system

Abstract

We provide the system. [Solution] A means of automatically collecting and integrating data provided by each vendor, A means of inputting collected data into a generative artificial intelligence model, A means of receiving questions from users and posing questions to a generative artificial intelligence model to generate responses, A means of presenting the generated response to the user, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Currently, many companies are conducting business using data provided by multiple vendors, but it is difficult to integrate these data and efficiently obtain the latest information. In addition, the search accuracy in existing information sources such as in-house Wikis is low, and it is difficult to quickly obtain the necessary information. For this reason, there is a problem that a lot of time and labor are spent on information collection and analysis, resulting in a decrease in work efficiency. Furthermore, there is a shortage of systems that can propose optimal options based on data spanning multiple vendors. As a result, there is a problem that the work efficiency and decision-making speed at the site are impaired.

Means for Solving the Problems

[0005] This invention provides a system that automatically collects and integrates data provided by various vendors. This system includes means for inputting the collected data into a generative artificial intelligence model, receiving questions from a user, and posing questions to the generative artificial intelligence model to generate responses. Furthermore, by including means for presenting the generated responses to the user, the system provides necessary information quickly and accurately. The invention also enhances data integrity and quality by tokenizing and normalizing the collected data before inputting it into the generative artificial intelligence model. Additionally, by including means for formatting the generated responses and presenting them in a user-friendly format, the system improves the readability of the information. In this way, users can quickly obtain the latest data across multiple vendors and make optimal decisions.

[0006] A "vendor" refers to a company or organization that provides a specific service or product.

[0007] "Data" refers to information that a system collects, processes, and analyzes.

[0008] "Collection" refers to the process of obtaining data from different sources.

[0009] "Integration" refers to the process of centrally managing multiple collected data sets.

[0010] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to generate appropriate responses to questions from large amounts of data.

[0011] "Tokenization" refers to the process of dividing text data into its smallest units, such as words or phrases.

[0012] "Normalization" refers to the process of standardizing and organizing the format and content of data.

[0013] A "question" refers to the text that a user enters to request information from the system.

[0014] "Response" refers to the information that a generative artificial intelligence model provides in response to a question.

[0015] A "user" refers to a person who operates a system and enters questions to obtain information.

[0016] A "terminal" refers to a device that a user uses to operate a system. [Brief explanation of the drawing]

[0017] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0018] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

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

[0020] In the following embodiments, a numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0021] In the following embodiments, a numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0022] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0023] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0025] [First Embodiment]

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

[0027] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0030] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0034] As shown in Figure 2, in the data processing device 12, specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0037] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0038] This invention is a system that automatically collects and integrates data provided by various vendors. The system inputs the collected data into a generative artificial intelligence model, receives questions from the user, and generates responses based on these questions. The generated responses are formatted and presented in a user-friendly format.

[0039] Data collection and integration

[0040] The server automatically collects system information, specifications, and parameter information from vendor APIs and FTP servers via the corporate network. For example, a scheduled job runs at midnight every day to retrieve the latest data and save it to the database. This data is also extracted from the company's internal wiki using an HTML parser and similarly stored in a centralized database. This data collection process ensures that the information is always up-to-date.

[0041] Data preprocessing and input to generative AI.

[0042] The collected data is tokenized and normalized by the server. For example, text data is split into words and phrases, normalized, and converted into a unified format. This pre-processed data is then input into a generative artificial intelligence model (e.g., GPT-4®). The server uses this AI model to generate the best possible response to the user's question based on the collected data.

[0043] User question reception and response generation

[0044] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for the base station and the best vendor." The terminal sends this input to a server, which analyzes the question using natural language processing. The analyzed question is then sent to a generative artificial intelligence model, which generates an appropriate response to the question. For example, a response like, "The latest parameter information is XX, and the best vendor is YY" might be generated.

[0045] Formatting and providing responses

[0046] The generated response is formatted by the server. The text formatting and grammar are corrected and converted into a user-friendly format. The formatted response is sent to the terminal and displayed to the user. For example, the information might be displayed on the screen in the format, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0047] Specific example

[0048] 1. The user enters "Please tell me the latest base station parameter information and best vendor" on their device.

[0049] 2. The terminal receives the input and sends the question to the server.

[0050] 3. The server analyzes the question using natural language processing and feeds it to a generative artificial intelligence model.

[0051] 4. The generative AI generates a response saying, "The latest parameter information is XX, and the optimal vendor is YY. The reason is that it has the highest performance."

[0052] 5. The server formats the generated response and sends it to the terminal.

[0053] 6. The user confirms the formatted response on their device screen.

[0054] This system allows users to quickly and accurately obtain the information they need, which is expected to improve work efficiency.

[0055] The following describes the processing flow.

[0056] Step 1:

[0057] The server automatically collects system information, specifications, and parameter information from APIs and FTP servers provided by various vendors via the corporate network. A scheduled job runs at midnight every day, retrieving this data and storing it in temporary storage.

[0058] Step 2:

[0059] The server uses an HTML parser to scrape and extract base station-related information from the company's internal wiki. This information is also stored in temporary storage.

[0060] Step 3:

[0061] The server integrates the collected data into a centralized management database. Normalization is performed to standardize the data format and eliminate redundant information. It also cross-references the data with existing database contents, replacing outdated information with the latest information.

[0062] Step 4:

[0063] The server performs tokenization and normalization of the integrated data for input into a generative artificial intelligence model. Text data is divided into words and phrases, normalized, and converted into a consistent format.

[0064] Step 5:

[0065] The server uses the AI ​​toolkit (API) interface to provide data to the generative artificial intelligence model. Requests to the generative AI are sent through the API client.

[0066] Step 6:

[0067] The user enters questions into the system via a terminal. For example, they might enter a question such as, "Please tell me the latest parameter information for base stations and the best vendor."

[0068] Step 7:

[0069] The terminal receives input and sends a question to the server. The server tokenizes this question and performs semantic analysis to understand the intent of the question.

[0070] Step 8:

[0071] The server sends the analyzed question to a generative artificial intelligence model. The generative AI model generates an appropriate response based on the data. For example, it might generate a response such as, "The latest parameter information is XX, and the best vendor is YY."

[0072] Step 9:

[0073] The server formats the generated response. The generated text is corrected to have the appropriate grammar and formatting.

[0074] Step 10:

[0075] The terminal displays a formatted response to the user. For example, the information might be presented on the screen in the format: "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0076] This entire process allows users to quickly and accurately obtain the information they need, resulting in improved work efficiency.

[0077] (Example 1)

[0078] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0079] In today's business environment, there is a need to quickly and accurately integrate various information provided by multiple suppliers and effectively respond to user inquiries. However, performing this manually is time-consuming and labor-intensive, and can lead to inaccuracies in information integration and response generation. Furthermore, there is a lack of integrated systems that consistently automate information collection, preprocessing, analysis, and response generation and formatting.

[0080] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0081] In this invention, the server includes means for automatically collecting and integrating information provided by each supplier, means for inputting the collected information into a generative artificial intelligence model, and means for receiving inquiries from users and posing inquiries to the generative artificial intelligence model to generate responses. This includes means for dividing the collected information into tokens, normalizing them and inputting them into the generative artificial intelligence model, means for formatting the generated responses and presenting them in a user-friendly format, means for using APIs, FTP servers, and HTML scraping when acquiring information, means for analyzing inquiries and inputting the analyzed inquiries as prompts into the generative artificial intelligence model, and means for using regular expressions when formatting the responses from the generative artificial intelligence model. This enables the entire process from information collection to presenting responses to users to be consistently automated, allowing for quick and accurate responses to user inquiries.

[0082] A "supplier" refers to a company or organization that provides specific goods or services.

[0083] "Information" includes data such as system information, specifications, and parameter information, and is provided by each supplier.

[0084] A "generative artificial intelligence model" is an artificial intelligence algorithm that generates natural language responses based on collected information and user inquiries.

[0085] "Tokenizing" refers to the process of breaking down collected text information into words and phrases.

[0086] "Normalization" refers to preprocessing that converts collected information into a unified format to make it easier to process.

[0087] A "prompt statement" is a sentence containing a question or instruction to be input into a generative artificial intelligence model.

[0088] "API" stands for Application Programming Interface, and it is a means of exchanging information with other software applications.

[0089] An "FTP server" is a server that executes a protocol for transferring files.

[0090] "HTML scraping" refers to a technique for automatically extracting necessary information from web pages.

[0091] "Natural language processing" refers to a set of techniques and algorithms that enable computers to understand, analyze, and generate human language.

[0092] A "regular expression" is a method of representing string patterns to perform string manipulation and searching efficiently.

[0093] An "inquiry" refers to a question or request for information that a user enters into a system.

[0094] This invention is a system that automatically collects and integrates information provided by various suppliers and generates responses to user inquiries using a generative artificial intelligence model. This system is implemented using the following hardware and software.

[0095] The server automatically collects information from supplier APIs and FTP servers via the corporate network. For example, it uses the Python requests library to access APIs and retrieve information. From FTP servers, it uses the Python ftplib library to download necessary files. BeautifulSoup is used to extract data from the company wiki using HTML scraping. This allows the server to store information in a centralized database. For example, a scheduled job runs at midnight every day to maintain the latest data.

[0096] The collected information is split into tokens and normalized by the server. Specifically, the NLTK library in Python is used to split the text data into words and phrases, convert all text to lowercase, and remove unnecessary symbols and spaces. Stemming and lemmatization are also applied to unify the words into their base forms. The information, after this preprocessing is complete, is then input into a generative artificial intelligence model (e.g., GPT-4).

[0097] The user enters a query into the system via a terminal. For example, they might enter a query such as, "Please tell me the latest parameter information and best vendor for base stations," and click the send button. The terminal sends this input to the server, which analyzes the query using a natural language processing library (e.g., spaCy). The analyzed query is then sent as a prompt to a generative artificial intelligence model, which generates the optimal response.

[0098] The generated response is formatted by the server. For example, Python regular expressions are used to correct the text formatting and grammar, transforming it into a user-friendly format. The formatted response is sent to the terminal and presented to the user. Specifically, a web browser might display something like, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0099] Specific example

[0100] For example, a user might type "Tell me the latest parameter information and best vendor for the base station" into their device. The device receives this input and sends a query to the server. The server analyzes the question using natural language processing and feeds it to a generative artificial intelligence model. The generative AI model generates a response such as "The latest parameter information is XX, and the best vendor is YY. The reason is that it has the highest performance." The server formats the generated response and sends it to the device. The user then checks the formatted response on their device screen.

[0101] Examples of prompt statements

[0102] "Please provide the latest parameter information for base stations and the best vendor."

[0103] "Please provide the system information for the server you are currently using."

[0104] "Please provide the latest specifications from supplier XX."

[0105] This invention allows users to quickly and accurately obtain the information they need, which is expected to improve work efficiency.

[0106] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0107] Step 1: Gathering information from suppliers

[0108] The server uses the corporate network to collect information from supplier APIs and FTP servers. Specifically, it uses the Python requests library to send HTTP GET requests and retrieve system information, specifications, and parameter information from the API. For FTP servers, it uses the Python ftplib library to connect and download the latest CSV file. It uses API endpoints and FTP server connection information as input and obtains the retrieved JSON data and CSV files as output.

[0109] Step 2: Extracting data from the company wiki

[0110] The server extracts information from the company's internal wiki using HTML scraping. It uses the Python BeautifulSoup library to send HTTP GET requests to specific page URLs and parse the HTML. The internal wiki URL is used as input, and the necessary text information is extracted as output.

[0111] Step 3: Tokenizing and normalizing the data

[0112] The collected information is tokenized and normalized by the server. Specifically, the NLTK library in Python is used to split the text data into words and phrases, convert all text to lowercase, and remove unnecessary symbols and spaces. Furthermore, stemming and lemmatization are applied to unify the words into their base forms. The collected text data is used as input, and tokenized and normalized text data is obtained as output.

[0113] Step 4: Receiving user inquiries

[0114] The user inputs a query into the system via the terminal. They enter a specific inquiry, such as "Please tell me the latest base station parameter information and the best vendor," and click the send button. The user's query text is used as input, and the output is a transmission request from the terminal to the server.

[0115] Step 5: Analyze the inquiry

[0116] The terminal sends the received query to the server, which then parses the query using a natural language processing library (e.g., spaCy). The user's query text is used as input, and the parsed query content is obtained as output.

[0117] Step 6: Sending prompts to the generative AI model

[0118] The server sends the analyzed query as a prompt to a generative artificial intelligence model. Specifically, it inputs the prompt to a generative AI model such as GPT-4, which then generates the optimal response. The analyzed query is used as input, and the generated response is obtained as output.

[0119] Step 7: Formatting the response

[0120] The generated response is formatted by the server. Python regular expressions are used to correct the text formatting and grammar, transforming it into a user-friendly format. The response text from a generative AI model is used as input, and the formatted response is obtained as output.

[0121] Step 8: Presenting a response to the user

[0122] The formatted response is sent from the server to the terminal, which then displays it to the user. Specifically, the web browser will display "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." The formatted response text is used as input, and the output is text in a format that the user can view on the screen.

[0123] (Application Example 1)

[0124] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0125] In large-scale production facilities such as factories, effectively managing operational data and maintenance information for each piece of equipment and machinery, and providing it to on-site workers quickly and accurately, is a critical challenge. However, the data provided by various suppliers is diverse, and centrally collecting and integrating this data to provide appropriate answers to workers' questions is not easy. In such a situation, on-site operational efficiency may decrease, and the risk of maintenance delays and equipment failures may increase. To solve this, a system is needed that effectively manages data and provides workers with the necessary information in a timely manner.

[0126] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0127] In this invention, the server includes means for automatically collecting and integrating data provided by each supplier, means for inputting the collected data into a generative artificial intelligence model, means for receiving questions from users and posing questions to the generative artificial intelligence model to generate responses, means for presenting the generated responses to the user, means for periodically collecting operational data and maintenance information of each device and equipment in the factory and storing it in an integrated database, and means for inputting questions into the system through a terminal used by the user and generating appropriate responses based on those questions. As a result, factory workers can quickly obtain the latest operational status and maintenance information of each device and equipment and take necessary actions quickly.

[0128] "Each supplier" refers to a company or organization that provides the equipment, parts, or data necessary for a factory or industry.

[0129] "Means of automated data collection and integration" refers to devices or software that perform the process of acquiring data provided by suppliers on a regular or real-time basis and storing it in an integrated database for centralized management.

[0130] A "generative artificial intelligence model" refers to an artificial intelligence model used for natural language generation and response generation, specifically employing advanced language models such as GPT (Generative Pre-trained Transducer).

[0131] "A means of receiving questions from users and posing those questions to a generative artificial intelligence model to generate responses" refers to a device or software that receives questions input by users, appropriately analyzes those questions, inputs them into a generative artificial intelligence model, and executes the process of generating appropriate responses.

[0132] "Means of presenting the generated response to the user" refers to devices or software that visually or audibly display the response generated by the artificial intelligence model on the user's device.

[0133] "Operational data" refers to information such as the operating status, performance data, and operation history of each piece of equipment and machinery in a production facility.

[0134] "Maintenance information" refers to inspection records, repair history, and information on scheduled maintenance work for each device and piece of equipment.

[0135] "Means of storing data in an integrated database" refers to a database system that stores and centrally manages diverse collected data, making it available for organizational use.

[0136] "User-used devices" refer to devices used by factory workers to input questions and receive responses, and specifically include smart glasses and head-mounted displays.

[0137] The embodiments for carrying out this invention are shown below.

[0138] System Overview

[0139] This system periodically collects operational data and maintenance information from various devices and equipment within the factory and stores it in an integrated database. It then uses a generative artificial intelligence model to provide optimal responses to user inquiries. The system primarily consists of a "server," a "terminal," and a "generative artificial intelligence model."

[0140] Data collection and integration

[0141] The server automatically collects and centrally integrates data provided by each supplier. Specifically, it periodically retrieves data from API and FTP servers and stores it in an integrated database. Operational data and maintenance information are also collected and integrated at this time.

[0142] Data preprocessing and input to generative AI.

[0143] The collected data is tokenized and normalized by the server. For example, to input into generative artificial intelligence models such as GPT-4, the collected data is divided into words and phrases and converted into a unified format.

[0144] User question reception and response generation

[0145] The user inputs a question into the system using a device (such as smart glasses or a head-mounted display). For example, they might input, "Please tell me the latest maintenance information and recommended actions." The device sends the input to the server, which uses a generative artificial intelligence model to generate an appropriate response.

[0146] Formatting and providing responses

[0147] The generated response is formatted by the server and presented to the user in an easy-to-understand format. For example, it might be presented as, "Latest maintenance information is XX, recommended action is YY, reason is ZZ." The formatted response is sent to the terminal, where the user can confirm it visually or audibly.

[0148] Specific example

[0149] A factory worker uses smart glasses to ask, "What is the latest maintenance information and recommended action?" This question is sent to a server, and a generative artificial intelligence model (such as GPT-4) generates an appropriate response based on the collected and integrated data. For example, it might generate "The latest maintenance information is XX, and the recommended action is YY," which is then displayed on the user's smart glasses.

[0150] Example of a prompt

[0151] "Please provide the latest maintenance information and recommended actions."

[0152] This configuration allows factory workers to quickly obtain the latest operational status and maintenance information for each piece of equipment and machinery, and to take necessary actions promptly.

[0153] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0154] Step 1:

[0155] The server collects data provided by each supplier. This involves using APIs and FTP servers to retrieve operational data and maintenance information. The collected data is stored in an integrated database for centralized management. The input is the data provided by each supplier, and the output is the data stored in the integrated database.

[0156] Step 2:

[0157] The server tokenizes and normalizes the collected data. This involves splitting text data into words and phrases and converting it into a unified format. Specifically, it formats the data so that it can be input into a generative artificial intelligence model (e.g., GPT-4). The input is the collected raw data, and the output is the tokenized and normalized data.

[0158] Step 3:

[0159] The user enters a question using a terminal. For example, they might enter a natural language question such as, "Please tell me the latest maintenance information and recommended actions." The input is the user's question, and the output is the request data sent from the terminal to the server.

[0160] Step 4:

[0161] The terminal receives the user's question and sends it to the server. In this step, the user's question is properly communicated to the server. The input is the user's question, and the output is the request data sent to the server.

[0162] Step 5:

[0163] The server processes user questions and performs analysis to prepare them for input into a generative artificial intelligence model. Specifically, it uses natural language processing to analyze the questions and converts them into a format suitable for the generative AI model. The input is the user's question data, and the output is the analyzed question that is input into the generative AI model.

[0164] Step 6:

[0165] The server uses a generative artificial intelligence model to generate responses to user questions. For example, it generates answers based on input questions using GPT-4. The input is parsed question data, and the output is the generated response text.

[0166] Step 7:

[0167] The server formats the generated response and converts it into a user-friendly format. This process involves grammar checking and text formatting adjustments. The input is the generated response text, and the output is the formatted response text.

