system

Abstract

Provide a system. A database for storing private information within a company; a terminal for receiving a question from a user; a server that generates answers to questions based on the database and public information; a Retrieval Augmented Generation model that combines private and public information in the database to generate answers to questions; A system with a means for providing generated answers to a user.

Description

[Technical Field]

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

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

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

[0004] Current technology does not provide a means to provide quick answers to questions that utilize internal, non-public information. Therefore, when responding to inquiries about system specifications or source code, personnel must either read through a vast amount of documentation or consult with other experts. This makes the search and response process extremely time-consuming and labor-intensive, resulting in reduced work efficiency. The present invention was devised to solve these problems. [Means for solving the problem]

[0005] The present invention solves the above problems by the following means: A system including a database that stores internal, non-public information from a company, a terminal that receives questions from users, a server that generates answers to the questions based on the database and public information, a Retrieval Augmented Generation (RAG) model that generates answers to questions by combining the non-public information in the database with public information, and a means for providing the generated answers to users. This system enables quick and accurate answers that utilize internal, non-public information from a company, significantly improving the work efficiency of system personnel.

[0006] "Internal non-public information" refers to highly confidential information that is held internally by a company and is not publicly available, and primarily includes system specifications and source code.

[0007] A "database" is a collection of data for efficiently storing, searching, and managing specific information, and in the present invention is used to store private information within a company.

[0008] The "terminal that receives a question from a user" is a device and software that allows a user to input a question to the system, and has the function of transferring the question to the server.

[0009] A "server" is a computer system that receives questions from users, generates answers based on a database and public information, and transmits the generated answers to the user terminals.

[0010] "Public Information" is information available from the Internet or other publicly available sources that may be combined with non-public information to generate an answer to a question.

[0011] The "Retrieval Augmented Generation (RAG) model" is a machine learning model that searches for related information (Retrieval) and generates answers (Generation) based on that information when generating answers to questions.

[0012] The "means for providing the generated answer to the user" is a function including hardware and software for transmitting the answer generated by the server to the user terminal.

[0013] The "means for retrieving non-public information related to the question from the database" is a function that includes algorithms and processing procedures for searching for non-public information in the database based on the content of the question and extracting relevant information.

[0014] "System specifications and source code" refers to the technical documentation and program code that describes in detail how the system operates and is part of the company's internal, non-public information. [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 a system that generates answers to user questions by combining private and public information within a company. This system is composed of a database, a terminal, and a server.

[0037] First, the server collects private information from within the company and stores it in a database. Private information includes system source code and specifications. The database is designed to efficiently manage and search private information.

[0038] When using the system, the user inputs a question through the terminal. The terminal receives the user's question and transfers it to the server. This question triggers a match with the data in the system.

[0039] The server analyzes the question received from the device and searches for relevant data from private information in the database using a search algorithm based on the content of the question. In addition, the server also retrieves public information and uses these sources in combination to generate an answer to the question.

[0040] To generate answers, a technology called the Retrieval Augmented Generation (RAG) model is used. The RAG model first retrieves information related to the question from databases and public information, and then generates an answer based on that information. This model makes it possible to provide the best answer to the question.

[0041] The generated answer is sent from the server to the terminal and displayed to the user, who can instantly obtain specific answers about the system through the terminal.

[0042] As a specific example, if a user asks, "What does the add function in utils.py do?", the server searches the database for the appropriate source code sections and specifications, combines them with public information, and generates an answer. In this case, the server generates the answer, "The add function takes two arguments and returns their sum," and provides it to the user through the terminal.

[0043] By implementing this system, system personnel can quickly and accurately obtain information about system specifications and source code, which contributes to a significant improvement in work efficiency.

[0044] The processing flow will be explained below.

[0045] Step 1:

[0046] The server collects private information from within the company and stores it in a database. The private information includes system source code and specifications. The database is designed to efficiently manage and search private information.

[0047] Step 2:

[0048] The user inputs a question about the system through a terminal, which is processed as a string of characters.

[0049] Step 3:

[0050] The terminal receives a question entered by the user and forwards the question to the server, which then formats the user's input appropriately and sends it to the server.

[0051] Step 4:

[0052] The server analyzes the question received from the device using natural language processing (NLP) technology to extract the intent and keywords of the question.

[0053] Step 5:

[0054] Based on the parsed question, the server searches for relevant data from private information in a database using algorithms to identify sections of source code or specifications that match the query.

[0055] Step 6:

[0056] The server also retrieves public information related to the question, which is collected from technical documents, forums, and other sources on the Internet.

[0057] Step 7:

[0058] The server generates answers to questions by combining private information in the database with the public information it has retrieved, using a Retrieval Augmented Generation (RAG) model.

[0059] Step 8:

[0060] The server then sends the generated answer to the terminal, where it is properly formatted and presented in a user-friendly form.

[0061] Step 9:

[0062] The terminal displays the answer received from the server to the user, and the user can instantly obtain specific answers about the system through the terminal.

[0063] This series of steps allows the user to quickly and accurately obtain information about the system specifications and source code.

[0064] Example 1

[0065] 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."

[0066] There is a lot of private information within companies, and efficient use of this information can improve business efficiency, but previous systems lacked the means to properly integrate this information and generate optimal answers to user questions.In addition, there was no way to properly combine public and private information, which meant that users could not be provided with sufficient information.

[0067] 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.

[0068] In this invention, the server includes information storage means, question receiving means, answer generating means, information integrating means, and answer providing means, which allows the server to combine private and public information within the company to generate optimal answers and provide them to users.

[0069] "Information storage means" refers to systems and databases for collecting, efficiently storing, and managing non-public information within a company.

[0070] The "question receiving means" refers to an interface for receiving a question from a user and transferring it to the server in an appropriate format.

[0071] "Answer generation means" refers to algorithms and technologies for generating optimal answers based on received questions, using private and public information within the company.

[0072] "Information synthesis means" refers to means for combining private and public information to generate an integrated answer to a question.

[0073] The "answer providing means" refers to a means for presenting the generated answer to the user.

[0074] The present invention relates to a system that generates answers to user questions by combining private and public information within a company. This system is composed of an information storage means, a question receiving means, an answer generating means, an information integration means, and an answer providing means.

[0075] First, the server collects private information from within the company and stores it in a database, which is a means of storing information.Private information includes system program code and specifications.This database is designed to efficiently manage and search private information.

[0076] The user inputs a question through the terminal. The terminal receives the user's question and sends it to the server as a question receiving means. This question becomes a trigger for matching with the data inside the system.

[0077] The server uses a natural language processing library (e.g., spaCy or Transformers) to analyze the question, and then searches for relevant private information from a database, which is an information storage means, based on the analyzed question.

[0078] Additionally, the server uses APIs on the Internet to gather public information, for example, by retrieving the required information from specific technology sites or open source repositories.

[0079] After obtaining the private and public information, the server uses the Retrieval Augmented Generation (RAG) model as an information integration method to integrate each piece of information and generate the optimal answer. This model is an advanced algorithm that generates answers based on information related to the question.

[0080] The generated answer is sent from the server to the terminal, and the terminal displays it to the user as a means of providing the answer. As a specific example, if a user asks, "What does the add function in utils.py do?", the server searches the database for the appropriate program code section and specification, and combines it with public information to generate an answer. In this case, the server generates the answer, "The add function takes two arguments and returns their sum," and provides it to the user via the terminal.

[0081] An example of a prompt to input to a generative AI model is:

[0082] "A user has a system-related question. The question is: 'What does the add function in utils.py do?' Please provide a detailed answer using relevant parts of your database and publicly available information."

[0083] As described above, the present invention realizes a system that utilizes both private and public information within a company in an integrated manner to provide quick and accurate answers to user questions. This allows system personnel to efficiently obtain appropriate information, and is expected to significantly improve work efficiency.

[0084] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0085] Step 1:

[0086] The user inputs a question through the terminal.

[0087] (Input) User question (e.g., "What does the add function in utils.py do?")

[0088] (Operation) A user enters a question into the question input form using a dedicated application or a web browser. The user enters specific questions about the question in text format.

[0089] (Output) The user's question is saved in text format on the terminal.

[0090] Step 2:

[0091] The terminal sends a question to the server.

[0092] (Input) User question text (e.g., "What does the add function in utils.py do?")

[0093] (Behavior) When the user clicks the submit button, an HTTP POST request is generated and sent to the server's API endpoint (e.g., "api.example.com / query").

[0094] (Output) The server receives the user's question text.

[0095] Step 3:

[0096] The server receives and parses the query.

[0097] (Input) User question text (e.g., "What does the add function in utils.py do?")

[0098] (Operation) The server uses a natural language processing library (e.g., spaCy, Transformers) to analyze the question text received. It performs processes such as text tokenization, part-of-speech tagging, and semantic analysis. As a result of the analysis, "utils.py" and the "add function" are identified.

[0099] (Output) The parsed question data (e.g., "File name: utils.py", "Function name: add function") is generated.

[0100] Step 4:

[0101] The server searches the database for relevant non-public information.

[0102] (Input) Parsed question data (e.g., "File name: utils.py", "Function name: add function")

[0103] (Operation) The server uses an SQL query to retrieve the relevant program code or specification section from a database (e.g., MySQL®). Specifically, it executes the query "SELECT FROM SourceCode WHERE fileName='utils.py' AND functionName='add'".

[0104] (Output) Related private information (e.g., "source code of the add function" and "specification of the add function") is obtained.

[0105] Step 5:

[0106] The server collects public information from external APIs.

[0107] (Input) Analyzed question data and non-public information

[0108] (Operation) The server calls APIs on the Internet to collect public information. For example, it uses the GitHub API to obtain public repository information related to the "utils.py" file and the "add function."

[0109] (Output) Public information (e.g., "Related information about the GitHub repository") is obtained.

[0110] Step 6:

[0111] The server uses the RAG model to generate answers to questions.

[0112] (Input) Private and Public Information

[0113] (Operation) The server inputs the acquired private and public information into the RAG model and generates an integrated answer. Using the generative AI model, the optimal answer to the question (e.g., "The add function takes two arguments and returns their sum") is generated.

[0114] (Output) A generated answer (e.g., "The add function takes two arguments and returns their sum") is generated.

[0115] Step 7:

[0116] The server generates a response and sends it to the terminal.

[0117] (Input) Generated answer (e.g., "The add function takes two arguments and returns their sum")

[0118] (Operation) The server sends the generated answer to the terminal as an HTTP response. The response data is usually sent in JSON format.

[0119] (Output) The terminal receives the generated response.

[0120] Step 8:

[0121] The terminal displays the answer to the user.

[0122] (Input) Generated answer (e.g., "The add function takes two arguments and returns their sum")

[0123] (Operation) The device analyzes the received answer and displays it in a designated area on the web page or in the application, allowing the user to visually confirm the answer.

[0124] (Output) The user can check the generated answer through the terminal.

[0125] (Application example 1)

[0126] 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."

[0127] Conventional systems do not combine private and public information within a company to search and generate answers, making it difficult to quickly and accurately provide the information users are looking for. Furthermore, it is difficult to provide answers to security-related questions by appropriately integrating internal information with the latest external information. The present invention aims to solve these problems and provide a system that quickly provides the information users are looking for.

[0128] 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.

[0129] In this invention, the server includes a data storage means for storing private information within a company, a communication device means for receiving questions from users, a processing device means for generating answers to questions based on the data storage means and public information, a generative AI model-based means for exploring and generating answers to questions by combining private information and public information in the data storage means, an interface means for providing the generated answers to users, a means for acquiring security-related information from public information and combining it with internal information to generate answers, and a means for generating answers based on security guidelines, policies, and the latest security trends. This makes it possible to quickly provide optimal answers that integrate internal information and external information.

[0130] "Data storage means" means a device that has the function of storing, efficiently managing, and searching non-public information within a company.

[0131] The "communication device means" is a device for receiving a question from a user and transferring the question to a server.

[0132] "Processing device means" refers to a device for generating answers to questions based on the data storage means and public information.

[0133] A "generative AI model-based method" is a method that uses an artificial intelligence model to explore and generate answers to questions by combining private and public information.

[0134] An "interface means" is a device or method for providing generated answers to a user.

[0135] "Public Information" is information that is generally accessible outside the company, including security-related news and best practices.

[0136] "Security guidelines" are documents that outline security guidelines and rules established within a company.

[0137] "Security Trends" is information that shows the latest security trends and developments.

[0138] The present invention is a system that combines private information and public information within a company to generate answers to questions from users. Specific embodiments are described below.

[0139] The server uses data storage means to store the company's internal non-public information, such as internal security guidelines and policies, the latest security trends, etc. The data storage means is also provided to efficiently manage and search the non-public information.

[0140] The user inputs a question through a communication device, and the terminal receives the user's question and transfers it to the server.

[0141] The server uses the processing device means to analyze the received question and generate an answer based on the private and public information in the data storage means. Specifically, the server uses means based on a generative AI model to explore and generate an answer to the question. The model utilizes a Retrieval Augmented Generation (RAG) model to retrieve information related to the question from the data storage means and public information and generate an answer based on that information.

[0142] The generated answers are provided to the user via an interface means, which may include a terminal and a display device, and allows the user to intuitively view the generated answers.

[0143] It also includes a means to retrieve security-related information from public sources and combine it with internal information to generate answers, ensuring that answers to security questions are provided while taking into account the latest security trends and best practices.

[0144] For example, if a user asks, "What are the latest ransomware attack countermeasures?", the server generates an answer based on security guidelines in an internal database and the latest public information. Through analysis by the generative AI model, the server can provide the answer, "To counter the latest ransomware, we recommend taking regular backups and implementing multi-factor authentication."

[0145] An example of a prompt sentence could be in the following text format:

[0146] "What are the latest ransomware attack countermeasures?"

[0147] In this way, the present invention enables the user to quickly obtain the most appropriate answer that integrates internal information and external information.

[0148] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0149] Step 1:

[0150] The terminal receives a question from the user. The received question is saved as a prompt, which triggers the processing of the entire system.

[0151] Input: The question the user types into the terminal

[0152] Output: Saved prompt statement

[0153] Step 2:

[0154] The terminal forwards the received question to the server, which then starts the next process based on the prompt.

[0155] Input: prompt statement

[0156] Output: Prompt text forwarded to the server

[0157] Step 3:

[0158] The server parses the prompt and retrieves relevant non-public information from a data storage means, including internal security guidelines and policies.

[0159] Input: prompt statement

[0160] Output: Retrieved private information

[0161] Step 4:

[0162] The server calls external APIs to obtain the latest security-related information from public information sources, and the obtained information is managed as public information.

[0163] Input: prompt statement

[0164] Output: Public information obtained

[0165] Step 5:

[0166] The server uses a generative AI model (RAG model) to combine private and public information to generate the best answer to the question.

[0167] Input: Private information, Public information, Prompt text

[0168] Output: The generated answer

[0169] Step 6:

[0170] The server sends the generated answers to the device, where they are presented to the user in an intuitive format.

[0171] Input: Generated Answer

[0172] Output: Answer sent to terminal

[0173] Step 7:

[0174] The terminal displays the received answer on the display device, and the user thereby obtains the answer to the question.

[0175] Input: The answer received from the server

[0176] Output: The answer displayed to the user

[0177] This processing flow allows users to quickly obtain optimal answers that integrate internal and external information.

[0178] 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.

[0179] The present invention relates to a system that generates answers to user questions by combining private and public information within a company, and further includes an emotion engine that recognizes the user's emotions and adjusts the content of the answers. This system is composed of a database, a terminal, a server, and an emotion engine.

[0180] First, the server collects private information from within the company and stores it in a database. Private information includes system source code and specifications. This database is designed to efficiently manage and search private information.

