System

Abstract

A system is provided.SOLUTION: A system comprising: means for providing information in shelter life using a generative model; means for constructing a theme-specific Q & A database; means for reading the Q & A database into the generative model and customizing the Q & A database to a specific theme; means for providing an interface for receiving a question from a user; means for generating an optimal answer to the question of the user using the generative model; and means for transmitting the generated answer to a terminal of the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In recent years, there has been an increase in cases where natural disasters have forced many people to live in evacuation shelters. Living in evacuation shelters can be a major problem, with a lack of information, psychological stress, and inconveniences in daily life. However, the environment for disaster victims to quickly and accurately obtain the information they need is not yet fully developed. People tend to forget the lessons learned from past disasters, and they have difficulty selecting and judging the reliability of information, making it difficult to obtain practical and useful information. To solve these issues, there is a need for the development of a system that can quickly and accurately provide disaster victims with the information they need. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means. First, it includes a means for providing information on life in a shelter using a generative model, a means for building a thematic Q&A database, and a means for loading this into a generative model and customizing it to a specific theme. Furthermore, it includes a means for providing an interface for accepting questions from users, and a means for generating optimal answers to the user's questions using the generative model. Finally, it provides a system including a means for transmitting the generated answers to the user's terminal. This system allows users to quickly and accurately obtain the information they need, improving the quality of life in a shelter.

[0006] A "generative model" is a type of artificial intelligence technique used to generate new information or answers based on specific input data.

[0007] "Life in a shelter" refers to a situation where people leave their homes due to a natural disaster or other reason and temporarily live in a designated evacuation shelter.

[0008] A "topical Q&A database" is a database that stores pairs of questions related to a specific topic and their answers.

[0009] "Customization" is the process of providing more appropriate information by adjusting the generative model to suit a specific purpose or theme.

[0010] "Interface" refers to the operation screen, input form, etc. that allow the user to interact with the system.

[0011] "Means for accepting questions" refers to the functions and mechanisms that allow users to input their own questions or problems into the system.

[0012] The "best answer" is the provision of the most appropriate and useful information or advice to the user's question.

[0013] "Terminal" refers to the device (e.g., smartphone, tablet, computer) used by a user to access the system.

[0014] "Means for sending" refers to the communication means or function for sending the generated answer to the user's terminal. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] The present invention relates to a system for providing information related to life in an evacuation shelter using a generative model. Specific embodiments of the system are described below.

[0037] System Overview

[0038] The system is based on a server-side generative model and provides a user-friendly interface. The server collects data from reliable sources and builds a topic-specific Q&A database. The generative model learns from this database and provides optimal answers to user questions.

[0039] System configuration

[0040] Data collection and registration

[0041] The server collects information related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. The collected data is then categorized by topic, organized into a Q&A format, and registered in a database. Topics include health, education, and psychological support.

[0042] Training a generative model

[0043] The server uses the constructed Q&A database to train a generative model. The generative model utilizes natural language processing technology and is able to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0044] User Interface

[0045] The user interface is provided in a form that is easily accessible to users (e.g., smartphone app, web portal), and includes a question input form and options by category, making it intuitive to operate.

[0046] Question and Answer Process

[0047] When a user enters a question and presses the send button, the device sends the question to the server, which uses the generative model to instantly generate the best answer to the received question and sends it back to the user's device. This answer may also include additional related or supporting information as needed.

[0048] Specific examples

[0049] Example 1: Questions about breastfeeding at evacuation shelters

[0050] User: The user types a question: "How should breastfeeding be handled in an evacuation shelter?"

[0051] Terminal: The terminal sends a question to the server.

[0052] Server: The server uses the generative model to analyze the question and generate an appropriate answer, such as "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, with as much privacy as possible. If necessary, ask the evacuation center manager to provide separate spaces."

[0053] Terminal: The terminal displays the generated answer to the user.

[0054] Example 2: Questions about psychological stress

[0055] User: The user types a question: "Please tell me how to take care of myself mentally while living in an evacuation shelter."

[0056] Terminal: The terminal sends a question to the server.

[0057] Server: The server uses the generative model to generate the optimal answer, such as, "Regular rest and relaxation are important when living in an evacuation shelter. It is a good idea to incorporate relaxation techniques and mindfulness practice. Also, if you are experiencing severe psychological stress, we recommend that you seek professional counseling."

[0058] Terminal: The terminal displays the generated answer to the user.

[0059] In this way, the system of the present invention allows users to quickly and accurately obtain information about life in a shelter, thereby improving their quality of life.

[0060] The processing flow will be explained below.

[0061] Step 1: Collect and organize your data

[0062] The server collects information about life in evacuation shelters from reliable sources (e.g., official websites, documents issued by specialized organizations).

[0063] The server categorizes the collected information by topic, for example into categories such as health, education, and psychological support.

[0064] The server creates specific question-and-answer pairs for each topic, building a Q&A format database.

[0065] Step 2: Training the generative model

[0066] The server trains the generative model using a Q&A database.

[0067] The server applies the trained generative model, enabling the model to generate the best answer to the user's question.

[0068] Training is conducted regularly and new information and data is added.

[0069] Step 3: Prepare the user interface

[0070] The device provides a user-friendly interface, accessible through a smartphone app or web portal.

[0071] The terminal provides a question input form and a category selection menu, allowing users to easily input questions.

[0072] Step 4: Receiving questions from users

[0073] The user enters a question through the interface and presses the submit button.

[0074] The terminal transmits the user's question data to the server.

[0075] Step 5: Parsing the question and generating an answer

[0076] The server analyzes the received question data and passes it to the generative model.

[0077] The server uses the generative model to generate optimal answers to the user's questions.

[0078] If necessary, the server will search for additional information (e.g., related resources or supporting information) and append it to the answer.

[0079] Step 6: Submit your response

[0080] The server sends the generated answer and any additional information required to the user's terminal.

[0081] The terminal displays the received answer to the user.

[0082] Step 7: Gather user feedback

[0083] The terminal collects user feedback and sends it to the server.

[0084] The server analyzes the collected feedback and uses it to train the next generative model.

[0085] Through this series of steps, the system of the present invention can quickly and accurately provide information about life in an evacuation shelter, thereby contributing to improving the lives of users.

[0086] Example 1

[0087] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0088] Obtaining information during evacuation shelter life is difficult, and one of the challenges is the lack of timely provision of appropriate information tailored to individual needs and problems. This is due to limited resources, issues with the reliability of information sources, and the difficulty of accurately categorizing and providing information. Furthermore, a system that combines high accuracy and versatility is required to address a wide range of information needs, including those related to psychological stress and health issues.

[0089] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0090] In this invention, the server includes means for collecting data from reliable information sources, means for constructing a Q&A database categorized by theme, means for loading the Q&A database into a generative model to learn information, means for training the generative model to generate optimal answers to user questions, means for providing an interface for accepting questions from users, means for transmitting the generated answers to the user's terminal, and means for displaying the generated answers on the user's terminal, thereby enabling users to quickly and accurately obtain a wide range of information related to life in a shelter.

[0091] "Reliable sources" are institutions or platforms that provide accurate and up-to-date information, such as official websites or documents issued by specialized organizations.

[0092] A "Q&A database" is a database in which questions and their answers are formalized and stored, and each question and answer is categorized by topic.

[0093] A "generative model" is a machine learning model that uses natural language processing technology to generate optimal answers to user questions.

[0094] An "interface" is the means by which users access the system and input questions, and may be provided in the form of a smartphone app or web portal.

[0095] A "terminal" is a device that a user uses to use an interface, such as a smartphone or computer.

[0096] "Natural language processing technology" is a technology for processing human language using a computer, and is a technique for analyzing, understanding, and generating text.

[0097] "Additional related or supporting information" is supplemental content, such as more detailed information or advice, that accompanies the answer to the user's question.

[0098] The present invention relates to a system for providing information related to life in an evacuation shelter using a generative AI model. Specific embodiments for carrying out the present invention will be described below.

[0099] Overall system picture

[0100] The system is based on a generative AI model running on the server side and has a user-accessible interface. The server collects data from reliable sources and builds a Q&A database categorized by topic. This database is used to train the generative AI model, which generates optimal answers to user questions. The system also has a function for sending and displaying answers to the user's device.

[0101] Data collection and registration

[0102] The server collects data related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. Specifically, it automatically obtains the latest information using RSS feeds and APIs. The collected data is categorized by topic, organized into a Q&A format, and entered into a database. This database is divided into themes such as health, education, and psychological support.

[0103] Training a generative model

[0104] The server uses the constructed Q&A database to train a generative AI model. One example of a generative AI model used is GPT-4. This model utilizes natural language processing technology, enabling it to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0105] Providing a user interface

[0106] The user interface is provided in a form that is easy for users to access (e.g., smartphone app, web portal), using technologies such as React Native and React.js. The interface includes a question input form and options by category, making it intuitive to use.

[0107] Question and Answer Process

[0108] When a user enters a question into the interface and presses the send button, the device sends the question to the server. The server uses a generative AI model to generate the best answer for the received question. The answer may include additional related or supporting information as needed. The generated answer is then displayed to the user through the device.

[0109] Specific examples

[0110] Example 1: Questions about breastfeeding at evacuation shelters

[0111] User: The user types a question: "How should breastfeeding be handled in an evacuation shelter?"

[0112] Terminal: The terminal sends a question to the server.

[0113] Server: The server uses a generative AI model to analyze the question and generate an appropriate answer, such as "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, with as much privacy as possible. If necessary, ask the evacuation center manager to provide separate spaces."

[0114] Terminal: The terminal displays the generated answer to the user.

[0115] Example 2: Questions about psychological stress

[0116] User: The user types a question: "Please tell me how to take care of myself mentally while living in an evacuation shelter."

[0117] Terminal: The terminal sends a question to the server.

[0118] Server: The server uses a generative AI model to generate optimal answers, such as, "Regular rest and relaxation are important when living in an evacuation shelter. It's a good idea to incorporate relaxation techniques and mindfulness exercises. If you're experiencing significant psychological stress, we recommend seeking professional counseling."

[0119] Terminal: The terminal displays the generated answer to the user.

[0120] In this way, the system of the present invention allows users to quickly and accurately obtain information about life in a shelter, thereby improving their quality of life.

[0121] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0122] Step 1: Data collection step

[0123] The server automatically collects data related to shelter life from reliable sources (e.g., official websites, documents issued by specialized organizations), using RSS feeds and APIs to obtain the latest information.

[0124] Input: Source URL or API endpoint

[0125] Specific operation: The server sends an "HTTP request" to obtain data from the information source.

[0126] Output: Retrieved data (e.g. HTML, PDF, JSON)

[0127] Step 2: Data classification and registration step

[0128] The server categorizes the collected data by topic, organizes it into a Q&A format, and registers it in a database. Data categorization is performed using tagging technology.

[0129] Input: Collected data

[0130] How it works: The server uses natural language processing technology to analyze the data and classify it into categories such as "health," "education," and "psychological support." The classified data is then converted into a Q&A format and stored in a database.

[0131] Output: Q&A database

[0132] Step 3: Data preparation steps for the generative model

[0133] The server extracts data from the Q&A database and formats it as a training dataset for the generative model.

[0134] Input: Q&A database

[0135] What happens: The server extracts the Q&A data from the database and converts it into an appropriate format (e.g., JSON).

[0136] Output: Training dataset

[0137] Step 4: Training the generative model

[0138] The server uses the extracted data to train a generative model, such as GPT-4.

[0139] Input: Training dataset

[0140] Specific operation: The server executes a Python script and performs processing to train a generative model (GPT-4).

[0141] Output: A trained generative model

[0142] Step 5: Model Update Step

[0143] The server periodically retrieves new data and retrains the model with the updated data.

[0144] Input: New data and existing Q&A database

[0145] What happens: The server integrates the newly collected data with the existing database and retrains the generative model.

[0146] Output: Updated generative model

[0147] Step 6: Interface development step

[0148] The server develops the user interface in a form that is easy for users to access, using technologies such as React Native and React.js.

[0149] Input: Interface design and functional requirements

[0150] What it does: A developer uses React.js to build a web portal and adds a question form and options based on categories.

[0151] Output: The finished user interface

[0152] Step 7: Interface Operation Steps

[0153] The server monitors the operational interfaces and performs maintenance and updates as needed.

[0154] Input: User feedback and operational data

[0155] What it does: The server periodically collects user feedback and makes bug fixes and feature improvements.

[0156] Output: User interface in operation

[0157] Step 8: Question input step

[0158] The user accesses the interface and enters a question. The question entry form includes fields for selecting a category and entering specific questions.

[0159] Input: User question

[0160] What happens: The user types into the interface, "How should breastfeeding be done in an evacuation shelter?"

[0161] Output: Question data

[0162] Step 9: Submit a question

[0163] The device sends the questions entered by the user to the server, and the data is sent in JSON format.

[0164] Input: User question data

[0165] Specific operation: The device sends a POST request to the API endpoint.

[0166] Output: Query data to the server

[0167] Step 10: Question Analysis Step

[0168] The server analyzes the received question and formats it as data to be input into the generative model.

[0169] Input: Received question data

[0170] Specific behavior: The server parses and preprocesses the received query (e.g., tokenization and normalization).

[0171] Output: Well-formed question data

[0172] Step 11: Answer generation step

[0173] The server uses the generative model to generate optimal answers to questions.

[0174] Input: Well-formed question data

[0175] How it works: The server inputs question data into the generative model and generates an answer, such as, "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible. If necessary, ask the evacuation center manager to provide a separate space."

[0176] Output: Response data

[0177] Step 12: Send response step

[0178] The server then sends the generated response to the user's device. The data is sent in JSON format.

[0179] Input: Response data

[0180] Specific operation: The server sends a POST request to the API endpoint and delivers the response to the device.

[0181] Output: Answer data to the user's device

[0182] Step 13: Answer display step

[0183] The terminal displays the received response to the user, and the interface formats and displays the received data.

[0184] Input: Received response data

[0185] Specific operation: The device displays the received response on the user interface, saying, "Breastfeeding at the evacuation center will be conducted in the same way as normal breastfeeding, and privacy will be ensured as much as possible. If necessary, ask the evacuation center administrator to provide a separate space."

[0186] Output: Answers displayed on many users' terminals

[0187] (Application example 1)

[0188] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0189] Conventional information provision systems for life in evacuation shelters often do not have a dedicated platform, making it difficult for users to quickly and accurately obtain the information they need. Furthermore, in emergencies and disasters, there is a need to provide security information and specific advice on evacuation routes, but no system exists that can meet this need.

[0190] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0191] In this invention, the server includes: means for providing information on life in a shelter using a generative model; means for building a themed Q&A database; means for loading the Q&A database into a generative model and customizing it to a specific theme; means for providing an interface for accepting questions from users; means for generating optimal answers to user questions using the generative model; means for transmitting the generated answers to the user's terminal; means for acquiring disaster safety information such as security information from a reliable information source in JSON format; and means for the generative model to provide the necessary answers based on the acquired data and support evacuation behavior in an emergency, thereby enabling users to obtain quick and accurate information via their smartphones or other terminals.

[0192] A "generative model" is an algorithm that generates new information or answers based on the data provided.

[0193] "Shelter life" refers to the living environment and lifestyle in a temporary shelter set up in the event of a disaster or emergency.

[0194] An "information providing system" is a computer system for providing necessary information to users.

[0195] A "topical Q&A database" is a database that organizes and stores data in the form of questions and answers for specific themes.

[0196] An "interface" is the means or screen through which a user interacts with a computer system or application.

[0197] "Natural language processing" is a technology that uses computers to analyze, process, and generate natural language used by humans.

[0198] "Security information" refers to various information provided to ensure the safety of users, and particularly includes information in emergencies and disasters.

[0199] The "JSON format" is a common format for expressing data in text format and is one of the formats widely used for data exchange.

[0200] System Overview

[0201] The present invention relates to a system that uses generative models to provide information on life in evacuation shelters and security during emergencies. The system is based on a generative model executed on the server side and has an interface that is easily accessible to users.

[0202] Data collection and registration

[0203] The server collects information related to shelter life and security in JSON format from reliable sources, then categorizes the collected data by topic, organizes it into a Q&A format, and registers it in a database that covers topics such as health, education, psychological support, and security information.

[0204] Training a generative model

[0205] The server uses the constructed Q&A database to train a generative model. This generative model utilizes natural language processing technology to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0206] User Interface

[0207] The user interface is provided as a smartphone app or web portal. The interface is intuitive and includes a question input form and category-based options. Users can enter their questions through this interface and immediately obtain the information they need.

[0208] Question and Answer Process

[0209] When a user enters a question and presses the send button, the device sends the question to the server, which uses the generative model to generate the best answer to the question and sends it back to the user's device. This answer may also include additional relevant or security information, if necessary.

[0210] Specific examples

[0211] Example 1: Selection of evacuation routes

[0212] User: The user types a question: "What is the best evacuation route in case of a flood?"

[0213] Terminal: The terminal sends a question to the server.

[0214] Server: The server uses the generative model to analyze the question and generate an appropriate answer, such as "In the event of a flood, it is best to evacuate to higher ground. The nearest evacuation shelter is the one on the higher ground in XX Park."

[0215] Terminal: The terminal displays the generated answer to the user.

[0216] Example 2: Emergency contact method

[0217] User: The user types a question: "How do I contact my family in case of an earthquake?"

[0218] Terminal: The terminal sends a question to the server.

[0219] Server: The server uses the generative model to generate the optimal answer. For example, it might generate an answer such as, "Since congestion is expected during an earthquake, it's best to communicate using text messages or social media. It's also a good idea to decide on a meeting place at an evacuation shelter."

[0220] Terminal: The terminal displays the generated answer to the user.

[0221] Hardware and software used

[0222] This system mainly performs data processing on the server side. The server requires a high-performance processor and a large amount of memory, so it is recommended to use a cloud service (e.g., Amazon Web Services, Google Cloud Platform). To run the generative model, we use a library specifically designed for natural language processing (e.g., Transformers by Hugging Face).

[0223] Prompt Sentence Examples

[0224] Here are some examples of prompts:

[0225] "What is the safest evacuation route in the event of a flood?"

[0226] "What is the best way to contact my family in the event of an earthquake?"

[0227] "Please tell me how to take care of your mental health while living in an evacuation shelter."

[0228] As described above, the present invention allows users to quickly and accurately obtain necessary information in the event of an emergency or disaster.

[0229] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0230] Step 1:

[0231] The server collects information related to shelter life and security from reliable sources in JSON format. This allows it to obtain the necessary data from multiple official websites and documents issued by specialized organizations. The input is the URL of each source, and the output is the retrieved JSON data. The server then organizes the information into a database.

[0232] Step 2:

[0233] The server categorizes the collected data by theme, formats it in Q&A format, and registers it in a database. The input is the JSON data obtained in step 1, and the output is a database organized in Q&A format. The server uses natural language processing technology to categorize each piece of information by theme.

[0234] Step 3:

[0235] The server trains a generative model using a Q&A database it has built. The input is the Q&A database, and the output is the trained generative model. The server uses libraries (e.g., Hugging Face Transformers) to build a generative model customized for a specific theme.

[0236] Step 4:

[0237] A user inputs a question through a smartphone app or web portal. The input is the user's question, and the output is the question data. The device provides the user interface, and the user intuitively operates it to input the question.

[0238] Step 5:

[0239] The terminal sends the user's query data to the server. The input is the query data entered by the user in step 4, and the output is the query data sent to the server. The terminal sends the data to the server via the network.

[0240] Step 6:

[0241] The server uses a generative model to generate the best answer to the received question. The input is the user's question data and a Q&A database, and the output is the generated answer. The server uses natural language processing technology to have the model analyze the user's question and generate an answer based on related information.

[0242] Step 7:

[0243] The server sends the generated answer to the user's terminal. The input is the generated answer, and the output is the answer sent to the user's terminal. The server sends the answer data to the terminal via the network, and the terminal receives it.

[0244] Step 8:

[0245] The terminal displays the received answer to the user. The input is the answer data received from the server, and the output is the answer displayed to the user. The terminal displays the answer through a user interface, and the user confirms it.

[0246] As a result, the user can quickly obtain accurate information.