[0168] Step 8:

[0169] The server sends the formatted response to the terminal. In this step, the formatted response is transmitted to the user's terminal. The input is the formatted response text, and the output is the response data sent to the terminal.

[0170] Step 9:

[0171] The user views the response generated and formatted by a generative artificial intelligence model through their device. For example, it might be displayed on smart glasses or a head-mounted display. The input is the response data sent to the device, and the output is the response displayed to the user.

[0172] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0173] This invention combines an emotion engine with a system that automatically collects and integrates data provided by various vendors and responds to user questions based on a generative artificial intelligence model. The system tokenizes and normalizes the collected data and inputs it into the generative AI model. It also receives user questions, poses questions to the generative AI model to generate responses, formats those responses, and presents them to the user. Furthermore, it uses the emotion engine to recognize the user's emotions, adjusts the response based on those emotions, and visualizes and presents the emotions to the user as needed.

[0174] Data collection and integration

[0175] The server automatically collects system information, specifications, and parameter information from various vendors' APIs and FTP servers via the corporate network. This data is retrieved periodically and stored in a centralized database. Additionally, an HTML parser is used to scrape base station-related information from the company wiki, and this data is also integrated into the centralized database. Normalization is performed to standardize the data format and eliminate redundant information.

[0176] Data preprocessing and input to generative AI.

[0177] The collected data is tokenized and normalized by the server. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency. Once these preprocessing steps are complete, the data is input into a generative artificial intelligence model (e.g., GPT-4). The server provides the data to the generative AI using an API, and the model generates appropriate responses to the user's questions.

[0178] User question reception and response generation

[0179] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for the base station and the best vendor." The terminal sends this input to a server, which analyzes the question using natural language processing. The analyzed question is then sent to a generative artificial intelligence model, which generates an appropriate response to the question. For example, a response like, "The latest parameter information is XX, and the best vendor is YY" might be generated.

[0180] Adjusting responses using an emotional engine

[0181] The generated responses are analyzed by an emotion engine before being formatted by the server. The emotion engine recognizes emotions from the text contained in the user's question and adjusts the response of the generative artificial intelligence model. For example, if the user expresses anxiety, elements that provide reassurance will be added to the response.

[0182] Formatting and providing responses

[0183] The responses, refined by the emotion engine, are formatted by the server. The generated text is corrected to the appropriate grammar and format. The formatted responses are sent to the terminal and displayed to the user. For example, information might be presented on the screen in the form of "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." Furthermore, the user's emotional state is visualized, and feedback is provided based on this.

[0184] Specific example

[0185] 1. The user enters "Please tell me the latest base station parameter information and best vendor" on their device.

[0186] 2. The terminal receives the question and sends it to the server.

[0187] 3. The server analyzes the question using natural language processing and sends it to a generative artificial intelligence model.

[0188] 4. The generative AI generates a response saying, "The latest parameter information is XX, and the best vendor is YY. The reason is that it has the highest performance."

[0189] 5. The server passes the generated response to the emotion engine, which analyzes the user's emotions and adjusts the response accordingly.

[0190] 6. The server sends a formatted response to the terminal for the user to confirm.

[0191] Thus, by providing responses that take into account the user's emotions, the present invention enables the provision of more appropriate and satisfying information.

[0192] The following describes the processing flow.

[0193] Step 1:

[0194] The server automatically collects system information, specifications, and parameter information from various vendors' APIs and FTP servers via the corporate network. For example, a scheduled job runs at midnight every day to retrieve the latest data. The collected data is stored in temporary storage.

[0195] Step 2:

[0196] The server uses an HTML parser to scrape and extract base station-related information from the company's internal wiki. This information is also stored in temporary storage.

[0197] Step 3:

[0198] The server integrates the collected data into a centralized management database. It performs normalization to standardize the data format and eliminate duplicate and redundant information. It also cross-references the data with existing database contents, replacing outdated information with the latest information.

[0199] Step 4:

[0200] The server performs tokenization and normalization processes on the data. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency.

[0201] Step 5:

[0202] The server uses an API to input tokenized and normalized data into a generative artificial intelligence model. Data is sent to the generative AI model via the API client, and the model is trained based on this data.

[0203] Step 6:

[0204] The user enters questions into the system via a terminal. For example, they might enter a question such as, "Please tell me the latest parameter information for base stations and the best vendor."

[0205] Step 7:

[0206] The terminal receives the entered question and sends it to the server. The server uses natural language processing to analyze the question and extract the necessary information.

[0207] Step 8:

[0208] The server sends the analyzed question to a generative artificial intelligence model. The generative AI model generates an appropriate response based on the data. For example, it might generate a response such as, "The latest parameter information is XX, and the best vendor is YY."

[0209] Step 9:

[0210] The server sends the generated response to the emotion engine, which performs sentiment analysis based on the user's input. The emotion engine recognizes the user's emotions and adjusts the response accordingly. For example, if the user expresses anxiety, elements that provide reassurance will be added to the response.

[0211] Step 10:

[0212] The server formats the responses that have been refined by the emotion engine. It corrects the grammar and formatting of the generated text and prepares it for the final form.

[0213] Step 11:

[0214] The terminal displays formatted responses to the user. For example, information might be presented in the form of, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance," and the user's emotional state may also be visualized.

[0215] This entire process allows users to quickly and accurately obtain the information they need, leading to improved work efficiency. Furthermore, the introduction of an emotion engine makes responses more user-friendly.

[0216] (Example 2)

[0217] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0218] In modern information systems, it is crucial to appropriately collect and integrate data from a wide variety of vendors and to effectively utilize generative artificial intelligence models to respond to user inquiries. However, conventional systems have limitations in improving the user experience because they lack sufficient preprocessing, such as normalization and tokenization of collected data, and are unable to provide responses that take user emotions into consideration. Furthermore, there is a need to present the generated responses in an easily understandable format, and conventional methods also have room for improvement in this regard.

[0219] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting and integrating data provided by each information provider, means for tokenizing and normalizing the collected data and inputting it into a generative artificial intelligence model, means for receiving questions from the user and posing questions to the generative artificial intelligence model to generate responses, means for recognizing the user's emotions and adjusting the responses, and means for formatting the generated responses and presenting them in a format that is easy for the user to understand. This enables efficient integration of data from multiple information sources, high-quality response generation using a generative artificial intelligence model, and the provision of responses that take into account the user's emotions, thereby improving the user experience.

[0220] "Information provider" refers to an institution or organization that provides data or information.

[0221] A "generative artificial intelligence model" is a type of artificial intelligence that generates new data or responses based on given input data.

[0222] "Collection" refers to the process of gathering scattered data into one place.

[0223] "Integration" refers to the process of centralizing collected data and compiling it into a consistent format.

[0224] "Tokenization" refers to the process of dividing text data into smaller units such as words and phrases.

[0225] "Normalization" refers to the process of transforming data to make it consistent.

[0226] A "question" refers to a question that a user enters into the system.

[0227] "Response" refers to the answer that the system generates in response to a question.

[0228] "Emotion" refers to the user's psychological state in response to questions and answers.

[0229] "Adjustment" refers to the process of modifying the generated response based on the user's emotions.

[0230] "Formatting" refers to the process of arranging a response to conform to the appropriate grammar and format.

[0231] "Format" refers to the specific layout and structure used when displaying data or information.

[0232] This invention relates to a system that automatically collects and integrates data provided by various information providers and responds to user questions using a generative artificial intelligence model. Furthermore, this system has the feature of analyzing the user's emotions using an emotion engine and adjusting its response based on those emotions.

[0233] Data collection and integration

[0234] The server has the capability to automatically collect data from each information provider's API or FTP server. This data includes system information, specifications, and parameter information, and is retrieved periodically. The data is then stored in a centralized database. The server also uses an HTML parser to scrape base station-related information from the company's internal wiki and integrates this into the centralized database as well. This standardizes the format of data from each information provider and performs normalization to eliminate redundant information.

[0235] Data preprocessing and input to generative artificial intelligence models

[0236] The collected data undergoes a tokenization and normalization process by the server. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency. Once these preprocessing steps are complete, the data is input into a generative artificial intelligence model (e.g., GPT-4). The server provides the data to the generative AI using an API, and the model generates appropriate responses to the user's questions.

[0237] As a concrete example, the server calls the information provider's API at 1 AM every day to obtain the latest information and integrate it into a centralized management database. For instance, it obtains the "latest parameter information for base stations," normalizes it, and then inputs it into GPT-4.

[0238] User question reception and analysis

[0239] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for base stations and the best vendor." The terminal sends this input to the server. The server receives the question and analyzes it using natural language processing techniques. The analyzed question is sent to a generative artificial intelligence model, which generates a response to the question.

[0240] For example, if a user enters "Tell me the latest parameter information for the base station and the best vendor," the server will analyze this question and generate a response such as "The latest parameter information is XX, and the best vendor is YY."

[0241] Adjusting responses using an emotional engine

[0242] The generated response is passed by the server to the sentiment engine. The sentiment engine recognizes the emotions contained in the user's question and adjusts the generated response accordingly. For example, for a user who is feeling anxious, elements that provide reassurance will be added to the response.

[0243] For example, if a user asks a question that includes anxiety, such as, "I have a very important meeting, but I urgently need the base station parameter information. Please let me know the latest information as soon as possible," the server's emotion engine will recognize this as an emotion of anxiety and generate a response such as, "The latest parameter information is XX, and the best vendor is YY. Don't worry, this information will be useful for your meeting."

[0244] Formatting and providing responses

[0245] The responses, refined by the emotion engine, are formatted by the server. The generated text is then corrected to the appropriate grammar and format before being sent to the terminal and displayed to the user. For example, information might be presented on the screen in the form of "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." Furthermore, the user's emotional state is visualized, and feedback is provided based on this.

[0246] Examples of prompt statements

[0247] The following are examples of prompts to input into a generative AI model.

[0248] Input: Please provide the latest parameter information and best vendor for base stations.

[0249] Prompt message: "The user wants to know the latest base station parameter information and the best vendor. The parameter information is XX, and the best vendor is YY. The reason is that it offers the highest performance. Add any elements that would reassure the user."

[0250] Thus, the present invention provides an advanced information system that takes user emotions into consideration and provides appropriate responses.

[0251] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0252] Step 1:

[0253] The server collects data from each information provider's API or FTP server. It uses information from each information provider's API endpoint or FTP server as input. The acquired data (system information, specifications, parameter information, etc.) is stored in a centralized database as output. Specifically, the server periodically sends requests to the API or FTP server according to scheduled tasks, receives response data, and writes it to the database.

[0254] Step 2:

[0255] The server uses an HTML parser to scrape information from the company's internal wiki. It uses the URL of the page to be scraped as input. The extracted data is added to a centralized management database as output. Specifically, the server sends an HTTP request to retrieve the internal wiki page, uses the HTML parser to extract the necessary information, and saves it to the database.

[0256] Step 3:

[0257] The server tokenizes and normalizes the collected data. It uses data stored in a centralized database as input. As output, it converts the tokenized and normalized data into a format that can be input into a generative artificial intelligence model. Specifically, the server reads data from the database and processes it using a tokenizer and normalization library.

[0258] Step 4:

[0259] The user enters the question via their device. The question text entered by the user is used as input. The question text is sent to the server as output. Specifically, the user enters the question into the input form in their web browser and clicks the submit button.

[0260] Step 5:

[0261] The terminal receives the user's question and sends it to the server. It uses the user's question text as input and sends the question text to the server's API endpoint as output. Specifically, the terminal sends the entered question to the server via the API.

[0262] Step 6:

[0263] The server sends a question to a generative AI model for analysis and generation of a response. It uses the user's question text as input and retrieves the response text from the generative AI model as output. Specifically, the server converts the question into a prompt and sends it to the generative AI model's API, then receives the generated response.

[0264] Step 7:

[0265] The server analyzes and adjusts responses using an emotion engine. It uses response text obtained from a generative artificial intelligence model as input. It obtains response text adjusted based on emotion as output. Specifically, the server inputs the response text into an emotion analysis library and adjusts the response based on the analysis results.

[0266] Step 8:

[0267] The server formats the adjusted response and sends it to the terminal. It uses the response text adjusted by the sentiment engine as input. It sends the formatted response text to the terminal as output. Specifically, the server uses a text formatting library to format the response and sends it to the terminal.

[0268] Step 9:

[0269] The terminal displays a formatted response to the user. It uses the formatted response text sent from the server as input. It displays the response on the screen as output. Specifically, the terminal binds the received response text to an element in the web browser and displays it on the screen.

[0270] As a concrete example, a user might input "Tell me the latest parameter information for the base station and the best vendor" into their terminal. The server then goes through the processes of collection, integration, analysis, generation, adjustment, and formatting, and finally displays on the terminal "The latest parameter information is XX, and the best vendor is YY."

[0271] (Application Example 2)

[0272] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0273] In content distribution services, a challenge exists in providing users with the most suitable content based on their emotional state, as this makes it difficult to make more appropriate recommendations. Traditional systems recommend content without considering user emotions, which may result in insufficient user satisfaction.

[0274] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting and integrating data provided by each supplier, means for inputting the collected data into a generative artificial intelligence model, means for receiving questions from the user and posing questions to the generative artificial intelligence model to generate responses, means for recognizing the user's emotions using an emotion analysis engine and adjusting the response based on those emotions, and means for formatting the adjusted response and presenting it to the user in an easy-to-understand format with added emotional information. This enables more personalized and appropriate content recommendations based on the user's emotional state.

[0275] "Each supplier" refers to companies, organizations, etc., that provide various services or products.

[0276] A "generative artificial intelligence model" refers to an algorithm or system that uses natural language processing to generate appropriate responses based on user input.

[0277] "Methods for tokenizing data" refer to algorithms and processes for dividing collected text data into individual words or phrases.

[0278] "Methods of normalizing data" refer to the process of converting data into a standard format in order to maintain consistency and integrity.

[0279] An "emotion analysis engine" refers to an algorithm or system that analyzes and recognizes emotions from a user's text input.

[0280] The "means for recognizing the user's emotions" refers to the process of identifying the emotional state from the user's input using an emotion analysis engine.

[0281] The "means for adjusting the response" refers to the process for modifying and optimizing the generated response based on the user's emotions.

[0282] The "means for shaping the response" refers to the process for making the generated response grammatically correct and converting it into a form that is easy for the user to understand.

[0283] The "means for adding and presenting emotion information" refers to the process or mechanism for adding and presenting feedback based on the user's emotional state.

[0284] The present invention describes a system for recommending appropriate content based on the user's emotional state. This system automatically collects data from each provider, integrates it, and inputs it into a generative artificial intelligence model. Furthermore, it combines an emotion analysis engine and has a function for adjusting responses based on the user's emotional state.

[0285] Hardware and software to be used

[0286] Hardware: [[ID=U=26]]

[0287] User terminal (smartphone)

[0288] Server

[0289] Software:

[0290] Emotion analysis engine (Hugging Face's transformers library)

[0291] Natural language processing and generative artificial intelligence model (OpenAI (registered trademark) GPT-3 (registered trademark))

[0292] For data collection, an HTTP request library (e.g., requests) is used.

[0293] For data processing, use Python.

[0294] Data collection and integration

[0295] The server automatically collects data provided by each supplier. The data is retrieved from the suppliers' APIs and FTP servers and integrated and stored in a centralized database. During this process, the collected data is tokenized and normalized.

[0296] Question reception and sentiment analysis

[0297] The user enters a question in natural language through their device. This question is sent to a server, where an emotion analysis engine analyzes the user's emotional state. The system then determines whether the user's emotional state is positive or negative.

[0298] Input and response generation for generative AI models

[0299] The server inputs tokenized and normalized data into a generative artificial intelligence model (e.g., GPT-3) to generate prompt sentences that take into account the user's emotional state. The generative AI model then generates an appropriate response based on these prompts.

[0300] Example of a prompt:

[0301] The user is in a negative state and asks the following question: "I've been feeling down lately, do you have any movie recommendations?" What would be the best content?

[0302] Coordinating and providing responses

[0303] The generated response is adjusted based on the user's sentiment by the sentiment analysis engine. For example, if the user is feeling down, content that can boost the mood is recommended. Finally, the formatted response is sent to the user terminal and presented to the user with added sentiment information.

[0304] Specific Example

[0305] When the user asks, "I've been feeling down lately. Are there any recommended movies?", this question is sent from the terminal to the server. The server performs sentiment analysis and recognizes a negative sentiment state. Then, the following prompt sentence is generated:

[0306] The user is in a NEGATIVE state and asks the following question: "I've been feeling down lately. Are there any recommended movies?". What is the optimal content?

[0307] Based on this prompt sentence, the generative artificial intelligence model recommends an appropriate movie and also explains the reason. For example, a response like "The recommended movie is 'The Best Way to Find a Wonderful Life'. The touching story will warm your heart" is generated. In this way, optimal content considering the user's sentiment is proposed to the user.

[0308] The above are the embodiments of the present invention. By using the present invention, more appropriate content recommendation based on the user's sentiment state can be realized.

[0309] The flow of the specific process in Application Example 2 will be described using FIG. 14.

[0310] Step 1:

[0311] The server automatically collects and integrates data provided by each supplier. Input data is obtained from the suppliers' APIs and FTP servers. This data is stored in a centralized database. Specifically, it periodically retrieves data using an HTTP request library and stores it in the server's database.

[0312] Step 2:

[0313] The server tokenizes and normalizes the collected data. The input is raw text data, and the output is data converted into a format suitable for generative artificial intelligence models. Specifically, it uses Python's natural language processing library to split the text into words and phrases and convert them into a standard format.

[0314] Step 3:

[0315] The user enters a question through their terminal. For example, they might enter, "I've been feeling down lately, do you have any movie recommendations?" This input is sent directly to the server. The input is the user's natural language question text.

[0316] Step 4:

[0317] The server analyzes the user's question using an emotion analysis engine. The input is the user's question text, and the output is the user's emotional state (positive or negative). Specifically, the Hugging Face transformers library is used to determine the emotion from the text.

[0318] Step 5:

[0319] The server generates prompts for a generative artificial intelligence model, taking the user's question and emotional state as input. The input is the user's question text and emotional state, and the output is a prompt. For example, it might generate a prompt such as, "The user is in a NEGATIVE state and asks the following question: 'I've been feeling down lately, do you have any movie recommendations?' What would be the best content?"

[0320] Step 6:

[0321] A generative artificial intelligence model (e.g., GPT-3) is input with a prompt sentence and generates a response. The input is the prompt sentence, and the output is the generated response text. Specifically, the server uses the OpenAI API to send the prompt sentence and receive optimal content recommendations.

[0322] Step 7:

[0323] The server adjusts the generated response based on its sentiment analysis engine. The input is the generated response text and the user's emotional state, and the output is the adjusted response text. For example, if the user is in a negative emotional state, the response will include comforting content.

