System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting natural disasters in real time; means for training a generative AI model using the collected natural disasters; means for predicting a probability of occurrence of disasters using the generative AI model; means for reflecting the predicted probability of occurrence of disasters in a map service; and means for providing shelter information and information for preparing for disasters.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, natural disasters such as earthquakes, fires, and tsunamis have been increasing, making them difficult to predict. Furthermore, many local residents are unable to grasp the probability of a disaster occurring in their area in real time, and therefore often lack adequate disaster preparedness. This makes it difficult to take prompt and appropriate action when a disaster occurs, potentially resulting in greater damage. Therefore, the purpose of this invention is to build a system that predicts and visualizes the probability of a natural disaster occurring in real time, thereby providing local residents with appropriate preparations and evacuation information. [Means for solving the problem]

[0005] The present invention solves the problems by the following means. Specifically, the system includes a means for collecting natural disaster data in real time, a means for training a generative AI model using the collected natural disaster data, a means for predicting the probability of disaster occurrence using the generative AI model, a means for reflecting the predicted disaster occurrence probability in a map service, and a means for providing evacuation shelter information and disaster preparation information (Claim 1). Furthermore, by providing a means for storing past natural disaster data in a database and creating a data set necessary for the generative AI model to learn (Claim 2), more accurate predictions are possible. Furthermore, by providing a means for collecting evacuation shelter information from databases of local governments and government agencies and providing users with the latest evacuation shelter information in real time (Claim 3), the system helps users evacuate appropriately and quickly.

[0006] "Real-time" refers to a state in which data is collected and processed as it occurs and reflected without delay.

[0007] "Natural disaster data" refers to information related to natural disasters such as earthquakes, fires, and tsunamis, and includes attributes such as date, time, location, and scale.

[0008] A "generative AI model" is a model that uses artificial intelligence technology to analyze data and make predictions, learning patterns and trends from past data.

[0009] "Disaster occurrence probability" is the probability that a natural disaster may occur in a specific area or period of time.

[0010] A "mapping service" is an online platform that visually displays geographic information, typically accessed via a web browser or mobile app.

[0011] "Evacuation shelter information" refers to information about facilities and locations where local residents can safely evacuate in the event that evacuation is necessary.

[0012] "Disaster preparedness information" refers to information such as items, methods, and guidelines for action that should be prepared in advance in the event of a disaster.

[0013] A database is a system that stores and manages data in a structured format, allowing you to quickly search and retrieve the information you need.

[0014] "Learning" refers to the process by which a generative AI model finds patterns and rules from given data, and is an important step in improving the model's predictive accuracy.

[0015] "Prediction" is the process of estimating future events or conditions based on past data and current conditions.

[0016] "Users" refer to local residents and related parties who use this system, and are the entities that receive disaster information, evacuation shelter information, etc. through the map service.

[0017] "Local governments and government agencies" refers to local administrative bodies and government departments that are public institutions that respond to disasters and manage evacuation shelters. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] This section describes an embodiment of the present invention. This system is composed of a server, a terminal, and a user component, and by these components working together, it is possible to collect, analyze, and predict natural disaster data in real time and visualize it through a map service.

[0040] Server Roles

[0041] The server first collects various natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. The collected data is then stored in a database and organized in an appropriate format.

[0042] The server then uses a generative AI model to train this data. The model is trained using past disaster data so that it can predict the probability of disaster occurrence with high accuracy. This training process requires preprocessing the data, such as normalization and missing data completion.

[0043] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. These prediction results are then stored in the database again for use in the map service.

[0044] The server then sends the predicted disaster probability data to the map service's API, which plots disaster risk information for each region on a map. The marks on the map use different colors and symbols to visualize the level of disaster risk.

[0045] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0046] Device Role

[0047] The device is the interface through which users access the map service. When a user accesses the map service via a web browser or mobile app, the device displays the latest disaster risk information and evacuation shelter information.

[0048] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0049] Marks on the map displayed on the device can be clicked to view details, and route guidance to evacuation shelters is also provided to support specific evacuation actions.

[0050] User Roles

[0051] Users can use the map service to check disaster risks and evacuation shelter information for their area, and if necessary, prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0052] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0053] Specific examples

[0054] As a specific example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0055] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0056] Users access the map service and check the warning symbols for the Tokyo area. When users click on the symbol, information about the nearest evacuation shelter is displayed, retrieved from the server. At the same time, specific preparation information is also provided, such as a list of items needed in an emergency evacuation bag and how to check on the safety of family members.

[0057] This will allow users to prepare for appropriate evacuation actions and evacuate quickly and safely in the event of an earthquake. This system is a powerful tool for promoting quick and appropriate responses before a disaster occurs.

[0058] The processing flow will be explained below.

[0059] Step 1:

[0060] Server: Periodically sends requests to APIs for various natural disaster data (earthquakes, fires, tsunamis, etc.) to collect new data. The collected data is formatted and stored in a database.

[0061] Step 2:

[0062] Server: Extracts past natural disaster data from the database and preprocesses it as a training dataset for the generative AI model. Preprocessing includes normalizing the data and filling in missing data.