[0247] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0248] The present invention relates to a system that uses a generative model to provide information related to life in a shelter, and by combining it with an emotion engine, provides optimal information according to the user's emotional state. Specific embodiments of this system are described below.

[0249] System Overview

[0250] This system is based on a generative model, emotion engine, and user interface running on the server side. The server collects data from reliable sources and builds a topic-specific Q&A database. The generative model learns from this database and provides optimal answers to user questions. The emotion engine also analyzes the user's emotions and appropriately adjusts the generated answers based on those emotions.

[0251] System configuration

[0252] Data collection and registration

[0253] The server collects information related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. The collected data is then categorized by topic, organized into a Q&A format, and registered in a database. Topics include health, education, and psychological support.

[0254] Training a generative model

[0255] The server uses the constructed Q&A database to train a generative model. The generative model utilizes natural language processing technology and is able to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0256] Emotion Engine Integration

[0257] The emotion engine analyzes text entered through the user interface and recognizes the user's emotions. For example, it can detect emotions such as gratitude, anger, and anxiety from specific keywords and contexts contained in the user's text.

[0258] User Interface

[0259] The user interface is provided in a form that is easily accessible to users (e.g., a smartphone app or web portal). The interface is intuitive and includes a question input form, category selection menu, etc. It also provides an optional field for inputting user emotions.

[0260] Question and Answer Process

[0261] When a user enters a question and presses the send button, the device sends the question to the server. The server uses the generative model to instantly generate the best answer for the received question, adjusting the answer according to the user's emotions as recognized by the emotion engine. For example, if the user is feeling anxious, the answer may be more polite and reassuring. The answer may also include additional related or supporting information as needed.

[0262] Specific examples

[0263] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[0264] User: The user types, "How should I breastfeed at the evacuation shelter? I'm worried."

[0265] Device: The device sends the question and emotional expression to the server.

[0266] Server: The server uses a generative model to analyze the question and generate an appropriate answer. For example, it might generate an answer such as, "Breastfeeding at the evacuation center should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible. If necessary, ask the evacuation center manager to provide a separate space." The emotion engine recognizes the user's "worry" emotion and adds a message such as, "Don't worry, we'll support you."

[0267] Terminal: The terminal displays the generated answer to the user.

[0268] Example 2: Questions about psychological stress and emotion recognition

[0269] User: The user types, "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious."

[0270] Device: The device sends the question and emotional expression to the server.

[0271] Server: The server uses a generative model to generate the optimal answer. For example, it might generate an answer such as, "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness practice. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The emotion engine recognizes the emotion "very anxious" and adds a message to the answer such as, "We understand your anxiety. Please don't suffer alone."

[0272] Terminal: The terminal displays the generated answer to the user.

[0273] In this way, by combining a generative model and an emotion engine, the system of the present invention can provide appropriate information according to the user's emotional state, thereby improving the quality of information acquisition during life in an evacuation shelter and supporting the user's daily life.

[0274] The processing flow will be explained below.

[0275] Step 1: Collect and organize your data

[0276] The server collects information about life in evacuation shelters from reliable sources (e.g., official websites, documents issued by specialized organizations).

[0277] The server categorizes the collected information by topic, for example into categories such as health, education, and psychological support.

[0278] The server creates specific question-and-answer pairs for each topic, building a Q&A format database.

[0279] Step 2: Training the generative model

[0280] The server trains the generative model using a Q&A database.

[0281] The server applies the trained generative model, enabling the model to generate the best answer to the user's question.

[0282] Training is conducted regularly and new information and data is added.

[0283] Step 3: Training and Integrating the Emotion Engine

[0284] The server trains the emotion engine using the user's text data.

[0285] The server integrates the emotion engine with the generative model to recognize emotions from user questions.

[0286] The emotion engine includes algorithms that detect emotions from specific keywords and contexts.

[0287] Step 4: Prepare the User Interface

[0288] The device provides a user-friendly interface, accessible through a smartphone app or web portal.

[0289] The terminal provides a question input form and a category selection menu, allowing users to easily input questions.

[0290] The device provides an emotion input option, allowing the user to explicitly input emotions.

[0291] Step 5: Receiving user questions

[0292] The user enters a question and emotion through the interface and presses the send button.

[0293] The terminal transmits the user's question data and emotion data to the server.

[0294] Step 6: Question and Sentiment Analysis

[0295] The server analyzes the received question data and emotion data.

[0296] The server passes the question data to the generative model to generate the optimal answer.

[0297] The server passes the emotion data to the emotion engine, which analyzes the user's emotional state.

[0298] Step 7: Adjust your responses based on emotion

[0299] The server adjusts the generated answer based on the emotion recognized by the emotion engine.

[0300] For example, if the user is feeling anxious, add reassuring language to the answer.

[0301] Step 8: Submit your response

[0302] The server transmits the generated answer and the adjusted answer based on the emotion to the user's terminal.

[0303] The terminal displays the received answer to the user.

[0304] Step 9: Gather user feedback

[0305] The terminal collects user feedback and sends it to the server.

[0306] The server analyzes the collected feedback and uses it to train the generative model and emotion engine next time.

[0307] Specific examples

[0308] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[0309] Step 5:

[0310] The user types, "How should I breastfeed at an evacuation shelter? I'm worried," and presses the send button.

[0311] Step 6:

[0312] The terminal transmits the question and emotion data to the server.

[0313] The server passes the question data to a generative model to generate the optimal answer.

[0314] The server passes the emotional data "I'm worried" to the emotion engine, which then recognizes the emotion.

[0315] Step 7:

[0316] The server generates an answer: "Breastfeeding in evacuation shelters should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible." The answer is then adjusted using the emotion engine.

[0317] The user is worried, so you add "Don't worry, we're here to help" to your answer.

[0318] Step 8:

[0319] The server sends the adjusted response to the user's terminal.

[0320] The terminal displays the generated answer to the user.

[0321] Example 2: Questions about psychological stress and emotion recognition

[0322] Step 5:

[0323] The user types, "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious," and presses the send button.

[0324] Step 6:

[0325] The terminal transmits the question and emotion data to the server.

[0326] The server passes the question data to a generative model to generate the optimal answer.

[0327] The server passes the emotional data "I'm very anxious" to the emotion engine, which then recognizes the emotion.

[0328] Step 7:

[0329] The server generates an answer: "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness exercises. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The answer is then adjusted using the emotion engine.

[0330] The user is feeling very anxious, so add "We understand your anxiety. Please don't go through it alone" to your response.

[0331] Step 8:

[0332] The server sends the adjusted response to the user's terminal.

[0333] The terminal displays the generated answer to the user.

[0334] In this way, by combining a generative model and an emotion engine, the system of the present invention can provide appropriate information according to the user's emotional state, thereby improving the quality of information acquisition during life in an evacuation shelter and supporting the user's daily life.

[0335] Example 2

[0336] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0337] When living in an evacuation shelter, there is a need for a method that allows users to easily obtain the information they need. In particular, it is important to provide optimal information according to the user's emotional state. However, conventional systems have difficulty providing information that reflects the user's emotions, and have not been able to provide sufficient psychological support during evacuation shelter life.

[0338] The identification process 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 providing information on life in a shelter using a generative model, means for constructing a thematic information base, means for loading the information base into a generative model and customizing it to a specific theme, means for providing a user interface for accepting questions from users, means for generating optimal answers to the user's questions using the generative model, means for transmitting the generated answers to the user's information device, and means for analyzing the user's emotions using an emotion analysis engine and reflecting the emotions in the answers. This makes it possible to provide optimal information according to the user's emotional state.

[0339] 1. "Generative model" is an artificial intelligence technique for generating natural language responses from collected data.

[0340] 2. The "Thematic Information Base" is a database constructed by classifying information related to life in evacuation shelters by theme.

[0341] 3. "Customization" means tailoring information to a specific topic or user needs and converting it into an appropriate format.

[0342] 4. "User interface" refers to interactive mechanisms such as input forms and menu screens that allow users to interact with a system.

[0343] 5. "User's information equipment" refers to electronic devices used by the User, such as smartphones, tablets, and personal computers.

[0344] 6. "Sentiment analysis engine" is software that analyzes text entered by a user and identifies the emotions contained therein.

[0345] 7. "Natural language processing" is a technology that allows computers to understand and generate human language.

[0346] 8. "Life in a shelter" refers to a situation in which people temporarily live in a shelter due to a disaster or other reason.

[0347] The present invention is a system that provides information related to life in an evacuation shelter, and by combining a generative model and an emotion analysis engine, it is possible to provide optimal information according to the user's emotional state. This system has the following configuration.

[0348] Data collection and database construction

[0349] The server collects data on life in evacuation shelters from government agencies and other reliable sources. The collected data is categorized by theme, such as "health," "education," and "psychological support." The server then organizes the data into a Q&A format and registers it in a database as a thematic information base. This database is managed using a relational database management system such as MySQL.

[0350] Training a generative model

[0351] The server uses the thematic information base to train a generative model (such as GPT-3). This generative model utilizes natural language processing techniques and is capable of generating optimal answers to user questions. Model training is performed using machine learning frameworks such as TensorFlow and PyTorch. The model is regularly updated to reflect the latest information.

[0352] Sentiment analysis engine integration

[0353] The sentiment analysis engine analyzes the text entered by the user through the user interface and recognizes the user's emotions. For example, the sentiment analysis engine identifies emotions such as "joy," "sadness," "anger," and "anxiety" from specific keywords and context. NLP libraries (Natural Language Toolkit and SpaCy) are used for this analysis.

[0354] Providing a user interface

[0355] The user interface is provided in the form of a smartphone app or web portal that is easily accessible to users. This interface is built using React Native and React.js and includes a question input form, a category selection menu, and an optional field for inputting user sentiment.

[0356] Question and Answer Process

[0357] When a user enters a question and presses the send button, the device sends the question and emotional expression to the server. The server uses a generative model to generate an optimal answer to the received question and adjusts the answer according to the user's emotions recognized by the emotion analysis engine. For example, if the user is feeling anxious, the server adds a reassuring message to the answer, such as "Don't worry, appropriate measures are being taken."

[0358] Specific examples

[0359] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[0360] User: Type "How should I breastfeed at the evacuation shelter? I'm worried."

[0361] Device: Sends questions and emotional expressions to the server.

[0362] Server: Using a generative model, it generates answers such as, "Breastfeeding at the evacuation center should be handled as normal, and privacy should be maintained as much as possible. If necessary, please ask the evacuation center manager to provide a separate space for the baby." The sentiment analysis engine recognizes the emotion "worry" and adds messages such as, "Don't worry, we're here to support you."

[0363] Terminal: Displays the generated answer to the user.

[0364] Example 2: Questions about psychological stress and emotion recognition

[0365] User: Type "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious."

[0366] Device: Sends questions and emotional expressions to the server.

[0367] Server: Using a generative model, the server generates answers such as, "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness practice. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The sentiment analysis engine recognizes the emotion "very anxious" and adds messages such as, "We understand your anxiety. Please don't bear it alone."

[0368] Terminal: Displays the generated answer to the user.

[0369] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0370] Step 1: Data collection and database registration

[0371] The server periodically crawls data related to shelter life from government agencies and reliable websites, using web scraping technology to collect reliable information.

[0372] The server automatically categorizes the crawled data into themes, such as "health," "education," and "psychological support."

[0373] The server converts the classified data into Q&A format and registers it in a MySQL database. The input data is the collected text information, and the output data is a database organized in Q&A format.

[0374] Step 2: Training the generative model

[0375] The server trains a generative model (such as GPT-3) using a Q&A dataset stored in the database. This process uses natural language processing techniques, and the model is trained to generate appropriate answers to questions.

[0376] The server evaluates the training results and, if they are insufficient, adds data and performs training again.,The input data is a Q&A format database, and the output data is,the trained generative model.

[0377] The server periodically adds new information to the database and updates the generative model.

[0378] Step 3: Integrating a sentiment analysis engine

[0379] The server embeds a sentiment analysis engine into the user interface, where a sentiment analysis algorithm analyzes the text data to identify sentiment.

[0380] The server analyzes the text entered by the user in real time and identifies emotions such as "joy," "sadness," "anger," and "anxiety" based on keywords and context. The input data is the text entered by the user, and the output data is the result of identifying the emotion.

[0381] Test the accuracy of your sentiment analysis engine and fine-tune the algorithm as needed.

[0382] Step 4: Providing a User Interface

[0383] The server provides users with smartphone apps using React Native and web portals using React.js.

[0384] The terminal displays a question input form and a category selection menu to the user. The input data is the user's operation information, and the output data is the interface display content.

[0385] The user enters a question and enters an emotion in the emotion input field.

[0386] Step 5: Question-answering process

[0387] The user enters a question and presses the send button. For example, "How should I breastfeed at an evacuation shelter? I'm worried."

[0388] The terminal sends the input question and emotional expression to the server. The input data is the user's question and emotional information, and the output data is the information to be sent to the server.

[0389] The server analyzes the question using a generative model and generates an optimal answer. For example, it generates an answer such as, "Breastfeeding at evacuation shelters should be done in the same way as normal breastfeeding, and privacy should be maintained as much as possible. If necessary, please ask the shelter manager to provide a separate space." The input data is the user's question text, and the output data is the generated answer.

[0390] The server uses a sentiment analysis engine to tailor the generated answer depending on the user's emotions, for example adding a message like "Don't worry, we're here to help" if the user is feeling anxious.

[0391] The server sends the final answer to the terminal. The input data is the adjusted answer, and the output data is the information to be sent to the terminal.

[0392] The device will then display the received response to the user, such as, "Breastfeeding at the evacuation shelter should be done in the same way as normal breastfeeding, and please ensure privacy as much as possible. If necessary, please ask the shelter manager to provide a separate space. Don't worry, we'll support you."

[0393] (Application example 2)

[0394] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0395] Conventional food delivery systems only suggest generic menus without considering the user's emotional state. As a result, they are unable to suggest appropriate menus that reflect the user's psychological state, making it difficult to sufficiently improve satisfaction. There is a need for a system that can solve this problem and make highly personalized menu suggestions that correspond to the user's emotional state.

[0396] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for providing information on life in a shelter using a generative model, means for constructing a themed Q&A database, means for loading the Q&A database into a generative model and customizing it to a specific theme, means for providing an interface for accepting questions from users, means for generating optimal answers to user questions using the generative model, means for transmitting the generated answers to the user's terminal, means for analyzing the user's emotions using an emotion engine, and means for adjusting the generated answers depending on the user's emotional state. This makes it possible to propose optimal menus depending on the user's emotional state.

[0397] - A "generative model" is an algorithm that learns from large amounts of data and generates appropriate outputs for given inputs.

[0398] The "Topical Q&A Database" is a database that contains questions and answers related to different themes.

[0399] An "interface" is a means for exchanging data between a user and a system.

[0400] An "emotion engine" is an analysis system that analyzes emotions from user input and recognizes specific emotional states.

[0401] "Analysis" is the process of examining data and understanding its structure and meaning.

[0402] "Terminal" refers to a device through which a user inputs and receives information.

[0403] "Food delivery" refers to the general service of ordering and having food delivered.

[0404] "Menu suggestions" refers to presenting appropriate meal options to the user.

[0405] "Personalization" means customizing something individually to suit a specific user.

[0406] System configuration

[0407] The present invention includes a generative model, an emotion engine, a user interface, and a server-based system for controlling them.

[0408] Collect data from reliable sources and build a thematic Q&A database.

[0409] The generative model learns from this database and generates optimal answers to user questions.

[0410] The emotion engine analyzes the user's input text and recognizes the emotional state.

[0411] The user interface is provided as a smartphone app, allowing intuitive user interaction.

[0412] Program processing

[0413] 1. Data Collection:

[0414] The server automatically collects information related to shelter life and food delivery from official websites and documents issued by specialized organizations.

[0415] The collected data is categorized by theme and registered in a database in Q&A format.

[0416] 2. Training the generative model:

[0417] The server uses the constructed Q&A database to train a generative model (e.g., OpenAI's GPT-3).

[0418] The model will be updated periodically to reflect new information.

[0419] 3. Emotion engine integration:

[0420] The emotion engine analyzes the user's input text and identifies an emotional state.

[0421] For example, it can recognize emotions such as "fatigue" and "stress" from specific keywords and context within the text.

[0422] 4. User Interface:

[0423] It is provided as a smartphone app, allowing users to enter and submit questions.

[0424] The app includes a question form, a category selection menu, and an emotion input option.

[0425] 5. Question and Answer Process:

[0426] When a user inputs a question and presses the send button, the terminal sends the question to the server.

[0427] The server uses a generative model to generate optimal answers and adjusts the answers according to the user's emotional state, as recognized by the emotion engine.

[0428] The generated answer is sent to the user's terminal and displayed.

[0429] Specific examples

[0430] 1. Example 1: Menu suggestions for fatigued users

[0431] User: Type, "I'm feeling tired today, what foods will make me feel better?"

[0432] Server: The emotion engine recognizes the emotion "fatigue," and the generative model generates an appropriate answer: "To recover from fatigue, we recommend fruits rich in vitamin C and lean meat, which contains iron. Also, don't forget to get adequate rest."

[0433] 2. Example 2: Menu suggestions for stressed users

[0434] User: Type "I'm stressed and want some food to help me relax."

[0435] Server: The emotion engine recognizes the emotion "stress," and the generative model generates an appropriate answer: "Herbal tea and yogurt are recommended for relieving stress. Dark chocolate also helps you relax."

[0436] Example prompt sentence:

[0437] 1. "User Question: I'm feeling tired today. What foods will make me feel better?"

[0438] "User question: I'm tired today, what foods will make me feel better?\nUser emotion: Fatigue\nGenerate the best answer."

[0439] 2. "User Question: I'm stressed and want some food to help me relax."

[0440] "User question: I'm stressed and want some food to help me relax.\nUser emotion: Stress\nGenerate the best answer."

[0441] In this way, the invention combines an emotion engine and a generative model to provide optimal information according to the user's emotional state.

[0442] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0443] Step 1:

[0444] The server automatically collects information related to shelter life and food delivery from official websites and documents issued by specialized organizations. It uses APIs or scraping technology to extract data and categorize it by topic. The collected data is then registered in a database. The input is data obtained from reliable sources, and the output is a database in Q&A format, categorized by topic.

[0445] Step 2:

[0446] The server uses the constructed Q&A database to train a generative model (e.g., OpenAI's GPT-3). In this process, the generative model learns from the data using natural language processing so that it can generate appropriate answers to questions. The input is the information in the Q&A database, and the output is the trained generative model.

[0447] Step 3:

[0448] The server uses an emotion engine to analyze emotions from the user's input text. Specifically, the emotion engine analyzes specific keywords and contexts in the text and identifies the emotional state as "fatigue" or "stress," etc. The input is the user's input text, and the output is the analyzed emotional information.

[0449] Step 4:

[0450] The user interface (smartphone app) accepts questions from users. Questions are entered into an input form and sent to the server by pressing the send button. The input is the user's question text, and the output is the question data sent to the server.

[0451] Step 5:

[0452] When the server receives a user's question, it uses a generative model to generate an optimal answer. It then adjusts the answer based on the user's emotional state, as recognized by the emotion engine. For example, if the user is feeling tired, the answer may include specific suggestions such as "foods that will help recover from fatigue." The input is the user's question data and emotional information, and the output is the generated answer.

[0453] Step 6:

[0454] The generated answer is sent from the server to the user's device (smartphone). The user's device displays the received answer. The input is the generated answer data from the server, and the output is the answer displayed on the user's device.

[0455] Specific actions

[0456] Step 1: For example, the server scrapes health information from the official WHO website and categorizes it by topic, such as "health" or "psychological support."

[0457] Step 2: Using the GPT-3 model, we train the model by learning from the collected Q&A data on themes such as "health" and "psychological support."

[0458] Step 3: The emotion engine analyzes the emotion “fatigue” from the text entered by the user: “I feel tired today, please tell me what foods will make me feel better.”

[0459] Step 4: The user interface receives and sends the question "I'm stressed and want some food to help me relax" from the user.

[0460] Step 5: The server generates a response to the user in the "stressed" state, suggesting "herbal tea and yogurt" as "foods that help relieve stress."