[0324] Step 8:

[0325] The server formats the adjusted response and presents it in a user-friendly format. The input is the adjusted response text, and the output is the final text displayed to the user. Specifically, it performs grammar checks and formatting corrections to make it easy for the user to understand.

[0326] Step 9:

[0327] The refined response is sent to the terminal and presented to the user. The input is the formatted final text, and the output is a content recommendation displayed on the user's terminal screen. For example, a specific response such as "The recommended movie is 'The Bucket List.' It's a touching story that will warm your heart" might be displayed.

[0328] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0329] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0330] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0331] [Second Embodiment]

[0332] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0333] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0334] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0335] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0336] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0337] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0338] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0339] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0340] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0341] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0342] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0343] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0344] This invention is a system that automatically collects and integrates data provided by various vendors. The system inputs the collected data into a generative artificial intelligence model, receives questions from the user, and generates responses based on these questions. The generated responses are formatted and presented in a user-friendly format.

[0345] Data collection and integration

[0346] The server automatically collects system information, specifications, and parameter information from vendor APIs and FTP servers via the corporate network. For example, a scheduled job runs at midnight every day to retrieve the latest data and save it to the database. This data is also extracted from the company's internal wiki using an HTML parser and similarly stored in a centralized database. This data collection process ensures that the information is always up-to-date.

[0347] Data preprocessing and input to generative AI.

[0348] The collected data is tokenized and normalized by the server. For example, text data is split into words and phrases, normalized, and converted into a unified format. This pre-processed data is then input into a generative artificial intelligence model (e.g., GPT-4). The server uses this AI model to generate the best possible response to the user's question based on the collected data.

[0349] User question reception and response generation

[0350] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for the base station and the best vendor." The terminal sends this input to a server, which analyzes the question using natural language processing. The analyzed question is then sent to a generative artificial intelligence model, which generates an appropriate response to the question. For example, a response like, "The latest parameter information is XX, and the best vendor is YY" might be generated.

[0351] Formatting and providing responses

[0352] The generated response is formatted by the server. The text formatting and grammar are corrected and converted into a user-friendly format. The formatted response is sent to the terminal and displayed to the user. For example, the information might be displayed on the screen in the format, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0353] Specific example

[0354] 1. The user enters "Please tell me the latest base station parameter information and best vendor" on their device.

[0355] 2. The terminal receives the input and sends the question to the server.

[0356] 3. The server analyzes the question using natural language processing and feeds it to a generative artificial intelligence model.

[0357] 4. The generative AI generates a response saying, "The latest parameter information is XX, and the optimal vendor is YY. The reason is that it has the highest performance."

[0358] 5. The server formats the generated response and sends it to the terminal.

[0359] 6. The user confirms the formatted response on their device screen.

[0360] This system allows users to quickly and accurately obtain the information they need, which is expected to improve work efficiency.

[0361] The following describes the processing flow.

[0362] Step 1:

[0363] The server automatically collects system information, specifications, and parameter information from APIs and FTP servers provided by various vendors via the corporate network. A scheduled job runs at midnight every day, retrieving this data and storing it in temporary storage.

[0364] Step 2:

[0365] The server uses an HTML parser to scrape and extract base station-related information from the company's internal wiki. This information is also stored in temporary storage.

[0366] Step 3:

[0367] The server integrates the collected data into a centralized management database. Normalization is performed to standardize the data format and eliminate redundant information. It also cross-references the data with existing database contents, replacing outdated information with the latest information.

[0368] Step 4:

[0369] The server performs tokenization and normalization of the integrated data for input into a generative artificial intelligence model. Text data is divided into words and phrases, normalized, and converted into a consistent format.

[0370] Step 5:

[0371] The server uses the AI ​​toolkit (API) interface to provide data to the generative artificial intelligence model. Requests to the generative AI are sent through the API client.

[0372] Step 6:

[0373] The user enters questions into the system via a terminal. For example, they might enter a question such as, "Please tell me the latest parameter information for base stations and the best vendor."

[0374] Step 7:

[0375] The terminal receives input and sends a question to the server. The server tokenizes this question and performs semantic analysis to understand the intent of the question.

[0376] Step 8:

[0377] The server sends the analyzed question to a generative artificial intelligence model. The generative AI model generates an appropriate response based on the data. For example, it might generate a response such as, "The latest parameter information is XX, and the best vendor is YY."

[0378] Step 9:

[0379] The server formats the generated response. The generated text is corrected to have the appropriate grammar and formatting.

[0380] Step 10:

[0381] The terminal displays a formatted response to the user. For example, the information might be presented on the screen in the format: "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0382] This entire process allows users to quickly and accurately obtain the information they need, resulting in improved work efficiency.

[0383] (Example 1)

[0384] Next, we will describe Example 1. 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."

[0385] In today's business environment, there is a need to quickly and accurately integrate various information provided by multiple suppliers and effectively respond to user inquiries. However, performing this manually is time-consuming and labor-intensive, and can lead to inaccuracies in information integration and response generation. Furthermore, there is a lack of integrated systems that consistently automate information collection, preprocessing, analysis, and response generation and formatting.

[0386] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0387] In this invention, the server includes means for automatically collecting and integrating information provided by each supplier, means for inputting the collected information into a generative artificial intelligence model, and means for receiving inquiries from users and posing inquiries to the generative artificial intelligence model to generate responses. This includes means for dividing the collected information into tokens, normalizing them and inputting them into the generative artificial intelligence model, means for formatting the generated responses and presenting them in a user-friendly format, means for using APIs, FTP servers, and HTML scraping when acquiring information, means for analyzing inquiries and inputting the analyzed inquiries as prompts into the generative artificial intelligence model, and means for using regular expressions when formatting the responses from the generative artificial intelligence model. This enables the entire process from information collection to presenting responses to users to be consistently automated, allowing for quick and accurate responses to user inquiries.

[0388] A "supplier" refers to a company or organization that provides specific goods or services.

[0389] "Information" includes data such as system information, specifications, and parameter information, and is provided by each supplier.

[0390] A "generative artificial intelligence model" is an artificial intelligence algorithm that generates natural language responses based on collected information and user inquiries.

[0391] "Tokenizing" refers to the process of breaking down collected text information into words and phrases.

[0392] "Normalization" refers to preprocessing that converts collected information into a unified format to make it easier to process.

[0393] A "prompt statement" is a sentence containing a question or instruction to be input into a generative artificial intelligence model.

[0394] "API" stands for Application Programming Interface, and it is a means of exchanging information with other software applications.

[0395] An "FTP server" is a server that executes a protocol for transferring files.

[0396] "HTML scraping" refers to a technique for automatically extracting necessary information from web pages.

[0397] "Natural language processing" refers to a set of techniques and algorithms that enable computers to understand, analyze, and generate human language.

[0398] A "regular expression" is a method of representing string patterns to perform string manipulation and searching efficiently.

[0399] An "inquiry" refers to a question or request for information that a user enters into a system.

[0400] This invention is a system that automatically collects and integrates information provided by various suppliers and generates responses to user inquiries using a generative artificial intelligence model. This system is implemented using the following hardware and software.

[0401] The server automatically collects information from supplier APIs and FTP servers via the corporate network. For example, it uses the Python requests library to access APIs and retrieve information. From FTP servers, it uses the Python ftplib library to download necessary files. BeautifulSoup is used to extract data from the company wiki using HTML scraping. This allows the server to store information in a centralized database. For example, a scheduled job runs at midnight every day to maintain the latest data.

[0402] The collected information is split into tokens and normalized by the server. Specifically, the NLTK library in Python is used to split the text data into words and phrases, convert all text to lowercase, and remove unnecessary symbols and spaces. Stemming and lemmatization are also applied to unify the words into their base forms. The information, after this preprocessing is complete, is then input into a generative artificial intelligence model (e.g., GPT-4).

[0403] The user enters a query into the system via a terminal. For example, they might enter a query such as, "Please tell me the latest parameter information and best vendor for base stations," and click the send button. The terminal sends this input to the server, which analyzes the query using a natural language processing library (e.g., spaCy). The analyzed query is then sent as a prompt to a generative artificial intelligence model, which generates the optimal response.

[0404] The generated response is formatted by the server. For example, Python regular expressions are used to correct the text formatting and grammar, transforming it into a user-friendly format. The formatted response is sent to the terminal and presented to the user. Specifically, a web browser might display something like, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0405] Specific example

[0406] For example, a user might type "Tell me the latest parameter information and best vendor for the base station" into their device. The device receives this input and sends a query to the server. The server analyzes the question using natural language processing and feeds it to a generative artificial intelligence model. The generative AI model generates a response such as "The latest parameter information is XX, and the best vendor is YY. The reason is that it has the highest performance." The server formats the generated response and sends it to the device. The user then checks the formatted response on their device screen.

[0407] Examples of prompt statements

[0408] "Please provide the latest parameter information for base stations and the best vendor."

[0409] "Please provide the system information for the server you are currently using."

[0410] "Please provide the latest specifications from supplier XX."

[0411] This invention allows users to quickly and accurately obtain the information they need, which is expected to improve work efficiency.

[0412] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0413] Step 1: Gathering information from suppliers

[0414] The server uses the corporate network to collect information from supplier APIs and FTP servers. Specifically, it uses the Python requests library to send HTTP GET requests and retrieve system information, specifications, and parameter information from the API. For FTP servers, it uses the Python ftplib library to connect and download the latest CSV file. It uses API endpoints and FTP server connection information as input and obtains the retrieved JSON data and CSV files as output.

[0415] Step 2: Extracting data from the company wiki

[0416] The server extracts information from the company's internal wiki using HTML scraping. It uses the Python BeautifulSoup library to send HTTP GET requests to specific page URLs and parse the HTML. The internal wiki URL is used as input, and the necessary text information is extracted as output.

[0417] Step 3: Tokenizing and normalizing the data

[0418] The collected information is tokenized and normalized by the server. Specifically, the NLTK library in Python is used to split the text data into words and phrases, convert all text to lowercase, and remove unnecessary symbols and spaces. Furthermore, stemming and lemmatization are applied to unify the words into their base forms. The collected text data is used as input, and tokenized and normalized text data is obtained as output.

[0419] Step 4: Receiving user inquiries

[0420] The user inputs a query into the system via the terminal. They enter a specific inquiry, such as "Please tell me the latest base station parameter information and the best vendor," and click the send button. The user's query text is used as input, and the output is a transmission request from the terminal to the server.

[0421] Step 5: Analyze the inquiry

[0422] The terminal sends the received query to the server, which then parses the query using a natural language processing library (e.g., spaCy). The user's query text is used as input, and the parsed query content is obtained as output.

[0423] Step 6: Sending prompts to the generative AI model

[0424] The server sends the analyzed query as a prompt to a generative artificial intelligence model. Specifically, it inputs the prompt to a generative AI model such as GPT-4, which then generates the optimal response. The analyzed query is used as input, and the generated response is obtained as output.

[0425] Step 7: Formatting the response

[0426] The generated response is formatted by the server. Python regular expressions are used to correct the text formatting and grammar, transforming it into a user-friendly format. The response text from a generative AI model is used as input, and the formatted response is obtained as output.

[0427] Step 8: Presenting a response to the user

[0428] The formatted response is sent from the server to the terminal, which then displays it to the user. Specifically, the web browser will display "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." The formatted response text is used as input, and the output is text in a format that the user can view on the screen.

[0429] (Application Example 1)

[0430] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0431] In large-scale production facilities such as factories, effectively managing operational data and maintenance information for each piece of equipment and machinery, and providing it to on-site workers quickly and accurately, is a critical challenge. However, the data provided by various suppliers is diverse, and centrally collecting and integrating this data to provide appropriate answers to workers' questions is not easy. In such a situation, on-site operational efficiency may decrease, and the risk of maintenance delays and equipment failures may increase. To solve this, a system is needed that effectively manages data and provides workers with the necessary information in a timely manner.

[0432] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0433] In this invention, the server includes means for automatically collecting and integrating data provided by each supplier, means for inputting the collected data into a generative artificial intelligence model, means for receiving questions from users and posing questions to the generative artificial intelligence model to generate responses, means for presenting the generated responses to the user, means for periodically collecting operational data and maintenance information of each device and equipment in the factory and storing it in an integrated database, and means for inputting questions into the system through a terminal used by the user and generating appropriate responses based on those questions. As a result, factory workers can quickly obtain the latest operational status and maintenance information of each device and equipment and take necessary actions quickly.

[0434] "Each supplier" refers to a company or organization that provides the equipment, parts, or data necessary for a factory or industry.

[0435] "Means of automated data collection and integration" refers to devices or software that perform the process of acquiring data provided by suppliers on a regular or real-time basis and storing it in an integrated database for centralized management.

[0436] A "generative artificial intelligence model" refers to an artificial intelligence model used for natural language generation and response generation, specifically employing advanced language models such as GPT (Generative Pre-trained Transducer).

[0437] "A means of receiving questions from users and posing those questions to a generative artificial intelligence model to generate responses" refers to a device or software that receives questions input by users, appropriately analyzes those questions, inputs them into a generative artificial intelligence model, and executes the process of generating appropriate responses.

[0438] "Means of presenting the generated response to the user" refers to devices or software that visually or audibly display the response generated by the artificial intelligence model on the user's device.

[0439] "Operational data" refers to information such as the operating status, performance data, and operation history of each piece of equipment and machinery in a production facility.

[0440] "Maintenance information" refers to inspection records, repair history, and information on scheduled maintenance work for each device and piece of equipment.

[0441] "Means of storing data in an integrated database" refers to a database system that stores and centrally manages diverse collected data, making it available for organizational use.

[0442] "User-used devices" refer to devices used by factory workers to input questions and receive responses, and specifically include smart glasses and head-mounted displays.

[0443] The embodiments for carrying out this invention are shown below.

[0444] System Overview

[0445] This system periodically collects operational data and maintenance information from various devices and equipment within the factory and stores it in an integrated database. It then uses a generative artificial intelligence model to provide optimal responses to user inquiries. The system primarily consists of a "server," a "terminal," and a "generative artificial intelligence model."

[0446] Data collection and integration

[0447] The server automatically collects and centrally integrates data provided by each supplier. Specifically, it periodically retrieves data from API and FTP servers and stores it in an integrated database. Operational data and maintenance information are also collected and integrated at this time.

[0448] Data preprocessing and input to generative AI.

[0449] The collected data is tokenized and normalized by the server. For example, to input into generative artificial intelligence models such as GPT-4, the collected data is divided into words and phrases and converted into a unified format.

[0450] User question reception and response generation

[0451] The user inputs a question into the system using a device (such as smart glasses or a head-mounted display). For example, they might input, "Please tell me the latest maintenance information and recommended actions." The device sends the input to the server, which uses a generative artificial intelligence model to generate an appropriate response.

[0452] Formatting and providing responses

[0453] The generated response is formatted by the server and presented to the user in an easy-to-understand format. For example, it might be presented as, "Latest maintenance information is XX, recommended action is YY, reason is ZZ." The formatted response is sent to the terminal, where the user can confirm it visually or audibly.

[0454] Specific example

[0455] A factory worker uses smart glasses to ask, "What is the latest maintenance information and recommended action?" This question is sent to a server, and a generative artificial intelligence model (such as GPT-4) generates an appropriate response based on the collected and integrated data. For example, it might generate "The latest maintenance information is XX, and the recommended action is YY," which is then displayed on the user's smart glasses.

[0456] Example of a prompt

[0457] "Please provide the latest maintenance information and recommended actions."

[0458] This configuration allows factory workers to quickly obtain the latest operational status and maintenance information for each piece of equipment and machinery, and to take necessary actions promptly.

[0459] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0460] Step 1:

[0461] The server collects data provided by each supplier. This involves using APIs and FTP servers to retrieve operational data and maintenance information. The collected data is stored in an integrated database for centralized management. The input is the data provided by each supplier, and the output is the data stored in the integrated database.

[0462] Step 2:

[0463] The server tokenizes and normalizes the collected data. This involves splitting text data into words and phrases and converting it into a unified format. Specifically, it formats the data so that it can be input into a generative artificial intelligence model (e.g., GPT-4). The input is the collected raw data, and the output is the tokenized and normalized data.

[0464] Step 3:

[0465] The user enters a question using a terminal. For example, they might enter a natural language question such as, "Please tell me the latest maintenance information and recommended actions." The input is the user's question, and the output is the request data sent from the terminal to the server.

[0466] Step 4:

[0467] The terminal receives the user's question and sends it to the server. In this step, the user's question is properly communicated to the server. The input is the user's question, and the output is the request data sent to the server.

[0468] Step 5:

[0469] The server processes user questions and performs analysis to prepare them for input into a generative artificial intelligence model. Specifically, it uses natural language processing to analyze the questions and converts them into a format suitable for the generative AI model. The input is the user's question data, and the output is the analyzed question that is input into the generative AI model.

[0470] Step 6:

[0471] The server uses a generative artificial intelligence model to generate responses to user questions. For example, it generates answers based on input questions using GPT-4. The input is parsed question data, and the output is the generated response text.

[0472] Step 7:

[0473] The server formats the generated response and converts it into a user-friendly format. This process involves grammar checking and text formatting adjustments. The input is the generated response text, and the output is the formatted response text.

[0474] Step 8:

[0475] The server sends the formatted response to the terminal. In this step, the formatted response is transmitted to the user's terminal. The input is the formatted response text, and the output is the response data sent to the terminal.

[0476] Step 9:

[0477] The user views the response generated and formatted by a generative artificial intelligence model through their device. For example, it might be displayed on smart glasses or a head-mounted display. The input is the response data sent to the device, and the output is the response displayed to the user.

[0478] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0479] This invention combines an emotion engine with a system that automatically collects and integrates data provided by various vendors and responds to user questions based on a generative artificial intelligence model. The system tokenizes and normalizes the collected data and inputs it into the generative AI model. It also receives user questions, poses questions to the generative AI model to generate responses, formats those responses, and presents them to the user. Furthermore, it uses the emotion engine to recognize the user's emotions, adjusts the response based on those emotions, and visualizes and presents the emotions to the user as needed.

[0480] Data collection and integration

[0481] The server automatically collects system information, specifications, and parameter information from various vendors' APIs and FTP servers via the corporate network. This data is retrieved periodically and stored in a centralized database. Additionally, an HTML parser is used to scrape base station-related information from the company wiki, and this data is also integrated into the centralized database. Normalization is performed to standardize the data format and eliminate redundant information.

[0482] Data preprocessing and input to generative AI.

[0483] The collected data is tokenized and normalized by the server. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency. Once these preprocessing steps are complete, the data is input into a generative artificial intelligence model (e.g., GPT-4). The server provides the data to the generative AI using an API, and the model generates appropriate responses to the user's questions.

[0484] User question reception and response generation

[0485] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for the base station and the best vendor." The terminal sends this input to a server, which analyzes the question using natural language processing. The analyzed question is then sent to a generative artificial intelligence model, which generates an appropriate response to the question. For example, a response like, "The latest parameter information is XX, and the best vendor is YY" might be generated.