[0063] Step 3:

[0064] Server: The preprocessed dataset is used to train a generative AI model, for example, by training the model using a machine learning framework (such as TensorFlow or PyTorch).

[0065] Step 4:

[0066] Server: Using the generative AI model that has completed training, new disaster data is input in real time to predict the probability of disaster occurrence. The prediction results are stored in a database.

[0067] Step 5:

[0068] Server: Connects to the Yahoo Maps API and reflects predicted disaster probability data on the map. Specifically, areas with high disaster risk are marked with warning marks or color-coded.

[0069] Step 6:

[0070] Device: When a user accesses Yahoo! Maps using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on a map.

[0071] Step 7:

[0072] Device: When a user clicks on a warning mark on the map, evacuation shelter information and disaster preparedness advice retrieved from the server are displayed, such as the location of the nearest evacuation shelter and a list of emergency supplies to take with you.

[0073] Step 8:

[0074] Server: Collects the latest evacuation shelter information from databases of local governments and government agencies and stores it in a database. This information is also reflected in the map service and made available to users.

[0075] Step 9:

[0076] User: Make necessary preparations based on the displayed disaster risk information and evacuation shelter information. For example, prepare an emergency kit and check evacuation routes with family members.

[0077] Step 10:

[0078] Server: Collects various feedback and log data to help improve our services. Analyzes user access history and click data to identify new features and service improvements.

[0079] Example 1

[0080] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0081] In modern society, the frequency and scale of damage caused by natural disasters continue to increase. Conventional disaster information systems have difficulty collecting real-time data and making highly accurate predictions, which means users are unable to obtain the information they need to take prompt and appropriate action. In addition, updates to evacuation shelter information and disaster preparation information are often delayed, preventing users from acting based on the latest information. Therefore, there is a need to build a system that can quickly and accurately predict disaster risks and provide users with useful information.

[0082] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0083] In this invention, the server includes means for collecting natural disaster data in real time, means for storing the collected natural disaster data in a database and organizing it while maintaining consistency, means for performing data preprocessing using the collected natural disaster data, means for training a generative AI model using the preprocessed data, means for predicting disaster occurrence probabilities in real time using the generative AI model, means for re-storing the predicted disaster occurrence probabilities in the database, means for reflecting the predicted disaster occurrence probabilities in a map service, means for collecting evacuation shelter information from local governments and government agencies and storing it in the database, means for providing evacuation shelter information and disaster preparation information to users, and terminal interface means for users to check the evacuation shelter information and disaster occurrence probability information. This allows users to quickly and accurately obtain predicted disaster risk information and take appropriate evacuation actions and preparations.

[0084] A "server" is a computer system that has the ability to provide various services and data to other computers over a network.

[0085] "Natural disaster data" refers to data that includes information on natural phenomena such as earthquakes, tsunamis, and fires.

[0086] A "database" is a system for efficiently storing, retrieving, and updating collected and organized data.

[0087] "Data preprocessing" is the process of converting data into a format suitable for analysis and learning models, and completing and normalizing missing values.

[0088] A "generative AI model" is an artificial intelligence model that learns from collected data and makes predictions about new data.

[0089] "Disaster occurrence probability" is a predicted value that indicates the probability that a specific natural disaster will occur within a certain period of time.

[0090] A "map service" is a web service or application that visually displays geographic information and various data.

[0091] "Evacuation shelter information" is information about places where people can evacuate in the event of a disaster.

[0092] A "terminal interface" is an operation screen for a device or application that allows a user to access a system and obtain information.

[0093] "Local governments and government agencies" refers to local government and national public institutions, which are organizations that provide various public services and information.

[0094] The following describes in detail an embodiment of the present invention. This system is composed of a server, a terminal, and a user component, and by these components working together, it is possible to collect, analyze, and predict natural disaster data in real time, and visualize it through a map service.

[0095] Server Roles

[0096] The server first collects natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. For example, it uses the Japan Meteorological Agency's API to obtain earthquake information. Requests are sent at regular intervals, and JSON-formatted data is returned as a response.

[0097] The server then stores the collected data in a database, which can be structured for consistency and organized by time and location, for example, using MongoDB.

[0098] The server then performs data preprocessing, including normalizing the data and imputing missing values. Normalization aligns the scale of the data, allowing for efficient training of generative AI models. For example, Pandas is used to normalize the "magnitude" scale from 0 to 1.

[0099] The server then uses the data to train a generative AI model. To train the model using data from past disasters, the server uses a machine learning framework such as TensorFlow to build a long short-term memory (LSTM) network and train the model on earthquake data.

[0100] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. The prediction results are then stored in the database again, ready for use with the map service.

[0101] The server then sends the prediction results to the API of the map service, which plots disaster risk information for each region on a map, with high-risk areas displayed in red and low-risk areas in green.

[0102] In addition, the server collects evacuation shelter information from databases of local governments and government agencies, stores the latest evacuation shelter information in the database, and sends this information to the map service so that users can check it.

