System
The system addresses inefficiencies in generative AI by generating initial and detailed responses based on user queries, allowing users to gain deeper insights without additional questions, thus enhancing user experience.
Patent Information
- Application Number
- JP2024115270
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2026-01-29
AI Technical Summary
Current generative AI systems require users to repeatedly ask questions for detailed information, leading to inefficiency and inconvenience.
A system that includes a first generative AI to generate an initial response, determines if a detailed response is needed, and uses a second generative AI to provide deeper information if required, enhancing user experience by reducing the need for additional queries.
Enables users to efficiently acquire and deeply understand information without repetitive questioning, improving user experience.
Smart Images

Figure 2026014273000001_ABST
Abstract
Description
[Technical Field]
[0001] The technology of the present disclosure relates to a system. [Background technology]
[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]
[0004] In today's information utilization, users are required to use generative artificial intelligence (AI) to efficiently and deeply understand information. However, with current generative AI systems, if users need more detailed information, they must continually ask additional questions themselves, which detracts from the user experience. Furthermore, the responses from generative AI often do not always contain the detailed information the user needs, which often causes inconvenience to users. There is a need for a system that can solve these problems and allow users to easily dig deeper into information. [Means for solving the problem]
[0005] This invention provides a system that includes a first generative artificial intelligence means for accepting a question from a user and generating an initial response based on the question. Next, a means for determining whether a detailed response is necessary based on the content of the initial response is provided, and if a detailed response is necessary, a second generative artificial intelligence means for generating a detailed response based on the initial response is used. Finally, a system is provided that includes a means for returning the initial response or detailed response to the user, thereby enabling the user to understand information efficiently and deeply. This reduces the burden on the user and improves the user experience.
[0006] "User" means any person or entity that uses the System to enter questions and receive responses.
[0007] "Means for accepting questions" refers to an interface or program that receives questions sent by users in digital form and processes them appropriately.
[0008] "Generative artificial intelligence means" refers to algorithms or programs that use machine learning or deep learning to generate sentences or responses based on input data.
[0009] "Initial Response" refers to the first basic answer generated in response to a question submitted by a user.
[0010] An "elaborate response" is a follow-up response that builds on the initial response and includes more in-depth information or clarification.
[0011] "Generating an initial response based on a question" refers to the process by which a generative artificial intelligence generates an initial answer using a question received from a user as input.
[0012] "Means for determining whether a detailed response is required" refers to an algorithm or program that analyzes the content of a user's question or initial response and determines whether a more detailed answer is required.
[0013] "Second generative artificial intelligence means for generating a detailed response" refers to a generative artificial intelligence algorithm or program for generating deeper information or a more detailed explanation based on the initial response.
[0014] "Means for sending back" refers to an interface or program for displaying and notifying the user of the generated response. [Brief explanation of the drawings]
[0015] [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
[0016] 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.
[0017] First, the terms used in the following description will be explained.
[0018] 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).
[0019] 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.
[0020] 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.
[0021] 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.
[0022] 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."
[0023] [First embodiment]
[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.
[0025] 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.
[0026] 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).
[0027] 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.
[0028] 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.
[0029] 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.
[0030] 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.
[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.
[0032] 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.
[0033] 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.
[0034] 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.
[0035] 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."
[0036] The present invention relates to an interactive information exchange system with a user that uses generative artificial intelligence, allowing the user to efficiently acquire information and, in some cases, obtain in-depth, detailed information.
[0037] The main components of the system are:
[0038] 1. User Interface: Provides an interface through which users can enter questions and receive responses from the system. This can be a web, mobile, or desktop application. Users enter questions through this interface.
[0039] 2. Server: The server plays a central role in receiving and processing queries from users. The specific processes performed by the server are explained below.
[0040] Accepting a question: The server accepts a question submitted by the user via an HTTP POST request.
[0041] Generate an initial response: The server passes the received question to a first generative artificial intelligence means for generating an initial response, which uses a machine learning algorithm to generate a reasonable answer to the question.
[0042] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[0043] Generate detailed response: If it is determined that more detailed information is needed, the server passes the initial response to a second generative AI means to generate a more detailed response. This second generative AI has an algorithm that provides more in-depth information based on the initial response.
[0044] Returning a response: The server returns the generated initial or detailed response to the user interface, through which the user can receive the response.
[0045] 3. Terminal: A terminal is a device that runs the user interface. It sends questions entered by the user to the server and displays the responses sent back from the server. Below is a concrete example of how the system works.
[0046] Specific examples
[0047] User: Type the question "How does AI work?"
[0048] On your device: Capture this question and send it to your server as an HTTP POST request.
[0049] server:
[0050] A question is received, and an initial response is generated using a first generative artificial intelligence means, the initial response being "AI learns using algorithms and data."
[0051] Detects whether the question contains the keyword "details" or "more details." Since it does not contain them in this case, a detailed response will not be generated.
[0052] Sends the initial response back to the user.
[0053] Device: Receives the response from the server and displays the answer to the user: "AI learns using algorithms and data."
[0054] If the user requests more detailed information, for example by entering "Tell me how AI works. Please give me more details," the server will determine that more information is needed and will use a second generative artificial intelligence means to generate a detailed response: "The AI's learning process is achieved by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns," which will be generated and sent back to the user.
[0055] This system allows users to efficiently obtain the information they need and gain a deeper understanding without asking additional questions.
[0056] The processing flow will be explained below.
[0057] Step 1:
[0058] The user enters a question. The user enters a question using the user interface displayed on their device. For example, they might enter, "Tell me how AI works."
[0059] Step 2:
[0060] The device captures the user's question and sends it to the server as an HTTP POST request, which contains the question in JSON format.
[0061] Step 3:
[0062] The server receives the request from the device and extracts the question from the request body. Since the server uses the Flask framework, it extracts the question using request.json.
[0063] Step 4:
[0064] The server passes the extracted question to a first generative AI means (e.g., GPT-3) to generate an initial response. The server then calls the generative AI's API to obtain the response corresponding to the question.
[0065] Step 5:
[0066] The server obtains the initial response and temporarily stores it, for example, "AI uses algorithms and data to learn."
[0067] Step 6:
[0068] The server detects whether the user's question contains the keywords "details" or "more details." This is done using a string search algorithm.
[0069] Step 7:
[0070] If it is determined that the keyword is included, the server passes the initial response to the second generative artificial intelligence means to generate a detailed response based on the initial response, and then calls the API of the generative artificial intelligence again to obtain a response containing more in-depth information.
[0071] Step 8:
[0072] The server retrieves the detailed response and saves it as the final response, which is generated as follows: "The AI learning process is done by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[0073] Step 9:
[0074] The server returns the generated initial or detailed response in JSON format to the terminal, and constructs the final response as a JSON response and returns it as an HTTP response.
[0075] Step 10:
[0076] The terminal receives the response from the server, extracts the response text from the response data, and processes the text to display it on the user interface.
[0077] Step 11:
[0078] The terminal displays the extracted response text on the user interface, allowing the user to check the displayed answer and obtain the necessary information.
[0079] Example 1
[0080] 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."
[0081] Modern users want to efficiently obtain a large amount of information at once, but existing information processing systems only provide an initial response and then require users to enter a question again to obtain more detailed information. Such systems require users to pose additional questions, reducing the efficiency of information acquisition. Therefore, a new system is needed that allows users to quickly and efficiently obtain the information they need.
[0082] 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.
[0083] In this invention, the server includes a means for accepting input from a user, a generative AI model for generating an initial response based on the input, a means for determining the need for a detailed response based on the initial response, a second generative AI model for generating a detailed response when it is determined that the detailed response is needed, and a means for returning the generated response to the user. This allows the user to efficiently obtain the information they need in the first question, and to quickly obtain detailed information as needed.
[0084] A "user" is an entity that uses the system to enter questions and obtain information.
[0085] "Input" is information that refers to questions or requests that a user makes to a system.
[0086] A "server" is a central computer system that receives input from users, processes it, and generates a response.
[0087] A "generative AI model" is an artificial intelligence system that uses machine learning algorithms to generate rational answers based on user input.
[0088] An "initial response" is the first answer a generative AI model generates in response to a user's input.
[0089] The "means for determining the need for a detailed response" is an algorithm that determines whether the user needs more detailed information based on the initial response.
[0090] The "second generative AI model" is an artificial intelligence system with machine learning algorithms to provide deeper information that is used when more detailed information is deemed necessary.
[0091] A "means for returning a response" is a communication means for sending the initial response or detailed response generated by the server back to the user interface.
[0092] MODE FOR CARRYING OUT THE INVENTION
[0093] The present invention relates to an interactive information exchange system with a user that uses generative artificial intelligence, which allows the user to efficiently acquire information and, in some cases, obtain in-depth, detailed information.
[0094] The main components of the system are:
[0095] 1. User Interface: Provides an interface for users to enter questions and receive responses from the system. This user interface can take the form of a web application, mobile application, or desktop application. Users enter questions through this interface.
[0096] 2. Server: The server plays a central role in receiving and processing input from users. The specific processes performed by the server are as follows:
[0097] Accepting a Question: The server accepts input submitted by the user. This acceptance is done via an HTTP POST request.
[0098] Generate initial response: The server passes the received input to a first generative AI model to generate an initial response. This generative AI model uses machine learning algorithms to generate a reasonable answer to the input.
[0099] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's input contains the keywords "details" or "more details."
[0100] Generate a detailed response: If more detailed information is required, the server passes the initial response to a second generative AI model, which generates a more detailed response. This second generative AI model has an algorithm that provides more in-depth information based on the initial response.
[0101] Returning a response: The server returns the generated initial or detailed response to the user interface, through which the user can receive the response.
[0102] 3. Terminal: A terminal is a device that runs a user interface, sends questions entered by the user to the server, and displays the responses sent back by the server.
[0103] Specific examples
[0104] User: Type the question "How does AI work?"
[0105] On your device: Capture this question and send it to your server as an HTTP POST request.
[0106] server:
[0107] A question is received and an initial response is generated using a first generative AI model, with the initial response being "AI learns using algorithms and data."
[0108] Detects whether the question contains the keyword "details" or "more details." Since it does not contain them in this case, a detailed response will not be generated.
[0109] Sends the initial response back to the user.
[0110] Device: Receives the response from the server and displays the answer to the user: "AI learns using algorithms and data."
[0111] If the user requests more detailed information, for example by typing "Tell me how AI works. Please tell me more," the server determines that more information is needed and uses a second generative AI model to generate a detailed response. The detailed response, "AI's learning process is achieved by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns," is generated and sent back to the user. This system allows users to efficiently obtain the information they need and gain a deeper understanding without having to ask additional questions.
[0112] The flow of the identification process in the first embodiment will be described with reference to FIG.
[0113] Step 1:
[0114] The user enters a question.
[0115] Input: The question text that the user types into the system's interface. Example: "Tell me how AI works."
[0116] Specific action: Enter a question into the text box in the user interface (web app, mobile app, desktop app) and press the "Submit" button.
[0117] Output: The question text is sent to the terminal.
[0118] Step 2:
[0119] The device captures the input and sends it to the server.
[0120] Input: The question text from the user.
[0121] What happens: The user interface takes the input data, includes the question text in the body of an HTTP POST request, and sends it to a specific endpoint on the server.
[0122] Output: The question text is passed to the server in an HTTP POST request.
[0123] Step 3:
[0124] The server receives the query.
[0125] Input: The question text from the HTTP POST request.
[0126] Specific behavior: The server receives the request and extracts the question text from request.body.
[0127] Output: The question text is stored as data for processing on the server.
[0128] Step 4:
[0129] The server generates an initial response.
[0130] Input: The user's question text.
[0131] Specific operation: The question text is passed to the generative AI model as a prompt to generate an initial response. The generation process calls the AI model's API, sends data, and retrieves the response text.
[0132] Output: An initial response text is generated. Example: "AI uses algorithms and data to learn."
[0133] Step 5:
[0134] The server determines whether a detailed response is required.
[0135] Input: The initial response text and the user's question text.
[0136] What it does: It uses a keyword detection algorithm to check whether the user's question text contains the keywords "details" or "more details."
[0137] Output: The result of whether a detailed response is required. If yes, it is true; otherwise, it is false.
[0138] Step 6:
[0139] The server generates a detailed response (if required).
[0140] Input: Initial response text and whether a detailed response is required.
[0141] Specific operation: If a detailed response is required, the initial response text is passed as a prompt to the second generative AI model to generate a detailed response. This process is also performed by calling the AI model's API.
[0142] Output: A detailed response text is generated. Example: "The AI learning process involves applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[0143] Step 7:
[0144] The server sends back a response.
[0145] Input: Initial response text or detailed response text.
[0146] Specific operation: The generated response is serialized as an HTTP response and sent back to the user's device. When creating a response, it is common to use a format such as JSON.
[0147] Output: The response text is sent back to the terminal.
[0148] Step 8:
[0149] The terminal displays the response to the user.
[0150] Input: The response text sent back by the server.
[0151] What happens: The user interface receives the HTTP response, parses it, and displays the response text in a display area in the browser or application.
[0152] Output: The user can see the system's response visually, such as in a text box, saying, "AI uses algorithms and data to learn."
[0153] (Application example 1)
[0154] 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."
[0155] In today's world, users want to access news articles and information efficiently and in detail, but there are challenges: searching for information and digging deeper is time-consuming, and it is difficult to obtain all the information needed at once. Furthermore, there is a lack of systems that can instantly provide the detailed information users desire. This situation leads to a poor user experience and a decrease in the efficiency of information gathering.
[0156] 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.
[0157] In this invention, the server includes means for receiving a question from a user, first artificial intelligence generative means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second artificial intelligence generative means for generating a detailed response based on the initial response if it is determined that a detailed response is necessary, means for returning the initial response or the detailed response to the user, and means for providing news articles and information to the user. This allows the user to obtain an initial response to a question and, if necessary, to immediately receive detailed information, thereby realizing efficient and in-depth information gathering.
[0158] "User" means an individual or entity that utilizes the System to enter questions and receive responses.
[0159] A "question" is a question or request for information that a user enters into the system.
[0160] An "initial response" is the first response that a generative artificial intelligence generates based on a user's question.
[0161] "Generative AI" is an AI system that uses machine learning algorithms to generate responses to questions or information.
[0162] A "detailed response" is a response that is generated to provide more in-depth information based on the initial response.
[0163] The "server" is a central computer system that receives a user's question, generates a response using generative artificial intelligence, and sends the response back to the user.
[0164] "News Article" means news or information content provided to Users by the System.
[0165] "Information" is a general term for the data and knowledge that users want to obtain through questions.
[0166] "Details" are additional, more specific explanations or data about the specific information the user is seeking.
[0167] "Delivery method" refers to the method or technology the system uses to convey news stories and information to users.
[0168] The present invention is implemented as an interactive information exchange system using generative artificial intelligence. A detailed implementation method of this system will be described below.
[0169] System Overview
[0170] The system mainly consists of a user interface, a server, and a terminal. The user interface is implemented as a mobile application, and is a means for users to input questions and receive responses from the system.
[0171] Server Roles
[0172] The server plays a central role in receiving and processing user-submitted questions, specifically:
[0173] 1. Accepting questions
[0174] The questions entered by the user are received as HTTP POST requests, which are handled on the server using a web framework such as Flask.
[0175] 2. Generate an initial response
[0176] The server then asks the question to OpenAI's generative AI model (e.g., GPT model) and generates an initial response. The prompt is "Q: User's question\nA:".
[0177] 3. Determining whether a detailed response is necessary
[0178] After generating the initial response, the server detects whether the user's question contains specific keywords such as "details" or "more details," and determines whether a detailed response is required. This decision algorithm is implemented by a script that runs on the server side.
[0179] 4. Generate a detailed response
[0180] If the initial response does not satisfy the user's needs, the server again uses the generative AI model to generate a detailed response, this time with the prompt "Expand upon the following information: initial response."
[0181] 5. Returning the Response
[0182] The generated initial or detailed response is sent back to the user interface and displayed to the user.
[0183] Hardware and Software Details
[0184] Server: Web server using Flask
[0185] Generative AI models: OpenAI's GPT-based models (e.g., text-davinci-003)
[0186] API: OpenAI API
[0187] Specific examples
[0188] 1. User Question: "What is the history of artificial intelligence?"
[0189] 2. Server processing:
[0190] Initial response: "Artificial intelligence dates back to the 1950s."
[0191] If the user types "Tell me more," the detailed response is: "The history of artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has followed since then."
[0192] This system allows users to get an initial response to their questions and instantly receive more detailed information as needed, enabling efficient and in-depth information gathering. The following example prompts are used to set the input format for the generative AI model:
[0193] Example prompts for generating initial responses:
[0194] Q: What is the history of artificial intelligence?
[0195] A:
[0196] Example prompt for detailed response generation:
[0197] Expand upon the following information: The history of artificial intelligence dates back to the 1950s.
[0198] The flow of the specific processing in the application example 1 will be described with reference to FIG.
[0199] Step 1:
[0200] A user enters a question through a smartphone app, which the user interface captures and sends as an HTTP POST request to the server.
[0201] Input: User question (e.g., "What is the history of artificial intelligence?")
[0202] Output: HTTP POST request sent to the server
[0203] Step 2:
[0204] The server analyzes the received HTTP POST request and extracts the question.
[0205] Input: HTTP POST request
[0206] Output: Question (e.g. "What is the history of artificial intelligence?")
[0207] Step 3:
[0208] To generate an initial response based on the question, the server passes the question to a generative AI model (OpenAI's GPT-based model). The prompt used here is "Q: User's question\nA:". The generative AI generates an initial response based on this prompt.
[0209] Input: Question
[0210] Output: Initial response (e.g., "Artificial intelligence dates back to the 1950s.")
[0211] Step 4:
[0212] Based on the initial response generated, the server determines whether more information is needed, using an algorithm that detects whether the question contains the keywords "details" or "more details."
[0213] Input: Question and initial response
[0214] Output: Result of the judgment whether a detailed response is required or not (e.g., Detailed response required)
[0215] Step 5:
[0216] If a detailed response is required, the server again uses the generative AI model to generate a detailed response, with the prompt "Expand upon the following information: Initial response."
[0217] Input: Initial response
[0218] Output: Detailed response (e.g., "The history of artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has been done since then.")