[0461] Step 6: The generated answer is displayed in the smartphone app for the user to review.

[0462] In this way, by using the trained generative model and emotion engine, personalized food menu suggestions that are in line with the user's emotional state can be realized.

[0463] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0464] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0465] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0466] [Second embodiment]

[0467] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0468] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0469] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0470] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0471] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0472] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0473] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0474] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0475] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0476] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0477] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0478] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0479] The present invention relates to a system for providing information related to life in an evacuation shelter using a generative model. Specific embodiments of the system are described below.

[0480] System Overview

[0481] The system is based on a server-side generative model and provides a user-friendly interface. The server collects data from reliable sources and builds a topic-specific Q&A database. The generative model learns from this database and provides optimal answers to user questions.

[0482] System configuration

[0483] Data collection and registration

[0484] The server collects information related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. The collected data is then categorized by topic, organized into a Q&A format, and registered in a database. Topics include health, education, and psychological support.

[0485] Training a generative model

[0486] The server uses the constructed Q&A database to train a generative model. The generative model utilizes natural language processing technology and is able to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0487] User Interface

[0488] The user interface is provided in a form that is easily accessible to users (e.g., smartphone app, web portal), and includes a question input form and options by category, making it intuitive to operate.

[0489] Question and Answer Process

[0490] When a user enters a question and presses the send button, the device sends the question to the server, which uses the generative model to instantly generate the best answer to the received question and sends it back to the user's device. This answer may also include additional related or supporting information as needed.

[0491] Specific examples

[0492] Example 1: Questions about breastfeeding at evacuation shelters

[0493] User: The user types a question: "How should breastfeeding be handled in an evacuation shelter?"

[0494] Terminal: The terminal sends a question to the server.

[0495] Server: The server uses the generative model to analyze the question and generate an appropriate answer, such as "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, with as much privacy as possible. If necessary, ask the evacuation center manager to provide separate spaces."

[0496] Terminal: The terminal displays the generated answer to the user.

[0497] Example 2: Questions about psychological stress

[0498] User: The user types a question: "Please tell me how to take care of myself mentally while living in an evacuation shelter."

[0499] Terminal: The terminal sends a question to the server.

[0500] Server: The server uses the generative model to generate the optimal answer, such as, "Regular rest and relaxation are important when living in an evacuation shelter. It is a good idea to incorporate relaxation techniques and mindfulness practice. Also, if you are experiencing severe psychological stress, we recommend that you seek professional counseling."

[0501] Terminal: The terminal displays the generated answer to the user.

[0502] In this way, the system of the present invention allows users to quickly and accurately obtain information about life in a shelter, thereby improving their quality of life.

[0503] The processing flow will be explained below.

[0504] Step 1: Collect and organize your data

[0505] The server collects information about life in evacuation shelters from reliable sources (e.g., official websites, documents issued by specialized organizations).

[0506] The server categorizes the collected information by topic, for example into categories such as health, education, and psychological support.

[0507] The server creates specific question-and-answer pairs for each topic, building a Q&A format database.

[0508] Step 2: Training the generative model

[0509] The server trains the generative model using a Q&A database.

[0510] The server applies the trained generative model, enabling the model to generate the best answer to the user's question.

[0511] Training is conducted regularly and new information and data is added.

[0512] Step 3: Prepare the user interface

[0513] The device provides a user-friendly interface, accessible through a smartphone app or web portal.

[0514] The terminal provides a question input form and a category selection menu, allowing users to easily input questions.

[0515] Step 4: Receiving questions from users

[0516] The user enters a question through the interface and presses the submit button.

[0517] The terminal transmits the user's question data to the server.

[0518] Step 5: Parsing the question and generating an answer

[0519] The server analyzes the received question data and passes it to the generative model.

[0520] The server uses the generative model to generate optimal answers to the user's questions.

[0521] If necessary, the server will search for additional information (e.g., related resources or supporting information) and append it to the answer.

[0522] Step 6: Submit your response

[0523] The server sends the generated answer and any additional information required to the user's terminal.

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

[0525] Step 7: Gather user feedback

[0526] The terminal collects user feedback and sends it to the server.

[0527] The server analyzes the collected feedback and uses it to train the next generative model.

[0528] Through this series of steps, the system of the present invention can quickly and accurately provide information about life in an evacuation shelter, thereby contributing to improving the lives of users.

[0529] Example 1

[0530] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0531] Obtaining information during evacuation shelter life is difficult, and one of the challenges is the lack of timely provision of appropriate information tailored to individual needs and problems. This is due to limited resources, issues with the reliability of information sources, and the difficulty of accurately categorizing and providing information. Furthermore, a system that combines high accuracy and versatility is required to address a wide range of information needs, including those related to psychological stress and health issues.

[0532] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0533] In this invention, the server includes means for collecting data from reliable information sources, means for constructing a Q&A database categorized by theme, means for loading the Q&A database into a generative model to learn information, means for training the generative model to generate optimal answers to user questions, means for providing an interface for accepting questions from users, means for transmitting the generated answers to the user's terminal, and means for displaying the generated answers on the user's terminal, thereby enabling users to quickly and accurately obtain a wide range of information related to life in a shelter.

[0534] "Reliable sources" are institutions or platforms that provide accurate and up-to-date information, such as official websites or documents issued by specialized organizations.

[0535] A "Q&A database" is a database in which questions and their answers are formalized and stored, and each question and answer is categorized by topic.

[0536] A "generative model" is a machine learning model that uses natural language processing technology to generate optimal answers to user questions.

[0537] An "interface" is the means by which users access the system and input questions, and may be provided in the form of a smartphone app or web portal.

[0538] A "terminal" is a device that a user uses to use an interface, such as a smartphone or computer.

[0539] "Natural language processing technology" is a technology for processing human language using a computer, and is a technique for analyzing, understanding, and generating text.

[0540] "Additional related or supporting information" is supplemental content, such as more detailed information or advice, that accompanies the answer to the user's question.

[0541] The present invention relates to a system for providing information related to life in an evacuation shelter using a generative AI model. Specific embodiments for carrying out the present invention will be described below.

[0542] Overall system picture

[0543] The system is based on a generative AI model running on the server side and has a user-accessible interface. The server collects data from reliable sources and builds a Q&A database categorized by topic. This database is used to train the generative AI model, which generates optimal answers to user questions. The system also has a function for sending and displaying answers to the user's device.

[0544] Data collection and registration

[0545] The server collects data related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. Specifically, it automatically obtains the latest information using RSS feeds and APIs. The collected data is categorized by topic, organized into a Q&A format, and entered into a database. This database is divided into themes such as health, education, and psychological support.

[0546] Training a generative model

[0547] The server uses the constructed Q&A database to train a generative AI model. One example of a generative AI model used is GPT-4. This model utilizes natural language processing technology, enabling it to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0548] Providing a user interface

[0549] The user interface is provided in a form that is easy for users to access (e.g., smartphone app, web portal), using technologies such as React Native and React.js. The interface includes a question input form and options by category, making it intuitive to use.

[0550] Question and Answer Process

[0551] When a user enters a question into the interface and presses the send button, the device sends the question to the server. The server uses a generative AI model to generate the best answer for the received question. The answer may include additional related or supporting information as needed. The generated answer is then displayed to the user through the device.

[0552] Specific examples

[0553] Example 1: Questions about breastfeeding at evacuation shelters

[0554] User: The user types a question: "How should breastfeeding be handled in an evacuation shelter?"

[0555] Terminal: The terminal sends a question to the server.

[0556] Server: The server uses a generative AI model to analyze the question and generate an appropriate answer, such as "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, with as much privacy as possible. If necessary, ask the evacuation center manager to provide separate spaces."

[0557] Terminal: The terminal displays the generated answer to the user.

[0558] Example 2: Questions about psychological stress

[0559] User: The user types a question: "Please tell me how to take care of myself mentally while living in an evacuation shelter."

[0560] Terminal: The terminal sends a question to the server.

[0561] Server: The server uses a generative AI model to generate optimal answers, such as, "Regular rest and relaxation are important when living in an evacuation shelter. It's a good idea to incorporate relaxation techniques and mindfulness exercises. If you're experiencing significant psychological stress, we recommend seeking professional counseling."

[0562] Terminal: The terminal displays the generated answer to the user.

[0563] In this way, the system of the present invention allows users to quickly and accurately obtain information about life in a shelter, thereby improving their quality of life.

[0564] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0565] Step 1: Data collection step

[0566] The server automatically collects data related to shelter life from reliable sources (e.g., official websites, documents issued by specialized organizations), using RSS feeds and APIs to obtain the latest information.

[0567] Input: Source URL or API endpoint

[0568] Specific operation: The server sends an "HTTP request" to obtain data from the information source.

[0569] Output: Retrieved data (e.g. HTML, PDF, JSON)

[0570] Step 2: Data classification and registration step

[0571] The server categorizes the collected data by topic, organizes it into a Q&A format, and registers it in a database. Data categorization is performed using tagging technology.

[0572] Input: Collected data

[0573] How it works: The server uses natural language processing technology to analyze the data and classify it into categories such as "health," "education," and "psychological support." The classified data is then converted into a Q&A format and stored in a database.

[0574] Output: Q&A database

[0575] Step 3: Data preparation steps for the generative model

[0576] The server extracts data from the Q&A database and formats it as a training dataset for the generative model.

[0577] Input: Q&A database

[0578] What happens: The server extracts the Q&A data from the database and converts it into an appropriate format (e.g., JSON).

[0579] Output: Training dataset

[0580] Step 4: Training the generative model

[0581] The server uses the extracted data to train a generative model, such as GPT-4.

[0582] Input: Training dataset

[0583] Specific operation: The server executes a Python script and performs processing to train a generative model (GPT-4).

[0584] Output: A trained generative model

[0585] Step 5: Model Update Step

[0586] The server periodically retrieves new data and retrains the model with the updated data.

[0587] Input: New data and existing Q&A database

[0588] What happens: The server integrates the newly collected data with the existing database and retrains the generative model.

[0589] Output: Updated generative model

[0590] Step 6: Interface development step

[0591] The server develops the user interface in a form that is easy for users to access, using technologies such as React Native and React.js.

[0592] Input: Interface design and functional requirements

[0593] What it does: A developer uses React.js to build a web portal and adds a question form and options based on categories.

[0594] Output: The finished user interface

[0595] Step 7: Interface Operation Steps

[0596] The server monitors the operational interfaces and performs maintenance and updates as needed.

[0597] Input: User feedback and operational data

[0598] What it does: The server periodically collects user feedback and makes bug fixes and feature improvements.

[0599] Output: User interface in operation

[0600] Step 8: Question input step

[0601] The user accesses the interface and enters a question. The question entry form includes fields for selecting a category and entering specific questions.

[0602] Input: User question

[0603] What happens: The user types into the interface, "How should breastfeeding be done in an evacuation shelter?"

[0604] Output: Question data

[0605] Step 9: Submit a question

[0606] The device sends the questions entered by the user to the server, and the data is sent in JSON format.

[0607] Input: User question data

[0608] Specific operation: The device sends a POST request to the API endpoint.

[0609] Output: Query data to the server

[0610] Step 10: Question Analysis Step

[0611] The server analyzes the received question and formats it as data to be input into the generative model.

[0612] Input: Received question data

[0613] Specific behavior: The server parses and preprocesses the received query (e.g., tokenization and normalization).

[0614] Output: Well-formed question data

[0615] Step 11: Answer generation step

[0616] The server uses the generative model to generate optimal answers to questions.

[0617] Input: Well-formed question data

[0618] How it works: The server inputs question data into the generative model and generates an answer, such as, "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible. If necessary, ask the evacuation center manager to provide a separate space."

[0619] Output: Response data

[0620] Step 12: Send response step

[0621] The server then sends the generated response to the user's device. The data is sent in JSON format.

[0622] Input: Response data

[0623] Specific operation: The server sends a POST request to the API endpoint and delivers the response to the device.

[0624] Output: Answer data to the user's device

[0625] Step 13: Answer display step

[0626] The terminal displays the received response to the user, and the interface formats and displays the received data.

[0627] Input: Received response data

[0628] Specific operation: The device displays the received response on the user interface, saying, "Breastfeeding at the evacuation center will be conducted in the same way as normal breastfeeding, and privacy will be ensured as much as possible. If necessary, ask the evacuation center administrator to provide a separate space."

[0629] Output: Answers displayed on many users' terminals

[0630] (Application example 1)

[0631] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0632] Conventional information provision systems for life in evacuation shelters often do not have a dedicated platform, making it difficult for users to quickly and accurately obtain the information they need. Furthermore, in emergencies and disasters, there is a need to provide security information and specific advice on evacuation routes, but no system exists that can meet this need.

[0633] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0634] In this invention, the server includes: means for providing information on life in a shelter using a generative model; means for building a themed Q&A database; means for loading the Q&A database into a generative model and customizing it to a specific theme; means for providing an interface for accepting questions from users; means for generating optimal answers to user questions using the generative model; means for transmitting the generated answers to the user's terminal; means for acquiring disaster safety information such as security information from a reliable information source in JSON format; and means for the generative model to provide the necessary answers based on the acquired data and support evacuation behavior in an emergency, thereby enabling users to obtain quick and accurate information via their smartphones or other terminals.

[0635] A "generative model" is an algorithm that generates new information or answers based on the data provided.

[0636] "Shelter life" refers to the living environment and lifestyle in a temporary shelter set up in the event of a disaster or emergency.

[0637] An "information providing system" is a computer system for providing necessary information to users.

[0638] A "topical Q&A database" is a database that organizes and stores data in the form of questions and answers for specific themes.

[0639] An "interface" is the means or screen through which a user interacts with a computer system or application.

[0640] "Natural language processing" is a technology that uses computers to analyze, process, and generate natural language used by humans.

[0641] "Security information" refers to various information provided to ensure the safety of users, and particularly includes information in emergencies and disasters.

[0642] The "JSON format" is a common format for expressing data in text format and is one of the formats widely used for data exchange.

[0643] System Overview

[0644] The present invention relates to a system that uses generative models to provide information on life in evacuation shelters and security during emergencies. The system is based on a generative model executed on the server side and has an interface that is easily accessible to users.

[0645] Data collection and registration

[0646] The server collects information related to shelter life and security in JSON format from reliable sources, then categorizes the collected data by topic, organizes it into a Q&A format, and registers it in a database that covers topics such as health, education, psychological support, and security information.

[0647] Training a generative model

[0648] The server uses the constructed Q&A database to train a generative model. This generative model utilizes natural language processing technology to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0649] User Interface

[0650] The user interface is provided as a smartphone app or web portal. The interface is intuitive and includes a question input form and category-based options. Users can enter their questions through this interface and immediately obtain the information they need.

[0651] Question and Answer Process

[0652] When a user enters a question and presses the send button, the device sends the question to the server, which uses the generative model to generate the best answer to the question and sends it back to the user's device. This answer may also include additional relevant or security information, if necessary.

[0653] Specific examples

[0654] Example 1: Selection of evacuation routes

[0655] User: The user types a question: "What is the best evacuation route in case of a flood?"

[0656] Terminal: The terminal sends a question to the server.

[0657] Server: The server uses the generative model to analyze the question and generate an appropriate answer, such as "In the event of a flood, it is best to evacuate to higher ground. The nearest evacuation shelter is the one on the higher ground in XX Park."

[0658] Terminal: The terminal displays the generated answer to the user.

[0659] Example 2: Emergency contact method

[0660] User: The user types a question: "How do I contact my family in case of an earthquake?"

[0661] Terminal: The terminal sends a question to the server.

[0662] Server: The server uses the generative model to generate the optimal answer. For example, it might generate an answer such as, "Since congestion is expected during an earthquake, it's best to communicate using text messages or social media. It's also a good idea to decide on a meeting place at an evacuation shelter."

[0663] Terminal: The terminal displays the generated answer to the user.

[0664] Hardware and software used

[0665] This system mainly performs data processing on the server side. The server requires a high-performance processor and a large amount of memory, so it is recommended to use a cloud service (e.g., Amazon Web Services, Google Cloud Platform). To run the generative model, we use a library specifically designed for natural language processing (e.g., Transformers by Hugging Face).

[0666] Prompt Sentence Examples

[0667] Here are some examples of prompts:

[0668] "What is the safest evacuation route in the event of a flood?"

[0669] "What is the best way to contact my family in the event of an earthquake?"

[0670] "Please tell me how to take care of your mental health while living in an evacuation shelter."

[0671] As described above, the present invention allows users to quickly and accurately obtain necessary information in the event of an emergency or disaster.

[0672] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0673] Step 1:

[0674] The server collects information related to shelter life and security from reliable sources in JSON format. This allows it to obtain the necessary data from multiple official websites and documents issued by specialized organizations. The input is the URL of each source, and the output is the retrieved JSON data. The server then organizes the information into a database.

[0675] Step 2:

[0676] The server categorizes the collected data by theme, formats it in Q&A format, and registers it in a database. The input is the JSON data obtained in step 1, and the output is a database organized in Q&A format. The server uses natural language processing technology to categorize each piece of information by theme.

[0677] Step 3:

[0678] The server trains a generative model using a Q&A database it has built. The input is the Q&A database, and the output is the trained generative model. The server uses libraries (e.g., Hugging Face Transformers) to build a generative model customized for a specific theme.

[0679] Step 4:

[0680] A user inputs a question through a smartphone app or web portal. The input is the user's question, and the output is the question data. The device provides the user interface, and the user intuitively operates it to input the question.

[0681] Step 5:

[0682] The terminal sends the user's query data to the server. The input is the query data entered by the user in step 4, and the output is the query data sent to the server. The terminal sends the data to the server via the network.

[0683] Step 6:

[0684] The server uses a generative model to generate the best answer to the received question. The input is the user's question data and a Q&A database, and the output is the generated answer. The server uses natural language processing technology to have the model analyze the user's question and generate an answer based on related information.

[0685] Step 7:

[0686] The server sends the generated answer to the user's terminal. The input is the generated answer, and the output is the answer sent to the user's terminal. The server sends the answer data to the terminal via the network, and the terminal receives it.

[0687] Step 8:

[0688] The terminal displays the received answer to the user. The input is the answer data received from the server, and the output is the answer displayed to the user. The terminal displays the answer through a user interface, and the user confirms it.

[0689] As a result, the user can quickly obtain accurate information.

[0690] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0691] The present invention relates to a system that uses a generative model to provide information related to life in a shelter, and by combining it with an emotion engine, provides optimal information according to the user's emotional state. Specific embodiments of this system are described below.

[0692] System Overview

[0693] This system is based on a generative model, emotion engine, and user interface running on the server side. The server collects data from reliable sources and builds a topic-specific Q&A database. The generative model learns from this database and provides optimal answers to user questions. The emotion engine also analyzes the user's emotions and appropriately adjusts the generated answers based on those emotions.

[0694] System configuration

[0695] Data collection and registration

[0696] The server collects information related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. The collected data is then categorized by topic, organized into a Q&A format, and registered in a database. Topics include health, education, and psychological support.

[0697] Training a generative model

[0698] The server uses the constructed Q&A database to train a generative model. The generative model utilizes natural language processing technology and is able to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0699] Emotion Engine Integration

[0700] The emotion engine analyzes text entered through the user interface and recognizes the user's emotions. For example, it can detect emotions such as gratitude, anger, and anxiety from specific keywords and contexts contained in the user's text.

[0701] User Interface

[0702] The user interface is provided in a form that is easily accessible to users (e.g., a smartphone app or web portal). The interface is intuitive and includes a question input form, category selection menu, etc. It also provides an optional field for inputting user emotions.

[0703] Question and Answer Process

[0704] When a user enters a question and presses the send button, the device sends the question to the server. The server uses the generative model to instantly generate the best answer for the received question, adjusting the answer according to the user's emotions as recognized by the emotion engine. For example, if the user is feeling anxious, the answer may be more polite and reassuring. The answer may also include additional related or supporting information as needed.

[0705] Specific examples

[0706] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[0707] User: The user types, "How should I breastfeed at the evacuation shelter? I'm worried."

[0708] Device: The device sends the question and emotional expression to the server.