[0103] Device Role

[0104] The device acts as an interface for using the map service. When a user accesses the map service via a web browser or mobile app, the device displays the latest disaster risk information and evacuation shelter information.

[0105] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0106] The marks displayed on the device's map can be clicked to view detailed information. Route guidance to evacuation shelters is also provided, supporting specific evacuation actions.

[0107] User Roles

[0108] Users can use the map service to check disaster risk information and evacuation shelter information for their area. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0109] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake occurring, they can check to see if they have prepared an emergency kit and how to contact their family members. They can also check the route to the nearest evacuation shelter in advance, allowing them to act quickly and calmly in the event of an emergency.

[0110] Prompt Sentence Examples

[0111] Example prompts to input to a generative AI model:

[0112] "Using the past earthquake data below, please predict the probability of an earthquake occurring within the next 24 hours. This data is limited to the Tokyo area."

[0113] This system is a powerful tool for promoting rapid and appropriate responses before a disaster occurs, and by each component working together, users can obtain the information they need in a timely manner and act safely.

[0114] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0115] Step 1:

[0116] The server collects natural disaster data. Specifically, it periodically sends requests to an API that provides data on earthquakes, fires, tsunamis, etc. In this process, the input is the API endpoint URL, and the output is natural disaster data in JSON format. For example, it sends a request to "https: / / api.kishou.go.jp / earthquake" and receives the returned data.

[0117] Step 2:

[0118] The server stores the collected natural disaster data in a database. The input is the collected data in JSON format, and the output is the organized data stored in the database. Specifically, it uses MongoDB to insert earthquake data into the "earthquake_data" collection, which contains fields such as "timestamp," "location," and "magnitude."

[0119] Step 3:

[0120] The server performs data preprocessing. The input is natural disaster data retrieved from the database, and the output is the preprocessed data. Specifically, it uses Pandas to normalize the data and convert the values ​​of the "magnitude" field into the range of 0 to 1. It also imputes missing values ​​with appropriate values ​​if any.

[0121] Step 4:

[0122] The server trains a generative AI model using the preprocessed data. The input is the preprocessed data, and the output is a trained generative AI model. Using TensorFlow, an LSTM network is constructed and the model is trained using past earthquake data. Specifically, the LSTM model is trained using earthquake data from the past five years.

[0123] Step 5:

[0124] The server uses a trained generative AI model to predict the probability of a disaster. The input is newly collected natural disaster data, and the output is the predicted probability of a disaster. New earthquake data is input into the model, and the probability of an earthquake occurring within the next 24 hours is calculated.

[0125] Step 6:

[0126] The server stores the prediction results in the database again. The input is the predicted probability of a disaster occurring, and the output is the prediction result stored in the database. Specifically, the server inserts the predicted earthquake occurrence probability into the "predictions" collection.

[0127] Step 7:

[0128] The server sends the prediction results to the map service's API. The input is the prediction results obtained from the database, and the output is the disaster risk information reflected in the map service. Using the Google Maps API, disaster risk data for each region is plotted on a map. Areas with high risk are displayed in red, and areas with low risk are displayed in green.

[0129] Step 8:

[0130] The terminal is an interface for using map services. The input is the user's search query, and the output is disaster risk information and evacuation shelter information displayed on the map. Specifically, when a user searches for a specific area, disaster information and the probability of disaster occurrence related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0131] Step 9:

[0132] Users use map services to check disaster risk information and evacuation shelter information for the area in which they live. The input is various operations performed by the user (for example, searching for an area or clicking a mark), and the output is related disaster information and evacuation shelter information. Specifically, when a user clicks on a warning mark on the map, information on the nearest evacuation shelter and preparation information such as a "list of items necessary for an emergency evacuation bag" and "how to check on the safety of family members" is displayed.

[0133] (Application example 1)

[0134] 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."

[0135] In recent years, the increasing number of natural disasters has created a need for real-time collection, analysis, and prediction of disaster information. However, existing systems do not provide sufficient, fast, and accurate information when a disaster occurs, which often delays evacuation and preparation. Furthermore, the provision of specific evacuation instructions and disaster prevention guidelines to users is insufficient, making it difficult to respond to disasters effectively.

[0136] 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.

[0137] In this invention, the server includes means for collecting natural disaster data in real time, means for training a generative AI model using the collected natural disaster data, means for predicting the probability of a disaster occurring using the generative AI model, means for reflecting the predicted probability of a disaster occurring in a map service, means for providing evacuation shelter information and disaster preparation information, means for providing disaster notifications in real time on the terminal, means for displaying route guidance to evacuation shelters on the terminal, and means for presenting a disaster action guide and an emergency kit list on the terminal. This allows users to obtain quick and accurate disaster information and provides specific evacuation instructions and disaster prevention guidelines, enabling appropriate disaster response.

[0138] "Means for collecting natural disaster data in real time" refers to technological means for instantly obtaining information on natural disasters such as earthquakes, fires, and tsunamis.

[0139] "Means for training a generative AI model using collected natural disaster data" refers to technical means for training a generative AI model based on acquired natural disaster data.