[0219] Step 6:
[0220] The server returns the generated initial response or detailed response to the user interface as an HTTP response.
[0221] Input: Initial response or detailed response
[0222] Output: HTTP response
[0223] Step 7:
[0224] The user interface displays the received responses and provides information to the user.
[0225] Input: Initial response or detailed response
[0226] Output: The response shown to the user (e.g., "Artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has followed since then.")
[0227] Through these steps, users can quickly obtain detailed information efficiently.
[0228] 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.
[0229] The present invention relates to a user-interactive information exchange system that combines generative artificial intelligence and an emotion engine, allowing users to efficiently acquire information while adjusting the information according to their emotional state.
[0230] The main components of the system are:
[0231] 1. User Interface: Provides an interface through which users can enter questions and receive responses from the system. This can be a web, mobile, or desktop application. Users enter questions through this interface.
[0232] 2. Server: The server plays a central role in receiving and processing questions and emotional states from users. The specific processing performed by the server is described below.
[0233] Question and emotion reception: The server receives the question and emotion data submitted by the user via an HTTP POST request.
[0234] Detecting emotional states using an emotion engine: The server uses an emotion engine to analyze the emotional state from the user's input data and detect the emotional state. The emotion engine can perform text analysis and voice analysis.
[0235] Generate initial response: The server passes the question to a first generative artificial intelligence means to generate an initial response. The generative artificial intelligence uses machine learning algorithms to generate a rational answer to the question.
[0236] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[0237] Generate detailed response: If it is determined that more detailed information is needed, the server passes the initial response to the second generative AI means to generate a detailed response based on the initial response, and then calls the generative AI API again to obtain a response containing more in-depth information.
[0238] Emotion-based response adjustment: The server adjusts the initial and detailed responses based on the detected emotional state. For example, if the user's emotional state is detected as negative, the server adjusts the response to be more empathetic and encouraging.
[0239] Returning a response: The server returns the generated initial or detailed response in JSON format to the terminal. It then constructs the final response as a JSON response and returns it as an HTTP response.
[0240] 3. Terminal: The terminal is the device that runs the user interface. It transmits the questions and emotion data entered by the user to the server and displays the responses sent back from the server. A specific example of the system's operation is explained below.
[0241] Specific examples
[0242] User: Enters the question "How does AI work?" and the emotion engine detects the user's emotional state as "neutral" as they enter text.
[0243] Device: Capture this question and emotion data and send it to the server as an HTTP POST request.
[0244] server:
[0245] The question and emotion data are received, and an initial response is generated using a first generative artificial intelligence means, with the initial response being "AI learns using algorithms and data."
[0246] The question does not contain the keywords "details" or "more details", so no detailed response will be generated.
[0247] The emotion engine detects the user's emotional state as "neutral," so the response is sent back to the user without any tailoring.
[0248] Device: Receives the response from the server and displays the answer to the user: "AI uses algorithms and data to learn."
[0249] If a user has a negative emotional state in the context of requesting more detailed information, for example, by typing "Tell me how AI works. Please tell me more," and the emotion engine detects that the user's emotional state is "negative," the server will not only generate more detailed information but also adjust the response to be more empathetic and encouraging. In this way, it is possible to further improve the user experience by providing a customized response according to the user's emotional state.
[0250] The processing flow will be explained below.
[0251] Step 1:
[0252] The user types in a question. The user types in "Tell me how AI works" using the user interface displayed on their device.
[0253] Step 2:
[0254] The emotion engine analyzes the user's input and detects the user's emotional state. Based on the user's text input, the emotion engine recognizes the user's emotional state as "neutral."
[0255] Step 3:
[0256] The device captures the user's question and emotional state and sends it to the server as an HTTP POST request, which contains the question and emotional state in JSON format.
[0257] Step 4:
[0258] The server receives the request from the device and extracts the question and emotional state from the request body. Since the server uses the Flask framework, it extracts data using request.json.
[0259] Step 5:
[0260] The server passes the extracted question to a first generative AI means (e.g., GPT-3) to generate an initial response. The server then calls the generative AI's API and obtains the initial response: "AI learns using algorithms and data."
[0261] Step 6:
[0262] The server detects whether the user's question contains the keywords "details" or "more details." In this case, these keywords are not included, so the initial response is sufficient.
[0263] Step 7:
[0264] The server adjusts the response appropriately based on the user's emotional state, which is detected by the emotion engine as "neutral." In this case, since the emotional state is "neutral," no particular adjustment of the response is made.
[0265] Step 8:
[0266] The server returns the generated initial response to the device as a JSON response, saying, "AI uses algorithms and data to learn."
[0267] Step 9:
[0268] The terminal receives the response from the server, extracts the response text from the response data, and processes the text to display it on the user interface.
[0269] Step 10:
[0270] The device displays the extracted response text, "AI uses algorithms and data to learn," on the user interface. The user confirms the displayed answer and obtains the information.
[0271] Example (detailed response)
[0272] User: Enters the question "How does AI work? Please tell me more." As the text is entered, the emotion engine detects the user's emotional state as "negative."
[0273] Device: Capture this question and emotion data and send it to the server as an HTTP POST request.
[0274] server:
[0275] It receives the question and emotion data and uses a first generative artificial intelligence means to generate an initial response: "AI learns using algorithms and data."
[0276] The question contains the keyword "details" or "more details," so the server determines that more information is required.
[0277] To generate a detailed response based on the initial response, the initial response is passed to a second generative artificial intelligence means, which obtains a detailed response that states, "The AI learning process involves applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[0278] The emotion engine detects that the user's emotional state is "negative," so the detailed response is adjusted to be more empathetic and encouraging.
[0279] The adjusted detailed response is sent back to the device as a JSON response.
[0280] Terminal: Receives the response from the server, extracts the response text from the response data, and displays it in the user interface.
[0281] User: Check out the tailored detailed response, "The AI learning process works by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns. Don't worry, you'll understand too." and get informed.
[0282] This system allows users to efficiently obtain the information they need in a manner that is appropriately tailored to their emotions.
[0283] Example 2
[0284] 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."
[0285] Conventional information exchange systems are required to generate appropriate responses to user questions, but they lack the ability to adjust responses based on the user's emotional state. In particular, when a user is in a negative emotional state, they are unable to generate responses that are in line with that state, which can lead to a decline in the quality of the user experience. Another issue is the inability to generate efficient responses to users' requests for more detailed information.
[0286] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.
[0287] In this invention, the server includes means for receiving a question from a user, first generative artificial intelligence means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second generative artificial intelligence means for generating a detailed response based on the initial response if it is determined that the detailed response is necessary, means for using an emotion engine for detecting the emotional state of the user, means for adjusting the initial response or the detailed response based on the emotional state, and means for returning the initial response or the adjusted detailed response to the user. This makes it possible to provide a customized response according to the emotional state of the user and efficiently generate a response even for users who request detailed information.
[0288] "Means for accepting questions from users" refers to the interface or process by which the system receives questions entered by users.
[0289] The "first generative artificial intelligence means" refers to a machine learning algorithm or generative AI model for generating an initial response based on a user's question.
[0290] "Initial response" refers to the answer that is first generated by the first generative artificial intelligence means in response to the user's question.
[0291] "Means for determining whether a detailed response is required" refers to algorithms or programs that determine whether detailed information is required based on conditions such as whether the user's question contains specific keywords.
[0292] The "second generative artificial intelligence means" refers to a machine learning algorithm or generative AI model that generates more detailed information based on the initial response.
[0293] A "detailed response" is a response containing more detailed information provided to a user based on an initial response.
[0294] An "emotion engine" refers to software or algorithms that analyze and detect a user's emotional state from input data.
[0295] "Emotional state" refers to the emotional state (e.g., neutral, negative, positive) that the user is in when entering the question.
[0296] "Response tailoring" means a process or algorithm that tailors the content of an initial response or detailed response based on the detected emotional state to make it more relevant to the user.
[0297] "Means for returning a response to the user" refers to the communications means or protocol for constructing the generated response and returning it to the user's terminal.
[0298] This invention is a user-interactive information exchange system that combines generative artificial intelligence and an emotion engine. This system allows users to efficiently acquire information while adjusting the information according to their emotional state.
[0299] System configuration
[0300] The main components of the system are:
[0301] 1. User Interface:
[0302] Provide an interface, which may be a web, mobile, or desktop application, through which users can enter questions and receive responses from the system.
[0303] 2. Server:
[0304] The server plays a central role in receiving and processing questions and emotional states from users. The server performs the following specific processes:
[0305] Questions and comments welcome:
[0306] The server receives the question and emotion data submitted by the user via an HTTP POST request.
[0307] Emotion engine detects emotional states:
[0308] The server uses an emotion engine to analyze and detect the user's emotional state from the input data. The emotion engine performs text analysis and voice analysis.
[0309] Generate the initial response:
[0310] The server passes the question to a first generative artificial intelligence means to generate an initial response, which uses machine learning algorithms to generate a rational answer to the question.
[0311] Detailed response required:
[0312] Based on the initial response, the server determines whether more information is needed, using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[0313] Generate a detailed response:
[0314] If it is determined that more detailed information is needed, the server passes the initial response to a second generative AI means to generate a detailed response, which then invokes the generative AI API again to obtain a response containing more in-depth information.
[0315] Adjusting responses based on emotions:
[0316] The server adjusts the initial and detailed responses based on the detected emotional state, for example, adjusting the response to be more empathetic and encouraging if the user's emotional state is detected to be negative.
[0317] Returning a response:
[0318] The server returns the generated initial or detailed response to the terminal in JSON format, and constructs the final response as a JSON response and returns it as an HTTP response.
[0319] 3. Terminal:
[0320] A terminal is a device that executes a user interface. It transmits questions and emotion data entered by the user to a server and displays the responses returned by the server. The terminal has a program for displaying the responses received from the server.
[0321] Specific examples
[0322] A concrete example of the operation of the system is given below.
[0323] 1. User:
[0324] The user inputs the question, "Please tell me how AI works." The emotion engine detects the user's emotional state as "neutral" as they input the text.
[0325] 2. Terminal:
[0326] The device captures this question and emotion data and sends it to the server as an HTTP POST request.
[0327] 3. Server:
[0328] The question and emotion data are received, and an initial response is generated using a first generative artificial intelligence means, for example, the initial response is generated as "AI learns using algorithms and data."
[0329] The question does not contain the keywords "details" or "more details", so no detailed response will be generated.
[0330] The emotion engine detects the user's emotional state as "neutral," so the response is sent back to the user without any adjustments.
[0331] 4. Terminal:
[0332] The device receives a response from the server and displays the answer to the user: "AI learns using algorithms and data."
[0333] Prompt Sentence Examples
[0334] Here is an example of an input prompt for a generative AI model:
[0335] "When a user asks 'Tell me how AI works,' the emotion engine detects the user's emotion as 'neutral.' Please generate a reasonable answer to this question."
[0336] Using these prompts, the generative artificial intelligence generates appropriate responses to the user's questions.
[0337] The flow of the identification process in the second embodiment will be described with reference to FIG.
[0338] Step 1:
[0339] Users can input questions through the interface, for example, in a text box on the device's web browser. When a user types, "Please tell me how AI works," the text data is sent to the device.
[0340] Step 2:
[0341] The device receives the input question and passes it to an emotion engine to detect the user's emotional state. For example, the user's emotional state may be determined to be "neutral" through text analysis. This emotion data is then sent to the server along with the input text.
[0342] Step 3:
[0343] The server receives the HTTP POST request and stores the user's question and emotion data in an internal data store for processing, so that the question and emotion state data are managed within the server.
[0344] Step 4:
[0345] The server calls the emotion engine and re-analyzes the user's emotional state from the text and voice data. For example, it confirms the previously received "neutral" emotional state. The result of this emotion analysis is recorded on the server.
[0346] Step 5:
[0347] The server passes the question to a first generative artificial intelligence means, which generates an initial response. For example, the initial response may be "AI learns using algorithms and data." This is generated using a specific prompt. An example prompt is: "When a user asks 'How does AI work?' please generate a reasonable answer."
[0348] Step 6:
[0349] The server determines whether more detailed information is needed based on the initial response and the question. For example, it analyzes whether the question contains keywords such as "details" or "more details." If the keywords are included, a more detailed response is generated. In this process, the initial response is used as input data.
[0350] Step 7:
[0351] If it is determined that more information is needed, the server passes the initial response to a second generative artificial intelligence means to generate a detailed response. For example, a detailed response such as "The AI automatically learns and predicts by combining data and algorithms" may be generated. A specific prompt sentence is again used for this generation. An example of a prompt sentence is: "Please provide a more detailed explanation based on this initial response."
[0352] Step 8:
[0353] The server adjusts the response based on the detected emotional state. Specifically, if the emotion engine detects a user's emotional state as negative, the response will be changed to something more friendly and encouraging. For example, "This technology is very interesting and worth learning."
[0354] Step 9:
[0355] The server constructs the final response in JSON format and sends it back to the device as an HTTP response, such as "AI automatically learns and predicts by combining data and algorithms. This technology is very interesting and worth learning."
[0356] Step 10:
[0357] The device parses the JSON response received from the server and displays it to the user. For example, a message might appear on the web browser saying, "AI automatically learns and predicts by combining data and algorithms. This technology is very interesting and worth learning."
[0358] (Application example 2)
[0359] 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."
[0360] Conventional information exchange systems often return mechanical responses to user questions, making it difficult to respond flexibly to the user's emotions and circumstances. Furthermore, providing information without considering the user's emotional state can potentially detract from the user experience. In particular, if a system is unable to respond appropriately to a user who is in a negative emotional state, this can lead to a decline in user satisfaction and willingness to use the system.
[0361] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for accepting a question from a user, first generative artificial intelligence means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second generative artificial intelligence means for generating a detailed response based on the initial response when it is determined that a detailed response is necessary, means for returning the initial response or the detailed response to the user, and emotion analysis means for detecting the user's emotional state and adjusting the response content based on the emotional state. This makes it possible to provide flexible and appropriate information according to the user's emotional state, thereby improving the user experience.
[0362] The "means for accepting questions from the user" is an interface for receiving questions posed by the user as input.
[0363] "First generative artificial intelligence means for generating an initial response based on a question" refers to a generative artificial intelligence algorithm for creating an initial response to a question entered by a user.
[0364] The "means for determining whether a detailed response is required based on an initial response" is an algorithm for determining whether further detailed information is required based on the generated initial response.
[0365] "A second generative artificial intelligence means for generating a detailed response based on an initial response when it is determined that a detailed response is necessary" refers to a generative artificial intelligence algorithm for generating a detailed response containing deeper information based on the initial response.
[0366] The "means for returning an initial response or a detailed response to a user" refers to a means for presenting the generated initial response or detailed response to a user.
[0367] "Emotion analysis means for detecting the emotional state of a user and adjusting the response content based on the emotional state" refers to an algorithm or device that analyzes the emotional state of a user and appropriately adjusts the response content based on that emotion.
[0368] The present invention relates to an emotion-responsive content delivery system that adjusts response content based on the emotional state of a user. A basic system configuration and processing method for implementing the present invention will be described.
[0369] System Configuration
[0370] The system consists of the following main components:
[0371] 1. User Interface: The interface through which the user enters questions and receives responses from the system. This can take the form of a web application, a mobile application, or a desktop application.
[0372] 2. Server: Provides the core functions of the system and performs the following specific processes:
[0373] Question acceptance method: Accepts questions from users. This is done via an HTTP POST request.
[0374] Sentiment analysis: Detecting the user's emotional state from their questions and other input data. Emotional state analysis is done using text and speech analysis algorithms.
[0375] Initial response generation means (first generative artificial intelligence means): Generate an initial response to the user's question. This is done using a generative artificial intelligence model (e.g., GPT-3).
[0376] Detailed response decision means: Based on the initial response, a decision is made as to whether a detailed response is necessary. This decision is made using an algorithm that detects specific keywords contained in the user's question.
[0377] Detailed response generation means (second generative artificial intelligence means): When it is determined that a detailed response is necessary, a detailed response is generated based on the initial response.
[0378] Response adjustment measures: Adjust the response content based on the results of emotion analysis. If the emotional state is negative, change the response to something more empathetic and encouraging.
[0379] Response return method: The generated response is returned to the user, also as an HTTP response.
[0380] Processing example
[0381] Here is a concrete example of how this can be done:
[0382] 1. User enters question:
[0383] The user inputs a question through the application, such as "Tell me today's news. I want some encouragement." The user's emotional state is detected as "positive."
[0384] 2. Sending questions and emotional states
[0385] The device captures this question and emotional state data and sends it to the server as an HTTP POST request.
[0386] 3. Generating an initial response
[0387] The server receives the question and generates an initial response using a generative artificial intelligence model (e.g., GPT-3). The initial response generated is "We have great news today! A Japanese athlete won a medal!"
[0388] 4. Adjusting responses based on emotional state
[0389] Since the emotion analysis means detects the emotion as "positive," the response is sent back to the user without any special adjustments.
[0390] 5. Viewing the Response
[0391] The terminal receives the response from the server and displays the generated response to the user.
[0392] Prompt Sentence Examples
[0393] "Tell me the news today. I want some encouragement."
[0394] "Tell me how AI works. Give me the details."
[0395] "I've been feeling really down lately. I'd love some words of encouragement."
[0396] This enables flexible and appropriate information provision according to the user's emotional state, improving the user experience.
[0397] The flow of the specific processing in the application example 2 will be described with reference to FIG.
[0398] Step 1:
[0399] A user inputs a question. For example, the user inputs "Tell me what's in the news today. I need some encouragement." This input is made through the user interface and captured as input data.
[0400] Step 2:
[0401] The device collects emotional data along with questions from the user. The device uses an emotion analysis method to analyze the user's emotion from the input text and determines it as "positive." This data is sent to the server as an HTTP POST request.
[0402] Step 3:
[0403] The server receives the question and emotion data. The server receives the HTTP request and extracts the question text and emotion data. This data is passed to the next processing step.
[0404] Step 4:
[0405] The server generates an initial response using a generative artificial intelligence model. The server inputs the user's question into a first generative artificial intelligence means (e.g., GPT-3) and generates an initial response such as, "We have great news for you today! A Japanese athlete won a medal!"