[0181] When using the system, the user inputs a question to the system through the terminal. When the terminal receives the user's question, it forwards it to the server. The question is processed as input to the data and emotion engine within the system.

[0182] The server analyzes the query received from the device and searches for relevant data from private and public information in the database. The server uses a search algorithm based on the content of the query to search for relevant data. The server also retrieves public information and uses these sources in combination to generate an answer to the query.

[0183] To generate answers, a technology called the Retrieval Augmented Generation (RAG) model is used. The RAG model first retrieves information related to the question from databases and public information, and then generates an answer based on that information. This model makes it possible to provide the best answer to the question.

[0184] Furthermore, the present invention includes an emotion engine that identifies the user's emotional state. This emotion engine recognizes emotions by analyzing the tone and content of the user's input (e.g., question) or biometric information (e.g., voice, facial expression). This allows the system to provide a response in a gentle tone appropriate to the user's state if the user is feeling stressed or anxious.

[0185] As a specific example, if a user asks "What does the add function in utils.py do?" and the emotion engine recognizes that the user is feeling stressed, the server will generate a response such as "The add function takes two arguments and returns their sum. Is there anything else I can help you with?" In this way, the emotion engine can respond in a way that takes into account the user's emotional state.

[0186] The generated answers are sent from the server to the terminal and displayed to the user, allowing the user to quickly and accurately obtain specific answers about the system through the terminal, and to receive appropriate responses that take into account their own emotional state.

[0187] By introducing this system, system personnel will not only be able to quickly and accurately obtain information about system specifications and source code, but will also be able to respond according to the user's emotional state, contributing to a significant improvement in work efficiency.

[0188] The processing flow will be explained below.

[0189] Step 1:

[0190] The server collects private information from within the company and stores it in a database. The private information includes system source code and specifications. The database is designed to efficiently manage and search this private information.

[0191] Step 2:

[0192] The user types a question about the system via the terminal, for example, "What does the add function in utils.py do?"

[0193] Step 3:

[0194] The terminal receives the question entered by the user and transfers the question to the server, where the content of the question is sent as a character string.

[0195] Step 4:

[0196] The device detects the user's emotional state using an emotion engine that analyzes the user's voice and text emotions, for example, recognizing through voice analysis that the user is feeling stressed.

[0197] Step 5:

[0198] The server analyzes the question received from the device using natural language processing (NLP) technology, extracts the intent and keywords of the question, and generates an appropriate search query.

[0199] Step 6:

[0200] The server searches for relevant data (e.g., source code and specifications) from private information in the database based on the analyzed question, and integrates multiple related data as needed.

[0201] Step 7:

[0202] The server also retrieves public information related to the question from the Internet, including technical documentation, forum posts, blog articles, etc.

[0203] Step 8:

[0204] The server generates answers to questions based on the private information in the database and the public information it retrieves using a Retrieval Augmented Generation (RAG) model, which combines the retrieved information to generate the best answer.

[0205] Step 9:

[0206] The server takes into account the user's emotional state as detected by the emotion engine and adjusts the tone and content of the response: for example, if the user is feeling stressed, the response will be more polite and include additional explanations if necessary.

[0207] Step 10:

[0208] The server then sends the generated answer to the device, which is then formatted appropriately and presented in a way that is easy for the user to understand.

[0209] Step 11:

[0210] The terminal displays the answer it receives from the server to the user, for example, "The add function takes two arguments and returns their sum. Is there anything else I can help you with?"

[0211] This flow allows users to quickly and accurately obtain answers regarding system specifications and source code, and the emotion engine enables responses based on the user's emotional state, providing a better user experience.

[0212] Example 2

[0213] 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."

[0214] Modern companies need to efficiently utilize both private and public information to respond to user questions quickly and accurately. However, there is a lack of ways to adjust responses while taking into account the user's emotional state, which makes it difficult to improve user satisfaction and work efficiency. In particular, when a user is feeling stressed or anxious, a response that adapts to that state is required, but this has been difficult to achieve with conventional systems.

[0215] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a database storing private internal company information, a terminal receiving questions from users, means for generating answers to questions based on the database and public information, an information retrieval expansion and generation model for generating answers to questions by combining private and public information in the database, an emotion engine for recognizing the user's emotional state and adjusting the content of the answer, and means for providing the generated answer to the user. As a result, when a user inputs a question, not only is an optimal answer provided that efficiently combines private internal company information and public information, but also a gentle response that corresponds to the user's emotional state is possible. This improves user satisfaction and business efficiency.

[0216] "Internal non-public information" refers to information that is not publicly available within a company, including system source code, technical specifications, internal documents, etc.

[0217] A "database" is an information collection and management system established to efficiently manage and search confidential information within a company.

[0218] A "terminal" is a device through which a user inputs questions about the system, and includes PCs, smartphones, etc.

[0219] The "server" is the central computer of the system that analyzes questions received from terminals and generates answers to the questions based on private and public information in the database.

[0220] "Public Information" is information obtained from an external source, such as the Internet, and made available to the public.

[0221] "Information Retrieval Augmented Generative Model" refers to a technology that combines private and public information in a database to generate answers to questions, and includes generative algorithms that utilize natural language processing.

[0222] The "emotion engine" is a system that recognizes the user's emotional state from their input and biometric information, and adjusts the response content according to that emotional state.

[0223] "Natural language processing technology" refers to technology that enables computers to understand and process human language, and includes text analysis and language models.

[0224] "Biometric information" refers to information obtained from the user's body, such as their voice and facial expressions, and is data used to recognize their emotional state.

[0225] The present invention relates to a system that efficiently utilizes private and public information within a company to provide quick and accurate answers to user questions. This system is composed of a database, terminals, a server, and an emotion engine.

[0226] First, the server collects private information from within the company. This private information includes system source code, technical specifications, internal documents, etc. The collected private information is stored in a database using SQL Server. The database is set up to efficiently manage and search private information.

[0227] Next, the user inputs a question about the system through a terminal (PC or smartphone). For example, they input a question like, "What does the add function in utils.py do?" The terminal receives the question entered by the user and transfers it to the server via an HTTP request.

[0228] The server analyzes the received question using natural language processing (NLP) technology. Here, an NLP library such as SpaCy is used. Based on the analysis results, the server searches for relevant data from private information in the database and external public information. ElasticSearch (registered trademark) is used for the search.

[0229] The server generates an answer using a Retrieval Augmented Generation (RAG) model based on relevant data. The generative AI model used is the commonly used GPT-3 (registered trademark). The server incorporates information related to the question into a prompt sentence and inputs it into the RAG model to generate the optimal answer.

[0230] Furthermore, the emotion engine recognizes the user's emotional state from their input and biometric information (e.g., voice, facial expression). This engine uses, for example, Dialogflow. If the user is feeling stressed or anxious, the server adjusts the tone of the response to a gentler tone depending on the user's emotional state.

[0231] The generated answer is sent from the server to the terminal and displayed to the user, for example, an adapted answer like "The add function takes two arguments and returns their sum. Is there anything else I can help you with?"

[0232] Examples of specific prompts include:

[0233] "If a user is experiencing a stress level from the system and asks, 'What does the add function in utils.py do?', formulate your answer in a gentle tone."

[0234] In this way, users can not only obtain information from the system quickly and accurately, but also receive appropriate responses that take into account their emotional state, which greatly improves user satisfaction and business efficiency.

[0235] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0236] Step 1:

[0237] The server collects confidential internal information from the company and stores it in a database. Specific inputs include system source code, technical specifications, and internal documents, and receives this information as Excel or PDF files. The server analyzes these files, extracts relevant items, and stores them in a database using SQL Server. The output is a database containing the confidential information.

[0238] Step 2:

[0239] The user uses the terminal to input a question about the system. An example of input is "What does the add function in utils.py do?" The terminal transfers the question data entered by the user to the server as an HTTP request. The output here is the question data transferred to the server.

[0240] Step 3:

[0241] The server analyzes questions received from the terminal. The input is the question text, which is analyzed using natural language processing (NLP) technology. SpaCy is used as the NLP library, and keywords from the question are extracted as the analysis result. The output is the extracted keywords and their related information.

[0242] Step 4:

[0243] Based on the analysis results, the server searches for relevant data from private information in the database and external public information. The specific input is the extracted keywords. The server uses ElasticSearch to quickly search for relevant information in the database and on the Internet. The output is a dataset of the searched private and public information.

[0244] Step 5:

[0245] The server generates an answer using a Retrieval Augmented Generation (RAG) model based on the retrieved information. The input is the retrieved dataset, and a generative AI model (e.g., GPT-3) is used to create a prompt sentence, which is then input to the model. The output is the generated answer text.

[0246] Step 6:

[0247] The emotion engine recognizes the user's emotional state based on their input and biometric information. The input is the user's question text and biometric information (e.g., voice, facial expression). The emotion engine analyzes this information and identifies the user's emotional state (e.g., stress, anxiety). The output is the recognized emotional state data.

[0248] Step 7:

[0249] The server adjusts the answer content based on the emotional state data obtained from the emotion engine. The input is the generated answer text and the user's emotional state data. The server adjusts the tone of the answer, such as softening it, depending on the emotional state. The output is the adjusted optimal answer text.

[0250] Step 8:

[0251] The server sends the adjusted answer to the terminal. The input is the adjusted answer text. The terminal displays the received answer to the user. For example, the answer "The add function takes two arguments and returns their sum. Is there anything else I can help you with?" The output is the answer text displayed to the user.

[0252] Through these processing steps, the system can efficiently utilize both private and public information within a company to provide optimal answers to users' questions. It can also adjust responses based on the user's emotional state, contributing to improved user satisfaction.

[0253] (Application example 2)

[0254] 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."

[0255] Modern companies manage vast amounts of private and public information, but lack systems for effectively utilizing that information. Furthermore, when users ask questions, it is difficult to provide appropriate answers quickly, and the lack of consideration for the user's emotional state can negatively impact the user experience. This invention aims to provide a system that combines private and public information within a company to not only provide optimal answers to users' questions but also generate answers that take into account the user's emotional state.

[0256] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0257] In this invention, the server includes an information storage device that stores private information within the company, an input device that receives questions from users, an information processing device that generates answers to the questions based on the information storage device and public information, an information retrieval generation model that generates answers to the questions by combining the private information and public information in the information storage device, a display device that provides the generated answers to the users, and an emotion recognition device that identifies the emotional state of the users and adjusts the content of the answers. This makes it possible to effectively utilize the vast amount of information within the company and provide appropriate and prompt answers to users' questions, as well as flexibly respond to answers that take the users' emotions into consideration.

[0258] An "information storage device" is a device for storing confidential information within a company and is used to efficiently manage and search for information.

[0259] An "input device" is a device for receiving questions from a user, and is a device that plays a role in inputting user questions into the system.

[0260] An "information processing device" is a device that generates answers to questions based on information storage devices and public information, and is a device used to analyze information and provide optimal answers.

[0261] The term "information retrieval generation model" refers to a model that generates answers to questions by combining private and public information in an information storage device, and in particular refers to the Retrieval Augmented Generation (RAG) model.

[0262] A "display device" is a device for providing the generated answer to the user, and enabling the user to visually confirm the answer.

[0263] An "emotion recognition device" is a device that identifies the emotional state of a user and adjusts the content of a response, and includes technology for analyzing the user's emotions.

[0264] "Non-public information" refers to information that is managed internally by a company and is not made public, including internal documents, source code, specifications, etc.

[0265] "Public information" refers to information that is publicly available, including materials on the Internet and public documents.

[0266] The present invention is a system that generates answers to user questions by combining private and public information within a company, and further includes an emotion engine that recognizes the user's emotions and adjusts the content of the answers. This system is composed of an information storage device, an input device, an information processing device, an information retrieval generation model, a display device, and an emotion recognition device.

[0267] First, the information storage device collects and stores confidential information from within a company for efficient management and retrieval. Non-public information includes system source code and specifications. This information storage device allows the company's internal information to be organized and maintained.

[0268] Next, the user inputs a question into the system through an input device, which receives the question and forwards it to the information processing device, where it is processed as an input for the information retrieval generative model.

[0269] The information processing device analyzes the question received from the input device and searches for related data from private information and public information in the information storage device. The related data is searched for using a search algorithm based on the content of the question.

[0270] The information processing device uses an information retrieval generation model (particularly the Retrieval Augmented Generation (RAG) model) to generate answers to questions. This model retrieves information related to the question from an information storage device and public information, and generates an optimal answer based on that information.

[0271] Furthermore, the emotion recognition device identifies the user's emotional state. This emotion recognition device recognizes emotions by analyzing the tone and content of the user's input (e.g., question) or biometric information (e.g., voice, facial expression). This allows it to provide a response in a gentle tone that reflects the user's state, such as if the user is feeling stressed or anxious.

[0272] For example, if a user asks the question "Please tell me about the system's security settings," and the emotion recognition device recognizes that the user is feeling stressed, the information processing device will generate a response such as, "Don't worry. The system's security settings are configured individually for each user, and detailed instructions are provided in section 5.3 of the Security Policy Manual."

[0273] The generated answers are provided to the user via a display device. Through this system, the user can quickly and accurately obtain specific answers to their questions, and can also receive appropriate responses that take into account their own emotional state.

[0274] By introducing this system, companies can not only effectively utilize internal information and provide prompt and appropriate answers to questions, but also increase user satisfaction.

[0275] Specific prompt examples

[0276] User Question: "What are the security settings on my system?"

[0277] Emotional state: "When the emotion recognition device recognizes that the user is feeling stressed"

[0278] Tailored answer: "Don't worry, system security settings are configured individually for each user and detailed instructions are provided in section 5.3 of the Security Policy Manual."

[0279] This makes it possible to provide the most appropriate answer to the question and to realize a flexible response that takes into consideration the user's feelings.

[0280] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0281] Step 1:

[0282] A user inputs a question using an input device such as a smartphone.

[0283] Input: User question (e.g., "What are the security settings on my system?")

[0284] Output: User question data

[0285] The user inputs a question through the input device, which is recorded in the input device, and the question data is sent to the next processing step.

[0286] Step 2:

[0287] The terminal receives a question from the user and transfers it to the information processing device.

[0288] Input: User question data

[0289] Output: Data transferred to information processing device

[0290] The terminal receives the query data and prepares it for transmission to the information processing device, where data format conversion and encoding are performed.

[0291] Step 3:

[0292] The information processing device analyzes the question data and searches for related private information and public information.

[0293] Input: Question data

[0294] Output: Relevant data (company internal private information, public information)

[0295] The information processing device analyzes the question data and searches for relevant information from databases and the Internet, using a search algorithm to extract information that best suits the question.

[0296] Step 4:

[0297] The information processing device generates an optimal answer using an information retrieval generation model.

[0298] Input: Question data, related data (private information, public information)

[0299] Output: The generated answer

[0300] The information retrieval generative model (RAG model) takes relevant data as input and generates the best answer to the question, using advanced natural language processing to ensure data consistency and integrity.

[0301] Step 5:

[0302] The emotion recognizer analyzes the user's emotional state and adjusts the response as needed.

[0303] Input: User question data (tone, content), biometric information (e.g., voice, facial expression)

[0304] Output: Adjusted answer

[0305] The emotion recognition device analyzes the user's emotions and adjusts the tone of the response to be gentler if the user is feeling stressed or anxious. It also uses biometric information for more accurate emotion recognition.

[0306] Step 6:

[0307] The adjusted answer is provided to the user via a display device.

[0308] Input: Adjusted Answer

[0309] Output: What the user sees

[0310] The generated answer is sent to a display device and displayed to the user, thereby enabling the user to quickly obtain an appropriate answer.