[0140] "Means for predicting the probability of disaster occurrence using a generative AI model" refers to a technical means for utilizing a trained generative AI model to quantify and predict the possibility of a disaster occurring.

[0141] "Means for reflecting the predicted probability of disaster occurrence on the map service" refers to the technical means for displaying the probability of disaster occurrence predicted by the generative AI model on the map service.

[0142] "Means for providing information on evacuation shelters and disaster preparation information" refers to technical means for providing users with information on the locations of evacuation shelters and the equipment needed in the event of a disaster.

[0143] "Means for providing real-time disaster notifications on devices" refers to technical means for sending users immediate notifications about disaster occurrences through devices such as smartphones or smart glasses.

[0144] "Means for displaying route guidance to a shelter on a device" refers to a technical means for displaying specific directions for a user to safely reach a shelter on a device such as a smartphone or smart glasses.

[0145] "Means for presenting disaster action guidelines and emergency bag lists on devices" refers to technological means for displaying specific actions to be taken and lists of necessary items to bring in the event of a disaster on devices such as smartphones or smart glasses.

[0146] An embodiment of the present invention will now be described. This system is comprised of a server, a terminal, and a user component, which work together to enable a rapid and appropriate response to natural disasters.

[0147] Server Roles

[0148] The server first collects natural disaster data in real time. This data is obtained through APIs related to disasters such as earthquakes, fires, and tsunamis. The collected data is then stored in a database and organized in an appropriate format. For example, this process involves using the Python requests library to obtain data from external APIs and storing the data in MySQL or PostgreSQL.

[0149] The server then uses this data to train a generative AI model. The model is trained using past disaster data to accurately predict the probability of a disaster. This training process involves preprocessing, such as normalizing the data and filling in missing data.

[0150] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. The prediction results are then stored in a database and sent to the map service's API, which visualizes disaster risk on a map.

[0151] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service API so that users can check it.

[0152] Device Role

[0153] The device is the interface through which users use the map service. When users access the map service using a web browser or mobile app, the latest disaster risk information and evacuation shelter information are displayed. For example, when a user searches for their local area, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information such as the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies is also provided.

[0154] The map displayed on the device can be clicked to view details. Route guidance to evacuation shelters is also provided, supporting specific evacuation actions. Real-time notifications are sent via smartphones, smart glasses, and other devices in the event of a disaster such as an earthquake or fire.

[0155] User Roles

[0156] Users can use the map service to check disaster risks and evacuation shelter information in their area. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors. For example, if a user checks their area on the map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. In addition, by checking the route to the nearest evacuation shelter in advance, they can act quickly and safely in the event of an emergency.

[0157] Specific examples

[0158] As a specific example, when a user in Tokyo receives a notification informing them of an increased earthquake risk, the message "Earthquake risk is increasing in the Tokyo area. Please check for the latest information" is displayed on the user's smartphone. Route guidance to an evacuation shelter, a list of emergency kits, and a prompt such as "List of items necessary for an emergency kit" are also provided.

[0159] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0160] Step 1:

[0161] The server collects natural disaster data in real time from an external API. The server periodically sends requests to the API endpoint to get the latest disaster information such as earthquakes, fires, tsunamis, etc. In this step, the input is the data obtained from the API, and the output is the raw natural disaster data.

[0162] Step 2:

[0163] The server stores the collected natural disaster data in a database. The server performs preprocessing to standardize the data format and fill in missing data. This inputs normalized natural disaster data into the database. The input is raw natural disaster data, and the output is organized database entries.

[0164] Step 3:

[0165] The server trains the generative AI model using the compiled natural disaster data. The server supplies past disaster data to the generative AI model and trains the disaster occurrence prediction algorithm. The input is past disaster data, and the output is the trained generative AI model.

[0166] Step 4:

[0167] The server inputs new disaster data into the AI ​​model in real time to predict the probability of a disaster occurring. The model evaluates disaster risk based on the latest data and generates the results. The input is the latest disaster data, and the output is a predicted disaster probability.

[0168] Step 5:

[0169] The server reflects the predicted disaster occurrence probability on the map service. The server sends this information to the map service API, which provides data for visualizing disaster risk on a map. The input is the prediction result, and the output is disaster risk information plotted on the map service.

[0170] Step 6:

[0171] The server collects evacuation shelter information from databases of local governments and government agencies and stores it in a database. The server retrieves the latest evacuation shelter information and prepares it for users to access. The input is the latest information on evacuation shelters, and the output is a compiled evacuation shelter database entry.

[0172] Step 7:

[0173] When a user accesses the map service, the device displays the latest disaster risk information and evacuation shelter information. When a user searches for their area, disaster data and evacuation shelter information for that area are displayed on the screen. The input is the user's search query, and the output is disaster information and evacuation shelter information displayed on the device's display.

[0174] Step 8:

[0175] The device provides users with real-time disaster notifications. The server pushes disaster information to the device so that users can check it immediately. The input is newly collected disaster data, and the output is notifications on the device.

[0176] Step 9:

[0177] The device displays route guidance to the nearest evacuation shelter. When the user clicks on the evacuation shelter marking, the device displays the route using GPS and map service APIs. The input is the user's current location and the location information of the evacuation shelter, and the output is the route guidance displayed on the device.