[0486] Adjusting responses using an emotional engine

[0487] The generated responses are analyzed by an emotion engine before being formatted by the server. The emotion engine recognizes emotions from the text contained in the user's question and adjusts the response of the generative artificial intelligence model. For example, if the user expresses anxiety, elements that provide reassurance will be added to the response.

[0488] Formatting and providing responses

[0489] The responses, refined by the emotion engine, are formatted by the server. The generated text is corrected to the appropriate grammar and format. The formatted responses are sent to the terminal and displayed to the user. For example, information might be presented on the screen in the form of "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." Furthermore, the user's emotional state is visualized, and feedback is provided based on this.

[0490] Specific example

[0491] 1. The user enters "Please tell me the latest base station parameter information and best vendor" on their device.

[0492] 2. The terminal receives the question and sends it to the server.

[0493] 3. The server analyzes the question using natural language processing and sends it to a generative artificial intelligence model.

[0494] 4. The generative AI generates a response saying, "The latest parameter information is XX, and the best vendor is YY. The reason is that it has the highest performance."

[0495] 5. The server passes the generated response to the emotion engine, which analyzes the user's emotions and adjusts the response accordingly.

[0496] 6. The server sends a formatted response to the terminal for the user to confirm.

[0497] Thus, by providing responses that take into account the user's emotions, the present invention enables the provision of more appropriate and satisfying information.

[0498] The following describes the processing flow.

[0499] Step 1:

[0500] The server automatically collects system information, specifications, and parameter information from various vendors' APIs and FTP servers via the corporate network. For example, a scheduled job runs at midnight every day to retrieve the latest data. The collected data is stored in temporary storage.

[0501] Step 2:

[0502] The server uses an HTML parser to scrape and extract base station-related information from the company's internal wiki. This information is also stored in temporary storage.

[0503] Step 3:

[0504] The server integrates the collected data into a centralized management database. It performs normalization to standardize the data format and eliminate duplicate and redundant information. It also cross-references the data with existing database contents, replacing outdated information with the latest information.

[0505] Step 4:

[0506] The server performs tokenization and normalization processes on the data. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency.

[0507] Step 5:

[0508] The server uses an API to input tokenized and normalized data into a generative artificial intelligence model. Data is sent to the generative AI model via the API client, and the model is trained based on this data.

[0509] Step 6:

[0510] The user enters questions into the system via a terminal. For example, they might enter a question such as, "Please tell me the latest parameter information for base stations and the best vendor."

[0511] Step 7:

[0512] The terminal receives the entered question and sends it to the server. The server uses natural language processing to analyze the question and extract the necessary information.

[0513] Step 8:

[0514] The server sends the analyzed question to a generative artificial intelligence model. The generative AI model generates an appropriate response based on the data. For example, it might generate a response such as, "The latest parameter information is XX, and the best vendor is YY."

[0515] Step 9:

[0516] The server sends the generated response to the emotion engine, which performs sentiment analysis based on the user's input. The emotion engine recognizes the user's emotions and adjusts the response accordingly. For example, if the user expresses anxiety, elements that provide reassurance will be added to the response.

[0517] Step 10:

[0518] The server formats the responses that have been refined by the emotion engine. It corrects the grammar and formatting of the generated text and prepares it for the final form.

[0519] Step 11:

[0520] The terminal displays formatted responses to the user. For example, information might be presented in the form of, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance," and the user's emotional state may also be visualized.

[0521] This entire process allows users to quickly and accurately obtain the information they need, leading to improved work efficiency. Furthermore, the introduction of an emotion engine makes responses more user-friendly.

[0522] (Example 2)

[0523] Next, we will describe Example 2. 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".

[0524] In modern information systems, it is crucial to appropriately collect and integrate data from a wide variety of vendors and to effectively utilize generative artificial intelligence models to respond to user inquiries. However, conventional systems have limitations in improving the user experience because they lack sufficient preprocessing, such as normalization and tokenization of collected data, and are unable to provide responses that take user emotions into consideration. Furthermore, there is a need to present the generated responses in an easily understandable format, and conventional methods also have room for improvement in this regard.

[0525] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting and integrating data provided by each information provider, means for tokenizing and normalizing the collected data and inputting it into a generative artificial intelligence model, means for receiving questions from the user and posing questions to the generative artificial intelligence model to generate responses, means for recognizing the user's emotions and adjusting the responses, and means for formatting the generated responses and presenting them in a format that is easy for the user to understand. This enables efficient integration of data from multiple information sources, high-quality response generation using a generative artificial intelligence model, and the provision of responses that take into account the user's emotions, thereby improving the user experience.

[0526] "Information provider" refers to an institution or organization that provides data or information.

[0527] A "generative artificial intelligence model" is a type of artificial intelligence that generates new data or responses based on given input data.

[0528] "Collection" refers to the process of gathering scattered data into one place.

[0529] "Integration" refers to the process of centralizing collected data and compiling it into a consistent format.

[0530] "Tokenization" refers to the process of dividing text data into smaller units such as words and phrases.

[0531] "Normalization" refers to the process of transforming data to make it consistent.

[0532] A "question" refers to a question that a user enters into the system.

[0533] "Response" refers to the answer that the system generates in response to a question.

[0534] "Emotion" refers to the user's psychological state in response to questions and answers.

[0535] "Adjustment" refers to the process of modifying the generated response based on the user's emotions.

[0536] "Formatting" refers to the process of arranging a response to conform to the appropriate grammar and format.

[0537] "Format" refers to the specific layout and structure used when displaying data or information.

[0538] This invention relates to a system that automatically collects and integrates data provided by various information providers and responds to user questions using a generative artificial intelligence model. Furthermore, this system has the feature of analyzing the user's emotions using an emotion engine and adjusting its response based on those emotions.

[0539] Data collection and integration

[0540] The server has the capability to automatically collect data from each information provider's API or FTP server. This data includes system information, specifications, and parameter information, and is retrieved periodically. The data is then stored in a centralized database. The server also uses an HTML parser to scrape base station-related information from the company's internal wiki and integrates this into the centralized database as well. This standardizes the format of data from each information provider and performs normalization to eliminate redundant information.

[0541] Data preprocessing and input to generative artificial intelligence models

[0542] The collected data undergoes a tokenization and normalization process by the server. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency. Once these preprocessing steps are complete, the data is input into a generative artificial intelligence model (e.g., GPT-4). The server provides the data to the generative AI using an API, and the model generates appropriate responses to the user's questions.

[0543] As a concrete example, the server calls the information provider's API at 1 AM every day to obtain the latest information and integrate it into a centralized management database. For instance, it obtains the "latest parameter information for base stations," normalizes it, and then inputs it into GPT-4.

[0544] User question reception and analysis

[0545] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for base stations and the best vendor." The terminal sends this input to the server. The server receives the question and analyzes it using natural language processing techniques. The analyzed question is sent to a generative artificial intelligence model, which generates a response to the question.

[0546] For example, if a user enters "Tell me the latest parameter information for the base station and the best vendor," the server will analyze this question and generate a response such as "The latest parameter information is XX, and the best vendor is YY."

[0547] Adjusting responses using an emotional engine

[0548] The generated response is passed by the server to the sentiment engine. The sentiment engine recognizes the emotions contained in the user's question and adjusts the generated response accordingly. For example, for a user who is feeling anxious, elements that provide reassurance will be added to the response.

[0549] For example, if a user asks a question that includes anxiety, such as, "I have a very important meeting, but I urgently need the base station parameter information. Please let me know the latest information as soon as possible," the server's emotion engine will recognize this as an emotion of anxiety and generate a response such as, "The latest parameter information is XX, and the best vendor is YY. Don't worry, this information will be useful for your meeting."

[0550] Formatting and providing responses

[0551] The responses, refined by the emotion engine, are formatted by the server. The generated text is then corrected to the appropriate grammar and format before being sent to the terminal and displayed to the user. For example, information might be presented on the screen in the form of "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." Furthermore, the user's emotional state is visualized, and feedback is provided based on this.

[0552] Examples of prompt statements

[0553] The following are examples of prompts to input into a generative AI model.

[0554] Input: Please provide the latest parameter information and best vendor for base stations.

[0555] Prompt message: "The user wants to know the latest base station parameter information and the best vendor. The parameter information is XX, and the best vendor is YY. The reason is that it offers the highest performance. Add any elements that would reassure the user."

[0556] Thus, the present invention provides an advanced information system that takes user emotions into consideration and provides appropriate responses.

[0557] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0558] Step 1:

[0559] The server collects data from each information provider's API or FTP server. It uses information from each information provider's API endpoint or FTP server as input. The acquired data (system information, specifications, parameter information, etc.) is stored in a centralized database as output. Specifically, the server periodically sends requests to the API or FTP server according to scheduled tasks, receives response data, and writes it to the database.

[0560] Step 2:

[0561] The server uses an HTML parser to scrape information from the company's internal wiki. It uses the URL of the page to be scraped as input. The extracted data is added to a centralized management database as output. Specifically, the server sends an HTTP request to retrieve the internal wiki page, uses the HTML parser to extract the necessary information, and saves it to the database.

[0562] Step 3:

[0563] The server tokenizes and normalizes the collected data. It uses data stored in a centralized database as input. As output, it converts the tokenized and normalized data into a format that can be input into a generative artificial intelligence model. Specifically, the server reads data from the database and processes it using a tokenizer and normalization library.

[0564] Step 4:

[0565] The user enters the question via their device. The question text entered by the user is used as input. The question text is sent to the server as output. Specifically, the user enters the question into the input form in their web browser and clicks the submit button.

[0566] Step 5:

[0567] The terminal receives the user's question and sends it to the server. It uses the user's question text as input and sends the question text to the server's API endpoint as output. Specifically, the terminal sends the entered question to the server via the API.

[0568] Step 6:

[0569] The server sends a question to a generative AI model for analysis and generation of a response. It uses the user's question text as input and retrieves the response text from the generative AI model as output. Specifically, the server converts the question into a prompt and sends it to the generative AI model's API, then receives the generated response.

[0570] Step 7:

[0571] The server analyzes and adjusts responses using an emotion engine. It uses response text obtained from a generative artificial intelligence model as input. It obtains response text adjusted based on emotion as output. Specifically, the server inputs the response text into an emotion analysis library and adjusts the response based on the analysis results.

[0572] Step 8:

[0573] The server formats the adjusted response and sends it to the terminal. It uses the response text adjusted by the sentiment engine as input. It sends the formatted response text to the terminal as output. Specifically, the server uses a text formatting library to format the response and sends it to the terminal.

[0574] Step 9:

[0575] The terminal displays a formatted response to the user. It uses the formatted response text sent from the server as input. It displays the response on the screen as output. Specifically, the terminal binds the received response text to an element in the web browser and displays it on the screen.

[0576] As a concrete example, a user might input "Tell me the latest parameter information for the base station and the best vendor" into their terminal. The server then goes through the processes of collection, integration, analysis, generation, adjustment, and formatting, and finally displays on the terminal "The latest parameter information is XX, and the best vendor is YY."

[0577] (Application Example 2)

[0578] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0579] In content distribution services, a challenge exists in providing users with the most suitable content based on their emotional state, as this makes it difficult to make more appropriate recommendations. Traditional systems recommend content without considering user emotions, which may result in insufficient user satisfaction.

[0580] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting and integrating data provided by each supplier, means for inputting the collected data into a generative artificial intelligence model, means for receiving questions from the user and posing questions to the generative artificial intelligence model to generate responses, means for recognizing the user's emotions using an emotion analysis engine and adjusting the response based on those emotions, and means for formatting the adjusted response and presenting it to the user in an easy-to-understand format with added emotional information. This enables more personalized and appropriate content recommendations based on the user's emotional state.

[0581] "Each supplier" refers to companies, organizations, etc., that provide various services or products.

[0582] A "generative artificial intelligence model" refers to an algorithm or system that uses natural language processing to generate appropriate responses based on user input.

[0583] "Methods for tokenizing data" refer to algorithms and processes for dividing collected text data into individual words or phrases.

[0584] "Methods of normalizing data" refer to the process of converting data into a standard format in order to maintain consistency and integrity.

[0585] An "emotion analysis engine" refers to an algorithm or system that analyzes and recognizes emotions from a user's text input.

[0586] "Means of recognizing user emotions" refers to the process of identifying a user's emotional state from their input using an emotion analysis engine.

[0587] "Means of adjusting responses" refers to the process of modifying and optimizing generated responses based on the user's emotions.

[0588] "Methods for formatting responses" refer to the process of making the generated response grammatically correct and converting it into a format that is easy for the user to understand.

[0589] "Means of presenting with added emotional information" refers to processes and mechanisms for presenting feedback based on the user's emotional state.

[0590] This invention describes a system that recommends appropriate content based on the user's emotional state. This system automatically collects data from various suppliers, integrates it, and inputs it into a generative artificial intelligence model. Furthermore, it incorporates an emotion analysis engine and has the function of adjusting the response based on the user's emotional state.

[0591] Hardware and software to be used

[0592] Hardware:

[0593] User device (smartphone)

[0594] server

[0595] software:

[0596] Emotion analysis engine (Hugging Face's transformers library)

[0597] Natural language processing and generative artificial intelligence models (OpenAI GPT-3)

[0598] For data collection, an HTTP request library (e.g., requests) is used.

[0599] For data processing, use Python.

[0600] Data collection and integration

[0601] The server automatically collects data provided by each supplier. The data is retrieved from the suppliers' APIs and FTP servers and integrated and stored in a centralized database. During this process, the collected data is tokenized and normalized.

[0602] Question reception and sentiment analysis

[0603] The user enters a question in natural language through their device. This question is sent to a server, where an emotion analysis engine analyzes the user's emotional state. The system then determines whether the user's emotional state is positive or negative.

[0604] Input and response generation for generative AI models

[0605] The server inputs tokenized and normalized data into a generative artificial intelligence model (e.g., GPT-3) to generate prompt sentences that take into account the user's emotional state. The generative AI model then generates an appropriate response based on these prompts.

[0606] Example of a prompt:

[0607] The user is in a negative state and asks the following question: "I've been feeling down lately, do you have any movie recommendations?" What would be the best content?

[0608] Coordinating and providing responses

[0609] The generated response is adjusted by an emotion analysis engine based on the user's emotions. For example, if the user is feeling down, it will recommend content that will lift their spirits. Finally, the refined response is sent to the user's device and presented to the user with added emotional information.

[0610] Specific example

[0611] When a user asks, "I've been feeling down lately, do you have any movie recommendations?", this question is sent from the device to the server. The server performs sentiment analysis and recognizes a negative emotional state. Then, a prompt message like the following is generated:

[0612] The user is in a negative state and asks the following question: "I've been feeling down lately, do you have any movie recommendations?" What would be the best content?

[0613] The generative artificial intelligence model uses this prompt to recommend an appropriate movie and explain the reason. For example, it might generate a response like, "My recommended movie is 'The Bucket List.' Its moving story will warm your heart." This ensures that the user is offered content that takes their emotions into consideration.

[0614] The above describes embodiments of the present invention. By using the present invention, it is possible to realize more appropriate content recommendations based on the user's emotional state.

[0615] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0616] Step 1:

[0617] The server automatically collects and integrates data provided by each supplier. Input data is obtained from the suppliers' APIs and FTP servers. This data is stored in a centralized database. Specifically, it periodically retrieves data using an HTTP request library and stores it in the server's database.

[0618] Step 2:

[0619] The server tokenizes and normalizes the collected data. The input is raw text data, and the output is data converted into a format suitable for generative artificial intelligence models. Specifically, it uses Python's natural language processing library to split the text into words and phrases and convert them into a standard format.

[0620] Step 3:

[0621] The user enters a question through their terminal. For example, they might enter, "I've been feeling down lately, do you have any movie recommendations?" This input is sent directly to the server. The input is the user's natural language question text.

[0622] Step 4:

[0623] The server analyzes the user's question using an emotion analysis engine. The input is the user's question text, and the output is the user's emotional state (positive or negative). Specifically, the Hugging Face transformers library is used to determine the emotion from the text.

[0624] Step 5:

[0625] The server generates prompts for a generative artificial intelligence model, taking the user's question and emotional state as input. The input is the user's question text and emotional state, and the output is a prompt. For example, it might generate a prompt such as, "The user is in a NEGATIVE state and asks the following question: 'I've been feeling down lately, do you have any movie recommendations?' What would be the best content?"

[0626] Step 6:

[0627] A generative artificial intelligence model (e.g., GPT-3) is input with a prompt sentence and generates a response. The input is the prompt sentence, and the output is the generated response text. Specifically, the server uses the OpenAI API to send the prompt sentence and receive optimal content recommendations.

[0628] Step 7:

[0629] The server adjusts the generated response based on its sentiment analysis engine. The input is the generated response text and the user's emotional state, and the output is the adjusted response text. For example, if the user is in a negative emotional state, the response will include comforting content.

[0630] Step 8:

[0631] The server formats the adjusted response and presents it in a user-friendly format. The input is the adjusted response text, and the output is the final text displayed to the user. Specifically, it performs grammar checks and formatting corrections to make it easy for the user to understand.

[0632] Step 9:

[0633] The refined response is sent to the terminal and presented to the user. The input is the formatted final text, and the output is a content recommendation displayed on the user's terminal screen. For example, a specific response such as "The recommended movie is 'The Bucket List.' It's a touching story that will warm your heart" might be displayed.

[0634] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0635] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0636] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0637] [Third Embodiment]

[0638] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0639] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0640] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0641] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0642] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0643] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0644] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0645] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0646] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0647] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0648] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0649] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0650] This invention is a system that automatically collects and integrates data provided by various vendors. The system inputs the collected data into a generative artificial intelligence model, receives questions from the user, and generates responses based on these questions. The generated responses are formatted and presented in a user-friendly format.

[0651] Data collection and integration

[0652] The server automatically collects system information, specifications, and parameter information from vendor APIs and FTP servers via the corporate network. For example, a scheduled job runs at midnight every day to retrieve the latest data and save it to the database. This data is also extracted from the company's internal wiki using an HTML parser and similarly stored in a centralized database. This data collection process ensures that the information is always up-to-date.

[0653] Data preprocessing and input to generative AI.

[0654] The collected data is tokenized and normalized by the server. For example, text data is split into words and phrases, normalized, and converted into a unified format. This pre-processed data is then input into a generative artificial intelligence model (e.g., GPT-4). The server uses this AI model to generate the best possible response to the user's question based on the collected data.

[0655] User question reception and response generation

[0656] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for the base station and the best vendor." The terminal sends this input to a server, which analyzes the question using natural language processing. The analyzed question is then sent to a generative artificial intelligence model, which generates an appropriate response to the question. For example, a response like, "The latest parameter information is XX, and the best vendor is YY" might be generated.