[0709] Server: The server uses a generative model to analyze the question and generate an appropriate answer. For example, it might generate an answer such as, "Breastfeeding at the evacuation center should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible. If necessary, ask the evacuation center manager to provide a separate space." The emotion engine recognizes the user's "worry" emotion and adds a message such as, "Don't worry, we'll support you."

[0710] Terminal: The terminal displays the generated answer to the user.

[0711] Example 2: Questions about psychological stress and emotion recognition

[0712] User: The user types, "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious."

[0713] Device: The device sends the question and emotional expression to the server.

[0714] Server: The server uses a generative model to generate the optimal answer. For example, it might generate an answer such as, "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness practice. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The emotion engine recognizes the emotion "very anxious" and adds a message to the answer such as, "We understand your anxiety. Please don't suffer alone."

[0715] Terminal: The terminal displays the generated answer to the user.

[0716] In this way, by combining a generative model and an emotion engine, the system of the present invention can provide appropriate information according to the user's emotional state, thereby improving the quality of information acquisition during life in an evacuation shelter and supporting the user's daily life.

[0717] The processing flow will be explained below.

[0718] Step 1: Collect and organize your data

[0719] The server collects information about life in evacuation shelters from reliable sources (e.g., official websites, documents issued by specialized organizations).

[0720] The server categorizes the collected information by topic, for example into categories such as health, education, and psychological support.

[0721] The server creates specific question-and-answer pairs for each topic, building a Q&A format database.

[0722] Step 2: Training the generative model

[0723] The server trains the generative model using a Q&A database.

[0724] The server applies the trained generative model, enabling the model to generate the best answer to the user's question.

[0725] Training is conducted regularly and new information and data is added.

[0726] Step 3: Training and Integrating the Emotion Engine

[0727] The server trains the emotion engine using the user's text data.

[0728] The server integrates the emotion engine with the generative model to recognize emotions from user questions.

[0729] The emotion engine includes algorithms that detect emotions from specific keywords and contexts.

[0730] Step 4: Prepare the User Interface

[0731] The device provides a user-friendly interface, accessible through a smartphone app or web portal.

[0732] The terminal provides a question input form and a category selection menu, allowing users to easily input questions.

[0733] The device provides an emotion input option, allowing the user to explicitly input emotions.

[0734] Step 5: Receiving user questions

[0735] The user enters a question and emotion through the interface and presses the send button.

[0736] The terminal transmits the user's question data and emotion data to the server.

[0737] Step 6: Question and Sentiment Analysis

[0738] The server analyzes the received question data and emotion data.

[0739] The server passes the question data to the generative model to generate the optimal answer.

[0740] The server passes the emotion data to the emotion engine, which analyzes the user's emotional state.

[0741] Step 7: Adjust your responses based on emotion

[0742] The server adjusts the generated answer based on the emotion recognized by the emotion engine.

[0743] For example, if the user is feeling anxious, add reassuring language to the answer.

[0744] Step 8: Submit your response

[0745] The server transmits the generated answer and the adjusted answer based on the emotion to the user's terminal.

[0746] The terminal displays the received answer to the user.

[0747] Step 9: Gather user feedback

[0748] The terminal collects user feedback and sends it to the server.

[0749] The server analyzes the collected feedback and uses it to train the generative model and emotion engine next time.

[0750] Specific examples

[0751] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[0752] Step 5:

[0753] The user types, "How should I breastfeed at an evacuation shelter? I'm worried," and presses the send button.

[0754] Step 6:

[0755] The terminal transmits the question and emotion data to the server.

[0756] The server passes the question data to a generative model to generate the optimal answer.

[0757] The server passes the emotional data "I'm worried" to the emotion engine, which then recognizes the emotion.

[0758] Step 7:

[0759] The server generates an answer: "Breastfeeding in evacuation shelters should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible." The answer is then adjusted using the emotion engine.

[0760] The user is worried, so you add "Don't worry, we're here to help" to your answer.

[0761] Step 8:

[0762] The server sends the adjusted response to the user's terminal.

[0763] The terminal displays the generated answer to the user.

[0764] Example 2: Questions about psychological stress and emotion recognition

[0765] Step 5:

[0766] The user types, "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious," and presses the send button.

[0767] Step 6:

[0768] The terminal transmits the question and emotion data to the server.

[0769] The server passes the question data to a generative model to generate the optimal answer.

[0770] The server passes the emotional data "I'm very anxious" to the emotion engine, which then recognizes the emotion.

[0771] Step 7:

[0772] The server generates an answer: "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness exercises. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The answer is then adjusted using the emotion engine.

[0773] The user is feeling very anxious, so add "We understand your anxiety. Please don't go through it alone" to your response.

[0774] Step 8:

[0775] The server sends the adjusted response to the user's terminal.

[0776] The terminal displays the generated answer to the user.

[0777] In this way, by combining a generative model and an emotion engine, the system of the present invention can provide appropriate information according to the user's emotional state, thereby improving the quality of information acquisition during life in an evacuation shelter and supporting the user's daily life.

[0778] Example 2

[0779] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0780] When living in an evacuation shelter, there is a need for a method that allows users to easily obtain the information they need. In particular, it is important to provide optimal information according to the user's emotional state. However, conventional systems have difficulty providing information that reflects the user's emotions, and have not been able to provide sufficient psychological support during evacuation shelter life.

[0781] The identification process 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 providing information on life in a shelter using a generative model, means for constructing a thematic information base, means for loading the information base into a generative model and customizing it to a specific theme, means for providing a user interface for accepting questions from users, means for generating optimal answers to the user's questions using the generative model, means for transmitting the generated answers to the user's information device, and means for analyzing the user's emotions using an emotion analysis engine and reflecting the emotions in the answers. This makes it possible to provide optimal information according to the user's emotional state.

[0782] 1. "Generative model" is an artificial intelligence technique for generating natural language responses from collected data.

[0783] 2. The "Thematic Information Base" is a database constructed by classifying information related to life in evacuation shelters by theme.

[0784] 3. "Customization" means tailoring information to a specific topic or user needs and converting it into an appropriate format.

[0785] 4. "User interface" refers to interactive mechanisms such as input forms and menu screens that allow users to interact with a system.

[0786] 5. "User's information equipment" refers to electronic devices used by the User, such as smartphones, tablets, and personal computers.

[0787] 6. "Sentiment analysis engine" is software that analyzes text entered by a user and identifies the emotions contained therein.

[0788] 7. "Natural language processing" is a technology that allows computers to understand and generate human language.

[0789] 8. "Life in a shelter" refers to a situation in which people temporarily live in a shelter due to a disaster or other reason.

[0790] The present invention is a system that provides information related to life in an evacuation shelter, and by combining a generative model and an emotion analysis engine, it is possible to provide optimal information according to the user's emotional state. This system has the following configuration.

[0791] Data collection and database construction

[0792] The server collects data on life in evacuation shelters from government agencies and other reliable sources. The collected data is categorized by theme, such as "health," "education," and "psychological support." The server then organizes the data into a Q&A format and registers it in a database as a thematic information base. This database is managed using a relational database management system such as MySQL.

[0793] Training a generative model

[0794] The server uses the thematic information base to train a generative model (such as GPT-3). This generative model utilizes natural language processing techniques and is capable of generating optimal answers to user questions. Model training is performed using machine learning frameworks such as TensorFlow and PyTorch. The model is regularly updated to reflect the latest information.

[0795] Sentiment analysis engine integration

[0796] The sentiment analysis engine analyzes the text entered by the user through the user interface and recognizes the user's emotions. For example, the sentiment analysis engine identifies emotions such as "joy," "sadness," "anger," and "anxiety" from specific keywords and context. NLP libraries (Natural Language Toolkit and SpaCy) are used for this analysis.

[0797] Providing a user interface

[0798] The user interface is provided in the form of a smartphone app or web portal that is easily accessible to users. This interface is built using React Native and React.js and includes a question input form, a category selection menu, and an optional field for inputting user sentiment.

[0799] Question and Answer Process

[0800] When a user enters a question and presses the send button, the device sends the question and emotional expression to the server. The server uses a generative model to generate an optimal answer to the received question and adjusts the answer according to the user's emotions recognized by the emotion analysis engine. For example, if the user is feeling anxious, the server adds a reassuring message to the answer, such as "Don't worry, appropriate measures are being taken."

[0801] Specific examples

[0802] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[0803] User: Type "How should I breastfeed at the evacuation shelter? I'm worried."

[0804] Device: Sends questions and emotional expressions to the server.

[0805] Server: Using a generative model, it generates answers such as, "Breastfeeding at the evacuation center should be handled as normal, and privacy should be maintained as much as possible. If necessary, please ask the evacuation center manager to provide a separate space for the baby." The sentiment analysis engine recognizes the emotion "worry" and adds messages such as, "Don't worry, we're here to support you."

[0806] Terminal: Displays the generated answer to the user.

[0807] Example 2: Questions about psychological stress and emotion recognition

[0808] User: Type "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious."

[0809] Device: Sends questions and emotional expressions to the server.

[0810] Server: Using a generative model, the server generates answers such as, "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness practice. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The sentiment analysis engine recognizes the emotion "very anxious" and adds messages such as, "We understand your anxiety. Please don't bear it alone."

[0811] Terminal: Displays the generated answer to the user.

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

[0813] Step 1: Data collection and database registration

[0814] The server periodically crawls data related to shelter life from government agencies and reliable websites, using web scraping technology to collect reliable information.

[0815] The server automatically categorizes the crawled data into themes, such as "health," "education," and "psychological support."

[0816] The server converts the classified data into Q&A format and registers it in a MySQL database. The input data is the collected text information, and the output data is a database organized in Q&A format.

[0817] Step 2: Training the generative model

[0818] The server trains a generative model (such as GPT-3) using a Q&A dataset stored in the database. This process uses natural language processing techniques, and the model is trained to generate appropriate answers to questions.

[0819] The server evaluates the training results and, if they are insufficient, adds data and performs training again.,The input data is a Q&A format database, and the output data is,the trained generative model.

[0820] The server periodically adds new information to the database and updates the generative model.

[0821] Step 3: Integrating a sentiment analysis engine

[0822] The server embeds a sentiment analysis engine into the user interface, where a sentiment analysis algorithm analyzes the text data to identify sentiment.

[0823] The server analyzes the text entered by the user in real time and identifies emotions such as "joy," "sadness," "anger," and "anxiety" based on keywords and context. The input data is the text entered by the user, and the output data is the result of identifying the emotion.

[0824] Test the accuracy of your sentiment analysis engine and fine-tune the algorithm as needed.

[0825] Step 4: Providing a User Interface

[0826] The server provides users with smartphone apps using React Native and web portals using React.js.

[0827] The terminal displays a question input form and a category selection menu to the user. The input data is the user's operation information, and the output data is the interface display content.

[0828] The user enters a question and enters an emotion in the emotion input field.

[0829] Step 5: Question-answering process

[0830] The user enters a question and presses the send button. For example, "How should I breastfeed at an evacuation shelter? I'm worried."

[0831] The terminal sends the input question and emotional expression to the server. The input data is the user's question and emotional information, and the output data is the information to be sent to the server.

[0832] The server analyzes the question using a generative model and generates an optimal answer. For example, it generates an answer such as, "Breastfeeding at evacuation shelters should be done in the same way as normal breastfeeding, and privacy should be maintained as much as possible. If necessary, please ask the shelter manager to provide a separate space." The input data is the user's question text, and the output data is the generated answer.

[0833] The server uses a sentiment analysis engine to tailor the generated answer depending on the user's emotions, for example adding a message like "Don't worry, we're here to help" if the user is feeling anxious.

[0834] The server sends the final answer to the terminal. The input data is the adjusted answer, and the output data is the information to be sent to the terminal.

[0835] The device will then display the received response to the user, such as, "Breastfeeding at the evacuation shelter should be done in the same way as normal breastfeeding, and please ensure privacy as much as possible. If necessary, please ask the shelter manager to provide a separate space. Don't worry, we'll support you."

[0836] (Application example 2)

[0837] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0838] Conventional food delivery systems only suggest generic menus without considering the user's emotional state. As a result, they are unable to suggest appropriate menus that reflect the user's psychological state, making it difficult to sufficiently improve satisfaction. There is a need for a system that can solve this problem and make highly personalized menu suggestions that correspond to the user's emotional state.

[0839] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for providing information on life in a shelter using a generative model, means for constructing a themed Q&A database, means for loading the Q&A database into a generative model and customizing it to a specific theme, means for providing an interface for accepting questions from users, means for generating optimal answers to user questions using the generative model, means for transmitting the generated answers to the user's terminal, means for analyzing the user's emotions using an emotion engine, and means for adjusting the generated answers depending on the user's emotional state. This makes it possible to propose optimal menus depending on the user's emotional state.

[0840] - A "generative model" is an algorithm that learns from large amounts of data and generates appropriate outputs for given inputs.

[0841] The "Topical Q&A Database" is a database that contains questions and answers related to different themes.

[0842] An "interface" is a means for exchanging data between a user and a system.

[0843] An "emotion engine" is an analysis system that analyzes emotions from user input and recognizes specific emotional states.

[0844] "Analysis" is the process of examining data and understanding its structure and meaning.

[0845] "Terminal" refers to a device through which a user inputs and receives information.

[0846] "Food delivery" refers to the general service of ordering and having food delivered.

[0847] "Menu suggestions" refers to presenting appropriate meal options to the user.

[0848] "Personalization" means customizing something individually to suit a specific user.

[0849] System configuration

[0850] The present invention includes a generative model, an emotion engine, a user interface, and a server-based system for controlling them.

[0851] Collect data from reliable sources and build a thematic Q&A database.

[0852] The generative model learns from this database and generates optimal answers to user questions.

[0853] The emotion engine analyzes the user's input text and recognizes the emotional state.

[0854] The user interface is provided as a smartphone app, allowing intuitive user interaction.

[0855] Program processing

[0856] 1. Data Collection:

[0857] The server automatically collects information related to shelter life and food delivery from official websites and documents issued by specialized organizations.

[0858] The collected data is categorized by theme and registered in a database in Q&A format.

[0859] 2. Training the generative model:

[0860] The server uses the constructed Q&A database to train a generative model (e.g., OpenAI's GPT-3).

[0861] The model will be updated periodically to reflect new information.

[0862] 3. Emotion engine integration:

[0863] The emotion engine analyzes the user's input text and identifies an emotional state.

[0864] For example, it can recognize emotions such as "fatigue" and "stress" from specific keywords and context within the text.

[0865] 4. User Interface:

[0866] It is provided as a smartphone app, allowing users to enter and submit questions.

[0867] The app includes a question form, a category selection menu, and an emotion input option.

[0868] 5. Question and Answer Process:

[0869] When a user inputs a question and presses the send button, the terminal sends the question to the server.

[0870] The server uses a generative model to generate optimal answers and adjusts the answers according to the user's emotional state, as recognized by the emotion engine.

[0871] The generated answer is sent to the user's terminal and displayed.

[0872] Specific examples

[0873] 1. Example 1: Menu suggestions for fatigued users

[0874] User: Type, "I'm feeling tired today, what foods will make me feel better?"

[0875] Server: The emotion engine recognizes the emotion "fatigue," and the generative model generates an appropriate answer: "To recover from fatigue, we recommend fruits rich in vitamin C and lean meat, which contains iron. Also, don't forget to get adequate rest."

[0876] 2. Example 2: Menu suggestions for stressed users

[0877] User: Type "I'm stressed and want some food to help me relax."

[0878] Server: The emotion engine recognizes the emotion "stress," and the generative model generates an appropriate answer: "Herbal tea and yogurt are recommended for relieving stress. Dark chocolate also helps you relax."

[0879] Example prompt sentence:

[0880] 1. "User Question: I'm feeling tired today. What foods will make me feel better?"

[0881] "User question: I'm tired today, what foods will make me feel better?\nUser emotion: Fatigue\nGenerate the best answer."

[0882] 2. "User Question: I'm stressed and want some food to help me relax."

[0883] "User question: I'm stressed and want some food to help me relax.\nUser emotion: Stress\nGenerate the best answer."

[0884] In this way, the invention combines an emotion engine and a generative model to provide optimal information according to the user's emotional state.

[0885] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0886] Step 1:

[0887] The server automatically collects information related to shelter life and food delivery from official websites and documents issued by specialized organizations. It uses APIs or scraping technology to extract data and categorize it by topic. The collected data is then registered in a database. The input is data obtained from reliable sources, and the output is a database in Q&A format, categorized by topic.

[0888] Step 2:

[0889] The server uses the constructed Q&A database to train a generative model (e.g., OpenAI's GPT-3). In this process, the generative model learns from the data using natural language processing so that it can generate appropriate answers to questions. The input is the information in the Q&A database, and the output is the trained generative model.

[0890] Step 3:

[0891] The server uses an emotion engine to analyze emotions from the user's input text. Specifically, the emotion engine analyzes specific keywords and contexts in the text and identifies the emotional state as "fatigue" or "stress," etc. The input is the user's input text, and the output is the analyzed emotional information.

[0892] Step 4:

[0893] The user interface (smartphone app) accepts questions from users. Questions are entered into an input form and sent to the server by pressing the send button. The input is the user's question text, and the output is the question data sent to the server.

[0894] Step 5:

[0895] When the server receives a user's question, it uses a generative model to generate an optimal answer. It then adjusts the answer based on the user's emotional state, as recognized by the emotion engine. For example, if the user is feeling tired, the answer may include specific suggestions such as "foods that will help recover from fatigue." The input is the user's question data and emotional information, and the output is the generated answer.

[0896] Step 6:

[0897] The generated answer is sent from the server to the user's device (smartphone). The user's device displays the received answer. The input is the generated answer data from the server, and the output is the answer displayed on the user's device.

[0898] Specific actions

[0899] Step 1: For example, the server scrapes health information from the official WHO website and categorizes it by topic, such as "health" or "psychological support."

[0900] Step 2: Using the GPT-3 model, we train the model by learning from the collected Q&A data on themes such as "health" and "psychological support."

[0901] Step 3: The emotion engine analyzes the emotion “fatigue” from the text entered by the user: “I feel tired today, please tell me what foods will make me feel better.”

[0902] Step 4: The user interface receives and sends the question "I'm stressed and want some food to help me relax" from the user.

[0903] Step 5: The server generates a response to the user in the "stressed" state, suggesting "herbal tea and yogurt" as "foods that help relieve stress."

[0904] Step 6: The generated answer is displayed in the smartphone app for the user to review.

[0905] In this way, by using the trained generative model and emotion engine, personalized food menu suggestions that are in line with the user's emotional state can be realized.

[0906] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0907] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0908] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0909] [Third embodiment]

[0910] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0911] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0912] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0913] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0914] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0915] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0916] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0917] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0918] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0919] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0920] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0921] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0922] The present invention relates to a system for providing information related to life in an evacuation shelter using a generative model. Specific embodiments of the system are described below.

[0923] System Overview

[0924] The system is based on a server-side generative model and provides a user-friendly interface. The server collects data from reliable sources and builds a topic-specific Q&A database. The generative model learns from this database and provides optimal answers to user questions.

[0925] System configuration

[0926] Data collection and registration

[0927] The server collects information related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. The collected data is then categorized by topic, organized into a Q&A format, and registered in a database. Topics include health, education, and psychological support.

[0928] Training a generative model

[0929] The server uses the constructed Q&A database to train a generative model. The generative model utilizes natural language processing technology and is able to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0930] User Interface

[0931] The user interface is provided in a form that is easily accessible to users (e.g., smartphone app, web portal), and includes a question input form and options by category, making it intuitive to operate.

[0932] Question and Answer Process

[0933] When a user enters a question and presses the send button, the device sends the question to the server, which uses the generative model to instantly generate the best answer to the received question and sends it back to the user's device. This answer may also include additional related or supporting information as needed.

[0934] Specific examples

[0935] Example 1: Questions about breastfeeding at evacuation shelters

[0936] User: The user types a question: "How should breastfeeding be handled in an evacuation shelter?"

[0937] Terminal: The terminal sends a question to the server.

[0938] Server: The server uses the generative model to analyze the question and generate an appropriate answer, such as "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, with as much privacy as possible. If necessary, ask the evacuation center manager to provide separate spaces."