[0178] Step 10:

[0179] The device presents a list of action guidelines and emergency kits for emergencies. When the user checks the disaster information, a specific action plan and a list of necessary items are displayed. The input is predicted disaster risk information, and the output is the display of the action guidelines and list.

[0180] 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.

[0181] An embodiment of the present invention will now be described. The system of the present invention is composed of a server, a terminal, and a user component, and by combining it with an emotion engine that recognizes the user's emotions, it is possible to collect, analyze, and predict natural disaster data in real time and visualize it through a map service. This system not only provides disaster information and evacuation shelter information, but also provides customized information and messages according to the user's emotional state.

[0182] Server Roles

[0183] The server first collects various natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. The collected data is then stored in a database and organized in an appropriate format.

[0184] The server then uses a generative AI model to train this data. The model is trained using past disaster data so that it can predict the probability of disaster occurrence with high accuracy. This training process requires preprocessing the data, such as normalization and missing data completion.

[0185] New data is input in real time using the generative AI model to predict the probability of a disaster occurring. The prediction results are then stored in a database for use in the map service.

[0186] The server then sends the predicted disaster probability data to the map service's API, which plots disaster risk information for each region on a map. Marks on the map use different colors and symbols to visualize the level of disaster risk.

[0187] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0188] The role of the emotional engine

[0189] The emotion engine recognizes the user's emotional state and provides customized messages and advice based on that information. When the user browses disaster information, the emotion engine collects and analyzes emotional data via the camera and microphone. Using the results of this analysis, if the user is feeling anxious or tense, it displays a reassuring message.

[0190] Device Role

[0191] The device is the interface through which users use the map service. When a user accesses the map service using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. Furthermore, information based on the user's emotional state is also displayed based on the analysis results of the emotion engine.

[0192] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0193] Marks on the map displayed on the device can be clicked to view details, and route guidance to evacuation shelters is also provided to support specific evacuation actions.

[0194] User Roles

[0195] Users can use the map service to check disaster risks and evacuation shelter information for their area. They can also receive information customized based on their emotional state using the emotion engine. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0196] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0197] Specific examples

[0198] As a specific example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0199] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0200] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice through the camera and microphone to recognize their current emotional state. If the user feels anxious or nervous, it displays a reassuring message such as, "Please remain calm. There is a shelter nearby. We will show you the route to the nearest shelter."

[0201] Users access the map service and check the warning mark for the Tokyo area. When the user clicks on the warning mark, information about the nearest evacuation shelter retrieved from the server is displayed. At the same time, specific preparation information such as a list of items needed in an emergency evacuation bag and how to check on the safety of family members is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0202] In this way, by combining the emotion engine, support tailored to the user's emotional state can be provided, enabling more effective disaster response. This system is a powerful tool for encouraging prompt and appropriate responses before a disaster occurs.

[0203] The processing flow will be explained below.

[0204] Step 1:

[0205] Server: Periodically sends requests to APIs for various natural disaster data (earthquakes, fires, tsunamis, etc.) to collect new data. The collected data is formatted and stored in a database.

[0206] Step 2:

[0207] Server: Extracts past natural disaster data from the database and preprocesses it as a training dataset for the generative AI model. Preprocessing includes normalizing the data and filling in missing data.

[0208] Step 3:

[0209] Server: The preprocessed dataset is used to train a generative AI model, for example, by training the model using a machine learning framework (such as TensorFlow or PyTorch).

[0210] Step 4:

[0211] Server: Using the generative AI model that has completed training, new disaster data is input in real time to predict the probability of disaster occurrence. The prediction results are stored in a database.

[0212] Step 5:

[0213] Server: Connects to the Yahoo Maps API and reflects predicted disaster probability data on the map. Specifically, areas with high disaster risk are marked with warning marks or color-coded.

[0214] Step 6:

[0215] Device: When a user accesses Yahoo! Maps using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on a map.

[0216] Step 7:

[0217] Device: When a user clicks on a warning mark on the map, evacuation shelter information and disaster preparedness advice retrieved from the server are displayed, such as the location of the nearest evacuation shelter and a list of emergency supplies to take with you.

[0218] Step 8:

[0219] Server: Collects the latest evacuation shelter information from databases of local governments and government agencies and stores it in a database. This information is also reflected in the map service and made available to users.

[0220] Step 9:

[0221] Emotion engine: When a user uses the map service, the system collects emotional data from the user via the camera and microphone. For example, it uses facial recognition and voice analysis to assess whether the user is feeling anxious or nervous.

[0222] Step 10:

[0223] Emotion engine: Analyzes collected emotional data and generates customized messages and advice based on the user's emotional state. For example, if the user is feeling anxious, the engine generates a message such as, "Please remain calm. Information about the nearest evacuation shelter will be displayed."

[0224] Step 11:

[0225] Terminal: Displays customized messages generated by the emotion engine to the user, providing messages that reassure the user and specific guidelines for action.