[0657] Formatting and providing responses

[0658] The generated response is formatted by the server. The text formatting and grammar are corrected and converted into a user-friendly format. The formatted response is sent to the terminal and displayed to the user. For example, the information might be displayed on the screen in the format, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0659] Specific example

[0660] 1. The user enters "Please tell me the latest base station parameter information and best vendor" on their device.

[0661] 2. The terminal receives the input and sends the question to the server.

[0662] 3. The server analyzes the question using natural language processing and feeds it to a generative artificial intelligence model.

[0663] 4. The generative AI generates a response saying, "The latest parameter information is XX, and the optimal vendor is YY. The reason is that it has the highest performance."

[0664] 5. The server formats the generated response and sends it to the terminal.

[0665] 6. The user confirms the formatted response on their device screen.

[0666] This system allows users to quickly and accurately obtain the information they need, which is expected to improve work efficiency.

[0667] The following describes the processing flow.

[0668] Step 1:

[0669] The server automatically collects system information, specifications, and parameter information from APIs and FTP servers provided by various vendors via the corporate network. A scheduled job runs at midnight every day, retrieving this data and storing it in temporary storage.

[0670] Step 2:

[0671] The server uses an HTML parser to scrape and extract base station-related information from the company's internal wiki. This information is also stored in temporary storage.

[0672] Step 3:

[0673] The server integrates the collected data into a centralized management database. Normalization is performed to standardize the data format and eliminate redundant information. It also cross-references the data with existing database contents, replacing outdated information with the latest information.

[0674] Step 4:

[0675] The server performs tokenization and normalization of the integrated data for input into a generative artificial intelligence model. Text data is divided into words and phrases, normalized, and converted into a consistent format.

[0676] Step 5:

[0677] The server uses the AI ​​toolkit (API) interface to provide data to the generative artificial intelligence model. Requests to the generative AI are sent through the API client.

[0678] Step 6:

[0679] The user enters questions into the system via a terminal. For example, they might enter a question such as, "Please tell me the latest parameter information for base stations and the best vendor."

[0680] Step 7:

[0681] The terminal receives input and sends a question to the server. The server tokenizes this question and performs semantic analysis to understand the intent of the question.

[0682] Step 8:

[0683] The server sends the analyzed question to a generative artificial intelligence model. The generative AI model generates an appropriate response based on the data. For example, it might generate a response such as, "The latest parameter information is XX, and the best vendor is YY."

[0684] Step 9:

[0685] The server formats the generated response. The generated text is corrected to have the appropriate grammar and formatting.

[0686] Step 10:

[0687] The terminal displays a formatted response to the user. For example, the information might be presented on the screen in the format: "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0688] This entire process allows users to quickly and accurately obtain the information they need, resulting in improved work efficiency.

[0689] (Example 1)

[0690] Next, we will describe Example 1. 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."

[0691] In today's business environment, there is a need to quickly and accurately integrate various information provided by multiple suppliers and effectively respond to user inquiries. However, performing this manually is time-consuming and labor-intensive, and can lead to inaccuracies in information integration and response generation. Furthermore, there is a lack of integrated systems that consistently automate information collection, preprocessing, analysis, and response generation and formatting.

[0692] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0693] In this invention, the server includes means for automatically collecting and integrating information provided by each supplier, means for inputting the collected information into a generative artificial intelligence model, and means for receiving inquiries from users and posing inquiries to the generative artificial intelligence model to generate responses. This includes means for dividing the collected information into tokens, normalizing them and inputting them into the generative artificial intelligence model, means for formatting the generated responses and presenting them in a user-friendly format, means for using APIs, FTP servers, and HTML scraping when acquiring information, means for analyzing inquiries and inputting the analyzed inquiries as prompts into the generative artificial intelligence model, and means for using regular expressions when formatting the responses from the generative artificial intelligence model. This enables the entire process from information collection to presenting responses to users to be consistently automated, allowing for quick and accurate responses to user inquiries.

[0694] A "supplier" refers to a company or organization that provides specific goods or services.

[0695] "Information" includes data such as system information, specifications, and parameter information, and is provided by each supplier.

[0696] A "generative artificial intelligence model" is an artificial intelligence algorithm that generates natural language responses based on collected information and user inquiries.

[0697] "Tokenizing" refers to the process of breaking down collected text information into words and phrases.

[0698] "Normalization" refers to preprocessing that converts collected information into a unified format to make it easier to process.

[0699] A "prompt statement" is a sentence containing a question or instruction to be input into a generative artificial intelligence model.

[0700] "API" stands for Application Programming Interface, and it is a means of exchanging information with other software applications.

[0701] An "FTP server" is a server that executes a protocol for transferring files.

[0702] "HTML scraping" refers to a technique for automatically extracting necessary information from web pages.

[0703] "Natural language processing" refers to a set of techniques and algorithms that enable computers to understand, analyze, and generate human language.

[0704] A "regular expression" is a method of representing string patterns to perform string manipulation and searching efficiently.

[0705] An "inquiry" refers to a question or request for information that a user enters into a system.

[0706] This invention is a system that automatically collects and integrates information provided by various suppliers and generates responses to user inquiries using a generative artificial intelligence model. This system is implemented using the following hardware and software.

[0707] The server automatically collects information from supplier APIs and FTP servers via the corporate network. For example, it uses the Python requests library to access APIs and retrieve information. From FTP servers, it uses the Python ftplib library to download necessary files. BeautifulSoup is used to extract data from the company wiki using HTML scraping. This allows the server to store information in a centralized database. For example, a scheduled job runs at midnight every day to maintain the latest data.

[0708] The collected information is split into tokens and normalized by the server. Specifically, the NLTK library in Python is used to split the text data into words and phrases, convert all text to lowercase, and remove unnecessary symbols and spaces. Stemming and lemmatization are also applied to unify the words into their base forms. The information, after this preprocessing is complete, is then input into a generative artificial intelligence model (e.g., GPT-4).

[0709] The user enters a query into the system via a terminal. For example, they might enter a query such as, "Please tell me the latest parameter information and best vendor for base stations," and click the send button. The terminal sends this input to the server, which analyzes the query using a natural language processing library (e.g., spaCy). The analyzed query is then sent as a prompt to a generative artificial intelligence model, which generates the optimal response.

[0710] The generated response is formatted by the server. For example, Python regular expressions are used to correct the text formatting and grammar, transforming it into a user-friendly format. The formatted response is sent to the terminal and presented to the user. Specifically, a web browser might display something like, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0711] Specific example

[0712] For example, a user might type "Tell me the latest parameter information and best vendor for the base station" into their device. The device receives this input and sends a query to the server. The server analyzes the question using natural language processing and feeds it to a generative artificial intelligence model. The generative AI model generates a response such as "The latest parameter information is XX, and the best vendor is YY. The reason is that it has the highest performance." The server formats the generated response and sends it to the device. The user then checks the formatted response on their device screen.

[0713] Examples of prompt statements

[0714] "Please provide the latest parameter information for base stations and the best vendor."

[0715] "Please provide the system information for the server you are currently using."

[0716] "Please provide the latest specifications from supplier XX."

[0717] This invention allows users to quickly and accurately obtain the information they need, which is expected to improve work efficiency.

[0718] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0719] Step 1: Gathering information from suppliers

[0720] The server uses the corporate network to collect information from supplier APIs and FTP servers. Specifically, it uses the Python requests library to send HTTP GET requests and retrieve system information, specifications, and parameter information from the API. For FTP servers, it uses the Python ftplib library to connect and download the latest CSV file. It uses API endpoints and FTP server connection information as input and obtains the retrieved JSON data and CSV files as output.

[0721] Step 2: Extracting data from the company wiki

[0722] The server extracts information from the company's internal wiki using HTML scraping. It uses the Python BeautifulSoup library to send HTTP GET requests to specific page URLs and parse the HTML. The internal wiki URL is used as input, and the necessary text information is extracted as output.

[0723] Step 3: Tokenizing and normalizing the data

[0724] The collected information is tokenized and normalized by the server. Specifically, the NLTK library in Python is used to split the text data into words and phrases, convert all text to lowercase, and remove unnecessary symbols and spaces. Furthermore, stemming and lemmatization are applied to unify the words into their base forms. The collected text data is used as input, and tokenized and normalized text data is obtained as output.

[0725] Step 4: Receiving user inquiries

[0726] The user inputs a query into the system via the terminal. They enter a specific inquiry, such as "Please tell me the latest base station parameter information and the best vendor," and click the send button. The user's query text is used as input, and the output is a transmission request from the terminal to the server.

[0727] Step 5: Analyze the inquiry

[0728] The terminal sends the received query to the server, which then parses the query using a natural language processing library (e.g., spaCy). The user's query text is used as input, and the parsed query content is obtained as output.

[0729] Step 6: Sending prompts to the generative AI model

[0730] The server sends the analyzed query as a prompt to a generative artificial intelligence model. Specifically, it inputs the prompt to a generative AI model such as GPT-4, which then generates the optimal response. The analyzed query is used as input, and the generated response is obtained as output.

[0731] Step 7: Formatting the response

[0732] The generated response is formatted by the server. Python regular expressions are used to correct the text formatting and grammar, transforming it into a user-friendly format. The response text from a generative AI model is used as input, and the formatted response is obtained as output.

[0733] Step 8: Presenting a response to the user

[0734] The formatted response is sent from the server to the terminal, which then displays it to the user. Specifically, the web browser will display "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." The formatted response text is used as input, and the output is text in a format that the user can view on the screen.

[0735] (Application Example 1)

[0736] Next, we will explain Application Example 1. In the following explanation, 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."

[0737] In large-scale production facilities such as factories, effectively managing operational data and maintenance information for each piece of equipment and machinery, and providing it to on-site workers quickly and accurately, is a critical challenge. However, the data provided by various suppliers is diverse, and centrally collecting and integrating this data to provide appropriate answers to workers' questions is not easy. In such a situation, on-site operational efficiency may decrease, and the risk of maintenance delays and equipment failures may increase. To solve this, a system is needed that effectively manages data and provides workers with the necessary information in a timely manner.

[0738] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0739] In this invention, the server includes means for automatically collecting and integrating data provided by each supplier, means for inputting the collected data into a generative artificial intelligence model, means for receiving questions from users and posing questions to the generative artificial intelligence model to generate responses, means for presenting the generated responses to the user, means for periodically collecting operational data and maintenance information of each device and equipment in the factory and storing it in an integrated database, and means for inputting questions into the system through a terminal used by the user and generating appropriate responses based on those questions. As a result, factory workers can quickly obtain the latest operational status and maintenance information of each device and equipment and take necessary actions quickly.

[0740] "Each supplier" refers to a company or organization that provides the equipment, parts, or data necessary for a factory or industry.

[0741] "Means of automated data collection and integration" refers to devices or software that perform the process of acquiring data provided by suppliers on a regular or real-time basis and storing it in an integrated database for centralized management.

[0742] A "generative artificial intelligence model" refers to an artificial intelligence model used for natural language generation and response generation, specifically employing advanced language models such as GPT (Generative Pre-trained Transducer).

[0743] "A means of receiving questions from users and posing those questions to a generative artificial intelligence model to generate responses" refers to a device or software that receives questions input by users, appropriately analyzes those questions, inputs them into a generative artificial intelligence model, and executes the process of generating appropriate responses.

[0744] "Means of presenting the generated response to the user" refers to devices or software that visually or audibly display the response generated by the artificial intelligence model on the user's device.

[0745] "Operational data" refers to information such as the operating status, performance data, and operation history of each piece of equipment and machinery in a production facility.

[0746] "Maintenance information" refers to inspection records, repair history, and information on scheduled maintenance work for each device and piece of equipment.

[0747] "Means of storing data in an integrated database" refers to a database system that stores and centrally manages diverse collected data, making it available for organizational use.

[0748] "User-used devices" refer to devices used by factory workers to input questions and receive responses, and specifically include smart glasses and head-mounted displays.

[0749] The embodiments for carrying out this invention are shown below.

[0750] System Overview

[0751] This system periodically collects operational data and maintenance information from various devices and equipment within the factory and stores it in an integrated database. It then uses a generative artificial intelligence model to provide optimal responses to user inquiries. The system primarily consists of a "server," a "terminal," and a "generative artificial intelligence model."

[0752] Data collection and integration

[0753] The server automatically collects and centrally integrates data provided by each supplier. Specifically, it periodically retrieves data from API and FTP servers and stores it in an integrated database. Operational data and maintenance information are also collected and integrated at this time.

[0754] Data preprocessing and input to generative AI.

[0755] The collected data is tokenized and normalized by the server. For example, to input into generative artificial intelligence models such as GPT-4, the collected data is divided into words and phrases and converted into a unified format.

[0756] User question reception and response generation

[0757] The user inputs a question into the system using a device (such as smart glasses or a head-mounted display). For example, they might input, "Please tell me the latest maintenance information and recommended actions." The device sends the input to the server, which uses a generative artificial intelligence model to generate an appropriate response.

[0758] Formatting and providing responses

[0759] The generated response is formatted by the server and presented to the user in an easy-to-understand format. For example, it might be presented as, "Latest maintenance information is XX, recommended action is YY, reason is ZZ." The formatted response is sent to the terminal, where the user can confirm it visually or audibly.

[0760] Specific example

[0761] A factory worker uses smart glasses to ask, "What is the latest maintenance information and recommended action?" This question is sent to a server, and a generative artificial intelligence model (such as GPT-4) generates an appropriate response based on the collected and integrated data. For example, it might generate "The latest maintenance information is XX, and the recommended action is YY," which is then displayed on the user's smart glasses.

[0762] Example of a prompt

[0763] "Please provide the latest maintenance information and recommended actions."

[0764] This configuration allows factory workers to quickly obtain the latest operational status and maintenance information for each piece of equipment and machinery, and to take necessary actions promptly.

[0765] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0766] Step 1:

[0767] The server collects data provided by each supplier. This involves using APIs and FTP servers to retrieve operational data and maintenance information. The collected data is stored in an integrated database for centralized management. The input is the data provided by each supplier, and the output is the data stored in the integrated database.

[0768] Step 2:

[0769] The server tokenizes and normalizes the collected data. This involves splitting text data into words and phrases and converting it into a unified format. Specifically, it formats the data so that it can be input into a generative artificial intelligence model (e.g., GPT-4). The input is the collected raw data, and the output is the tokenized and normalized data.

[0770] Step 3:

[0771] The user enters a question using a terminal. For example, they might enter a natural language question such as, "Please tell me the latest maintenance information and recommended actions." The input is the user's question, and the output is the request data sent from the terminal to the server.

[0772] Step 4:

[0773] The terminal receives the user's question and sends it to the server. In this step, the user's question is properly communicated to the server. The input is the user's question, and the output is the request data sent to the server.

[0774] Step 5:

[0775] The server processes user questions and performs analysis to prepare them for input into a generative artificial intelligence model. Specifically, it uses natural language processing to analyze the questions and converts them into a format suitable for the generative AI model. The input is the user's question data, and the output is the analyzed question that is input into the generative AI model.

[0776] Step 6:

[0777] The server uses a generative artificial intelligence model to generate responses to user questions. For example, it generates answers based on input questions using GPT-4. The input is parsed question data, and the output is the generated response text.

[0778] Step 7:

[0779] The server formats the generated response and converts it into a user-friendly format. This process involves grammar checking and text formatting adjustments. The input is the generated response text, and the output is the formatted response text.

[0780] Step 8:

[0781] The server sends the formatted response to the terminal. In this step, the formatted response is transmitted to the user's terminal. The input is the formatted response text, and the output is the response data sent to the terminal.

[0782] Step 9:

[0783] The user views the response generated and formatted by a generative artificial intelligence model through their device. For example, it might be displayed on smart glasses or a head-mounted display. The input is the response data sent to the device, and the output is the response displayed to the user.

[0784] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0785] This invention combines an emotion engine with a system that automatically collects and integrates data provided by various vendors and responds to user questions based on a generative artificial intelligence model. The system tokenizes and normalizes the collected data and inputs it into the generative AI model. It also receives user questions, poses questions to the generative AI model to generate responses, formats those responses, and presents them to the user. Furthermore, it uses the emotion engine to recognize the user's emotions, adjusts the response based on those emotions, and visualizes and presents the emotions to the user as needed.

[0786] Data collection and integration

[0787] The server automatically collects system information, specifications, and parameter information from various vendors' APIs and FTP servers via the corporate network. This data is retrieved periodically and stored in a centralized database. Additionally, an HTML parser is used to scrape base station-related information from the company wiki, and this data is also integrated into the centralized database. Normalization is performed to standardize the data format and eliminate redundant information.

[0788] Data preprocessing and input to generative AI.

[0789] The collected data is tokenized and normalized by the server. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency. Once these preprocessing steps are complete, the data is input into a generative artificial intelligence model (e.g., GPT-4). The server provides the data to the generative AI using an API, and the model generates appropriate responses to the user's questions.

[0790] User question reception and response generation

[0791] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for the base station and the best vendor." The terminal sends this input to a server, which analyzes the question using natural language processing. The analyzed question is then sent to a generative artificial intelligence model, which generates an appropriate response to the question. For example, a response like, "The latest parameter information is XX, and the best vendor is YY" might be generated.

[0792] Adjusting responses using an emotional engine

[0793] The generated responses are analyzed by an emotion engine before being formatted by the server. The emotion engine recognizes emotions from the text contained in the user's question and adjusts the response of the generative artificial intelligence model. For example, if the user expresses anxiety, elements that provide reassurance will be added to the response.

[0794] Formatting and providing responses

[0795] The responses, refined by the emotion engine, are formatted by the server. The generated text is corrected to the appropriate grammar and format. The formatted responses are sent to the terminal and displayed to the user. For example, information might be presented on the screen in the form of "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." Furthermore, the user's emotional state is visualized, and feedback is provided based on this.

[0796] Specific example

[0797] 1. The user enters "Please tell me the latest base station parameter information and best vendor" on their device.

[0798] 2. The terminal receives the question and sends it to the server.

[0799] 3. The server analyzes the question using natural language processing and sends it to a generative artificial intelligence model.

[0800] 4. The generative AI generates a response saying, "The latest parameter information is XX, and the best vendor is YY. The reason is that it has the highest performance."

[0801] 5. The server passes the generated response to the emotion engine, which analyzes the user's emotions and adjusts the response accordingly.

[0802] 6. The server sends a formatted response to the terminal for the user to confirm.

[0803] Thus, by providing responses that take into account the user's emotions, the present invention enables the provision of more appropriate and satisfying information.

[0804] The following describes the processing flow.

[0805] Step 1:

[0806] The server automatically collects system information, specifications, and parameter information from various vendors' APIs and FTP servers via the corporate network. For example, a scheduled job runs at midnight every day to retrieve the latest data. The collected data is stored in temporary storage.

[0807] Step 2:

[0808] The server uses an HTML parser to scrape and extract base station-related information from the company's internal wiki. This information is also stored in temporary storage.