[0939] Terminal: The terminal displays the generated answer to the user.

[0940] Example 2: Questions about psychological stress

[0941] User: The user types a question: "Please tell me how to take care of myself mentally while living in an evacuation shelter."

[0942] Terminal: The terminal sends a question to the server.

[0943] Server: The server uses the generative model to generate the optimal answer, such as, "Regular rest and relaxation are important when living in an evacuation shelter. It is a good idea to incorporate relaxation techniques and mindfulness practice. Also, if you are experiencing severe psychological stress, we recommend that you seek professional counseling."

[0944] Terminal: The terminal displays the generated answer to the user.

[0945] In this way, the system of the present invention allows users to quickly and accurately obtain information about life in a shelter, thereby improving their quality of life.

[0946] The processing flow will be explained below.

[0947] Step 1: Collect and organize your data

[0948] The server collects information about life in evacuation shelters from reliable sources (e.g., official websites, documents issued by specialized organizations).

[0949] The server categorizes the collected information by topic, for example into categories such as health, education, and psychological support.

[0950] The server creates specific question-and-answer pairs for each topic, building a Q&A format database.

[0951] Step 2: Training the generative model

[0952] The server trains the generative model using a Q&A database.

[0953] The server applies the trained generative model, enabling the model to generate the best answer to the user's question.

[0954] Training is conducted regularly and new information and data is added.

[0955] Step 3: Prepare the user interface

[0956] The device provides a user-friendly interface, accessible through a smartphone app or web portal.

[0957] The terminal provides a question input form and a category selection menu, allowing users to easily input questions.

[0958] Step 4: Receiving questions from users

[0959] The user enters a question through the interface and presses the submit button.

[0960] The terminal transmits the user's question data to the server.

[0961] Step 5: Parsing the question and generating an answer

[0962] The server analyzes the received question data and passes it to the generative model.

[0963] The server uses the generative model to generate optimal answers to the user's questions.

[0964] If necessary, the server will search for additional information (e.g., related resources or supporting information) and append it to the answer.

[0965] Step 6: Submit your response

[0966] The server sends the generated answer and any additional information required to the user's terminal.

[0967] The terminal displays the received answer to the user.

[0968] Step 7: Gather user feedback

[0969] The terminal collects user feedback and sends it to the server.

[0970] The server analyzes the collected feedback and uses it to train the next generative model.

[0971] Through this series of steps, the system of the present invention can quickly and accurately provide information about life in an evacuation shelter, thereby contributing to improving the lives of users.

[0972] Example 1

[0973] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0974] Obtaining information during evacuation shelter life is difficult, and one of the challenges is the lack of timely provision of appropriate information tailored to individual needs and problems. This is due to limited resources, issues with the reliability of information sources, and the difficulty of accurately categorizing and providing information. Furthermore, a system that combines high accuracy and versatility is required to address a wide range of information needs, including those related to psychological stress and health issues.

[0975] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0976] In this invention, the server includes means for collecting data from reliable information sources, means for constructing a Q&A database categorized by theme, means for loading the Q&A database into a generative model to learn information, means for training the generative model to generate optimal answers to user questions, means for providing an interface for accepting questions from users, means for transmitting the generated answers to the user's terminal, and means for displaying the generated answers on the user's terminal, thereby enabling users to quickly and accurately obtain a wide range of information related to life in a shelter.

[0977] "Reliable sources" are institutions or platforms that provide accurate and up-to-date information, such as official websites or documents issued by specialized organizations.

[0978] A "Q&A database" is a database in which questions and their answers are formalized and stored, and each question and answer is categorized by topic.

[0979] A "generative model" is a machine learning model that uses natural language processing technology to generate optimal answers to user questions.

[0980] An "interface" is the means by which users access the system and input questions, and may be provided in the form of a smartphone app or web portal.

[0981] A "terminal" is a device that a user uses to use an interface, such as a smartphone or computer.

[0982] "Natural language processing technology" is a technology for processing human language using a computer, and is a technique for analyzing, understanding, and generating text.

[0983] "Additional related or supporting information" is supplemental content, such as more detailed information or advice, that accompanies the answer to the user's question.

[0984] The present invention relates to a system for providing information related to life in an evacuation shelter using a generative AI model. Specific embodiments for carrying out the present invention will be described below.

[0985] Overall system picture

[0986] The system is based on a generative AI model running on the server side and has a user-accessible interface. The server collects data from reliable sources and builds a Q&A database categorized by topic. This database is used to train the generative AI model, which generates optimal answers to user questions. The system also has a function for sending and displaying answers to the user's device.

[0987] Data collection and registration

[0988] The server collects data related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. Specifically, it automatically obtains the latest information using RSS feeds and APIs. The collected data is categorized by topic, organized into a Q&A format, and entered into a database. This database is divided into themes such as health, education, and psychological support.

[0989] Training a generative model

[0990] The server uses the constructed Q&A database to train a generative AI model. One example of a generative AI model used is GPT-4. This model utilizes natural language processing technology, enabling it to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[0991] Providing a user interface

[0992] The user interface is provided in a form that is easy for users to access (e.g., smartphone app, web portal), using technologies such as React Native and React.js. The interface includes a question input form and options by category, making it intuitive to use.

[0993] Question and Answer Process

[0994] When a user enters a question into the interface and presses the send button, the device sends the question to the server. The server uses a generative AI model to generate the best answer for the received question. The answer may include additional related or supporting information as needed. The generated answer is then displayed to the user through the device.

[0995] Specific examples

[0996] Example 1: Questions about breastfeeding at evacuation shelters

[0997] User: The user types a question: "How should breastfeeding be handled in an evacuation shelter?"

[0998] Terminal: The terminal sends a question to the server.

[0999] Server: The server uses a generative AI model to analyze the question and generate an appropriate answer, such as "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, with as much privacy as possible. If necessary, ask the evacuation center manager to provide separate spaces."

[1000] Terminal: The terminal displays the generated answer to the user.

[1001] Example 2: Questions about psychological stress

[1002] User: The user types a question: "Please tell me how to take care of myself mentally while living in an evacuation shelter."

[1003] Terminal: The terminal sends a question to the server.

[1004] Server: The server uses a generative AI model to generate optimal answers, such as, "Regular rest and relaxation are important when living in an evacuation shelter. It's a good idea to incorporate relaxation techniques and mindfulness exercises. If you're experiencing significant psychological stress, we recommend seeking professional counseling."

[1005] Terminal: The terminal displays the generated answer to the user.

[1006] In this way, the system of the present invention allows users to quickly and accurately obtain information about life in a shelter, thereby improving their quality of life.

[1007] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1008] Step 1: Data collection step

[1009] The server automatically collects data related to shelter life from reliable sources (e.g., official websites, documents issued by specialized organizations), using RSS feeds and APIs to obtain the latest information.

[1010] Input: Source URL or API endpoint

[1011] Specific operation: The server sends an "HTTP request" to obtain data from the information source.

[1012] Output: Retrieved data (e.g. HTML, PDF, JSON)

[1013] Step 2: Data classification and registration step

[1014] The server categorizes the collected data by topic, organizes it into a Q&A format, and registers it in a database. Data categorization is performed using tagging technology.

[1015] Input: Collected data

[1016] How it works: The server uses natural language processing technology to analyze the data and classify it into categories such as "health," "education," and "psychological support." The classified data is then converted into a Q&A format and stored in a database.

[1017] Output: Q&A database

[1018] Step 3: Data preparation steps for the generative model

[1019] The server extracts data from the Q&A database and formats it as a training dataset for the generative model.

[1020] Input: Q&A database

[1021] What happens: The server extracts the Q&A data from the database and converts it into an appropriate format (e.g., JSON).

[1022] Output: Training dataset

[1023] Step 4: Training the generative model

[1024] The server uses the extracted data to train a generative model, such as GPT-4.

[1025] Input: Training dataset

[1026] Specific operation: The server executes a Python script and performs processing to train a generative model (GPT-4).

[1027] Output: A trained generative model

[1028] Step 5: Model Update Step

[1029] The server periodically retrieves new data and retrains the model with the updated data.

[1030] Input: New data and existing Q&A database

[1031] What happens: The server integrates the newly collected data with the existing database and retrains the generative model.

[1032] Output: Updated generative model

[1033] Step 6: Interface development step

[1034] The server develops the user interface in a form that is easy for users to access, using technologies such as React Native and React.js.

[1035] Input: Interface design and functional requirements

[1036] What it does: A developer uses React.js to build a web portal and adds a question form and options based on categories.

[1037] Output: The finished user interface

[1038] Step 7: Interface Operation Steps

[1039] The server monitors the operational interfaces and performs maintenance and updates as needed.

[1040] Input: User feedback and operational data

[1041] What it does: The server periodically collects user feedback and makes bug fixes and feature improvements.

[1042] Output: User interface in operation

[1043] Step 8: Question input step

[1044] The user accesses the interface and enters a question. The question entry form includes fields for selecting a category and entering specific questions.

[1045] Input: User question

[1046] What happens: The user types into the interface, "How should breastfeeding be done in an evacuation shelter?"

[1047] Output: Question data

[1048] Step 9: Submit a question

[1049] The device sends the questions entered by the user to the server, and the data is sent in JSON format.

[1050] Input: User question data

[1051] Specific operation: The device sends a POST request to the API endpoint.

[1052] Output: Query data to the server

[1053] Step 10: Question Analysis Step

[1054] The server analyzes the received question and formats it as data to be input into the generative model.

[1055] Input: Received question data

[1056] Specific behavior: The server parses and preprocesses the received query (e.g., tokenization and normalization).

[1057] Output: Well-formed question data

[1058] Step 11: Answer generation step

[1059] The server uses the generative model to generate optimal answers to questions.

[1060] Input: Well-formed question data

[1061] How it works: The server inputs question data into the generative model and generates an answer, such as, "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible. If necessary, ask the evacuation center manager to provide a separate space."

[1062] Output: Response data

[1063] Step 12: Send response step

[1064] The server then sends the generated response to the user's device. The data is sent in JSON format.

[1065] Input: Response data

[1066] Specific operation: The server sends a POST request to the API endpoint and delivers the response to the device.

[1067] Output: Answer data to the user's device

[1068] Step 13: Answer display step

[1069] The terminal displays the received response to the user, and the interface formats and displays the received data.

[1070] Input: Received response data

[1071] Specific operation: The device displays the received response on the user interface, saying, "Breastfeeding at the evacuation center will be conducted in the same way as normal breastfeeding, and privacy will be ensured as much as possible. If necessary, ask the evacuation center administrator to provide a separate space."

[1072] Output: Answers displayed on many users' terminals

[1073] (Application example 1)

[1074] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1075] Conventional information provision systems for life in evacuation shelters often do not have a dedicated platform, making it difficult for users to quickly and accurately obtain the information they need. Furthermore, in emergencies and disasters, there is a need to provide security information and specific advice on evacuation routes, but no system exists that can meet this need.

[1076] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1077] In this invention, the server includes: means for providing information on life in a shelter using a generative model; means for building a themed Q&A database; means for loading the Q&A database into a generative model and customizing it to a specific theme; means for providing an interface for accepting questions from users; means for generating optimal answers to user questions using the generative model; means for transmitting the generated answers to the user's terminal; means for acquiring disaster safety information such as security information from a reliable information source in JSON format; and means for the generative model to provide the necessary answers based on the acquired data and support evacuation behavior in an emergency, thereby enabling users to obtain quick and accurate information via their smartphones or other terminals.

[1078] A "generative model" is an algorithm that generates new information or answers based on the data provided.

[1079] "Shelter life" refers to the living environment and lifestyle in a temporary shelter set up in the event of a disaster or emergency.

[1080] An "information providing system" is a computer system for providing necessary information to users.

[1081] A "topical Q&A database" is a database that organizes and stores data in the form of questions and answers for specific themes.

[1082] An "interface" is the means or screen through which a user interacts with a computer system or application.

[1083] "Natural language processing" is a technology that uses computers to analyze, process, and generate natural language used by humans.

[1084] "Security information" refers to various information provided to ensure the safety of users, and particularly includes information in emergencies and disasters.

[1085] The "JSON format" is a common format for expressing data in text format and is one of the formats widely used for data exchange.

[1086] System Overview

[1087] The present invention relates to a system that uses generative models to provide information on life in evacuation shelters and security during emergencies. The system is based on a generative model executed on the server side and has an interface that is easily accessible to users.

[1088] Data collection and registration

[1089] The server collects information related to shelter life and security in JSON format from reliable sources, then categorizes the collected data by topic, organizes it into a Q&A format, and registers it in a database that covers topics such as health, education, psychological support, and security information.

[1090] Training a generative model

[1091] The server uses the constructed Q&A database to train a generative model. This generative model utilizes natural language processing technology to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[1092] User Interface

[1093] The user interface is provided as a smartphone app or web portal. The interface is intuitive and includes a question input form and category-based options. Users can enter their questions through this interface and immediately obtain the information they need.

[1094] Question and Answer Process

[1095] When a user enters a question and presses the send button, the device sends the question to the server, which uses the generative model to generate the best answer to the question and sends it back to the user's device. This answer may also include additional relevant or security information, if necessary.

[1096] Specific examples

[1097] Example 1: Selection of evacuation routes

[1098] User: The user types a question: "What is the best evacuation route in case of a flood?"

[1099] Terminal: The terminal sends a question to the server.

[1100] Server: The server uses the generative model to analyze the question and generate an appropriate answer, such as "In the event of a flood, it is best to evacuate to higher ground. The nearest evacuation shelter is the one on the higher ground in XX Park."

[1101] Terminal: The terminal displays the generated answer to the user.

[1102] Example 2: Emergency contact method

[1103] User: The user types a question: "How do I contact my family in case of an earthquake?"

[1104] Terminal: The terminal sends a question to the server.

[1105] Server: The server uses the generative model to generate the optimal answer. For example, it might generate an answer such as, "Since congestion is expected during an earthquake, it's best to communicate using text messages or social media. It's also a good idea to decide on a meeting place at an evacuation shelter."

[1106] Terminal: The terminal displays the generated answer to the user.

[1107] Hardware and software used

[1108] This system mainly performs data processing on the server side. The server requires a high-performance processor and a large amount of memory, so it is recommended to use a cloud service (e.g., Amazon Web Services, Google Cloud Platform). To run the generative model, we use a library specifically designed for natural language processing (e.g., Transformers by Hugging Face).

[1109] Prompt Sentence Examples

[1110] Here are some examples of prompts:

[1111] "What is the safest evacuation route in the event of a flood?"

[1112] "What is the best way to contact my family in the event of an earthquake?"

[1113] "Please tell me how to take care of your mental health while living in an evacuation shelter."

[1114] As described above, the present invention allows users to quickly and accurately obtain necessary information in the event of an emergency or disaster.

[1115] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1116] Step 1:

[1117] The server collects information related to shelter life and security from reliable sources in JSON format. This allows it to obtain the necessary data from multiple official websites and documents issued by specialized organizations. The input is the URL of each source, and the output is the retrieved JSON data. The server then organizes the information into a database.

[1118] Step 2:

[1119] The server categorizes the collected data by theme, formats it in Q&A format, and registers it in a database. The input is the JSON data obtained in step 1, and the output is a database organized in Q&A format. The server uses natural language processing technology to categorize each piece of information by theme.

[1120] Step 3:

[1121] The server trains a generative model using a Q&A database it has built. The input is the Q&A database, and the output is the trained generative model. The server uses libraries (e.g., Hugging Face Transformers) to build a generative model customized for a specific theme.

[1122] Step 4:

[1123] A user inputs a question through a smartphone app or web portal. The input is the user's question, and the output is the question data. The device provides the user interface, and the user intuitively operates it to input the question.

[1124] Step 5:

[1125] The terminal sends the user's query data to the server. The input is the query data entered by the user in step 4, and the output is the query data sent to the server. The terminal sends the data to the server via the network.

[1126] Step 6:

[1127] The server uses a generative model to generate the best answer to the received question. The input is the user's question data and a Q&A database, and the output is the generated answer. The server uses natural language processing technology to have the model analyze the user's question and generate an answer based on related information.

[1128] Step 7:

[1129] The server sends the generated answer to the user's terminal. The input is the generated answer, and the output is the answer sent to the user's terminal. The server sends the answer data to the terminal via the network, and the terminal receives it.

[1130] Step 8:

[1131] The terminal displays the received answer to the user. The input is the answer data received from the server, and the output is the answer displayed to the user. The terminal displays the answer through a user interface, and the user confirms it.

[1132] As a result, the user can quickly obtain accurate information.

[1133] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1134] The present invention relates to a system that uses a generative model to provide information related to life in a shelter, and by combining it with an emotion engine, provides optimal information according to the user's emotional state. Specific embodiments of this system are described below.

[1135] System Overview

[1136] This system is based on a generative model, emotion engine, and user interface running on the server side. The server collects data from reliable sources and builds a topic-specific Q&A database. The generative model learns from this database and provides optimal answers to user questions. The emotion engine also analyzes the user's emotions and appropriately adjusts the generated answers based on those emotions.

[1137] System configuration

[1138] Data collection and registration

[1139] The server collects information related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. The collected data is then categorized by topic, organized into a Q&A format, and registered in a database. Topics include health, education, and psychological support.

[1140] Training a generative model

[1141] The server uses the constructed Q&A database to train a generative model. The generative model utilizes natural language processing technology and is able to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[1142] Emotion Engine Integration

[1143] The emotion engine analyzes text entered through the user interface and recognizes the user's emotions. For example, it can detect emotions such as gratitude, anger, and anxiety from specific keywords and contexts contained in the user's text.

[1144] User Interface

[1145] The user interface is provided in a form that is easily accessible to users (e.g., a smartphone app or web portal). The interface is intuitive and includes a question input form, category selection menu, etc. It also provides an optional field for inputting user emotions.

[1146] Question and Answer Process

[1147] When a user enters a question and presses the send button, the device sends the question to the server. The server uses the generative model to instantly generate the best answer for the received question, adjusting the answer according to the user's emotions as recognized by the emotion engine. For example, if the user is feeling anxious, the answer may be more polite and reassuring. The answer may also include additional related or supporting information as needed.

[1148] Specific examples

[1149] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[1150] User: The user types, "How should I breastfeed at the evacuation shelter? I'm worried."

[1151] Device: The device sends the question and emotional expression to the server.

[1152] Server: The server uses a generative model to analyze the question and generate an appropriate answer. For example, it might generate an answer such as, "Breastfeeding at the evacuation center should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible. If necessary, ask the evacuation center manager to provide a separate space." The emotion engine recognizes the user's "worry" emotion and adds a message such as, "Don't worry, we'll support you."

[1153] Terminal: The terminal displays the generated answer to the user.

[1154] Example 2: Questions about psychological stress and emotion recognition

[1155] User: The user types, "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious."

[1156] Device: The device sends the question and emotional expression to the server.

[1157] Server: The server uses a generative model to generate the optimal answer. For example, it might generate an answer such as, "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness practice. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The emotion engine recognizes the emotion "very anxious" and adds a message to the answer such as, "We understand your anxiety. Please don't suffer alone."

[1158] Terminal: The terminal displays the generated answer to the user.

[1159] In this way, by combining a generative model and an emotion engine, the system of the present invention can provide appropriate information according to the user's emotional state, thereby improving the quality of information acquisition during life in an evacuation shelter and supporting the user's daily life.

[1160] The processing flow will be explained below.

[1161] Step 1: Collect and organize your data

[1162] The server collects information about life in evacuation shelters from reliable sources (e.g., official websites, documents issued by specialized organizations).

[1163] The server categorizes the collected information by topic, for example into categories such as health, education, and psychological support.

[1164] The server creates specific question-and-answer pairs for each topic, building a Q&A format database.

[1165] Step 2: Training the generative model

[1166] The server trains the generative model using a Q&A database.

[1167] The server applies the trained generative model, enabling the model to generate the best answer to the user's question.

[1168] Training is conducted regularly and new information and data is added.

[1169] Step 3: Training and Integrating the Emotion Engine

[1170] The server trains the emotion engine using the user's text data.

[1171] The server integrates the emotion engine with the generative model to recognize emotions from user questions.