[0406] Step 5:
[0407] The server determines whether a detailed response is required. The server checks whether the question contains certain keywords (e.g., "details," "more details"), and in this case, since the keywords are not included, it determines that a detailed response is not required.
[0408] Step 6:
[0409] The server adjusts the response content based on the emotional state. Because the emotional state is analyzed as "positive" by the emotion analysis means, the server uses the generated initial response as is without any special adjustment.
[0410] Step 7:
[0411] The server returns the final response. The server structures the initial response it generated in JSON format and returns it to the device as an HTTP response.
[0412] Step 8:
[0413] The terminal receives the response and displays it to the user. The terminal receives the HTTP response from the server and displays the received response data in the user interface. The user sees the message, "We have great news for you today! Japanese athletes have won medals!"
[0414] 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.
[0415] 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.
[0416] 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.
[0417] [Second embodiment]
[0418] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.
[0419] 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.
[0420] 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).
[0421] 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.
[0422] 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.
[0423] 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).
[0424] 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.
[0425] 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.
[0426] 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.
[0427] 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.
[0428] 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.
[0429] 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."
[0430] The present invention relates to an interactive information exchange system with a user that uses generative artificial intelligence, allowing the user to efficiently acquire information and, in some cases, obtain in-depth, detailed information.
[0431] The main components of the system are:
[0432] 1. User Interface: Provides an interface through which users can enter questions and receive responses from the system. This can be a web, mobile, or desktop application. Users enter questions through this interface.
[0433] 2. Server: The server plays a central role in receiving and processing queries from users. The specific processes performed by the server are explained below.
[0434] Accepting a question: The server accepts a question submitted by the user via an HTTP POST request.
[0435] Generate an initial response: The server passes the received question to a first generative artificial intelligence means for generating an initial response, which uses a machine learning algorithm to generate a reasonable answer to the question.
[0436] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[0437] Generate detailed response: If it is determined that more detailed information is needed, the server passes the initial response to a second generative AI means to generate a more detailed response. This second generative AI has an algorithm that provides more in-depth information based on the initial response.
[0438] Returning a response: The server returns the generated initial or detailed response to the user interface, through which the user can receive the response.
[0439] 3. Terminal: A terminal is a device that runs the user interface. It sends questions entered by the user to the server and displays the responses sent back from the server. Below is a concrete example of how the system works.
[0440] Specific examples
[0441] User: Type the question "How does AI work?"
[0442] On your device: Capture this question and send it to your server as an HTTP POST request.
[0443] server:
[0444] A question is received, and an initial response is generated using a first generative artificial intelligence means, the initial response being "AI learns using algorithms and data."
[0445] Detects whether the question contains the keyword "details" or "more details." Since it does not contain them in this case, a detailed response will not be generated.
[0446] Sends the initial response back to the user.
[0447] Device: Receives the response from the server and displays the answer to the user: "AI learns using algorithms and data."
[0448] If the user requests more detailed information, for example by entering "Tell me how AI works. Please give me more details," the server will determine that more information is needed and will use a second generative artificial intelligence means to generate a detailed response: "The AI's learning process is achieved by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns," which will be generated and sent back to the user.
[0449] This system allows users to efficiently obtain the information they need and gain a deeper understanding without asking additional questions.
[0450] The processing flow will be explained below.
[0451] Step 1:
[0452] The user enters a question. The user enters a question using the user interface displayed on their device. For example, they might enter, "Tell me how AI works."
[0453] Step 2:
[0454] The device captures the user's question and sends it to the server as an HTTP POST request, which contains the question in JSON format.
[0455] Step 3:
[0456] The server receives the request from the device and extracts the question from the request body. Since the server uses the Flask framework, it extracts the question using request.json.
[0457] Step 4:
[0458] The server passes the extracted question to a first generative AI means (e.g., GPT-3) to generate an initial response. The server then calls the generative AI's API to obtain the response corresponding to the question.
[0459] Step 5:
[0460] The server obtains the initial response and temporarily stores it, for example, "AI uses algorithms and data to learn."
[0461] Step 6:
[0462] The server detects whether the user's question contains the keywords "details" or "more details." This is done using a string search algorithm.
[0463] Step 7:
[0464] If it is determined that the keyword is included, the server passes the initial response to the second generative artificial intelligence means to generate a detailed response based on the initial response, and then calls the API of the generative artificial intelligence again to obtain a response containing more in-depth information.
[0465] Step 8:
[0466] The server retrieves the detailed response and saves it as the final response, which is generated as follows: "The AI learning process is done by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[0467] Step 9:
[0468] The server returns the generated initial or detailed response in JSON format to the terminal, and constructs the final response as a JSON response and returns it as an HTTP response.
[0469] Step 10:
[0470] The terminal receives the response from the server, extracts the response text from the response data, and processes the text to display it on the user interface.
[0471] Step 11:
[0472] The terminal displays the extracted response text on the user interface, allowing the user to check the displayed answer and obtain the necessary information.
[0473] Example 1
[0474] 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."
[0475] Modern users want to efficiently obtain a large amount of information at once, but existing information processing systems only provide an initial response and then require users to enter a question again to obtain more detailed information. Such systems require users to pose additional questions, reducing the efficiency of information acquisition. Therefore, a new system is needed that allows users to quickly and efficiently obtain the information they need.
[0476] 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.
[0477] In this invention, the server includes a means for accepting input from a user, a generative AI model for generating an initial response based on the input, a means for determining the need for a detailed response based on the initial response, a second generative AI model for generating a detailed response when it is determined that the detailed response is needed, and a means for returning the generated response to the user. This allows the user to efficiently obtain the information they need in the first question, and to quickly obtain detailed information as needed.
[0478] A "user" is an entity that uses the system to enter questions and obtain information.
[0479] "Input" is information that refers to questions or requests that a user makes to a system.
[0480] A "server" is a central computer system that receives input from users, processes it, and generates a response.
[0481] A "generative AI model" is an artificial intelligence system that uses machine learning algorithms to generate rational answers based on user input.
[0482] An "initial response" is the first answer a generative AI model generates in response to a user's input.
[0483] The "means for determining the need for a detailed response" is an algorithm that determines whether the user needs more detailed information based on the initial response.
[0484] The "second generative AI model" is an artificial intelligence system with machine learning algorithms to provide deeper information that is used when more detailed information is deemed necessary.
[0485] A "means for returning a response" is a communication means for sending the initial response or detailed response generated by the server back to the user interface.
[0486] MODE FOR CARRYING OUT THE INVENTION
[0487] The present invention relates to an interactive information exchange system with a user that uses generative artificial intelligence, which allows the user to efficiently acquire information and, in some cases, obtain in-depth, detailed information.
[0488] The main components of the system are:
[0489] 1. User Interface: Provides an interface for users to enter questions and receive responses from the system. This user interface can take the form of a web application, mobile application, or desktop application. Users enter questions through this interface.
[0490] 2. Server: The server plays a central role in receiving and processing input from users. The specific processes performed by the server are as follows:
[0491] Accepting a Question: The server accepts input submitted by the user. This acceptance is done via an HTTP POST request.
[0492] Generate initial response: The server passes the received input to a first generative AI model to generate an initial response. This generative AI model uses machine learning algorithms to generate a reasonable answer to the input.
[0493] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's input contains the keywords "details" or "more details."
[0494] Generate a detailed response: If more detailed information is required, the server passes the initial response to a second generative AI model, which generates a more detailed response. This second generative AI model has an algorithm that provides more in-depth information based on the initial response.
[0495] Returning a response: The server returns the generated initial or detailed response to the user interface, through which the user can receive the response.
[0496] 3. Terminal: A terminal is a device that runs a user interface, sends questions entered by the user to the server, and displays the responses sent back by the server.
[0497] Specific examples
[0498] User: Type the question "How does AI work?"
[0499] On your device: Capture this question and send it to your server as an HTTP POST request.
[0500] server:
[0501] A question is received and an initial response is generated using a first generative AI model, with the initial response being "AI learns using algorithms and data."
[0502] Detects whether the question contains the keyword "details" or "more details." Since it does not contain them in this case, a detailed response will not be generated.
[0503] Sends the initial response back to the user.
[0504] Device: Receives the response from the server and displays the answer to the user: "AI learns using algorithms and data."
[0505] If the user requests more detailed information, for example by typing "Tell me how AI works. Please tell me more," the server determines that more information is needed and uses a second generative AI model to generate a detailed response. The detailed response, "AI's learning process is achieved by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns," is generated and sent back to the user. This system allows users to efficiently obtain the information they need and gain a deeper understanding without having to ask additional questions.
[0506] The flow of the identification process in the first embodiment will be described with reference to FIG.
[0507] Step 1:
[0508] The user enters a question.
[0509] Input: The question text that the user types into the system's interface. Example: "Tell me how AI works."
[0510] Specific action: Enter a question into the text box in the user interface (web app, mobile app, desktop app) and press the "Submit" button.
[0511] Output: The question text is sent to the terminal.
[0512] Step 2:
[0513] The device captures the input and sends it to the server.
[0514] Input: The question text from the user.
[0515] What happens: The user interface takes the input data, includes the question text in the body of an HTTP POST request, and sends it to a specific endpoint on the server.
[0516] Output: The question text is passed to the server in an HTTP POST request.
[0517] Step 3:
[0518] The server receives the query.
[0519] Input: The question text from the HTTP POST request.
[0520] Specific behavior: The server receives the request and extracts the question text from request.body.
[0521] Output: The question text is stored as data for processing on the server.
[0522] Step 4:
[0523] The server generates an initial response.
[0524] Input: The user's question text.
[0525] Specific operation: The question text is passed to the generative AI model as a prompt to generate an initial response. The generation process calls the AI model's API, sends data, and retrieves the response text.
[0526] Output: An initial response text is generated. Example: "AI uses algorithms and data to learn."
[0527] Step 5:
[0528] The server determines whether a detailed response is required.
[0529] Input: The initial response text and the user's question text.
[0530] What it does: It uses a keyword detection algorithm to check whether the user's question text contains the keywords "details" or "more details."
[0531] Output: The result of whether a detailed response is required. If yes, it is true; otherwise, it is false.
[0532] Step 6:
[0533] The server generates a detailed response (if required).
[0534] Input: Initial response text and whether a detailed response is required.
[0535] Specific operation: If a detailed response is required, the initial response text is passed as a prompt to the second generative AI model to generate a detailed response. This process is also performed by calling the AI model's API.
[0536] Output: A detailed response text is generated. Example: "The AI learning process involves applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[0537] Step 7:
[0538] The server sends back a response.
[0539] Input: Initial response text or detailed response text.
[0540] Specific operation: The generated response is serialized as an HTTP response and sent back to the user's device. When creating a response, it is common to use a format such as JSON.
[0541] Output: The response text is sent back to the terminal.
[0542] Step 8:
[0543] The terminal displays the response to the user.
[0544] Input: The response text sent back by the server.
[0545] What happens: The user interface receives the HTTP response, parses it, and displays the response text in a display area in the browser or application.
[0546] Output: The user can see the system's response visually, such as in a text box, saying, "AI uses algorithms and data to learn."
[0547] (Application example 1)
[0548] 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."
[0549] In today's world, users want to access news articles and information efficiently and in detail, but there are challenges: searching for information and digging deeper is time-consuming, and it is difficult to obtain all the information needed at once. Furthermore, there is a lack of systems that can instantly provide the detailed information users desire. This situation leads to a poor user experience and a decrease in the efficiency of information gathering.
[0550] 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.
[0551] In this invention, the server includes means for receiving a question from a user, first artificial intelligence generative means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second artificial intelligence generative means for generating a detailed response based on the initial response if it is determined that a detailed response is necessary, means for returning the initial response or the detailed response to the user, and means for providing news articles and information to the user. This allows the user to obtain an initial response to a question and, if necessary, to immediately receive detailed information, thereby realizing efficient and in-depth information gathering.
[0552] "User" means an individual or entity that utilizes the System to enter questions and receive responses.
[0553] A "question" is a question or request for information that a user enters into the system.
[0554] An "initial response" is the first response that a generative artificial intelligence generates based on a user's question.
[0555] "Generative AI" is an AI system that uses machine learning algorithms to generate responses to questions or information.
[0556] A "detailed response" is a response that is generated to provide more in-depth information based on the initial response.
[0557] The "server" is a central computer system that receives a user's question, generates a response using generative artificial intelligence, and sends the response back to the user.
[0558] "News Article" means news or information content provided to Users by the System.
[0559] "Information" is a general term for the data and knowledge that users want to obtain through questions.
[0560] "Details" are additional, more specific explanations or data about the specific information the user is seeking.
[0561] "Delivery method" refers to the method or technology the system uses to convey news stories and information to users.
[0562] The present invention is implemented as an interactive information exchange system using generative artificial intelligence. A detailed implementation method of this system will be described below.
[0563] System Overview
[0564] The system mainly consists of a user interface, a server, and a terminal. The user interface is implemented as a mobile application, and is a means for users to input questions and receive responses from the system.
[0565] Server Roles
[0566] The server plays a central role in receiving and processing user-submitted questions, specifically:
[0567] 1. Accepting questions
[0568] The questions entered by the user are received as HTTP POST requests, which are handled on the server using a web framework such as Flask.
[0569] 2. Generate an initial response
[0570] The server then asks the question to OpenAI's generative AI model (e.g., GPT model) and generates an initial response. The prompt is "Q: User's question\nA:".
[0571] 3. Determining whether a detailed response is necessary
[0572] After generating the initial response, the server detects whether the user's question contains specific keywords such as "details" or "more details," and determines whether a detailed response is required. This decision algorithm is implemented by a script that runs on the server side.
[0573] 4. Generate a detailed response
[0574] If the initial response does not satisfy the user's needs, the server again uses the generative AI model to generate a detailed response, this time with the prompt "Expand upon the following information: initial response."
[0575] 5. Returning the Response
[0576] The generated initial or detailed response is sent back to the user interface and displayed to the user.
[0577] Hardware and Software Details
[0578] Server: Web server using Flask
[0579] Generative AI models: OpenAI's GPT-based models (e.g., text-davinci-003)
[0580] API: OpenAI API
[0581] Specific examples
[0582] 1. User Question: "What is the history of artificial intelligence?"
[0583] 2. Server processing:
[0584] Initial response: "Artificial intelligence dates back to the 1950s."
[0585] If the user types "Tell me more," the detailed response is: "The history of artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has followed since then."
[0586] This system allows users to get an initial response to their questions and instantly receive more detailed information as needed, enabling efficient and in-depth information gathering. The following example prompts are used to set the input format for the generative AI model:
[0587] Example prompts for generating initial responses:
[0588] Q: What is the history of artificial intelligence?
[0589] A:
[0590] Example prompt for detailed response generation:
[0591] Expand upon the following information: The history of artificial intelligence dates back to the 1950s.
[0592] The flow of the specific processing in the application example 1 will be described with reference to FIG.
[0593] Step 1:
[0594] A user enters a question through a smartphone app, which the user interface captures and sends as an HTTP POST request to the server.
[0595] Input: User question (e.g., "What is the history of artificial intelligence?")
[0596] Output: HTTP POST request sent to the server
[0597] Step 2:
[0598] The server analyzes the received HTTP POST request and extracts the question.
[0599] Input: HTTP POST request
[0600] Output: Question (e.g. "What is the history of artificial intelligence?")
[0601] Step 3:
[0602] To generate an initial response based on the question, the server passes the question to a generative AI model (OpenAI's GPT-based model). The prompt used here is "Q: User's question\nA:". The generative AI generates an initial response based on this prompt.
[0603] Input: Question
[0604] Output: Initial response (e.g., "Artificial intelligence dates back to the 1950s.")
[0605] Step 4:
[0606] Based on the initial response generated, the server determines whether more information is needed, using an algorithm that detects whether the question contains the keywords "details" or "more details."
[0607] Input: Question and initial response
[0608] Output: Result of the judgment whether a detailed response is required or not (e.g., Detailed response required)
[0609] Step 5:
[0610] If a detailed response is required, the server again uses the generative AI model to generate a detailed response, with the prompt "Expand upon the following information: Initial response."
[0611] Input: Initial response
[0612] Output: Detailed response (e.g., "The history of artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has been done since then.")
[0613] Step 6:
[0614] The server returns the generated initial response or detailed response to the user interface as an HTTP response.
[0615] Input: Initial response or detailed response
[0616] Output: HTTP response
[0617] Step 7:
[0618] The user interface displays the received responses and provides information to the user.
[0619] Input: Initial response or detailed response
[0620] Output: The response shown to the user (e.g., "Artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has followed since then.")
[0621] Through these steps, users can quickly obtain detailed information efficiently.
[0622] 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.
[0623] The present invention relates to a user-interactive information exchange system that combines generative artificial intelligence and an emotion engine, allowing users to efficiently acquire information while adjusting the information according to their emotional state.
[0624] The main components of the system are:
[0625] 1. User Interface: Provides an interface through which users can enter questions and receive responses from the system. This can be a web, mobile, or desktop application. Users enter questions through this interface.
[0626] 2. Server: The server plays a central role in receiving and processing questions and emotional states from users. The specific processing performed by the server is described below.
[0627] Question and emotion reception: The server receives the question and emotion data submitted by the user via an HTTP POST request.
[0628] Detecting emotional states using an emotion engine: The server uses an emotion engine to analyze the emotional state from the user's input data and detect the emotional state. The emotion engine can perform text analysis and voice analysis.
[0629] Generate initial response: The server passes the question to a first generative artificial intelligence means to generate an initial response. The generative artificial intelligence uses machine learning algorithms to generate a rational answer to the question.
[0630] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[0631] Generate detailed response: If it is determined that more detailed information is needed, the server passes the initial response to the second generative AI means to generate a detailed response based on the initial response, and then calls the generative AI API again to obtain a response containing more in-depth information.
[0632] Emotion-based response adjustment: The server adjusts the initial and detailed responses based on the detected emotional state. For example, if the user's emotional state is detected as negative, the server adjusts the response to be more empathetic and encouraging.
[0633] Returning a response: The server returns the generated initial or detailed response in JSON format to the terminal. It then constructs the final response as a JSON response and returns it as an HTTP response.
[0634] 3. Terminal: The terminal is the device that runs the user interface. It transmits the questions and emotion data entered by the user to the server and displays the responses sent back from the server. A specific example of the system's operation is explained below.