[0809] Step 3:

[0810] The server integrates the collected data into a centralized management database. It performs normalization to standardize the data format and eliminate duplicate and redundant information. It also cross-references the data with existing database contents, replacing outdated information with the latest information.

[0811] Step 4:

[0812] The server performs tokenization and normalization processes on the data. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency.

[0813] Step 5:

[0814] The server uses an API to input tokenized and normalized data into a generative artificial intelligence model. Data is sent to the generative AI model via the API client, and the model is trained based on this data.

[0815] Step 6:

[0816] The user enters questions into the system via a terminal. For example, they might enter a question such as, "Please tell me the latest parameter information for base stations and the best vendor."

[0817] Step 7:

[0818] The terminal receives the entered question and sends it to the server. The server uses natural language processing to analyze the question and extract the necessary information.

[0819] Step 8:

[0820] The server sends the analyzed question to a generative artificial intelligence model. The generative AI model generates an appropriate response based on the data. For example, it might generate a response such as, "The latest parameter information is XX, and the best vendor is YY."

[0821] Step 9:

[0822] The server sends the generated response to the emotion engine, which performs sentiment analysis based on the user's input. The emotion engine recognizes the user's emotions and adjusts the response accordingly. For example, if the user expresses anxiety, elements that provide reassurance will be added to the response.

[0823] Step 10:

[0824] The server formats the responses that have been refined by the emotion engine. It corrects the grammar and formatting of the generated text and prepares it for the final form.

[0825] Step 11:

[0826] The terminal displays formatted responses to the user. For example, information might be presented in the form of, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance," and the user's emotional state may also be visualized.

[0827] This entire process allows users to quickly and accurately obtain the information they need, leading to improved work efficiency. Furthermore, the introduction of an emotion engine makes responses more user-friendly.

[0828] (Example 2)

[0829] Next, we will describe Example 2. 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."

[0830] In modern information systems, it is crucial to appropriately collect and integrate data from a wide variety of vendors and to effectively utilize generative artificial intelligence models to respond to user inquiries. However, conventional systems have limitations in improving the user experience because they lack sufficient preprocessing, such as normalization and tokenization of collected data, and are unable to provide responses that take user emotions into consideration. Furthermore, there is a need to present the generated responses in an easily understandable format, and conventional methods also have room for improvement in this regard.

[0831] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting and integrating data provided by each information provider, means for tokenizing and normalizing the collected data and inputting it into a generative artificial intelligence model, means for receiving questions from the user and posing questions to the generative artificial intelligence model to generate responses, means for recognizing the user's emotions and adjusting the responses, and means for formatting the generated responses and presenting them in a format that is easy for the user to understand. This enables efficient integration of data from multiple information sources, high-quality response generation using a generative artificial intelligence model, and the provision of responses that take into account the user's emotions, thereby improving the user experience.

[0832] "Information provider" refers to an institution or organization that provides data or information.

[0833] A "generative artificial intelligence model" is a type of artificial intelligence that generates new data or responses based on given input data.

[0834] "Collection" refers to the process of gathering scattered data into one place.

[0835] "Integration" refers to the process of centralizing collected data and compiling it into a consistent format.

[0836] "Tokenization" refers to the process of dividing text data into smaller units such as words and phrases.

[0837] "Normalization" refers to the process of transforming data to make it consistent.

[0838] A "question" refers to a question that a user enters into the system.

[0839] "Response" refers to the answer that the system generates in response to a question.

[0840] "Emotion" refers to the user's psychological state in response to questions and answers.

[0841] "Adjustment" refers to the process of modifying the generated response based on the user's emotions.

[0842] "Formatting" refers to the process of arranging a response to conform to the appropriate grammar and format.

[0843] "Format" refers to the specific layout and structure used when displaying data or information.

[0844] This invention relates to a system that automatically collects and integrates data provided by various information providers and responds to user questions using a generative artificial intelligence model. Furthermore, this system has the feature of analyzing the user's emotions using an emotion engine and adjusting its response based on those emotions.

[0845] Data collection and integration

[0846] The server has the capability to automatically collect data from each information provider's API or FTP server. This data includes system information, specifications, and parameter information, and is retrieved periodically. The data is then stored in a centralized database. The server also uses an HTML parser to scrape base station-related information from the company's internal wiki and integrates this into the centralized database as well. This standardizes the format of data from each information provider and performs normalization to eliminate redundant information.

[0847] Data preprocessing and input to generative artificial intelligence models

[0848] The collected data undergoes a tokenization and normalization process by the server. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency. Once these preprocessing steps are complete, the data is input into a generative artificial intelligence model (e.g., GPT-4). The server provides the data to the generative AI using an API, and the model generates appropriate responses to the user's questions.

[0849] As a concrete example, the server calls the information provider's API at 1 AM every day to obtain the latest information and integrate it into a centralized management database. For instance, it obtains the "latest parameter information for base stations," normalizes it, and then inputs it into GPT-4.

[0850] User question reception and analysis

[0851] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for base stations and the best vendor." The terminal sends this input to the server. The server receives the question and analyzes it using natural language processing techniques. The analyzed question is sent to a generative artificial intelligence model, which generates a response to the question.

[0852] For example, if a user enters "Tell me the latest parameter information for the base station and the best vendor," the server will analyze this question and generate a response such as "The latest parameter information is XX, and the best vendor is YY."

[0853] Adjusting responses using an emotional engine

[0854] The generated response is passed by the server to the sentiment engine. The sentiment engine recognizes the emotions contained in the user's question and adjusts the generated response accordingly. For example, for a user who is feeling anxious, elements that provide reassurance will be added to the response.

[0855] For example, if a user asks a question that includes anxiety, such as, "I have a very important meeting, but I urgently need the base station parameter information. Please let me know the latest information as soon as possible," the server's emotion engine will recognize this as an emotion of anxiety and generate a response such as, "The latest parameter information is XX, and the best vendor is YY. Don't worry, this information will be useful for your meeting."

[0856] Formatting and providing responses

[0857] The responses, refined by the emotion engine, are formatted by the server. The generated text is then corrected to the appropriate grammar and format before being sent to the terminal and displayed to the user. For example, information might be presented on the screen in the form of "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." Furthermore, the user's emotional state is visualized, and feedback is provided based on this.

[0858] Examples of prompt statements

[0859] The following are examples of prompts to input into a generative AI model.

[0860] Input: Please provide the latest parameter information and best vendor for base stations.

[0861] Prompt message: "The user wants to know the latest base station parameter information and the best vendor. The parameter information is XX, and the best vendor is YY. The reason is that it offers the highest performance. Add any elements that would reassure the user."

[0862] Thus, the present invention provides an advanced information system that takes user emotions into consideration and provides appropriate responses.

[0863] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0864] Step 1:

[0865] The server collects data from each information provider's API or FTP server. It uses information from each information provider's API endpoint or FTP server as input. The acquired data (system information, specifications, parameter information, etc.) is stored in a centralized database as output. Specifically, the server periodically sends requests to the API or FTP server according to scheduled tasks, receives response data, and writes it to the database.

[0866] Step 2:

[0867] The server uses an HTML parser to scrape information from the company's internal wiki. It uses the URL of the page to be scraped as input. The extracted data is added to a centralized management database as output. Specifically, the server sends an HTTP request to retrieve the internal wiki page, uses the HTML parser to extract the necessary information, and saves it to the database.

[0868] Step 3:

[0869] The server tokenizes and normalizes the collected data. It uses data stored in a centralized database as input. As output, it converts the tokenized and normalized data into a format that can be input into a generative artificial intelligence model. Specifically, the server reads data from the database and processes it using a tokenizer and normalization library.

[0870] Step 4:

[0871] The user enters the question via their device. The question text entered by the user is used as input. The question text is sent to the server as output. Specifically, the user enters the question into the input form in their web browser and clicks the submit button.

[0872] Step 5:

[0873] The terminal receives the user's question and sends it to the server. It uses the user's question text as input and sends the question text to the server's API endpoint as output. Specifically, the terminal sends the entered question to the server via the API.

[0874] Step 6:

[0875] The server sends a question to a generative AI model for analysis and generation of a response. It uses the user's question text as input and retrieves the response text from the generative AI model as output. Specifically, the server converts the question into a prompt and sends it to the generative AI model's API, then receives the generated response.

[0876] Step 7:

[0877] The server analyzes and adjusts responses using an emotion engine. It uses response text obtained from a generative artificial intelligence model as input. It obtains response text adjusted based on emotion as output. Specifically, the server inputs the response text into an emotion analysis library and adjusts the response based on the analysis results.

[0878] Step 8:

[0879] The server formats the adjusted response and sends it to the terminal. It uses the response text adjusted by the sentiment engine as input. It sends the formatted response text to the terminal as output. Specifically, the server uses a text formatting library to format the response and sends it to the terminal.

[0880] Step 9:

[0881] The terminal displays a formatted response to the user. It uses the formatted response text sent from the server as input. It displays the response on the screen as output. Specifically, the terminal binds the received response text to an element in the web browser and displays it on the screen.

[0882] As a concrete example, a user might input "Tell me the latest parameter information for the base station and the best vendor" into their terminal. The server then goes through the processes of collection, integration, analysis, generation, adjustment, and formatting, and finally displays on the terminal "The latest parameter information is XX, and the best vendor is YY."

[0883] (Application Example 2)

[0884] Next, we will explain Application Example 2. In the following explanation, 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."

[0885] In content distribution services, a challenge exists in providing users with the most suitable content based on their emotional state, as this makes it difficult to make more appropriate recommendations. Traditional systems recommend content without considering user emotions, which may result in insufficient user satisfaction.

[0886] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting and integrating data provided by each supplier, means for inputting the collected data into a generative artificial intelligence model, means for receiving questions from the user and posing questions to the generative artificial intelligence model to generate responses, means for recognizing the user's emotions using an emotion analysis engine and adjusting the response based on those emotions, and means for formatting the adjusted response and presenting it to the user in an easy-to-understand format with added emotional information. This enables more personalized and appropriate content recommendations based on the user's emotional state.

[0887] "Each supplier" refers to companies, organizations, etc., that provide various services or products.

[0888] A "generative artificial intelligence model" refers to an algorithm or system that uses natural language processing to generate appropriate responses based on user input.

[0889] "Methods for tokenizing data" refer to algorithms and processes for dividing collected text data into individual words or phrases.

[0890] "Methods of normalizing data" refer to the process of converting data into a standard format in order to maintain consistency and integrity.

[0891] An "emotion analysis engine" refers to an algorithm or system that analyzes and recognizes emotions from a user's text input.

[0892] "Means of recognizing user emotions" refers to the process of identifying a user's emotional state from their input using an emotion analysis engine.

[0893] "Means of adjusting responses" refers to the process of modifying and optimizing generated responses based on the user's emotions.

[0894] "Methods for formatting responses" refer to the process of making the generated response grammatically correct and converting it into a format that is easy for the user to understand.

[0895] "Means of presenting with added emotional information" refers to processes and mechanisms for presenting feedback based on the user's emotional state.

[0896] This invention describes a system that recommends appropriate content based on the user's emotional state. This system automatically collects data from various suppliers, integrates it, and inputs it into a generative artificial intelligence model. Furthermore, it incorporates an emotion analysis engine and has the function of adjusting the response based on the user's emotional state.

[0897] Hardware and software to be used

[0898] Hardware:

[0899] User device (smartphone)

[0900] server

[0901] software:

[0902] Emotion analysis engine (Hugging Face's transformers library)

[0903] Natural language processing and generative artificial intelligence models (OpenAI GPT-3)

[0904] For data collection, an HTTP request library (e.g., requests) is used.

[0905] For data processing, use Python.

[0906] Data collection and integration

[0907] The server automatically collects data provided by each supplier. The data is retrieved from the suppliers' APIs and FTP servers and integrated and stored in a centralized database. During this process, the collected data is tokenized and normalized.

[0908] Question reception and sentiment analysis

[0909] The user enters a question in natural language through their device. This question is sent to a server, where an emotion analysis engine analyzes the user's emotional state. The system then determines whether the user's emotional state is positive or negative.

[0910] Input and response generation for generative AI models

[0911] The server inputs tokenized and normalized data into a generative artificial intelligence model (e.g., GPT-3) to generate prompt sentences that take into account the user's emotional state. The generative AI model then generates an appropriate response based on these prompts.

[0912] Example of a prompt:

[0913] The user is in a negative state and asks the following question: "I've been feeling down lately, do you have any movie recommendations?" What would be the best content?

[0914] Coordinating and providing responses

[0915] The generated response is adjusted by an emotion analysis engine based on the user's emotions. For example, if the user is feeling down, it will recommend content that will lift their spirits. Finally, the refined response is sent to the user's device and presented to the user with added emotional information.

[0916] Specific example

[0917] When a user asks, "I've been feeling down lately, do you have any movie recommendations?", this question is sent from the device to the server. The server performs sentiment analysis and recognizes a negative emotional state. Then, a prompt message like the following is generated:

[0918] The user is in a negative state and asks the following question: "I've been feeling down lately, do you have any movie recommendations?" What would be the best content?

[0919] The generative artificial intelligence model uses this prompt to recommend an appropriate movie and explain the reason. For example, it might generate a response like, "My recommended movie is 'The Bucket List.' Its moving story will warm your heart." This ensures that the user is offered content that takes their emotions into consideration.

[0920] The above describes embodiments of the present invention. By using the present invention, it is possible to realize more appropriate content recommendations based on the user's emotional state.

[0921] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0922] Step 1:

[0923] The server automatically collects and integrates data provided by each supplier. Input data is obtained from the suppliers' APIs and FTP servers. This data is stored in a centralized database. Specifically, it periodically retrieves data using an HTTP request library and stores it in the server's database.

[0924] Step 2:

[0925] The server tokenizes and normalizes the collected data. The input is raw text data, and the output is data converted into a format suitable for generative artificial intelligence models. Specifically, it uses Python's natural language processing library to split the text into words and phrases and convert them into a standard format.

[0926] Step 3:

[0927] The user enters a question through their terminal. For example, they might enter, "I've been feeling down lately, do you have any movie recommendations?" This input is sent directly to the server. The input is the user's natural language question text.

[0928] Step 4:

[0929] The server analyzes the user's question using an emotion analysis engine. The input is the user's question text, and the output is the user's emotional state (positive or negative). Specifically, the Hugging Face transformers library is used to determine the emotion from the text.

[0930] Step 5:

[0931] The server generates prompts for a generative artificial intelligence model, taking the user's question and emotional state as input. The input is the user's question text and emotional state, and the output is a prompt. For example, it might generate a prompt such as, "The user is in a NEGATIVE state and asks the following question: 'I've been feeling down lately, do you have any movie recommendations?' What would be the best content?"

[0932] Step 6:

[0933] A generative artificial intelligence model (e.g., GPT-3) is input with a prompt sentence and generates a response. The input is the prompt sentence, and the output is the generated response text. Specifically, the server uses the OpenAI API to send the prompt sentence and receive optimal content recommendations.

[0934] Step 7:

[0935] The server adjusts the generated response based on its sentiment analysis engine. The input is the generated response text and the user's emotional state, and the output is the adjusted response text. For example, if the user is in a negative emotional state, the response will include comforting content.

[0936] Step 8:

[0937] The server formats the adjusted response and presents it in a user-friendly format. The input is the adjusted response text, and the output is the final text displayed to the user. Specifically, it performs grammar checks and formatting corrections to make it easy for the user to understand.

[0938] Step 9:

[0939] The refined response is sent to the terminal and presented to the user. The input is the formatted final text, and the output is a content recommendation displayed on the user's terminal screen. For example, a specific response such as "The recommended movie is 'The Bucket List.' It's a touching story that will warm your heart" might be displayed.

[0940] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0941] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0942] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0943] [Fourth Embodiment]

[0944] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0945] As shown in Figure 7, the 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.

[0946] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0947] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0948] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0949] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0950] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0951] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0952] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0953] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0954] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0955] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0956] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0957] This invention is a system that automatically collects and integrates data provided by various vendors. The system inputs the collected data into a generative artificial intelligence model, receives questions from the user, and generates responses based on these questions. The generated responses are formatted and presented in a user-friendly format.

[0958] Data collection and integration

[0959] The server automatically collects system information, specifications, and parameter information from vendor APIs and FTP servers via the corporate network. For example, a scheduled job runs at midnight every day to retrieve the latest data and save it to the database. This data is also extracted from the company's internal wiki using an HTML parser and similarly stored in a centralized database. This data collection process ensures that the information is always up-to-date.

[0960] Data preprocessing and input to generative AI.

[0961] The collected data is tokenized and normalized by the server. For example, text data is split into words and phrases, normalized, and converted into a unified format. This pre-processed data is then input into a generative artificial intelligence model (e.g., GPT-4). The server uses this AI model to generate the best possible response to the user's question based on the collected data.

[0962] User question reception and response generation

[0963] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for the base station and the best vendor." The terminal sends this input to a server, which analyzes the question using natural language processing. The analyzed question is then sent to a generative artificial intelligence model, which generates an appropriate response to the question. For example, a response like, "The latest parameter information is XX, and the best vendor is YY" might be generated.

[0964] Formatting and providing responses

[0965] The generated response is formatted by the server. The text formatting and grammar are corrected and converted into a user-friendly format. The formatted response is sent to the terminal and displayed to the user. For example, the information might be displayed on the screen in the format, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0966] Specific example

[0967] 1. The user enters "Please tell me the latest base station parameter information and best vendor" on their device.

[0968] 2. The terminal receives the input and sends the question to the server.

[0969] 3. The server analyzes the question using natural language processing and feeds it to a generative artificial intelligence model.

[0970] 4. The generative AI generates a response saying, "The latest parameter information is XX, and the optimal vendor is YY. The reason is that it has the highest performance."

[0971] 5. The server formats the generated response and sends it to the terminal.

[0972] 6. The user confirms the formatted response on their device screen.

[0973] This system allows users to quickly and accurately obtain the information they need, which is expected to improve work efficiency.

[0974] The following describes the processing flow.

[0975] Step 1:

[0976] The server automatically collects system information, specifications, and parameter information from APIs and FTP servers provided by various vendors via the corporate network. A scheduled job runs at midnight every day, retrieving this data and storing it in temporary storage.

[0977] Step 2:

[0978] The server uses an HTML parser to scrape and extract base station-related information from the company's internal wiki. This information is also stored in temporary storage.

[0979] Step 3:

[0980] The server integrates the collected data into a centralized management database. Normalization is performed to standardize the data format and eliminate redundant information. It also cross-references the data with existing database contents, replacing outdated information with the latest information.

[0981] Step 4:

[0982] The server performs tokenization and normalization of the integrated data for input into a generative artificial intelligence model. Text data is divided into words and phrases, normalized, and converted into a consistent format.

[0983] Step 5:

[0984] The server uses the AI ​​toolkit (API) interface to provide data to the generative artificial intelligence model. Requests to the generative AI are sent through the API client.