[1172] The emotion engine includes algorithms that detect emotions from specific keywords and contexts.

[1173] Step 4: Prepare the User Interface

[1174] The device provides a user-friendly interface, accessible through a smartphone app or web portal.

[1175] The terminal provides a question input form and a category selection menu, allowing users to easily input questions.

[1176] The device provides an emotion input option, allowing the user to explicitly input emotions.

[1177] Step 5: Receiving user questions

[1178] The user enters a question and emotion through the interface and presses the send button.

[1179] The terminal transmits the user's question data and emotion data to the server.

[1180] Step 6: Question and Sentiment Analysis

[1181] The server analyzes the received question data and emotion data.

[1182] The server passes the question data to the generative model to generate the optimal answer.

[1183] The server passes the emotion data to the emotion engine, which analyzes the user's emotional state.

[1184] Step 7: Adjust your responses based on emotion

[1185] The server adjusts the generated answer based on the emotion recognized by the emotion engine.

[1186] For example, if the user is feeling anxious, add reassuring language to the answer.

[1187] Step 8: Submit your response

[1188] The server transmits the generated answer and the adjusted answer based on the emotion to the user's terminal.

[1189] The terminal displays the received answer to the user.

[1190] Step 9: Gather user feedback

[1191] The terminal collects user feedback and sends it to the server.

[1192] The server analyzes the collected feedback and uses it to train the generative model and emotion engine next time.

[1193] Specific examples

[1194] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[1195] Step 5:

[1196] The user types, "How should I breastfeed at an evacuation shelter? I'm worried," and presses the send button.

[1197] Step 6:

[1198] The terminal transmits the question and emotion data to the server.

[1199] The server passes the question data to a generative model to generate the optimal answer.

[1200] The server passes the emotional data "I'm worried" to the emotion engine, which then recognizes the emotion.

[1201] Step 7:

[1202] The server generates an answer: "Breastfeeding in evacuation shelters should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible." The answer is then adjusted using the emotion engine.

[1203] The user is worried, so you add "Don't worry, we're here to help" to your answer.

[1204] Step 8:

[1205] The server sends the adjusted response to the user's terminal.

[1206] The terminal displays the generated answer to the user.

[1207] Example 2: Questions about psychological stress and emotion recognition

[1208] Step 5:

[1209] The user types, "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious," and presses the send button.

[1210] Step 6:

[1211] The terminal transmits the question and emotion data to the server.

[1212] The server passes the question data to a generative model to generate the optimal answer.

[1213] The server passes the emotional data "I'm very anxious" to the emotion engine, which then recognizes the emotion.

[1214] Step 7:

[1215] The server generates an answer: "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness exercises. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The answer is then adjusted using the emotion engine.

[1216] The user is feeling very anxious, so add "We understand your anxiety. Please don't go through it alone" to your response.

[1217] Step 8:

[1218] The server sends the adjusted response to the user's terminal.

[1219] The terminal displays the generated answer to the user.

[1220] In this way, by combining a generative model and an emotion engine, the system of the present invention can provide appropriate information according to the user's emotional state, thereby improving the quality of information acquisition during life in an evacuation shelter and supporting the user's daily life.

[1221] Example 2

[1222] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1223] When living in an evacuation shelter, there is a need for a method that allows users to easily obtain the information they need. In particular, it is important to provide optimal information according to the user's emotional state. However, conventional systems have difficulty providing information that reflects the user's emotions, and have not been able to provide sufficient psychological support during evacuation shelter life.

[1224] The identification process 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 providing information on life in a shelter using a generative model, means for constructing a thematic information base, means for loading the information base into a generative model and customizing it to a specific theme, means for providing a user interface for accepting questions from users, means for generating optimal answers to the user's questions using the generative model, means for transmitting the generated answers to the user's information device, and means for analyzing the user's emotions using an emotion analysis engine and reflecting the emotions in the answers. This makes it possible to provide optimal information according to the user's emotional state.

[1225] 1. "Generative model" is an artificial intelligence technique for generating natural language responses from collected data.

[1226] 2. The "Thematic Information Base" is a database constructed by classifying information related to life in evacuation shelters by theme.

[1227] 3. "Customization" means tailoring information to a specific topic or user needs and converting it into an appropriate format.

[1228] 4. "User interface" refers to interactive mechanisms such as input forms and menu screens that allow users to interact with a system.

[1229] 5. "User's information equipment" refers to electronic devices used by the User, such as smartphones, tablets, and personal computers.

[1230] 6. "Sentiment analysis engine" is software that analyzes text entered by a user and identifies the emotions contained therein.

[1231] 7. "Natural language processing" is a technology that allows computers to understand and generate human language.

[1232] 8. "Life in a shelter" refers to a situation in which people temporarily live in a shelter due to a disaster or other reason.

[1233] The present invention is a system that provides information related to life in an evacuation shelter, and by combining a generative model and an emotion analysis engine, it is possible to provide optimal information according to the user's emotional state. This system has the following configuration.

[1234] Data collection and database construction

[1235] The server collects data on life in evacuation shelters from government agencies and other reliable sources. The collected data is categorized by theme, such as "health," "education," and "psychological support." The server then organizes the data into a Q&A format and registers it in a database as a thematic information base. This database is managed using a relational database management system such as MySQL.

[1236] Training a generative model

[1237] The server uses the thematic information base to train a generative model (such as GPT-3). This generative model utilizes natural language processing techniques and is capable of generating optimal answers to user questions. Model training is performed using machine learning frameworks such as TensorFlow and PyTorch. The model is regularly updated to reflect the latest information.

[1238] Sentiment analysis engine integration

[1239] The sentiment analysis engine analyzes the text entered by the user through the user interface and recognizes the user's emotions. For example, the sentiment analysis engine identifies emotions such as "joy," "sadness," "anger," and "anxiety" from specific keywords and context. NLP libraries (Natural Language Toolkit and SpaCy) are used for this analysis.

[1240] Providing a user interface

[1241] The user interface is provided in the form of a smartphone app or web portal that is easily accessible to users. This interface is built using React Native and React.js and includes a question input form, a category selection menu, and an optional field for inputting user sentiment.

[1242] Question and Answer Process

[1243] When a user enters a question and presses the send button, the device sends the question and emotional expression to the server. The server uses a generative model to generate an optimal answer to the received question and adjusts the answer according to the user's emotions recognized by the emotion analysis engine. For example, if the user is feeling anxious, the server adds a reassuring message to the answer, such as "Don't worry, appropriate measures are being taken."

[1244] Specific examples

[1245] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[1246] User: Type "How should I breastfeed at the evacuation shelter? I'm worried."

[1247] Device: Sends questions and emotional expressions to the server.

[1248] Server: Using a generative model, it generates answers such as, "Breastfeeding at the evacuation center should be handled as normal, and privacy should be maintained as much as possible. If necessary, please ask the evacuation center manager to provide a separate space for the baby." The sentiment analysis engine recognizes the emotion "worry" and adds messages such as, "Don't worry, we're here to support you."

[1249] Terminal: Displays the generated answer to the user.

[1250] Example 2: Questions about psychological stress and emotion recognition

[1251] User: Type "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious."

[1252] Device: Sends questions and emotional expressions to the server.

[1253] Server: Using a generative model, the server generates answers such as, "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness practice. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The sentiment analysis engine recognizes the emotion "very anxious" and adds messages such as, "We understand your anxiety. Please don't bear it alone."

[1254] Terminal: Displays the generated answer to the user.

[1255] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1256] Step 1: Data collection and database registration

[1257] The server periodically crawls data related to shelter life from government agencies and reliable websites, using web scraping technology to collect reliable information.

[1258] The server automatically categorizes the crawled data into themes, such as "health," "education," and "psychological support."

[1259] The server converts the classified data into Q&A format and registers it in a MySQL database. The input data is the collected text information, and the output data is a database organized in Q&A format.

[1260] Step 2: Training the generative model

[1261] The server trains a generative model (such as GPT-3) using a Q&A dataset stored in a database. This process uses natural language processing techniques, and the model is trained to generate appropriate answers to questions.

[1262] The server evaluates the training results and, if they are insufficient, adds data and performs training again.,The input data is a Q&A format database, and the output data is,the trained generative model.

[1263] The server periodically adds new information to the database and updates the generative model.

[1264] Step 3: Integrating a sentiment analysis engine

[1265] The server embeds a sentiment analysis engine into the user interface, where a sentiment analysis algorithm analyzes the text data to identify sentiment.

[1266] The server analyzes the text entered by the user in real time and identifies emotions such as "joy," "sadness," "anger," and "anxiety" based on keywords and context. The input data is the text entered by the user, and the output data is the result of identifying the emotion.

[1267] Test the accuracy of your sentiment analysis engine and fine-tune the algorithm as needed.

[1268] Step 4: Providing a User Interface

[1269] The server provides users with smartphone apps using React Native and web portals using React.js.

[1270] The terminal displays a question input form and a category selection menu to the user. The input data is the user's operation information, and the output data is the interface display content.

[1271] The user enters a question and enters an emotion in the emotion input field.

[1272] Step 5: Question-answering process

[1273] The user enters a question and presses the send button. For example, "How should I breastfeed at an evacuation shelter? I'm worried."

[1274] The terminal sends the input question and emotional expression to the server. The input data is the user's question and emotional information, and the output data is the information to be sent to the server.

[1275] The server analyzes the question using a generative model and generates an optimal answer. For example, it generates an answer such as, "Breastfeeding at evacuation shelters should be done in the same way as normal breastfeeding, and privacy should be maintained as much as possible. If necessary, please ask the shelter manager to provide a separate space." The input data is the user's question text, and the output data is the generated answer.

[1276] The server uses a sentiment analysis engine to tailor the generated answer depending on the user's emotions, for example adding a message like "Don't worry, we're here to help" if the user is feeling anxious.

[1277] The server sends the final answer to the terminal. The input data is the adjusted answer, and the output data is the information to be sent to the terminal.

[1278] The device will then display the received response to the user, such as, "Breastfeeding at the evacuation shelter should be done in the same way as normal breastfeeding, and please ensure privacy as much as possible. If necessary, please ask the shelter manager to provide a separate space. Don't worry, we'll support you."

[1279] (Application example 2)

[1280] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1281] Conventional food delivery systems only suggest generic menus without considering the user's emotional state. As a result, they are unable to suggest appropriate menus that reflect the user's psychological state, making it difficult to sufficiently improve satisfaction. There is a need for a system that can solve this problem and make highly personalized menu suggestions that correspond to the user's emotional state.

[1282] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for providing information on life in a shelter using a generative model, means for constructing a themed Q&A database, means for loading the Q&A database into a generative model and customizing it to a specific theme, means for providing an interface for accepting questions from users, means for generating optimal answers to user questions using the generative model, means for transmitting the generated answers to the user's terminal, means for analyzing the user's emotions using an emotion engine, and means for adjusting the generated answers depending on the user's emotional state. This makes it possible to propose optimal menus depending on the user's emotional state.

[1283] - A "generative model" is an algorithm that learns from large amounts of data and generates appropriate outputs for given inputs.

[1284] The "Topical Q&A Database" is a database that contains questions and answers related to different themes.

[1285] An "interface" is a means for exchanging data between a user and a system.

[1286] An "emotion engine" is an analysis system that analyzes emotions from user input and recognizes specific emotional states.

[1287] "Analysis" is the process of examining data and understanding its structure and meaning.

[1288] "Terminal" refers to a device through which a user inputs and receives information.

[1289] "Food delivery" refers to the general service of ordering and having food delivered.

[1290] "Menu suggestions" refers to presenting appropriate meal options to the user.

[1291] "Personalization" means customizing something individually to suit a specific user.

[1292] System configuration

[1293] The present invention includes a generative model, an emotion engine, a user interface, and a server-based system for controlling them.

[1294] Collect data from reliable sources and build a thematic Q&A database.

[1295] The generative model learns from this database and generates optimal answers to user questions.

[1296] The emotion engine analyzes the user's input text and recognizes the emotional state.

[1297] The user interface is provided as a smartphone app, allowing intuitive user interaction.

[1298] Program processing

[1299] 1. Data Collection:

[1300] The server automatically collects information related to shelter life and food delivery from official websites and documents issued by specialized organizations.

[1301] The collected data is categorized by theme and registered in a database in Q&A format.

[1302] 2. Training the generative model:

[1303] The server uses the constructed Q&A database to train a generative model (e.g., OpenAI's GPT-3).

[1304] The model will be updated periodically to reflect new information.

[1305] 3. Emotion engine integration:

[1306] The emotion engine analyzes the user's input text and identifies an emotional state.

[1307] For example, it can recognize emotions such as "fatigue" and "stress" from specific keywords and context within the text.

[1308] 4. User Interface:

[1309] It is provided as a smartphone app, allowing users to enter and submit questions.

[1310] The app includes a question form, a category selection menu, and an emotion input option.

[1311] 5. Question and Answer Process:

[1312] When a user inputs a question and presses the send button, the terminal sends the question to the server.

[1313] The server uses a generative model to generate optimal answers and adjusts the answers according to the user's emotional state, as recognized by the emotion engine.

[1314] The generated answer is sent to the user's terminal and displayed.

[1315] Specific examples

[1316] 1. Example 1: Menu suggestions for fatigued users

[1317] User: Type, "I'm feeling tired today, what foods will make me feel better?"

[1318] Server: The emotion engine recognizes the emotion "fatigue," and the generative model generates an appropriate answer: "To recover from fatigue, we recommend fruits rich in vitamin C and lean meat, which contains iron. Also, don't forget to get adequate rest."

[1319] 2. Example 2: Menu suggestions for stressed users

[1320] User: Type "I'm stressed and want some food to help me relax."

[1321] Server: The emotion engine recognizes the emotion "stress," and the generative model generates an appropriate answer: "Herbal tea and yogurt are recommended for relieving stress. Dark chocolate also helps you relax."

[1322] Example prompt sentence:

[1323] 1. "User Question: I'm feeling tired today. What foods will make me feel better?"

[1324] "User question: I'm tired today, what foods will make me feel better?\nUser emotion: Fatigue\nGenerate the best answer."

[1325] 2. "User Question: I'm stressed and want some food to help me relax."

[1326] "User question: I'm stressed and want some food to help me relax.\nUser emotion: Stress\nGenerate the best answer."

[1327] In this way, the invention combines an emotion engine and a generative model to provide optimal information according to the user's emotional state.

[1328] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1329] Step 1:

[1330] The server automatically collects information related to shelter life and food delivery from official websites and documents issued by specialized organizations. It uses APIs or scraping technology to extract data and categorize it by topic. The collected data is then registered in a database. The input is data obtained from reliable sources, and the output is a database in Q&A format, categorized by topic.

[1331] Step 2:

[1332] The server uses the constructed Q&A database to train a generative model (e.g., OpenAI's GPT-3). In this process, the generative model learns from the data using natural language processing so that it can generate appropriate answers to questions. The input is the information in the Q&A database, and the output is the trained generative model.

[1333] Step 3:

[1334] The server uses an emotion engine to analyze emotions from the user's input text. Specifically, the emotion engine analyzes specific keywords and contexts in the text and identifies the emotional state as "fatigue" or "stress," etc. The input is the user's input text, and the output is the analyzed emotional information.

[1335] Step 4:

[1336] The user interface (smartphone app) accepts questions from users. Questions are entered into an input form and sent to the server by pressing the send button. The input is the user's question text, and the output is the question data sent to the server.

[1337] Step 5:

[1338] When the server receives a user's question, it uses a generative model to generate an optimal answer. It then adjusts the answer based on the user's emotional state, as recognized by the emotion engine. For example, if the user is feeling tired, the answer may include specific suggestions such as "foods that will help recover from fatigue." The input is the user's question data and emotional information, and the output is the generated answer.

[1339] Step 6:

[1340] The generated answer is sent from the server to the user's device (smartphone). The user's device displays the received answer. The input is the generated answer data from the server, and the output is the answer displayed on the user's device.

[1341] Specific actions

[1342] Step 1: For example, the server scrapes health information from the official WHO website and categorizes it by topic, such as "health" or "psychological support."

[1343] Step 2: Using the GPT-3 model, we train the model by learning from the collected Q&A data on themes such as "health" and "psychological support."

[1344] Step 3: The emotion engine analyzes the emotion “fatigue” from the text entered by the user: “I feel tired today, please tell me what foods will make me feel better.”

[1345] Step 4: The user interface receives and sends the question "I'm stressed and want some food to help me relax" from the user.

[1346] Step 5: The server generates a response to the user in the "stressed" state, suggesting "herbal tea and yogurt" as "foods that help relieve stress."

[1347] Step 6: The generated answer is displayed in the smartphone app for the user to review.

[1348] In this way, by using the trained generative model and emotion engine, personalized food menu suggestions that are in line with the user's emotional state can be realized.

[1349] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1350] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1351] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1352] [Fourth embodiment]

[1353] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1354] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1355] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1356] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1357] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1358] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1359] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1360] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1361] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1362] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1363] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1364] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1365] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1366] The present invention relates to a system for providing information related to life in an evacuation shelter using a generative model. Specific embodiments of the system are described below.

[1367] System Overview

[1368] The system is based on a server-side generative model and provides a user-friendly interface. The server collects data from reliable sources and builds a topic-specific Q&A database. The generative model learns from this database and provides optimal answers to user questions.

[1369] System configuration

[1370] Data collection and registration

[1371] The server collects information related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. The collected data is then categorized by topic, organized into a Q&A format, and registered in a database. Topics include health, education, and psychological support.

[1372] Training a generative model

[1373] The server uses the constructed Q&A database to train a generative model. The generative model utilizes natural language processing technology and is able to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[1374] User Interface

[1375] The user interface is provided in a form that is easily accessible to users (e.g., smartphone app, web portal), and includes a question input form and options by category, making it intuitive to operate.

[1376] Question and Answer Process

[1377] When a user enters a question and presses the send button, the device sends the question to the server, which uses the generative model to instantly generate the best answer to the received question and sends it back to the user's device. This answer may also include additional related or supporting information as needed.

[1378] Specific examples

[1379] Example 1: Questions about breastfeeding at evacuation shelters

[1380] User: The user types a question: "How should breastfeeding be handled in an evacuation shelter?"

[1381] Terminal: The terminal sends a question to the server.

[1382] Server: The server uses the generative model to analyze the question and generate an appropriate answer, such as "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, with as much privacy as possible. If necessary, ask the evacuation center manager to provide separate spaces."

[1383] Terminal: The terminal displays the generated answer to the user.

[1384] Example 2: Questions about psychological stress

[1385] User: The user types a question: "Please tell me how to take care of myself mentally while living in an evacuation shelter."

[1386] Terminal: The terminal sends a question to the server.

[1387] Server: The server uses the generative model to generate the optimal answer, such as, "Regular rest and relaxation are important when living in an evacuation shelter. It is a good idea to incorporate relaxation techniques and mindfulness practice. Also, if you are experiencing severe psychological stress, we recommend that you seek professional counseling."

[1388] Terminal: The terminal displays the generated answer to the user.

[1389] In this way, the system of the present invention allows users to quickly and accurately obtain information about life in a shelter, thereby improving their quality of life.

[1390] The processing flow will be explained below.

[1391] Step 1: Collect and organize your data

[1392] The server collects information about life in evacuation shelters from reliable sources (e.g., official websites, documents issued by specialized organizations).

[1393] The server categorizes the collected information by topic, for example into categories such as health, education, and psychological support.

[1394] The server creates specific question-and-answer pairs for each topic, building a Q&A format database.

[1395] Step 2: Training the generative model

[1396] The server trains the generative model using a Q&A database.

[1397] The server applies the trained generative model, enabling the model to generate the best answer to the user's question.

[1398] Training is conducted regularly and new information and data is added.

[1399] Step 3: Prepare the user interface

[1400] The device provides a user-friendly interface, accessible through a smartphone app or web portal.

[1401] The terminal provides a question input form and a category selection menu, allowing users to easily input questions.

[1402] Step 4: Receiving questions from users

[1403] The user enters a question through the interface and presses the submit button.

[1404] The terminal transmits the user's question data to the server.

[1405] Step 5: Parsing the question and generating an answer

[1406] The server analyzes the received question data and passes it to the generative model.