[0226] Step 12:

[0227] User: Make necessary preparations based on the displayed disaster risk information and evacuation shelter information. For example, prepare an emergency kit and check evacuation routes with family members.

[0228] Step 13:

[0229] Server: Collects various feedback and log data to help improve our services. Analyzes user access history and click data to identify new features and service improvements.

[0230] Example 2

[0231] 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."

[0232] Natural disasters have a significant impact on human lives and property, so providing fast and accurate information is essential. However, existing disaster information systems lack the ability to collect and analyze information in real time, and they do not provide customized information based on the user's emotional state. This makes it difficult for users to take appropriate action, and anxiety and confusion are likely to arise.

[0233] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0234] In this invention, the server includes means for collecting natural disaster data in real time, means for storing the collected natural disaster data in a database, means for preprocessing the data and training a generative AI model, means for predicting the probability of a disaster occurring using the generative AI model, means for reflecting the predicted probability of a disaster occurring in a map service, means for providing evacuation shelter information and disaster preparation information, and means for recognizing a user's emotional state and displaying a customized message according to that state. This allows users to receive accurate natural disaster information in real time, and also to receive appropriate guidelines for action and reassurance messages according to their emotional state.

[0235] "Means for collecting natural disaster data in real time" refers to a system or device for obtaining information on natural disasters such as earthquakes, fires, and tsunamis in a timely manner from external data sources.

[0236] "Means for storing collected natural disaster data in a database" refers to a database system and related technologies for efficiently managing and storing acquired natural disaster data and making it accessible later.

[0237] "Data preprocessing and means for training generative AI models" refers to techniques for properly preparing data and converting it into the format required for an AI model to learn, and then using AI techniques to train the model.

[0238] "Means for predicting the probability of disaster occurrence using a generative AI model" refers to technologies and systems that use a generative AI model to calculate the likelihood of future disaster occurrence based on collected natural disaster data.

[0239] "Means for reflecting the predicted probability of disaster occurrence in map services" refers to technology and systems that display the probability information of disaster occurrence predicted by an AI model on a visual map, making it easier for users to understand visually.

[0240] "Means for providing evacuation shelter information and disaster preparation information" refers to technologies and systems that provide users with specific information about where to evacuate and necessary preparations when a disaster occurs, before, during, or after the disaster.

[0241] "Means for recognizing the user's emotional state and displaying a customized message according to that state" refers to technology that analyzes the user's facial and voice data to determine their emotional state, and then displays individual reassuring messages or advice based on the results.

[0242] MODE FOR CARRYING OUT THE INVENTION

[0243] The system of the present invention is composed of a server, a terminal, and a user component, and by combining it with an emotion engine, it collects, analyzes, and predicts natural disaster data in real time, and visualizes it through a map service. A specific embodiment of this system is described in detail below.

[0244] Server Roles

[0245] The server first collects real-time data on natural disasters such as earthquakes, fires, and tsunamis. Specifically, the server periodically sends requests to external APIs (e.g., earthquake information APIs, weather information APIs) to obtain the latest disaster data. To do this, it uses a program library such as Python's requests library. The obtained data is received in JSON format or similar and stored in a database (e.g., MySQL, MongoDB).

[0246] The server then trains the generative AI model using this data. Data preprocessing involves normalizing the data and imputing missing data. This is done using data analysis libraries such as scikit-learn and pandas. The preprocessed data is then used to train the generative AI model using an AI framework such as TensorFlow or PyTorch.

[0247] Using a trained generative AI model, new data is input in real time to predict the probability of a disaster occurring. The prediction results are then stored in a database and sent to the map service's API (e.g., Google Maps API). Disaster risk information for each region is then plotted on a map. Marks on the map use different colors and symbols to visualize the level of disaster risk.

[0248] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0249] The role of the emotional engine

[0250] The emotion engine recognizes the user's emotional state and provides customized messages and advice based on that information. When the user browses disaster information, emotional data is collected via the camera and microphone and analyzed. This analysis uses facial expression recognition technology (e.g., Amazon Rekognition) and voice analysis technology (e.g., Google Cloud Speech-to-Text). Using the analysis results, if the user is feeling anxious or tense, a reassuring message is displayed.

[0251] Device Role

[0252] The device functions as an interface for users to use the map service. When a user accesses the map service using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. The device displays the map and information using web technologies such as HTML, CSS, and JavaScript. It also displays information according to the user's emotional state based on the analysis results of an emotion engine.

[0253] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided. Marks on the map can be clicked to view details. Route guidance to the evacuation shelter is also provided to support specific evacuation actions.

[0254] User Roles

[0255] Users can use the map service to check disaster risks and evacuation shelter information for their area. They can also receive information customized based on their emotional state using the emotion engine. This allows them to prepare for evacuation as needed and take appropriate measures in cooperation with their family and neighbors.

[0256] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0257] Specific examples

[0258] For example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0259] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0260] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice via the camera and microphone to recognize their current emotional state. If the user feels anxious or nervous, it displays reassuring messages such as "Please stay calm. There is a shelter nearby" or "We will show you the route to the nearest shelter."

[0261] Users access the map service and check the warning symbols for the Tokyo area. When the user clicks on the symbol, information about the nearest evacuation shelter is displayed, retrieved from the server. At the same time, specific information on preparations, such as a list of items needed in an emergency bag and how to check on the safety of family members, is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0262] Prompt Sentence Examples

[0263] Here are some example prompts to input to a generative AI model:

[0264] "Predict the probability of an earthquake occurring in the next 12 hours."

[0265] "Get the latest shelter information and display it to the user."

[0266] "Analyze the user's emotional state and display appropriate messages."

[0267] As described above, the present invention not only collects, predicts, and displays disaster information, but also supports appropriate behavior in emergencies by providing customized messages that take into account the user's emotional state.

[0268] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0269] Step 1:

[0270] The server sends requests to an external natural disaster data API to retrieve natural disaster data in real time. Specifically, it uses the Python requests library to send a GET request to the API. It uses the API's endpoint URL as input and obtains the retrieved data in JSON format as output. This data includes disaster information such as earthquakes, fires, and tsunamis.

[0271] Step 2:

[0272] The server stores the acquired natural disaster data in a database. Specifically, it uses MySQL or MongoDB to store the data. The acquired JSON data is used as input, and disaster data is saved in the database as output. The database includes information such as the type of disaster, time of occurrence, location, and magnitude.

[0273] Step 3:

[0274] The server preprocesses the data stored in the database. Specifically, it uses scikit-learn and pandas to normalize the data and impute missing data. It uses raw data from the database as input and obtains preprocessed data as output. This puts the data in a format suitable for training AI models.

[0275] Step 4:

[0276] The server uses the preprocessed data to train a generative AI model, specifically using TensorFlow or PyTorch. It uses the preprocessed data and labeled data as input and obtains a trained AI model as output. This AI model is trained to be used to predict the probability of disaster occurrence.

[0277] Step 5:

[0278] The server inputs new natural disaster data into the generative AI model to predict the probability of disaster occurrence. It uses real-time data as input and obtains the predicted probability of disaster occurrence as output. To do this, the AI ​​model performs predictive calculations to calculate the probability of the next possible disaster.

[0279] Step 6:

[0280] The server sends the predicted disaster probability to a map service API, which plots disaster risk information for each region on a map. It uses disaster probability data as input and obtains a visual plot on the map service as output. It places marks on the map using APIs such as Google Maps to display the risk level.

[0281] Step 7:

[0282] The server retrieves evacuation shelter information from local government and government agency databases and stores it in a database. It uses the endpoint URL of the external database as input and stores the retrieved evacuation shelter information as output in an internal database. This includes information such as the latest evacuation shelter locations and capacity.

[0283] Step 8:

[0284] The emotion engine collects emotion data from the camera and microphone when the user browses disaster information. It uses the user's facial expression and voice data as input and obtains emotion analysis results as output. This is done using facial expression recognition software (e.g., Amazon Rekognition) and voice analysis software (e.g., Google Cloud Speech-to-Text).

[0285] Step 9:

[0286] Based on the analysis results, the emotion engine generates and displays a reassuring message if the user feels anxious or tense. It uses the emotion analysis results as input and gets a customized reassuring message as output, which helps the user relax and take appropriate action.

[0287] Step 10:

[0288] The device displays disaster risk information and evacuation shelter information to the user. It uses disaster risk data and evacuation shelter information obtained from the server as input and displays them on a map service or information screen as output. Information is provided to the user in real time via a web browser or mobile app.

[0289] (Application example 2)

[0290] 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."

[0291] In recent years, the frequency of natural disasters has increased, making countermeasures an urgent need. Meanwhile, with the widespread adoption of autonomous vehicles, it is necessary to ensure the safe operation of vehicles and the peace of mind of users during disasters. However, conventional systems have struggled to efficiently collect and analyze disaster information in real time and provide specific disaster prevention measures. Furthermore, they have been unable to provide customized messages based on the user's emotional state, making it difficult to encourage appropriate responses when users feel anxious.

[0292] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0293] In this invention, the server includes means for collecting natural disaster data in real time, means for training a generative AI model using the collected natural disaster data, means for predicting the probability of disaster occurrence using the generative AI model, means for reflecting the predicted probability of disaster occurrence in a map service, means for providing evacuation shelter information and disaster preparation information, means for recognizing a user's emotional state and providing a customized message based thereon, and means for optimizing a safe driving route in the event of a natural disaster. This allows a user to obtain the latest disaster information in real time, ensure a safe driving route, and receive a customized message according to their emotional state, enabling them to use autonomous vehicles with peace of mind.

[0294] "Real-time" means instantly processing and reflecting events and information occurring at that time.

[0295] "Natural disaster data" refers to data that includes information on natural disasters such as earthquakes, tsunamis, and fires.

[0296] A "generative AI model" is a type of artificial intelligence that uses machine learning algorithms to learn from past data and make predictions and classifications based on new data.

[0297] "Probability of disaster occurrence" is an index that indicates the degree of possibility of a disaster occurring in a specific region or under specific conditions.