[0635] Specific examples
[0636] User: Enters the question "How does AI work?" and the emotion engine detects the user's emotional state as "neutral" as they enter text.
[0637] Device: Capture this question and emotion data and send it to the server as an HTTP POST request.
[0638] server:
[0639] The question and emotion data are received, and an initial response is generated using a first generative artificial intelligence means, with the initial response being "AI learns using algorithms and data."
[0640] The question does not contain the keywords "details" or "more details", so no detailed response will be generated.
[0641] The emotion engine detects the user's emotional state as "neutral," so the response is sent back to the user without any tailoring.
[0642] Device: Receives the response from the server and displays the answer to the user: "AI uses algorithms and data to learn."
[0643] If a user has a negative emotional state in the context of requesting more detailed information, for example, by typing "Tell me how AI works. Please tell me more," and the emotion engine detects that the user's emotional state is "negative," the server will not only generate more detailed information but also adjust the response to be more empathetic and encouraging. In this way, it is possible to further improve the user experience by providing a customized response according to the user's emotional state.
[0644] The processing flow will be explained below.
[0645] Step 1:
[0646] The user types in a question. The user types in "Tell me how AI works" using the user interface displayed on their device.
[0647] Step 2:
[0648] The emotion engine analyzes the user's input and detects the user's emotional state. Based on the user's text input, the emotion engine recognizes the user's emotional state as "neutral."
[0649] Step 3:
[0650] The device captures the user's question and emotional state and sends it to the server as an HTTP POST request, which contains the question and emotional state in JSON format.
[0651] Step 4:
[0652] The server receives the request from the device and extracts the question and emotional state from the request body. Since the server uses the Flask framework, it extracts data using request.json.
[0653] Step 5:
[0654] The server passes the extracted question to a first generative AI means (e.g., GPT-3) to generate an initial response. The server then calls the generative AI's API and obtains the initial response: "AI learns using algorithms and data."
[0655] Step 6:
[0656] The server detects whether the user's question contains the keywords "details" or "more details." In this case, these keywords are not included, so the initial response is sufficient.
[0657] Step 7:
[0658] The server adjusts the response appropriately based on the user's emotional state, which is detected by the emotion engine as "neutral." In this case, since the emotional state is "neutral," no particular adjustment of the response is made.
[0659] Step 8:
[0660] The server returns the generated initial response to the device as a JSON response, saying, "AI uses algorithms and data to learn."
[0661] Step 9:
[0662] The terminal receives the response from the server, extracts the response text from the response data, and processes the text to display it on the user interface.
[0663] Step 10:
[0664] The device displays the extracted response text, "AI uses algorithms and data to learn," on the user interface. The user confirms the displayed answer and obtains the information.
[0665] Example (detailed response)
[0666] User: Enters the question "How does AI work? Please tell me more." As the text is entered, the emotion engine detects the user's emotional state as "negative."
[0667] Device: Capture this question and emotion data and send it to the server as an HTTP POST request.
[0668] server:
[0669] It receives the question and emotion data and uses a first generative artificial intelligence means to generate an initial response: "AI learns using algorithms and data."
[0670] The question contains the keyword "details" or "more details," so the server determines that more information is required.
[0671] To generate a detailed response based on the initial response, the initial response is passed to a second generative artificial intelligence means, which obtains a detailed response that states, "The AI learning process involves applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[0672] The emotion engine detects that the user's emotional state is "negative," so the detailed response is adjusted to be more empathetic and encouraging.
[0673] The adjusted detailed response is sent back to the device as a JSON response.
[0674] Terminal: Receives the response from the server, extracts the response text from the response data, and displays it in the user interface.
[0675] User: Check out the tailored detailed response, "The AI learning process works by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns. Don't worry, you'll understand too." and get informed.
[0676] This system allows users to efficiently obtain the information they need in a manner that is appropriately tailored to their emotions.
[0677] Example 2
[0678] 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."
[0679] Conventional information exchange systems are required to generate appropriate responses to user questions, but they lack the ability to adjust responses based on the user's emotional state. In particular, when a user is in a negative emotional state, they are unable to generate responses that are in line with that state, which can lead to a decline in the quality of the user experience. Another issue is the inability to generate efficient responses to users' requests for more detailed information.
[0680] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.
[0681] In this invention, the server includes means for receiving a question from a user, first generative artificial intelligence means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second generative artificial intelligence means for generating a detailed response based on the initial response if it is determined that the detailed response is necessary, means for using an emotion engine for detecting the emotional state of the user, means for adjusting the initial response or the detailed response based on the emotional state, and means for returning the initial response or the adjusted detailed response to the user. This makes it possible to provide a customized response according to the emotional state of the user and efficiently generate a response even for users who request detailed information.
[0682] "Means for accepting questions from users" refers to the interface or process by which the system receives questions entered by users.
[0683] The "first generative artificial intelligence means" refers to a machine learning algorithm or generative AI model for generating an initial response based on a user's question.
[0684] "Initial response" refers to the answer that is first generated by the first generative artificial intelligence means in response to the user's question.
[0685] "Means for determining whether a detailed response is required" refers to algorithms or programs that determine whether detailed information is required based on conditions such as whether the user's question contains specific keywords.
[0686] The "second generative artificial intelligence means" refers to a machine learning algorithm or generative AI model that generates more detailed information based on the initial response.
[0687] A "detailed response" is a response containing more detailed information provided to a user based on an initial response.
[0688] An "emotion engine" refers to software or algorithms that analyze and detect a user's emotional state from input data.
[0689] "Emotional state" refers to the emotional state (e.g., neutral, negative, positive) that the user is in when entering the question.
[0690] "Response tailoring" means a process or algorithm that tailors the content of an initial response or detailed response based on the detected emotional state to make it more relevant to the user.
[0691] "Means for returning a response to the user" refers to the communications means or protocol for constructing the generated response and returning it to the user's terminal.
[0692] This invention is a user-interactive information exchange system that combines generative artificial intelligence and an emotion engine. This system allows users to efficiently acquire information while adjusting the information according to their emotional state.
[0693] System configuration
[0694] The main components of the system are:
[0695] 1. User Interface:
[0696] Provide an interface, which may be a web, mobile, or desktop application, through which users can enter questions and receive responses from the system.
[0697] 2. Server:
[0698] The server plays a central role in receiving and processing questions and emotional states from users. The server performs the following specific processes:
[0699] Questions and comments welcome:
[0700] The server receives the question and emotion data submitted by the user via an HTTP POST request.
[0701] Emotion engine detects emotional states:
[0702] The server uses an emotion engine to analyze and detect the user's emotional state from the input data. The emotion engine performs text analysis and voice analysis.
[0703] Generate the initial response:
[0704] The server passes the question to a first generative artificial intelligence means to generate an initial response, which uses machine learning algorithms to generate a rational answer to the question.
[0705] Detailed response required:
[0706] Based on the initial response, the server determines whether more information is needed, using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[0707] Generate a detailed response:
[0708] If it is determined that more detailed information is needed, the server passes the initial response to a second generative AI means to generate a detailed response, which then invokes the generative AI API again to obtain a response containing more in-depth information.
[0709] Adjusting responses based on emotions:
[0710] The server adjusts the initial and detailed responses based on the detected emotional state, for example, adjusting the response to be more empathetic and encouraging if the user's emotional state is detected to be negative.
[0711] Returning a response:
[0712] The server returns the generated initial or detailed response to the terminal in JSON format, and constructs the final response as a JSON response and returns it as an HTTP response.
[0713] 3. Terminal:
[0714] A terminal is a device that executes a user interface. It transmits questions and emotion data entered by the user to a server and displays the responses returned by the server. The terminal has a program for displaying the responses received from the server.
[0715] Specific examples
[0716] A concrete example of the operation of the system is given below.
[0717] 1. User:
[0718] The user inputs the question, "Please tell me how AI works." The emotion engine detects the user's emotional state as "neutral" as they input the text.
[0719] 2. Terminal:
[0720] The device captures this question and emotion data and sends it to the server as an HTTP POST request.
[0721] 3. Server:
[0722] The question and emotion data are received, and an initial response is generated using a first generative artificial intelligence means, for example, the initial response is generated as "AI learns using algorithms and data."
[0723] The question does not contain the keywords "details" or "more details", so no detailed response will be generated.
[0724] The emotion engine detects the user's emotional state as "neutral," so the response is sent back to the user without any adjustments.
[0725] 4. Terminal:
[0726] The device receives a response from the server and displays the answer to the user: "AI learns using algorithms and data."
[0727] Prompt Sentence Examples
[0728] Here is an example of an input prompt for a generative AI model:
[0729] "When a user asks 'Tell me how AI works,' the emotion engine detects the user's emotion as 'neutral.' Please generate a reasonable answer to this question."
[0730] Using these prompts, the generative artificial intelligence generates appropriate responses to the user's questions.
[0731] The flow of the identification process in the second embodiment will be described with reference to FIG.
[0732] Step 1:
[0733] Users can input questions through the interface, for example, in a text box on the device's web browser. When a user types, "Please tell me how AI works," the text data is sent to the device.
[0734] Step 2:
[0735] The device receives the input question and passes it to an emotion engine to detect the user's emotional state. For example, the user's emotional state may be determined to be "neutral" through text analysis. This emotion data is then sent to the server along with the input text.
[0736] Step 3:
[0737] The server receives the HTTP POST request and stores the user's question and emotion data in an internal data store for processing, so that the question and emotion state data are managed within the server.
[0738] Step 4:
[0739] The server calls the emotion engine and re-analyzes the user's emotional state from the text and voice data. For example, it confirms the previously received "neutral" emotional state. The result of this emotion analysis is recorded on the server.
[0740] Step 5:
[0741] The server passes the question to a first generative artificial intelligence means, which generates an initial response. For example, the initial response may be "AI learns using algorithms and data." This is generated using a specific prompt. An example prompt is: "When a user asks 'How does AI work?' please generate a reasonable answer."
[0742] Step 6:
[0743] The server determines whether more detailed information is needed based on the initial response and the question. For example, it analyzes whether the question contains keywords such as "details" or "more details." If the keywords are included, a more detailed response is generated. In this process, the initial response is used as input data.
[0744] Step 7:
[0745] If it is determined that more information is needed, the server passes the initial response to a second generative artificial intelligence means to generate a detailed response. For example, a detailed response such as "The AI automatically learns and predicts by combining data and algorithms" may be generated. A specific prompt sentence is again used for this generation. An example of a prompt sentence is: "Please provide a more detailed explanation based on this initial response."
[0746] Step 8:
[0747] The server adjusts the response based on the detected emotional state. Specifically, if the emotion engine detects a user's emotional state as negative, the response will be changed to something more friendly and encouraging. For example, "This technology is very interesting and worth learning."
[0748] Step 9:
[0749] The server constructs the final response in JSON format and sends it back to the device as an HTTP response, such as "AI automatically learns and predicts by combining data and algorithms. This technology is very interesting and worth learning."
[0750] Step 10:
[0751] The device parses the JSON response received from the server and displays it to the user. For example, a message might appear on the web browser saying, "AI automatically learns and predicts by combining data and algorithms. This technology is very interesting and worth learning."
[0752] (Application example 2)
[0753] 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."
[0754] Conventional information exchange systems often return mechanical responses to user questions, making it difficult to respond flexibly to the user's emotions and circumstances. Furthermore, providing information without considering the user's emotional state can potentially detract from the user experience. In particular, if a system is unable to respond appropriately to a user who is in a negative emotional state, this can lead to a decline in user satisfaction and willingness to use the system.
[0755] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for accepting a question from a user, first generative artificial intelligence means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second generative artificial intelligence means for generating a detailed response based on the initial response when it is determined that a detailed response is necessary, means for returning the initial response or the detailed response to the user, and emotion analysis means for detecting the user's emotional state and adjusting the response content based on the emotional state. This makes it possible to provide flexible and appropriate information according to the user's emotional state, thereby improving the user experience.
[0756] The "means for accepting questions from the user" is an interface for receiving questions posed by the user as input.
[0757] "First generative artificial intelligence means for generating an initial response based on a question" refers to a generative artificial intelligence algorithm for creating an initial response to a question entered by a user.
[0758] The "means for determining whether a detailed response is required based on an initial response" is an algorithm for determining whether further detailed information is required based on the generated initial response.
[0759] "A second generative artificial intelligence means for generating a detailed response based on an initial response when it is determined that a detailed response is necessary" refers to a generative artificial intelligence algorithm for generating a detailed response containing deeper information based on the initial response.
[0760] The "means for returning an initial response or a detailed response to a user" refers to a means for presenting the generated initial response or detailed response to a user.
[0761] "Emotion analysis means for detecting the emotional state of a user and adjusting the response content based on the emotional state" refers to an algorithm or device that analyzes the emotional state of a user and appropriately adjusts the response content based on that emotion.
[0762] The present invention relates to an emotion-responsive content delivery system that adjusts response content based on the emotional state of a user. A basic system configuration and processing method for implementing the present invention will be described.
[0763] System Configuration
[0764] The system consists of the following main components:
[0765] 1. User Interface: The interface through which the user enters questions and receives responses from the system. This can take the form of a web application, a mobile application, or a desktop application.
[0766] 2. Server: Provides the core functions of the system and performs the following specific processes:
[0767] Question acceptance method: Accepts questions from users. This is done via an HTTP POST request.
[0768] Sentiment analysis: Detecting the user's emotional state from their questions and other input data. Emotional state analysis is done using text and speech analysis algorithms.
[0769] Initial response generation means (first generative artificial intelligence means): Generate an initial response to the user's question. This is done using a generative artificial intelligence model (e.g., GPT-3).
[0770] Detailed response decision means: Based on the initial response, a decision is made as to whether a detailed response is necessary. This decision is made using an algorithm that detects specific keywords contained in the user's question.
[0771] Detailed response generation means (second generative artificial intelligence means): When it is determined that a detailed response is necessary, a detailed response is generated based on the initial response.
[0772] Response adjustment measures: Adjust the response content based on the results of emotion analysis. If the emotional state is negative, change the response to something more empathetic and encouraging.
[0773] Response return method: The generated response is returned to the user, also as an HTTP response.
[0774] Processing example
[0775] Here is a concrete example of how this can be done:
[0776] 1. User enters question:
[0777] The user inputs a question through the application, such as "Tell me today's news. I want some encouragement." The user's emotional state is detected as "positive."
[0778] 2. Sending questions and emotional states
[0779] The device captures this question and emotional state data and sends it to the server as an HTTP POST request.
[0780] 3. Generating an initial response
[0781] The server receives the question and generates an initial response using a generative artificial intelligence model (e.g., GPT-3). The initial response generated is "We have great news today! A Japanese athlete won a medal!"
[0782] 4. Adjusting responses based on emotional state
[0783] Since the emotion analysis means detects the emotion as "positive," the response is sent back to the user without any special adjustments.
[0784] 5. Viewing the Response
[0785] The terminal receives the response from the server and displays the generated response to the user.
[0786] Prompt Sentence Examples
[0787] "Tell me the news today. I want some encouragement."
[0788] "Tell me how AI works. Give me the details."
[0789] "I've been feeling really down lately. I'd love some words of encouragement."
[0790] This enables flexible and appropriate information provision according to the user's emotional state, improving the user experience.
[0791] The flow of the specific processing in the application example 2 will be described with reference to FIG.
[0792] Step 1:
[0793] A user inputs a question. For example, the user inputs "Tell me what's in the news today. I need some encouragement." This input is made through the user interface and captured as input data.
[0794] Step 2:
[0795] The device collects emotional data along with questions from the user. The device uses an emotion analysis method to analyze the user's emotion from the input text and determines it as "positive." This data is sent to the server as an HTTP POST request.
[0796] Step 3:
[0797] The server receives the question and emotion data. The server receives the HTTP request and extracts the question text and emotion data. This data is passed to the next processing step.
[0798] Step 4:
[0799] The server generates an initial response using a generative artificial intelligence model. The server inputs the user's question into a first generative artificial intelligence means (e.g., GPT-3) and generates an initial response such as, "We have great news for you today! A Japanese athlete won a medal!"
[0800] Step 5:
[0801] The server determines whether a detailed response is required. The server checks whether the question contains certain keywords (e.g., "details," "more details"), and in this case, since the keywords are not included, it determines that a detailed response is not required.
[0802] Step 6:
[0803] The server adjusts the response content based on the emotional state. Because the emotional state is analyzed as "positive" by the emotion analysis means, the server uses the generated initial response as is without any special adjustment.
[0804] Step 7:
[0805] The server returns the final response. The server structures the initial response it generated in JSON format and returns it to the device as an HTTP response.
[0806] Step 8:
[0807] The terminal receives the response and displays it to the user. The terminal receives the HTTP response from the server and displays the received response data in the user interface. The user sees the message, "We have great news for you today! Japanese athletes have won medals!"
[0808] 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.
[0809] 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.
[0810] 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.
[0811] [Third embodiment]
[0812] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.
[0813] 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.
[0814] 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).
[0815] 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.
[0816] 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.
[0817] 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).
[0818] 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.
[0819] 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.
[0820] 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.
[0821] 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.
[0822] 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.
[0823] 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."
[0824] The present invention relates to an interactive information exchange system with a user that uses generative artificial intelligence, allowing the user to efficiently acquire information and, in some cases, obtain in-depth, detailed information.
[0825] The main components of the system are:
[0826] 1. User Interface: Provides an interface through which users can enter questions and receive responses from the system. This can be a web, mobile, or desktop application. Users enter questions through this interface.
[0827] 2. Server: The server plays a central role in receiving and processing queries from users. The specific processes performed by the server are explained below.
[0828] Accepting a question: The server accepts a question submitted by the user via an HTTP POST request.
[0829] Generate an initial response: The server passes the received question to a first generative artificial intelligence means for generating an initial response, which uses a machine learning algorithm to generate a reasonable answer to the question.
[0830] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[0831] Generate detailed response: If it is determined that more detailed information is needed, the server passes the initial response to a second generative AI means to generate a more detailed response. This second generative AI has an algorithm that provides more in-depth information based on the initial response.
[0832] Returning a response: The server returns the generated initial or detailed response to the user interface, through which the user can receive the response.
[0833] 3. Terminal: A terminal is a device that runs the user interface. It sends questions entered by the user to the server and displays the responses sent back from the server. Below is a concrete example of how the system works.
[0834] Specific examples
[0835] User: Type the question "How does AI work?"
[0836] On your device: Capture this question and send it to your server as an HTTP POST request.
[0837] server:
[0838] A question is received, and an initial response is generated using a first generative artificial intelligence means, the initial response being "AI learns using algorithms and data."
[0839] Detects whether the question contains the keyword "details" or "more details." Since it does not contain them in this case, a detailed response will not be generated.
[0840] Sends the initial response back to the user.
[0841] Device: Receives the response from the server and displays the answer to the user: "AI learns using algorithms and data."
[0842] If the user requests more detailed information, for example by entering "Tell me how AI works. Please give me more details," the server will determine that more information is needed and will use a second generative artificial intelligence means to generate a detailed response: "The AI's learning process is achieved by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns," which will be generated and sent back to the user.
[0843] This system allows users to efficiently obtain the information they need and gain a deeper understanding without asking additional questions.
[0844] The processing flow will be explained below.
[0845] Step 1:
[0846] The user enters a question. The user enters a question using the user interface displayed on their device. For example, they might enter, "Tell me how AI works."
[0847] Step 2:
[0848] The device captures the user's question and sends it to the server as an HTTP POST request, which contains the question in JSON format.
[0849] Step 3:
[0850] The server receives the request from the device and extracts the question from the request body. Since the server uses the Flask framework, it extracts the question using request.json.
[0851] Step 4:
[0852] The server passes the extracted question to a first generative AI means (e.g., GPT-3) to generate an initial response. The server then calls the generative AI's API to obtain the response corresponding to the question.
[0853] Step 5:
[0854] The server obtains the initial response and temporarily stores it, for example, "AI uses algorithms and data to learn."
[0855] Step 6:
[0856] The server detects whether the user's question contains the keywords "details" or "more details." This is done using a string search algorithm.
[0857] Step 7:
[0858] If it is determined that the keyword is included, the server passes the initial response to the second generative artificial intelligence means to generate a detailed response based on the initial response, and then calls the API of the generative artificial intelligence again to obtain a response containing more in-depth information.
[0859] Step 8:
[0860] The server retrieves the detailed response and saves it as the final response, which is generated as follows: "The AI learning process is done by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[0861] Step 9:
[0862] The server returns the generated initial or detailed response in JSON format to the terminal, and constructs the final response as a JSON response and returns it as an HTTP response.
[0863] Step 10:
[0864] The terminal receives the response from the server, extracts the response text from the response data, and processes the text to display it on the user interface.
[0865] Step 11:
[0866] The terminal displays the extracted response text on the user interface, allowing the user to check the displayed answer and obtain the necessary information.
[0867] Example 1
[0868] 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."
[0869] Modern users want to efficiently obtain a large amount of information at once, but existing information processing systems only provide an initial response and then require users to enter a question again to obtain more detailed information. Such systems require users to pose additional questions, reducing the efficiency of information acquisition. Therefore, a new system is needed that allows users to quickly and efficiently obtain the information they need.
[0870] 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.
[0871] In this invention, the server includes a means for accepting input from a user, a generative AI model for generating an initial response based on the input, a means for determining the need for a detailed response based on the initial response, a second generative AI model for generating a detailed response when it is determined that the detailed response is needed, and a means for returning the generated response to the user. This allows the user to efficiently obtain the information they need in the first question, and to quickly obtain detailed information as needed.
[0872] A "user" is an entity that uses the system to enter questions and obtain information.
[0873] "Input" is information that refers to questions or requests that a user makes to a system.
[0874] A "server" is a central computer system that receives input from users, processes it, and generates a response.
[0875] A "generative AI model" is an artificial intelligence system that uses machine learning algorithms to generate rational answers based on user input.
[0876] An "initial response" is the first answer a generative AI model generates in response to a user's input.
[0877] The "means for determining the need for a detailed response" is an algorithm that determines whether the user needs more detailed information based on the initial response.
[0878] The "second generative AI model" is an artificial intelligence system with machine learning algorithms to provide deeper information that is used when more detailed information is deemed necessary.
[0879] A "means for returning a response" is a communication means for sending the initial response or detailed response generated by the server back to the user interface.
[0880] MODE FOR CARRYING OUT THE INVENTION
[0881] The present invention relates to an interactive information exchange system with a user that uses generative artificial intelligence, which allows the user to efficiently acquire information and, in some cases, obtain in-depth, detailed information.
[0882] The main components of the system are:
[0883] 1. User Interface: Provides an interface for users to enter questions and receive responses from the system. This user interface can take the form of a web application, mobile application, or desktop application. Users enter questions through this interface.
[0884] 2. Server: The server plays a central role in receiving and processing input from users. The specific processes performed by the server are as follows:
[0885] Accepting a Question: The server accepts input submitted by the user. This acceptance is done via an HTTP POST request.
[0886] Generate initial response: The server passes the received input to a first generative AI model to generate an initial response. This generative AI model uses machine learning algorithms to generate a reasonable answer to the input.
[0887] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's input contains the keywords "details" or "more details."
[0888] Generate a detailed response: If more detailed information is required, the server passes the initial response to a second generative AI model, which generates a more detailed response. This second generative AI model has an algorithm that provides more in-depth information based on the initial response.
[0889] Returning a response: The server returns the generated initial or detailed response to the user interface, through which the user can receive the response.
[0890] 3. Terminal: A terminal is a device that runs a user interface, sends questions entered by the user to the server, and displays the responses sent back by the server.
[0891] Specific examples
[0892] User: Type the question "How does AI work?"
[0893] On your device: Capture this question and send it to your server as an HTTP POST request.
[0894] server:
[0895] A question is received and an initial response is generated using a first generative AI model, with the initial response being "AI learns using algorithms and data."
[0896] Detects whether the question contains the keyword "details" or "more details." Since it does not contain them in this case, a detailed response will not be generated.
[0897] Sends the initial response back to the user.
[0898] Device: Receives the response from the server and displays the answer to the user: "AI learns using algorithms and data."
[0899] If the user requests more detailed information, for example by typing "Tell me how AI works. Please tell me more," the server determines that more information is needed and uses a second generative AI model to generate a detailed response. The detailed response, "AI's learning process is achieved by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns," is generated and sent back to the user. This system allows users to efficiently obtain the information they need and gain a deeper understanding without having to ask additional questions.
[0900] The flow of the identification process in the first embodiment will be described with reference to FIG.
[0901] Step 1:
[0902] The user enters a question.
[0903] Input: The question text that the user types into the system's interface. Example: "Tell me how AI works."
[0904] Specific action: Enter a question into the text box in the user interface (web app, mobile app, desktop app) and press the "Submit" button.
[0905] Output: The question text is sent to the terminal.
[0906] Step 2:
[0907] The device captures the input and sends it to the server.
[0908] Input: The question text from the user.
[0909] What happens: The user interface takes the input data, includes the question text in the body of an HTTP POST request, and sends it to a specific endpoint on the server.
[0910] Output: The question text is passed to the server in an HTTP POST request.
[0911] Step 3:
[0912] The server receives the query.
[0913] Input: The question text from the HTTP POST request.
[0914] Specific behavior: The server receives the request and extracts the question text from request.body.
[0915] Output: The question text is stored as data for processing on the server.
[0916] Step 4:
[0917] The server generates an initial response.
[0918] Input: The user's question text.
[0919] Specific operation: The question text is passed to the generative AI model as a prompt to generate an initial response. The generation process calls the AI model's API, sends data, and retrieves the response text.
[0920] Output: An initial response text is generated. Example: "AI uses algorithms and data to learn."
[0921] Step 5:
[0922] The server determines whether a detailed response is required.
[0923] Input: The initial response text and the user's question text.
[0924] What it does: It uses a keyword detection algorithm to check whether the user's question text contains the keywords "details" or "more details."
[0925] Output: The result of whether a detailed response is required. If yes, it is true; otherwise, it is false.
[0926] Step 6:
[0927] The server generates a detailed response (if required).
[0928] Input: Initial response text and whether a detailed response is required.
[0929] Specific operation: If a detailed response is required, the initial response text is passed as a prompt to the second generative AI model to generate a detailed response. This process is also performed by calling the AI model's API.
[0930] Output: A detailed response text is generated. Example: "The AI learning process involves applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[0931] Step 7:
[0932] The server sends back a response.
[0933] Input: Initial response text or detailed response text.
[0934] Specific operation: The generated response is serialized as an HTTP response and sent back to the user's device. When creating a response, it is common to use a format such as JSON.
[0935] Output: The response text is sent back to the terminal.
[0936] Step 8:
[0937] The terminal displays the response to the user.
[0938] Input: The response text sent back by the server.
[0939] What happens: The user interface receives the HTTP response, parses it, and displays the response text in a display area within the browser or application.
[0940] Output: The user can see the system's response visually, such as in a text box, saying, "AI uses algorithms and data to learn."
[0941] (Application example 1)
[0942] 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."
[0943] In today's world, users want to access news articles and information efficiently and in detail, but there are challenges: searching for information and digging deeper is time-consuming, and it is difficult to obtain all the information needed at once. Furthermore, there is a lack of systems that can instantly provide the detailed information users desire. This situation leads to a poor user experience and a decrease in the efficiency of information gathering.
[0944] 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.
[0945] In this invention, the server includes means for receiving a question from a user, first artificial intelligence generative means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second artificial intelligence generative means for generating a detailed response based on the initial response if it is determined that a detailed response is necessary, means for returning the initial response or the detailed response to the user, and means for providing news articles and information to the user. This allows the user to obtain an initial response to a question and, if necessary, to immediately receive detailed information, thereby realizing efficient and in-depth information gathering.
[0946] "User" means an individual or entity that utilizes the System to enter questions and receive responses.
[0947] A "question" is a question or request for information that a user enters into the system.
[0948] An "initial response" is the first response that a generative artificial intelligence generates based on a user's question.
[0949] "Generative AI" is an AI system that uses machine learning algorithms to generate responses to questions or information.
[0950] A "detailed response" is a response that is generated to provide more in-depth information based on the initial response.
[0951] The "server" is a central computer system that receives a user's question, generates a response using generative artificial intelligence, and sends the response back to the user.
[0952] "News Article" means news or information content provided to Users by the System.
[0953] "Information" is a general term for the data and knowledge that users want to obtain through questions.
[0954] "Details" are additional, more specific explanations or data about the specific information the user is seeking.
[0955] "Delivery method" refers to the method or technology the system uses to convey news stories and information to users.
[0956] The present invention is implemented as an interactive information exchange system using generative artificial intelligence. A detailed implementation method of this system will be described below.
[0957] System Overview
[0958] The system mainly consists of a user interface, a server, and a terminal. The user interface is implemented as a mobile application, and is a means for users to input questions and receive responses from the system.
[0959] Server Roles
[0960] The server plays a central role in receiving and processing user-submitted questions, specifically:
[0961] 1. Accepting questions
[0962] The questions entered by the user are received as HTTP POST requests, which are handled on the server using a web framework such as Flask.
[0963] 2. Generating an initial response
[0964] The server then asks the question to OpenAI's generative AI model (e.g., GPT model) and generates an initial response. The prompt is "Q: User's question\nA:".
[0965] 3. Determining whether a detailed response is necessary
[0966] After generating the initial response, the server detects whether the user's question contains specific keywords such as "details" or "more details," and determines whether a detailed response is required. This decision algorithm is implemented by a script that runs on the server side.
[0967] 4. Generate a detailed response
[0968] If the initial response does not satisfy the user's needs, the server again uses the generative AI model to generate a detailed response, this time with the prompt "Expand upon the following information: initial response."
[0969] 5. Returning the Response
[0970] The generated initial or detailed response is sent back to the user interface and displayed to the user.
[0971] Hardware and Software Details
[0972] Server: Web server using Flask
[0973] Generative AI models: OpenAI's GPT-based models (e.g., text-davinci-003)
[0974] API: OpenAI API
[0975] Specific examples
[0976] 1. User Question: "What is the history of artificial intelligence?"
[0977] 2. Server processing:
[0978] Initial response: "Artificial intelligence dates back to the 1950s."
[0979] If the user types "Tell me more," the detailed response is: "The history of artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has followed since then."
[0980] This system allows users to get an initial response to their questions and instantly receive more detailed information as needed, enabling efficient and in-depth information gathering. The following example prompts are used to set the input format for the generative AI model:
[0981] Example prompts for generating initial responses:
[0982] Q: What is the history of artificial intelligence?
[0983] A:
[0984] Example prompt for detailed response generation:
[0985] Expand upon the following information: The history of artificial intelligence dates back to the 1950s.
[0986] The flow of the specific processing in the application example 1 will be described with reference to FIG.
[0987] Step 1:
[0988] A user enters a question through a smartphone app, which the user interface captures and sends as an HTTP POST request to the server.
[0989] Input: User question (e.g., "What is the history of artificial intelligence?")
[0990] Output: HTTP POST request sent to the server
[0991] Step 2:
[0992] The server analyzes the received HTTP POST request and extracts the question.
[0993] Input: HTTP POST request
[0994] Output: Question (e.g. "What is the history of artificial intelligence?")
[0995] Step 3:
[0996] To generate an initial response based on the question, the server passes the question to a generative AI model (OpenAI's GPT-based model). The prompt used here is "Q: User's question\nA:". The generative AI generates an initial response based on this prompt.
[0997] Input: Question
[0998] Output: Initial response (e.g., "Artificial intelligence dates back to the 1950s.")
[0999] Step 4:
[1000] Based on the initial response generated, the server determines whether more information is needed, using an algorithm that detects whether the question contains the keywords "details" or "more details."
[1001] Input: Question and initial response
[1002] Output: Result of the judgment whether a detailed response is required or not (e.g., Detailed response required)
[1003] Step 5:
[1004] If a detailed response is required, the server again uses the generative AI model to generate a detailed response, with the prompt "Expand upon the following information: Initial response."
[1005] Input: Initial response
[1006] Output: Detailed response (e.g., "The history of artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has been done since then.")
[1007] Step 6:
[1008] The server returns the generated initial response or detailed response to the user interface as an HTTP response.
[1009] Input: Initial response or detailed response
[1010] Output: HTTP response
[1011] Step 7:
[1012] The user interface displays the received responses and provides information to the user.
[1013] Input: Initial response or detailed response
[1014] Output: The response shown to the user (e.g., "Artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has followed since then.")
[1015] Through these steps, users can quickly obtain detailed information efficiently.
[1016] 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.
[1017] The present invention relates to a user-interactive information exchange system that combines generative artificial intelligence and an emotion engine, allowing users to efficiently acquire information while adjusting the information according to their emotional state.
[1018] The main components of the system are:
[1019] 1. User Interface: Provides an interface through which users can enter questions and receive responses from the system. This can be a web, mobile, or desktop application. Users enter questions through this interface.
[1020] 2. Server: The server plays a central role in receiving and processing questions and emotional states from users. The specific processing performed by the server is described below.
[1021] Question and emotion reception: The server receives the question and emotion data submitted by the user via an HTTP POST request.
[1022] Detecting emotional states using an emotion engine: The server uses an emotion engine to analyze the emotional state from the user's input data and detect the emotional state. The emotion engine can perform text analysis and voice analysis.
[1023] Generate an initial response: The server passes the question to a first generative artificial intelligence means to generate an initial response. The generative artificial intelligence uses machine learning algorithms to generate a rational answer to the question.
[1024] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[1025] Generate detailed response: If it is determined that more detailed information is needed, the server passes the initial response to the second generative AI means to generate a detailed response based on the initial response, and then calls the generative AI API again to obtain a response containing more in-depth information.
[1026] Emotion-based response adjustment: The server adjusts the initial and detailed responses based on the detected emotional state. For example, if the user's emotional state is detected as negative, the server adjusts the response to be more empathetic and encouraging.
[1027] Returning a response: The server returns the generated initial or detailed response in JSON format to the terminal. It then constructs the final response as a JSON response and returns it as an HTTP response.
[1028] 3. Terminal: The terminal is the device that runs the user interface. It transmits the questions and emotion data entered by the user to the server and displays the responses sent back from the server. A specific example of the system's operation is explained below.
[1029] Specific examples
[1030] User: Enters the question "How does AI work?" and the emotion engine detects the user's emotional state as "neutral" as they enter text.
[1031] Device: Capture this question and emotion data and send it to the server as an HTTP POST request.
[1032] server:
[1033] The question and emotion data are received, and an initial response is generated using a first generative artificial intelligence means, with the initial response being "AI learns using algorithms and data."
[1034] The question does not contain the keywords "details" or "more details", so no detailed response will be generated.
[1035] The emotion engine detects the user's emotional state as "neutral," so the response is sent back to the user without any tailoring.
[1036] Device: Receives the response from the server and displays the answer to the user: "AI uses algorithms and data to learn."
[1037] If a user has a negative emotional state in the context of requesting more detailed information, for example, by typing "Tell me how AI works. Please tell me more," and the emotion engine detects that the user's emotional state is "negative," the server will not only generate more detailed information but also adjust the response to be more empathetic and encouraging. In this way, it is possible to further improve the user experience by providing a customized response according to the user's emotional state.
[1038] The processing flow will be explained below.
[1039] Step 1:
[1040] The user types in a question. The user types in "Tell me how AI works" using the user interface displayed on their device.
[1041] Step 2:
[1042] The emotion engine analyzes the user's input and detects the user's emotional state. Based on the user's text input, the emotion engine recognizes the user's emotional state as "neutral."
[1043] Step 3:
[1044] The device captures the user's question and emotional state and sends it to the server as an HTTP POST request, which contains the question and emotional state in JSON format.
[1045] Step 4:
[1046] The server receives the request from the device and extracts the question and emotional state from the request body. Since the server uses the Flask framework, it extracts data using request.json.
[1047] Step 5:
[1048] The server passes the extracted question to a first generative AI means (e.g., GPT-3) to generate an initial response. The server then calls the generative AI's API and obtains the initial response: "AI learns using algorithms and data."
[1049] Step 6:
[1050] The server detects whether the user's question contains the keywords "details" or "more details." In this case, these keywords are not included, so the initial response is sufficient.
[1051] Step 7:
[1052] The server adjusts the response appropriately based on the user's emotional state, which is detected by the emotion engine as "neutral." In this case, since the emotional state is "neutral," no particular adjustment of the response is made.
[1053] Step 8:
[1054] The server returns the generated initial response to the device as a JSON response, saying, "AI uses algorithms and data to learn."
[1055] Step 9:
[1056] The terminal receives the response from the server, extracts the response text from the response data, and processes the text to display it on the user interface.
[1057] Step 10:
[1058] The device displays the extracted response text, "AI uses algorithms and data to learn," on the user interface. The user confirms the displayed answer and obtains the information.
[1059] Example (detailed response)
[1060] User: Enters the question "How does AI work? Please tell me more." As the text is entered, the emotion engine detects the user's emotional state as "negative."
[1061] Device: Capture this question and emotion data and send it to the server as an HTTP POST request.
[1062] server:
[1063] It receives the question and emotion data and uses a first generative artificial intelligence means to generate an initial response: "AI learns using algorithms and data."
[1064] The question contains the keyword "details" or "more details," so the server determines that more information is required.
[1065] To generate a detailed response based on the initial response, the initial response is passed to a second generative artificial intelligence means, which obtains a detailed response that states, "The AI learning process involves applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[1066] The emotion engine detects that the user's emotional state is "negative," so the detailed response is adjusted to be more empathetic and encouraging.
[1067] The adjusted detailed response is sent back to the device as a JSON response.
[1068] Terminal: Receives the response from the server, extracts the response text from the response data, and displays it in the user interface.
[1069] User: Check out the tailored detailed response, "The AI learning process works by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns. Don't worry, you'll understand too." and get informed.
[1070] This system allows users to efficiently obtain the information they need in a manner that is appropriately tailored to their emotions.
[1071] Example 2
[1072] 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."
[1073] Conventional information exchange systems are required to generate appropriate responses to user questions, but they lack the ability to adjust responses based on the user's emotional state. In particular, when a user is in a negative emotional state, they are unable to generate responses that are in line with that state, which can lead to a decline in the quality of the user experience. Another issue is the inability to generate efficient responses to users' requests for more detailed information.
[1074] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.
[1075] In this invention, the server includes means for receiving a question from a user, first generative artificial intelligence means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second generative artificial intelligence means for generating a detailed response based on the initial response if it is determined that the detailed response is necessary, means for using an emotion engine for detecting the emotional state of the user, means for adjusting the initial response or the detailed response based on the emotional state, and means for returning the initial response or the adjusted detailed response to the user. This makes it possible to provide a customized response according to the emotional state of the user and efficiently generate a response even for users who request detailed information.
[1076] "Means for accepting questions from users" refers to the interface or process by which the system receives questions entered by users.
[1077] The "first generative artificial intelligence means" refers to a machine learning algorithm or generative AI model for generating an initial response based on a user's question.
[1078] "Initial response" refers to the answer that is first generated by the first generative artificial intelligence means in response to the user's question.
[1079] "Means for determining whether a detailed response is required" refers to algorithms or programs that determine whether detailed information is required based on conditions such as whether the user's question contains specific keywords.
[1080] The "second generative artificial intelligence means" refers to a machine learning algorithm or generative AI model that generates more detailed information based on the initial response.
[1081] A "detailed response" is a response containing more detailed information provided to a user based on an initial response.
[1082] An "emotion engine" refers to software or algorithms that analyze and detect a user's emotional state from input data.
[1083] "Emotional state" refers to the emotional state (e.g., neutral, negative, positive) that the user is in when entering the question.
[1084] "Response tailoring" means a process or algorithm that tailors the content of an initial response or detailed response based on the detected emotional state to make it more relevant to the user.
[1085] "Means for returning a response to the user" refers to the communications means or protocol for constructing the generated response and returning it to the user's terminal.
[1086] This invention is a user-interactive information exchange system that combines generative artificial intelligence and an emotion engine. This system allows users to efficiently acquire information while adjusting the information according to their emotional state.
[1087] System configuration
[1088] The main components of the system are:
[1089] 1. User Interface:
[1090] Provide an interface, which may be a web, mobile, or desktop application, through which users can enter questions and receive responses from the system.
[1091] 2. Server:
[1092] The server plays a central role in receiving and processing questions and emotional states from users. The server performs the following specific processes:
[1093] Questions and comments welcome:
[1094] The server receives the question and emotion data submitted by the user via an HTTP POST request.
[1095] Emotion engine detects emotional states:
[1096] The server uses an emotion engine to analyze and detect the user's emotional state from the input data. The emotion engine performs text analysis and voice analysis.
[1097] Generate the initial response:
[1098] The server passes the question to a first generative artificial intelligence means to generate an initial response, which uses machine learning algorithms to generate a rational answer to the question.
[1099] Detailed response required:
[1100] Based on the initial response, the server determines whether more information is needed, using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[1101] Generate a detailed response:
[1102] If it is determined that more detailed information is needed, the server passes the initial response to a second generative AI means to generate a detailed response, which then invokes the generative AI API again to obtain a response containing more in-depth information.
[1103] Adjusting responses based on emotions:
[1104] The server adjusts the initial and detailed responses based on the detected emotional state, for example, adjusting the response to be more empathetic and encouraging if the user's emotional state is detected to be negative.
[1105] Returning a response:
[1106] The server returns the generated initial or detailed response to the terminal in JSON format, and constructs the final response as a JSON response and returns it as an HTTP response.
[1107] 3. Terminal:
[1108] A terminal is a device that executes a user interface. It transmits questions and emotion data entered by the user to a server and displays the responses sent back from the server. The terminal has a program for displaying the responses received from the server.
[1109] Specific examples
[1110] A concrete example of the operation of the system is given below.
[1111] 1. User:
[1112] The user inputs the question, "Please tell me how AI works." The emotion engine detects the user's emotional state as "neutral" as they input the text.
[1113] 2. Terminal:
[1114] The device captures this question and emotion data and sends it to the server as an HTTP POST request.
[1115] 3. Server:
[1116] The question and emotion data are received, and an initial response is generated using a first generative artificial intelligence means, for example, the initial response is generated as "AI learns using algorithms and data."
[1117] The question does not contain the keywords "details" or "more details", so no detailed response will be generated.
[1118] The emotion engine detects the user's emotional state as "neutral," so the response is sent back to the user without any adjustments.
[1119] 4. Terminal:
[1120] The device receives a response from the server and displays the answer to the user: "AI learns using algorithms and data."
[1121] Prompt Sentence Examples
[1122] Here is an example of an input prompt for a generative AI model:
[1123] "When a user asks 'Tell me how AI works,' the emotion engine detects the user's emotion as 'neutral.' Please generate a reasonable answer to this question."
[1124] Using these prompts, the generative artificial intelligence generates appropriate responses to the user's questions.
[1125] The flow of the identification process in the second embodiment will be described with reference to FIG.
[1126] Step 1:
[1127] The user inputs questions through the interface, for example, in a text box on the device's web browser. When the user types, "Please tell me how AI works," the text data is sent to the device.
[1128] Step 2:
[1129] The device receives the input question and passes it to an emotion engine to detect the user's emotional state. For example, through text analysis, the user's emotional state may be determined to be "neutral." This emotion data is then sent to the server along with the input text.
[1130] Step 3:
[1131] The server receives the HTTP POST request and stores the user's question and emotion data in an internal data store for processing, so that the question and emotion state data are managed within the server.
[1132] Step 4:
[1133] The server calls the emotion engine and re-analyzes the user's emotional state from the text and voice data. For example, it confirms the previously received "neutral" emotional state. The result of this emotion analysis is recorded on the server.
[1134] Step 5:
[1135] The server passes the question to a first generative artificial intelligence means, which generates an initial response. For example, the initial response may be "AI uses algorithms and data to learn." This is generated using a specific prompt. An example prompt is: "When a user asks 'How does AI work?' please generate a reasonable answer."
[1136] Step 6:
[1137] The server determines whether more detailed information is needed based on the initial response and the question. For example, it analyzes whether the question contains keywords such as "details" or "more details." If the keywords are included, a more detailed response is generated. In this process, the initial response is used as input data.
[1138] Step 7:
[1139] If it is determined that more information is needed, the server passes the initial response to a second generative artificial intelligence means to generate a detailed response. For example, a detailed response such as "The AI automatically learns and predicts by combining data and algorithms" may be generated. A specific prompt sentence is again used for this generation. An example of a prompt sentence is: "Please provide a more detailed explanation based on this initial response."
[1140] Step 8:
[1141] The server adjusts the response based on the detected emotional state. Specifically, if the emotion engine detects a user's emotional state as negative, the response will be changed to something more friendly and encouraging. For example, "This technology is very interesting and worth learning."
[1142] Step 9:
[1143] The server constructs the final response in JSON format and sends it back to the device as an HTTP response, such as "AI automatically learns and predicts by combining data and algorithms. This technology is very interesting and worth learning."
[1144] Step 10:
[1145] The device parses the JSON response received from the server and displays it to the user. For example, a message might appear on the web browser saying, "AI automatically learns and predicts by combining data and algorithms. This technology is very interesting and worth learning."
[1146] (Application example 2)
[1147] 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."
[1148] Conventional information exchange systems often return mechanical responses to user questions, making it difficult to respond flexibly to the user's emotions and circumstances. Furthermore, providing information without considering the user's emotional state can potentially detract from the user experience. In particular, if a system is unable to respond appropriately to a user who is in a negative emotional state, this can lead to a decline in user satisfaction and willingness to use the system.
[1149] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for accepting a question from a user, first generative artificial intelligence means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second generative artificial intelligence means for generating a detailed response based on the initial response when it is determined that a detailed response is necessary, means for returning the initial response or the detailed response to the user, and emotion analysis means for detecting the user's emotional state and adjusting the response content based on the emotional state. This makes it possible to provide flexible and appropriate information according to the user's emotional state, thereby improving the user experience.
[1150] The "means for accepting questions from the user" is an interface for receiving questions posed by the user as input.
[1151] "First generative artificial intelligence means for generating an initial response based on a question" refers to a generative artificial intelligence algorithm for creating an initial response to a question entered by a user.
[1152] The "means for determining whether a detailed response is required based on an initial response" is an algorithm for determining whether further detailed information is required based on the generated initial response.
[1153] "A second generative artificial intelligence means for generating a detailed response based on an initial response when it is determined that a detailed response is necessary" refers to a generative artificial intelligence algorithm for generating a detailed response containing deeper information based on the initial response.
[1154] The "means for returning an initial response or a detailed response to a user" refers to a means for presenting the generated initial response or detailed response to a user.
[1155] "Emotion analysis means for detecting the emotional state of a user and adjusting the response content based on the emotional state" refers to an algorithm or device that analyzes the emotional state of a user and appropriately adjusts the response content based on that emotion.
[1156] The present invention relates to an emotion-responsive content delivery system that adjusts response content based on the emotional state of a user. A basic system configuration and processing method for implementing the present invention will be described.
[1157] System Configuration
[1158] The system consists of the following main components:
[1159] 1. User Interface: The interface through which the user enters questions and receives responses from the system. This can take the form of a web application, a mobile application, or a desktop application.
[1160] 2. Server: Provides the core functions of the system and performs the following specific processes:
[1161] Question acceptance method: Accepts questions from users. This is done via an HTTP POST request.
[1162] Sentiment analysis: Detecting the user's emotional state from their questions and other input data. Emotional state analysis is done using text and speech analysis algorithms.
[1163] Initial response generation means (first generative artificial intelligence means): Generate an initial response to the user's question. This is done using a generative artificial intelligence model (e.g., GPT-3).
[1164] Detailed response decision means: Based on the initial response, a decision is made as to whether a detailed response is necessary. This decision is made using an algorithm that detects specific keywords contained in the user's question.
[1165] Detailed response generation means (second generative artificial intelligence means): When it is determined that a detailed response is necessary, a detailed response is generated based on the initial response.
[1166] Response adjustment measures: Adjust the response content based on the results of emotion analysis. If the emotional state is negative, change the response to something more empathetic and encouraging.
[1167] Response return method: The generated response is returned to the user, also as an HTTP response.
[1168] Processing example
[1169] Here is a concrete example of how this can be done:
[1170] 1. User enters question:
[1171] The user inputs a question through the application, such as "Tell me today's news. I want some encouragement." The user's emotional state is detected as "positive."
[1172] 2. Sending questions and emotional states
[1173] The device captures this question and emotional state data and sends it to the server as an HTTP POST request.
[1174] 3. Generating an initial response
[1175] The server receives the question and generates an initial response using a generative artificial intelligence model (e.g., GPT-3). The initial response generated is "We have great news today! A Japanese athlete won a medal!"
[1176] 4. Adjusting responses based on emotional state
[1177] Since the emotion analysis means detects the emotion as "positive," the response is sent back to the user without any special adjustments.
[1178] 5. Viewing the Response
[1179] The terminal receives the response from the server and displays the generated response to the user.
[1180] Prompt Sentence Examples
[1181] "Tell me the news today. I want some encouragement."
[1182] "Tell me how AI works. Give me the details."
[1183] "I've been feeling really down lately. I'd love some words of encouragement."
[1184] This enables flexible and appropriate information provision according to the user's emotional state, improving the user experience.
[1185] The flow of the specific processing in the application example 2 will be described with reference to FIG.
[1186] Step 1:
[1187] A user inputs a question. For example, the user inputs "Tell me what's in the news today. I need some encouragement." This input is made through the user interface and captured as input data.
[1188] Step 2:
[1189] The device collects emotional data along with questions from the user. The device uses an emotion analysis method to analyze the user's emotion from the input text and determines it as "positive." This data is sent to the server as an HTTP POST request.
[1190] Step 3:
[1191] The server receives the question and emotion data. The server receives the HTTP request and extracts the question text and emotion data. This data is passed to the next processing step.
[1192] Step 4:
[1193] The server generates an initial response using a generative artificial intelligence model. The server inputs the user's question into a first generative artificial intelligence means (e.g., GPT-3) and generates an initial response such as, "We have great news for you today! A Japanese athlete won a medal!"
[1194] Step 5:
[1195] The server determines whether a detailed response is required. The server checks whether the question contains certain keywords (e.g., "details," "more details"), and in this case, since the keywords are not included, it determines that a detailed response is not required.
[1196] Step 6:
[1197] The server adjusts the response content based on the emotional state. Because the emotional state is analyzed as "positive" by the emotion analysis means, the server uses the generated initial response as is without any special adjustment.
[1198] Step 7:
[1199] The server returns the final response. The server structures the initial response it generated in JSON format and returns it to the device as an HTTP response.
[1200] Step 8:
[1201] The terminal receives the response and displays it to the user. The terminal receives the HTTP response from the server and displays the received response data in the user interface. The user sees the message, "We have great news for you today! Japanese athletes have won medals!"
[1202] 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.
[1203] 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.
[1204] 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.
[1205] [Fourth embodiment]
[1206] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.
[1207] 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.
[1208] 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).
[1209] 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.
[1210] 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.
[1211] 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).
[1212] 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.
[1213] 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.
[1214] 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.
[1215] 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.
[1216] 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.
[1217] 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.
[1218] 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."
[1219] The present invention relates to an interactive information exchange system with a user that uses generative artificial intelligence, allowing the user to efficiently acquire information and, in some cases, obtain in-depth, detailed information.
[1220] The main components of the system are:
[1221] 1. User Interface: Provides an interface through which users can enter questions and receive responses from the system. This can be a web, mobile, or desktop application. Users enter questions through this interface.
[1222] 2. Server: The server plays a central role in receiving and processing queries from users. The specific processes performed by the server are explained below.
[1223] Accepting a question: The server accepts a question submitted by the user via an HTTP POST request.
[1224] Generate an initial response: The server passes the received question to a first generative artificial intelligence means for generating an initial response, which uses a machine learning algorithm to generate a reasonable answer to the question.
[1225] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[1226] Generate detailed response: If it is determined that more detailed information is needed, the server passes the initial response to a second generative AI means to generate a more detailed response. This second generative AI has an algorithm that provides more in-depth information based on the initial response.
[1227] Returning a response: The server returns the generated initial or detailed response to the user interface, through which the user can receive the response.
[1228] 3. Terminal: A terminal is a device that runs the user interface. It sends questions entered by the user to the server and displays the responses sent back from the server. Below is a concrete example of how the system works.
[1229] Specific examples
[1230] User: Type the question "How does AI work?"
[1231] On your device: Capture this question and send it to your server as an HTTP POST request.
[1232] server:
[1233] A question is received, and an initial response is generated using a first generative artificial intelligence means, the initial response being "AI learns using algorithms and data."
[1234] Detects whether the question contains the keyword "details" or "more details." Since it does not contain them in this case, a detailed response will not be generated.
[1235] Sends the initial response back to the user.
[1236] Device: Receives the response from the server and displays the answer to the user: "AI learns using algorithms and data."
[1237] If the user requests more detailed information, for example by entering "Tell me how AI works. Please give me more details," the server will determine that more information is needed and will use a second generative artificial intelligence means to generate a detailed response: "The AI's learning process is achieved by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns," which will be generated and sent back to the user.
[1238] This system allows users to efficiently obtain the information they need and gain a deeper understanding without asking additional questions.
[1239] The processing flow will be explained below.
[1240] Step 1:
[1241] The user enters a question. The user enters a question using the user interface displayed on their device. For example, they might enter, "Tell me how AI works."
[1242] Step 2:
[1243] The device captures the user's question and sends it to the server as an HTTP POST request, which contains the question in JSON format.
[1244] Step 3:
[1245] The server receives the request from the device and extracts the question from the request body. Since the server uses the Flask framework, it extracts the question using request.json.
[1246] Step 4:
[1247] The server passes the extracted question to a first generative AI means (e.g., GPT-3) to generate an initial response. The server then calls the generative AI's API to obtain the response corresponding to the question.
[1248] Step 5:
[1249] The server obtains the initial response and temporarily stores it, for example, "AI uses algorithms and data to learn."
[1250] Step 6:
[1251] The server detects whether the user's question contains the keywords "details" or "more details." This is done using a string search algorithm.
[1252] Step 7:
[1253] If it is determined that the keyword is included, the server passes the initial response to the second generative artificial intelligence means to generate a detailed response based on the initial response, and then calls the API of the generative artificial intelligence again to obtain a response containing more in-depth information.
[1254] Step 8:
[1255] The server retrieves the detailed response and saves it as the final response, which is generated as follows: "The AI learning process is done by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[1256] Step 9:
[1257] The server returns the generated initial or detailed response in JSON format to the terminal, and constructs the final response as a JSON response and returns it as an HTTP response.
[1258] Step 10:
[1259] The terminal receives the response from the server, extracts the response text from the response data, and processes the text to display it on the user interface.
[1260] Step 11:
[1261] The terminal displays the extracted response text on the user interface, allowing the user to check the displayed answer and obtain the necessary information.
[1262] Example 1
[1263] 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."
[1264] Modern users want to efficiently obtain a large amount of information at once, but existing information processing systems only provide an initial response and then require users to enter a question again to obtain more detailed information. Such systems require users to pose additional questions, reducing the efficiency of information acquisition. Therefore, a new system is needed that allows users to quickly and efficiently obtain the information they need.
[1265] 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.
[1266] In this invention, the server includes a means for accepting input from a user, a generative AI model for generating an initial response based on the input, a means for determining the need for a detailed response based on the initial response, a second generative AI model for generating a detailed response when it is determined that the detailed response is needed, and a means for returning the generated response to the user. This allows the user to efficiently obtain the information they need in the first question, and to quickly obtain detailed information as needed.
[1267] A "user" is an entity that uses the system to enter questions and obtain information.
[1268] "Input" is information that refers to questions or requests that a user makes to a system.
[1269] A "server" is a central computer system that receives input from users, processes it, and generates a response.
[1270] A "generative AI model" is an artificial intelligence system that uses machine learning algorithms to generate rational answers based on user input.
[1271] An "initial response" is the first answer a generative AI model generates in response to a user's input.
[1272] The "means for determining the need for a detailed response" is an algorithm that determines whether the user needs more detailed information based on the initial response.
[1273] The "second generative AI model" is an artificial intelligence system with machine learning algorithms to provide deeper information that is used when more detailed information is deemed necessary.
[1274] A "means for returning a response" is a communication means for sending the initial response or detailed response generated by the server back to the user interface.
[1275] MODE FOR CARRYING OUT THE INVENTION
[1276] The present invention relates to an interactive information exchange system with a user that uses generative artificial intelligence, which allows the user to efficiently acquire information and, in some cases, obtain in-depth, detailed information.
[1277] The main components of the system are:
[1278] 1. User Interface: Provides an interface for users to enter questions and receive responses from the system. This user interface can take the form of a web application, mobile application, or desktop application. Users enter questions through this interface.
[1279] 2. Server: The server plays a central role in receiving and processing input from users. The specific processes performed by the server are as follows:
[1280] Accepting a Question: The server accepts input submitted by the user. This acceptance is done via an HTTP POST request.
[1281] Generate initial response: The server passes the received input to a first generative AI model to generate an initial response. This generative AI model uses machine learning algorithms to generate a reasonable answer to the input.
[1282] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's input contains the keywords "details" or "more details."
[1283] Generate a detailed response: If more detailed information is required, the server passes the initial response to a second generative AI model, which generates a more detailed response. This second generative AI model has an algorithm that provides more in-depth information based on the initial response.
[1284] Returning a response: The server returns the generated initial or detailed response to the user interface, through which the user can receive the response.
[1285] 3. Terminal: A terminal is a device that runs a user interface, sends questions entered by the user to the server, and displays the responses sent back by the server.
[1286] Specific examples
[1287] User: Type the question "How does AI work?"
[1288] On your device: Capture this question and send it to your server as an HTTP POST request.
[1289] server:
[1290] A question is received and an initial response is generated using a first generative AI model, with the initial response being "AI learns using algorithms and data."
[1291] Detects whether the question contains the keyword "details" or "more details." Since it does not contain them in this case, a detailed response will not be generated.
[1292] Sends the initial response back to the user.
[1293] Device: Receives the response from the server and displays the answer to the user: "AI learns using algorithms and data."
[1294] If the user requests more detailed information, for example by typing "Tell me how AI works. Please tell me more," the server determines that more information is needed and uses a second generative AI model to generate a detailed response. The detailed response, "AI's learning process is achieved by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns," is generated and sent back to the user. This system allows users to efficiently obtain the information they need and gain a deeper understanding without having to ask additional questions.
[1295] The flow of the identification process in the first embodiment will be described with reference to FIG.
[1296] Step 1:
[1297] The user enters a question.
[1298] Input: The question text that the user types into the system's interface. Example: "Tell me how AI works."
[1299] Specific action: Enter a question into the text box in the user interface (web app, mobile app, desktop app) and press the "Submit" button.
[1300] Output: The question text is sent to the terminal.
[1301] Step 2:
[1302] The device captures the input and sends it to the server.
[1303] Input: The question text from the user.
[1304] What happens: The user interface takes the input data, includes the question text in the body of an HTTP POST request, and sends it to a specific endpoint on the server.
[1305] Output: The question text is passed to the server in an HTTP POST request.
[1306] Step 3:
[1307] The server receives the query.
[1308] Input: The question text from the HTTP POST request.
[1309] Specific behavior: The server receives the request and extracts the question text from request.body.
[1310] Output: The question text is stored as data for processing on the server.
[1311] Step 4:
[1312] The server generates an initial response.
[1313] Input: The user's question text.
[1314] Specific operation: The question text is passed to the generative AI model as a prompt to generate an initial response. The generation process calls the AI model's API, sends data, and retrieves the response text.
[1315] Output: An initial response text is generated. Example: "AI uses algorithms and data to learn."
[1316] Step 5:
[1317] The server determines whether a detailed response is required.
[1318] Input: The initial response text and the user's question text.
[1319] What it does: It uses a keyword detection algorithm to check whether the user's question text contains the keywords "details" or "more details."
[1320] Output: The result of whether a detailed response is required. If yes, it is true; otherwise, it is false.
[1321] Step 6:
[1322] The server generates a detailed response (if required).
[1323] Input: Initial response text and whether a detailed response is required.
[1324] Specific operation: If a detailed response is required, the initial response text is passed as a prompt to the second generative AI model to generate a detailed response. This process is also performed by calling the AI model's API.
[1325] Output: A detailed response text is generated. Example: "The AI learning process involves applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[1326] Step 7:
[1327] The server sends back a response.
[1328] Input: Initial response text or detailed response text.
[1329] Specific operation: The generated response is serialized as an HTTP response and sent back to the user's device. When creating a response, it is common to use a format such as JSON.
[1330] Output: The response text is sent back to the terminal.
[1331] Step 8:
[1332] The terminal displays the response to the user.
[1333] Input: The response text sent back by the server.
[1334] What happens: The user interface receives the HTTP response, parses it, and displays the response text in a display area within the browser or application.
[1335] Output: The user can see the system's response visually, such as in a text box, saying, "AI uses algorithms and data to learn."
[1336] (Application example 1)
[1337] 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."
[1338] In today's world, users want to access news articles and information efficiently and in detail, but there are challenges: searching for information and digging deeper is time-consuming, and it is difficult to obtain all the information needed at once. Furthermore, there is a lack of systems that can instantly provide the detailed information users desire. This situation leads to a poor user experience and a decrease in the efficiency of information gathering.
[1339] 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.
[1340] In this invention, the server includes means for receiving a question from a user, first artificial intelligence generative means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second artificial intelligence generative means for generating a detailed response based on the initial response if it is determined that a detailed response is necessary, means for returning the initial response or the detailed response to the user, and means for providing news articles and information to the user. This allows the user to obtain an initial response to a question and, if necessary, to immediately receive detailed information, thereby realizing efficient and in-depth information gathering.
[1341] "User" means an individual or entity that utilizes the System to enter questions and receive responses.
[1342] A "question" is a question or request for information that a user enters into the system.
[1343] An "initial response" is the first response that a generative artificial intelligence generates based on a user's question.
[1344] "Generative AI" is an AI system that uses machine learning algorithms to generate responses to questions or information.
[1345] A "detailed response" is a response that is generated to provide more in-depth information based on the initial response.
[1346] The "server" is a central computer system that receives a user's question, generates a response using generative artificial intelligence, and sends the response back to the user.
[1347] "News Article" means news or information content provided to Users by the System.
[1348] "Information" is a general term for the data and knowledge that users want to obtain through questions.
[1349] "Details" are additional, more specific explanations or data about the specific information the user is seeking.
[1350] "Delivery method" refers to the method or technology the system uses to convey news stories and information to users.
[1351] The present invention is implemented as an interactive information exchange system using generative artificial intelligence. A detailed implementation method of this system will be described below.
[1352] System Overview
[1353] The system mainly consists of a user interface, a server, and a terminal. The user interface is implemented as a mobile application, and is a means for users to input questions and receive responses from the system.
[1354] Server Roles
[1355] The server plays a central role in receiving and processing user-submitted questions, specifically:
[1356] 1. Accepting questions
[1357] The questions entered by the user are received as HTTP POST requests, which are handled on the server using a web framework such as Flask.
[1358] 2. Generating an initial response
[1359] The server then asks the question to OpenAI's generative AI model (e.g., GPT model) and generates an initial response. The prompt is "Q: User's question\nA:".
[1360] 3. Determining whether a detailed response is necessary
[1361] After generating the initial response, the server detects whether the user's question contains specific keywords such as "details" or "more details," and determines whether a detailed response is required. This decision algorithm is implemented by a script that runs on the server side.
[1362] 4. Generate a detailed response
[1363] If the initial response does not satisfy the user's needs, the server again uses the generative AI model to generate a detailed response, this time with the prompt "Expand upon the following information: initial response."
[1364] 5. Returning the Response
[1365] The generated initial or detailed response is sent back to the user interface and displayed to the user.
[1366] Hardware and Software Details
[1367] Server: Web server using Flask
[1368] Generative AI models: OpenAI's GPT-based models (e.g., text-davinci-003)
[1369] API: OpenAI API
[1370] Specific examples
[1371] 1. User Question: "What is the history of artificial intelligence?"
[1372] 2. Server processing:
[1373] Initial response: "Artificial intelligence dates back to the 1950s."
[1374] If the user types "Tell me more," the detailed response is: "The history of artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has followed since then."
[1375] This system allows users to get an initial response to their questions and instantly receive more detailed information as needed, enabling efficient and in-depth information gathering. The following example prompts are used to set the input format for the generative AI model:
[1376] Example prompts for generating initial responses:
[1377] Q: What is the history of artificial intelligence?
[1378] A:
[1379] Example prompt for detailed response generation:
[1380] Expand upon the following information: The history of artificial intelligence dates back to the 1950s.
[1381] The flow of the specific processing in the application example 1 will be described with reference to FIG.
[1382] Step 1:
[1383] A user enters a question through a smartphone app, which the user interface captures and sends as an HTTP POST request to the server.
[1384] Input: User question (e.g., "What is the history of artificial intelligence?")
[1385] Output: HTTP POST request sent to the server
[1386] Step 2:
[1387] The server analyzes the received HTTP POST request and extracts the question.
[1388] Input: HTTP POST request
[1389] Output: Question (e.g. "What is the history of artificial intelligence?")
[1390] Step 3:
[1391] To generate an initial response based on the question, the server passes the question to a generative AI model (OpenAI's GPT-based model). The prompt used here is "Q: User's question\nA:". The generative AI generates an initial response based on this prompt.
[1392] Input: Question
[1393] Output: Initial response (e.g., "Artificial intelligence dates back to the 1950s.")
[1394] Step 4:
[1395] Based on the initial response generated, the server determines whether more information is needed, using an algorithm that detects whether the question contains the keywords "details" or "more details."
[1396] Input: Question and initial response
[1397] Output: Result of the judgment whether a detailed response is required or not (e.g., Detailed response required)
[1398] Step 5:
[1399] If a detailed response is required, the server again uses the generative AI model to generate a detailed response, with the prompt "Expand upon the following information: Initial response."
[1400] Input: Initial response
[1401] Output: Detailed response (e.g., "The history of artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has been done since then.")
[1402] Step 6:
[1403] The server returns the generated initial response or detailed response to the user interface as an HTTP response.
[1404] Input: Initial response or detailed response
[1405] Output: HTTP response
[1406] Step 7:
[1407] The user interface displays the received responses and provides information to the user.
[1408] Input: Initial response or detailed response
[1409] Output: The response shown to the user (e.g., "Artificial intelligence dates back to the 1950s, when the first neural network models were developed. Much research has followed since then.")
[1410] Through these steps, users can quickly obtain detailed information efficiently.
[1411] 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.
[1412] The present invention relates to a user-interactive information exchange system that combines generative artificial intelligence and an emotion engine, allowing users to efficiently acquire information while adjusting the information according to their emotional state.
[1413] The main components of the system are:
[1414] 1. User Interface: Provides an interface through which users can enter questions and receive responses from the system. This can be a web, mobile, or desktop application. Users enter questions through this interface.
[1415] 2. Server: The server plays a central role in receiving and processing questions and emotional states from users. The specific processing performed by the server is described below.
[1416] Question and emotion reception: The server receives the question and emotion data submitted by the user via an HTTP POST request.
[1417] Detecting emotional states using an emotion engine: The server uses an emotion engine to analyze the emotional state from the user's input data and detect the emotional state. The emotion engine can perform text analysis and voice analysis.
[1418] Generate an initial response: The server passes the question to a first generative artificial intelligence means to generate an initial response. The generative artificial intelligence uses machine learning algorithms to generate a rational answer to the question.
[1419] Determining if more information is needed: Based on the initial response, the server determines if more information is needed by using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[1420] Generate detailed response: If it is determined that more detailed information is needed, the server passes the initial response to the second generative AI means to generate a detailed response based on the initial response, and then calls the generative AI API again to obtain a response containing more in-depth information.
[1421] Emotion-based response adjustment: The server adjusts the initial and detailed responses based on the detected emotional state. For example, if the user's emotional state is detected as negative, the server adjusts the response to be more empathetic and encouraging.
[1422] Returning a response: The server returns the generated initial or detailed response in JSON format to the terminal. It then constructs the final response as a JSON response and returns it as an HTTP response.
[1423] 3. Terminal: The terminal is the device that runs the user interface. It transmits the questions and emotion data entered by the user to the server and displays the responses sent back from the server. A specific example of the system's operation is explained below.
[1424] Specific examples
[1425] User: Enters the question "How does AI work?" and the emotion engine detects the user's emotional state as "neutral" as they enter text.
[1426] Device: Capture this question and emotion data and send it to the server as an HTTP POST request.
[1427] server:
[1428] The question and emotion data are received, and an initial response is generated using a first generative artificial intelligence means, with the initial response being "AI learns using algorithms and data."
[1429] The question does not contain the keywords "details" or "more details", so no detailed response will be generated.
[1430] The emotion engine detects the user's emotional state as "neutral," so the response is sent back to the user without any tailoring.
[1431] Device: Receives the response from the server and displays the answer to the user: "AI uses algorithms and data to learn."
[1432] If a user has a negative emotional state in the context of requesting more detailed information, for example, by typing "Tell me how AI works. Please tell me more," and the emotion engine detects that the user's emotional state is "negative," the server will not only generate more detailed information but also adjust the response to be more empathetic and encouraging. In this way, it is possible to further improve the user experience by providing a customized response according to the user's emotional state.
[1433] The processing flow will be explained below.
[1434] Step 1:
[1435] The user types in a question. The user types in "Tell me how AI works" using the user interface displayed on their device.
[1436] Step 2:
[1437] The emotion engine analyzes the user's input and detects the user's emotional state. Based on the user's text input, the emotion engine recognizes the user's emotional state as "neutral."
[1438] Step 3:
[1439] The device captures the user's question and emotional state and sends it to the server as an HTTP POST request, which contains the question and emotional state in JSON format.
[1440] Step 4:
[1441] The server receives the request from the device and extracts the question and emotional state from the request body. Since the server uses the Flask framework, it extracts data using request.json.
[1442] Step 5:
[1443] The server passes the extracted question to a first generative AI means (e.g., GPT-3) to generate an initial response. The server then calls the generative AI's API and obtains the initial response: "AI learns using algorithms and data."
[1444] Step 6:
[1445] The server detects whether the user's question contains the keywords "details" or "more details." In this case, these keywords are not included, so the initial response is sufficient.
[1446] Step 7:
[1447] The server adjusts the response appropriately based on the user's emotional state, which is detected by the emotion engine as "neutral." In this case, since the emotional state is "neutral," no particular adjustment of the response is made.
[1448] Step 8:
[1449] The server returns the generated initial response to the device as a JSON response, saying, "AI uses algorithms and data to learn."
[1450] Step 9:
[1451] The terminal receives the response from the server, extracts the response text from the response data, and processes the text to display it on the user interface.
[1452] Step 10:
[1453] The device displays the extracted response text, "AI uses algorithms and data to learn," on the user interface. The user confirms the displayed answer and obtains the information.
[1454] Example (detailed response)
[1455] User: Enters the question "How does AI work? Please tell me more." As the text is entered, the emotion engine detects the user's emotional state as "negative."
[1456] Device: Capture this question and emotion data and send it to the server as an HTTP POST request.
[1457] server:
[1458] It receives the question and emotion data and uses a first generative artificial intelligence means to generate an initial response: "AI learns using algorithms and data."
[1459] The question contains the keyword "details" or "more details," so the server determines that more information is required.
[1460] To generate a detailed response based on the initial response, the initial response is passed to a second generative artificial intelligence means, which obtains a detailed response that states, "The AI learning process involves applying various algorithms (e.g., neural networks and decision trees) to data to find patterns."
[1461] The emotion engine detects that the user's emotional state is "negative," so the detailed response is adjusted to be more empathetic and encouraging.
[1462] The adjusted detailed response is sent back to the device as a JSON response.
[1463] Terminal: Receives the response from the server, extracts the response text from the response data, and displays it in the user interface.
[1464] User: Check out the tailored detailed response, "The AI learning process works by applying various algorithms (e.g., neural networks and decision trees) to data to find patterns. Don't worry, you'll understand too." and get informed.
[1465] This system allows users to efficiently obtain the information they need in a manner that is appropriately tailored to their emotions.
[1466] Example 2
[1467] 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."
[1468] Conventional information exchange systems are required to generate appropriate responses to user questions, but they lack the ability to adjust responses based on the user's emotional state. In particular, when a user is in a negative emotional state, they are unable to generate responses that are in line with that state, which can lead to a decline in the quality of the user experience. Another issue is the inability to generate efficient responses to users' requests for more detailed information.
[1469] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.
[1470] In this invention, the server includes means for receiving a question from a user, first generative artificial intelligence means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second generative artificial intelligence means for generating a detailed response based on the initial response if it is determined that the detailed response is necessary, means for using an emotion engine for detecting the emotional state of the user, means for adjusting the initial response or the detailed response based on the emotional state, and means for returning the initial response or the adjusted detailed response to the user. This makes it possible to provide a customized response according to the emotional state of the user and efficiently generate a response even for users who request detailed information.
[1471] "Means for accepting questions from users" refers to the interface or process by which the system receives questions entered by users.
[1472] The "first generative artificial intelligence means" refers to a machine learning algorithm or generative AI model for generating an initial response based on a user's question.
[1473] "Initial response" refers to the answer that is first generated by the first generative artificial intelligence means in response to the user's question.
[1474] "Means for determining whether a detailed response is required" refers to algorithms or programs that determine whether detailed information is required based on conditions such as whether the user's question contains specific keywords.
[1475] The "second generative artificial intelligence means" refers to a machine learning algorithm or generative AI model that generates more detailed information based on the initial response.
[1476] A "detailed response" is a response containing more detailed information provided to a user based on an initial response.
[1477] An "emotion engine" refers to software or algorithms that analyze and detect a user's emotional state from input data.
[1478] "Emotional state" refers to the emotional state (e.g., neutral, negative, positive) that the user is in when entering the question.
[1479] "Response tailoring" means a process or algorithm that tailors the content of an initial response or detailed response based on the detected emotional state to make it more relevant to the user.
[1480] "Means for returning a response to the user" refers to the communications means or protocol for constructing the generated response and returning it to the user's terminal.
[1481] This invention is a user-interactive information exchange system that combines generative artificial intelligence and an emotion engine. This system allows users to efficiently acquire information while adjusting the information according to their emotional state.
[1482] System configuration
[1483] The main components of the system are:
[1484] 1. User Interface:
[1485] Provide an interface, which may be a web, mobile, or desktop application, through which users can enter questions and receive responses from the system.
[1486] 2. Server:
[1487] The server plays a central role in receiving and processing questions and emotional states from users. The server performs the following specific processes:
[1488] Questions and comments welcome:
[1489] The server receives the question and emotion data submitted by the user via an HTTP POST request.
[1490] Emotion engine detects emotional states:
[1491] The server uses an emotion engine to analyze and detect the user's emotional state from the input data. The emotion engine performs text analysis and voice analysis.
[1492] Generate the initial response:
[1493] The server passes the question to a first generative artificial intelligence means to generate an initial response, which uses machine learning algorithms to generate a rational answer to the question.
[1494] Detailed response required:
[1495] Based on the initial response, the server determines whether more information is needed, using an algorithm that detects whether the user's question contains the keywords "details" or "more details."
[1496] Generate a detailed response:
[1497] If it is determined that more detailed information is needed, the server passes the initial response to a second generative AI means to generate a detailed response, which then invokes the generative AI API again to obtain a response containing more in-depth information.
[1498] Adjusting responses based on emotions:
[1499] The server adjusts the initial and detailed responses based on the detected emotional state, for example, adjusting the response to be more empathetic and encouraging if the user's emotional state is detected to be negative.
[1500] Returning a response:
[1501] The server returns the generated initial or detailed response to the terminal in JSON format, and constructs the final response as a JSON response and returns it as an HTTP response.
[1502] 3. Terminal:
[1503] A terminal is a device that executes a user interface. It transmits questions and emotion data entered by the user to a server and displays the responses sent back from the server. The terminal has a program for displaying the responses received from the server.
[1504] Specific examples
[1505] A concrete example of the operation of the system is given below.
[1506] 1. User:
[1507] The user inputs the question, "Please tell me how AI works." The emotion engine detects the user's emotional state as "neutral" as they input the text.
[1508] 2. Terminal:
[1509] The device captures this question and emotion data and sends it to the server as an HTTP POST request.
[1510] 3. Server:
[1511] The question and emotion data are received, and an initial response is generated using a first generative artificial intelligence means, for example, the initial response is "AI learns using algorithms and data."
[1512] The question does not contain the keywords "details" or "more details", so no detailed response will be generated.
[1513] The emotion engine detects the user's emotional state as "neutral," so the response is sent back to the user without any adjustments.
[1514] 4. Terminal:
[1515] The device receives a response from the server and displays the answer to the user: "AI learns using algorithms and data."
[1516] Prompt Sentence Examples
[1517] Here is an example of an input prompt for a generative AI model:
[1518] "When a user asks 'Tell me how AI works,' the emotion engine detects the user's emotion as 'neutral.' Please generate a reasonable answer to this question."
[1519] Using these prompts, the generative artificial intelligence generates appropriate responses to the user's questions.
[1520] The flow of the identification process in the second embodiment will be described with reference to FIG.
[1521] Step 1:
[1522] The user inputs questions through the interface, for example, in a text box on the device's web browser. When the user types, "Please tell me how AI works," the text data is sent to the device.
[1523] Step 2:
[1524] The device receives the input question and passes it to an emotion engine to detect the user's emotional state. For example, through text analysis, the user's emotional state may be determined to be "neutral." This emotion data is then sent to the server along with the input text.
[1525] Step 3:
[1526] The server receives the HTTP POST request and stores the user's question and emotion data in an internal data store for processing, so that the question and emotion state data are managed within the server.
[1527] Step 4:
[1528] The server calls the emotion engine and re-analyzes the user's emotional state from the text and voice data. For example, it confirms the previously received "neutral" emotional state. The result of this emotion analysis is recorded on the server.
[1529] Step 5:
[1530] The server passes the question to a first generative artificial intelligence means, which generates an initial response. For example, the initial response may be "AI learns using algorithms and data." This is generated using a specific prompt. An example prompt is: "When a user asks 'How does AI work?' please generate a reasonable answer."
[1531] Step 6:
[1532] The server determines whether more detailed information is needed based on the initial response and the question. For example, it analyzes whether the question contains keywords such as "details" or "more details." If the keywords are included, a more detailed response is generated. In this process, the initial response is used as input data.
[1533] Step 7:
[1534] If it is determined that more information is needed, the server passes the initial response to a second generative artificial intelligence means to generate a detailed response. For example, a detailed response such as "The AI automatically learns and predicts by combining data and algorithms" may be generated. A specific prompt sentence is again used for this generation. An example of a prompt sentence is: "Please provide a more detailed explanation based on this initial response."
[1535] Step 8:
[1536] The server adjusts the response based on the detected emotional state. Specifically, if the emotion engine detects a user's emotional state as negative, the response will be changed to something more friendly and encouraging. For example, "This technology is very interesting and worth learning."
[1537] Step 9:
[1538] The server constructs the final response in JSON format and sends it back to the device as an HTTP response, such as "AI automatically learns and predicts by combining data and algorithms. This technology is very interesting and worth learning."
[1539] Step 10:
[1540] The device parses the JSON response received from the server and displays it to the user. For example, a message might appear on the web browser saying, "AI automatically learns and predicts by combining data and algorithms. This technology is very interesting and worth learning."
[1541] (Application example 2)
[1542] 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."
[1543] Conventional information exchange systems often return mechanical responses to user questions, making it difficult to respond flexibly to the user's emotions and circumstances. Furthermore, providing information without considering the user's emotional state can potentially detract from the user experience. In particular, if a system is unable to respond appropriately to a user who is in a negative emotional state, this can lead to a decline in user satisfaction and willingness to use the system.
[1544] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for accepting a question from a user, first generative artificial intelligence means for generating an initial response based on the question, means for determining whether a detailed response is necessary based on the initial response, second generative artificial intelligence means for generating a detailed response based on the initial response when it is determined that a detailed response is necessary, means for returning the initial response or the detailed response to the user, and emotion analysis means for detecting the user's emotional state and adjusting the response content based on the emotional state. This makes it possible to provide flexible and appropriate information according to the user's emotional state, thereby improving the user experience.
[1545] The "means for accepting questions from the user" is an interface for receiving questions posed by the user as input.
[1546] "First generative artificial intelligence means for generating an initial response based on a question" refers to a generative artificial intelligence algorithm for creating an initial response to a question entered by a user.
[1547] The "means for determining whether a detailed response is required based on an initial response" is an algorithm for determining whether further detailed information is required based on the generated initial response.
[1548] "A second generative artificial intelligence means for generating a detailed response based on an initial response when it is determined that a detailed response is necessary" refers to a generative artificial intelligence algorithm for generating a detailed response containing deeper information based on the initial response.
[1549] The "means for returning an initial response or a detailed response to a user" refers to a means for presenting the generated initial response or detailed response to a user.
[1550] "Emotion analysis means for detecting the emotional state of a user and adjusting the response content based on the emotional state" refers to an algorithm or device that analyzes the emotional state of a user and appropriately adjusts the response content based on that emotion.
[1551] The present invention relates to an emotion-responsive content delivery system that adjusts response content based on the emotional state of a user. A basic system configuration and processing method for implementing the present invention will be described.
[1552] System Configuration
[1553] The system consists of the following main components:
[1554] 1. User Interface: The interface through which the user enters questions and receives responses from the system. This can take the form of a web application, a mobile application, or a desktop application.
[1555] 2. Server: Provides the core functions of the system and performs the following specific processes:
[1556] Question acceptance method: Accepts questions from users. This is done via an HTTP POST request.
[1557] Sentiment analysis: Detecting the user's emotional state from their questions and other input data. Emotional state analysis is done using text and speech analysis algorithms.
[1558] Initial response generation means (first generative artificial intelligence means): Generate an initial response to the user's question. This is done using a generative artificial intelligence model (e.g., GPT-3).
[1559] Detailed response decision means: Based on the initial response, a decision is made as to whether a detailed response is necessary. This decision is made using an algorithm that detects specific keywords contained in the user's question.
[1560] Detailed response generation means (second generative artificial intelligence means): When it is determined that a detailed response is necessary, a detailed response is generated based on the initial response.
[1561] Response adjustment measures: Adjust the response content based on the results of emotion analysis. If the emotional state is negative, change the response to something more empathetic and encouraging.
[1562] Response return method: The generated response is returned to the user, also as an HTTP response.
[1563] Processing example
[1564] Here is a concrete example of how this can be done:
[1565] 1. User enters question:
[1566] The user inputs a question through the application, such as "Tell me today's news. I want some encouragement." The user's emotional state is detected as "positive."
[1567] 2. Sending questions and emotional states
[1568] The device captures this question and emotional state data and sends it to the server as an HTTP POST request.
[1569] 3. Generating an initial response
[1570] The server receives the question and generates an initial response using a generative artificial intelligence model (e.g., GPT-3). The initial response generated is "We have great news today! A Japanese athlete won a medal!"
[1571] 4. Adjusting responses based on emotional state
[1572] Since the emotion analysis means detects the emotion as "positive," the response is sent back to the user without any special adjustments.
[1573] 5. Viewing the Response
[1574] The terminal receives the response from the server and displays the generated response to the user.
[1575] Prompt Sentence Examples
[1576] "Tell me the news today. I want some encouragement."
[1577] "Tell me how AI works. Give me the details."
[1578] "I've been feeling really down lately. I'd love some words of encouragement."
[1579] This makes it possible to provide flexible and appropriate information according to the user's emotional state, improving the user experience.
[1580] The flow of the specific processing in the application example 2 will be described with reference to FIG.
[1581] Step 1:
[1582] A user inputs a question. For example, the user inputs "Tell me what's in the news today. I need some encouragement." This input is made through the user interface and captured as input data.
[1583] Step 2:
[1584] The device collects emotional data along with questions from the user. The device uses an emotion analysis method to analyze the user's emotion from the input text and determines it as "positive." This data is sent to the server as an HTTP POST request.
[1585] Step 3:
[1586] The server receives the question and emotion data. The server receives the HTTP request and extracts the question text and emotion data. This data is passed to the next processing step.
[1587] Step 4:
[1588] The server generates an initial response using a generative artificial intelligence model. The server inputs the user's question into a first generative artificial intelligence means (e.g., GPT-3) and generates an initial response such as, "We have great news for you today! A Japanese athlete won a medal!"
[1589] Step 5:
[1590] The server determines whether a detailed response is required. The server checks whether the question contains certain keywords (e.g., "details," "more details"), and in this case, since the keywords are not included, it determines that a detailed response is not required.
[1591] Step 6:
[1592] The server adjusts the response content based on the emotional state. Because the emotional state is analyzed as "positive" by the emotion analysis means, the server uses the generated initial response as is without any special adjustment.
[1593] Step 7:
[1594] The server returns the final response. The server structures the initial response it generated in JSON format and returns it to the terminal as an HTTP response.
[1595] Step 8:
[1596] The terminal receives the response and displays it to the user. The terminal receives the HTTP response from the server and displays the received response data in the user interface. The user sees the message, "We have great news for you today! Japanese athletes have won medals!"
[1597] 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.
[1598] 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.
[1599] 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.
[1600] 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.
[1601] 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.
[1602] 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.
[1603] 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).
[1604] 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.
[1605] 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."
[1606] 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.
[1607] 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).
[1608] 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.
[1609] 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.
[1610] 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.
[1611] 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.
[1612] 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.
[1613] 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.
[1614] 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.
[1615] 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.
[1616] 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.
[1617] 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.
[1618] The following is further disclosed regarding the above embodiment.
[1619] (Claim 1)
[1620] A means of accepting questions from users;
[1621] a first generative artificial intelligence means for generating an initial response based on the question;
[1622] means for determining whether a detailed response is required based on the initial response;
[1623] a second artificial intelligence generating means for generating a detailed response based on the initial response when it is determined that the detailed response is necessary;
[1624] means for returning said initial response or detailed response to a user;
[1625] A system including:
[1626] (Claim 2)
[1627] 2. The system according to claim 1, wherein the means for determining whether a detailed response is required makes the determination by detecting whether a specific keyword is included in the user's question.
[1628] (Claim 3)
[1629] 2. The system according to claim 1, wherein the first generative artificial intelligence means and the second generative artificial intelligence means use the same generative artificial intelligence algorithm.
[1630] "Example 1"
[1631] (Claim 1)
[1632] a means for accepting input from a user;
[1633] a server that processes the input;
[1634] a generative AI model that generates an initial response based on the input;
[1635] means for determining the need for a detailed response based on the initial response;
[1636] a second generative AI model that generates a detailed response when it is determined that the detailed response is necessary; and
[1637] means for returning the generated response to a user;
[1638] A system including:
[1639] (Claim 2)
[1640] 2. The system of claim 1, wherein the means for determining the need for a detailed response includes an algorithm for detecting whether a particular keyword is included in the user's input.
[1641] (Claim 3)
[1642] The system of claim 1, wherein the generative AI model and the second generative AI model use the same generative artificial intelligence algorithm.
[1643] "Application Example 1"
[1644] (Claim 1)
[1645] A means of accepting questions from users;
[1646] a first generative artificial intelligence means for generating an initial response based on the question;
[1647] means for determining whether a detailed response is required based on the initial response;
[1648] a second artificial intelligence generating means for generating a detailed response based on the initial response when it is determined that the detailed response is necessary;
[1649] means for returning said initial response or detailed response to a user;
[1650] a means of providing news articles and information to users;
[1651] A system including:
[1652] (Claim 2)
[1653] 2. The system according to claim 1, wherein the means for determining whether a detailed response is required makes the determination by detecting whether a specific keyword is included in the user's question.
[1654] (Claim 3)
[1655] 2. The system according to claim 1, wherein the first generative artificial intelligence means and the second generative artificial intelligence means use the same generative artificial intelligence algorithm.
[1656] "Example 2: Combining Emotion Engines"
[1657] (Claim 1)
[1658] A means of accepting questions from users;
[1659] a first generative artificial intelligence means for generating an initial response based on the question;
[1660] means for determining whether a detailed response is required based on the initial response;
[1661] a second artificial intelligence generating means for generating a detailed response based on the initial response when it is determined that the detailed response is necessary;
[1662] a means for using an emotion engine to detect an emotional state of a user;
[1663] means for adjusting an initial response or a detailed response based on said emotional state;
[1664] means for returning said initial response or tailored detailed response to a user;
[1665] A system including:
[1666] (Claim 2)
[1667] 2. The system according to claim 1, wherein the means for determining whether a detailed response is required makes the determination by detecting whether a specific keyword is included in the user's question.
[1668] (Claim 3)
[1669] 2. The system according to claim 1, wherein the first generative artificial intelligence means and the second generative artificial intelligence means use the same generative artificial intelligence algorithm.
[1670] "Application example 2 when combining emotion engines"
[1671] (Claim 1)
[1672] A means of accepting questions from users;
[1673] a first generative artificial intelligence means for generating an initial response based on the question;
[1674] means for determining whether a detailed response is required based on the initial response;
[1675] a second artificial intelligence generating means for generating a detailed response based on the initial response when it is determined that the detailed response is necessary;
[1676] means for returning said initial response or detailed response to a user;
[1677] emotion analysis means for detecting an emotional state of a user and adjusting a response content based on the emotional state;
[1678] A system including:
[1679] (Claim 2)
[1680] The system of claim 1, wherein the means for determining whether a detailed response is necessary makes the determination by detecting whether a specific keyword is included in the user's question, and when the emotional state detected by the emotion analysis means exceeds a specific criterion.
[1681] (Claim 3)
[1682] 2. The system according to claim 1, wherein the first generative artificial intelligence means and the second generative artificial intelligence means use the same generative artificial intelligence algorithm. [Explanation of symbols]
[1683] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>
Claims
1. A means of accepting questions from users; a first generative artificial intelligence means for generating an initial response based on the question; means for determining whether a detailed response is required based on the initial response; a second artificial intelligence generating means for generating a detailed response based on the initial response when it is determined that the detailed response is necessary; means for returning said initial response or detailed response to a user; A system including:
2. 2. The system according to claim 1, wherein the means for determining whether a detailed response is required makes the determination by detecting whether a specific keyword is included in the user's question.
3. 2. The system according to claim 1, wherein said first generative artificial intelligence means and said second generative artificial intelligence means use the same generative artificial intelligence algorithm.
Citation Information
Patent Citations
Persona chatbot control method and system
JP2022180282A