[0311] Through the above steps, appropriate and prompt answers to the user's questions are provided, and at the same time, flexible responses that take into account the user's emotional state are realized.

[0312] 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.

[0313] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0314] 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.

[0315] [Second embodiment]

[0316] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0317] 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.

[0318] 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).

[0319] 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.

[0320] 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.

[0321] 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).

[0322] 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.

[0323] 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.

[0324] 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.

[0325] 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.

[0326] 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.

[0327] 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."

[0328] The present invention relates to a system that generates answers to user questions by combining private and public information within a company. This system is composed of a database, a terminal, and a server.

[0329] First, the server collects private information from within the company and stores it in a database. Private information includes system source code and specifications. The database is designed to efficiently manage and search private information.

[0330] When using the system, the user inputs a question through the terminal. The terminal receives the user's question and transfers it to the server. This question triggers a match with the data in the system.

[0331] The server analyzes the question received from the device and searches for relevant data from private information in the database using a search algorithm based on the content of the question. In addition, the server also retrieves public information and uses these sources in combination to generate an answer to the question.

[0332] To generate answers, a technology called the Retrieval Augmented Generation (RAG) model is used. The RAG model first retrieves information related to the question from databases and public information, and then generates an answer based on that information. This model makes it possible to provide the best answer to the question.

[0333] The generated answer is sent from the server to the terminal and displayed to the user, who can instantly obtain specific answers about the system through the terminal.

[0334] As a specific example, if a user asks, "What does the add function in utils.py do?", the server searches the database for the appropriate source code sections and specifications, combines them with public information, and generates an answer. In this case, the server generates the answer, "The add function takes two arguments and returns their sum," and provides it to the user through the terminal.

[0335] By implementing this system, system personnel can quickly and accurately obtain information about system specifications and source code, which contributes to a significant improvement in work efficiency.

[0336] The processing flow will be explained below.

[0337] Step 1:

[0338] The server collects private information from within the company and stores it in a database. The private information includes system source code and specifications. The database is designed to efficiently manage and search private information.

[0339] Step 2:

[0340] The user inputs a question about the system through a terminal, which is processed as a string of characters.

[0341] Step 3:

[0342] The terminal receives a question entered by the user and forwards the question to the server, which then formats the user's input appropriately and sends it to the server.

[0343] Step 4:

[0344] The server analyzes the question received from the device using natural language processing (NLP) technology to extract the intent and keywords of the question.

[0345] Step 5:

[0346] Based on the parsed question, the server searches for relevant data from private information in a database using algorithms to identify sections of source code or specifications that match the query.

[0347] Step 6:

[0348] The server also retrieves public information related to the question, which is collected from technical documents, forums, and other sources on the Internet.

[0349] Step 7:

[0350] The server generates answers to questions by combining private information in the database with the public information it has retrieved, using a Retrieval Augmented Generation (RAG) model.

[0351] Step 8:

[0352] The server then sends the generated answer to the terminal, where it is properly formatted and presented in a user-friendly form.

[0353] Step 9:

[0354] The terminal displays the answer received from the server to the user, and the user can instantly obtain specific answers about the system through the terminal.

[0355] This series of steps allows the user to quickly and accurately obtain information about the system specifications and source code.

[0356] Example 1

[0357] 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."

[0358] There is a lot of private information within companies, and efficient use of this information can improve business efficiency, but previous systems lacked the means to properly integrate this information and generate optimal answers to user questions.In addition, there was no way to properly combine public and private information, which meant that users could not be provided with sufficient information.

[0359] 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.

[0360] In this invention, the server includes information storage means, question receiving means, answer generating means, information integrating means, and answer providing means, which allows the server to combine private and public information within the company to generate optimal answers and provide them to users.

[0361] "Information storage means" refers to systems and databases for collecting, efficiently storing, and managing non-public information within a company.

[0362] The "question receiving means" refers to an interface for receiving a question from a user and transferring it to the server in an appropriate format.

[0363] "Answer generation means" refers to algorithms and technologies for generating optimal answers based on received questions, using private and public information within the company.

[0364] "Information synthesis means" refers to means for combining private and public information to generate an integrated answer to a question.

[0365] The "answer providing means" refers to a means for presenting the generated answer to the user.

[0366] The present invention relates to a system that generates answers to user questions by combining private and public information within a company. This system is composed of an information storage means, a question receiving means, an answer generating means, an information integration means, and an answer providing means.

[0367] First, the server collects private information from within the company and stores it in a database, which is a means of storing information.Private information includes system program code and specifications.This database is designed to efficiently manage and search private information.

[0368] The user inputs a question through the terminal. The terminal receives the user's question and sends it to the server as a question receiving means. This question becomes a trigger for matching with the data inside the system.

[0369] The server uses a natural language processing library (e.g., spaCy or Transformers) to analyze the question, and then searches for relevant private information from a database, which is an information storage means, based on the analyzed question.

[0370] Additionally, the server uses APIs on the Internet to gather public information, for example, by retrieving the required information from specific technology sites or open source repositories.

[0371] After obtaining the private and public information, the server uses the Retrieval Augmented Generation (RAG) model as an information integration method to integrate each piece of information and generate the optimal answer. This model is an advanced algorithm that generates answers based on information related to the question.

[0372] The generated answer is sent from the server to the terminal, and the terminal displays it to the user as a means of providing the answer. As a specific example, if a user asks, "What does the add function in utils.py do?", the server searches the database for the appropriate program code section and specification, and combines it with public information to generate an answer. In this case, the server generates the answer, "The add function takes two arguments and returns their sum," and provides it to the user via the terminal.

[0373] An example of a prompt to input to a generative AI model is:

[0374] "A user has a system-related question. The question is: 'What does the add function in utils.py do?' Please provide a detailed answer using relevant parts of your database and publicly available information."

[0375] As described above, the present invention realizes a system that utilizes both private and public information within a company in an integrated manner to provide quick and accurate answers to user questions. This allows system personnel to efficiently obtain appropriate information, and is expected to significantly improve work efficiency.

[0376] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0377] Step 1:

[0378] The user inputs a question through the terminal.

[0379] (Input) User question (e.g., "What does the add function in utils.py do?")

[0380] (Operation) A user enters a question into the question input form using a dedicated application or a web browser. The user enters specific questions about the question in text format.

[0381] (Output) The user's question is saved in text format on the terminal.

[0382] Step 2:

[0383] The terminal sends a question to the server.

[0384] (Input) User question text (e.g., "What does the add function in utils.py do?")

[0385] (Behavior) When the user clicks the submit button, an HTTP POST request is generated and sent to the server's API endpoint (e.g., "api.example.com / query").

[0386] (Output) The server receives the user's question text.

[0387] Step 3:

[0388] The server receives and parses the query.

[0389] (Input) User question text (e.g., "What does the add function in utils.py do?")

[0390] (Operation) The server uses a natural language processing library (e.g., spaCy, Transformers) to analyze the question text received. It performs processes such as text tokenization, part-of-speech tagging, and semantic analysis. As a result of the analysis, "utils.py" and the "add function" are identified.

[0391] (Output) The parsed question data (e.g., "File name: utils.py", "Function name: add function") is generated.

[0392] Step 4:

[0393] The server searches the database for relevant non-public information.

[0394] (Input) Parsed question data (e.g., "File name: utils.py", "Function name: add function")

[0395] (Behavior) The server uses an SQL query to retrieve the relevant program code or specification section from a database (e.g. MySQL), specifically, it executes the query "SELECT FROM SourceCode WHERE fileName='utils.py' AND functionName='add'".

[0396] (Output) Related private information (e.g., "source code of the add function" and "specification of the add function") is obtained.

[0397] Step 5:

[0398] The server collects public information from external APIs.

[0399] (Input) Analyzed question data and non-public information

[0400] (Operation) The server calls APIs on the Internet to collect public information. For example, it uses the GitHub API to obtain public repository information related to the "utils.py" file and the "add function."

[0401] (Output) Public information (e.g., "Related information about the GitHub repository") is obtained.

[0402] Step 6:

[0403] The server uses the RAG model to generate answers to questions.

[0404] (Input) Private and Public Information

[0405] (Operation) The server inputs the acquired private and public information into the RAG model and generates an integrated answer. Using the generative AI model, the optimal answer to the question (e.g., "The add function takes two arguments and returns their sum") is generated.

[0406] (Output) A generated answer (e.g., "The add function takes two arguments and returns their sum") is generated.

[0407] Step 7:

[0408] The server generates a response and sends it to the terminal.

[0409] (Input) Generated answer (e.g., "The add function takes two arguments and returns their sum")

[0410] (Operation) The server sends the generated answer to the terminal as an HTTP response. The response data is usually sent in JSON format.

[0411] (Output) The terminal receives the generated response.

[0412] Step 8:

[0413] The terminal displays the answer to the user.

[0414] (Input) Generated answer (e.g., "The add function takes two arguments and returns their sum")

[0415] (Operation) The device analyzes the received answer and displays it in a designated area on the web page or in the application, allowing the user to visually confirm the answer.

[0416] (Output) The user can check the generated answer through the terminal.

[0417] (Application example 1)

[0418] 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."

[0419] Conventional systems do not combine private and public information within a company to search and generate answers, making it difficult to quickly and accurately provide the information users are looking for. Furthermore, it is difficult to provide answers to security-related questions by appropriately integrating internal information with the latest external information. The present invention aims to solve these problems and provide a system that quickly provides the information users are looking for.

[0420] 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.

[0421] In this invention, the server includes a data storage means for storing private information within a company, a communication device means for receiving questions from users, a processing device means for generating answers to questions based on the data storage means and public information, a generative AI model-based means for exploring and generating answers to questions by combining private information and public information in the data storage means, an interface means for providing the generated answers to users, a means for acquiring security-related information from public information and combining it with internal information to generate answers, and a means for generating answers based on security guidelines, policies, and the latest security trends. This makes it possible to quickly provide optimal answers that integrate internal information and external information.

[0422] "Data storage means" means a device that has the function of storing, efficiently managing, and searching non-public information within a company.

[0423] The "communication device means" is a device for receiving a question from a user and transferring the question to a server.

[0424] "Processing device means" refers to a device for generating answers to questions based on the data storage means and public information.

[0425] A "generative AI model-based method" is a method that uses an artificial intelligence model to explore and generate answers to questions by combining private and public information.

[0426] An "interface means" is a device or method for providing generated answers to a user.

[0427] "Public Information" is information that is generally accessible outside the company, including security-related news and best practices.

[0428] "Security guidelines" are documents that outline security guidelines and rules established within a company.

[0429] "Security Trends" is information that shows the latest security trends and developments.

[0430] The present invention is a system that combines private information and public information within a company to generate answers to questions from users. Specific embodiments are described below.

[0431] The server uses data storage means to store the company's internal non-public information, such as internal security guidelines and policies, the latest security trends, etc. The data storage means is also provided to efficiently manage and search the non-public information.

[0432] The user inputs a question through a communication device, and the terminal receives the user's question and transfers it to the server.

[0433] The server uses the processing device means to analyze the received question and generate an answer based on the private and public information in the data storage means. Specifically, the server uses means based on a generative AI model to explore and generate an answer to the question. The model utilizes a Retrieval Augmented Generation (RAG) model to retrieve information related to the question from the data storage means and public information and generate an answer based on that information.

[0434] The generated answers are provided to the user via an interface means, which may include a terminal and a display device, and allows the user to intuitively view the generated answers.

[0435] It also includes a means to retrieve security-related information from public sources and combine it with internal information to generate answers, ensuring that answers to security questions are provided while taking into account the latest security trends and best practices.

[0436] For example, if a user asks, "What are the latest ransomware attack countermeasures?", the server generates an answer based on security guidelines in an internal database and the latest public information. Through analysis by the generative AI model, the server can provide the answer, "To counter the latest ransomware, we recommend taking regular backups and implementing multi-factor authentication."

[0437] An example of a prompt sentence could be in the following text format:

[0438] "What are the latest ransomware attack countermeasures?"

[0439] In this way, the present invention enables the user to quickly obtain the most appropriate answer that integrates internal information and external information.

[0440] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0441] Step 1:

[0442] The terminal receives a question from the user. The received question is saved as a prompt, which triggers the processing of the entire system.

[0443] Input: The question the user types into the terminal

[0444] Output: Saved prompt statement

[0445] Step 2:

[0446] The terminal forwards the received question to the server, which then starts the next process based on the prompt.

[0447] Input: prompt statement

[0448] Output: Prompt text forwarded to the server

[0449] Step 3:

[0450] The server parses the prompt and retrieves relevant non-public information from a data storage means, including internal security guidelines and policies.

[0451] Input: prompt statement

[0452] Output: Retrieved private information

[0453] Step 4:

[0454] The server calls external APIs to obtain the latest security-related information from public information sources, and the obtained information is managed as public information.

[0455] Input: prompt statement

[0456] Output: Public information obtained

[0457] Step 5:

[0458] The server uses a generative AI model (RAG model) to combine private and public information to generate the best answer to the question.

[0459] Input: Private information, Public information, Prompt text

[0460] Output: The generated answer

[0461] Step 6:

[0462] The server sends the generated answers to the device, where they are presented to the user in an intuitive format.

[0463] Input: Generated Answer

[0464] Output: Answer sent to terminal

[0465] Step 7:

[0466] The terminal displays the received answer on the display device, and the user thereby obtains the answer to the question.

[0467] Input: The answer received from the server

[0468] Output: The answer displayed to the user

[0469] This processing flow allows users to quickly obtain optimal answers that integrate internal and external information.

[0470] 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.

[0471] The present invention relates to a system that generates answers to user questions by combining private and public information within a company, and further includes an emotion engine that recognizes the user's emotions and adjusts the content of the answers. This system is composed of a database, a terminal, a server, and an emotion engine.

[0472] First, the server collects private information from within the company and stores it in a database. Private information includes system source code and specifications. This database is designed to efficiently manage and search private information.

[0473] When using the system, the user inputs a question to the system through the terminal. When the terminal receives the user's question, it forwards it to the server. The question is processed as input to the data and emotion engine within the system.

[0474] The server analyzes the query received from the device and searches for relevant data from private and public information in the database. The server uses a search algorithm based on the content of the query to search for relevant data. The server also retrieves public information and uses these sources in combination to generate an answer to the query.

[0475] To generate answers, a technology called the Retrieval Augmented Generation (RAG) model is used. The RAG model first retrieves information related to the question from databases and public information, and then generates an answer based on that information. This model makes it possible to provide the best answer to the question.

[0476] Furthermore, the present invention includes an emotion engine that identifies the user's emotional state. This emotion engine recognizes emotions by analyzing the tone and content of the user's input (e.g., question) or biometric information (e.g., voice, facial expression). This allows the system to provide a response in a gentle tone appropriate to the user's state if the user is feeling stressed or anxious.

[0477] As a specific example, if a user asks "What does the add function in utils.py do?" and the emotion engine recognizes that the user is feeling stressed, the server will generate a response such as "The add function takes two arguments and returns their sum. Is there anything else I can help you with?" In this way, the emotion engine can respond in a way that takes into account the user's emotional state.

[0478] The generated answers are sent from the server to the terminal and displayed to the user, allowing the user to quickly and accurately obtain specific answers about the system through the terminal, and to receive appropriate responses that take into account their own emotional state.

[0479] By introducing this system, system personnel will not only be able to quickly and accurately obtain information about system specifications and source code, but will also be able to respond according to the user's emotional state, contributing to a significant improvement in work efficiency.

[0480] The processing flow will be explained below.