[1407] The server uses the generative model to generate optimal answers to the user's questions.

[1408] If necessary, the server will search for additional information (e.g., related resources or supporting information) and append it to the answer.

[1409] Step 6: Submit your response

[1410] The server sends the generated answer and any additional information required to the user's terminal.

[1411] The terminal displays the received answer to the user.

[1412] Step 7: Gather user feedback

[1413] The terminal collects user feedback and sends it to the server.

[1414] The server analyzes the collected feedback and uses it to train the next generative model.

[1415] Through this series of steps, the system of the present invention can quickly and accurately provide information about life in an evacuation shelter, thereby contributing to improving the lives of users.

[1416] Example 1

[1417] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1418] Obtaining information during evacuation shelter life is difficult, and one of the challenges is the lack of timely provision of appropriate information tailored to individual needs and problems. This is due to limited resources, issues with the reliability of information sources, and the difficulty of accurately categorizing and providing information. Furthermore, a system that combines high accuracy and versatility is required to address a wide range of information needs, including those related to psychological stress and health issues.

[1419] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1420] In this invention, the server includes means for collecting data from reliable information sources, means for constructing a Q&A database categorized by theme, means for loading the Q&A database into a generative model to learn information, means for training the generative model to generate optimal answers to user questions, means for providing an interface for accepting questions from users, means for transmitting the generated answers to the user's terminal, and means for displaying the generated answers on the user's terminal, thereby enabling users to quickly and accurately obtain a wide range of information related to life in a shelter.

[1421] "Reliable sources" are institutions or platforms that provide accurate and up-to-date information, such as official websites or documents issued by specialized organizations.

[1422] A "Q&A database" is a database in which questions and their answers are formalized and stored, and each question and answer is categorized by topic.

[1423] A "generative model" is a machine learning model that uses natural language processing technology to generate optimal answers to user questions.

[1424] An "interface" is the means by which users access the system and input questions, and may be provided in the form of a smartphone app or web portal.

[1425] A "terminal" is a device that a user uses to use an interface, such as a smartphone or computer.

[1426] "Natural language processing technology" is a technology for processing human language using a computer, and is a technique for analyzing, understanding, and generating text.

[1427] "Additional related or supporting information" is supplemental content, such as more detailed information or advice, that accompanies the answer to the user's question.

[1428] The present invention relates to a system for providing information related to life in an evacuation shelter using a generative AI model. Specific embodiments for carrying out the present invention will be described below.

[1429] Overall system picture

[1430] The system is based on a generative AI model running on the server side and has a user-accessible interface. The server collects data from reliable sources and builds a Q&A database categorized by topic. This database is used to train the generative AI model, which generates optimal answers to user questions. The system also has a function for sending and displaying answers to the user's device.

[1431] Data collection and registration

[1432] The server collects data related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. Specifically, it automatically obtains the latest information using RSS feeds and APIs. The collected data is categorized by topic, organized into a Q&A format, and entered into a database. This database is divided into themes such as health, education, and psychological support.

[1433] Training a generative model

[1434] The server uses the constructed Q&A database to train a generative AI model. One example of a generative AI model used is GPT-4. This model utilizes natural language processing technology, enabling it to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[1435] Providing a user interface

[1436] The user interface is provided in a form that is easy for users to access (e.g., smartphone app, web portal), using technologies such as React Native and React.js. The interface includes a question input form and options by category, making it intuitive to use.

[1437] Question and Answer Process

[1438] When a user enters a question into the interface and presses the send button, the device sends the question to the server. The server uses a generative AI model to generate the best answer for the received question. The answer may include additional related or supporting information as needed. The generated answer is then displayed to the user through the device.

[1439] Specific examples

[1440] Example 1: Questions about breastfeeding at evacuation shelters

[1441] User: The user types a question: "How should breastfeeding be handled in an evacuation shelter?"

[1442] Terminal: The terminal sends a question to the server.

[1443] Server: The server uses a generative AI model to analyze the question and generate an appropriate answer, such as "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, with as much privacy as possible. If necessary, ask the evacuation center manager to provide separate spaces."

[1444] Terminal: The terminal displays the generated answer to the user.

[1445] Example 2: Questions about psychological stress

[1446] User: The user types a question: "Please tell me how to take care of myself mentally while living in an evacuation shelter."

[1447] Terminal: The terminal sends a question to the server.

[1448] Server: The server uses a generative AI model to generate optimal answers, such as, "Regular rest and relaxation are important when living in an evacuation shelter. It's a good idea to incorporate relaxation techniques and mindfulness exercises. If you're experiencing significant psychological stress, we recommend seeking professional counseling."

[1449] Terminal: The terminal displays the generated answer to the user.

[1450] In this way, the system of the present invention allows users to quickly and accurately obtain information about life in a shelter, thereby improving their quality of life.

[1451] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1452] Step 1: Data collection step

[1453] The server automatically collects data related to shelter life from reliable sources (e.g., official websites, documents issued by specialized organizations), using RSS feeds and APIs to obtain the latest information.

[1454] Input: Source URL or API endpoint

[1455] Specific operation: The server sends an "HTTP request" to obtain data from the information source.

[1456] Output: Retrieved data (e.g. HTML, PDF, JSON)

[1457] Step 2: Data classification and registration step

[1458] The server categorizes the collected data by topic, organizes it into a Q&A format, and registers it in a database. Data categorization is performed using tagging technology.

[1459] Input: Collected data

[1460] How it works: The server uses natural language processing technology to analyze the data and classify it into categories such as "health," "education," and "psychological support." The classified data is then converted into a Q&A format and stored in a database.

[1461] Output: Q&A database

[1462] Step 3: Data preparation steps for the generative model

[1463] The server extracts data from the Q&A database and formats it as a training dataset for the generative model.

[1464] Input: Q&A database

[1465] What happens: The server extracts the Q&A data from the database and converts it into an appropriate format (e.g., JSON).

[1466] Output: Training dataset

[1467] Step 4: Training the generative model

[1468] The server uses the extracted data to train a generative model, such as GPT-4.

[1469] Input: Training dataset

[1470] Specific operation: The server executes a Python script and performs processing to train a generative model (GPT-4).

[1471] Output: A trained generative model

[1472] Step 5: Model Update Step

[1473] The server periodically retrieves new data and retrains the model with the updated data.

[1474] Input: New data and existing Q&A database

[1475] What happens: The server integrates the newly collected data with the existing database and retrains the generative model.

[1476] Output: Updated generative model

[1477] Step 6: Interface development step

[1478] The server develops the user interface in a form that is easy for users to access, using technologies such as React Native and React.js.

[1479] Input: Interface design and functional requirements

[1480] What it does: A developer uses React.js to build a web portal and adds a question form and options based on categories.

[1481] Output: The finished user interface

[1482] Step 7: Interface Operation Steps

[1483] The server monitors the operational interfaces and performs maintenance and updates as needed.

[1484] Input: User feedback and operational data

[1485] What it does: The server periodically collects user feedback and makes bug fixes and feature improvements.

[1486] Output: User interface in operation

[1487] Step 8: Question input step

[1488] The user accesses the interface and enters a question. The question entry form includes fields for selecting a category and entering specific questions.

[1489] Input: User question

[1490] What happens: The user types into the interface, "How should breastfeeding be done in an evacuation shelter?"

[1491] Output: Question data

[1492] Step 9: Submit a question

[1493] The device sends the questions entered by the user to the server, and the data is sent in JSON format.

[1494] Input: User question data

[1495] Specific operation: The device sends a POST request to the API endpoint.

[1496] Output: Query data to the server

[1497] Step 10: Question Analysis Step

[1498] The server analyzes the received question and formats it as data to be input into the generative model.

[1499] Input: Received question data

[1500] Specific behavior: The server parses and preprocesses the received query (e.g., tokenization and normalization).

[1501] Output: Well-formed question data

[1502] Step 11: Answer generation step

[1503] The server uses the generative model to generate optimal answers to questions.

[1504] Input: Well-formed question data

[1505] How it works: The server inputs question data into the generative model and generates an answer, such as, "Breastfeeding at evacuation centers should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible. If necessary, ask the evacuation center manager to provide a separate space."

[1506] Output: Response data

[1507] Step 12: Send response step

[1508] The server then sends the generated response to the user's device. The data is sent in JSON format.

[1509] Input: Response data

[1510] Specific operation: The server sends a POST request to the API endpoint and delivers the response to the device.

[1511] Output: Answer data to the user's device

[1512] Step 13: Answer display step

[1513] The terminal displays the received response to the user, and the interface formats and displays the received data.

[1514] Input: Received response data

[1515] Specific operation: The device displays the received response on the user interface, saying, "Breastfeeding at the evacuation center will be conducted in the same way as normal breastfeeding, and privacy will be ensured as much as possible. If necessary, ask the evacuation center administrator to provide a separate space."

[1516] Output: Answers displayed on many users' terminals

[1517] (Application example 1)

[1518] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1519] Conventional information provision systems for life in evacuation shelters often do not have a dedicated platform, making it difficult for users to quickly and accurately obtain the information they need. Furthermore, in emergencies and disasters, there is a need to provide security information and specific advice on evacuation routes, but no system exists that can meet this need.

[1520] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1521] In this invention, the server includes: means for providing information on life in a shelter using a generative model; means for building a themed Q&A database; means for loading the Q&A database into a generative model and customizing it to a specific theme; means for providing an interface for accepting questions from users; means for generating optimal answers to user questions using the generative model; means for transmitting the generated answers to the user's terminal; means for acquiring disaster safety information such as security information from a reliable information source in JSON format; and means for the generative model to provide the necessary answers based on the acquired data and support evacuation behavior in an emergency, thereby enabling users to obtain quick and accurate information via their smartphones or other terminals.

[1522] A "generative model" is an algorithm that generates new information or answers based on the data provided.

[1523] "Shelter life" refers to the living environment and lifestyle in a temporary shelter set up in the event of a disaster or emergency.

[1524] An "information providing system" is a computer system for providing necessary information to users.

[1525] A "topical Q&A database" is a database that organizes and stores data in the form of questions and answers for specific themes.

[1526] An "interface" is the means or screen through which a user interacts with a computer system or application.

[1527] "Natural language processing" is a technology that uses computers to analyze, process, and generate natural language used by humans.

[1528] "Security information" refers to various information provided to ensure the safety of users, and particularly includes information in emergencies and disasters.

[1529] The "JSON format" is a common format for expressing data in text format and is one of the formats widely used for data exchange.

[1530] System Overview

[1531] The present invention relates to a system that uses generative models to provide information on life in evacuation shelters and security during emergencies. The system is based on a generative model executed on the server side and has an interface that is easily accessible to users.

[1532] Data collection and registration

[1533] The server collects information related to shelter life and security in JSON format from reliable sources, then categorizes the collected data by topic, organizes it into a Q&A format, and registers it in a database that covers topics such as health, education, psychological support, and security information.

[1534] Training a generative model

[1535] The server uses the constructed Q&A database to train a generative model. This generative model utilizes natural language processing technology to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[1536] User Interface

[1537] The user interface is provided as a smartphone app or web portal. The interface is intuitive and includes a question input form and category-based options. Users can enter their questions through this interface and immediately obtain the information they need.

[1538] Question and Answer Process

[1539] When a user enters a question and presses the send button, the device sends the question to the server, which uses the generative model to generate the best answer to the question and sends it back to the user's device. This answer may also include additional relevant or security information, if necessary.

[1540] Specific examples

[1541] Example 1: Selection of evacuation routes

[1542] User: The user types a question: "What is the best evacuation route in case of a flood?"

[1543] Terminal: The terminal sends a question to the server.

[1544] Server: The server uses the generative model to analyze the question and generate an appropriate answer, such as "In the event of a flood, it is best to evacuate to higher ground. The nearest evacuation shelter is the one on the higher ground in XX Park."

[1545] Terminal: The terminal displays the generated answer to the user.

[1546] Example 2: Emergency contact method

[1547] User: The user types a question: "How do I contact my family in case of an earthquake?"

[1548] Terminal: The terminal sends a question to the server.

[1549] Server: The server uses the generative model to generate the optimal answer. For example, it might generate an answer such as, "Since congestion is expected during an earthquake, it's best to communicate using text messages or social media. It's also a good idea to decide on a meeting place at an evacuation shelter."

[1550] Terminal: The terminal displays the generated answer to the user.

[1551] Hardware and software used

[1552] This system mainly performs data processing on the server side. The server requires a high-performance processor and a large amount of memory, so it is recommended to use a cloud service (e.g., Amazon Web Services, Google Cloud Platform). To run the generative model, we use a library specifically designed for natural language processing (e.g., Transformers by Hugging Face).

[1553] Prompt Sentence Examples

[1554] Here are some examples of prompts:

[1555] "What is the safest evacuation route in the event of a flood?"

[1556] "What is the best way to contact my family in the event of an earthquake?"

[1557] "Please tell me how to take care of your mental health while living in an evacuation shelter."

[1558] As described above, the present invention allows users to quickly and accurately obtain necessary information in the event of an emergency or disaster.

[1559] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1560] Step 1:

[1561] The server collects information related to shelter life and security from reliable sources in JSON format. This allows it to obtain the necessary data from multiple official websites and documents issued by specialized organizations. The input is the URL of each source, and the output is the retrieved JSON data. The server then organizes the information into a database.

[1562] Step 2:

[1563] The server categorizes the collected data by theme, formats it in Q&A format, and registers it in a database. The input is the JSON data obtained in step 1, and the output is a database organized in Q&A format. The server uses natural language processing technology to categorize each piece of information by theme.

[1564] Step 3:

[1565] The server trains a generative model using a Q&A database it has built. The input is the Q&A database, and the output is the trained generative model. The server uses libraries (e.g., Hugging Face Transformers) to build a generative model customized for a specific theme.

[1566] Step 4:

[1567] A user inputs a question through a smartphone app or web portal. The input is the user's question, and the output is the question data. The device provides the user interface, and the user intuitively operates it to input the question.

[1568] Step 5:

[1569] The terminal sends the user's query data to the server. The input is the query data entered by the user in step 4, and the output is the query data sent to the server. The terminal sends the data to the server via the network.

[1570] Step 6:

[1571] The server uses a generative model to generate the best answer to the received question. The input is the user's question data and a Q&A database, and the output is the generated answer. The server uses natural language processing technology to have the model analyze the user's question and generate an answer based on related information.

[1572] Step 7:

[1573] The server sends the generated answer to the user's terminal. The input is the generated answer, and the output is the answer sent to the user's terminal. The server sends the answer data to the terminal via the network, and the terminal receives it.

[1574] Step 8:

[1575] The terminal displays the received answer to the user. The input is the answer data received from the server, and the output is the answer displayed to the user. The terminal displays the answer through a user interface, and the user confirms it.

[1576] As a result, the user can quickly obtain accurate information.

[1577] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1578] The present invention relates to a system that uses a generative model to provide information related to life in a shelter, and by combining it with an emotion engine, provides optimal information according to the user's emotional state. Specific embodiments of this system are described below.

[1579] System Overview

[1580] This system is based on a generative model, emotion engine, and user interface running on the server side. The server collects data from reliable sources and builds a topic-specific Q&A database. The generative model learns from this database and provides optimal answers to user questions. The emotion engine also analyzes the user's emotions and appropriately adjusts the generated answers based on those emotions.

[1581] System configuration

[1582] Data collection and registration

[1583] The server collects information related to life in evacuation shelters from reliable sources, such as official websites and documents issued by specialized organizations. The collected data is then categorized by topic, organized into a Q&A format, and registered in a database. Topics include health, education, and psychological support.

[1584] Training a generative model

[1585] The server uses the constructed Q&A database to train a generative model. The generative model utilizes natural language processing technology and is able to generate optimal answers to user questions. The model is regularly updated to reflect new information.

[1586] Emotion Engine Integration

[1587] The emotion engine analyzes text entered through the user interface and recognizes the user's emotions. For example, it can detect emotions such as gratitude, anger, and anxiety from specific keywords and contexts contained in the user's text.

[1588] User Interface

[1589] The user interface is provided in a form that is easily accessible to users (e.g., a smartphone app or web portal). The interface is intuitive and includes a question input form, category selection menu, etc. It also provides an optional field for inputting user emotions.

[1590] Question and Answer Process

[1591] When a user enters a question and presses the send button, the device sends the question to the server. The server uses the generative model to instantly generate the best answer for the received question, adjusting the answer according to the user's emotions as recognized by the emotion engine. For example, if the user is feeling anxious, the answer may be more polite and reassuring. The answer may also include additional related or supporting information as needed.

[1592] Specific examples

[1593] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[1594] User: The user types, "How should I breastfeed at the evacuation shelter? I'm worried."

[1595] Device: The device sends the question and emotional expression to the server.

[1596] Server: The server uses a generative model to analyze the question and generate an appropriate answer. For example, it might generate an answer such as, "Breastfeeding at the evacuation center should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible. If necessary, ask the evacuation center manager to provide a separate space." The emotion engine recognizes the user's "worry" emotion and adds a message such as, "Don't worry, we'll support you."

[1597] Terminal: The terminal displays the generated answer to the user.

[1598] Example 2: Questions about psychological stress and emotion recognition

[1599] User: The user types, "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious."

[1600] Device: The device sends the question and emotional expression to the server.

[1601] Server: The server uses a generative model to generate the optimal answer. For example, it might generate an answer such as, "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness practice. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The emotion engine recognizes the emotion "very anxious" and adds a message to the answer such as, "We understand your anxiety. Please don't suffer alone."

[1602] Terminal: The terminal displays the generated answer to the user.

[1603] In this way, by combining a generative model and an emotion engine, the system of the present invention can provide appropriate information according to the user's emotional state, thereby improving the quality of information acquisition during life in an evacuation shelter and supporting the user's daily life.

[1604] The processing flow will be explained below.

[1605] Step 1: Collect and organize your data

[1606] The server collects information about life in evacuation shelters from reliable sources (e.g., official websites, documents issued by specialized organizations).

[1607] The server categorizes the collected information by topic, for example into categories such as health, education, and psychological support.

[1608] The server creates specific question-and-answer pairs for each topic, building a Q&A format database.

[1609] Step 2: Training the generative model

[1610] The server trains the generative model using a Q&A database.

[1611] The server applies the trained generative model, enabling the model to generate the best answer to the user's question.

[1612] Training is conducted regularly and new information and data is added.

[1613] Step 3: Training and Integrating the Emotion Engine

[1614] The server trains the emotion engine using the user's text data.

[1615] The server integrates the emotion engine with the generative model to recognize emotions from user questions.

[1616] The emotion engine includes algorithms that detect emotions from specific keywords and contexts.

[1617] Step 4: Prepare the User Interface

[1618] The device provides a user-friendly interface, accessible through a smartphone app or web portal.

[1619] The terminal provides a question input form and a category selection menu, allowing users to easily input questions.

[1620] The device provides an emotion input option, allowing the user to explicitly input emotions.

[1621] Step 5: Receiving user questions

[1622] The user enters a question and emotion through the interface and presses the send button.

[1623] The terminal transmits the user's question data and emotion data to the server.

[1624] Step 6: Question and Sentiment Analysis

[1625] The server analyzes the received question data and emotion data.

[1626] The server passes the question data to the generative model to generate the optimal answer.

[1627] The server passes the emotion data to the emotion engine, which analyzes the user's emotional state.

[1628] Step 7: Adjust your responses based on emotion

[1629] The server adjusts the generated answer based on the emotion recognized by the emotion engine.

[1630] For example, if the user is feeling anxious, add reassuring language to the answer.

[1631] Step 8: Submit your response

[1632] The server transmits the generated answer and the adjusted answer based on the emotion to the user's terminal.

[1633] The terminal displays the received answer to the user.

[1634] Step 9: Gather user feedback

[1635] The terminal collects user feedback and sends it to the server.

[1636] The server analyzes the collected feedback and uses it to train the generative model and emotion engine next time.

[1637] Specific examples

[1638] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[1639] Step 5:

[1640] The user types, "How should I breastfeed at an evacuation shelter? I'm worried," and presses the send button.

[1641] Step 6:

[1642] The terminal transmits the question and emotion data to the server.

[1643] The server passes the question data to a generative model to generate the optimal answer.

[1644] The server passes the emotional data "I'm worried" to the emotion engine, which then recognizes the emotion.

[1645] Step 7:

[1646] The server generates an answer: "Breastfeeding in evacuation shelters should be conducted in the same way as normal breastfeeding, and privacy should be ensured as much as possible." The answer is then adjusted using the emotion engine.

[1647] The user is worried, so you add "Don't worry, we're here to help" to your answer.

[1648] Step 8:

[1649] The server sends the adjusted response to the user's terminal.

[1650] The terminal displays the generated answer to the user.

[1651] Example 2: Questions about psychological stress and emotion recognition

[1652] Step 5:

[1653] The user types, "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious," and presses the send button.

[1654] Step 6:

[1655] The terminal transmits the question and emotion data to the server.

[1656] The server passes the question data to a generative model to generate the optimal answer.

[1657] The server passes the emotional data "I'm very anxious" to the emotion engine, which then recognizes the emotion.

[1658] Step 7:

[1659] The server generates an answer: "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness exercises. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The answer is then adjusted using the emotion engine.

[1660] The user is feeling very anxious, so add "We understand your anxiety. Please don't go through it alone" to your response.

[1661] Step 8:

[1662] The server sends the adjusted response to the user's terminal.

[1663] The terminal displays the generated answer to the user.

[1664] In this way, by combining a generative model and an emotion engine, the system of the present invention can provide appropriate information according to the user's emotional state, thereby improving the quality of information acquisition during life in an evacuation shelter and supporting the user's daily life.

[1665] Example 2

[1666] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1667] When living in an evacuation shelter, there is a need for a method that allows users to easily obtain the information they need. In particular, it is important to provide optimal information according to the user's emotional state. However, conventional systems have difficulty providing information that reflects the user's emotions, and have not been able to provide sufficient psychological support during evacuation shelter life.

[1668] The identification process 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 providing information on life in a shelter using a generative model, means for constructing a thematic information base, means for loading the information base into a generative model and customizing it to a specific theme, means for providing a user interface for accepting questions from users, means for generating optimal answers to the user's questions using the generative model, means for transmitting the generated answers to the user's information device, and means for analyzing the user's emotions using an emotion analysis engine and reflecting the emotions in the answers. This makes it possible to provide optimal information according to the user's emotional state.

[1669] 1. "Generative model" is an artificial intelligence technique for generating natural language responses from collected data.

[1670] 2. The "Thematic Information Base" is a database constructed by classifying information related to life in evacuation shelters by theme.

[1671] 3. "Customization" means tailoring information to a specific topic or user needs and converting it into an appropriate format.

[1672] 4. "User interface" refers to interactive mechanisms such as input forms and menu screens that allow users to interact with a system.

[1673] 5. "User's information equipment" refers to electronic devices used by the User, such as smartphones, tablets, and personal computers.

[1674] 6. "Sentiment analysis engine" is software that analyzes text entered by a user and identifies the emotions contained therein.

[1675] 7. "Natural language processing" is a technology that allows computers to understand and generate human language.

[1676] 8. "Life in a shelter" refers to a situation in which people temporarily live in a shelter due to a disaster or other reason.

[1677] The present invention is a system that provides information related to life in an evacuation shelter, and by combining a generative model and an emotion analysis engine, it is possible to provide optimal information according to the user's emotional state. This system has the following configuration.

[1678] Data collection and database construction

[1679] The server collects data on life in evacuation shelters from government agencies and other reliable sources. The collected data is categorized by theme, such as "health," "education," and "psychological support." The server then organizes the data into a Q&A format and registers it in a database as a thematic information base. This database is managed using a relational database management system such as MySQL.

[1680] Training a generative model

[1681] The server uses the thematic information base to train a generative model (such as GPT-3). This generative model utilizes natural language processing techniques and is capable of generating optimal answers to user questions. Model training is performed using machine learning frameworks such as TensorFlow and PyTorch. The model is regularly updated to reflect the latest information.

[1682] Sentiment analysis engine integration

[1683] The sentiment analysis engine analyzes the text entered by the user through the user interface and recognizes the user's emotions. For example, the sentiment analysis engine identifies emotions such as "joy," "sadness," "anger," and "anxiety" from specific keywords and context. NLP libraries (Natural Language Toolkit and SpaCy) are used for this analysis.

[1684] Providing a user interface

[1685] The user interface is provided in the form of a smartphone app or web portal that is easily accessible to users. This interface is built using React Native and React.js and includes a question input form, a category selection menu, and an optional field for inputting user sentiment.

[1686] Question and Answer Process

[1687] When a user enters a question and presses the send button, the device sends the question and emotional expression to the server. The server uses a generative model to generate an optimal answer to the received question and adjusts the answer according to the user's emotions recognized by the emotion analysis engine. For example, if the user is feeling anxious, the server adds a reassuring message to the answer, such as "Don't worry, appropriate measures are being taken."

[1688] Specific examples

[1689] Example 1: Questions about breastfeeding in evacuation shelters and emotion recognition

[1690] User: Type "How should I breastfeed at the evacuation shelter? I'm worried."

[1691] Device: Sends questions and emotional expressions to the server.

[1692] Server: Using a generative model, it generates answers such as, "Breastfeeding at the evacuation center should be handled as normal, and privacy should be maintained as much as possible. If necessary, please ask the evacuation center manager to provide a separate space for the baby." The sentiment analysis engine recognizes the emotion "worry" and adds messages such as, "Don't worry, we're here to support you."

[1693] Terminal: Displays the generated answer to the user.

[1694] Example 2: Questions about psychological stress and emotion recognition

[1695] User: Type "Please tell me how to take care of myself mentally while living in an evacuation shelter. I'm very anxious."

[1696] Device: Sends questions and emotional expressions to the server.

[1697] Server: Using a generative model, the server generates answers such as, "Regular rest and relaxation are important when living in a shelter. It's a good idea to incorporate relaxation techniques and mindfulness practice. If you're experiencing severe psychological stress, we recommend seeking professional counseling." The sentiment analysis engine recognizes the emotion "very anxious" and adds messages such as, "We understand your anxiety. Please don't bear it alone."

[1698] Terminal: Displays the generated answer to the user.

[1699] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1700] Step 1: Data collection and database registration

[1701] The server periodically crawls data related to shelter life from government agencies and reliable websites, using web scraping technology to collect reliable information.

[1702] The server automatically categorizes the crawled data into themes, such as "health," "education," and "psychological support."

[1703] The server converts the classified data into Q&A format and registers it in a MySQL database. The input data is the collected text information, and the output data is a database organized in Q&A format.

[1704] Step 2: Training the generative model

[1705] The server trains a generative model (such as GPT-3) using a Q&A dataset stored in the database. This process uses natural language processing techniques, and the model is trained to generate appropriate answers to questions.

[1706] The server evaluates the training results and, if they are insufficient, adds data and performs training again.,The input data is a Q&A format database, and the output data is,the trained generative model.

[1707] The server periodically adds new information to the database and updates the generative model.

[1708] Step 3: Integrating a sentiment analysis engine

[1709] The server embeds a sentiment analysis engine into the user interface, where a sentiment analysis algorithm analyzes the text data to identify sentiment.

[1710] The server analyzes the text entered by the user in real time and identifies emotions such as "joy," "sadness," "anger," and "anxiety" based on keywords and context. The input data is the text entered by the user, and the output data is the result of identifying the emotion.

[1711] Test the accuracy of your sentiment analysis engine and fine-tune the algorithm as needed.

[1712] Step 4: Providing a User Interface

[1713] The server provides users with smartphone apps using React Native and web portals using React.js.

[1714] The terminal displays a question input form and a category selection menu to the user. The input data is the user's operation information, and the output data is the interface display content.

[1715] The user enters a question and enters an emotion in the emotion input field.

[1716] Step 5: Question-answering process

[1717] The user enters a question and presses the send button. For example, "How should I breastfeed at an evacuation shelter? I'm worried."

[1718] The terminal sends the input question and emotional expression to the server. The input data is the user's question and emotional information, and the output data is the information to be sent to the server.

[1719] The server analyzes the question using a generative model and generates an optimal answer. For example, it generates an answer such as, "Breastfeeding at evacuation shelters should be done in the same way as normal breastfeeding, and privacy should be maintained as much as possible. If necessary, please ask the shelter manager to provide a separate space." The input data is the user's question text, and the output data is the generated answer.

[1720] The server uses a sentiment analysis engine to tailor the generated answer depending on the user's emotions, for example adding a message like "Don't worry, we're here to help" if the user is feeling anxious.

[1721] The server sends the final answer to the terminal. The input data is the adjusted answer, and the output data is the information to be sent to the terminal.

[1722] The device will then display the received response to the user, such as, "Breastfeeding at the evacuation shelter should be done in the same way as normal breastfeeding, and please ensure privacy as much as possible. If necessary, please ask the shelter manager to provide a separate space. Don't worry, we'll support you."

[1723] (Application example 2)

[1724] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1725] Conventional food delivery systems only suggest generic menus without considering the user's emotional state. As a result, they are unable to suggest appropriate menus that reflect the user's psychological state, making it difficult to sufficiently improve satisfaction. There is a need for a system that can solve this problem and make highly personalized menu suggestions that correspond to the user's emotional state.

[1726] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for providing information on life in a shelter using a generative model, means for constructing a themed Q&A database, means for loading the Q&A database into a generative model and customizing it to a specific theme, means for providing an interface for accepting questions from users, means for generating optimal answers to user questions using the generative model, means for transmitting the generated answers to the user's terminal, means for analyzing the user's emotions using an emotion engine, and means for adjusting the generated answers depending on the user's emotional state. This makes it possible to propose optimal menus depending on the user's emotional state.

[1727] - A "generative model" is an algorithm that learns from large amounts of data and generates appropriate outputs for given inputs.

[1728] The "Topical Q&A Database" is a database that contains questions and answers related to different themes.

[1729] An "interface" is a means for exchanging data between a user and a system.

[1730] An "emotion engine" is an analysis system that analyzes emotions from user input and recognizes specific emotional states.

[1731] "Analysis" is the process of examining data and understanding its structure and meaning.

[1732] "Terminal" refers to a device through which a user inputs and receives information.

[1733] "Food delivery" refers to the general service of ordering and having food delivered.

[1734] "Menu suggestions" refers to presenting appropriate meal options to the user.

[1735] "Personalization" means customizing something individually to suit a specific user.

[1736] System configuration

[1737] The present invention includes a generative model, an emotion engine, a user interface, and a server-based system for controlling them.

[1738] Collect data from reliable sources and build a thematic Q&A database.

[1739] The generative model learns from this database and generates optimal answers to user questions.

[1740] The emotion engine analyzes the user's input text and recognizes the emotional state.

[1741] The user interface is provided as a smartphone app, allowing intuitive user interaction.

[1742] Program processing

[1743] 1. Data Collection:

[1744] The server automatically collects information related to shelter life and food delivery from official websites and documents issued by specialized organizations.

[1745] The collected data is categorized by theme and registered in a database in Q&A format.

[1746] 2. Training the generative model:

[1747] The server uses the constructed Q&A database to train a generative model (e.g., OpenAI's GPT-3).

[1748] The model will be updated periodically to reflect new information.

[1749] 3. Emotion engine integration:

[1750] The emotion engine analyzes the user's input text and identifies an emotional state.

[1751] For example, it can recognize emotions such as "fatigue" and "stress" from specific keywords and context within the text.

[1752] 4. User Interface:

[1753] It is provided as a smartphone app, allowing users to enter and submit questions.

[1754] The app includes a question form, a category selection menu, and an emotion input option.

[1755] 5. Question and Answer Process:

[1756] When a user inputs a question and presses the send button, the terminal sends the question to the server.

[1757] The server uses a generative model to generate optimal answers and adjusts the answers according to the user's emotional state, as recognized by the emotion engine.

[1758] The generated answer is sent to the user's terminal and displayed.

[1759] Specific examples

[1760] 1. Example 1: Menu suggestions for fatigued users

[1761] User: Type, "I'm feeling tired today, what foods will make me feel better?"

[1762] Server: The emotion engine recognizes the emotion "fatigue," and the generative model generates an appropriate answer: "To recover from fatigue, we recommend fruits rich in vitamin C and lean meat, which contains iron. Also, don't forget to get adequate rest."

[1763] 2. Example 2: Menu suggestions for stressed users

[1764] User: Type "I'm stressed and want some food to help me relax."

[1765] Server: The emotion engine recognizes the emotion "stress," and the generative model generates an appropriate answer: "Herbal tea and yogurt are recommended for relieving stress. Dark chocolate also helps you relax."

[1766] Example prompt sentence:

[1767] 1. "User Question: I'm feeling tired today. What foods will make me feel better?"

[1768] "User question: I'm tired today, what foods will make me feel better?\nUser emotion: Fatigue\nGenerate the best answer."

[1769] 2. "User Question: I'm stressed and want some food to help me relax."

[1770] "User question: I'm stressed and want some food to help me relax.\nUser emotion: Stress\nGenerate the best answer."

[1771] In this way, the invention combines an emotion engine and a generative model to provide optimal information according to the user's emotional state.

[1772] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1773] Step 1:

[1774] The server automatically collects information related to shelter life and food delivery from official websites and documents issued by specialized organizations. It uses APIs or scraping technology to extract data and categorize it by topic. The collected data is then registered in a database. The input is data obtained from reliable sources, and the output is a database in Q&A format, categorized by topic.

[1775] Step 2:

[1776] The server uses the constructed Q&A database to train a generative model (e.g., OpenAI's GPT-3). In this process, the generative model learns from the data using natural language processing so that it can generate appropriate answers to questions. The input is the information in the Q&A database, and the output is the trained generative model.

[1777] Step 3:

[1778] The server uses an emotion engine to analyze emotions from the user's input text. Specifically, the emotion engine analyzes specific keywords and contexts in the text and identifies the emotional state as "fatigue" or "stress," etc. The input is the user's input text, and the output is the analyzed emotional information.

[1779] Step 4:

[1780] The user interface (smartphone app) accepts questions from users. Questions are entered into an input form and sent to the server by pressing the send button. The input is the user's question text, and the output is the question data sent to the server.

[1781] Step 5:

[1782] When the server receives a user's question, it uses a generative model to generate an optimal answer. It then adjusts the answer based on the user's emotional state, as recognized by the emotion engine. For example, if the user is feeling tired, the answer may include specific suggestions such as "foods that will help recover from fatigue." The input is the user's question data and emotional information, and the output is the generated answer.

[1783] Step 6:

[1784] The generated answer is sent from the server to the user's device (smartphone). The user's device displays the received answer. The input is the generated answer data from the server, and the output is the answer displayed on the user's device.

[1785] Specific actions

[1786] Step 1: For example, the server scrapes health information from the official WHO website and categorizes it by topic, such as "health" or "psychological support."

[1787] Step 2: Using the GPT-3 model, we train the model by learning from the collected Q&A data on themes such as "health" and "psychological support."

[1788] Step 3: The emotion engine analyzes the emotion “fatigue” from the text entered by the user: “I feel tired today, please tell me what foods will make me feel better.”

[1789] Step 4: The user interface receives and sends the question "I'm stressed and want some food to help me relax" from the user.

[1790] Step 5: The server generates a response to the user in the "stressed" state, suggesting "herbal tea and yogurt" as "foods that help relieve stress."

[1791] Step 6: The generated answer is displayed in the smartphone app for the user to review.

[1792] In this way, by using the trained generative model and emotion engine, personalized food menu suggestions that are in line with the user's emotional state can be realized.

[1793] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1794] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1795] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1796] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1797] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1798] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1799] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1800] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1801] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1802] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1803] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1804] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1805] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1806] 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.

[1807] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1808] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1809] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1810] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1811] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1812] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1813] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1814] The following is further disclosed regarding the above embodiment.

[1815] (Claim 1)

[1816] A means for providing information on life in a shelter using a generative model;

[1817] A means to build a thematic Q&A database;

[1818] A means for loading the Q&A database into a generative model and customizing it to a specific theme;

[1819] means for providing an interface for accepting questions from a user;

[1820] means for generating an optimal answer to a user's question using the generative model;

[1821] means for transmitting the generated answer to a user's terminal;

[1822] A system including:

[1823] (Claim 2)

[1824] 10. The system of claim 1, wherein the generated answer includes additional information.

[1825] (Claim 3)

[1826] 2. The system of claim 1, wherein the generative model uses natural language processing.

[1827] "Example 1"

[1828] (Claim 1)

[1829] The means to collect data from reliable sources;

[1830] A means to build a thematically categorized Q&A database;

[1831] A means for reading the Q&A database into a generative model and learning information;

[1832] means for training the generative model to generate optimal answers to user questions;

[1833] means for providing an interface for receiving questions from a user;

[1834] means for transmitting the generated answer to a user's terminal;

[1835] means for displaying the generated answer on a user's terminal;

[1836] A system including:

[1837] (Claim 2)

[1838] 10. The system of claim 1, wherein the generated answer includes additional related or supporting information.

[1839] (Claim 3)

[1840] 2. The system according to claim 1, wherein the generative model uses natural language processing techniques.

[1841] "Application Example 1"

[1842] (Claim 1)

[1843] A means for providing information on life in a shelter using a generative model;

[1844] A means to build a thematic Q&A database;

[1845] A means for loading the Q&A database into a generative model and customizing it to a specific theme;

[1846] means for providing an interface for accepting questions from a user;

[1847] means for generating an optimal answer to a user's question using the generative model;

[1848] means for transmitting the generated answer to a user's terminal;

[1849] A means to obtain disaster safety information such as security information from reliable information sources in JSON format, and

[1850] A means for a generative model to provide the necessary answers based on the acquired data and support evacuation behavior in an emergency.

[1851] A system including:

[1852] (Claim 2)

[1853] 10. The system of claim 1, wherein the generated answer includes additional information.

[1854] (Claim 3)

[1855] 2. The system of claim 1, wherein the generative model uses natural language processing.

[1856] "Example 2: Combining Emotion Engines"

[1857] (Claim 1)

[1858] A means for providing information on life in a shelter using a generative model;

[1859] means for building thematic information bases;

[1860] means for loading said information base into a generative model and customizing it to a particular theme;

[1861] means for providing a user interface for accepting questions from a user;

[1862] means for generating an optimal answer to a user's question using the generative model;

[1863] means for transmitting the generated response to a user's information device;

[1864] means for analyzing the user's emotions using an emotion analysis engine and reflecting the emotions in the response;

[1865] A system including:

[1866] (Claim 2)

[1867] 10. The system of claim 1, wherein the generated answer includes additional information.

[1868] (Claim 3)

[1869] 2. The system of claim 1, wherein the generative model uses natural language processing.

[1870] "Application example 2 when combining emotion engines"

[1871] (Claim 1)

[1872] A means for providing information on life in a shelter using a generative model;

[1873] A means to build a thematic Q&A database;

[1874] A means for loading the Q&A database into a generative model and customizing it to a specific theme;

[1875] means for providing an interface for accepting questions from a user;

[1876] means for generating an optimal answer to a user's question using the generative model;

[1877] means for transmitting the generated answer to a user's terminal;

[1878] means for analyzing a user's emotions using an emotion engine;

[1879] means for adjusting the generated response in response to the user's emotional state;

[1880] A system including:

[1881] (Claim 2)

[1882] 10. The system of claim 1, wherein the generated answer includes additional information.

[1883] (Claim 3)

[1884] 2. The system of claim 1, wherein the generative model uses natural language processing.

[1885] (Claim 4)

[1886] The system according to claim 1, characterized in that the application is food delivery.

[1887] (Claim 5)

[1888] 5. The system according to claim 4, wherein the system proposes an optimal menu according to the user's emotions. [Explanation of symbols]

[1889] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for providing information on life in a shelter using a generative model; A means to build a thematic Q&A database; A means for loading the Q&A database into a generative model and customizing it to a specific theme; means for providing an interface for accepting questions from a user; means for generating an optimal answer to a user's question using the generative model; means for transmitting the generated answer to a user's terminal; A system including:

2. 10. The system of claim 1, wherein the generated answer includes additional information.

3. The system of claim 1 , wherein the generative model uses natural language processing.

Images