[0985] Step 6:

[0986] The user enters questions into the system via a terminal. For example, they might enter a question such as, "Please tell me the latest parameter information for base stations and the best vendor."

[0987] Step 7:

[0988] The terminal receives input and sends a question to the server. The server tokenizes this question and performs semantic analysis to understand the intent of the question.

[0989] Step 8:

[0990] The server sends the analyzed question to a generative artificial intelligence model. The generative AI model generates an appropriate response based on the data. For example, it might generate a response such as, "The latest parameter information is XX, and the best vendor is YY."

[0991] Step 9:

[0992] The server formats the generated response. The generated text is corrected to have the appropriate grammar and formatting.

[0993] Step 10:

[0994] The terminal displays a formatted response to the user. For example, the information might be presented on the screen in the format: "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[0995] This entire process allows users to quickly and accurately obtain the information they need, resulting in improved work efficiency.

[0996] (Example 1)

[0997] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0998] In today's business environment, there is a need to quickly and accurately integrate various information provided by multiple suppliers and effectively respond to user inquiries. However, performing this manually is time-consuming and labor-intensive, and can lead to inaccuracies in information integration and response generation. Furthermore, there is a lack of integrated systems that consistently automate information collection, preprocessing, analysis, and response generation and formatting.

[0999] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1000] In this invention, the server includes means for automatically collecting and integrating information provided by each supplier, means for inputting the collected information into a generative artificial intelligence model, and means for receiving inquiries from users and posing inquiries to the generative artificial intelligence model to generate responses. This includes means for dividing the collected information into tokens, normalizing them and inputting them into the generative artificial intelligence model, means for formatting the generated responses and presenting them in a user-friendly format, means for using APIs, FTP servers, and HTML scraping when acquiring information, means for analyzing inquiries and inputting the analyzed inquiries as prompts into the generative artificial intelligence model, and means for using regular expressions when formatting the responses from the generative artificial intelligence model. This enables the entire process from information collection to presenting responses to users to be consistently automated, allowing for quick and accurate responses to user inquiries.

[1001] A "supplier" refers to a company or organization that provides specific goods or services.

[1002] "Information" includes data such as system information, specifications, and parameter information, and is provided by each supplier.

[1003] A "generative artificial intelligence model" is an artificial intelligence algorithm that generates natural language responses based on collected information and user inquiries.

[1004] "Tokenizing" refers to the process of breaking down collected text information into words and phrases.

[1005] "Normalization" refers to preprocessing that converts collected information into a unified format to make it easier to process.

[1006] A "prompt statement" is a sentence containing a question or instruction to be input into a generative artificial intelligence model.

[1007] "API" stands for Application Programming Interface, and it is a means of exchanging information with other software applications.

[1008] An "FTP server" is a server that executes a protocol for transferring files.

[1009] "HTML scraping" refers to a technique for automatically extracting necessary information from web pages.

[1010] "Natural language processing" refers to a set of techniques and algorithms that enable computers to understand, analyze, and generate human language.

[1011] A "regular expression" is a method of representing string patterns to perform string manipulation and searching efficiently.

[1012] An "inquiry" refers to a question or request for information that a user enters into a system.

[1013] This invention is a system that automatically collects and integrates information provided by various suppliers and generates responses to user inquiries using a generative artificial intelligence model. This system is implemented using the following hardware and software.

[1014] The server automatically collects information from supplier APIs and FTP servers via the corporate network. For example, it uses the Python requests library to access APIs and retrieve information. From FTP servers, it uses the Python ftplib library to download necessary files. BeautifulSoup is used to extract data from the company wiki using HTML scraping. This allows the server to store information in a centralized database. For example, a scheduled job runs at midnight every day to maintain the latest data.

[1015] The collected information is split into tokens and normalized by the server. Specifically, the NLTK library in Python is used to split the text data into words and phrases, convert all text to lowercase, and remove unnecessary symbols and spaces. Stemming and lemmatization are also applied to unify the words into their base forms. The information, after this preprocessing is complete, is then input into a generative artificial intelligence model (e.g., GPT-4).

[1016] The user enters a query into the system via a terminal. For example, they might enter a query such as, "Please tell me the latest parameter information and best vendor for base stations," and click the send button. The terminal sends this input to the server, which analyzes the query using a natural language processing library (e.g., spaCy). The analyzed query is then sent as a prompt to a generative artificial intelligence model, which generates the optimal response.

[1017] The generated response is formatted by the server. For example, Python regular expressions are used to correct the text formatting and grammar, transforming it into a user-friendly format. The formatted response is sent to the terminal and presented to the user. Specifically, a web browser might display something like, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance."

[1018] Specific example

[1019] For example, a user might type "Tell me the latest parameter information and best vendor for the base station" into their device. The device receives this input and sends a query to the server. The server analyzes the question using natural language processing and feeds it to a generative artificial intelligence model. The generative AI model generates a response such as "The latest parameter information is XX, and the best vendor is YY. The reason is that it has the highest performance." The server formats the generated response and sends it to the device. The user then checks the formatted response on their device screen.

[1020] Examples of prompt statements

[1021] "Please provide the latest parameter information for base stations and the best vendor."

[1022] "Please provide the system information for the server you are currently using."

[1023] "Please provide the latest specifications from supplier XX."

[1024] This invention allows users to quickly and accurately obtain the information they need, which is expected to improve work efficiency.

[1025] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1026] Step 1: Gathering information from suppliers

[1027] The server uses the corporate network to collect information from supplier APIs and FTP servers. Specifically, it uses the Python requests library to send HTTP GET requests and retrieve system information, specifications, and parameter information from the API. For FTP servers, it uses the Python ftplib library to connect and download the latest CSV file. It uses API endpoints and FTP server connection information as input and obtains the retrieved JSON data and CSV files as output.

[1028] Step 2: Extracting data from the company wiki

[1029] The server extracts information from the company's internal wiki using HTML scraping. It uses the Python BeautifulSoup library to send HTTP GET requests to specific page URLs and parse the HTML. The internal wiki URL is used as input, and the necessary text information is extracted as output.

[1030] Step 3: Tokenizing and normalizing the data

[1031] The collected information is tokenized and normalized by the server. Specifically, the NLTK library in Python is used to split the text data into words and phrases, convert all text to lowercase, and remove unnecessary symbols and spaces. Furthermore, stemming and lemmatization are applied to unify the words into their base forms. The collected text data is used as input, and tokenized and normalized text data is obtained as output.

[1032] Step 4: Receiving user inquiries

[1033] The user inputs a query into the system via the terminal. They enter a specific inquiry, such as "Please tell me the latest base station parameter information and the best vendor," and click the send button. The user's query text is used as input, and the output is a transmission request from the terminal to the server.

[1034] Step 5: Analyze the inquiry

[1035] The terminal sends the received query to the server, which then parses the query using a natural language processing library (e.g., spaCy). The user's query text is used as input, and the parsed query content is obtained as output.

[1036] Step 6: Sending prompts to the generative AI model

[1037] The server sends the analyzed query as a prompt to a generative artificial intelligence model. Specifically, it inputs the prompt to a generative AI model such as GPT-4, which then generates the optimal response. The analyzed query is used as input, and the generated response is obtained as output.

[1038] Step 7: Formatting the response

[1039] The generated response is formatted by the server. Python regular expressions are used to correct the text formatting and grammar, transforming it into a user-friendly format. The response text from a generative AI model is used as input, and the formatted response is obtained as output.

[1040] Step 8: Presenting a response to the user

[1041] The formatted response is sent from the server to the terminal, which then displays it to the user. Specifically, the web browser will display "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." The formatted response text is used as input, and the output is text in a format that the user can view on the screen.

[1042] (Application Example 1)

[1043] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1044] In large-scale production facilities such as factories, effectively managing operational data and maintenance information for each piece of equipment and machinery, and providing it to on-site workers quickly and accurately, is a critical challenge. However, the data provided by various suppliers is diverse, and centrally collecting and integrating this data to provide appropriate answers to workers' questions is not easy. In such a situation, on-site operational efficiency may decrease, and the risk of maintenance delays and equipment failures may increase. To solve this, a system is needed that effectively manages data and provides workers with the necessary information in a timely manner.

[1045] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1046] In this invention, the server includes means for automatically collecting and integrating data provided by each supplier, means for inputting the collected data into a generative artificial intelligence model, means for receiving questions from users and posing questions to the generative artificial intelligence model to generate responses, means for presenting the generated responses to the user, means for periodically collecting operational data and maintenance information of each device and equipment in the factory and storing it in an integrated database, and means for inputting questions into the system through a terminal used by the user and generating appropriate responses based on those questions. As a result, factory workers can quickly obtain the latest operational status and maintenance information of each device and equipment and take necessary actions quickly.

[1047] "Each supplier" refers to a company or organization that provides the equipment, parts, or data necessary for a factory or industry.

[1048] "Means of automated data collection and integration" refers to devices or software that perform the process of acquiring data provided by suppliers on a regular or real-time basis and storing it in an integrated database for centralized management.

[1049] A "generative artificial intelligence model" refers to an artificial intelligence model used for natural language generation and response generation, specifically employing advanced language models such as GPT (Generative Pre-trained Transducer).

[1050] "A means of receiving questions from users and posing those questions to a generative artificial intelligence model to generate responses" refers to a device or software that receives questions input by users, appropriately analyzes those questions, inputs them into a generative artificial intelligence model, and executes the process of generating appropriate responses.

[1051] "Means of presenting the generated response to the user" refers to devices or software that visually or audibly display the response generated by the artificial intelligence model on the user's device.

[1052] "Operational data" refers to information such as the operating status, performance data, and operation history of each piece of equipment and machinery in a production facility.

[1053] "Maintenance information" refers to inspection records, repair history, and information on scheduled maintenance work for each device and piece of equipment.

[1054] "Means of storing data in an integrated database" refers to a database system that stores and centrally manages diverse collected data, making it available for organizational use.

[1055] "User-used devices" refer to devices used by factory workers to input questions and receive responses, and specifically include smart glasses and head-mounted displays.

[1056] The embodiments for carrying out this invention are shown below.

[1057] System Overview

[1058] This system periodically collects operational data and maintenance information from various devices and equipment within the factory and stores it in an integrated database. It then uses a generative artificial intelligence model to provide optimal responses to user inquiries. The system primarily consists of a "server," a "terminal," and a "generative artificial intelligence model."

[1059] Data collection and integration

[1060] The server automatically collects and centrally integrates data provided by each supplier. Specifically, it periodically retrieves data from API and FTP servers and stores it in an integrated database. Operational data and maintenance information are also collected and integrated at this time.

[1061] Data preprocessing and input to generative AI.

[1062] The collected data is tokenized and normalized by the server. For example, to input into generative artificial intelligence models such as GPT-4, the collected data is divided into words and phrases and converted into a unified format.

[1063] User question reception and response generation

[1064] The user inputs a question into the system using a device (such as smart glasses or a head-mounted display). For example, they might input, "Please tell me the latest maintenance information and recommended actions." The device sends the input to the server, which uses a generative artificial intelligence model to generate an appropriate response.

[1065] Formatting and providing responses

[1066] The generated response is formatted by the server and presented to the user in an easy-to-understand format. For example, it might be presented as, "Latest maintenance information is XX, recommended action is YY, reason is ZZ." The formatted response is sent to the terminal, where the user can confirm it visually or audibly.

[1067] Specific example

[1068] A factory worker uses smart glasses to ask, "What is the latest maintenance information and recommended action?" This question is sent to a server, and a generative artificial intelligence model (such as GPT-4) generates an appropriate response based on the collected and integrated data. For example, it might generate "The latest maintenance information is XX, and the recommended action is YY," which is then displayed on the user's smart glasses.

[1069] Example of a prompt

[1070] "Please provide the latest maintenance information and recommended actions."

[1071] This configuration allows factory workers to quickly obtain the latest operational status and maintenance information for each piece of equipment and machinery, and to take necessary actions promptly.

[1072] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1073] Step 1:

[1074] The server collects data provided by each supplier. This involves using APIs and FTP servers to retrieve operational data and maintenance information. The collected data is stored in an integrated database for centralized management. The input is the data provided by each supplier, and the output is the data stored in the integrated database.

[1075] Step 2:

[1076] The server tokenizes and normalizes the collected data. This involves splitting text data into words and phrases and converting it into a unified format. Specifically, it formats the data so that it can be input into a generative artificial intelligence model (e.g., GPT-4). The input is the collected raw data, and the output is the tokenized and normalized data.

[1077] Step 3:

[1078] The user enters a question using a terminal. For example, they might enter a natural language question such as, "Please tell me the latest maintenance information and recommended actions." The input is the user's question, and the output is the request data sent from the terminal to the server.

[1079] Step 4:

[1080] The terminal receives the user's question and sends it to the server. In this step, the user's question is properly communicated to the server. The input is the user's question, and the output is the request data sent to the server.

[1081] Step 5:

[1082] The server processes user questions and performs analysis to prepare them for input into a generative artificial intelligence model. Specifically, it uses natural language processing to analyze the questions and converts them into a format suitable for the generative AI model. The input is the user's question data, and the output is the analyzed question that is input into the generative AI model.

[1083] Step 6:

[1084] The server uses a generative artificial intelligence model to generate responses to user questions. For example, it generates answers based on input questions using GPT-4. The input is parsed question data, and the output is the generated response text.

[1085] Step 7:

[1086] The server formats the generated response and converts it into a user-friendly format. This process involves grammar checking and text formatting adjustments. The input is the generated response text, and the output is the formatted response text.

[1087] Step 8:

[1088] The server sends the formatted response to the terminal. In this step, the formatted response is transmitted to the user's terminal. The input is the formatted response text, and the output is the response data sent to the terminal.

[1089] Step 9:

[1090] The user views the response generated and formatted by a generative artificial intelligence model through their device. For example, it might be displayed on smart glasses or a head-mounted display. The input is the response data sent to the device, and the output is the response displayed to the user.

[1091] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1092] This invention combines an emotion engine with a system that automatically collects and integrates data provided by various vendors and responds to user questions based on a generative artificial intelligence model. The system tokenizes and normalizes the collected data and inputs it into the generative AI model. It also receives user questions, poses questions to the generative AI model to generate responses, formats those responses, and presents them to the user. Furthermore, it uses the emotion engine to recognize the user's emotions, adjusts the response based on those emotions, and visualizes and presents the emotions to the user as needed.

[1093] Data collection and integration

[1094] The server automatically collects system information, specifications, and parameter information from various vendors' APIs and FTP servers via the corporate network. This data is retrieved periodically and stored in a centralized database. Additionally, an HTML parser is used to scrape base station-related information from the company wiki, and this data is also integrated into the centralized database. Normalization is performed to standardize the data format and eliminate redundant information.

[1095] Data preprocessing and input to generative AI.

[1096] The collected data is tokenized and normalized by the server. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency. Once these preprocessing steps are complete, the data is input into a generative artificial intelligence model (e.g., GPT-4). The server provides the data to the generative AI using an API, and the model generates appropriate responses to the user's questions.

[1097] User question reception and response generation

[1098] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for the base station and the best vendor." The terminal sends this input to a server, which analyzes the question using natural language processing. The analyzed question is then sent to a generative artificial intelligence model, which generates an appropriate response to the question. For example, a response like, "The latest parameter information is XX, and the best vendor is YY" might be generated.

[1099] Adjusting responses using an emotional engine

[1100] The generated responses are analyzed by an emotion engine before being formatted by the server. The emotion engine recognizes emotions from the text contained in the user's question and adjusts the response of the generative artificial intelligence model. For example, if the user expresses anxiety, elements that provide reassurance will be added to the response.

[1101] Formatting and providing responses

[1102] The responses, refined by the emotion engine, are formatted by the server. The generated text is corrected to the appropriate grammar and format. The formatted responses are sent to the terminal and displayed to the user. For example, information might be presented on the screen in the form of "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." Furthermore, the user's emotional state is visualized, and feedback is provided based on this.

[1103] Specific example

[1104] 1. The user enters "Please tell me the latest base station parameter information and best vendor" on their device.

[1105] 2. The terminal receives the question and sends it to the server.

[1106] 3. The server analyzes the question using natural language processing and sends it to a generative artificial intelligence model.

[1107] 4. The generative AI generates a response saying, "The latest parameter information is XX, and the best vendor is YY. The reason is that it has the highest performance."

[1108] 5. The server passes the generated response to the emotion engine, which analyzes the user's emotions and adjusts the response accordingly.

[1109] 6. The server sends a formatted response to the terminal for the user to confirm.

[1110] Thus, by providing responses that take into account the user's emotions, the present invention enables the provision of more appropriate and satisfying information.

[1111] The following describes the processing flow.

[1112] Step 1:

[1113] The server automatically collects system information, specifications, and parameter information from various vendors' APIs and FTP servers via the corporate network. For example, a scheduled job runs at midnight every day to retrieve the latest data. The collected data is stored in temporary storage.

[1114] Step 2:

[1115] The server uses an HTML parser to scrape and extract base station-related information from the company's internal wiki. This information is also stored in temporary storage.

[1116] Step 3:

[1117] The server integrates the collected data into a centralized management database. It performs normalization to standardize the data format and eliminate duplicate and redundant information. It also cross-references the data with existing database contents, replacing outdated information with the latest information.

[1118] Step 4:

[1119] The server performs tokenization and normalization processes on the data. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency.

[1120] Step 5:

[1121] The server uses an API to input tokenized and normalized data into a generative artificial intelligence model. Data is sent to the generative AI model via the API client, and the model is trained based on this data.

[1122] Step 6:

[1123] The user enters questions into the system via a terminal. For example, they might enter a question such as, "Please tell me the latest parameter information for base stations and the best vendor."

[1124] Step 7:

[1125] The terminal receives the entered question and sends it to the server. The server uses natural language processing to analyze the question and extract the necessary information.

[1126] Step 8:

[1127] The server sends the analyzed question to a generative artificial intelligence model. The generative AI model generates an appropriate response based on the data. For example, it might generate a response such as, "The latest parameter information is XX, and the best vendor is YY."

[1128] Step 9:

[1129] The server sends the generated response to the emotion engine, which performs sentiment analysis based on the user's input. The emotion engine recognizes the user's emotions and adjusts the response accordingly. For example, if the user expresses anxiety, elements that provide reassurance will be added to the response.

[1130] Step 10:

[1131] The server formats the responses that have been refined by the emotion engine. It corrects the grammar and formatting of the generated text and prepares it for the final form.

[1132] Step 11:

[1133] The terminal displays formatted responses to the user. For example, information might be presented in the form of, "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance," and the user's emotional state may also be visualized.

[1134] This entire process allows users to quickly and accurately obtain the information they need, leading to improved work efficiency. Furthermore, the introduction of an emotion engine makes responses more user-friendly.

[1135] (Example 2)

[1136] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1137] In modern information systems, it is crucial to appropriately collect and integrate data from a wide variety of vendors and to effectively utilize generative artificial intelligence models to respond to user inquiries. However, conventional systems have limitations in improving the user experience because they lack sufficient preprocessing, such as normalization and tokenization of collected data, and are unable to provide responses that take user emotions into consideration. Furthermore, there is a need to present the generated responses in an easily understandable format, and conventional methods also have room for improvement in this regard.

[1138] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting and integrating data provided by each information provider, means for tokenizing and normalizing the collected data and inputting it into a generative artificial intelligence model, means for receiving questions from the user and posing questions to the generative artificial intelligence model to generate responses, means for recognizing the user's emotions and adjusting the responses, and means for formatting the generated responses and presenting them in a format that is easy for the user to understand. This enables efficient integration of data from multiple information sources, high-quality response generation using a generative artificial intelligence model, and the provision of responses that take into account the user's emotions, thereby improving the user experience.

[1139] "Information provider" refers to an institution or organization that provides data or information.

[1140] A "generative artificial intelligence model" is a type of artificial intelligence that generates new data or responses based on given input data.

[1141] "Collection" refers to the process of gathering scattered data into one place.

[1142] "Integration" refers to the process of centralizing collected data and compiling it into a consistent format.

[1143] "Tokenization" refers to the process of dividing text data into smaller units such as words and phrases.

[1144] "Normalization" refers to the process of transforming data to make it consistent.

[1145] A "question" refers to a question that a user enters into the system.

[1146] "Response" refers to the answer that the system generates in response to a question.

[1147] "Emotion" refers to the user's psychological state in response to questions and answers.

[1148] "Adjustment" refers to the process of modifying the generated response based on the user's emotions.

[1149] "Formatting" refers to the process of arranging a response to conform to the appropriate grammar and format.

[1150] "Format" refers to the specific layout and structure used when displaying data or information.

[1151] This invention relates to a system that automatically collects and integrates data provided by various information providers and responds to user questions using a generative artificial intelligence model. Furthermore, this system has the feature of analyzing the user's emotions using an emotion engine and adjusting its response based on those emotions.

[1152] Data collection and integration

[1153] The server has the capability to automatically collect data from each information provider's API or FTP server. This data includes system information, specifications, and parameter information, and is retrieved periodically. The data is then stored in a centralized database. The server also uses an HTML parser to scrape base station-related information from the company's internal wiki and integrates this into the centralized database as well. This standardizes the format of data from each information provider and performs normalization to eliminate redundant information.

[1154] Data preprocessing and input to generative artificial intelligence models

[1155] The collected data undergoes a tokenization and normalization process by the server. Tokenization is the process of dividing text data into words and phrases, while normalization is a transformation process to maintain data consistency. Once these preprocessing steps are complete, the data is input into a generative artificial intelligence model (e.g., GPT-4). The server provides the data to the generative AI using an API, and the model generates appropriate responses to the user's questions.

[1156] As a concrete example, the server calls the information provider's API at 1 AM every day to obtain the latest information and integrate it into a centralized management database. For instance, it obtains the "latest parameter information for base stations," normalizes it, and then inputs it into GPT-4.

[1157] User question reception and analysis

[1158] The user inputs a question into the system via a terminal. For example, they might input a question like, "Please tell me the latest parameter information for base stations and the best vendor." The terminal sends this input to the server. The server receives the question and analyzes it using natural language processing techniques. The analyzed question is sent to a generative artificial intelligence model, which generates a response to the question.

[1159] For example, if a user enters "Tell me the latest parameter information for the base station and the best vendor," the server will analyze this question and generate a response such as "The latest parameter information is XX, and the best vendor is YY."

[1160] Adjusting responses using an emotional engine

[1161] The generated response is passed by the server to the sentiment engine. The sentiment engine recognizes the emotions contained in the user's question and adjusts the generated response accordingly. For example, for a user who is feeling anxious, elements that provide reassurance will be added to the response.

[1162] For example, if a user asks a question that includes anxiety, such as, "I have a very important meeting, but I urgently need the base station parameter information. Please let me know the latest information as soon as possible," the server's emotion engine will recognize this as an emotion of anxiety and generate a response such as, "The latest parameter information is XX, and the best vendor is YY. Don't worry, this information will be useful for your meeting."

[1163] Formatting and providing responses

[1164] The responses, refined by the emotion engine, are formatted by the server. The generated text is then corrected to the appropriate grammar and format before being sent to the terminal and displayed to the user. For example, information might be presented on the screen in the form of "Latest parameter information is XX, the best vendor is YY, because it offers the highest performance." Furthermore, the user's emotional state is visualized, and feedback is provided based on this.

[1165] Examples of prompt statements

[1166] The following are examples of prompts to input into a generative AI model.

[1167] Input: Please provide the latest parameter information and best vendor for base stations.

[1168] Prompt message: "The user wants to know the latest base station parameter information and the best vendor. The parameter information is XX, and the best vendor is YY. The reason is that it offers the highest performance. Add any elements that would reassure the user."

[1169] Thus, the present invention provides an advanced information system that takes user emotions into consideration and provides appropriate responses.

[1170] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1171] Step 1:

[1172] The server collects data from each information provider's API or FTP server. It uses information from each information provider's API endpoint or FTP server as input. The acquired data (system information, specifications, parameter information, etc.) is stored in a centralized database as output. Specifically, the server periodically sends requests to the API or FTP server according to scheduled tasks, receives response data, and writes it to the database.

[1173] Step 2:

[1174] The server uses an HTML parser to scrape information from the company's internal wiki. It uses the URL of the page to be scraped as input. The extracted data is added to a centralized management database as output. Specifically, the server sends an HTTP request to retrieve the internal wiki page, uses the HTML parser to extract the necessary information, and saves it to the database.

[1175] Step 3:

[1176] The server tokenizes and normalizes the collected data. It uses data stored in a centralized database as input. As output, it converts the tokenized and normalized data into a format that can be input into a generative artificial intelligence model. Specifically, the server reads data from the database and processes it using a tokenizer and normalization library.

[1177] Step 4:

[1178] The user enters the question via their device. The question text entered by the user is used as input. The question text is sent to the server as output. Specifically, the user enters the question into the input form in their web browser and clicks the submit button.

[1179] Step 5:

[1180] The terminal receives the user's question and sends it to the server. It uses the user's question text as input and sends the question text to the server's API endpoint as output. Specifically, the terminal sends the entered question to the server via the API.

[1181] Step 6:

[1182] The server sends a question to a generative AI model for analysis and generation of a response. It uses the user's question text as input and retrieves the response text from the generative AI model as output. Specifically, the server converts the question into a prompt and sends it to the generative AI model's API, then receives the generated response.

[1183] Step 7:

[1184] The server analyzes and adjusts responses using an emotion engine. It uses response text obtained from a generative artificial intelligence model as input. It obtains response text adjusted based on emotion as output. Specifically, the server inputs the response text into an emotion analysis library and adjusts the response based on the analysis results.

[1185] Step 8:

[1186] The server formats the adjusted response and sends it to the terminal. It uses the response text adjusted by the sentiment engine as input. It sends the formatted response text to the terminal as output. Specifically, the server uses a text formatting library to format the response and sends it to the terminal.

[1187] Step 9:

[1188] The terminal displays a formatted response to the user. It uses the formatted response text sent from the server as input. It displays the response on the screen as output. Specifically, the terminal binds the received response text to an element in the web browser and displays it on the screen.

[1189] As a concrete example, a user might input "Tell me the latest parameter information for the base station and the best vendor" into their terminal. The server then goes through the processes of collection, integration, analysis, generation, adjustment, and formatting, and finally displays on the terminal "The latest parameter information is XX, and the best vendor is YY."

[1190] (Application Example 2)

[1191] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1192] In content distribution services, a challenge exists in providing users with the most suitable content based on their emotional state, as this makes it difficult to make more appropriate recommendations. Traditional systems recommend content without considering user emotions, which may result in insufficient user satisfaction.

[1193] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting and integrating data provided by each supplier, means for inputting the collected data into a generative artificial intelligence model, means for receiving questions from the user and posing questions to the generative artificial intelligence model to generate responses, means for recognizing the user's emotions using an emotion analysis engine and adjusting the response based on those emotions, and means for formatting the adjusted response and presenting it to the user in an easy-to-understand format with added emotional information. This enables more personalized and appropriate content recommendations based on the user's emotional state.

[1194] "Each supplier" refers to companies, organizations, etc., that provide various services or products.

[1195] A "generative artificial intelligence model" refers to an algorithm or system that uses natural language processing to generate appropriate responses based on user input.

[1196] "Methods for tokenizing data" refer to algorithms and processes for dividing collected text data into individual words or phrases.

[1197] "Methods of normalizing data" refer to the process of converting data into a standard format in order to maintain consistency and integrity.

[1198] An "emotion analysis engine" refers to an algorithm or system that analyzes and recognizes emotions from a user's text input.

[1199] "Means of recognizing user emotions" refers to the process of identifying a user's emotional state from their input using an emotion analysis engine.

[1200] "Means of adjusting responses" refers to the process of modifying and optimizing generated responses based on the user's emotions.

[1201] "Methods for formatting responses" refer to the process of making the generated response grammatically correct and converting it into a format that is easy for the user to understand.

[1202] "Means of presenting with added emotional information" refers to processes and mechanisms for presenting feedback based on the user's emotional state.

[1203] This invention describes a system that recommends appropriate content based on the user's emotional state. This system automatically collects data from various suppliers, integrates it, and inputs it into a generative artificial intelligence model. Furthermore, it incorporates an emotion analysis engine and has the function of adjusting the response based on the user's emotional state.

[1204] Hardware and software to be used

[1205] Hardware:

[1206] User device (smartphone)

[1207] server

[1208] software:

[1209] Emotion analysis engine (Hugging Face's transformers library)

[1210] Natural language processing and generative artificial intelligence models (OpenAI GPT-3)

[1211] For data collection, an HTTP request library (e.g., requests) is used.

[1212] For data processing, use Python.

[1213] Data collection and integration

[1214] The server automatically collects data provided by each supplier. The data is retrieved from the suppliers' APIs and FTP servers and integrated and stored in a centralized database. During this process, the collected data is tokenized and normalized.

[1215] Question reception and sentiment analysis

[1216] The user enters a question in natural language through their device. This question is sent to a server, where an emotion analysis engine analyzes the user's emotional state. The system then determines whether the user's emotional state is positive or negative.

[1217] Input and response generation for generative AI models

[1218] The server inputs tokenized and normalized data into a generative artificial intelligence model (e.g., GPT-3) to generate prompt sentences that take into account the user's emotional state. The generative AI model then generates an appropriate response based on these prompts.

[1219] Example of a prompt:

[1220] The user is in a negative state and asks the following question: "I've been feeling down lately, do you have any movie recommendations?" What would be the best content?

[1221] Coordinating and providing responses

[1222] The generated response is adjusted by an emotion analysis engine based on the user's emotions. For example, if the user is feeling down, it will recommend content that will lift their spirits. Finally, the refined response is sent to the user's device and presented to the user with added emotional information.

[1223] Specific example

[1224] When a user asks, "I've been feeling down lately, do you have any movie recommendations?", this question is sent from the device to the server. The server performs sentiment analysis and recognizes a negative emotional state. Then, a prompt message like the following is generated:

[1225] The user is in a negative state and asks the following question: "I've been feeling down lately, do you have any movie recommendations?" What would be the best content?

[1226] The generative artificial intelligence model uses this prompt to recommend an appropriate movie and explain the reason. For example, it might generate a response like, "My recommended movie is 'The Bucket List.' Its moving story will warm your heart." This ensures that the user is offered content that takes their emotions into consideration.

[1227] The above describes embodiments of the present invention. By using the present invention, it is possible to realize more appropriate content recommendations based on the user's emotional state.

[1228] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1229] Step 1:

[1230] The server automatically collects and integrates data provided by each supplier. Input data is obtained from the suppliers' APIs and FTP servers. This data is stored in a centralized database. Specifically, it periodically retrieves data using an HTTP request library and stores it in the server's database.

[1231] Step 2:

[1232] The server tokenizes and normalizes the collected data. The input is raw text data, and the output is data converted into a format suitable for generative artificial intelligence models. Specifically, it uses Python's natural language processing library to split the text into words and phrases and convert them into a standard format.

[1233] Step 3:

[1234] The user enters a question through their terminal. For example, they might enter, "I've been feeling down lately, do you have any movie recommendations?" This input is sent directly to the server. The input is the user's natural language question text.

[1235] Step 4:

[1236] The server analyzes the user's question using an emotion analysis engine. The input is the user's question text, and the output is the user's emotional state (positive or negative). Specifically, the Hugging Face transformers library is used to determine the emotion from the text.

[1237] Step 5:

[1238] The server generates prompts for a generative artificial intelligence model, taking the user's question and emotional state as input. The input is the user's question text and emotional state, and the output is a prompt. For example, it might generate a prompt such as, "The user is in a NEGATIVE state and asks the following question: 'I've been feeling down lately, do you have any movie recommendations?' What would be the best content?"

[1239] Step 6:

[1240] A generative artificial intelligence model (e.g., GPT-3) is input with a prompt sentence and generates a response. The input is the prompt sentence, and the output is the generated response text. Specifically, the server uses the OpenAI API to send the prompt sentence and receive optimal content recommendations.

[1241] Step 7:

[1242] The server adjusts the generated response based on its sentiment analysis engine. The input is the generated response text and the user's emotional state, and the output is the adjusted response text. For example, if the user is in a negative emotional state, the response will include comforting content.

[1243] Step 8:

[1244] The server formats the adjusted response and presents it in a user-friendly format. The input is the adjusted response text, and the output is the final text displayed to the user. Specifically, it performs grammar checks and formatting corrections to make it easy for the user to understand.

[1245] Step 9:

[1246] The refined response is sent to the terminal and presented to the user. The input is the formatted final text, and the output is a content recommendation displayed on the user's terminal screen. For example, a specific response such as "The recommended movie is 'The Bucket List.' It's a touching story that will warm your heart" might be displayed.

[1247] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1248] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1249] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1250] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1251] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1252] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1253] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1254] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1255] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1256] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1257] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1258] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1259] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1260] 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.

[1261] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1262] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1263] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1264] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1265] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1266] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1267] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1268] The following is further disclosed regarding the embodiments described above.

[1269] (Claim 1)

[1270] A means of automatically collecting and integrating data provided by each vendor,

[1271] A means of inputting collected data into a generative artificial intelligence model,

[1272] A means of receiving questions from users and posing questions to a generative artificial intelligence model to generate responses,

[1273] A means of presenting the generated response to the user,

[1274] A system that includes this.

[1275] (Claim 2)

[1276] The system according to claim 1, which tokenizes and normalizes collected data and inputs it into a generative artificial intelligence model.

[1277] (Claim 3)

[1278] The system according to claim 1, which formats the generated response and presents it in a format that is easy for the user to understand.

[1279] "Example 1"

[1280] (Claim 1)

[1281] A means of automatically collecting and integrating information provided by each supplier,

[1282] A means of inputting collected information into a generative artificial intelligence model,

[1283] A means of receiving inquiries from users and posing those inquiries to a generative artificial intelligence model to generate answers,

[1284] A means of presenting the generated answer to the user,

[1285] A system that includes this.

[1286] (Claim 2)

[1287] The system according to claim 1, which divides collected information into tokens, normalizes them, and inputs them into a generative artificial intelligence model.

[1288] (Claim 3)

[1289] The system according to claim 1, which formats the generated response and presents it in a format that is easy for the user to understand.

[1290] (Claim 4)

[1291] The system according to claim 1, which uses an API, an FTP server, and HTML scraping to obtain information.

[1292] (Claim 5)

[1293] The system according to claim 1, comprising means for analyzing a query and inputting the analyzed query as a prompt to a generative artificial intelligence model.

[1294] (Claim 6)

[1295] The system according to claim 3, which uses regular expressions when formatting the response from a generative artificial intelligence model.

[1296] "Application Example 1"

[1297] (Claim 1)

[1298] A means of automatically collecting and integrating data provided by each supplier,

[1299] A means of inputting collected data into a generative artificial intelligence model,

[1300] A means of receiving questions from users and posing questions to a generative artificial intelligence model to generate responses,

[1301] A means of presenting the generated response to the user,

[1302] A means of periodically collecting operational data and maintenance information for each piece of equipment and machinery within the factory and storing it in an integrated database,

[1303] A means for inputting a question into the system through a terminal used by the user and generating an appropriate response based on that,

[1304] A system that includes this.

[1305] (Claim 2)

[1306] The system according to claim 1, which tokenizes and normalizes collected data and inputs it into a generative artificial intelligence model.

[1307] (Claim 3)

[1308] The system according to claim 1, which formats the generated response and presents it in a format that is easy for the user to understand.

[1309] "Example 2 of combining an emotion engine"

[1310] (Claim 1)

[1311] A means of automatically collecting and integrating data provided by each information provider,

[1312] A means of inputting collected data into a generative artificial intelligence model,

[1313] A means of receiving questions from users and posing questions to a generative artificial intelligence model to generate responses,

[1314] A means of presenting the generated response to the user,

[1315] A means of recognizing user emotions and adjusting responses,

[1316] A system that includes this.

[1317] (Claim 2)

[1318] The system according to claim 1, which tokenizes and normalizes collected data and inputs it into a generative artificial intelligence model.

[1319] (Claim 3)

[1320] The system according to claim 1, which formats the generated response and presents it in a format that is easy for the user to understand.

[1321] "Application example 2 when combining with an emotional engine"

[1322] (Claim 1)

[1323] A means of automatically collecting and integrating data provided by each supplier,

[1324] A means of inputting collected data into a generative artificial intelligence model,

[1325] A means of receiving questions from users and posing questions to a generative artificial intelligence model to generate responses,

[1326] A means of presenting the generated response to the user,

[1327] A means for recognizing a user's emotions using an emotion analysis engine and adjusting the response based on those emotions,

[1328] A means of shaping a refined response, adding emotional information to it, and presenting it to the user in an easily understandable format,

[1329] A system that includes this.

[1330] (Claim 2)

[1331] The system according to claim 1, which tokenizes and normalizes collected data and inputs it into a generative artificial intelligence model.

[1332] (Claim 3)

[1333] The system according to claim 1, which formats the generated response and adjusts the response using an emotion analysis engine before presenting it in a format that is easy for the user to understand. [Explanation of symbols]

[1334] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of automatically collecting and integrating data provided by each vendor, A means of inputting collected data into a generative artificial intelligence model, A means of receiving a question from a user and posing that question to a generative artificial intelligence model to generate a response, A means of presenting the generated response to the user, A system that includes this.

2. The system according to claim 1, which tokenizes and normalizes collected data and inputs it into a generative artificial intelligence model.

3. The system according to claim 1, which formats the generated response and presents it in a format that is easy for the user to understand.

Images