[0481] Step 1:

[0482] The server collects private information from within the company and stores it in a database. The private information includes system source code and specifications. The database is designed to efficiently manage and search this private information.

[0483] Step 2:

[0484] The user types a question about the system via the terminal, for example, "What does the add function in utils.py do?"

[0485] Step 3:

[0486] The terminal receives the question entered by the user and transfers the question to the server, where the content of the question is sent as a character string.

[0487] Step 4:

[0488] The device detects the user's emotional state using an emotion engine that analyzes the user's voice and text emotions, for example, recognizing through voice analysis that the user is feeling stressed.

[0489] Step 5:

[0490] The server analyzes the question received from the device using natural language processing (NLP) technology, extracts the intent and keywords of the question, and generates an appropriate search query.

[0491] Step 6:

[0492] The server searches for relevant data (e.g., source code and specifications) from private information in the database based on the analyzed question, and integrates multiple related data as needed.

[0493] Step 7:

[0494] The server also retrieves public information related to the question from the Internet, including technical documentation, forum posts, blog articles, etc.

[0495] Step 8:

[0496] The server generates answers to questions based on the private information in the database and the public information it retrieves using a Retrieval Augmented Generation (RAG) model, which combines the retrieved information to generate the best answer.

[0497] Step 9:

[0498] The server takes into account the user's emotional state as detected by the emotion engine and adjusts the tone and content of the response: for example, if the user is feeling stressed, the response will be more polite and include additional explanations if necessary.

[0499] Step 10:

[0500] The server then sends the generated answer to the device, which is then formatted appropriately and presented in a way that is easy for the user to understand.

[0501] Step 11:

[0502] The terminal displays the answer it receives from the server to the user, for example, "The add function takes two arguments and returns their sum. Is there anything else I can help you with?"

[0503] This flow allows users to quickly and accurately obtain answers regarding system specifications and source code, and the emotion engine enables responses based on the user's emotional state, providing a better user experience.

[0504] Example 2

[0505] 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."

[0506] Modern companies need to efficiently utilize both private and public information to respond to user questions quickly and accurately. However, there is a lack of ways to adjust responses while taking into account the user's emotional state, which makes it difficult to improve user satisfaction and work efficiency. In particular, when a user is feeling stressed or anxious, a response that adapts to that state is required, but this has been difficult to achieve with conventional systems.

[0507] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a database storing private internal company information, a terminal receiving questions from users, means for generating answers to questions based on the database and public information, an information retrieval expansion and generation model for generating answers to questions by combining private and public information in the database, an emotion engine for recognizing the user's emotional state and adjusting the content of the answer, and means for providing the generated answer to the user. As a result, when a user inputs a question, not only is an optimal answer provided that efficiently combines private internal company information and public information, but also a gentle response that corresponds to the user's emotional state is possible. This improves user satisfaction and business efficiency.

[0508] "Internal non-public information" refers to information that is not publicly available within a company, including system source code, technical specifications, internal documents, etc.

[0509] A "database" is an information collection and management system established to efficiently manage and search confidential information within a company.

[0510] A "terminal" is a device through which a user inputs questions about the system, and includes PCs, smartphones, etc.

[0511] The "server" is the central computer of the system that analyzes questions received from terminals and generates answers to the questions based on private and public information in the database.

[0512] "Public Information" is information obtained from an external source, such as the Internet, and made available to the public.

[0513] "Information Retrieval Augmented Generative Model" refers to a technology that combines private and public information in a database to generate answers to questions, and includes generative algorithms that utilize natural language processing.

[0514] The "emotion engine" is a system that recognizes the user's emotional state from their input and biometric information, and adjusts the response content according to that emotional state.

[0515] "Natural language processing technology" refers to technology that enables computers to understand and process human language, and includes text analysis and language models.

[0516] "Biometric information" refers to information obtained from the user's body, such as their voice and facial expressions, and is data used to recognize their emotional state.

[0517] The present invention relates to a system that efficiently utilizes private and public information within a company to provide quick and accurate answers to user questions. This system is composed of a database, terminals, a server, and an emotion engine.

[0518] First, the server collects private information from within the company. This private information includes system source code, technical specifications, internal documents, etc. The collected private information is stored in a database using SQL Server. The database is set up to efficiently manage and search private information.

[0519] Next, the user inputs a question about the system through a terminal (PC or smartphone). For example, they input a question like, "What does the add function in utils.py do?" The terminal receives the question entered by the user and transfers it to the server via an HTTP request.

[0520] The server analyzes the received question using natural language processing (NLP) techniques. An NLP library such as SpaCy is used. Based on the analysis results, the server searches for relevant data from private information in the database and external public information. ElasticSearch is used for the search.

[0521] The server generates an answer using a Retrieval Augmented Generation (RAG) model based on relevant data. The generative AI model used is the commonly used GPT-3. The server incorporates information related to the question into a prompt sentence and inputs it into the RAG model to generate the optimal answer.

[0522] Furthermore, the emotion engine recognizes the user's emotional state from their input and biometric information (e.g., voice, facial expression). This engine uses, for example, Dialogflow. If the user is feeling stressed or anxious, the server adjusts the tone of the response to a gentler tone depending on the user's emotional state.

[0523] The generated answer is sent from the server to the terminal and displayed to the user, for example, an adapted answer like "The add function takes two arguments and returns their sum. Is there anything else I can help you with?"

[0524] Examples of specific prompts include:

[0525] "If a user is experiencing a stress level from the system and asks, 'What does the add function in utils.py do?', formulate your answer in a gentle tone."

[0526] In this way, users can not only obtain information from the system quickly and accurately, but also receive appropriate responses that take into account their emotional state, which greatly improves user satisfaction and business efficiency.

[0527] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0528] Step 1:

[0529] The server collects confidential internal information from the company and stores it in a database. Specific inputs include system source code, technical specifications, and internal documents, and receives this information as Excel or PDF files. The server analyzes these files, extracts relevant items, and stores them in a database using SQL Server. The output is a database containing the confidential information.

[0530] Step 2:

[0531] The user uses the terminal to input a question about the system. An example of input is "What does the add function in utils.py do?" The terminal transfers the question data entered by the user to the server as an HTTP request. The output here is the question data transferred to the server.

[0532] Step 3:

[0533] The server analyzes questions received from the terminal. The input is the question text, which is analyzed using natural language processing (NLP) technology. SpaCy is used as the NLP library, and keywords from the question are extracted as the analysis result. The output is the extracted keywords and their related information.

[0534] Step 4:

[0535] Based on the analysis results, the server searches for relevant data from private information in the database and external public information. The specific input is the extracted keywords. The server uses ElasticSearch to quickly search for relevant information in the database and on the Internet. The output is a dataset of the searched private and public information.

[0536] Step 5:

[0537] The server generates an answer using a Retrieval Augmented Generation (RAG) model based on the retrieved information. The input is the retrieved dataset, and a generative AI model (e.g., GPT-3) is used to create a prompt sentence, which is then input to the model. The output is the generated answer text.

[0538] Step 6:

[0539] The emotion engine recognizes the user's emotional state based on their input and biometric information. The input is the user's question text and biometric information (e.g., voice, facial expression). The emotion engine analyzes this information and identifies the user's emotional state (e.g., stress, anxiety). The output is the recognized emotional state data.

[0540] Step 7:

[0541] The server adjusts the answer content based on the emotional state data obtained from the emotion engine. The input is the generated answer text and the user's emotional state data. The server adjusts the tone of the answer, such as softening it, depending on the emotional state. The output is the adjusted optimal answer text.

[0542] Step 8:

[0543] The server sends the adjusted answer to the terminal. The input is the adjusted answer text. The terminal displays the received answer to the user. For example, the answer "The add function takes two arguments and returns their sum. Is there anything else I can help you with?" The output is the answer text displayed to the user.

[0544] Through these processing steps, the system can efficiently utilize both private and public information within a company to provide optimal answers to users' questions. It can also adjust responses based on the user's emotional state, contributing to improved user satisfaction.

[0545] (Application example 2)

[0546] 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."

[0547] Modern companies manage vast amounts of private and public information, but lack systems for effectively utilizing that information. Furthermore, when users ask questions, it is difficult to provide appropriate answers quickly, and the lack of consideration for the user's emotional state can negatively impact the user experience. This invention aims to provide a system that combines private and public information within a company to not only provide optimal answers to users' questions but also generate answers that take into account the user's emotional state.

[0548] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0549] In this invention, the server includes an information storage device that stores private information within the company, an input device that receives questions from users, an information processing device that generates answers to the questions based on the information storage device and public information, an information retrieval generation model that generates answers to the questions by combining the private information and public information in the information storage device, a display device that provides the generated answers to the users, and an emotion recognition device that identifies the emotional state of the users and adjusts the content of the answers. This makes it possible to effectively utilize the vast amount of information within the company and provide appropriate and prompt answers to users' questions, as well as flexibly respond to answers that take the users' emotions into consideration.

[0550] An "information storage device" is a device for storing confidential information within a company and is used to efficiently manage and search for information.

[0551] An "input device" is a device for receiving questions from a user, and is a device that plays a role in inputting user questions into the system.

[0552] An "information processing device" is a device that generates answers to questions based on information storage devices and public information, and is a device used to analyze information and provide optimal answers.

[0553] The term "information retrieval generation model" refers to a model that generates answers to questions by combining private and public information in an information storage device, and in particular refers to the Retrieval Augmented Generation (RAG) model.

[0554] A "display device" is a device for providing the generated answer to the user, and enabling the user to visually confirm the answer.

[0555] An "emotion recognition device" is a device that identifies the emotional state of a user and adjusts the content of a response, and includes technology for analyzing the user's emotions.

[0556] "Non-public information" refers to information that is managed internally by a company and is not made public, including internal documents, source code, specifications, etc.

[0557] "Public information" refers to information that is publicly available, including materials on the Internet and public documents.

[0558] The present invention is a system that generates answers to user questions by combining private and public information within a company, and further includes an emotion engine that recognizes the user's emotions and adjusts the content of the answers. This system is composed of an information storage device, an input device, an information processing device, an information retrieval generation model, a display device, and an emotion recognition device.

[0559] First, the information storage device collects and stores confidential information from within a company for efficient management and retrieval. Non-public information includes system source code and specifications. This information storage device allows the company's internal information to be organized and maintained.

[0560] Next, the user inputs a question into the system through an input device, which receives the question and forwards it to the information processing device, where it is processed as an input for the information retrieval generative model.

[0561] The information processing device analyzes the question received from the input device and searches for related data from private information and public information in the information storage device. The related data is searched for using a search algorithm based on the content of the question.

[0562] The information processing device uses an information retrieval generation model (particularly the Retrieval Augmented Generation (RAG) model) to generate answers to questions. This model retrieves information related to the question from an information storage device and public information, and generates an optimal answer based on that information.

[0563] Furthermore, the emotion recognition device identifies the user's emotional state. This emotion recognition device recognizes emotions by analyzing the tone and content of the user's input (e.g., question) or biometric information (e.g., voice, facial expression). This allows it to provide a response in a gentle tone that reflects the user's state, such as if the user is feeling stressed or anxious.

[0564] For example, if a user asks the question "Please tell me about the system's security settings," and the emotion recognition device recognizes that the user is feeling stressed, the information processing device will generate a response such as, "Don't worry. The system's security settings are configured individually for each user, and detailed instructions are provided in section 5.3 of the Security Policy Manual."

[0565] The generated answers are provided to the user via a display device. Through this system, the user can quickly and accurately obtain specific answers to their questions, and can also receive appropriate responses that take into account their own emotional state.

[0566] By introducing this system, companies can not only effectively utilize internal information and provide prompt and appropriate answers to questions, but also increase user satisfaction.

[0567] Specific prompt examples

[0568] User Question: "What are the security settings on my system?"

[0569] Emotional state: "When the emotion recognition device recognizes that the user is feeling stressed"

[0570] Tailored answer: "Don't worry, system security settings are configured individually for each user and detailed instructions are provided in section 5.3 of the Security Policy Manual."

[0571] This makes it possible to provide the most appropriate answer to the question and to realize a flexible response that takes into consideration the user's feelings.

[0572] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0573] Step 1:

[0574] A user inputs a question using an input device such as a smartphone.

[0575] Input: User question (e.g., "What are the security settings on my system?")

[0576] Output: User question data

[0577] The user inputs a question through the input device, which is recorded in the input device, and the question data is sent to the next processing step.

[0578] Step 2:

[0579] The terminal receives a question from the user and transfers it to the information processing device.

[0580] Input: User question data

[0581] Output: Data transferred to information processing device

[0582] The terminal receives the query data and prepares it for transmission to the information processing device, where data format conversion and encoding are performed.

[0583] Step 3:

[0584] The information processing device analyzes the question data and searches for related private information and public information.

[0585] Input: Question data

[0586] Output: Relevant data (company internal private information, public information)

[0587] The information processing device analyzes the question data and searches for relevant information from databases and the Internet, using a search algorithm to extract information that best suits the question.

[0588] Step 4:

[0589] The information processing device generates an optimal answer using an information retrieval generation model.

[0590] Input: Question data, related data (private information, public information)

[0591] Output: The generated answer

[0592] The information retrieval generative model (RAG model) takes relevant data as input and generates the best answer to the question, using advanced natural language processing to ensure data consistency and integrity.

[0593] Step 5:

[0594] The emotion recognizer analyzes the user's emotional state and adjusts the response as needed.

[0595] Input: User question data (tone, content), biometric information (e.g., voice, facial expression)

[0596] Output: Adjusted answer

[0597] The emotion recognition device analyzes the user's emotions and adjusts the tone of the response to be gentler if the user is feeling stressed or anxious. It also uses biometric information for more accurate emotion recognition.

[0598] Step 6:

[0599] The adjusted answer is provided to the user via a display device.

[0600] Input: Adjusted Answer

[0601] Output: What the user sees

[0602] The generated answer is sent to a display device and displayed to the user, thereby enabling the user to quickly obtain an appropriate answer.

[0603] Through the above steps, appropriate and prompt answers to the user's questions are provided, and at the same time, flexible responses that take into account the user's emotional state are realized.

[0604] 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.

[0605] 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.

[0606] 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.

[0607] [Third embodiment]

[0608] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0609] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0610] 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).

[0611] 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.

[0612] 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.

[0613] 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).

[0614] 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.

[0615] 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.

[0616] 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.

[0617] 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.

[0618] 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.

[0619] 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."

[0620] The present invention relates to a system that generates answers to user questions by combining private and public information within a company. This system is composed of a database, a terminal, and a server.

[0621] First, the server collects private information from within the company and stores it in a database. Private information includes system source code and specifications. The database is designed to efficiently manage and search private information.

[0622] When using the system, the user inputs a question through the terminal. The terminal receives the user's question and transfers it to the server. This question triggers a match with the data in the system.

[0623] The server analyzes the question received from the device and searches for relevant data from private information in the database using a search algorithm based on the content of the question. In addition, the server also retrieves public information and uses these sources in combination to generate an answer to the question.

[0624] To generate answers, a technology called the Retrieval Augmented Generation (RAG) model is used. The RAG model first retrieves information related to the question from databases and public information, and then generates an answer based on that information. This model makes it possible to provide the best answer to the question.

[0625] The generated answer is sent from the server to the terminal and displayed to the user, who can instantly obtain specific answers about the system through the terminal.

[0626] As a specific example, if a user asks, "What does the add function in utils.py do?", the server searches the database for the appropriate source code sections and specifications, combines them with public information, and generates an answer. In this case, the server generates the answer, "The add function takes two arguments and returns their sum," and provides it to the user through the terminal.

[0627] By implementing this system, system personnel can quickly and accurately obtain information about system specifications and source code, which contributes to a significant improvement in work efficiency.

[0628] The processing flow will be explained below.

[0629] Step 1:

[0630] The server collects private information from within the company and stores it in a database. The private information includes system source code and specifications. The database is designed to efficiently manage and search private information.

[0631] Step 2:

[0632] The user inputs a question about the system through a terminal, which is processed as a string of characters.

[0633] Step 3:

[0634] The terminal receives a question entered by the user and forwards the question to the server, which then formats the user's input appropriately and sends it to the server.

[0635] Step 4:

[0636] The server analyzes the question received from the device using natural language processing (NLP) technology to extract the intent and keywords of the question.

[0637] Step 5:

[0638] Based on the parsed question, the server searches for relevant data from private information in a database using algorithms to identify sections of source code or specifications that match the query.

[0639] Step 6:

[0640] The server also retrieves public information related to the question, which is collected from technical documents, forums, and other sources on the Internet.

[0641] Step 7:

[0642] The server generates answers to questions by combining private information in the database with the public information it has retrieved, using a Retrieval Augmented Generation (RAG) model.

[0643] Step 8:

[0644] The server then sends the generated answer to the terminal, where it is properly formatted and presented in a user-friendly form.

[0645] Step 9:

[0646] The terminal displays the answer received from the server to the user, and the user can instantly obtain specific answers about the system through the terminal.

[0647] This series of steps allows the user to quickly and accurately obtain information about the system specifications and source code.

[0648] Example 1

[0649] 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."

[0650] There is a lot of private information within companies, and efficient use of this information can improve business efficiency, but previous systems lacked the means to properly integrate this information and generate optimal answers to user questions.In addition, there was no way to properly combine public and private information, which meant that users could not be provided with sufficient information.

[0651] 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.

[0652] In this invention, the server includes information storage means, question receiving means, answer generating means, information integrating means, and answer providing means, which allows the server to combine private and public information within the company to generate optimal answers and provide them to users.

[0653] "Information storage means" refers to systems and databases for collecting, efficiently storing, and managing non-public information within a company.

[0654] The "question receiving means" refers to an interface for receiving a question from a user and transferring it to the server in an appropriate format.

[0655] "Answer generation means" refers to algorithms and technologies for generating optimal answers based on received questions, using private and public information within the company.

[0656] "Information synthesis means" refers to means for combining private and public information to generate an integrated answer to a question.

[0657] The "answer providing means" refers to a means for presenting the generated answer to the user.

[0658] The present invention relates to a system that generates answers to user questions by combining private and public information within a company. This system is composed of an information storage means, a question receiving means, an answer generating means, an information integration means, and an answer providing means.

[0659] First, the server collects private information from within the company and stores it in a database, which is a means of storing information.Private information includes system program code and specifications.This database is designed to efficiently manage and search private information.

[0660] The user inputs a question through the terminal. The terminal receives the user's question and sends it to the server as a question receiving means. This question becomes a trigger for matching with the data inside the system.

[0661] The server uses a natural language processing library (e.g., spaCy or Transformers) to analyze the question, and then searches for relevant private information from a database, which is an information storage means, based on the analyzed question.

[0662] Additionally, the server uses APIs on the Internet to gather public information, for example, by retrieving the required information from specific technology sites or open source repositories.

[0663] After obtaining the private and public information, the server uses the Retrieval Augmented Generation (RAG) model as an information integration method to integrate each piece of information and generate the optimal answer. This model is an advanced algorithm that generates answers based on information related to the question.

[0664] The generated answer is sent from the server to the terminal, and the terminal displays it to the user as a means of providing the answer. As a specific example, if a user asks, "What does the add function in utils.py do?", the server searches the database for the appropriate program code section and specification, and combines it with public information to generate an answer. In this case, the server generates the answer, "The add function takes two arguments and returns their sum," and provides it to the user via the terminal.

[0665] An example of a prompt to input to a generative AI model is:

[0666] "A user has a system-related question. The question is: 'What does the add function in utils.py do?' Please provide a detailed answer using relevant parts of your database and publicly available information."

[0667] As described above, the present invention realizes a system that utilizes both private and public information within a company in an integrated manner to provide quick and accurate answers to user questions. This allows system personnel to efficiently obtain appropriate information, and is expected to significantly improve work efficiency.

[0668] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0669] Step 1:

[0670] The user inputs a question through the terminal.

[0671] (Input) User question (e.g., "What does the add function in utils.py do?")

[0672] (Operation) A user enters a question into the question input form using a dedicated application or a web browser. The user enters specific questions about the question in text format.

[0673] (Output) The user's question is saved in text format on the terminal.

[0674] Step 2:

[0675] The terminal sends a question to the server.

[0676] (Input) User question text (e.g., "What does the add function in utils.py do?")

[0677] (Behavior) When the user clicks the submit button, an HTTP POST request is generated and sent to the server's API endpoint (e.g., "api.example.com / query").

[0678] (Output) The server receives the user's question text.

[0679] Step 3:

[0680] The server receives and parses the query.

[0681] (Input) User question text (e.g., "What does the add function in utils.py do?")

[0682] (Operation) The server uses a natural language processing library (e.g., spaCy, Transformers) to analyze the question text received. It performs processes such as text tokenization, part-of-speech tagging, and semantic analysis. As a result of the analysis, "utils.py" and the "add function" are identified.

[0683] (Output) The parsed question data (e.g., "File name: utils.py", "Function name: add function") is generated.

[0684] Step 4:

[0685] The server searches the database for relevant non-public information.

[0686] (Input) Parsed question data (e.g., "File name: utils.py", "Function name: add function")

[0687] (Behavior) The server uses an SQL query to retrieve the relevant program code or specification section from a database (e.g. MySQL), specifically, it executes the query "SELECT FROM SourceCode WHERE fileName='utils.py' AND functionName='add'".

[0688] (Output) Related private information (e.g., "source code of the add function" and "specification of the add function") is obtained.

[0689] Step 5:

[0690] The server collects public information from external APIs.

[0691] (Input) Analyzed question data and non-public information

[0692] (Operation) The server calls APIs on the Internet to collect public information. For example, it uses the GitHub API to obtain public repository information related to the "utils.py" file and the "add function."

[0693] (Output) Public information (e.g., "Related information about the GitHub repository") is obtained.

[0694] Step 6:

[0695] The server uses the RAG model to generate answers to questions.

[0696] (Input) Private and Public Information

[0697] (Operation) The server inputs the acquired private and public information into the RAG model and generates an integrated answer. Using the generative AI model, the optimal answer to the question (e.g., "The add function takes two arguments and returns their sum") is generated.

[0698] (Output) A generated answer (e.g., "The add function takes two arguments and returns their sum") is generated.

[0699] Step 7:

[0700] The server generates a response and sends it to the terminal.

[0701] (Input) Generated answer (e.g., "The add function takes two arguments and returns their sum")

[0702] (Operation) The server sends the generated answer to the terminal as an HTTP response. The response data is usually sent in JSON format.

[0703] (Output) The terminal receives the generated response.

[0704] Step 8:

[0705] The terminal displays the answer to the user.

[0706] (Input) Generated answer (e.g., "The add function takes two arguments and returns their sum")

[0707] (Operation) The device analyzes the received answer and displays it in a designated area on the web page or in the application, allowing the user to visually confirm the answer.

[0708] (Output) The user can check the generated answer through the terminal.

[0709] (Application example 1)

[0710] 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."

[0711] Conventional systems do not combine private and public information within a company to search and generate answers, making it difficult to quickly and accurately provide the information users are looking for. Furthermore, it is difficult to provide answers to security-related questions by appropriately integrating internal information with the latest external information. The present invention aims to solve these problems and provide a system that quickly provides the information users are looking for.

[0712] 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.

[0713] In this invention, the server includes a data storage means for storing private information within a company, a communication device means for receiving questions from users, a processing device means for generating answers to questions based on the data storage means and public information, a generative AI model-based means for exploring and generating answers to questions by combining private information and public information in the data storage means, an interface means for providing the generated answers to users, a means for acquiring security-related information from public information and combining it with internal information to generate answers, and a means for generating answers based on security guidelines, policies, and the latest security trends. This makes it possible to quickly provide optimal answers that integrate internal information and external information.

[0714] "Data storage means" means a device that has the function of storing, efficiently managing, and searching non-public information within a company.

[0715] The "communication device means" is a device for receiving a question from a user and transferring the question to a server.

[0716] "Processing device means" refers to a device for generating answers to questions based on the data storage means and public information.

[0717] A "generative AI model-based method" is a method that uses an artificial intelligence model to explore and generate answers to questions by combining private and public information.

[0718] An "interface means" is a device or method for providing generated answers to a user.

[0719] "Public Information" is information that is generally accessible outside the company, including security-related news and best practices.

[0720] "Security guidelines" are documents that outline security guidelines and rules established within a company.

[0721] "Security Trends" is information that shows the latest security trends and developments.

[0722] The present invention is a system that combines private information and public information within a company to generate answers to questions from users. Specific embodiments are described below.

[0723] The server uses data storage means to store the company's internal non-public information, such as internal security guidelines and policies, the latest security trends, etc. The data storage means is also provided to efficiently manage and search the non-public information.

[0724] The user inputs a question through a communication device, and the terminal receives the user's question and transfers it to the server.

[0725] The server uses the processing device means to analyze the received question and generate an answer based on the private and public information in the data storage means. Specifically, the server uses means based on a generative AI model to explore and generate an answer to the question. The model utilizes a Retrieval Augmented Generation (RAG) model to retrieve information related to the question from the data storage means and public information and generate an answer based on that information.

[0726] The generated answers are provided to the user via an interface means, which may include a terminal and a display device, and allows the user to intuitively view the generated answers.

[0727] It also includes a means to retrieve security-related information from public sources and combine it with internal information to generate answers, ensuring that answers to security questions are provided while taking into account the latest security trends and best practices.

[0728] For example, if a user asks, "What are the latest ransomware attack countermeasures?", the server generates an answer based on security guidelines in an internal database and the latest public information. Through analysis by the generative AI model, the server can provide the answer, "To counter the latest ransomware, we recommend taking regular backups and implementing multi-factor authentication."

[0729] An example of a prompt sentence could be in the following text format:

[0730] "What are the latest ransomware attack countermeasures?"

[0731] In this way, the present invention enables the user to quickly obtain the most appropriate answer that integrates internal information and external information.

[0732] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0733] Step 1:

[0734] The terminal receives a question from the user. The received question is saved as a prompt, which triggers the processing of the entire system.

[0735] Input: The question the user types into the terminal

[0736] Output: Saved prompt statement

[0737] Step 2:

[0738] The terminal forwards the received question to the server, which then starts the next process based on the prompt.

[0739] Input: prompt statement

[0740] Output: Prompt text forwarded to the server

[0741] Step 3:

[0742] The server parses the prompt and retrieves relevant non-public information from a data storage means, including internal security guidelines and policies.

[0743] Input: prompt statement

[0744] Output: Retrieved private information

[0745] Step 4:

[0746] The server calls external APIs to obtain the latest security-related information from public information sources, and the obtained information is managed as public information.

[0747] Input: prompt statement

[0748] Output: Public information obtained

[0749] Step 5:

[0750] The server uses a generative AI model (RAG model) to combine private and public information to generate the best answer to the question.

[0751] Input: Private information, Public information, Prompt text

[0752] Output: The generated answer

[0753] Step 6:

[0754] The server sends the generated answers to the device, where they are presented to the user in an intuitive format.

[0755] Input: Generated Answer

[0756] Output: Answer sent to terminal

[0757] Step 7:

[0758] The terminal displays the received answer on the display device, and the user thereby obtains the answer to the question.

[0759] Input: The answer received from the server

[0760] Output: The answer displayed to the user

[0761] This processing flow allows users to quickly obtain optimal answers that integrate internal and external information.

[0762] 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.

[0763] The present invention relates to a system that generates answers to user questions by combining private and public information within a company, and further includes an emotion engine that recognizes the user's emotions and adjusts the content of the answers. This system is composed of a database, a terminal, a server, and an emotion engine.

[0764] First, the server collects private information from within the company and stores it in a database. Private information includes system source code and specifications. This database is designed to efficiently manage and search private information.

[0765] When using the system, the user inputs a question to the system through the terminal. When the terminal receives the user's question, it forwards it to the server. The question is processed as input to the data and emotion engine within the system.

[0766] The server analyzes the query received from the device and searches for relevant data from private and public information in the database. The server uses a search algorithm based on the content of the query to search for relevant data. The server also retrieves public information and uses these sources in combination to generate an answer to the query.

[0767] To generate answers, a technology called the Retrieval Augmented Generation (RAG) model is used. The RAG model first retrieves information related to the question from databases and public information, and then generates an answer based on that information. This model makes it possible to provide the best answer to the question.

[0768] Furthermore, the present invention includes an emotion engine that identifies the user's emotional state. This emotion engine recognizes emotions by analyzing the tone and content of the user's input (e.g., question) or biometric information (e.g., voice, facial expression). This allows the system to provide a response in a gentle tone appropriate to the user's state if the user is feeling stressed or anxious.

[0769] As a specific example, if a user asks "What does the add function in utils.py do?" and the emotion engine recognizes that the user is feeling stressed, the server will generate a response such as "The add function takes two arguments and returns their sum. Is there anything else I can help you with?" In this way, the emotion engine can respond in a way that takes into account the user's emotional state.

[0770] The generated answers are sent from the server to the terminal and displayed to the user, allowing the user to quickly and accurately obtain specific answers about the system through the terminal, and to receive appropriate responses that take into account their own emotional state.

[0771] By introducing this system, system personnel will not only be able to quickly and accurately obtain information about system specifications and source code, but will also be able to respond according to the user's emotional state, contributing to a significant improvement in work efficiency.

[0772] The processing flow will be explained below.

[0773] Step 1:

[0774] The server collects private information from within the company and stores it in a database. The private information includes system source code and specifications. The database is designed to efficiently manage and search this private information.

[0775] Step 2:

[0776] The user types a question about the system via the terminal, for example, "What does the add function in utils.py do?"

[0777] Step 3:

[0778] The terminal receives the question entered by the user and transfers the question to the server, where the content of the question is sent as a character string.

[0779] Step 4:

[0780] The device detects the user's emotional state using an emotion engine that analyzes the user's voice and text emotions, for example, recognizing through voice analysis that the user is feeling stressed.

[0781] Step 5:

[0782] The server analyzes the question received from the device using natural language processing (NLP) technology, extracts the intent and keywords of the question, and generates an appropriate search query.

[0783] Step 6:

[0784] The server searches for relevant data (e.g., source code and specifications) from private information in the database based on the analyzed question, and integrates multiple related data as needed.

[0785] Step 7:

[0786] The server also retrieves public information related to the question from the Internet, including technical documentation, forum posts, blog articles, etc.

[0787] Step 8:

[0788] The server generates answers to questions based on the private information in the database and the public information it retrieves using a Retrieval Augmented Generation (RAG) model, which combines the retrieved information to generate the best answer.

[0789] Step 9:

[0790] The server takes into account the user's emotional state as detected by the emotion engine and adjusts the tone and content of the response: for example, if the user is feeling stressed, the response will be more polite and include additional explanations if necessary.

[0791] Step 10:

[0792] The server then sends the generated answer to the device, which is then formatted appropriately and presented in a way that is easy for the user to understand.

[0793] Step 11:

[0794] The terminal displays the answer it receives from the server to the user, for example, "The add function takes two arguments and returns their sum. Is there anything else I can help you with?"

[0795] This flow allows users to quickly and accurately obtain answers regarding system specifications and source code, and the emotion engine enables responses based on the user's emotional state, providing a better user experience.