[0298] A "map service" is a service that provides visual geographic information and displays location and route information.

[0299] "Evacuation shelter information" is information about safe places to evacuate to in the event of a disaster.

[0300] "Disaster preparedness information" refers to information on specific methods of response and items to prepare in the event of a disaster.

[0301] "Emotional state" refers to the emotions and mental state that a user is feeling at that time.

[0302] A "customized message" is information or a message that is optimized to a user's individual needs and emotional state.

[0303] A "safe route during a natural disaster" is a route that allows safe travel while avoiding danger even in the event of a natural disaster.

[0304] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

[0305] System Overview

[0306] This system, which consists of a server, terminals, and users, collects natural disaster information in real time and provides users with appropriate countermeasures.The system also uses a generative AI model to predict the probability of disaster occurrence and provides customized messages according to the user's emotional state, supporting the safe operation of autonomous vehicles.

[0307] Server Roles

[0308] The server collects natural disaster data in real time. Specifically, it periodically sends requests to APIs provided by data providers for earthquakes, fires, tsunamis, etc. to obtain new data. The obtained data is stored in a database and organized appropriately.

[0309] The server then uses a generative AI model to train this data and predict the probability of a disaster occurring. During this training process, data preprocessing is performed, including normalization and missing data completion. The prediction results are then stored in the database again and sent to the map service's API. This allows the disaster risk for each region to be visualized on a map. Marks on the map use different colors and symbols to indicate the level of disaster risk.

[0310] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies and updates it in real time, which is also sent to the map service for users to view.

[0311] The role of the emotional engine

[0312] The emotion engine recognizes the user's emotional state and provides customized messages based on that. When a user browses disaster information, emotional data is collected and analyzed via the camera and microphone. Based on the analysis results, if the user is feeling anxious or tense, a reassuring message is displayed.

[0313] Device Role

[0314] The device is the interface through which users use the map service. When a user accesses the map service through a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. Furthermore, information corresponding to the user's emotional state is also displayed based on the analysis results of the emotion engine.

[0315] Disaster information and occurrence probability related to the area searched by the user are displayed on a map. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided. Marks on the map can be clicked to check details, and route guidance to the evacuation shelter is also provided.

[0316] User Roles

[0317] Users can use this system to check disaster risks and evacuation shelter information for their area. They can also receive customized information based on their emotional state through the emotion engine, which allows them to make appropriate evacuation preparations and take appropriate evacuation actions.

[0318] Specific examples

[0319] For example, if a prediction is made that there is a high probability of a major earthquake occurring in Tokyo, the server will train the generative AI model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and send this prediction result to the map service API, which will display a red warning mark in the Tokyo area.

[0320] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice via the camera and microphone to recognize their current emotional state. If the user is feeling anxious or nervous, it displays a reassuring message such as, "Please remain calm. There is a shelter nearby. We will show you the route to the nearest shelter."

[0321] Users access the map service and check the warning symbols for the Tokyo area. Clicking on the symbol displays information about the nearest evacuation shelter. At the same time, specific information on preparations, such as a list of items needed in an emergency bag and how to check on the safety of family members, is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0322] Example prompt sentence:

[0323] If an earthquake is predicted to occur in the Tokyo area, the system will display routes to the nearest evacuation shelter, check the user's emotional state, and display reassuring messages if they feel anxious. Meanwhile, the system will safely optimize the routes of autonomous vehicles.

[0324] As described above, this system obtains disaster information in real time and provides customized messages according to the user's emotional state, thereby creating an environment in which users can use self-driving vehicles with peace of mind.

[0325] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0326] Step 1:

[0327] The server collects natural disaster data in real time from an external API. It sends API requests and stores the acquired data in a database. The input is the natural disaster data acquired from the API, and the output is the organized data stored in the database. The API requests are programmed to be made periodically.

[0328] Step 2:

[0329] The server trains the generative AI model using natural disaster data stored in a database. Data preprocessing involves normalization and missing data completion. The input is past disaster data obtained from the database, and the output is the trained generative AI model. This makes it possible to predict the probability of disaster occurrence with high accuracy.

[0330] Step 3:

[0331] The server uses the trained generative AI model to input new data in real time and predict the probability of a disaster. The input is the latest natural disaster data, and the output is data indicating the probability of a disaster occurring. The results are then stored in a database and sent to the map service's API.

[0332] Step 4:

[0333] The server visualizes the disaster risk for each region on a map based on the data sent to the map service's API. Marks on the map use different colors and symbols to indicate the level of disaster risk. The input is disaster probability data, and the output is visually plotted map data.

[0334] Step 5:

[0335] The server retrieves the latest evacuation shelter information from local government and government agency databases and stores it in the database. The input is the evacuation shelter information retrieved from the external database, and the output is the updated evacuation shelter data. This information is also sent to the map service so that users can view it.

[0336] Step 6:

[0337] The device accesses the map service through a web browser or mobile app to obtain disaster risk information for the user's current location or a specified area. The input is the user's location information and data from the map service API, and the output is a disaster risk map displayed on the device.

[0338] