​and sends it back to the server.

[0150] The server then sends the answer received from the generative AI model back to the device.

[0151] The terminal displays the answer to the user.

[0152] In this way, the present invention can quickly provide accurate information based on knowledge in each specialized field in response to a user request.

[0153] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0154] Step 1:

[0155] The user inputs a request into the terminal. For example, the user might input "Tell me about the law." This request is captured as text data. The terminal captures this text data and formats it for further processing. The formatted data is converted to JSON format and prepared for sending to the server.

[0156] Step 2:

[0157] The terminal sends the formatted request data to the server. Specifically, it is sent as an HTTP POST request. Here, the input is the formatted text data, and the output is the request data received by the server.

[0158] Step 3:

[0159] The server parses the request data received from the terminal. The server uses natural language processing software (e.g., spaCy) to tokenize the request text, tag it with parts of speech, and analyze dependencies. At this point, the input is the received text data, and the output is the syntax information of the parsed request.

[0160] Step 4:

[0161] The server uses the TF-IDF algorithm to extract important keywords from the request text. For example, the keyword "law" is extracted. The input is the parsed text data, and the output is the extracted keywords.

[0162] Step 5:

[0163] The server uses a machine learning model to identify the appropriate field of expertise based on the extracted keywords. For example, the keyword "law" identifies the field of "law." The machine learning model is pre-trained using training data from multiple fields. The input is the extracted keywords, and the output is the identified field of expertise.

[0164] Step 6:

[0165] The server forwards the request to the generative AI model corresponding to the identified area of ​​expertise. This request includes the user's question. For example, a request such as "Teach me about law" is sent to a generative AI model specializing in law. The input is the identified area of ​​expertise and the request, and the output is the request sent to the generative AI model.

[0166] Step 7:

[0167] The generative AI model consults a database of expert knowledge and generates an answer based on the request, for example, "Law is a system for defining social rules and norms." The input is the request sent to the generative AI model, and the output is the generated answer.

[0168] Step 8:

[0169] The generative AI model returns the generated answer to the server. The server formats the received answer and prepares it for sending back to the user. Specifically, it converts it into HTML format, etc. The input is the generated answer, and the output is the formatted answer.

[0170] Step 9:

[0171] The server sends the formatted response to the terminal. Specifically, it is sent as an HTTP response. The input is the formatted response data, and the output is the response data received by the terminal.

[0172] Step 10:

[0173] The device displays the answer received from the server to the user. Specifically, it displays it on the user interface of the browser or app. The input is the received answer data, and the output is the answer displayed to the user.

[0174] The specific actions and data flow performed at each step enable rapid and accurate provision of information in response to user requests.

[0175] (Application example 1)

[0176] 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."

[0177] In conventional systems, when users sought specialized information, the process of using multiple specialized knowledge databases and AI was cumbersome, making it difficult to provide fast and accurate information. Furthermore, when obtaining specialized knowledge related to a specific field, there was a lack of a way to make appropriate inquiries to specialized AI. This led to the problem that customers in physical stores could not immediately obtain detailed information about specific products or services.

[0178] 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.

[0179] In this invention, the server includes means for receiving a request from a user, means for analyzing the received request and identifying an appropriate specialty based on keywords, means for forwarding the request to a specialized AI having a specialized knowledge database corresponding to the identified specialty, means for returning an answer obtained from the specialized AI to the user, means for associating the requested information with a specific area and providing specialized knowledge in the specific area, and means for providing the answer to the user using a mobile terminal, thereby enabling users to quickly and accurately obtain detailed specialized information about specific products and services in physical stores.

[0180] The "means for receiving a request from a user" is an interface device that inputs questions or requests entered by a user into the system.

[0181] "Means for analyzing the content of the received request and identifying the appropriate specialty based on keywords" refers to the process of analyzing the user's request using natural language processing technology, extracting important keywords, and determining the specialty to consult.

[0182] "Means for forwarding requests to specialized AI with a database of specialized knowledge corresponding to the identified area of ​​expertise" refers to the process of sending a user's request to an AI system that has a database of relevant specialized knowledge based on the area of ​​expertise identified as a result of the analysis.

[0183] "Means for returning answers obtained from specialized AI to the user" refers to a communication means for conveying answers generated by specialized AI to the user.

[0184] "Means of relating requested information to a specific area and providing expertise in that specific area" refers to the process of analyzing the content of a user's request and providing information specific to related physical stores and areas.

[0185] "Means for providing answers to users using mobile devices" refers to a system for displaying answers from specialized AI to users via portable electronic devices such as smartphones and tablets.

[0186] To implement this invention, it is necessary to build a system that receives user requests using a smartphone application, analyzes them, and queries the appropriate AI specialist. This system includes the following elements:

[0187] Hardware and Software Overview

[0188] User device (smartphone): Used as a device for users to input questions.

[0189] Server: Analyzes requests, identifies areas of expertise, queries specialized AI, and returns answers.

[0190] Specialized AI server: A server on which AI with knowledge in each specialized field runs.

[0191] The application running on the device functions as follows:

[0192] User request received

[0193] The user launches the smartphone app and inputs the information or question they want to know, for example, by entering a prompt such as "Please tell me about the nutritional value of this food."

[0194] Request analysis and expertise identification

[0195] The server analyzes the received request using natural language processing technology (e.g., various natural language processing libraries and frameworks). It extracts important keywords from the request and identifies the appropriate field of expertise. For example, if the keyword "nutrition" is extracted, it determines that the request is related to "nutrition."

[0196] Specialized AI Inquiry

[0197] The user's request is forwarded to the specialized AI, which has a database of specialized knowledge corresponding to the specified specialized field. The server sends an inquiry to the specialized AI server using an HTTP request or the like.

[0198] Obtaining the answer and returning it to the user

[0199] The specialized AI generates an appropriate answer from its own knowledge base based on the received request. For example, it may provide information such as, "This food contains 200 kcal per 100 g and is rich in vitamins A and C." The server then sends the answer obtained from the specialized AI back to the user's smartphone and displays it.

[0200] Examples of concrete examples and prompts

[0201] As a concrete example, consider the case where a user types in "I would like to know more about the material of this furniture." In this case, the server extracts the keyword "furniture" and queries a specialized AI that specializes in furniture materials. The specialized AI then generates and returns an answer such as "This furniture is made of oak, which is durable and beautiful."

[0202] Example prompt sentence:

[0203] "Please tell me about the nutritional value of this food."

[0204] "I'd like to know more about the materials used in this furniture."

[0205] In this way, customers can quickly and accurately obtain detailed, specialized information about specific products and services even in physical stores.

[0206] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0207] Step 1:

[0208] User request received

[0209] The user launches the application on their device and inputs the information or question they want to know. The input data is sent to the server via the smartphone application. This input is a prompt sentence, such as "Please tell me about the nutritional value of this food."

[0210] Step 2:

[0211] Sending the request

[0212] The terminal receives the user's request and sends it to the server, which receives the request. The input here is the user's request, and the output is the data sent to the server.

[0213] Step 3:

[0214] Request Analysis

[0215] The server analyzes the received request using natural language processing technology. Specifically, it uses an NLP library (e.g., spaCy or NLTK) to extract important keywords from the request. The input is the user's request data, and the output is the extracted keywords.

[0216] Step 4:

[0217] Specialized field identification

[0218] The server identifies the appropriate specialty based on the keywords extracted in step 3. For example, if the keyword "nutrition" is included, it is associated with the "nutrition" specialty. The input is the extracted keywords, and the output is the identified specialty.

[0219] Step 5:

[0220] Inquiry to specialized AI

[0221] The server forwards the request to a specialized AI server that has an expert knowledge database corresponding to the specified specialized field. The user's request is sent to the specialized AI using an HTTP request, etc. The input is the specified specialized field and the user request, and the output is the request sent to the specialized AI.

[0222] Step 6:

[0223] Answer generation by specialized AI

[0224] Specialized AI generates appropriate answers from its own knowledge base based on the received request. For example, a nutrition AI might provide information such as "This food contains 200 kcal per 100 g and is rich in vitamins A and C." The input is the user request and the specialized knowledge base, and the output is the generated answer.

[0225] Step 7:

[0226] Returning the answer to the server

[0227] The specialized AI sends the generated answer back to the server. The server receives the answer data from the specialized AI. The input is the answer from the specialized AI, and the output is the data received by the server.

[0228] Step 8:

[0229] Returning the answer to the user

[0230] The server sends the received answer back to the device and displays it to the user. The user can view the answer from the specialized AI through a smartphone application. The input is the answer data from the specialized AI, and the output is displayed on the user's device.

[0231] 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.

[0232] The present invention is a system that integrates multiple specialized AIs and queries AIs in the appropriate specialized field in response to a user request, combined with an emotion engine that recognizes the user's emotions. This system includes the following means.

[0233] Program processing

[0234] 1. Request receiving method:

[0235] The terminal receives a request from the user as input. For example, the user inputs a question into the terminal such as "Tell me about medicine."

[0236] The terminal processes this user request and prepares it for transmission to the server.

[0237] 2. Emotion recognition means:

[0238] The terminal includes an emotion engine for analyzing the user's emotion from text, voice, or image data contained in the user's request.

[0239] The emotion engine recognizes the user's emotions (e.g., joy, anxiety, anger, etc.) and sends this information to the server.

[0240] 3. Request Analysis and Expertise Identification Methods:

[0241] The server analyzes the request received from the device and extracts important keywords from the question, for example, checking whether the keyword "medicine" is included.

[0242] The server then identifies the appropriate specialty based on the extracted keywords, in this case the "medical" specialty since "medicine" is included.

[0243] 4. Expert knowledge database and expert AI:

[0244] The server queries a database with specialized AIs corresponding to the identified specialty, in this case, a specialized AI specialized in the medical field.

[0245] The specialized AI analyzes the request against its knowledge base and generates an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[0246] 5. Emotion-based response adjustment measures:

[0247] The server adjusts the tone and content of the response to the request based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling "anxious," the server will respond in a gentler tone.

[0248] 6. Response return method:

[0249] The server prepares the answer received from the specialized AI to be sent back to the user's device.

[0250] The server generates an HTTP response containing the appropriately tailored answer and sends it to the terminal.

[0251] 7. Answer display method:

[0252] The terminal analyzes the HTTP response received from the server and extracts the response content.

[0253] The terminal displays the extracted answers to the user.

[0254] Specific examples

[0255] Example 1: Medical Inquiry (User is Anxious)

[0256] The user types "Teach me about medicine" into the terminal.

[0257] The device passes this request to the emotion engine, which recognizes the user's emotion as "anxiety."

[0258] The terminal transmits the user's request and emotion information to the server.

[0259] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0260] The server forwards the request to a specialized AI in the medical field.

[0261] Specialized AI generates appropriate answers based on your request.

[0262] The server adjusts the tone and content of the response based on the user's emotional information, and reconstructs the response in gentler language to ease the user's anxiety.

[0263] The server returns the reconstructed response to the terminal.

[0264] The device displays the answer to the user, providing a sense of security.

[0265] Example 2: Legal enquiry (if the user is angry)

[0266] A user types into a terminal, "Tell me about the law."

[0267] The device passes this request to the emotion engine, which recognizes the user's emotion as "anger."

[0268] The terminal transmits the user's request and emotion information to the server.

[0269] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0270] The server forwards the request to a specialized AI in the legal field.

[0271] Specialized AI generates appropriate answers based on your request.

[0272] The server reconstructs a response that encourages calmness based on the user's emotional information.

[0273] The server returns the reconstructed response to the terminal.

[0274] The device displays the answer to the user and encourages them to remain calm.

[0275] In this way, the present invention can provide optimal information to a user by recognizing the user's emotions and adjusting responses based on those emotions.

[0276] The processing flow will be explained below.

[0277] Step 1:

[0278] The user types "Teach me about medicine" into the terminal. When the user presses the send button, the terminal receives this request.

[0279] Step 2:

[0280] The device passes the text included in the user's request to the emotion engine, which analyzes the text and recognizes the user's emotion. For example, it detects from the text that the user is feeling "anxiety."

[0281] Step 3:

[0282] The device generates an HTTP request by combining the user's request and the recognized emotion information, and sends it to the server.

[0283] Step 4:

[0284] The server receives the HTTP request, analyzes the request body, extracts the user's question, and extracts the keyword "medicine."

[0285] Step 5:

[0286] The server identifies the appropriate specialty based on the extracted keywords. In this case, since the keyword "medicine" is included, it identifies the "medical care" specialty.

[0287] Step 6:

[0288] The server selects a specialized AI corresponding to the specified specialty field. In this case, it selects a specialized AI specialized in the medical field.

[0289] Step 7:

[0290] To forward the request to the selected specialized AI, the server calls the medical_ai query method and passes the request content.

[0291] Step 8:

[0292] The specialized AI receives the request and uses its knowledge base to generate an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[0293] Step 9:

[0294] The server takes into account the user's emotional information based on the answers received from the specialized AI. For example, if the user is feeling "anxiety," it will adjust the tone and content of the answer to make it more friendly and reassuring.

[0295] Step 10:

[0296] The server prepares an appropriately tailored answer to send back to the user's device, generating an HTTP response and including the answer.

[0297] Step 11:

[0298] The server sends an HTTP response to the device.

[0299] Step 12:

[0300] The terminal analyzes the HTTP response received from the server, extracts the answer, and displays the extracted answer to the user.

[0301] Step 13:

[0302] The user can check the answer displayed on the device and obtain information about their question. In addition, the answer is sensitive to the user's feelings, so the user feels reassured and satisfied.

[0303] Example 2

[0304] 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."

[0305] Conventional systems were able to identify the appropriate area of ​​expertise for user requests and generate answers using specialized AI, but they were unable to adjust the answers to take the user's emotions into account, making it difficult to provide optimal information according to the user's emotional state.

[0306] 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 receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty field based on keywords, means for transferring the request to a specialized AI having an expert knowledge database corresponding to the identified specialty field, means for returning an answer obtained from the specialized AI to the user, means for recognizing the user's emotion, and means for adjusting the tone and content of the answer based on the recognized emotion. This makes it possible to provide optimal information according to the user's emotional state.

[0307] The "means for receiving requests from a user" refers to a device or software that includes a process for receiving inquiries or instructions given by a user through a terminal as input data and converting them into a form that can be sent to a server.

[0308] The "means for analyzing the received request content and identifying the appropriate specialty based on keywords" refers to a device or software that analyzes the text data of the request, extracts important words and phrases, and identifies the relevant specialty based on those words and phrases.

[0309] A "means for forwarding requests to specialized AI with a database of specialized knowledge corresponding to an identified area of ​​expertise" is a device or software that includes a process for identifying an area of ​​expertise and then sending the request data to an AI system with specialized knowledge corresponding to that area.

[0310] A "means for returning answers obtained from specialized AI to the user" is a device or software that includes a process for receiving answers generated by specialized AI, converting them into an appropriate format, and returning them to the user.

[0311] "Means for recognizing user emotions" refers to devices or software that include technologies and processes for analyzing and evaluating emotions from user request data and identifying the user's emotional state.

[0312] "Means for adjusting the tone and content of a response based on the recognized emotion" refers to a device or software that includes a process for appropriately adjusting the tone and expression of a generated response sentence in accordance with the recognized emotion of the user.

[0313] The present invention is a system for generating an appropriate response to a request from a user that takes into consideration the user's feelings. Specific means for implementing this system will be described below.

[0314] 1. Request Receiving Method

[0315] The terminal receives a request input by a user. Specifically, the terminal captures the user's request in text format and prepares it for transmission to the server. For example, the user inputs "Teach me about medicine" into the terminal.

[0316] 2. Emotion recognition means

[0317] The device passes the user's request text to an emotion recognition engine, which uses natural language processing (NLP) techniques to analyze the user's emotions from the text. The emotion engine identifies emotions such as anxiety, joy, and anger. Publicly available NLP libraries and APIs can be used as emotion recognition engines.

[0318] 3. Request Analysis and Specialty Identification Methods

[0319] The server analyzes requests received from the device and extracts important keywords from them. For example, it detects the keyword "medicine." Text mining technology is used to analyze the request. Based on the analyzed keywords, the server identifies the appropriate specialty. For example, if "medicine" is included, it references a database to identify the corresponding "medical" specialty.

[0320] 4. Expert knowledge database and expert AI

[0321] The server queries a database containing specialized AI corresponding to the identified specialty. In the case of the medical field, a specialized AI with medical knowledge is selected. This specialized AI analyzes the request content based on a pre-built knowledge base and generates an appropriate answer. The specialized AI used can be based on a publicly available generative AI model.

[0322] 5. Emotion-Based Response Modification

[0323] The server adjusts the tone and content of the generated response based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, the server adjusts the response to use gentler language. For example, the server changes the response to something like, "Don't worry. Medicine is the scientific study of human health and disease."

[0324] 6. Response return method

[0325] The server prepares the answer received from the expert AI for return to the user's device and generates an HTTP response, which includes the adjusted answer and is sent to the device.

[0326] 7. Answer display means

[0327] The device analyzes the HTTP response received from the server and extracts the answer. It then displays the answer to the user using a user interface (UI). Specifically, the answer is displayed in a chat window on the screen.

[0328] Specific examples

[0329] Example 1: Medical Inquiry (User is Anxious)

[0330] The user types "Teach me about medicine" into the terminal.

[0331] The device passes this request to the emotion engine, which recognizes the user's emotion as "anxiety."

[0332] The terminal transmits the user's request and emotion information to the server.

[0333] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0334] The server forwards the request to a specialized AI in the medical field.

[0335] Specialized AI generates appropriate answers based on your request.

[0336] The server adjusts the tone and content of the response based on the user's emotional information, reframing the response in a gentler way to ease the "anxiety": "Don't worry. Medicine is the scientific study of human health and disease."

[0337] The server returns the reconstructed response to the terminal.

[0338] The device displays the answer to the user, providing a sense of security.

[0339] Example 2: Legal enquiry (if the user is angry)

[0340] A user types into a terminal, "Tell me about the law."

[0341] The device passes this request to the emotion engine, which recognizes the user's emotion as "anger."

[0342] The terminal transmits the user's request and emotion information to the server.

[0343] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0344] The server forwards the request to a specialized AI in the legal field.

[0345] Specialized AI generates appropriate answers based on your request.

[0346] The server reconstructs a response that encourages calmness based on the user's emotional information: "Please stay calm. Laws are an important system for determining social rules."

[0347] The server returns the reconstructed response to the terminal.

[0348] The device displays the answer to the user and encourages them to remain calm.

[0349] In this way, the present invention can provide optimal information to a user by recognizing the user's emotions and adjusting responses based on those emotions.

[0350] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0351] Step 1: How to receive requests

[0352] A user types text into a terminal, for example, "Teach me about medicine."

[0353] Input: The request text entered by the user.

[0354] The terminal receives a request input by a user and stores it in memory as text data.

[0355] Output: The data where the input text is saved.

[0356] Specific behavior: The device displays a confirmation pop-up message "Processing your request..." to the user.

[0357] Step 2: Emotion recognition tools

[0358] The device passes the saved request text to the emotion engine.

[0359] Input: The saved request text.

[0360] The emotion engine uses NLP techniques to analyze user emotions from text, specifically by using publicly available NLP libraries to analyze text and determine emotional states.

[0361] Output: Emotional information (e.g., anxiety, joy, anger, etc.).

[0362] Specific behavior: While the device is performing emotion recognition, it displays a status bar and tells the user, "Analyzing emotions..."

[0363] Step 3: Request analysis and domain identification measures

[0364] The server analyzes the request text and emotion information received from the terminal.

[0365] Input: Request text and emotion information.

[0366] The server uses text mining technology to extract important keywords from the request, for example, the keyword "medicine."

[0367] Output: Identified keywords (e.g., "medicine").

[0368] Specific behavior: The server records the progress in a log file while analyzing keywords. It writes "Keyword 'medicine' found" in the log.

[0369] Step 4: Expert knowledge database and expert AI

[0370] The server identifies appropriate specialties based on the identified keywords.

[0371] Input: Identified keyword (e.g., medicine).

[0372] The server selects a specialized AI from the database that corresponds to the specified field. In the case of the medical field, a specialized AI with medical knowledge is selected.

[0373] Output: Selected specialized AI.

[0374] Specific behavior: The server logs the progress information as "specialty 'Medical' identified."

[0375] Step 5: Route the request to our expert AI

[0376] The server forwards the request to the selected specialized AI.

[0377] Input: Request text and selected expert AI.

[0378] The specialized AI analyzes the request and uses its knowledge base to generate an answer, such as "Medicine is the scientific study of human health and disease."

[0379] Output: The generated answer.

[0380] Specific operation: The specialized AI records the detailed step-by-step process of generating an answer in a log file. "Answer generation: 'Medicine is concerned with the health and disease of the human body...'"

[0381] Step 6: Emotion-Based Response Modification Measures

[0382] The server adjusts the tone and content of the generated response based on the user's emotional information provided by the emotion engine.

[0383] Input: Generated answers and sentiment information.

[0384] The server uses an emotion-adjustment algorithm to adjust the response, for example adding a gentler tone to the sentence, "Don't worry. Medicine is the scientific study of human health and disease."

[0385] Output: The adjusted answer.

[0386] Specific behavior: The server adjusts the response based on the user's emotion and logs the emotion adjustment point. "Change the tone to a gentler tone depending on the user's anxiety."

[0387] Step 7: Response return method

[0388] The server generates the adjusted answer as an HTTP response.

[0389] Input: Adjusted answer text.

[0390] The server generates an HTTP response including appropriate header information and sends it to the terminal.

[0391] Output: The generated HTTP response.

[0392] Specific behavior: When the server generates a response, it measures the response time and records it in the log. "Response time: 120ms"

[0393] Step 8: Displaying the Answers

[0394] The terminal analyzes the HTTP response received from the server and displays the extracted answer content to the user.

[0395] Input: HTTP response.

[0396] The device will then display the answer using a user interface (UI), for example, in a chat window on the screen: "Don't worry. Medicine is the scientific study of human health and disease."

[0397] Output: The answer displayed to the user.

[0398] What it does: When the device displays the final answer on the screen, it displays a highlighted message to the user: "Answer received!"

[0399] (Application example 2)

[0400] 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."

[0401] Conventional systems simply provide uniform responses without considering the user's emotions, making it difficult to provide appropriate information according to the emotional state of each individual user. Furthermore, security services, in particular, are required to respond quickly and appropriately to users' anxiety and anger, and it is important to respond in a way that is sensitive to the user's emotions. This will improve user satisfaction and enhance trust in security.

[0402] 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.

[0403] In this invention, the server includes means for receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty based on keywords, means for transferring the request to a specialized AI having an expert knowledge database corresponding to the identified specialty, means for returning an answer obtained from the specialized AI to the user, means for recognizing the user's emotion from text, voice, or image data included in the user's request, and means for adjusting the answer based on the recognized emotion, thereby making it possible to recognize the user's emotion and provide an appropriate and personalized answer based on that emotion.

[0404] A "request receiving means" is a device or system that has the function of receiving a request from a user.

[0405] The "request analysis means" is a system that has the function of analyzing the content of a received request and extracting important keywords and contexts.

[0406] The "specialty field identification means" is a system having a function of identifying an appropriate specialty field based on the keywords extracted by the request analysis means.

[0407] An "expert knowledge database" is a database that accumulates knowledge and information in a specific specialized field, and is the knowledge base that specialized AI references.

[0408] "Specialized AI" is artificial intelligence that has been trained to specialize in a specific field of expertise, and is an algorithm that uses a database of specialized knowledge to generate answers to user requests.

[0409] The "answer return means" is a system that has the function of returning the answers obtained from the specialized AI to the user.

[0410] The "emotion recognition means" is a system that has the function of analyzing the user's emotions from the text, voice, or image data included in the user's request.

[0411] The "answer adjustment means" is a system that has the function of adjusting the tone and content of the answer based on the emotional information recognized by the emotion recognition means.

[0412] In this invention, a system is constructed that recognizes a user's emotions, queries an AI in an appropriate field of expertise, and provides an answer according to the emotion. Specific embodiments are shown below.

[0413] composition

[0414] The system includes the following hardware and software:

[0415] 1. Terminal: A device that receives requests from users and recognizes their emotions. This can be a smartphone, smart glasses, or a head-mounted display.

[0416] 2. Emotion Engine: Software that analyzes emotions from text and voice data contained in user requests, using natural language processing libraries such as TextBlob.

[0417] 3. Server: The back-end system that analyzes the request, identifies the appropriate area of ​​expertise, and queries the specialized AI.

[0418] 4. Specialized AI: AI trained to specialize in a specific field, referencing a database of specialized knowledge, such as the medical field or security field.

[0419] 5. Answer Adjustment Module: A software module that adjusts answers based on emotion recognition results.

[0420] Processing flow

[0421] 1. The device receives a request from the user in text or voice format, for example, "I think I have a virus on my computer. What should I do?"

[0422] 2. The emotion engine installed on the device analyzes the text and voice data contained in the request and classifies the user's emotions into categories such as "joy," "anxiety," and "anger."

[0423] 3. This emotion information and the user's request are sent to the server.

[0424] 4. The server analyzes the request and extracts relevant keywords, using TextBlob or other natural language processing tools.

[0425] 5. Identify appropriate specializations (e.g., "antivirus," "medical") based on the extracted keywords.

[0426] 6. The request is forwarded to a specialized AI in the identified area of ​​expertise, which then consults a dedicated database to generate an appropriate response.

[0427] 7. Adjust responses from specialized AI based on emotion, for example, reframing a response to a user who is feeling anxious in a more reassuring tone or context.

[0428] 8. The reconstructed answer is sent back to the terminal and displayed to the user.

[0429] Specific examples

[0430] Example 1

[0431] If a user types into their device, "My PC has been running slow lately. Is it because of a virus?":

[0432] The emotion engine recognizes this as "anxiety."

[0433] Extract "virus" and "PC" using keyword extraction.

[0434] Make an inquiry to the "antivirus specialist AI."

[0435] Tailor your answers to reduce anxiety based on your emotions.

[0436] User: "My PC has been running slow lately. Could it be because of a virus?"

[0437] Terminal: "A virus may be the cause of your PC's slow performance. We recommend that you perform virus scans regularly. If you are concerned, start with some simple measures. Please feel free to contact us."

[0438] By implementing this invention, it becomes possible to provide appropriate information that is in tune with the user's emotions, thereby improving user satisfaction and trust.

[0439] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0440] Step 1:

[0441] The terminal receives a request from a user.

[0442] Input: User text or voice input (e.g., "I think I have a virus on my computer. What should I do?")

[0443] Output: User request data

[0444] Specific operation: The device receives user requests through input devices such as a microphone or keyboard. In the case of voice input, it uses speech recognition software to convert the request into text.

[0445] Step 2:

[0446] The device's emotion engine analyzes the received request and recognizes the user's emotions.

[0447] Input: User request data

[0448] Output: User's emotion data (e.g., "anxiety")

[0449] What it does: The emotion engine (e.g., the TextBlob library) performs text analysis and calculates polarity to identify emotions such as "joy," "anxiety," or "anger."

[0450] Step 3:

[0451] The terminal transmits the user's emotion information and request data to the server.

[0452] Input: User request data, emotion data

[0453] Output: Request sent to the server and emotion information

[0454] Specific operation: The device sends request data and emotion data to the server via a network connection using a protocol such as HTTP.

[0455] Step 4:

[0456] The server analyzes the received request and extracts important keywords.

[0457] Input: Request data

[0458] Output: Extracted keywords (e.g. "virus", "PC")

[0459] Specific operation: The server uses a natural language processing tool (e.g., TextBlob) to extract important keywords from the request data.

[0460] Step 5:

[0461] The server identifies the appropriate specialty based on the extracted keywords.

[0462] Input: Extracted keywords

[0463] Output: Identified specialty (e.g. "Antivirus")

[0464] What it does: The server matches the keywords against a database and ruleset to determine the appropriate specialty (e.g., "antivirus").

[0465] Step 6:

[0466] The server forwards the request to the appropriate specialized AI.

[0467] Input: User request data, identified expertise

[0468] Output: Answer data from specialized AI

[0469] Specific operation: The server calls the specialized AI's API and sends the request data in the appropriate format. The specialized AI then references a dedicated database and generates an answer.

[0470] Step 7:

[0471] The server tailors the response based on the sentiment information.

[0472] Input: Answer data from specialized AI, emotional data

[0473] Output: Tailored response data (e.g., tone and content to reduce anxiety)

[0474] Specific operation: The server takes into account the emotional data and customizes the response, such as changing it to a gentler tone.

[0475] Step 8:

[0476] The server sends a tailored response back to the terminal.

[0477] Input: Adjusted response data

[0478] Output: Send the answer to the terminal

[0479] Specific operation: The server sends the adjusted response data to the terminal using the HTTP protocol or the like.

[0480] Step 9:

[0481] The terminal displays the received response to the user.

[0482] Input: Response data from the server

[0483] Output: The answer that is displayed to the user

[0484] Specific operation: The device converts the response data into a format that is easy to display and presents it to the user on the screen or via audio.

[0485] The above steps realize a system that provides appropriate answers that are in tune with the user's emotions.

[0486] 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.

[0487] 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.

[0488] 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.

[0489] [Second embodiment]

[0490] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0491] 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.

[0492] 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).

[0493] 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.

[0494] 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.

[0495] 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).

[0496] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0497] 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.

[0498] 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.

[0499] 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.

[0500] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0501] 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."

[0502] The present invention provides a system that integrates multiple specialized AIs and queries an AI with an appropriate specialty in response to a user request. This system includes the following means.

[0503] Program processing

[0504] 1. Request receiving method:

[0505] The terminal receives a request from the user as input. For example, the user inputs a question into the terminal such as "Tell me about medicine."

[0506] The terminal processes this user request and prepares it for transmission to the server.

[0507] 2. Request Analysis and Expertise Identification Methods:

[0508] The server analyzes the request received from the device. The server analyzes the request content and extracts important keywords from it. For example, it checks whether the keyword "medicine" is included.

[0509] The server then identifies the appropriate specialty based on the extracted keywords, in this case the "medical" specialty because "medicine" is included.

[0510] 3. Expert knowledge database and expert AI:

[0511] The server queries a database with specialized AIs corresponding to the identified specialty, in this case, a specialized AI specialized in the medical field.

[0512] Specialized AI refers to a database containing a large amount of medical-related data and information and generates appropriate answers to user questions, such as "Medicine is the scientific study of human health and disease."

[0513] 4. Response return method:

[0514] The server sends the answer received from the specialized AI back to the user's device.

[0515] The terminal displays the answer received from the server to the user.

[0516] Specific examples

[0517] Example 1: Medical enquiry

[0518] The user types "Teach me about medicine" into the terminal.

[0519] The terminal sends this request to the server.

[0520] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0521] The server forwards the request to a specialized AI in the medical field.

[0522] Based on the request received, the specialized AI generates an answer from its knowledge base: "Medicine is the scientific study of human health and disease."

[0523] The server then sends the answer received from the specialized AI back to the device.

[0524] The terminal displays the answer to the user.

[0525] Example 2: Legal enquiry

[0526] A user types into a terminal, "Tell me about the law."

[0527] The terminal sends this request to the server.

[0528] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0529] The server forwards the request to a specialized AI in the legal field.

[0530] Based on the request received, the specialized AI generates an answer from its own knowledge base: "Law is a system for establishing social rules and norms."

[0531] The server then sends the answer received from the specialized AI back to the device.

[0532] The terminal displays the answer to the user.

[0533] In this way, the present invention provides accurate information based on knowledge of each specialized field in response to user requests.

[0534] The processing flow will be explained below.

[0535] Step 1:

[0536] The user types "Teach me about medicine" into the terminal. When the user presses the send button, the terminal receives this request.

[0537] Step 2:

[0538] The terminal converts the user's request into an HTTP request and prepares it for transmission to the server.

[0539] Step 3:

[0540] The terminal sends an HTTP request to the server (for example, sends the request using the POST method).

[0541] Step 4:

[0542] The server receives the HTTP request, analyzes the request body, and extracts the user's question.

[0543] Step 5:

[0544] The server analyzes the question and extracts important keywords, for example, checking whether the keyword "medicine" is included.

[0545] Step 6:

[0546] The server identifies the appropriate specialty based on the extracted keywords. In this case, since "medicine" is included, it identifies the "medical" specialty.

[0547] Step 7:

[0548] The server selects a specialized AI corresponding to the specified specialty field. In this case, it selects a specialized AI specialized in the medical field.

[0549] Step 8:

[0550] The server forwards the request to the selected specialized AI. The server calls the medical_ai's query method and passes the request content.

[0551] Step 9:

[0552] The specialized AI receives the request and uses its knowledge base to generate an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[0553] Step 10:

[0554] The specialized AI returns the generated answer to the server, which then prepares the answer received from the specialized AI for sending back to the user's device.

[0555] Step 11:

[0556] The server generates an HTTP response for the user's device, includes the answer, and sends this HTTP response to the device.

[0557] Step 12:

[0558] The terminal analyzes the HTTP response received from the server, extracts the answer, and displays the extracted answer to the user.

[0559] Step 13:

[0560] The user can read the answers displayed on the terminal and obtain the desired information.

[0561] Example 1

[0562] 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."

[0563] In today's information society, users need to quickly and accurately obtain detailed information in various specialized fields. However, in current systems, AIs that provide specialized information in specific fields are distributed across each field, making it difficult for users to obtain appropriate information tailored to their respective fields. Furthermore, due to low accuracy in analyzing request content, it may not be possible to identify the exact specialized field the user is looking for. Therefore, there is a need for a system that can quickly provide appropriate answers to user requests.

[0564] 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.

[0565] In this invention, the server includes means for receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty based on keywords, means for transferring the request to a generative AI model having an expert knowledge database corresponding to the identified specialty, and means for returning an answer obtained from the generative AI model to the user. This makes it possible to integrate AI models specialized in different specialty areas and provide information on the appropriate specialty in response to a user request quickly and accurately.

[0566] A "user" is a person who requests information from the system.

[0567] A "request" is an inquiry that a user makes to the system requesting information or services.

[0568] The "means for receiving" is a mechanism or method for capturing requests from users and incorporating them into the system.

[0569] The "analyzing means" refers to a technique or method for analyzing a received request and extracting keywords contained therein.

[0570] "Keywords" are important words or phrases in a request that identify an area of ​​expertise.

[0571] A "discipline" is an area in which information or knowledge falls into a particular category or field.

[0572] A "generative AI model" is an artificial intelligence model that generates and analyzes information based on large amounts of data.

[0573] A "transfer means" is a method or technique for sending a request to a generative AI model in a specified area of ​​expertise.

[0574] A "means for returning" is a mechanism or method for returning answers received from a generative AI model to a user.

[0575] "Network communications" refers to communications technologies that connect computers and devices to send and receive data.

[0576] The present invention is a system that queries AI with an appropriate field of expertise in response to a user request. The system receives the user's request as input, analyzes its content, and identifies the appropriate field of expertise based on keywords. The system then forwards the request to a generative AI model corresponding to the identified field of expertise and returns the generated answer to the user. Specific embodiments of the system are described below.

[0577] Hardware and software used

[0578] 1. Terminal: A device through which a user inputs a request. Examples include smartphones, tablets, and computers.

[0579] 2. Server: A central processing unit responsible for analyzing requests, identifying areas of expertise, forwarding requests to generative AI models, and returning answers. The server has a high-bandwidth network connection and high-performance computing capabilities. For example, it can be a virtual server provided by a cloud service provider (e.g., AWS, Azure).

[0580] 3. Generative AI model: This is an artificial intelligence model that generates information according to a specific field of expertise. For example, large-scale language models such as GPT-3 and BERT can be used.

[0581] 4. Natural language processing software: Used to analyze requests and extract keywords. Examples include Python's NLTK and spaCy.

[0582] System Operation Overview

[0583] 1. Request received:

[0584] The user types a request into the terminal, for example, "Tell me about the law."

[0585] The device receives the request and converts it to a format for sending to the server. Specifically, it converts the text to JSON format and sends it to the server as an HTTP request.

[0586] 2. Request analysis and expertise identification:

[0587] The server analyzes the incoming request, using natural language processing software (e.g., spaCy) to tokenize the request, tag it with parts of speech, and analyze dependencies.

[0588] The server extracts important keywords using a keyword extraction algorithm (e.g., TF-IDF). For example, it detects the keyword "law."

[0589] The server identifies the appropriate field of expertise (in this case, "law") based on the extracted keywords, and classifies it using a machine learning model (e.g., sklearn classifier).

[0590] 3. Querying the database of expert knowledge and expert AI:

[0591] The server forwards the request to the generative AI model in the specified domain of expertise. Specifically, it sends the request as an API request to the generative AI model.

[0592] The generative AI model refers to a knowledge database in that field and generates an appropriate answer, such as "Law is a system for defining social rules and norms."

[0593] The generative AI model sends the generated answer back to the server.

[0594] 4. Return of Response:

[0595] The server formats the answer received from the generative AI model and sends it to the user's device, for example by converting it into HTML format.

[0596] The device displays the response received from the server to the user, for example, on the browser or app screen.

[0597] Specific examples

[0598] Example 1: Medical enquiry

[0599] The user types "Teach me about medicine" into the terminal.

[0600] The terminal sends this request to the server.

[0601] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0602] The server forwards the request to a generative AI model in the medical field.

[0603] The generative AI model generates the answer "Medicine is the scientific study of human health and disease" and sends it back to the server.

[0604] The server then sends the answer received from the generative AI model back to the device.

[0605] The terminal displays the answer to the user.

[0606] Example 2: Legal enquiry

[0607] A user types into a terminal, "Tell me about the law."

[0608] The terminal sends this request to the server.

[0609] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0610] The server forwards the request to a generative AI model in the legal domain.

[0611] The generative AI model generates the answer, "Law is a system for defining social rules and norms," ​​and sends it back to the server.

[0612] The server then sends the answer received from the generative AI model back to the device.

[0613] The terminal displays the answer to the user.

[0614] In this way, the present invention can quickly provide accurate information based on knowledge in each specialized field in response to a user request.

[0615] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0616] Step 1:

[0617] The user inputs a request into the terminal. For example, the user might input "Tell me about the law." This request is captured as text data. The terminal captures this text data and formats it for further processing. The formatted data is converted to JSON format and prepared for sending to the server.

[0618] Step 2:

[0619] The terminal sends the formatted request data to the server. Specifically, it is sent as an HTTP POST request. Here, the input is the formatted text data, and the output is the request data received by the server.

[0620] Step 3:

[0621] The server parses the request data received from the terminal. The server uses natural language processing software (e.g., spaCy) to tokenize the request text, tag it with parts of speech, and analyze dependencies. At this point, the input is the received text data, and the output is the syntax information of the parsed request.

[0622] Step 4:

[0623] The server uses the TF-IDF algorithm to extract important keywords from the request text. For example, the keyword "law" is extracted. The input is the parsed text data, and the output is the extracted keywords.

[0624] Step 5:

[0625] The server uses a machine learning model to identify the appropriate field of expertise based on the extracted keywords. For example, the keyword "law" identifies the field of "law." The machine learning model is pre-trained using training data from multiple fields. The input is the extracted keywords, and the output is the identified field of expertise.

[0626] Step 6:

[0627] The server forwards the request to the generative AI model corresponding to the identified area of ​​expertise. This request includes the user's question. For example, a request such as "Teach me about law" is sent to a generative AI model specializing in law. The input is the identified area of ​​expertise and the request, and the output is the request sent to the generative AI model.

[0628] Step 7:

[0629] The generative AI model consults a database of expert knowledge and generates an answer based on the request, for example, "Law is a system for defining social rules and norms." The input is the request sent to the generative AI model, and the output is the generated answer.

[0630] Step 8:

[0631] The generative AI model returns the generated answer to the server. The server formats the received answer and prepares it for sending back to the user. Specifically, it converts it into HTML format, etc. The input is the generated answer, and the output is the formatted answer.

[0632] Step 9:

[0633] The server sends the formatted response to the terminal. Specifically, it is sent as an HTTP response. The input is the formatted response data, and the output is the response data received by the terminal.

[0634] Step 10:

[0635] The device displays the answer received from the server to the user. Specifically, it displays it on the user interface of the browser or app. The input is the received answer data, and the output is the answer displayed to the user.

[0636] The specific actions and data flow performed at each step enable rapid and accurate provision of information in response to user requests.

[0637] (Application example 1)

[0638] 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."

[0639] In conventional systems, when users sought specialized information, the process of using multiple specialized knowledge databases and AI was cumbersome, making it difficult to provide fast and accurate information. Furthermore, when obtaining specialized knowledge related to a specific field, there was a lack of a way to make appropriate inquiries to specialized AI. This led to the problem that customers in physical stores could not immediately obtain detailed information about specific products or services.

[0640] 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.

[0641] In this invention, the server includes means for receiving a request from a user, means for analyzing the received request and identifying an appropriate specialty based on keywords, means for forwarding the request to a specialized AI having a specialized knowledge database corresponding to the identified specialty, means for returning an answer obtained from the specialized AI to the user, means for associating the requested information with a specific area and providing specialized knowledge in the specific area, and means for providing the answer to the user using a mobile terminal, thereby enabling users to quickly and accurately obtain detailed specialized information about specific products and services in physical stores.

[0642] The "means for receiving a request from a user" is an interface device that inputs questions or requests entered by a user into the system.

[0643] "Means for analyzing the content of the received request and identifying the appropriate specialty based on keywords" refers to the process of analyzing the user's request using natural language processing technology, extracting important keywords, and determining the specialty to consult.

[0644] "Means for forwarding requests to specialized AI with a database of specialized knowledge corresponding to the identified area of ​​expertise" refers to the process of sending a user's request to an AI system that has a database of relevant specialized knowledge based on the area of ​​expertise identified as a result of the analysis.

[0645] "Means for returning answers obtained from specialized AI to the user" refers to a communication means for conveying answers generated by specialized AI to the user.

[0646] "Means of relating requested information to a specific area and providing expertise in that specific area" refers to the process of analyzing the content of a user's request and providing information specific to related physical stores and areas.

[0647] "Means for providing answers to users using mobile devices" refers to a system for displaying answers from specialized AI to users via portable electronic devices such as smartphones and tablets.

[0648] To implement this invention, it is necessary to build a system that receives user requests using a smartphone application, analyzes them, and queries the appropriate AI specialist. This system includes the following elements:

[0649] Hardware and Software Overview

[0650] User device (smartphone): Used as a device for users to input questions.

[0651] Server: Analyzes requests, identifies areas of expertise, queries specialized AI, and returns answers.

[0652] Specialized AI server: A server on which AI with knowledge in each specialized field runs.

[0653] The application running on the device functions as follows:

[0654] User request received

[0655] The user launches the smartphone app and inputs the information or question they want to know, for example, by entering a prompt such as "Please tell me about the nutritional value of this food."

[0656] Request analysis and expertise identification

[0657] The server analyzes the received request using natural language processing technology (e.g., various natural language processing libraries and frameworks). It extracts important keywords from the request and identifies the appropriate field of expertise. For example, if the keyword "nutrition" is extracted, it determines that the request is related to "nutrition."

[0658] Specialized AI Inquiry

[0659] The user's request is forwarded to the specialized AI, which has a database of specialized knowledge corresponding to the specified specialized field. The server sends an inquiry to the specialized AI server using an HTTP request or the like.

[0660] Obtaining the answer and returning it to the user

[0661] The specialized AI generates an appropriate answer from its own knowledge base based on the received request. For example, it may provide information such as, "This food contains 200 kcal per 100 g and is rich in vitamins A and C." The server then sends the answer obtained from the specialized AI back to the user's smartphone and displays it.

[0662] Examples of concrete examples and prompts

[0663] As a concrete example, consider the case where a user types in "I would like to know more about the material of this furniture." In this case, the server extracts the keyword "furniture" and queries a specialized AI that specializes in furniture materials. The specialized AI then generates and returns an answer such as "This furniture is made of oak, which is durable and beautiful."

[0664] Example prompt sentence:

[0665] "Please tell me about the nutritional value of this food."

[0666] "I'd like to know more about the materials used in this furniture."

[0667] In this way, customers can quickly and accurately obtain detailed, specialized information about specific products and services even in physical stores.

[0668] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0669] Step 1:

[0670] User request received

[0671] The user launches the application on their device and inputs the information or question they want to know. The input data is sent to the server via the smartphone application. This input is a prompt sentence, such as "Please tell me about the nutritional value of this food."

[0672] Step 2:

[0673] Sending the request

[0674] The terminal receives the user's request and sends it to the server, which receives the request. The input here is the user's request, and the output is the data sent to the server.

[0675] Step 3:

[0676] Request Analysis

[0677] The server analyzes the received request using natural language processing technology. Specifically, it uses an NLP library (e.g., spaCy or NLTK) to extract important keywords from the request. The input is the user's request data, and the output is the extracted keywords.

[0678] Step 4:

[0679] Specialized field identification

[0680] The server identifies the appropriate specialty based on the keywords extracted in step 3. For example, if the keyword "nutrition" is included, it is associated with the "nutrition" specialty. The input is the extracted keywords, and the output is the identified specialty.

[0681] Step 5:

[0682] Inquiry to specialized AI

[0683] The server forwards the request to a specialized AI server that has an expert knowledge database corresponding to the specified specialized field. The user's request is sent to the specialized AI using an HTTP request, etc. The input is the specified specialized field and the user request, and the output is the request sent to the specialized AI.

[0684] Step 6:

[0685] Answer generation by specialized AI

[0686] Specialized AI generates appropriate answers from its own knowledge base based on the received request. For example, a nutrition AI might provide information such as "This food contains 200 kcal per 100 g and is rich in vitamins A and C." The input is the user request and the specialized knowledge base, and the output is the generated answer.

[0687] Step 7:

[0688] Returning the answer to the server

[0689] The specialized AI sends the generated answer back to the server. The server receives the answer data from the specialized AI. The input is the answer from the specialized AI, and the output is the data received by the server.

[0690] Step 8:

[0691] Returning the answer to the user

[0692] The server sends the received answer back to the device and displays it to the user. The user can view the answer from the specialized AI through a smartphone application. The input is the answer data from the specialized AI, and the output is displayed on the user's device.

[0693] 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.

[0694] The present invention is a system that integrates multiple specialized AIs and queries AIs in the appropriate specialized field in response to a user request, combined with an emotion engine that recognizes the user's emotions. This system includes the following means.

[0695] Program processing

[0696] 1. Request receiving method:

[0697] The terminal receives a request from the user as input. For example, the user inputs a question into the terminal such as "Tell me about medicine."

[0698] The terminal processes this user request and prepares it for transmission to the server.

[0699] 2. Emotion recognition means:

[0700] The terminal includes an emotion engine for analyzing the user's emotion from text, voice, or image data contained in the user's request.

[0701] The emotion engine recognizes the user's emotions (e.g., joy, anxiety, anger, etc.) and sends this information to the server.

[0702] 3. Request Analysis and Expertise Identification Methods:

[0703] The server analyzes the request received from the device and extracts important keywords from the question, for example, checking whether the keyword "medicine" is included.

[0704] The server then identifies the appropriate specialty based on the extracted keywords, in this case the "medical" specialty since "medicine" is included.

[0705] 4. Expert knowledge database and expert AI:

[0706] The server queries a database with specialized AIs corresponding to the identified specialty, in this case, a specialized AI specialized in the medical field.

[0707] The specialized AI analyzes the request against its knowledge base and generates an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[0708] 5. Emotion-based response adjustment measures:

[0709] The server adjusts the tone and content of the response to the request based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling "anxious," the server will respond in a gentler tone.

[0710] 6. Response return method:

[0711] The server prepares the answer received from the specialized AI to be sent back to the user's device.

[0712] The server generates an HTTP response containing the appropriately tailored answer and sends it to the terminal.

[0713] 7. Answer display method:

[0714] The terminal analyzes the HTTP response received from the server and extracts the response content.

[0715] The terminal displays the extracted answers to the user.

[0716] Specific examples

[0717] Example 1: Medical Inquiry (User is Anxious)

[0718] The user types "Teach me about medicine" into the terminal.

[0719] The device passes this request to the emotion engine, which recognizes the user's emotion as "anxiety."

[0720] The terminal transmits the user's request and emotion information to the server.

[0721] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0722] The server forwards the request to a specialized AI in the medical field.

[0723] Specialized AI generates appropriate answers based on your request.

[0724] The server adjusts the tone and content of the response based on the user's emotional information, and reconstructs the response in gentler language to ease the user's anxiety.

[0725] The server returns the reconstructed response to the terminal.

[0726] The device displays the answer to the user, providing a sense of security.

[0727] Example 2: Legal enquiry (if the user is angry)

[0728] A user types into a terminal, "Tell me about the law."

[0729] The device passes this request to the emotion engine, which recognizes the user's emotion as "anger."

[0730] The terminal transmits the user's request and emotion information to the server.

[0731] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0732] The server forwards the request to a specialized AI in the legal field.

[0733] Specialized AI generates appropriate answers based on your request.

[0734] The server reconstructs a response that encourages calmness based on the user's emotional information.

[0735] The server returns the reconstructed response to the terminal.

[0736] The device displays the answer to the user and encourages them to remain calm.

[0737] In this way, the present invention can provide optimal information to a user by recognizing the user's emotions and adjusting responses based on those emotions.

[0738] The processing flow will be explained below.

[0739] Step 1:

[0740] The user types "Teach me about medicine" into the terminal. When the user presses the send button, the terminal receives this request.

[0741] Step 2:

[0742] The device passes the text included in the user's request to the emotion engine, which analyzes the text and recognizes the user's emotion. For example, it detects from the text that the user is feeling "anxiety."

[0743] Step 3:

[0744] The device generates an HTTP request by combining the user's request and the recognized emotion information, and sends it to the server.

[0745] Step 4:

[0746] The server receives the HTTP request, analyzes the request body, extracts the user's question, and extracts the keyword "medicine."

[0747] Step 5:

[0748] The server identifies the appropriate specialty based on the extracted keywords. In this case, since the keyword "medicine" is included, it identifies the "medical care" specialty.

[0749] Step 6:

[0750] The server selects a specialized AI corresponding to the specified specialty field. In this case, it selects a specialized AI specialized in the medical field.

[0751] Step 7:

[0752] To forward the request to the selected specialized AI, the server calls the medical_ai query method and passes the request content.

[0753] Step 8:

[0754] The specialized AI receives the request and uses its knowledge base to generate an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[0755] Step 9:

[0756] The server takes into account the user's emotional information based on the answers received from the specialized AI. For example, if the user is feeling "anxiety," it will adjust the tone and content of the answer to make it more friendly and reassuring.

[0757] Step 10:

[0758] The server prepares an appropriately tailored answer to send back to the user's device, generating an HTTP response and including the answer.

[0759] Step 11:

[0760] The server sends an HTTP response to the device.

[0761] Step 12:

[0762] The terminal analyzes the HTTP response received from the server, extracts the answer, and displays the extracted answer to the user.

[0763] Step 13:

[0764] The user can check the answer displayed on the device and obtain information about their question. In addition, the answer is sensitive to the user's feelings, so the user feels reassured and satisfied.

[0765] Example 2

[0766] 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."

[0767] Conventional systems were able to identify the appropriate area of ​​expertise for user requests and generate answers using specialized AI, but they were unable to adjust the answers to take the user's emotions into account, making it difficult to provide optimal information according to the user's emotional state.

[0768] 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 receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty field based on keywords, means for transferring the request to a specialized AI having an expert knowledge database corresponding to the identified specialty field, means for returning an answer obtained from the specialized AI to the user, means for recognizing the user's emotion, and means for adjusting the tone and content of the answer based on the recognized emotion. This makes it possible to provide optimal information according to the user's emotional state.

[0769] The "means for receiving requests from a user" refers to a device or software that includes a process for receiving inquiries or instructions given by a user through a terminal as input data and converting them into a form that can be sent to a server.

[0770] The "means for analyzing the received request content and identifying the appropriate specialty based on keywords" refers to a device or software that analyzes the text data of the request, extracts important words and phrases, and identifies the relevant specialty based on those words and phrases.

[0771] A "means for forwarding requests to specialized AI with a database of specialized knowledge corresponding to an identified area of ​​expertise" is a device or software that includes a process for identifying an area of ​​expertise and then sending the request data to an AI system with specialized knowledge corresponding to that area.

[0772] A "means for returning answers obtained from specialized AI to the user" is a device or software that includes a process for receiving answers generated by specialized AI, converting them into an appropriate format, and returning them to the user.

[0773] "Means for recognizing user emotions" refers to devices or software that include technologies and processes for analyzing and evaluating emotions from user request data and identifying the user's emotional state.

[0774] "Means for adjusting the tone and content of a response based on the recognized emotion" refers to a device or software that includes a process for appropriately adjusting the tone and expression of a generated response sentence in accordance with the recognized emotion of the user.

[0775] The present invention is a system for generating an appropriate response to a request from a user that takes into consideration the user's feelings. Specific means for implementing this system will be described below.

[0776] 1. Request Receiving Method

[0777] The terminal receives a request input by a user. Specifically, the terminal captures the user's request in text format and prepares it for transmission to the server. For example, the user inputs "Teach me about medicine" into the terminal.

[0778] 2. Emotion recognition means

[0779] The device passes the user's request text to an emotion recognition engine, which uses natural language processing (NLP) techniques to analyze the user's emotions from the text. The emotion engine identifies emotions such as anxiety, joy, and anger. Publicly available NLP libraries and APIs can be used as emotion recognition engines.

[0780] 3. Request Analysis and Specialty Identification Methods

[0781] The server analyzes requests received from the device and extracts important keywords from them. For example, it detects the keyword "medicine." Text mining technology is used to analyze the request. Based on the analyzed keywords, the server identifies the appropriate specialty. For example, if "medicine" is included, it references a database to identify the corresponding "medical" specialty.

[0782] 4. Expert knowledge database and expert AI

[0783] The server queries a database containing specialized AI corresponding to the identified specialty. In the case of the medical field, a specialized AI with medical knowledge is selected. This specialized AI analyzes the request content based on a pre-built knowledge base and generates an appropriate answer. The specialized AI used can be based on a publicly available generative AI model.

[0784] 5. Emotion-Based Response Modification

[0785] The server adjusts the tone and content of the generated response based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, the server adjusts the response to use gentler language. For example, the server changes the response to something like, "Don't worry. Medicine is the scientific study of human health and disease."

[0786] 6. Response return method

[0787] The server prepares the answer received from the expert AI for return to the user's device and generates an HTTP response, which includes the adjusted answer and is sent to the device.

[0788] 7. Answer display means

[0789] The device analyzes the HTTP response received from the server and extracts the answer. It then displays the answer to the user using a user interface (UI). Specifically, the answer is displayed in a chat window on the screen.

[0790] Specific examples

[0791] Example 1: Medical Inquiry (User is Anxious)

[0792] The user types "Teach me about medicine" into the terminal.

[0793] The device passes this request to the emotion engine, which recognizes the user's emotion as "anxiety."

[0794] The terminal transmits the user's request and emotion information to the server.

[0795] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0796] The server forwards the request to a specialized AI in the medical field.

[0797] Specialized AI generates appropriate answers based on your request.

[0798] The server adjusts the tone and content of the response based on the user's emotional information, reframing the response in a gentler way to ease the "anxiety": "Don't worry. Medicine is the scientific study of human health and disease."

[0799] The server returns the reconstructed response to the terminal.

[0800] The device displays the answer to the user, providing a sense of security.

[0801] Example 2: Legal enquiry (if the user is angry)

[0802] A user types into a terminal, "Tell me about the law."

[0803] The device passes this request to the emotion engine, which recognizes the user's emotion as "anger."

[0804] The terminal transmits the user's request and emotion information to the server.

[0805] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0806] The server forwards the request to a specialized AI in the legal field.

[0807] Specialized AI generates appropriate answers based on your request.

[0808] The server reconstructs a response that encourages calmness based on the user's emotional information: "Please stay calm. Laws are an important system for determining social rules."

[0809] The server returns the reconstructed response to the terminal.

[0810] The device displays the answer to the user and encourages them to remain calm.

[0811] In this way, the present invention can provide optimal information to a user by recognizing the user's emotions and adjusting responses based on those emotions.

[0812] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0813] Step 1: How to receive requests

[0814] A user types text into a terminal, for example, "Teach me about medicine."

[0815] Input: The request text entered by the user.

[0816] The terminal receives a request input by a user and stores it in memory as text data.

[0817] Output: The data where the input text is saved.

[0818] Specific behavior: The device displays a confirmation pop-up message "Processing your request..." to the user.

[0819] Step 2: Emotion recognition tools

[0820] The device passes the saved request text to the emotion engine.

[0821] Input: The saved request text.

[0822] The emotion engine uses NLP techniques to analyze user emotions from text, specifically by using publicly available NLP libraries to analyze text and determine emotional states.

[0823] Output: Emotional information (e.g., anxiety, joy, anger, etc.).

[0824] Specific behavior: While the device is performing emotion recognition, it displays a status bar and tells the user, "Analyzing emotions..."

[0825] Step 3: Request analysis and domain identification measures

[0826] The server analyzes the request text and emotion information received from the terminal.

[0827] Input: Request text and emotion information.

[0828] The server uses text mining technology to extract important keywords from the request, for example, the keyword "medicine."

[0829] Output: Identified keywords (e.g., "medicine").

[0830] Specific behavior: The server records the progress in a log file while analyzing keywords. It writes "Keyword 'medicine' found" in the log.

[0831] Step 4: Expert knowledge database and expert AI

[0832] The server identifies appropriate specialties based on the identified keywords.

[0833] Input: Identified keyword (e.g., medicine).

[0834] The server selects a specialized AI from the database that corresponds to the specified field. In the case of the medical field, a specialized AI with medical knowledge is selected.

[0835] Output: Selected specialized AI.

[0836] Specific behavior: The server logs the progress information as "specialty 'Medical' identified."

[0837] Step 5: Route the request to our expert AI

[0838] The server forwards the request to the selected specialized AI.

[0839] Input: Request text and selected expert AI.

[0840] The specialized AI analyzes the request and uses its knowledge base to generate an answer, such as "Medicine is the scientific study of human health and disease."

[0841] Output: The generated answer.

[0842] Specific operation: The specialized AI records the detailed step-by-step process of generating an answer in a log file. "Answer generation: 'Medicine is concerned with the health and disease of the human body...'"

[0843] Step 6: Emotion-Based Response Modification Measures

[0844] The server adjusts the tone and content of the generated response based on the user's emotional information provided by the emotion engine.

[0845] Input: Generated answers and sentiment information.

[0846] The server uses an emotion-adjustment algorithm to adjust the response, for example adding a gentler tone to the sentence, "Don't worry. Medicine is the scientific study of human health and disease."

[0847] Output: The adjusted answer.

[0848] Specific behavior: The server adjusts the response based on the user's emotion and logs the emotion adjustment point. "Change the tone to a gentler tone depending on the user's anxiety."

[0849] Step 7: Response return method

[0850] The server generates the adjusted answer as an HTTP response.

[0851] Input: Adjusted answer text.

[0852] The server generates an HTTP response including appropriate header information and sends it to the terminal.

[0853] Output: The generated HTTP response.

[0854] Specific behavior: When the server generates a response, it measures the response time and records it in the log. "Response time: 120ms"

[0855] Step 8: Displaying the Answers

[0856] The terminal analyzes the HTTP response received from the server and displays the extracted answer content to the user.

[0857] Input: HTTP response.

[0858] The device will then display the answer using a user interface (UI), for example, in a chat window on the screen: "Don't worry. Medicine is the scientific study of human health and disease."

[0859] Output: The answer displayed to the user.

[0860] What it does: When the device displays the final answer on the screen, it displays a highlighted message to the user: "Answer received!"

[0861] (Application example 2)

[0862] 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."

[0863] Conventional systems simply provide uniform responses without considering the user's emotions, making it difficult to provide appropriate information according to the emotional state of each individual user. Furthermore, security services, in particular, are required to respond quickly and appropriately to users' anxiety and anger, and it is important to respond in a way that is sensitive to the user's emotions. This will improve user satisfaction and enhance trust in security.

[0864] 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.

[0865] In this invention, the server includes means for receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty based on keywords, means for transferring the request to a specialized AI having an expert knowledge database corresponding to the identified specialty, means for returning an answer obtained from the specialized AI to the user, means for recognizing the user's emotion from text, voice, or image data included in the user's request, and means for adjusting the answer based on the recognized emotion, thereby making it possible to recognize the user's emotion and provide an appropriate and personalized answer based on that emotion.

[0866] A "request receiving means" is a device or system that has the function of receiving a request from a user.

[0867] The "request analysis means" is a system that has the function of analyzing the content of a received request and extracting important keywords and contexts.

[0868] The "specialty field identification means" is a system having a function of identifying an appropriate specialty field based on the keywords extracted by the request analysis means.

[0869] An "expert knowledge database" is a database that accumulates knowledge and information in a specific specialized field, and is the knowledge base that specialized AI references.

[0870] "Specialized AI" is artificial intelligence that has been trained to specialize in a specific field of expertise, and is an algorithm that uses a database of specialized knowledge to generate answers to user requests.

[0871] The "answer return means" is a system that has the function of returning the answers obtained from the specialized AI to the user.

[0872] The "emotion recognition means" is a system that has the function of analyzing the user's emotions from the text, voice, or image data included in the user's request.

[0873] The "answer adjustment means" is a system that has the function of adjusting the tone and content of the answer based on the emotional information recognized by the emotion recognition means.

[0874] In this invention, a system is constructed that recognizes a user's emotions, queries an AI in an appropriate field of expertise, and provides an answer according to the emotion. Specific embodiments are shown below.

[0875] composition

[0876] The system includes the following hardware and software:

[0877] 1. Terminal: A device that receives requests from users and recognizes their emotions. This can be a smartphone, smart glasses, or a head-mounted display.

[0878] 2. Emotion Engine: Software that analyzes emotions from text and voice data contained in user requests, using natural language processing libraries such as TextBlob.

[0879] 3. Server: The back-end system that analyzes the request, identifies the appropriate area of ​​expertise, and queries the specialized AI.

[0880] 4. Specialized AI: AI trained to specialize in a specific field, referencing a database of specialized knowledge, such as the medical field or security field.

[0881] 5. Answer Adjustment Module: A software module that adjusts answers based on emotion recognition results.

[0882] Processing flow

[0883] 1. The device receives a request from the user in text or voice format, for example, "I think I have a virus on my computer. What should I do?"

[0884] 2. The emotion engine installed on the device analyzes the text and voice data contained in the request and classifies the user's emotions into categories such as "joy," "anxiety," and "anger."

[0885] 3. This emotion information and the user's request are sent to the server.

[0886] 4. The server analyzes the request and extracts relevant keywords, using TextBlob or other natural language processing tools.

[0887] 5. Identify appropriate specializations (e.g., "antivirus," "medical") based on the extracted keywords.

[0888] 6. The request is forwarded to a specialized AI in the identified area of ​​expertise, which then consults a dedicated database to generate an appropriate response.

[0889] 7. Adjust responses from specialized AI based on emotion, for example, reframing a response to a user who is feeling anxious in a more reassuring tone or context.

[0890] 8. The reconstructed answer is sent back to the terminal and displayed to the user.

[0891] Specific examples

[0892] Example 1

[0893] If a user types into their device, "My PC has been running slow lately. Is it because of a virus?":

[0894] The emotion engine recognizes this as "anxiety."

[0895] Extract "virus" and "PC" using keyword extraction.

[0896] Make an inquiry to the "antivirus specialist AI."

[0897] Tailor your answers to reduce anxiety based on your emotions.

[0898] User: "My PC has been running slow lately. Could it be because of a virus?"

[0899] Terminal: "A virus may be the cause of your PC's slow performance. We recommend that you perform virus scans regularly. If you are concerned, start with some simple measures. Please feel free to contact us."

[0900] By implementing this invention, it becomes possible to provide appropriate information that is in tune with the user's emotions, thereby improving user satisfaction and trust.

[0901] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0902] Step 1:

[0903] The terminal receives a request from a user.

[0904] Input: User text or voice input (e.g., "I think I have a virus on my computer. What should I do?")

[0905] Output: User request data

[0906] Specific operation: The device receives user requests through input devices such as a microphone or keyboard. In the case of voice input, it uses speech recognition software to convert the request into text.

[0907] Step 2:

[0908] The device's emotion engine analyzes the received request and recognizes the user's emotions.

[0909] Input: User request data

[0910] Output: User's emotion data (e.g., "anxiety")

[0911] What it does: The emotion engine (e.g., the TextBlob library) performs text analysis and calculates polarity to identify emotions such as "joy," "anxiety," or "anger."

[0912] Step 3:

[0913] The terminal transmits the user's emotion information and request data to the server.

[0914] Input: User request data, emotion data

[0915] Output: Request sent to the server and emotion information

[0916] Specific operation: The device sends request data and emotion data to the server via a network connection using a protocol such as HTTP.

[0917] Step 4:

[0918] The server analyzes the received request and extracts important keywords.

[0919] Input: Request data

[0920] Output: Extracted keywords (e.g. "virus", "PC")

[0921] Specific operation: The server uses a natural language processing tool (e.g., TextBlob) to extract important keywords from the request data.

[0922] Step 5:

[0923] The server identifies the appropriate specialty based on the extracted keywords.

[0924] Input: Extracted keywords

[0925] Output: Identified specialty (e.g. "Antivirus")

[0926] What it does: The server matches the keywords against a database and ruleset to determine the appropriate specialty (e.g., "antivirus").

[0927] Step 6:

[0928] The server forwards the request to the appropriate specialized AI.

[0929] Input: User request data, identified expertise

[0930] Output: Answer data from specialized AI

[0931] Specific operation: The server calls the specialized AI's API and sends the request data in the appropriate format. The specialized AI then references a dedicated database and generates an answer.

[0932] Step 7:

[0933] The server tailors the response based on the sentiment information.

[0934] Input: Answer data from specialized AI, emotional data

[0935] Output: Tailored response data (e.g., tone and content to reduce anxiety)

[0936] Specific operation: The server takes into account the emotional data and customizes the response, such as changing it to a gentler tone.

[0937] Step 8:

[0938] The server sends a tailored response back to the terminal.

[0939] Input: Adjusted response data

[0940] Output: Send the answer to the terminal

[0941] Specific operation: The server sends the adjusted response data to the terminal using the HTTP protocol or the like.

[0942] Step 9:

[0943] The terminal displays the received response to the user.

[0944] Input: Response data from the server

[0945] Output: The answer that is displayed to the user

[0946] Specific operation: The device converts the response data into a format that is easy to display and presents it to the user on the screen or via audio.

[0947] The above steps realize a system that provides appropriate answers that are in tune with the user's emotions.

[0948] 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.

[0949] 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.

[0950] 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.

[0951] [Third embodiment]

[0952] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0953] 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.

[0954] 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).

[0955] 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.

[0956] 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.

[0957] 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).

[0958] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0959] 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.

[0960] 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.

[0961] 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.

[0962] 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.

[0963] 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."

[0964] The present invention provides a system that integrates multiple specialized AIs and queries an AI with an appropriate specialty in response to a user request. This system includes the following means.

[0965] Program processing

[0966] 1. Request receiving method:

[0967] The terminal receives a request from the user as input. For example, the user inputs a question into the terminal such as "Tell me about medicine."

[0968] The terminal processes this user request and prepares it for transmission to the server.

[0969] 2. Request Analysis and Expertise Identification Methods:

[0970] The server analyzes the request received from the device. The server analyzes the request content and extracts important keywords from it. For example, it checks whether the keyword "medicine" is included.

[0971] The server then identifies the appropriate specialty based on the extracted keywords, in this case the "medical" specialty because "medicine" is included.

[0972] 3. Expert knowledge database and expert AI:

[0973] The server queries a database with specialized AIs corresponding to the identified specialty, in this case, a specialized AI specialized in the medical field.

[0974] Specialized AI refers to a database containing a large amount of medical-related data and information and generates appropriate answers to user questions, such as "Medicine is the scientific study of human health and disease."

[0975] 4. Response return method:

[0976] The server sends the answer received from the specialized AI back to the user's device.

[0977] The terminal displays the answer received from the server to the user.

[0978] Specific examples

[0979] Example 1: Medical enquiry

[0980] The user types "Teach me about medicine" into the terminal.

[0981] The terminal sends this request to the server.

[0982] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[0983] The server forwards the request to a specialized AI in the medical field.

[0984] Based on the request received, the specialized AI generates an answer from its knowledge base: "Medicine is the scientific study of human health and disease."

[0985] The server then sends the answer received from the specialized AI back to the device.

[0986] The terminal displays the answer to the user.

[0987] Example 2: Legal enquiry

[0988] A user types into a terminal, "Tell me about the law."

[0989] The terminal sends this request to the server.

[0990] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[0991] The server forwards the request to a specialized AI in the legal field.

[0992] Based on the request received, the specialized AI generates an answer from its own knowledge base: "Law is a system for establishing social rules and norms."

[0993] The server then sends the answer received from the specialized AI back to the device.

[0994] The terminal displays the answer to the user.

[0995] In this way, the present invention provides accurate information based on knowledge of each specialized field in response to user requests.

[0996] The processing flow will be explained below.

[0997] Step 1:

[0998] The user types "Teach me about medicine" into the terminal. When the user presses the send button, the terminal receives this request.

[0999] Step 2:

[1000] The terminal converts the user's request into an HTTP request and prepares it for transmission to the server.

[1001] Step 3:

[1002] The terminal sends an HTTP request to the server (for example, sends the request using the POST method).

[1003] Step 4:

[1004] The server receives the HTTP request, analyzes the request body, and extracts the user's question.

[1005] Step 5:

[1006] The server analyzes the question and extracts important keywords, for example, checking whether the keyword "medicine" is included.

[1007] Step 6:

[1008] The server identifies the appropriate specialty based on the extracted keywords. In this case, since "medicine" is included, it identifies the "medical" specialty.

[1009] Step 7:

[1010] The server selects a specialized AI corresponding to the specified specialty field. In this case, it selects a specialized AI specialized in the medical field.

[1011] Step 8:

[1012] The server forwards the request to the selected specialized AI. The server calls the medical_ai's query method and passes the request content.

[1013] Step 9:

[1014] The specialized AI receives the request and uses its knowledge base to generate an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[1015] Step 10:

[1016] The specialized AI returns the generated answer to the server, which then prepares the answer received from the specialized AI for sending back to the user's device.

[1017] Step 11:

[1018] The server generates an HTTP response for the user's device, includes the answer, and sends this HTTP response to the device.

[1019] Step 12:

[1020] The terminal analyzes the HTTP response received from the server, extracts the answer, and displays the extracted answer to the user.

[1021] Step 13:

[1022] The user can read the answers displayed on the terminal and obtain the desired information.

[1023] Example 1

[1024] 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."

[1025] In today's information society, users need to quickly and accurately obtain detailed information in various specialized fields. However, in current systems, AIs that provide specialized information in specific fields are distributed across each field, making it difficult for users to obtain appropriate information tailored to their respective fields. Furthermore, due to low accuracy in analyzing request content, it may not be possible to identify the exact specialized field the user is looking for. Therefore, there is a need for a system that can quickly provide appropriate answers to user requests.

[1026] 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.

[1027] In this invention, the server includes means for receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty based on keywords, means for transferring the request to a generative AI model having an expert knowledge database corresponding to the identified specialty, and means for returning an answer obtained from the generative AI model to the user. This makes it possible to integrate AI models specialized in different specialty areas and provide information on the appropriate specialty in response to a user request quickly and accurately.

[1028] A "user" is a person who requests information from the system.

[1029] A "request" is an inquiry that a user makes to the system requesting information or services.

[1030] The "means for receiving" is a mechanism or method for capturing requests from users and incorporating them into the system.

[1031] The "analyzing means" refers to a technique or method for analyzing a received request and extracting keywords contained therein.

[1032] "Keywords" are important words or phrases in a request that identify an area of ​​expertise.

[1033] A "discipline" is an area in which information or knowledge falls into a particular category or field.

[1034] A "generative AI model" is an artificial intelligence model that generates and analyzes information based on large amounts of data.

[1035] A "transfer means" is a method or technique for sending a request to a generative AI model in a specified area of ​​expertise.

[1036] A "means for returning" is a mechanism or method for returning answers received from a generative AI model to a user.

[1037] "Network communications" refers to communications technologies that connect computers and devices to send and receive data.

[1038] The present invention is a system that queries AI with an appropriate field of expertise in response to a user request. The system receives the user's request as input, analyzes its content, and identifies the appropriate field of expertise based on keywords. The system then forwards the request to a generative AI model corresponding to the identified field of expertise and returns the generated answer to the user. Specific embodiments of the system are described below.

[1039] Hardware and software used

[1040] 1. Terminal: A device through which a user inputs a request. Examples include smartphones, tablets, and computers.

[1041] 2. Server: A central processing unit responsible for analyzing requests, identifying areas of expertise, forwarding requests to generative AI models, and returning answers. The server has a high-bandwidth network connection and high-performance computing capabilities. For example, it can be a virtual server provided by a cloud service provider (e.g., AWS, Azure).

[1042] 3. Generative AI model: This is an artificial intelligence model that generates information according to a specific field of expertise. For example, large-scale language models such as GPT-3 and BERT can be used.

[1043] 4. Natural language processing software: Used to analyze requests and extract keywords. Examples include Python's NLTK and spaCy.

[1044] System Operation Overview

[1045] 1. Request received:

[1046] The user types a request into the terminal, for example, "Tell me about the law."

[1047] The device receives the request and converts it to a format for sending to the server. Specifically, it converts the text to JSON format and sends it to the server as an HTTP request.

[1048] 2. Request analysis and expertise identification:

[1049] The server analyzes the incoming request, using natural language processing software (e.g., spaCy) to tokenize the request, tag it with parts of speech, and analyze dependencies.

[1050] The server extracts important keywords using a keyword extraction algorithm (e.g., TF-IDF). For example, it detects the keyword "law."

[1051] The server identifies the appropriate field of expertise (in this case, "law") based on the extracted keywords, and classifies it using a machine learning model (e.g., sklearn classifier).

[1052] 3. Querying the database of expert knowledge and expert AI:

[1053] The server forwards the request to the generative AI model in the specified domain of expertise. Specifically, it sends the request as an API request to the generative AI model.

[1054] The generative AI model refers to a knowledge database in that field and generates an appropriate answer, such as "Law is a system for defining social rules and norms."

[1055] The generative AI model sends the generated answer back to the server.

[1056] 4. Return of Response:

[1057] The server formats the answer received from the generative AI model and sends it to the user's device, for example by converting it into HTML format.

[1058] The device displays the response received from the server to the user, for example, on the browser or app screen.

[1059] Specific examples

[1060] Example 1: Medical enquiry

[1061] The user types "Teach me about medicine" into the terminal.

[1062] The terminal sends this request to the server.

[1063] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[1064] The server forwards the request to a generative AI model in the medical field.

[1065] The generative AI model generates the answer "Medicine is the scientific study of human health and disease" and sends it back to the server.

[1066] The server then sends the answer received from the generative AI model back to the device.

[1067] The terminal displays the answer to the user.

[1068] Example 2: Legal enquiry

[1069] A user types into a terminal, "Tell me about the law."

[1070] The terminal sends this request to the server.

[1071] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[1072] The server forwards the request to a generative AI model in the legal domain.

[1073] The generative AI model generates the answer, "Law is a system for defining social rules and norms," ​​and sends it back to the server.

[1074] The server then sends the answer received from the generative AI model back to the device.

[1075] The terminal displays the answer to the user.

[1076] In this way, the present invention can quickly provide accurate information based on knowledge in each specialized field in response to a user request.

[1077] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1078] Step 1:

[1079] The user inputs a request into the terminal. For example, the user might input "Tell me about the law." This request is captured as text data. The terminal captures this text data and formats it for further processing. The formatted data is converted to JSON format and prepared for sending to the server.

[1080] Step 2:

[1081] The terminal sends the formatted request data to the server. Specifically, it is sent as an HTTP POST request. Here, the input is the formatted text data, and the output is the request data received by the server.

[1082] Step 3:

[1083] The server parses the request data received from the terminal. The server uses natural language processing software (e.g., spaCy) to tokenize the request text, tag it with parts of speech, and analyze dependencies. At this point, the input is the received text data, and the output is the syntax information of the parsed request.

[1084] Step 4:

[1085] The server uses the TF-IDF algorithm to extract important keywords from the request text. For example, the keyword "law" is extracted. The input is the parsed text data, and the output is the extracted keywords.

[1086] Step 5:

[1087] The server uses a machine learning model to identify the appropriate field of expertise based on the extracted keywords. For example, the keyword "law" identifies the field of "law." The machine learning model is pre-trained using training data from multiple fields. The input is the extracted keywords, and the output is the identified field of expertise.

[1088] Step 6:

[1089] The server forwards the request to the generative AI model corresponding to the identified area of ​​expertise. This request includes the user's question. For example, a request such as "Teach me about law" is sent to a generative AI model specializing in law. The input is the identified area of ​​expertise and the request, and the output is the request sent to the generative AI model.

[1090] Step 7:

[1091] The generative AI model consults a database of expert knowledge and generates an answer based on the request, for example, "Law is a system for defining social rules and norms." The input is the request sent to the generative AI model, and the output is the generated answer.

[1092] Step 8:

[1093] The generative AI model returns the generated answer to the server. The server formats the received answer and prepares it for sending back to the user. Specifically, it converts it into HTML format, etc. The input is the generated answer, and the output is the formatted answer.

[1094] Step 9:

[1095] The server sends the formatted response to the terminal. Specifically, it is sent as an HTTP response. The input is the formatted response data, and the output is the response data received by the terminal.

[1096] Step 10:

[1097] The device displays the answer received from the server to the user. Specifically, it displays it on the user interface of the browser or app. The input is the received answer data, and the output is the answer displayed to the user.

[1098] The specific actions and data flow performed at each step enable rapid and accurate provision of information in response to user requests.

[1099] (Application example 1)

[1100] 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."

[1101] In conventional systems, when users sought specialized information, the process of using multiple specialized knowledge databases and AI was cumbersome, making it difficult to provide fast and accurate information. Furthermore, when obtaining specialized knowledge related to a specific field, there was a lack of a way to make appropriate inquiries to specialized AI. This led to the problem that customers in physical stores could not immediately obtain detailed information about specific products or services.

[1102] 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.

[1103] In this invention, the server includes means for receiving a request from a user, means for analyzing the received request and identifying an appropriate specialty based on keywords, means for forwarding the request to a specialized AI having a specialized knowledge database corresponding to the identified specialty, means for returning an answer obtained from the specialized AI to the user, means for associating the requested information with a specific area and providing specialized knowledge in the specific area, and means for providing the answer to the user using a mobile terminal, thereby enabling users to quickly and accurately obtain detailed specialized information about specific products and services in physical stores.

[1104] The "means for receiving a request from a user" is an interface device that inputs questions or requests entered by a user into the system.

[1105] "Means for analyzing the content of the received request and identifying the appropriate specialty based on keywords" refers to the process of analyzing the user's request using natural language processing technology, extracting important keywords, and determining the specialty to consult.

[1106] "Means for forwarding requests to specialized AI with a database of specialized knowledge corresponding to the identified area of ​​expertise" refers to the process of sending a user's request to an AI system that has a database of relevant specialized knowledge based on the area of ​​expertise identified as a result of the analysis.

[1107] "Means for returning answers obtained from specialized AI to the user" refers to a communication means for conveying answers generated by specialized AI to the user.

[1108] "Means of relating requested information to a specific area and providing expertise in that specific area" refers to the process of analyzing the content of a user's request and providing information specific to related physical stores and areas.

[1109] "Means for providing answers to users using mobile devices" refers to a system for displaying answers from specialized AI to users via portable electronic devices such as smartphones and tablets.

[1110] To implement this invention, it is necessary to build a system that receives user requests using a smartphone application, analyzes them, and queries the appropriate AI specialist. This system includes the following elements:

[1111] Hardware and Software Overview

[1112] User device (smartphone): Used as a device for users to input questions.

[1113] Server: Analyzes requests, identifies areas of expertise, queries specialized AI, and returns answers.

[1114] Specialized AI server: A server on which AI with knowledge in each specialized field runs.

[1115] The application running on the device functions as follows:

[1116] User request received

[1117] The user launches the smartphone app and inputs the information or question they want to know, for example, by entering a prompt such as "Please tell me about the nutritional value of this food."

[1118] Request analysis and expertise identification

[1119] The server analyzes the received request using natural language processing technology (e.g., various natural language processing libraries and frameworks). It extracts important keywords from the request and identifies the appropriate field of expertise. For example, if the keyword "nutrition" is extracted, it determines that the request is related to "nutrition."

[1120] Specialized AI Inquiry

[1121] The user's request is forwarded to the specialized AI, which has a database of specialized knowledge corresponding to the specified specialized field. The server sends an inquiry to the specialized AI server using an HTTP request or the like.

[1122] Obtaining the answer and returning it to the user

[1123] The specialized AI generates an appropriate answer from its own knowledge base based on the received request. For example, it may provide information such as, "This food contains 200 kcal per 100 g and is rich in vitamins A and C." The server then sends the answer obtained from the specialized AI back to the user's smartphone and displays it.

[1124] Examples of concrete examples and prompts

[1125] As a concrete example, consider the case where a user types in "I would like to know more about the material of this furniture." In this case, the server extracts the keyword "furniture" and queries a specialized AI that specializes in furniture materials. The specialized AI then generates and returns an answer such as "This furniture is made of oak, which is durable and beautiful."

[1126] Example prompt sentence:

[1127] "Please tell me about the nutritional value of this food."

[1128] "I'd like to know more about the materials used in this furniture."

[1129] In this way, customers can quickly and accurately obtain detailed, specialized information about specific products and services even in physical stores.

[1130] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1131] Step 1:

[1132] User request received

[1133] The user launches the application on their device and inputs the information or question they want to know. The input data is sent to the server via the smartphone application. This input is a prompt sentence, such as "Please tell me about the nutritional value of this food."

[1134] Step 2:

[1135] Sending the request

[1136] The terminal receives the user's request and sends it to the server, which receives the request. The input here is the user's request, and the output is the data sent to the server.

[1137] Step 3:

[1138] Request Analysis

[1139] The server analyzes the received request using natural language processing technology. Specifically, it uses an NLP library (e.g., spaCy or NLTK) to extract important keywords from the request. The input is the user's request data, and the output is the extracted keywords.

[1140] Step 4:

[1141] Specialized field identification

[1142] The server identifies the appropriate specialty based on the keywords extracted in step 3. For example, if the keyword "nutrition" is included, it is associated with the "nutrition" specialty. The input is the extracted keywords, and the output is the identified specialty.

[1143] Step 5:

[1144] Inquiry to specialized AI

[1145] The server forwards the request to a specialized AI server that has an expert knowledge database corresponding to the specified specialized field. The user's request is sent to the specialized AI using an HTTP request, etc. The input is the specified specialized field and the user request, and the output is the request sent to the specialized AI.

[1146] Step 6:

[1147] Answer generation by specialized AI

[1148] Specialized AI generates appropriate answers from its own knowledge base based on the received request. For example, a nutrition AI might provide information such as "This food contains 200 kcal per 100 g and is rich in vitamins A and C." The input is the user request and the specialized knowledge base, and the output is the generated answer.

[1149] Step 7:

[1150] Returning the answer to the server

[1151] The specialized AI sends the generated answer back to the server. The server receives the answer data from the specialized AI. The input is the answer from the specialized AI, and the output is the data received by the server.

[1152] Step 8:

[1153] Returning the answer to the user

[1154] The server sends the received answer back to the device and displays it to the user. The user can view the answer from the specialized AI through a smartphone application. The input is the answer data from the specialized AI, and the output is displayed on the user's device.

[1155] 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.

[1156] The present invention is a system that integrates multiple specialized AIs and queries AIs in the appropriate specialized field in response to a user request, combined with an emotion engine that recognizes the user's emotions. This system includes the following means.

[1157] Program processing

[1158] 1. Request receiving method:

[1159] The terminal receives a request from the user as input. For example, the user inputs a question into the terminal such as "Tell me about medicine."

[1160] The terminal processes this user request and prepares it for transmission to the server.

[1161] 2. Emotion recognition means:

[1162] The terminal includes an emotion engine for analyzing the user's emotion from text, voice, or image data contained in the user's request.

[1163] The emotion engine recognizes the user's emotions (e.g., joy, anxiety, anger, etc.) and sends this information to the server.

[1164] 3. Request Analysis and Expertise Identification Methods:

[1165] The server analyzes the request received from the device and extracts important keywords from the question, for example, checking whether the keyword "medicine" is included.

[1166] The server then identifies the appropriate specialty based on the extracted keywords, in this case the "medical" specialty since "medicine" is included.

[1167] 4. Expert knowledge database and expert AI:

[1168] The server queries a database with specialized AIs corresponding to the identified specialty, in this case, a specialized AI specialized in the medical field.

[1169] The specialized AI analyzes the request against its knowledge base and generates an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[1170] 5. Emotion-based response adjustment measures:

[1171] The server adjusts the tone and content of the response to the request based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling "anxious," the server will respond in a gentler tone.

[1172] 6. Response return method:

[1173] The server prepares the answer received from the specialized AI to be sent back to the user's device.

[1174] The server generates an HTTP response containing the appropriately tailored answer and sends it to the terminal.

[1175] 7. Answer display method:

[1176] The terminal analyzes the HTTP response received from the server and extracts the response content.

[1177] The terminal displays the extracted answers to the user.

[1178] Specific examples

[1179] Example 1: Medical Inquiry (User is Anxious)

[1180] The user types "Teach me about medicine" into the terminal.

[1181] The device passes this request to the emotion engine, which recognizes the user's emotion as "anxiety."

[1182] The terminal transmits the user's request and emotion information to the server.

[1183] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[1184] The server forwards the request to a specialized AI in the medical field.

[1185] Specialized AI generates appropriate answers based on your request.

[1186] The server adjusts the tone and content of the response based on the user's emotional information, and reconstructs the response in gentler language to ease the user's anxiety.

[1187] The server returns the reconstructed response to the terminal.

[1188] The device displays the answer to the user, providing a sense of security.

[1189] Example 2: Legal enquiry (if the user is angry)

[1190] A user types into a terminal, "Tell me about the law."

[1191] The device passes this request to the emotion engine, which recognizes the user's emotion as "anger."

[1192] The terminal transmits the user's request and emotion information to the server.

[1193] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[1194] The server forwards the request to a specialized AI in the legal field.

[1195] Specialized AI generates appropriate answers based on your request.

[1196] The server reconstructs a response that encourages calmness based on the user's emotional information.

[1197] The server returns the reconstructed response to the terminal.

[1198] The device displays the answer to the user and encourages them to remain calm.

[1199] In this way, the present invention can provide optimal information to a user by recognizing the user's emotions and adjusting responses based on those emotions.

[1200] The processing flow will be explained below.

[1201] Step 1:

[1202] The user types "Teach me about medicine" into the terminal. When the user presses the send button, the terminal receives this request.

[1203] Step 2:

[1204] The device passes the text included in the user's request to the emotion engine, which analyzes the text and recognizes the user's emotion. For example, it detects from the text that the user is feeling "anxiety."

[1205] Step 3:

[1206] The device generates an HTTP request by combining the user's request and the recognized emotion information, and sends it to the server.

[1207] Step 4:

[1208] The server receives the HTTP request, analyzes the request body, extracts the user's question, and extracts the keyword "medicine."

[1209] Step 5:

[1210] The server identifies the appropriate specialty based on the extracted keywords. In this case, since the keyword "medicine" is included, it identifies the "medical care" specialty.

[1211] Step 6:

[1212] The server selects a specialized AI corresponding to the specified specialty field. In this case, it selects a specialized AI specialized in the medical field.

[1213] Step 7:

[1214] To forward the request to the selected specialized AI, the server calls the medical_ai query method and passes the request content.

[1215] Step 8:

[1216] The specialized AI receives the request and uses its knowledge base to generate an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[1217] Step 9:

[1218] The server takes into account the user's emotional information based on the answers received from the specialized AI. For example, if the user is feeling "anxiety," it will adjust the tone and content of the answer to make it more friendly and reassuring.

[1219] Step 10:

[1220] The server prepares an appropriately tailored answer to send back to the user's device, generating an HTTP response and including the answer.

[1221] Step 11:

[1222] The server sends an HTTP response to the device.

[1223] Step 12:

[1224] The terminal analyzes the HTTP response received from the server, extracts the answer, and displays the extracted answer to the user.

[1225] Step 13:

[1226] The user can check the answer displayed on the device and obtain information about their question. In addition, the answer is sensitive to the user's feelings, so the user feels reassured and satisfied.

[1227] Example 2

[1228] 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."

[1229] Conventional systems were able to identify the appropriate area of ​​expertise for user requests and generate answers using specialized AI, but they were unable to adjust the answers to take the user's emotions into account, making it difficult to provide optimal information according to the user's emotional state.

[1230] 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 receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty field based on keywords, means for transferring the request to a specialized AI having an expert knowledge database corresponding to the identified specialty field, means for returning an answer obtained from the specialized AI to the user, means for recognizing the user's emotion, and means for adjusting the tone and content of the answer based on the recognized emotion. This makes it possible to provide optimal information according to the user's emotional state.

[1231] The "means for receiving requests from a user" refers to a device or software that includes a process for receiving inquiries or instructions given by a user through a terminal as input data and converting them into a form that can be sent to a server.

[1232] The "means for analyzing the received request content and identifying the appropriate specialty based on keywords" refers to a device or software that analyzes the text data of the request, extracts important words and phrases, and identifies the relevant specialty based on those words and phrases.

[1233] A "means for forwarding requests to specialized AI with a database of specialized knowledge corresponding to an identified area of ​​expertise" is a device or software that includes a process for identifying an area of ​​expertise and then sending the request data to an AI system with specialized knowledge corresponding to that area.

[1234] A "means for returning answers obtained from specialized AI to the user" is a device or software that includes a process for receiving answers generated by specialized AI, converting them into an appropriate format, and returning them to the user.

[1235] "Means for recognizing user emotions" refers to devices or software that include technologies and processes for analyzing and evaluating emotions from user request data and identifying the user's emotional state.

[1236] "Means for adjusting the tone and content of a response based on the recognized emotion" refers to a device or software that includes a process for appropriately adjusting the tone and expression of a generated response sentence in accordance with the recognized emotion of the user.

[1237] The present invention is a system for generating an appropriate response to a request from a user that takes into consideration the user's feelings. Specific means for implementing this system will be described below.

[1238] 1. Request Receiving Method

[1239] The terminal receives a request input by a user. Specifically, the terminal captures the user's request in text format and prepares it for transmission to the server. For example, the user inputs "Teach me about medicine" into the terminal.

[1240] 2. Emotion recognition means

[1241] The device passes the user's request text to an emotion recognition engine, which uses natural language processing (NLP) techniques to analyze the user's emotions from the text. The emotion engine identifies emotions such as anxiety, joy, and anger. Publicly available NLP libraries and APIs can be used as emotion recognition engines.

[1242] 3. Request Analysis and Specialty Identification Methods

[1243] The server analyzes requests received from the device and extracts important keywords from them. For example, it detects the keyword "medicine." Text mining technology is used to analyze the request. Based on the analyzed keywords, the server identifies the appropriate specialty. For example, if "medicine" is included, it references a database to identify the corresponding "medical" specialty.

[1244] 4. Expert knowledge database and expert AI

[1245] The server queries a database containing specialized AI corresponding to the identified specialty. In the case of the medical field, a specialized AI with medical knowledge is selected. This specialized AI analyzes the request content based on a pre-built knowledge base and generates an appropriate answer. The specialized AI used can be based on a publicly available generative AI model.

[1246] 5. Emotion-Based Response Modification

[1247] The server adjusts the tone and content of the generated response based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, the server adjusts the response to use gentler language. For example, the server changes the response to something like, "Don't worry. Medicine is the scientific study of human health and disease."

[1248] 6. Response return method

[1249] The server prepares the answer received from the expert AI for return to the user's device and generates an HTTP response, which includes the adjusted answer and is sent to the device.

[1250] 7. Answer display means

[1251] The device analyzes the HTTP response received from the server and extracts the answer. It then displays the answer to the user using a user interface (UI). Specifically, the answer is displayed in a chat window on the screen.

[1252] Specific examples

[1253] Example 1: Medical Inquiry (User is Anxious)

[1254] The user types "Teach me about medicine" into the terminal.

[1255] The device passes this request to the emotion engine, which recognizes the user's emotion as "anxiety."

[1256] The terminal transmits the user's request and emotion information to the server.

[1257] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[1258] The server forwards the request to a specialized AI in the medical field.

[1259] Specialized AI generates appropriate answers based on your request.

[1260] The server adjusts the tone and content of the response based on the user's emotional information, reframing the response in a gentler way to ease the "anxiety": "Don't worry. Medicine is the scientific study of human health and disease."

[1261] The server returns the reconstructed response to the terminal.

[1262] The device displays the answer to the user, providing a sense of security.

[1263] Example 2: Legal enquiry (if the user is angry)

[1264] A user types into a terminal, "Tell me about the law."

[1265] The device passes this request to the emotion engine, which recognizes the user's emotion as "anger."

[1266] The terminal transmits the user's request and emotion information to the server.

[1267] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[1268] The server forwards the request to a specialized AI in the legal field.

[1269] Specialized AI generates appropriate answers based on your request.

[1270] The server reconstructs a response that encourages calmness based on the user's emotional information: "Please stay calm. Laws are an important system for determining social rules."

[1271] The server returns the reconstructed response to the terminal.

[1272] The device displays the answer to the user and encourages them to remain calm.

[1273] In this way, the present invention can provide optimal information to a user by recognizing the user's emotions and adjusting responses based on those emotions.

[1274] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1275] Step 1: How to receive requests

[1276] A user types text into a terminal, for example, "Teach me about medicine."

[1277] Input: The request text entered by the user.

[1278] The terminal receives a request input by a user and stores it in memory as text data.

[1279] Output: The data where the input text is saved.

[1280] Specific behavior: The device displays a confirmation pop-up message "Processing your request..." to the user.

[1281] Step 2: Emotion recognition tools

[1282] The device passes the saved request text to the emotion engine.

[1283] Input: The saved request text.

[1284] The emotion engine uses NLP techniques to analyze user emotions from text, specifically by using publicly available NLP libraries to analyze text and determine emotional states.

[1285] Output: Emotional information (e.g., anxiety, joy, anger, etc.).

[1286] Specific behavior: While the device is performing emotion recognition, it displays a status bar and tells the user, "Analyzing emotions..."

[1287] Step 3: Request analysis and domain identification measures

[1288] The server analyzes the request text and emotion information received from the terminal.

[1289] Input: Request text and emotion information.

[1290] The server uses text mining technology to extract important keywords from the request, for example, the keyword "medicine."

[1291] Output: Identified keywords (e.g., "medicine").

[1292] Specific behavior: The server records the progress in a log file while analyzing keywords. It writes "Keyword 'medicine' found" in the log.

[1293] Step 4: Expert knowledge database and expert AI

[1294] The server identifies appropriate specialties based on the identified keywords.

[1295] Input: Identified keyword (e.g., medicine).

[1296] The server selects a specialized AI from the database that corresponds to the specified field. In the case of the medical field, a specialized AI with medical knowledge is selected.

[1297] Output: Selected specialized AI.

[1298] Specific behavior: The server logs the progress information as "specialty 'Medical' identified."

[1299] Step 5: Route the request to our expert AI

[1300] The server forwards the request to the selected specialized AI.

[1301] Input: Request text and selected expert AI.

[1302] The specialized AI analyzes the request and uses its knowledge base to generate an answer, such as "Medicine is the scientific study of human health and disease."

[1303] Output: The generated answer.

[1304] Specific operation: The specialized AI records the detailed step-by-step process of generating an answer in a log file. "Answer generation: 'Medicine is concerned with the health and disease of the human body...'"

[1305] Step 6: Emotion-Based Response Modification Measures

[1306] The server adjusts the tone and content of the generated response based on the user's emotional information provided by the emotion engine.

[1307] Input: Generated answers and sentiment information.

[1308] The server uses an emotion-adjustment algorithm to adjust the response, for example adding a gentler tone to the sentence, "Don't worry. Medicine is the scientific study of human health and disease."

[1309] Output: The adjusted answer.

[1310] Specific behavior: The server adjusts the response based on the user's emotion and logs the emotion adjustment point. "Change the tone to a gentler tone depending on the user's anxiety."

[1311] Step 7: Response return method

[1312] The server generates the adjusted answer as an HTTP response.

[1313] Input: Adjusted answer text.

[1314] The server generates an HTTP response including appropriate header information and sends it to the terminal.

[1315] Output: The generated HTTP response.

[1316] Specific behavior: When the server generates a response, it measures the response time and records it in the log. "Response time: 120ms"

[1317] Step 8: Displaying the Answers

[1318] The terminal analyzes the HTTP response received from the server and displays the extracted answer content to the user.

[1319] Input: HTTP response.

[1320] The device will then display the answer using a user interface (UI), for example, in a chat window on the screen: "Don't worry. Medicine is the scientific study of human health and disease."

[1321] Output: The answer displayed to the user.

[1322] What it does: When the device displays the final answer on the screen, it displays a highlighted message to the user: "Answer received!"

[1323] (Application example 2)

[1324] 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."

[1325] Conventional systems simply provide uniform responses without considering the user's emotions, making it difficult to provide appropriate information according to the emotional state of each individual user. Furthermore, security services, in particular, are required to respond quickly and appropriately to users' anxiety and anger, and it is important to respond in a way that is sensitive to the user's emotions. This will improve user satisfaction and enhance trust in security.

[1326] 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.

[1327] In this invention, the server includes means for receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty based on keywords, means for transferring the request to a specialized AI having an expert knowledge database corresponding to the identified specialty, means for returning an answer obtained from the specialized AI to the user, means for recognizing the user's emotion from text, voice, or image data included in the user's request, and means for adjusting the answer based on the recognized emotion, thereby making it possible to recognize the user's emotion and provide an appropriate and personalized answer based on that emotion.

[1328] A "request receiving means" is a device or system that has the function of receiving a request from a user.

[1329] The "request analysis means" is a system that has the function of analyzing the content of a received request and extracting important keywords and contexts.

[1330] The "specialty field identification means" is a system having a function of identifying an appropriate specialty field based on the keywords extracted by the request analysis means.

[1331] An "expert knowledge database" is a database that accumulates knowledge and information in a specific specialized field, and is the knowledge base that specialized AI references.

[1332] "Specialized AI" is artificial intelligence that has been trained to specialize in a specific field of expertise, and is an algorithm that uses a database of specialized knowledge to generate answers to user requests.

[1333] The "answer return means" is a system that has the function of returning the answers obtained from the specialized AI to the user.

[1334] The "emotion recognition means" is a system that has the function of analyzing the user's emotions from the text, voice, or image data included in the user's request.

[1335] The "answer adjustment means" is a system that has the function of adjusting the tone and content of the answer based on the emotional information recognized by the emotion recognition means.

[1336] In this invention, a system is constructed that recognizes a user's emotions, queries an AI in an appropriate field of expertise, and provides an answer according to the emotion. Specific embodiments are shown below.

[1337] composition

[1338] The system includes the following hardware and software:

[1339] 1. Terminal: A device that receives requests from users and recognizes their emotions. This can be a smartphone, smart glasses, or a head-mounted display.

[1340] 2. Emotion Engine: Software that analyzes emotions from text and voice data contained in user requests, using natural language processing libraries such as TextBlob.

[1341] 3. Server: The back-end system that analyzes the request, identifies the appropriate area of ​​expertise, and queries the specialized AI.

[1342] 4. Specialized AI: AI trained to specialize in a specific field, referencing a database of specialized knowledge, such as the medical field or security field.

[1343] 5. Answer Adjustment Module: A software module that adjusts answers based on emotion recognition results.

[1344] Processing flow

[1345] 1. The device receives a request from the user in text or voice format, for example, "I think I have a virus on my computer. What should I do?"

[1346] 2. The emotion engine installed on the device analyzes the text and voice data contained in the request and classifies the user's emotions into categories such as "joy," "anxiety," and "anger."

[1347] 3. This emotion information and the user's request are sent to the server.

[1348] 4. The server analyzes the request and extracts relevant keywords, using TextBlob or other natural language processing tools.

[1349] 5. Identify appropriate specializations (e.g., "antivirus," "medical") based on the extracted keywords.

[1350] 6. The request is forwarded to a specialized AI in the identified area of ​​expertise, which then consults a dedicated database to generate an appropriate response.

[1351] 7. Adjust responses from specialized AI based on emotion, for example, reframing a response to a user who is feeling anxious in a more reassuring tone or context.

[1352] 8. The reconstructed answer is sent back to the terminal and displayed to the user.

[1353] Specific examples

[1354] Example 1

[1355] If a user types into their device, "My PC has been running slow lately. Is it because of a virus?":

[1356] The emotion engine recognizes this as "anxiety."

[1357] Extract "virus" and "PC" using keyword extraction.

[1358] Make an inquiry to the "antivirus specialist AI."

[1359] Tailor your answers to reduce anxiety based on your emotions.

[1360] User: "My PC has been running slow lately. Could it be because of a virus?"

[1361] Terminal: "A virus may be the cause of your PC's slow performance. We recommend that you perform virus scans regularly. If you are concerned, start with some simple measures. Please feel free to contact us."

[1362] By implementing this invention, it becomes possible to provide appropriate information that is in tune with the user's emotions, thereby improving user satisfaction and trust.

[1363] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1364] Step 1:

[1365] The terminal receives a request from a user.

[1366] Input: User text or voice input (e.g., "I think I have a virus on my computer. What should I do?")

[1367] Output: User request data

[1368] Specific operation: The device receives user requests through input devices such as a microphone or keyboard. In the case of voice input, it uses speech recognition software to convert the request into text.

[1369] Step 2:

[1370] The device's emotion engine analyzes the received request and recognizes the user's emotions.

[1371] Input: User request data

[1372] Output: User's emotion data (e.g., "anxiety")

[1373] What it does: The emotion engine (e.g., the TextBlob library) performs text analysis and calculates polarity to identify emotions such as "joy," "anxiety," or "anger."

[1374] Step 3:

[1375] The terminal transmits the user's emotion information and request data to the server.

[1376] Input: User request data, emotion data

[1377] Output: Request sent to the server and emotion information

[1378] Specific operation: The device sends request data and emotion data to the server via a network connection using a protocol such as HTTP.

[1379] Step 4:

[1380] The server analyzes the received request and extracts important keywords.

[1381] Input: Request data

[1382] Output: Extracted keywords (e.g. "virus", "PC")

[1383] Specific operation: The server uses a natural language processing tool (e.g., TextBlob) to extract important keywords from the request data.

[1384] Step 5:

[1385] The server identifies the appropriate specialty based on the extracted keywords.

[1386] Input: Extracted keywords

[1387] Output: Identified specialty (e.g. "Antivirus")

[1388] What it does: The server matches the keywords against a database and ruleset to determine the appropriate specialty (e.g., "antivirus").

[1389] Step 6:

[1390] The server forwards the request to the appropriate specialized AI.

[1391] Input: User request data, identified expertise

[1392] Output: Answer data from specialized AI

[1393] Specific operation: The server calls the specialized AI's API and sends the request data in the appropriate format. The specialized AI then references a dedicated database and generates an answer.

[1394] Step 7:

[1395] The server tailors the response based on the sentiment information.

[1396] Input: Answer data from specialized AI, emotional data

[1397] Output: Tailored response data (e.g., tone and content to reduce anxiety)

[1398] Specific operation: The server takes into account the emotional data and customizes the response, such as changing it to a gentler tone.

[1399] Step 8:

[1400] The server sends a tailored response back to the terminal.

[1401] Input: Adjusted response data

[1402] Output: Send the answer to the terminal

[1403] Specific operation: The server sends the adjusted response data to the terminal using the HTTP protocol or the like.

[1404] Step 9:

[1405] The terminal displays the received response to the user.

[1406] Input: Response data from the server

[1407] Output: The answer that is displayed to the user

[1408] Specific operation: The device converts the response data into a format that is easy to display and presents it to the user on the screen or via audio.

[1409] The above steps realize a system that provides appropriate answers that are in tune with the user's emotions.

[1410] 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.

[1411] 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.

[1412] 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.

[1413] [Fourth embodiment]

[1414] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1415] 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.

[1416] 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).

[1417] 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.

[1418] 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.

[1419] 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).

[1420] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1421] 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.

[1422] 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.

[1423] 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.

[1424] 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.

[1425] 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.

[1426] 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."

[1427] The present invention provides a system that integrates multiple specialized AIs and queries an AI with an appropriate specialty in response to a user request. This system includes the following means.

[1428] Program processing

[1429] 1. Request receiving method:

[1430] The terminal receives a request from the user as input. For example, the user inputs a question into the terminal such as "Tell me about medicine."

[1431] The terminal processes this user request and prepares it for transmission to the server.

[1432] 2. Request Analysis and Expertise Identification Methods:

[1433] The server analyzes the request received from the device. The server analyzes the request content and extracts important keywords from it. For example, it checks whether the keyword "medicine" is included.

[1434] The server then identifies the appropriate specialty based on the extracted keywords, in this case the "medical" specialty because "medicine" is included.

[1435] 3. Expert knowledge database and expert AI:

[1436] The server queries a database with specialized AIs corresponding to the identified specialty, in this case, a specialized AI specialized in the medical field.

[1437] Specialized AI refers to a database containing a large amount of medical-related data and information and generates appropriate answers to user questions, such as "Medicine is the scientific study of human health and disease."

[1438] 4. Response return method:

[1439] The server sends the answer received from the specialized AI back to the user's device.

[1440] The terminal displays the answer received from the server to the user.

[1441] Specific examples

[1442] Example 1: Medical enquiry

[1443] The user types "Teach me about medicine" into the terminal.

[1444] The terminal sends this request to the server.

[1445] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[1446] The server forwards the request to a specialized AI in the medical field.

[1447] Based on the request received, the specialized AI generates an answer from its knowledge base: "Medicine is the scientific study of human health and disease."

[1448] The server then sends the answer received from the specialized AI back to the device.

[1449] The terminal displays the answer to the user.

[1450] Example 2: Legal enquiry

[1451] A user types into a terminal, "Tell me about the law."

[1452] The terminal sends this request to the server.

[1453] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[1454] The server forwards the request to a specialized AI in the legal field.

[1455] Based on the request received, the specialized AI generates an answer from its own knowledge base: "Law is a system for establishing social rules and norms."

[1456] The server then sends the answer received from the specialized AI back to the device.

[1457] The terminal displays the answer to the user.

[1458] In this way, the present invention provides accurate information based on knowledge of each specialized field in response to user requests.

[1459] The processing flow will be explained below.

[1460] Step 1:

[1461] The user types "Teach me about medicine" into the terminal. When the user presses the send button, the terminal receives this request.

[1462] Step 2:

[1463] The terminal converts the user's request into an HTTP request and prepares it for transmission to the server.

[1464] Step 3:

[1465] The terminal sends an HTTP request to the server (for example, sends the request using the POST method).

[1466] Step 4:

[1467] The server receives the HTTP request, analyzes the request body, and extracts the user's question.

[1468] Step 5:

[1469] The server analyzes the question and extracts important keywords, for example, checking whether the keyword "medicine" is included.

[1470] Step 6:

[1471] The server identifies the appropriate specialty based on the extracted keywords. In this case, since "medicine" is included, it identifies the "medical" specialty.

[1472] Step 7:

[1473] The server selects a specialized AI corresponding to the specified specialty field. In this case, it selects a specialized AI specialized in the medical field.

[1474] Step 8:

[1475] The server forwards the request to the selected specialized AI. The server calls the medical_ai's query method and passes the request content.

[1476] Step 9:

[1477] The specialized AI receives the request and uses its knowledge base to generate an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[1478] Step 10:

[1479] The specialized AI returns the generated answer to the server, which then prepares the answer received from the specialized AI for sending back to the user's device.

[1480] Step 11:

[1481] The server generates an HTTP response for the user's device, includes the answer, and sends this HTTP response to the device.

[1482] Step 12:

[1483] The terminal analyzes the HTTP response received from the server, extracts the answer, and displays the extracted answer to the user.

[1484] Step 13:

[1485] The user can read the answers displayed on the terminal and obtain the desired information.

[1486] Example 1

[1487] 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."

[1488] In today's information society, users need to quickly and accurately obtain detailed information in various specialized fields. However, in current systems, AIs that provide specialized information in specific fields are distributed across each field, making it difficult for users to obtain appropriate information tailored to their respective fields. Furthermore, due to low accuracy in analyzing request content, it may not be possible to identify the exact specialized field the user is looking for. Therefore, there is a need for a system that can quickly provide appropriate answers to user requests.

[1489] 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.

[1490] In this invention, the server includes means for receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty based on keywords, means for transferring the request to a generative AI model having an expert knowledge database corresponding to the identified specialty, and means for returning an answer obtained from the generative AI model to the user. This makes it possible to integrate AI models specialized in different specialty areas and provide information on the appropriate specialty in response to a user request quickly and accurately.

[1491] A "user" is a person who requests information from the system.

[1492] A "request" is an inquiry that a user makes to the system requesting information or services.

[1493] The "means for receiving" is a mechanism or method for capturing requests from users and incorporating them into the system.

[1494] The "analyzing means" refers to a technique or method for analyzing a received request and extracting keywords contained therein.

[1495] "Keywords" are important words or phrases in a request that identify an area of ​​expertise.

[1496] A "discipline" is an area in which information or knowledge falls into a particular category or field.

[1497] A "generative AI model" is an artificial intelligence model that generates and analyzes information based on large amounts of data.

[1498] A "transfer means" is a method or technique for sending a request to a generative AI model in a specified area of ​​expertise.

[1499] A "means for returning" is a mechanism or method for returning answers received from a generative AI model to a user.

[1500] "Network communications" refers to communications technologies that connect computers and devices to send and receive data.

[1501] The present invention is a system that queries AI with an appropriate field of expertise in response to a user request. The system receives the user's request as input, analyzes its content, and identifies the appropriate field of expertise based on keywords. The system then forwards the request to a generative AI model corresponding to the identified field of expertise and returns the generated answer to the user. Specific embodiments of the system are described below.

[1502] Hardware and software used

[1503] 1. Terminal: A device through which a user inputs a request. Examples include smartphones, tablets, and computers.

[1504] 2. Server: A central processing unit responsible for analyzing requests, identifying areas of expertise, forwarding requests to generative AI models, and returning answers. The server has a high-bandwidth network connection and high-performance computing capabilities. For example, it can be a virtual server provided by a cloud service provider (e.g., AWS, Azure).

[1505] 3. Generative AI model: This is an artificial intelligence model that generates information according to a specific field of expertise. For example, large-scale language models such as GPT-3 and BERT can be used.

[1506] 4. Natural language processing software: Used to analyze requests and extract keywords. Examples include Python's NLTK and spaCy.

[1507] System Operation Overview

[1508] 1. Request received:

[1509] The user types a request into the terminal, for example, "Tell me about the law."

[1510] The device receives the request and converts it to a format for sending to the server. Specifically, it converts the text to JSON format and sends it to the server as an HTTP request.

[1511] 2. Request analysis and expertise identification:

[1512] The server analyzes the incoming request, using natural language processing software (e.g., spaCy) to tokenize the request, tag it with parts of speech, and analyze dependencies.

[1513] The server extracts important keywords using a keyword extraction algorithm (e.g., TF-IDF). For example, it detects the keyword "law."

[1514] The server identifies the appropriate field of expertise (in this case, "law") based on the extracted keywords, and classifies it using a machine learning model (e.g., sklearn classifier).

[1515] 3. Querying the database of expert knowledge and expert AI:

[1516] The server forwards the request to the generative AI model in the specified domain of expertise. Specifically, it sends the request as an API request to the generative AI model.

[1517] The generative AI model refers to a knowledge database in that field and generates an appropriate answer, such as "Law is a system for defining social rules and norms."

[1518] The generative AI model sends the generated answer back to the server.

[1519] 4. Return of Response:

[1520] The server formats the answer received from the generative AI model and sends it to the user's device, for example by converting it into HTML format.

[1521] The device displays the response received from the server to the user, for example, on the browser or app screen.

[1522] Specific examples

[1523] Example 1: Medical enquiry

[1524] The user types "Teach me about medicine" into the terminal.

[1525] The terminal sends this request to the server.

[1526] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[1527] The server forwards the request to a generative AI model in the medical field.

[1528] The generative AI model generates the answer "Medicine is the scientific study of human health and disease" and sends it back to the server.

[1529] The server then sends the answer received from the generative AI model back to the device.

[1530] The terminal displays the answer to the user.

[1531] Example 2: Legal enquiry

[1532] A user types into a terminal, "Tell me about the law."

[1533] The terminal sends this request to the server.

[1534] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[1535] The server forwards the request to a generative AI model in the legal domain.

[1536] The generative AI model generates the answer, "Law is a system for defining social rules and norms," ​​and sends it back to the server.

[1537] The server then sends the answer received from the generative AI model back to the device.

[1538] The terminal displays the answer to the user.

[1539] In this way, the present invention can quickly provide accurate information based on knowledge in each specialized field in response to a user request.

[1540] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1541] Step 1:

[1542] The user inputs a request into the terminal. For example, the user might input "Tell me about the law." This request is captured as text data. The terminal captures this text data and formats it for further processing. The formatted data is converted to JSON format and prepared for sending to the server.

[1543] Step 2:

[1544] The terminal sends the formatted request data to the server. Specifically, it is sent as an HTTP POST request. Here, the input is the formatted text data, and the output is the request data received by the server.

[1545] Step 3:

[1546] The server parses the request data received from the terminal. The server uses natural language processing software (e.g., spaCy) to tokenize the request text, tag it with parts of speech, and analyze dependencies. At this point, the input is the received text data, and the output is the syntax information of the parsed request.

[1547] Step 4:

[1548] The server uses the TF-IDF algorithm to extract important keywords from the request text. For example, the keyword "law" is extracted. The input is the parsed text data, and the output is the extracted keywords.

[1549] Step 5:

[1550] The server uses a machine learning model to identify the appropriate field of expertise based on the extracted keywords. For example, the keyword "law" identifies the field of "law." The machine learning model is pre-trained using training data from multiple fields. The input is the extracted keywords, and the output is the identified field of expertise.

[1551] Step 6:

[1552] The server forwards the request to the generative AI model corresponding to the identified area of ​​expertise. This request includes the user's question. For example, a request such as "Teach me about law" is sent to a generative AI model specializing in law. The input is the identified area of ​​expertise and the request, and the output is the request sent to the generative AI model.

[1553] Step 7:

[1554] The generative AI model consults a database of expert knowledge and generates an answer based on the request, for example, "Law is a system for defining social rules and norms." The input is the request sent to the generative AI model, and the output is the generated answer.

[1555] Step 8:

[1556] The generative AI model returns the generated answer to the server. The server formats the received answer and prepares it for sending back to the user. Specifically, it converts it into HTML format, etc. The input is the generated answer, and the output is the formatted answer.

[1557] Step 9:

[1558] The server sends the formatted response to the terminal. Specifically, it is sent as an HTTP response. The input is the formatted response data, and the output is the response data received by the terminal.

[1559] Step 10:

[1560] The device displays the answer received from the server to the user. Specifically, it displays it on the user interface of the browser or app. The input is the received answer data, and the output is the answer displayed to the user.

[1561] The specific actions and data flow performed at each step enable rapid and accurate provision of information in response to user requests.

[1562] (Application example 1)

[1563] 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."

[1564] In conventional systems, when users sought specialized information, the process of using multiple specialized knowledge databases and AI was cumbersome, making it difficult to provide fast and accurate information. Furthermore, when obtaining specialized knowledge related to a specific field, there was a lack of a way to make appropriate inquiries to specialized AI. This led to the problem that customers in physical stores could not immediately obtain detailed information about specific products or services.

[1565] 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.

[1566] In this invention, the server includes means for receiving a request from a user, means for analyzing the received request and identifying an appropriate specialty based on keywords, means for forwarding the request to a specialized AI having a specialized knowledge database corresponding to the identified specialty, means for returning an answer obtained from the specialized AI to the user, means for associating the requested information with a specific area and providing specialized knowledge in the specific area, and means for providing the answer to the user using a mobile terminal, thereby enabling users to quickly and accurately obtain detailed specialized information about specific products and services in physical stores.

[1567] The "means for receiving a request from a user" is an interface device that inputs questions or requests entered by a user into the system.

[1568] "Means for analyzing the content of the received request and identifying the appropriate specialty based on keywords" refers to the process of analyzing the user's request using natural language processing technology, extracting important keywords, and determining the specialty to consult.

[1569] "Means for forwarding requests to specialized AI with a database of specialized knowledge corresponding to the identified area of ​​expertise" refers to the process of sending a user's request to an AI system that has a database of relevant specialized knowledge based on the area of ​​expertise identified as a result of the analysis.

[1570] "Means for returning answers obtained from specialized AI to the user" refers to a communication means for conveying answers generated by specialized AI to the user.

[1571] "Means of relating requested information to a specific area and providing expertise in that specific area" refers to the process of analyzing the content of a user's request and providing information specific to related physical stores and areas.

[1572] "Means for providing answers to users using mobile devices" refers to a system for displaying answers from specialized AI to users via portable electronic devices such as smartphones and tablets.

[1573] To implement this invention, it is necessary to build a system that receives user requests using a smartphone application, analyzes them, and queries the appropriate AI specialist. This system includes the following elements:

[1574] Hardware and Software Overview

[1575] User device (smartphone): Used as a device for users to input questions.

[1576] Server: Analyzes requests, identifies areas of expertise, queries specialized AI, and returns answers.

[1577] Specialized AI server: A server on which AI with knowledge in each specialized field runs.

[1578] The application running on the device functions as follows:

[1579] User request received

[1580] The user launches the smartphone app and inputs the information or question they want to know, for example, by entering a prompt such as "Please tell me about the nutritional value of this food."

[1581] Request analysis and expertise identification

[1582] The server analyzes the received request using natural language processing technology (e.g., various natural language processing libraries and frameworks). It extracts important keywords from the request and identifies the appropriate field of expertise. For example, if the keyword "nutrition" is extracted, it determines that the request is related to "nutrition."

[1583] Specialized AI Inquiry

[1584] The user's request is forwarded to the specialized AI, which has a database of specialized knowledge corresponding to the specified specialized field. The server sends an inquiry to the specialized AI server using an HTTP request or the like.

[1585] Obtaining the answer and returning it to the user

[1586] The specialized AI generates an appropriate answer from its own knowledge base based on the received request. For example, it may provide information such as, "This food contains 200 kcal per 100 g and is rich in vitamins A and C." The server then sends the answer obtained from the specialized AI back to the user's smartphone and displays it.

[1587] Examples of concrete examples and prompts

[1588] As a concrete example, consider the case where a user types in "I would like to know more about the material of this furniture." In this case, the server extracts the keyword "furniture" and queries a specialized AI that specializes in furniture materials. The specialized AI then generates and returns an answer such as "This furniture is made of oak, which is durable and beautiful."

[1589] Example prompt sentence:

[1590] "Please tell me about the nutritional value of this food."

[1591] "I'd like to know more about the materials used in this furniture."

[1592] In this way, customers can quickly and accurately obtain detailed, specialized information about specific products and services even in physical stores.

[1593] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1594] Step 1:

[1595] User request received

[1596] The user launches the application on their device and inputs the information or question they want to know. The input data is sent to the server via the smartphone application. This input is a prompt sentence, such as "Please tell me about the nutritional value of this food."

[1597] Step 2:

[1598] Sending the request

[1599] The terminal receives the user's request and sends it to the server, which receives the request. The input here is the user's request, and the output is the data sent to the server.

[1600] Step 3:

[1601] Request Analysis

[1602] The server analyzes the received request using natural language processing technology. Specifically, it uses an NLP library (e.g., spaCy or NLTK) to extract important keywords from the request. The input is the user's request data, and the output is the extracted keywords.

[1603] Step 4:

[1604] Specialized field identification

[1605] The server identifies the appropriate specialty based on the keywords extracted in step 3. For example, if the keyword "nutrition" is included, it is associated with the "nutrition" specialty. The input is the extracted keywords, and the output is the identified specialty.

[1606] Step 5:

[1607] Inquiry to specialized AI

[1608] The server forwards the request to a specialized AI server that has an expert knowledge database corresponding to the specified specialized field. The user's request is sent to the specialized AI using an HTTP request, etc. The input is the specified specialized field and the user request, and the output is the request sent to the specialized AI.

[1609] Step 6:

[1610] Answer generation by specialized AI

[1611] Specialized AI generates appropriate answers from its own knowledge base based on the received request. For example, a nutrition AI might provide information such as "This food contains 200 kcal per 100 g and is rich in vitamins A and C." The input is the user request and the specialized knowledge base, and the output is the generated answer.

[1612] Step 7:

[1613] Returning the answer to the server

[1614] The specialized AI sends the generated answer back to the server. The server receives the answer data from the specialized AI. The input is the answer from the specialized AI, and the output is the data received by the server.

[1615] Step 8:

[1616] Returning the answer to the user

[1617] The server sends the received answer back to the device and displays it to the user. The user can view the answer from the specialized AI through a smartphone application. The input is the answer data from the specialized AI, and the output is displayed on the user's device.

[1618] 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.

[1619] The present invention is a system that integrates multiple specialized AIs and queries AIs in the appropriate specialized field in response to a user request, combined with an emotion engine that recognizes the user's emotions. This system includes the following means.

[1620] Program processing

[1621] 1. Request receiving method:

[1622] The terminal receives a request from the user as input. For example, the user inputs a question into the terminal such as "Tell me about medicine."

[1623] The terminal processes this user request and prepares it for transmission to the server.

[1624] 2. Emotion recognition means:

[1625] The terminal includes an emotion engine for analyzing the user's emotion from text, voice, or image data contained in the user's request.

[1626] The emotion engine recognizes the user's emotions (e.g., joy, anxiety, anger, etc.) and sends this information to the server.

[1627] 3. Request Analysis and Expertise Identification Methods:

[1628] The server analyzes the request received from the device and extracts important keywords from the question, for example, checking whether the keyword "medicine" is included.

[1629] The server then identifies the appropriate specialty based on the extracted keywords, in this case the "medical" specialty since "medicine" is included.

[1630] 4. Expert knowledge database and expert AI:

[1631] The server queries a database with specialized AIs corresponding to the identified specialty, in this case, a specialized AI specialized in the medical field.

[1632] The specialized AI analyzes the request against its knowledge base and generates an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[1633] 5. Emotion-based response adjustment measures:

[1634] The server adjusts the tone and content of the response to the request based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling "anxious," the server will respond in a gentler tone.

[1635] 6. Response return method:

[1636] The server prepares the answer received from the specialized AI to be sent back to the user's device.

[1637] The server generates an HTTP response containing the appropriately tailored answer and sends it to the terminal.

[1638] 7. Answer display method:

[1639] The terminal analyzes the HTTP response received from the server and extracts the response content.

[1640] The terminal displays the extracted answers to the user.

[1641] Specific examples

[1642] Example 1: Medical Inquiry (User is Anxious)

[1643] The user types "Teach me about medicine" into the terminal.

[1644] The device passes this request to the emotion engine, which recognizes the user's emotion as "anxiety."

[1645] The terminal transmits the user's request and emotion information to the server.

[1646] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[1647] The server forwards the request to a specialized AI in the medical field.

[1648] Specialized AI generates appropriate answers based on your request.

[1649] The server adjusts the tone and content of the response based on the user's emotional information, and reconstructs the response in gentler language to ease the user's anxiety.

[1650] The server returns the reconstructed response to the terminal.

[1651] The device displays the answer to the user, providing a sense of security.

[1652] Example 2: Legal enquiry (if the user is angry)

[1653] A user types into a terminal, "Tell me about the law."

[1654] The device passes this request to the emotion engine, which recognizes the user's emotion as "anger."

[1655] The terminal transmits the user's request and emotion information to the server.

[1656] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[1657] The server forwards the request to a specialized AI in the legal field.

[1658] Specialized AI generates appropriate answers based on your request.

[1659] The server reconstructs a response that encourages calmness based on the user's emotional information.

[1660] The server returns the reconstructed response to the terminal.

[1661] The device displays the answer to the user and encourages them to remain calm.

[1662] In this way, the present invention can provide optimal information to a user by recognizing the user's emotions and adjusting responses based on those emotions.

[1663] The processing flow will be explained below.

[1664] Step 1:

[1665] The user types "Teach me about medicine" into the terminal. When the user presses the send button, the terminal receives this request.

[1666] Step 2:

[1667] The device passes the text included in the user's request to the emotion engine, which analyzes the text and recognizes the user's emotion. For example, it detects from the text that the user is feeling "anxiety."

[1668] Step 3:

[1669] The device generates an HTTP request by combining the user's request and the recognized emotion information, and sends it to the server.

[1670] Step 4:

[1671] The server receives the HTTP request, analyzes the request body, extracts the user's question, and extracts the keyword "medicine."

[1672] Step 5:

[1673] The server identifies the appropriate specialty based on the extracted keywords. In this case, since the keyword "medicine" is included, it identifies the "medical care" specialty.

[1674] Step 6:

[1675] The server selects a specialized AI corresponding to the specified specialty field. In this case, it selects a specialized AI specialized in the medical field.

[1676] Step 7:

[1677] To forward the request to the selected specialized AI, the server calls the medical_ai query method and passes the request content.

[1678] Step 8:

[1679] The specialized AI receives the request and uses its knowledge base to generate an appropriate answer, such as "Medicine is the scientific study of human health and disease."

[1680] Step 9:

[1681] The server takes into account the user's emotional information based on the answers received from the specialized AI. For example, if the user is feeling "anxiety," it will adjust the tone and content of the answer to make it more friendly and reassuring.

[1682] Step 10:

[1683] The server prepares an appropriately tailored answer to send back to the user's device, generating an HTTP response and including the answer.

[1684] Step 11:

[1685] The server sends an HTTP response to the device.

[1686] Step 12:

[1687] The terminal analyzes the HTTP response received from the server, extracts the answer, and displays the extracted answer to the user.

[1688] Step 13:

[1689] The user can check the answer displayed on the device and obtain information about their question. In addition, the answer is sensitive to the user's feelings, so the user feels reassured and satisfied.

[1690] Example 2

[1691] 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."

[1692] Conventional systems were able to identify the appropriate area of ​​expertise for user requests and generate answers using specialized AI, but they were unable to adjust the answers to take the user's emotions into account, making it difficult to provide optimal information according to the user's emotional state.

[1693] 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 receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty field based on keywords, means for transferring the request to a specialized AI having an expert knowledge database corresponding to the identified specialty field, means for returning an answer obtained from the specialized AI to the user, means for recognizing the user's emotion, and means for adjusting the tone and content of the answer based on the recognized emotion. This makes it possible to provide optimal information according to the user's emotional state.

[1694] The "means for receiving requests from a user" refers to a device or software that includes a process for receiving inquiries or instructions given by a user through a terminal as input data and converting them into a form that can be sent to a server.

[1695] The "means for analyzing the received request content and identifying the appropriate specialty based on keywords" refers to a device or software that analyzes the text data of the request, extracts important words and phrases, and identifies the relevant specialty based on those words and phrases.

[1696] A "means for forwarding requests to specialized AI with a database of specialized knowledge corresponding to an identified area of ​​expertise" is a device or software that includes a process for identifying an area of ​​expertise and then sending the request data to an AI system with specialized knowledge corresponding to that area.

[1697] A "means for returning answers obtained from specialized AI to the user" is a device or software that includes a process for receiving answers generated by specialized AI, converting them into an appropriate format, and returning them to the user.

[1698] "Means for recognizing user emotions" refers to devices or software that include technologies and processes for analyzing and evaluating emotions from user request data and identifying the user's emotional state.

[1699] "Means for adjusting the tone and content of a response based on the recognized emotion" refers to a device or software that includes a process for appropriately adjusting the tone and expression of a generated response sentence in accordance with the recognized emotion of the user.

[1700] The present invention is a system for generating an appropriate response to a request from a user that takes into consideration the user's feelings. Specific means for implementing this system will be described below.

[1701] 1. Request Receiving Method

[1702] The terminal receives a request input by a user. Specifically, the terminal captures the user's request in text format and prepares it for transmission to the server. For example, the user inputs "Teach me about medicine" into the terminal.

[1703] 2. Emotion recognition means

[1704] The device passes the user's request text to an emotion recognition engine, which uses natural language processing (NLP) techniques to analyze the user's emotions from the text. The emotion engine identifies emotions such as anxiety, joy, and anger. Publicly available NLP libraries and APIs can be used as emotion recognition engines.

[1705] 3. Request Analysis and Specialty Identification Methods

[1706] The server analyzes requests received from the device and extracts important keywords from them. For example, it detects the keyword "medicine." Text mining technology is used to analyze the request. Based on the analyzed keywords, the server identifies the appropriate specialty. For example, if "medicine" is included, it references a database to identify the corresponding "medical" specialty.

[1707] 4. Expert knowledge database and expert AI

[1708] The server queries a database containing specialized AI corresponding to the identified specialty. In the case of the medical field, a specialized AI with medical knowledge is selected. This specialized AI analyzes the request content based on a pre-built knowledge base and generates an appropriate answer. The specialized AI used can be based on a publicly available generative AI model.

[1709] 5. Emotion-Based Response Modification

[1710] The server adjusts the tone and content of the generated response based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, the server adjusts the response to use gentler language. For example, the server changes the response to something like, "Don't worry. Medicine is the scientific study of human health and disease."

[1711] 6. Response return method

[1712] The server prepares the answer received from the expert AI for return to the user's device and generates an HTTP response, which includes the adjusted answer and is sent to the device.

[1713] 7. Answer display means

[1714] The device analyzes the HTTP response received from the server and extracts the answer. It then displays the answer to the user using a user interface (UI). Specifically, the answer is displayed in a chat window on the screen.

[1715] Specific examples

[1716] Example 1: Medical Inquiry (User is Anxious)

[1717] The user types "Teach me about medicine" into the terminal.

[1718] The device passes this request to the emotion engine, which recognizes the user's emotion as "anxiety."

[1719] The terminal transmits the user's request and emotion information to the server.

[1720] The server analyzes the request and identifies the "medical" field based on the keyword "medicine."

[1721] The server forwards the request to a specialized AI in the medical field.

[1722] Specialized AI generates appropriate answers based on your request.

[1723] The server adjusts the tone and content of the response based on the user's emotional information, reframing the response in a gentler way to ease the "anxiety": "Don't worry. Medicine is the scientific study of human health and disease."

[1724] The server returns the reconstructed response to the terminal.

[1725] The device displays the answer to the user, providing a sense of security.

[1726] Example 2: Legal enquiry (if the user is angry)

[1727] A user types into a terminal, "Tell me about the law."

[1728] The device passes this request to the emotion engine, which recognizes the user's emotion as "anger."

[1729] The terminal transmits the user's request and emotion information to the server.

[1730] The server analyzes the request and identifies the "legal" field based on the keyword "legal."

[1731] The server forwards the request to a specialized AI in the legal field.

[1732] Specialized AI generates appropriate answers based on your request.

[1733] The server reconstructs a response that encourages calmness based on the user's emotional information: "Please stay calm. Laws are an important system for determining social rules."

[1734] The server returns the reconstructed response to the terminal.

[1735] The device displays the answer to the user and encourages them to remain calm.

[1736] In this way, the present invention can provide optimal information to a user by recognizing the user's emotions and adjusting responses based on those emotions.

[1737] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1738] Step 1: How to receive requests

[1739] A user types text into a terminal, for example, "Teach me about medicine."

[1740] Input: The request text entered by the user.

[1741] The terminal receives a request input by a user and stores it in memory as text data.

[1742] Output: The data where the input text is saved.

[1743] Specific behavior: The device displays a confirmation pop-up message "Processing your request..." to the user.

[1744] Step 2: Emotion recognition tools

[1745] The device passes the saved request text to the emotion engine.

[1746] Input: The saved request text.

[1747] The emotion engine uses NLP techniques to analyze user emotions from text, specifically by using publicly available NLP libraries to analyze text and determine emotional states.

[1748] Output: Emotional information (e.g., anxiety, joy, anger, etc.).

[1749] Specific behavior: While the device is performing emotion recognition, it displays a status bar and tells the user, "Analyzing emotions..."

[1750] Step 3: Request analysis and domain identification measures

[1751] The server analyzes the request text and emotion information received from the terminal.

[1752] Input: Request text and emotion information.

[1753] The server uses text mining technology to extract important keywords from the request, for example, the keyword "medicine."

[1754] Output: Identified keywords (e.g., "medicine").

[1755] Specific behavior: The server records the progress in a log file while analyzing keywords. It writes "Keyword 'medicine' found" in the log.

[1756] Step 4: Expert knowledge database and expert AI

[1757] The server identifies appropriate specialties based on the identified keywords.

[1758] Input: Identified keyword (e.g., medicine).

[1759] The server selects a specialized AI from the database that corresponds to the specified field. In the case of the medical field, a specialized AI with medical knowledge is selected.

[1760] Output: Selected specialized AI.

[1761] Specific behavior: The server logs the progress information as "specialty 'Medical' identified."

[1762] Step 5: Route the request to our expert AI

[1763] The server forwards the request to the selected specialized AI.

[1764] Input: Request text and selected expert AI.

[1765] The specialized AI analyzes the request and uses its knowledge base to generate an answer, such as "Medicine is the scientific study of human health and disease."

[1766] Output: The generated answer.

[1767] Specific operation: The specialized AI records the detailed step-by-step process of generating an answer in a log file. "Answer generation: 'Medicine is concerned with the health and disease of the human body...'"

[1768] Step 6: Emotion-Based Response Modification Measures

[1769] The server adjusts the tone and content of the generated response based on the user's emotional information provided by the emotion engine.

[1770] Input: Generated answers and sentiment information.

[1771] The server uses an emotion-adjustment algorithm to adjust the response, for example adding a gentler tone to the sentence, "Don't worry. Medicine is the scientific study of human health and disease."

[1772] Output: The adjusted answer.

[1773] Specific behavior: The server adjusts the response based on the user's emotion and logs the emotion adjustment point. "Change the tone to a gentler tone depending on the user's anxiety."

[1774] Step 7: Response return method

[1775] The server generates the adjusted answer as an HTTP response.

[1776] Input: Adjusted answer text.

[1777] The server generates an HTTP response including appropriate header information and sends it to the terminal.

[1778] Output: The generated HTTP response.

[1779] Specific behavior: When the server generates a response, it measures the response time and records it in the log. "Response time: 120ms"

[1780] Step 8: Displaying the Answers

[1781] The terminal analyzes the HTTP response received from the server and displays the extracted answer content to the user.

[1782] Input: HTTP response.

[1783] The device will then display the answer using a user interface (UI), for example, in a chat window on the screen: "Don't worry. Medicine is the scientific study of human health and disease."

[1784] Output: The answer displayed to the user.

[1785] What it does: When the device displays the final answer on the screen, it displays a highlighted message to the user: "Answer received!"

[1786] (Application example 2)

[1787] 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."

[1788] Conventional systems simply provide uniform responses without considering the user's emotions, making it difficult to provide appropriate information according to the emotional state of each individual user. Furthermore, security services, in particular, are required to respond quickly and appropriately to users' anxiety and anger, and it is important to respond in a way that is sensitive to the user's emotions. This will improve user satisfaction and enhance trust in security.

[1789] 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.

[1790] In this invention, the server includes means for receiving a request from a user, means for analyzing the content of the received request and identifying an appropriate specialty based on keywords, means for transferring the request to a specialized AI having an expert knowledge database corresponding to the identified specialty, means for returning an answer obtained from the specialized AI to the user, means for recognizing the user's emotion from text, voice, or image data included in the user's request, and means for adjusting the answer based on the recognized emotion, thereby making it possible to recognize the user's emotion and provide an appropriate and personalized answer based on that emotion.

[1791] A "request receiving means" is a device or system that has the function of receiving a request from a user.

[1792] The "request analysis means" is a system that has the function of analyzing the content of a received request and extracting important keywords and contexts.

[1793] The "specialty field identification means" is a system having a function of identifying an appropriate specialty field based on the keywords extracted by the request analysis means.

[1794] An "expert knowledge database" is a database that accumulates knowledge and information in a specific specialized field, and is the knowledge base that specialized AI references.

[1795] "Specialized AI" is artificial intelligence that has been trained to specialize in a specific field of expertise, and is an algorithm that uses a database of specialized knowledge to generate answers to user requests.

[1796] The "answer return means" is a system that has the function of returning the answers obtained from the specialized AI to the user.

[1797] The "emotion recognition means" is a system that has the function of analyzing the user's emotions from the text, voice, or image data included in the user's request.

[1798] The "answer adjustment means" is a system that has the function of adjusting the tone and content of the answer based on the emotional information recognized by the emotion recognition means.

[1799] In this invention, a system is constructed that recognizes a user's emotions, queries an AI in an appropriate field of expertise, and provides an answer according to the emotion. Specific embodiments are shown below.

[1800] composition

[1801] The system includes the following hardware and software:

[1802] 1. Terminal: A device that receives requests from users and recognizes their emotions. This can be a smartphone, smart glasses, or a head-mounted display.

[1803] 2. Emotion Engine: Software that analyzes emotions from text and voice data contained in user requests, using natural language processing libraries such as TextBlob.

[1804] 3. Server: The back-end system that analyzes the request, identifies the appropriate area of ​​expertise, and queries the specialized AI.

[1805] 4. Specialized AI: AI trained to specialize in a specific field, referencing a database of specialized knowledge, such as the medical field or security field.

[1806] 5. Answer Adjustment Module: A software module that adjusts answers based on emotion recognition results.

[1807] Processing flow

[1808] 1. The device receives a request from the user in text or voice format, for example, "I think I have a virus on my computer. What should I do?"

[1809] 2. The emotion engine installed on the device analyzes the text and voice data contained in the request and classifies the user's emotions into categories such as "joy," "anxiety," and "anger."

[1810] 3. This emotion information and the user's request are sent to the server.

[1811] 4. The server analyzes the request and extracts relevant keywords, using TextBlob or other natural language processing tools.

[1812] 5. Identify appropriate specializations (e.g., "antivirus," "medical") based on the extracted keywords.

[1813] 6. The request is forwarded to a specialized AI in the identified area of ​​expertise, which then consults a dedicated database to generate an appropriate response.

[1814] 7. Adjust responses from specialized AI based on emotion, for example, reframing a response to a user who is feeling anxious in a more reassuring tone or context.

[1815] 8. The reconstructed answer is sent back to the terminal and displayed to the user.

[1816] Specific examples

[1817] Example 1

[1818] If a user types into their device, "My PC has been running slow lately. Is it because of a virus?":

[1819] The emotion engine recognizes this as "anxiety."

[1820] Extract "virus" and "PC" using keyword extraction.

[1821] Make an inquiry to the "antivirus specialist AI."

[1822] Tailor your answers to reduce anxiety based on your emotions.

[1823] User: "My PC has been running slow lately. Could it be because of a virus?"

[1824] Terminal: "A virus may be the cause of your PC's slow performance. We recommend that you perform virus scans regularly. If you are concerned, start with some simple measures. Please feel free to contact us."

[1825] By implementing this invention, it becomes possible to provide appropriate information that is in tune with the user's emotions, thereby improving user satisfaction and trust.

[1826] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1827] Step 1:

[1828] The terminal receives a request from a user.

[1829] Input: User text or voice input (e.g., "I think I have a virus on my computer. What should I do?")

[1830] Output: User request data

[1831] Specific operation: The device receives user requests through input devices such as a microphone or keyboard. In the case of voice input, it uses speech recognition software to convert the request into text.

[1832] Step 2:

[1833] The device's emotion engine analyzes the received request and recognizes the user's emotions.

[1834] Input: User request data

[1835] Output: User's emotion data (e.g., "anxiety")

[1836] What it does: The emotion engine (e.g., the TextBlob library) performs text analysis and calculates polarity to identify emotions such as "joy," "anxiety," or "anger."

[1837] Step 3:

[1838] The terminal transmits the user's emotion information and request data to the server.

[1839] Input: User request data, emotion data

[1840] Output: Request sent to the server and emotion information

[1841] Specific operation: The device sends request data and emotion data to the server via a network connection using a protocol such as HTTP.

[1842] Step 4:

[1843] The server analyzes the received request and extracts important keywords.

[1844] Input: Request data

[1845] Output: Extracted keywords (e.g. "virus", "PC")

[1846] Specific operation: The server uses a natural language processing tool (e.g., TextBlob) to extract important keywords from the request data.

[1847] Step 5:

[1848] The server identifies the appropriate specialty based on the extracted keywords.

[1849] Input: Extracted keywords

[1850] Output: Identified specialty (e.g. "Antivirus")

[1851] What it does: The server matches the keywords against a database and ruleset to determine the appropriate specialty (e.g., "antivirus").

[1852] Step 6:

[1853] The server forwards the request to the appropriate specialized AI.

[1854] Input: User request data, identified expertise

[1855] Output: Answer data from specialized AI

[1856] Specific operation: The server calls the specialized AI's API and sends the request data in the appropriate format. The specialized AI then references a dedicated database and generates an answer.

[1857] Step 7:

[1858] The server tailors the response based on the sentiment information.

[1859] Input: Answer data from specialized AI, emotional data

[1860] Output: Tailored response data (e.g., tone and content to reduce anxiety)

[1861] Specific operation: The server takes into account the emotional data and customizes the response, such as changing it to a gentler tone.

[1862] Step 8:

[1863] The server sends a tailored response back to the terminal.

[1864] Input: Adjusted response data

[1865] Output: Send the answer to the terminal

[1866] Specific operation: The server sends the adjusted response data to the terminal using the HTTP protocol or the like.

[1867] Step 9:

[1868] The terminal displays the received response to the user.

[1869] Input: Response data from the server

[1870] Output: The answer that is displayed to the user

[1871] Specific operation: The device converts the response data into a format that is easy to display and presents it to the user on the screen or via audio.

[1872] The above steps realize a system that provides appropriate answers that are in tune with the user's emotions.

[1873] 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.

[1874] 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.

[1875] 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.

[1876] 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.

[1877] 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.

[1878] 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.

[1879] 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).

[1880] 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.

[1881] 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."

[1882] 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.

[1883] 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).

[1884] 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.

[1885] 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.

[1886] 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.

[1887] 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.

[1888] 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.

[1889] 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.

[1890] 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.

[1891] 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.

[1892] 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.

[1893] 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.

[1894] The following is further disclosed regarding the above embodiment.

[1895] (Claim 1)

[1896] means for receiving a request from a user;

[1897] a means for analyzing the received request and identifying the appropriate area of ​​expertise based on keywords;

[1898] a means for forwarding the request to a specialized AI having a database of expertise corresponding to the identified area of ​​expertise;

[1899] A means for returning the answers obtained from the specialized AI to the user;

[1900] A system including:

[1901] (Claim 2)

[1902] The system according to claim 1, characterized in that there are multiple specialized AIs, each specialized AI specializing in a different field of expertise.

[1903] (Claim 3)

[1904] 2. The system according to claim 1, further comprising natural language processing means for identifying a field of expertise using keywords contained in the request content.

[1905] "Example 1"

[1906] (Claim 1)

[1907] means for receiving a request from a user;

[1908] a means for analyzing the received request and identifying the appropriate area of ​​expertise based on keywords;

[1909] means for forwarding the request to a generative AI model having an expert knowledge database corresponding to the identified area of ​​expertise;

[1910] a means for returning the answer obtained from the generative AI model to the user;

[1911] means for transmitting and receiving data between these means via network communication;

[1912] A system including:

[1913] (Claim 2)

[1914] The system of claim 1, characterized in that there are multiple generative AI models, each of which specializes in a different field of expertise.

[1915] (Claim 3)

[1916] 2. The system according to claim 1, further comprising natural language processing means for identifying a field of expertise using keywords contained in the request content.

[1917] "Application Example 1"

[1918] (Claim 1)

[1919] means for receiving a request from a user;

[1920] a means for analyzing the received request and identifying the appropriate area of ​​expertise based on keywords;

[1921] a means for forwarding the request to a specialized AI having a database of expertise corresponding to the identified area of ​​expertise;

[1922] A means for returning the answers obtained from the specialized AI to the user;

[1923] means for relating the requested information to a particular domain and providing expertise in that particular domain;

[1924] The system includes a means for providing answers to a user using a mobile terminal.

[1925] (Claim 2)

[1926] The system according to claim 1, characterized in that there are multiple specialized AIs, each specialized AI specializing in a different field of expertise.

[1927] (Claim 3)

[1928] 2. The system according to claim 1, further comprising natural language processing means for identifying a field of expertise using keywords contained in the request content.

[1929] "Example 2: Combining Emotion Engines"

[1930] (Claim 1)

[1931] means for receiving a request from a user;

[1932] a means for analyzing the received request and identifying the appropriate area of ​​expertise based on keywords;

[1933] a means for forwarding the request to a specialized AI having a database of expertise corresponding to the identified area of ​​expertise;

[1934] A means for returning the answers obtained from the specialized AI to the user;

[1935] means for recognizing a user's emotion;

[1936] a means of adjusting the tone and content of responses based on perceived emotions;

[1937] A system including:

[1938] (Claim 2)

[1939] The system according to claim 1, characterized in that there are multiple specialized AIs, each specialized AI specializing in a different field of expertise.

[1940] (Claim 3)

[1941] 2. The system according to claim 1, further comprising natural language processing means for identifying a field of expertise using keywords contained in the request content.

[1942] "Application example 2 when combining emotion engines"

[1943] (Claim 1)

[1944] means for receiving a request from a user;

[1945] a means for analyzing the received request and identifying the appropriate area of ​​expertise based on keywords;

[1946] a means for forwarding the request to a specialized AI having a database of expertise corresponding to the identified area of ​​expertise;

[1947] A means for returning the answers obtained from the specialized AI to the user;

[1948] means for recognizing a user's emotion from text, voice, or image data included in the user's request;

[1949] a means of adjusting responses based on perceived emotions;

[1950] A system including:

[1951] (Claim 2)

[1952] The system according to claim 1, characterized in that there are multiple specialized AIs, each specialized AI specializing in a different field of expertise.

[1953] (Claim 3)

[1954] 2. The system according to claim 1, further comprising natural language processing means for identifying a field of expertise using keywords contained in the request content.

[1955] (Claim 4)

[1956] The system described in claim 1 is characterized in that it provides answers to users in an appropriate tone and content from AI with expertise in the appropriate field based on emotion recognition. [Explanation of symbols]

[1957] 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. means for receiving a request from a user; a means for analyzing the received request and identifying the appropriate area of ​​expertise based on keywords; a means for forwarding the request to a specialized AI having a database of expertise corresponding to the identified area of ​​expertise; A means for returning the answers obtained from the specialized AI to the user; A system including:

2. 2. The system according to claim 1, characterized in that there are a plurality of specialized AIs, each specialized AI specializing in a different field of expertise.

3. 2. The system according to claim 1, further comprising natural language processing means for identifying a field of expertise using keywords contained in the request content.

Images