[0796] Example 2

[0797] 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."

[0798] Modern companies need to efficiently utilize both private and public information to respond to user questions quickly and accurately. However, there is a lack of ways to adjust responses while taking into account the user's emotional state, which makes it difficult to improve user satisfaction and work efficiency. In particular, when a user is feeling stressed or anxious, a response that adapts to that state is required, but this has been difficult to achieve with conventional systems.

[0799] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a database storing private internal company information, a terminal receiving questions from users, means for generating answers to questions based on the database and public information, an information retrieval expansion and generation model for generating answers to questions by combining private and public information in the database, an emotion engine for recognizing the user's emotional state and adjusting the content of the answer, and means for providing the generated answer to the user. As a result, when a user inputs a question, not only is an optimal answer provided that efficiently combines private internal company information and public information, but also a gentle response that corresponds to the user's emotional state is possible. This improves user satisfaction and business efficiency.

[0800] "Internal non-public information" refers to information that is not publicly available within a company, including system source code, technical specifications, internal documents, etc.

[0801] A "database" is an information collection and management system established to efficiently manage and search confidential information within a company.

[0802] A "terminal" is a device through which a user inputs questions about the system, and includes PCs, smartphones, etc.

[0803] The "server" is the central computer of the system that analyzes questions received from terminals and generates answers to the questions based on private and public information in the database.

[0804] "Public Information" is information obtained from an external source, such as the Internet, and made available to the public.

[0805] "Information Retrieval Augmented Generative Model" refers to a technology that combines private and public information in a database to generate answers to questions, and includes generative algorithms that utilize natural language processing.

[0806] The "emotion engine" is a system that recognizes the user's emotional state from their input and biometric information, and adjusts the response content according to that emotional state.

[0807] "Natural language processing technology" refers to technology that enables computers to understand and process human language, and includes text analysis and language models.

[0808] "Biometric information" refers to information obtained from the user's body, such as their voice and facial expressions, and is data used to recognize their emotional state.

[0809] The present invention relates to a system that efficiently utilizes private and public information within a company to provide quick and accurate answers to user questions. This system is composed of a database, terminals, a server, and an emotion engine.

[0810] First, the server collects private information from within the company. This private information includes system source code, technical specifications, internal documents, etc. The collected private information is stored in a database using SQL Server. The database is set up to efficiently manage and search private information.

[0811] Next, the user inputs a question about the system through a terminal (PC or smartphone). For example, they input a question like, "What does the add function in utils.py do?" The terminal receives the question entered by the user and transfers it to the server via an HTTP request.

[0812] The server analyzes the received question using natural language processing (NLP) techniques. An NLP library such as SpaCy is used. Based on the analysis results, the server searches for relevant data from private information in the database and external public information. ElasticSearch is used for the search.

[0813] The server generates an answer using a Retrieval Augmented Generation (RAG) model based on relevant data. The generative AI model used is the commonly used GPT-3. The server incorporates information related to the question into a prompt sentence and inputs it into the RAG model to generate the optimal answer.

[0814] Furthermore, the emotion engine recognizes the user's emotional state from their input and biometric information (e.g., voice, facial expression). This engine uses, for example, Dialogflow. If the user is feeling stressed or anxious, the server adjusts the tone of the response to a gentler tone depending on the user's emotional state.

[0815] The generated answer is sent from the server to the terminal and displayed to the user, for example, an adapted answer like "The add function takes two arguments and returns their sum. Is there anything else I can help you with?"

[0816] Examples of specific prompts include:

[0817] "If a user is experiencing a stress level from the system and asks, 'What does the add function in utils.py do?', formulate your answer in a gentle tone."

[0818] In this way, users can not only obtain information from the system quickly and accurately, but also receive appropriate responses that take into account their emotional state, which greatly improves user satisfaction and business efficiency.

[0819] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0820] Step 1:

[0821] The server collects confidential internal information from the company and stores it in a database. Specific inputs include system source code, technical specifications, and internal documents, and receives this information as Excel or PDF files. The server analyzes these files, extracts relevant items, and stores them in a database using SQL Server. The output is a database containing the confidential information.

[0822] Step 2:

[0823] The user uses the terminal to input a question about the system. An example of input is "What does the add function in utils.py do?" The terminal transfers the question data entered by the user to the server as an HTTP request. The output here is the question data transferred to the server.

[0824] Step 3:

[0825] The server analyzes questions received from the terminal. The input is the question text, which is analyzed using natural language processing (NLP) technology. SpaCy is used as the NLP library, and keywords from the question are extracted as the analysis result. The output is the extracted keywords and their related information.

[0826] Step 4:

[0827] Based on the analysis results, the server searches for relevant data from private information in the database and external public information. The specific input is the extracted keywords. The server uses ElasticSearch to quickly search for relevant information in the database and on the Internet. The output is a dataset of the searched private and public information.

[0828] Step 5:

[0829] The server generates an answer using a Retrieval Augmented Generation (RAG) model based on the retrieved information. The input is the retrieved dataset, and a generative AI model (e.g., GPT-3) is used to create a prompt sentence, which is then input to the model. The output is the generated answer text.

[0830] Step 6:

[0831] The emotion engine recognizes the user's emotional state based on their input and biometric information. The input is the user's question text and biometric information (e.g., voice, facial expression). The emotion engine analyzes this information and identifies the user's emotional state (e.g., stress, anxiety). The output is the recognized emotional state data.

[0832] Step 7:

[0833] The server adjusts the answer content based on the emotional state data obtained from the emotion engine. The input is the generated answer text and the user's emotional state data. The server adjusts the tone of the answer, such as softening it, depending on the emotional state. The output is the adjusted optimal answer text.

[0834] Step 8:

[0835] The server sends the adjusted answer to the terminal. The input is the adjusted answer text. The terminal displays the received answer to the user. For example, the answer "The add function takes two arguments and returns their sum. Is there anything else I can help you with?" The output is the answer text displayed to the user.

[0836] Through these processing steps, the system can efficiently utilize both private and public information within a company to provide optimal answers to users' questions. It can also adjust responses based on the user's emotional state, contributing to improved user satisfaction.

[0837] (Application example 2)

[0838] 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."

[0839] Modern companies manage vast amounts of private and public information, but lack systems for effectively utilizing that information. Furthermore, when users ask questions, it is difficult to provide appropriate answers quickly, and the lack of consideration for the user's emotional state can negatively impact the user experience. This invention aims to provide a system that combines private and public information within a company to not only provide optimal answers to users' questions but also generate answers that take into account the user's emotional state.

[0840] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0841] In this invention, the server includes an information storage device that stores private information within the company, an input device that receives questions from users, an information processing device that generates answers to the questions based on the information storage device and public information, an information retrieval generation model that generates answers to the questions by combining the private information and public information in the information storage device, a display device that provides the generated answers to the users, and an emotion recognition device that identifies the emotional state of the users and adjusts the content of the answers. This makes it possible to effectively utilize the vast amount of information within the company and provide appropriate and prompt answers to users' questions, as well as flexibly respond to answers that take the users' emotions into consideration.

[0842] An "information storage device" is a device for storing confidential information within a company and is used to efficiently manage and search for information.

[0843] An "input device" is a device for receiving questions from a user, and is a device that plays a role in inputting user questions into the system.

[0844] An "information processing device" is a device that generates answers to questions based on information storage devices and public information, and is a device used to analyze information and provide optimal answers.

[0845] The term "information retrieval generation model" refers to a model that generates answers to questions by combining private and public information in an information storage device, and in particular refers to the Retrieval Augmented Generation (RAG) model.

[0846] A "display device" is a device for providing the generated answer to the user, and enabling the user to visually confirm the answer.

[0847] An "emotion recognition device" is a device that identifies the emotional state of a user and adjusts the content of a response, and includes technology for analyzing the user's emotions.

[0848] "Non-public information" refers to information that is managed internally by a company and is not made public, including internal documents, source code, specifications, etc.

[0849] "Public information" refers to information that is publicly available, including materials on the Internet and public documents.

[0850] The present invention is a system that generates answers to user questions by combining private and public information within a company, and further includes an emotion engine that recognizes the user's emotions and adjusts the content of the answers. This system is composed of an information storage device, an input device, an information processing device, an information retrieval generation model, a display device, and an emotion recognition device.

[0851] First, the information storage device collects and stores confidential information from within a company for efficient management and retrieval. Non-public information includes system source code and specifications. This information storage device allows the company's internal information to be organized and maintained.

[0852] Next, the user inputs a question into the system through an input device, which receives the question and forwards it to the information processing device, where it is processed as an input for the information retrieval generative model.

[0853] The information processing device analyzes the question received from the input device and searches for related data from private information and public information in the information storage device. The related data is searched for using a search algorithm based on the content of the question.

[0854] The information processing device uses an information retrieval generation model (particularly the Retrieval Augmented Generation (RAG) model) to generate answers to questions. This model retrieves information related to the question from an information storage device and public information, and generates an optimal answer based on that information.

[0855] Furthermore, the emotion recognition device identifies the user's emotional state. This emotion recognition device recognizes emotions by analyzing the tone and content of the user's input (e.g., question) or biometric information (e.g., voice, facial expression). This allows it to provide a response in a gentle tone that reflects the user's state, such as if the user is feeling stressed or anxious.

[0856] For example, if a user asks the question "Please tell me about the system's security settings," and the emotion recognition device recognizes that the user is feeling stressed, the information processing device will generate a response such as, "Don't worry. The system's security settings are configured individually for each user, and detailed instructions are provided in section 5.3 of the Security Policy Manual."

[0857] The generated answers are provided to the user via a display device. Through this system, the user can quickly and accurately obtain specific answers to their questions, and can also receive appropriate responses that take into account their own emotional state.

[0858] By introducing this system, companies can not only effectively utilize internal information and provide prompt and appropriate answers to questions, but also increase user satisfaction.

[0859] Specific prompt examples

[0860] User Question: "What are the security settings on my system?"

[0861] Emotional state: "When the emotion recognition device recognizes that the user is feeling stressed"

[0862] Tailored answer: "Don't worry, system security settings are configured individually for each user and detailed instructions are provided in section 5.3 of the Security Policy Manual."

[0863] This makes it possible to provide the most appropriate answer to the question and to realize a flexible response that takes into consideration the user's feelings.

[0864] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0865] Step 1:

[0866] A user inputs a question using an input device such as a smartphone.

[0867] Input: User question (e.g., "What are the security settings on my system?")

[0868] Output: User question data

[0869] The user inputs a question through the input device, which is recorded in the input device, and the question data is sent to the next processing step.

[0870] Step 2:

[0871] The terminal receives a question from the user and transfers it to the information processing device.

[0872] Input: User question data

[0873] Output: Data transferred to information processing device

[0874] The terminal receives the query data and prepares it for transmission to the information processing device, where data format conversion and encoding are performed.

[0875] Step 3:

[0876] The information processing device analyzes the question data and searches for related private information and public information.

[0877] Input: Question data

[0878] Output: Relevant data (company internal private information, public information)

[0879] The information processing device analyzes the question data and searches for relevant information from databases and the Internet, using a search algorithm to extract information that best suits the question.

[0880] Step 4:

[0881] The information processing device generates an optimal answer using an information retrieval generation model.

[0882] Input: Question data, related data (private information, public information)

[0883] Output: The generated answer

[0884] The information retrieval generative model (RAG model) takes relevant data as input and generates the best answer to the question, using advanced natural language processing to ensure data consistency and integrity.

[0885] Step 5:

[0886] The emotion recognizer analyzes the user's emotional state and adjusts the response as needed.

[0887] Input: User question data (tone, content), biometric information (e.g., voice, facial expression)

[0888] Output: Adjusted answer

[0889] The emotion recognition device analyzes the user's emotions and adjusts the tone of the response to be gentler if the user is feeling stressed or anxious. It also uses biometric information for more accurate emotion recognition.

[0890] Step 6:

[0891] The adjusted answer is provided to the user via a display device.

[0892] Input: Adjusted Answer

[0893] Output: What the user sees

[0894] The generated answer is sent to a display device and displayed to the user, thereby enabling the user to quickly obtain an appropriate answer.

[0895] Through the above steps, appropriate and prompt answers to the user's questions are provided, and at the same time, flexible responses that take into account the user's emotional state are realized.

[0896] 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.

[0897] 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.

[0898] 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.

[0899] [Fourth embodiment]

[0900] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0901] 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.

[0902] 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).

[0903] 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.

[0904] 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.

[0905] 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).

[0906] 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.

[0907] 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.

[0908] 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.

[0909] 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.

[0910] 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.

[0911] 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.

[0912] 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."

[0913] The present invention relates to a system that generates answers to user questions by combining private and public information within a company. This system is composed of a database, a terminal, and a server.

[0914] First, the server collects private information from within the company and stores it in a database. Private information includes system source code and specifications. The database is designed to efficiently manage and search private information.

[0915] When using the system, the user inputs a question through the terminal. The terminal receives the user's question and transfers it to the server. This question triggers a match with the data in the system.

[0916] The server analyzes the question received from the device and searches for relevant data from private information in the database using a search algorithm based on the content of the question. In addition, the server also retrieves public information and uses these sources in combination to generate an answer to the question.

[0917] To generate answers, a technology called the Retrieval Augmented Generation (RAG) model is used. The RAG model first retrieves information related to the question from databases and public information, and then generates an answer based on that information. This model makes it possible to provide the best answer to the question.

[0918] The generated answer is sent from the server to the terminal and displayed to the user, who can instantly obtain specific answers about the system through the terminal.

[0919] As a specific example, if a user asks, "What does the add function in utils.py do?", the server searches the database for the appropriate source code sections and specifications, combines them with public information, and generates an answer. In this case, the server generates the answer, "The add function takes two arguments and returns their sum," and provides it to the user through the terminal.

[0920] By implementing this system, system personnel can quickly and accurately obtain information about system specifications and source code, which contributes to a significant improvement in work efficiency.

[0921] The processing flow will be explained below.

[0922] Step 1:

[0923] The server collects private information from within the company and stores it in a database. The private information includes system source code and specifications. The database is designed to efficiently manage and search private information.

[0924] Step 2:

[0925] The user inputs a question about the system through a terminal, which is processed as a string of characters.

[0926] Step 3:

[0927] The terminal receives a question entered by the user and forwards the question to the server, which then formats the user's input appropriately and sends it to the server.

[0928] Step 4:

[0929] The server analyzes the question received from the device using natural language processing (NLP) technology to extract the intent and keywords of the question.

[0930] Step 5:

[0931] Based on the parsed question, the server searches for relevant data from private information in a database using algorithms to identify sections of source code or specifications that match the query.

[0932] Step 6:

[0933] The server also retrieves public information related to the question, which is collected from technical documents, forums, and other sources on the Internet.

[0934] Step 7:

[0935] The server generates answers to questions by combining private information in the database with the public information it has retrieved, using a Retrieval Augmented Generation (RAG) model.

[0936] Step 8:

[0937] The server then sends the generated answer to the terminal, where it is properly formatted and presented in a user-friendly form.

[0938] Step 9:

[0939] The terminal displays the answer received from the server to the user, and the user can instantly obtain specific answers about the system through the terminal.

[0940] This series of steps allows the user to quickly and accurately obtain information about the system specifications and source code.

[0941] Example 1

[0942] 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."

[0943] There is a lot of private information within companies, and efficient use of this information can improve business efficiency, but previous systems lacked the means to properly integrate this information and generate optimal answers to user questions.In addition, there was no way to properly combine public and private information, which meant that users could not be provided with sufficient information.

[0944] 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.

[0945] In this invention, the server includes information storage means, question receiving means, answer generating means, information integrating means, and answer providing means, which allows the server to combine private and public information within the company to generate optimal answers and provide them to users.

[0946] "Information storage means" refers to systems and databases for collecting, efficiently storing, and managing non-public information within a company.

[0947] The "question receiving means" refers to an interface for receiving a question from a user and transferring it to the server in an appropriate format.

[0948] "Answer generation means" refers to algorithms and technologies for generating optimal answers based on received questions, using private and public information within the company.

[0949] "Information synthesis means" refers to means for combining private and public information to generate an integrated answer to a question.

[0950] The "answer providing means" refers to a means for presenting the generated answer to the user.

[0951] The present invention relates to a system that generates answers to user questions by combining private and public information within a company. This system is composed of an information storage means, a question receiving means, an answer generating means, an information integration means, and an answer providing means.

[0952] First, the server collects private information from within the company and stores it in a database, which is a means of storing information.Private information includes system program code and specifications.This database is designed to efficiently manage and search private information.

[0953] The user inputs a question through the terminal. The terminal receives the user's question and sends it to the server as a question receiving means. This question becomes a trigger for matching with the data inside the system.

[0954] The server uses a natural language processing library (e.g., spaCy or Transformers) to analyze the question, and then searches for relevant private information from a database, which is an information storage means, based on the analyzed question.

[0955] Additionally, the server uses APIs on the Internet to gather public information, for example, by retrieving the required information from specific technology sites or open source repositories.

[0956] After obtaining the private and public information, the server uses the Retrieval Augmented Generation (RAG) model as an information integration method to integrate each piece of information and generate the optimal answer. This model is an advanced algorithm that generates answers based on information related to the question.

[0957] The generated answer is sent from the server to the terminal, and the terminal displays it to the user as a means of providing the answer. As a specific example, if a user asks, "What does the add function in utils.py do?", the server searches the database for the appropriate program code section and specification, and combines it with public information to generate an answer. In this case, the server generates the answer, "The add function takes two arguments and returns their sum," and provides it to the user via the terminal.

[0958] An example of a prompt to input to a generative AI model is:

[0959] "A user has a system-related question. The question is: 'What does the add function in utils.py do?' Please provide a detailed answer using relevant parts of your database and publicly available information."

[0960] As described above, the present invention realizes a system that utilizes both private and public information within a company in an integrated manner to provide quick and accurate answers to user questions. This allows system personnel to efficiently obtain appropriate information, and is expected to significantly improve work efficiency.

[0961] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0962] Step 1:

[0963] The user inputs a question through the terminal.

[0964] (Input) User question (e.g., "What does the add function in utils.py do?")

[0965] (Operation) A user enters a question into the question input form using a dedicated application or a web browser. The user enters specific questions about the question in text format.

[0966] (Output) The user's question is saved in text format on the terminal.

[0967] Step 2:

[0968] The terminal sends a question to the server.

[0969] (Input) User question text (e.g., "What does the add function in utils.py do?")

[0970] (Behavior) When the user clicks the submit button, an HTTP POST request is generated and sent to the server's API endpoint (e.g., "api.example.com / query").

[0971] (Output) The server receives the user's question text.

[0972] Step 3:

[0973] The server receives and parses the query.

[0974] (Input) User question text (e.g., "What does the add function in utils.py do?")

[0975] (Operation) The server uses a natural language processing library (e.g., spaCy, Transformers) to analyze the question text received. It performs processes such as text tokenization, part-of-speech tagging, and semantic analysis. As a result of the analysis, "utils.py" and the "add function" are identified.

[0976] (Output) The parsed question data (e.g., "File name: utils.py", "Function name: add function") is generated.

[0977] Step 4:

[0978] The server searches the database for relevant non-public information.

[0979] (Input) Parsed question data (e.g., "File name: utils.py", "Function name: add function")

[0980] (Behavior) The server uses an SQL query to retrieve the relevant program code or specification section from a database (e.g. MySQL), specifically, it executes the query "SELECT FROM SourceCode WHERE fileName='utils.py' AND functionName='add'".

[0981] (Output) Related private information (e.g., "source code of the add function" and "specification of the add function") is obtained.

[0982] Step 5:

[0983] The server collects public information from external APIs.

[0984] (Input) Analyzed question data and non-public information

[0985] (Operation) The server calls APIs on the Internet to collect public information. For example, it uses the GitHub API to obtain public repository information related to the "utils.py" file and the "add function."

[0986] (Output) Public information (e.g., "Related information about the GitHub repository") is obtained.

[0987] Step 6:

[0988] The server uses the RAG model to generate answers to questions.

[0989] (Input) Private and Public Information

[0990] (Operation) The server inputs the acquired private and public information into the RAG model and generates an integrated answer. Using the generative AI model, the optimal answer to the question (e.g., "The add function takes two arguments and returns their sum") is generated.

[0991] (Output) A generated answer (e.g., "The add function takes two arguments and returns their sum") is generated.

[0992] Step 7:

[0993] The server generates a response and sends it to the terminal.

[0994] (Input) Generated answer (e.g., "The add function takes two arguments and returns their sum")

[0995] (Operation) The server sends the generated answer to the terminal as an HTTP response. The response data is usually sent in JSON format.

[0996] (Output) The terminal receives the generated response.

[0997] Step 8:

[0998] The terminal displays the answer to the user.

[0999] (Input) Generated answer (e.g., "The add function takes two arguments and returns their sum")

[1000] (Operation) The device analyzes the received answer and displays it in a designated area on the web page or in the application, allowing the user to visually confirm the answer.

[1001] (Output) The user can check the generated answer through the terminal.

[1002] (Application example 1)

[1003] 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."

[1004] Conventional systems do not combine private and public information within a company to search and generate answers, making it difficult to quickly and accurately provide the information users are looking for. Furthermore, it is difficult to provide answers to security-related questions by appropriately integrating internal information with the latest external information. The present invention aims to solve these problems and provide a system that quickly provides the information users are looking for.

[1005] 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.

[1006] In this invention, the server includes a data storage means for storing private information within a company, a communication device means for receiving questions from users, a processing device means for generating answers to questions based on the data storage means and public information, a generative AI model-based means for exploring and generating answers to questions by combining private information and public information in the data storage means, an interface means for providing the generated answers to users, a means for acquiring security-related information from public information and combining it with internal information to generate answers, and a means for generating answers based on security guidelines, policies, and the latest security trends. This makes it possible to quickly provide optimal answers that integrate internal information and external information.

[1007] "Data storage means" means a device that has the function of storing, efficiently managing, and searching non-public information within a company.

[1008] The "communication device means" is a device for receiving a question from a user and transferring the question to a server.

[1009] "Processing device means" refers to a device for generating answers to questions based on the data storage means and public information.

[1010] A "generative AI model-based method" is a method that uses an artificial intelligence model to explore and generate answers to questions by combining private and public information.

[1011] An "interface means" is a device or method for providing generated answers to a user.

[1012] "Public Information" is information that is generally accessible outside the company, including security-related news and best practices.

[1013] "Security guidelines" are documents that outline security guidelines and rules established within a company.

[1014] "Security Trends" is information that shows the latest security trends and developments.

[1015] The present invention is a system that combines private information and public information within a company to generate answers to questions from users. Specific embodiments are described below.

[1016] The server uses data storage means to store the company's internal non-public information, such as internal security guidelines and policies, the latest security trends, etc. The data storage means is also provided to efficiently manage and search the non-public information.

[1017] The user inputs a question through a communication device, and the terminal receives the user's question and transfers it to the server.

[1018] The server uses the processing device means to analyze the received question and generate an answer based on the private and public information in the data storage means. Specifically, the server uses means based on a generative AI model to explore and generate an answer to the question. The model utilizes a Retrieval Augmented Generation (RAG) model to retrieve information related to the question from the data storage means and public information and generate an answer based on that information.

[1019] The generated answers are provided to the user via an interface means, which may include a terminal and a display device, and allows the user to intuitively view the generated answers.

[1020] It also includes a means to retrieve security-related information from public sources and combine it with internal information to generate answers, ensuring that answers to security questions are provided while taking into account the latest security trends and best practices.

[1021] For example, if a user asks, "What are the latest ransomware attack countermeasures?", the server generates an answer based on security guidelines in an internal database and the latest public information. Through analysis by the generative AI model, the server can provide the answer, "To counter the latest ransomware, we recommend taking regular backups and implementing multi-factor authentication."

[1022] An example of a prompt sentence could be in the following text format:

[1023] "What are the latest ransomware attack countermeasures?"

[1024] In this way, the present invention enables the user to quickly obtain the most appropriate answer that integrates internal information and external information.

[1025] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1026] Step 1:

[1027] The terminal receives a question from the user. The received question is saved as a prompt, which triggers the processing of the entire system.

[1028] Input: The question the user types into the terminal

[1029] Output: Saved prompt statement

[1030] Step 2:

[1031] The terminal forwards the received question to the server, which then starts the next process based on the prompt.

[1032] Input: prompt statement

[1033] Output: Prompt text forwarded to the server

[1034] Step 3:

[1035] The server parses the prompt and retrieves relevant non-public information from a data storage means, including internal security guidelines and policies.

[1036] Input: prompt statement

[1037] Output: Retrieved private information

[1038] Step 4:

[1039] The server calls external APIs to obtain the latest security-related information from public information sources, and the obtained information is managed as public information.

[1040] Input: prompt statement

[1041] Output: Public information obtained

[1042] Step 5:

[1043] The server uses a generative AI model (RAG model) to combine private and public information to generate the best answer to the question.

[1044] Input: Private information, Public information, Prompt text

[1045] Output: The generated answer

[1046] Step 6:

[1047] The server sends the generated answers to the device, where they are presented to the user in an intuitive format.

[1048] Input: Generated Answer

[1049] Output: Answer sent to terminal

[1050] Step 7:

[1051] The terminal displays the received answer on the display device, and the user thereby obtains the answer to the question.

[1052] Input: The answer received from the server

[1053] Output: The answer displayed to the user

[1054] This processing flow allows users to quickly obtain optimal answers that integrate internal and external information.

[1055] 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.

[1056] The present invention relates to a system that generates answers to user questions by combining private and public information within a company, and further includes an emotion engine that recognizes the user's emotions and adjusts the content of the answers. This system is composed of a database, a terminal, a server, and an emotion engine.

[1057] First, the server collects private information from within the company and stores it in a database. Private information includes system source code and specifications. This database is designed to efficiently manage and search private information.

[1058] When using the system, the user inputs a question to the system through the terminal. When the terminal receives the user's question, it forwards it to the server. The question is processed as input to the data and emotion engine within the system.

[1059] The server analyzes the query received from the device and searches for relevant data from private and public information in the database. The server uses a search algorithm based on the content of the query to search for relevant data. The server also retrieves public information and uses these sources in combination to generate an answer to the query.

[1060] To generate answers, a technology called the Retrieval Augmented Generation (RAG) model is used. The RAG model first retrieves information related to the question from databases and public information, and then generates an answer based on that information. This model makes it possible to provide the best answer to the question.

[1061] Furthermore, the present invention includes an emotion engine that identifies the user's emotional state. This emotion engine recognizes emotions by analyzing the tone and content of the user's input (e.g., question) or biometric information (e.g., voice, facial expression). This allows the system to provide a response in a gentle tone appropriate to the user's state if the user is feeling stressed or anxious.

[1062] As a specific example, if a user asks "What does the add function in utils.py do?" and the emotion engine recognizes that the user is feeling stressed, the server will generate a response such as "The add function takes two arguments and returns their sum. Is there anything else I can help you with?" In this way, the emotion engine can respond in a way that takes into account the user's emotional state.

[1063] The generated answers are sent from the server to the terminal and displayed to the user, allowing the user to quickly and accurately obtain specific answers about the system through the terminal, and to receive appropriate responses that take into account their own emotional state.

[1064] By introducing this system, system personnel will not only be able to quickly and accurately obtain information about system specifications and source code, but will also be able to respond according to the user's emotional state, contributing to a significant improvement in work efficiency.

[1065] The processing flow will be explained below.

[1066] Step 1:

[1067] The server collects private information from within the company and stores it in a database. The private information includes system source code and specifications. The database is designed to efficiently manage and search this private information.

[1068] Step 2:

[1069] The user types a question about the system via the terminal, for example, "What does the add function in utils.py do?"

[1070] Step 3:

[1071] The terminal receives the question entered by the user and transfers the question to the server, where the content of the question is sent as a character string.

[1072] Step 4:

[1073] The device detects the user's emotional state using an emotion engine that analyzes the user's voice and text emotions, for example, recognizing through voice analysis that the user is feeling stressed.

[1074] Step 5:

[1075] The server analyzes the question received from the device using natural language processing (NLP) technology, extracts the intent and keywords of the question, and generates an appropriate search query.

[1076] Step 6:

[1077] The server searches for relevant data (e.g., source code and specifications) from private information in the database based on the analyzed question, and integrates multiple related data as needed.

[1078] Step 7:

[1079] The server also retrieves public information related to the question from the Internet, including technical documentation, forum posts, blog articles, etc.

[1080] Step 8:

[1081] The server generates answers to questions based on the private information in the database and the public information it retrieves using a Retrieval Augmented Generation (RAG) model, which combines the retrieved information to generate the best answer.

[1082] Step 9:

[1083] The server takes into account the user's emotional state as detected by the emotion engine and adjusts the tone and content of the response: for example, if the user is feeling stressed, the response will be more polite and include additional explanations if necessary.

[1084] Step 10:

[1085] The server then sends the generated answer to the device, which is then formatted appropriately and presented in a way that is easy for the user to understand.

[1086] Step 11:

[1087] The terminal displays the answer it receives from the server to the user, for example, "The add function takes two arguments and returns their sum. Is there anything else I can help you with?"

[1088] This flow allows users to quickly and accurately obtain answers regarding system specifications and source code, and the emotion engine enables responses based on the user's emotional state, providing a better user experience.

[1089] Example 2

[1090] 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."

[1091] Modern companies need to efficiently utilize both private and public information to respond to user questions quickly and accurately. However, there is a lack of ways to adjust responses while taking into account the user's emotional state, which makes it difficult to improve user satisfaction and work efficiency. In particular, when a user is feeling stressed or anxious, a response that adapts to that state is required, but this has been difficult to achieve with conventional systems.

[1092] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a database storing private internal company information, a terminal receiving questions from users, means for generating answers to questions based on the database and public information, an information retrieval expansion and generation model for generating answers to questions by combining private and public information in the database, an emotion engine for recognizing the user's emotional state and adjusting the content of the answer, and means for providing the generated answer to the user. As a result, when a user inputs a question, not only is an optimal answer provided that efficiently combines private internal company information and public information, but also a gentle response that corresponds to the user's emotional state is possible. This improves user satisfaction and business efficiency.

[1093] "Internal non-public information" refers to information that is not publicly available within a company, including system source code, technical specifications, internal documents, etc.

[1094] A "database" is an information collection and management system established to efficiently manage and search confidential information within a company.

[1095] A "terminal" is a device through which a user inputs questions about the system, and includes PCs, smartphones, etc.

[1096] The "server" is the central computer of the system that analyzes questions received from terminals and generates answers to the questions based on private and public information in the database.

[1097] "Public Information" is information obtained from an external source, such as the Internet, and made available to the public.

[1098] "Information Retrieval Augmented Generative Model" refers to a technology that combines private and public information in a database to generate answers to questions, and includes generative algorithms that utilize natural language processing.

[1099] The "emotion engine" is a system that recognizes the user's emotional state from their input and biometric information, and adjusts the response content according to that emotional state.

[1100] "Natural language processing technology" refers to technology that enables computers to understand and process human language, and includes text analysis and language models.

[1101] "Biometric information" refers to information obtained from the user's body, such as their voice and facial expressions, and is data used to recognize their emotional state.

[1102] The present invention relates to a system that efficiently utilizes private and public information within a company to provide quick and accurate answers to user questions. This system is composed of a database, terminals, a server, and an emotion engine.

[1103] First, the server collects private information from within the company. This private information includes system source code, technical specifications, internal documents, etc. The collected private information is stored in a database using SQL Server. The database is set up to efficiently manage and search private information.

[1104] Next, the user inputs a question about the system through a terminal (PC or smartphone). For example, they input a question like, "What does the add function in utils.py do?" The terminal receives the question entered by the user and transfers it to the server via an HTTP request.

[1105] The server analyzes the received question using natural language processing (NLP) techniques. An NLP library such as SpaCy is used. Based on the analysis results, the server searches for relevant data from private information in the database and external public information. ElasticSearch is used for the search.

[1106] The server generates an answer using a Retrieval Augmented Generation (RAG) model based on relevant data. The generative AI model used is the commonly used GPT-3. The server incorporates information related to the question into a prompt sentence and inputs it into the RAG model to generate the optimal answer.

[1107] Furthermore, the emotion engine recognizes the user's emotional state from their input and biometric information (e.g., voice, facial expression). This engine uses, for example, Dialogflow. If the user is feeling stressed or anxious, the server adjusts the tone of the response to a gentler tone depending on the user's emotional state.

[1108] The generated answer is sent from the server to the terminal and displayed to the user, for example, an adapted answer like "The add function takes two arguments and returns their sum. Is there anything else I can help you with?"

[1109] Examples of specific prompts include:

[1110] "If a user is experiencing a stress level from the system and asks, 'What does the add function in utils.py do?', formulate your answer in a gentle tone."

[1111] In this way, users can not only obtain information from the system quickly and accurately, but also receive appropriate responses that take into account their emotional state, which greatly improves user satisfaction and business efficiency.

[1112] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1113] Step 1:

[1114] The server collects confidential internal information from the company and stores it in a database. Specific inputs include system source code, technical specifications, and internal documents, and receives this information as Excel or PDF files. The server analyzes these files, extracts relevant items, and stores them in a database using SQL Server. The output is a database containing the confidential information.

[1115] Step 2:

[1116] The user uses the terminal to input a question about the system. An example of input is "What does the add function in utils.py do?" The terminal transfers the question data entered by the user to the server as an HTTP request. The output here is the question data transferred to the server.

[1117] Step 3:

[1118] The server analyzes questions received from the terminal. The input is the question text, which is analyzed using natural language processing (NLP) technology. SpaCy is used as the NLP library, and keywords from the question are extracted as the analysis result. The output is the extracted keywords and their related information.

[1119] Step 4:

[1120] Based on the analysis results, the server searches for relevant data from private information in the database and external public information. The specific input is the extracted keywords. The server uses ElasticSearch to quickly search for relevant information in the database and on the Internet. The output is a dataset of the searched private and public information.

[1121] Step 5:

[1122] The server generates an answer using a Retrieval Augmented Generation (RAG) model based on the retrieved information. The input is the retrieved dataset, and a generative AI model (e.g., GPT-3) is used to create a prompt sentence, which is then input to the model. The output is the generated answer text.

[1123] Step 6:

[1124] The emotion engine recognizes the user's emotional state based on their input and biometric information. The input is the user's question text and biometric information (e.g., voice, facial expression). The emotion engine analyzes this information and identifies the user's emotional state (e.g., stress, anxiety). The output is the recognized emotional state data.

[1125] Step 7:

[1126] The server adjusts the answer content based on the emotional state data obtained from the emotion engine. The input is the generated answer text and the user's emotional state data. The server adjusts the tone of the answer, such as softening it, depending on the emotional state. The output is the adjusted optimal answer text.

[1127] Step 8:

[1128] The server sends the adjusted answer to the terminal. The input is the adjusted answer text. The terminal displays the received answer to the user. For example, the answer "The add function takes two arguments and returns their sum. Is there anything else I can help you with?" The output is the answer text displayed to the user.

[1129] Through these processing steps, the system can efficiently utilize both private and public information within a company to provide optimal answers to users' questions. It can also adjust responses based on the user's emotional state, contributing to improved user satisfaction.

[1130] (Application example 2)

[1131] 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."

[1132] Modern companies manage vast amounts of private and public information, but lack systems for effectively utilizing that information. Furthermore, when users ask questions, it is difficult to provide appropriate answers quickly, and the lack of consideration for the user's emotional state can negatively impact the user experience. This invention aims to provide a system that combines private and public information within a company to not only provide optimal answers to users' questions but also generate answers that take into account the user's emotional state.

[1133] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1134] In this invention, the server includes an information storage device that stores private information within the company, an input device that receives questions from users, an information processing device that generates answers to the questions based on the information storage device and public information, an information retrieval generation model that generates answers to the questions by combining the private information and public information in the information storage device, a display device that provides the generated answers to the users, and an emotion recognition device that identifies the emotional state of the users and adjusts the content of the answers. This makes it possible to effectively utilize the vast amount of information within the company and provide appropriate and prompt answers to users' questions, as well as flexibly respond to answers that take the users' emotions into consideration.

[1135] An "information storage device" is a device for storing confidential information within a company and is used to efficiently manage and search for information.

[1136] An "input device" is a device for receiving questions from a user, and is a device that plays a role in inputting user questions into the system.

[1137] An "information processing device" is a device that generates answers to questions based on information storage devices and public information, and is a device used to analyze information and provide optimal answers.

[1138] The term "information retrieval generation model" refers to a model that generates answers to questions by combining private and public information in an information storage device, and in particular refers to the Retrieval Augmented Generation (RAG) model.

[1139] A "display device" is a device for providing the generated answer to the user, and enabling the user to visually confirm the answer.

[1140] An "emotion recognition device" is a device that identifies the emotional state of a user and adjusts the content of a response, and includes technology for analyzing the user's emotions.

[1141] "Non-public information" refers to information that is managed internally by a company and is not made public, including internal documents, source code, specifications, etc.

[1142] "Public information" refers to information that is publicly available, including materials on the Internet and public documents.

[1143] The present invention is a system that generates answers to user questions by combining private and public information within a company, and further includes an emotion engine that recognizes the user's emotions and adjusts the content of the answers. This system is composed of an information storage device, an input device, an information processing device, an information retrieval generation model, a display device, and an emotion recognition device.

[1144] First, the information storage device collects and stores confidential information from within a company for efficient management and retrieval. Non-public information includes system source code and specifications. This information storage device allows the company's internal information to be organized and maintained.

[1145] Next, the user inputs a question into the system through an input device, which receives the question and forwards it to the information processing device, where it is processed as an input for the information retrieval generative model.

[1146] The information processing device analyzes the question received from the input device and searches for related data from private information and public information in the information storage device. The related data is searched for using a search algorithm based on the content of the question.

[1147] The information processing device uses an information retrieval generation model (particularly the Retrieval Augmented Generation (RAG) model) to generate answers to questions. This model retrieves information related to the question from an information storage device and public information, and generates an optimal answer based on that information.

[1148] Furthermore, the emotion recognition device identifies the user's emotional state. This emotion recognition device recognizes emotions by analyzing the tone and content of the user's input (e.g., question) or biometric information (e.g., voice, facial expression). This allows it to provide a response in a gentle tone that reflects the user's state, such as if the user is feeling stressed or anxious.

[1149] For example, if a user asks the question "Please tell me about the system's security settings," and the emotion recognition device recognizes that the user is feeling stressed, the information processing device will generate a response such as, "Don't worry. The system's security settings are configured individually for each user, and detailed instructions are provided in section 5.3 of the Security Policy Manual."

[1150] The generated answers are provided to the user via a display device. Through this system, the user can quickly and accurately obtain specific answers to their questions, and can also receive appropriate responses that take into account their own emotional state.

[1151] By introducing this system, companies can not only effectively utilize internal information and provide prompt and appropriate answers to questions, but also increase user satisfaction.

[1152] Specific prompt examples

[1153] User Question: "What are the security settings on my system?"

[1154] Emotional state: "When the emotion recognition device recognizes that the user is feeling stressed"

[1155] Tailored answer: "Don't worry, system security settings are configured individually for each user and detailed instructions are provided in section 5.3 of the Security Policy Manual."

[1156] This makes it possible to provide the most appropriate answer to the question and to realize a flexible response that takes into consideration the user's feelings.

[1157] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1158] Step 1:

[1159] A user inputs a question using an input device such as a smartphone.

[1160] Input: User question (e.g., "What are the security settings on my system?")

[1161] Output: User question data

[1162] The user inputs a question through the input device, which is recorded in the input device, and the question data is sent to the next processing step.

[1163] Step 2:

[1164] The terminal receives a question from the user and transfers it to the information processing device.

[1165] Input: User question data

[1166] Output: Data transferred to information processing device

[1167] The terminal receives the query data and prepares it for transmission to the information processing device, where data format conversion and encoding are performed.

[1168] Step 3:

[1169] The information processing device analyzes the question data and searches for related private information and public information.

[1170] Input: Question data

[1171] Output: Relevant data (company internal private information, public information)

[1172] The information processing device analyzes the question data and searches for relevant information from databases and the Internet, using a search algorithm to extract information that best suits the question.

[1173] Step 4:

[1174] The information processing device generates an optimal answer using an information retrieval generation model.

[1175] Input: Question data, related data (private information, public information)

[1176] Output: The generated answer

[1177] The information retrieval generative model (RAG model) takes relevant data as input and generates the best answer to the question, using advanced natural language processing to ensure data consistency and integrity.

[1178] Step 5:

[1179] The emotion recognizer analyzes the user's emotional state and adjusts the response as needed.

[1180] Input: User question data (tone, content), biometric information (e.g., voice, facial expression)

[1181] Output: Adjusted answer

[1182] The emotion recognition device analyzes the user's emotions and adjusts the tone of the response to be gentler if the user is feeling stressed or anxious. It also uses biometric information for more accurate emotion recognition.

[1183] Step 6:

[1184] The adjusted answer is provided to the user via a display device.

[1185] Input: Adjusted Answer

[1186] Output: What the user sees

[1187] The generated answer is sent to a display device and displayed to the user, thereby enabling the user to quickly obtain an appropriate answer.

[1188] Through the above steps, appropriate and prompt answers to the user's questions are provided, and at the same time, flexible responses that take into account the user's emotional state are realized.

[1189] 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.

[1190] 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.

[1191] 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.

[1192] 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.

[1193] 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.

[1194] 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.

[1195] 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).

[1196] 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.

[1197] 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."

[1198] 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.

[1199] 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).

[1200] 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.

[1201] 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.

[1202] 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.

[1203] 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.

[1204] 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.

[1205] 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.

[1206] 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.

[1207] 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.

[1208] 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.

[1209] 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.

[1210] The following is further disclosed regarding the above embodiment.

[1211] (Claim 1)

[1212] A database that stores internal, non-public information of a company;

[1213] a terminal for receiving a question from a user;

[1214] a server that generates answers to questions based on the database and public information;

[1215] a Retrieval Augmented Generation (RAG) model that combines private and public information in the database to generate answers to questions;

[1216] A system with a means for providing generated answers to a user.

[1217] (Claim 2)

[1218] 10. The system of claim 1, further comprising means for retrieving non-public information related to the question from a database and using the retrieved information to generate an answer.

[1219] (Claim 3)

[1220] 10. The system of claim 1, wherein the generated answers are based on specifications and source code of an internal company system.

[1221] "Example 1"

[1222] (Claim 1)

[1223] an information storage means for storing internal non-public information of a company;

[1224] a question receiving means for receiving a question from a user;

[1225] an answer generation means for generating an answer to a question based on the information storage means and the public information;

[1226] an information integration means for combining the private information and the public information in the information storage means to generate an answer to the question;

[1227] A system including an answer providing means for providing a generated answer to a user.

[1228] (Claim 2)

[1229] 2. The system according to claim 1, further comprising information acquisition means for acquiring non-public information related to the question from the information storage means and generating an answer using the acquired information.

[1230] (Claim 3)

[1231] 10. The system of claim 1, wherein the generated answers are based on the specifications and program code of the company's internal systems.

[1232] "Application Example 1"

[1233] (Claim 1)

[1234] a data storage means for storing internal non-public information of the company;

[1235] a communication device means for receiving a question from a user;

[1236] a processing unit for generating answers to questions based on the data storage unit and public information;

[1237] means for exploring and generating answers to questions based on a generative AI model by combining private and public information in said data storage means;

[1238] an interface means for providing the generated answers to a user;

[1239] a means for obtaining security-related information from public sources and combining it with internal information to generate a response;

[1240] A system that includes a means to ensure that the answers generated are based on security guidelines, policies, and the latest security trends.

[1241] (Claim 2)

[1242] 10. The system of claim 1, further comprising means for retrieving non-public information related to the question from the data storage means and using the retrieved information to generate the answer.

[1243] (Claim 3)

[1244] 10. The system of claim 1, wherein the generated answers are based on an enterprise's internal operational guidelines and code base.

[1245] "Example 2: Combining Emotion Engines"

[1246] (Claim 1)

[1247] A database that stores internal, non-public information of a company;

[1248] a terminal for receiving a question from a user;

[1249] a server that generates answers to questions based on the database and public information;

[1250] an information retrieval augmentation and generative model that combines private and public information in the database to generate answers to questions;

[1251] an emotion engine that recognizes the user's emotional state and adjusts the content of the response;

[1252] A system with a means for providing generated answers to a user.

[1253] (Claim 2)

[1254] 10. The system of claim 1, further comprising means for retrieving non-public information related to the question from a database and using the retrieved information to generate an answer.

[1255] (Claim 3)

[1256] 10. The system of claim 1, wherein the generated answers are based on specifications and source code of an internal company system.

[1257] (Claim 4)

[1258] 10. The system of claim 1, wherein the server further comprises means for analyzing the user's question and retrieving relevant information using natural language processing techniques.

[1259] (Claim 5)

[1260] 2. The system according to claim 1, further comprising means for the emotion engine to recognize an emotional state from the user's input content and biometric information, and to adjust the response in accordance with the recognized emotional state.

[1261] "Application example 2 when combining emotion engines"

[1262] (Claim 1)

[1263] an information storage device for storing private information within a company;

[1264] an input device for receiving a question from a user;

[1265] an information processing device that generates an answer to a question based on the information storage device and public information;

[1266] an information retrieval generation model that combines private information and public information in the information storage device to generate answers to questions;

[1267] a display device for providing the generated answers to the user; and

[1268] A system equipped with an emotion recognition device that identifies the user's emotional state and adjusts the content of the response.

[1269] (Claim 2)

[1270] 10. The system of claim 1, further comprising means for retrieving non-public information related to the question from an information store and using the retrieved information to generate an answer.

[1271] (Claim 3)

[1272] 10. The system of claim 1, wherein the generated answers are based on an internal company information specification and information processing code. [Explanation of symbols]

[1273] 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 database that stores internal, non-public information of a company; a terminal for receiving a question from a user; a server that generates answers to questions based on the database and public information; a Retrieval Augmented Generation model that combines private and public information in the database to generate answers to questions; A system with a means for providing generated answers to a user.

2. 10. The system of claim 1, further comprising means for retrieving non-public information related to the question from a database and using the retrieved information to generate an answer.

3. 10. The system of claim 1, wherein the generated answers are based on specifications and source code of an internal company system.

Images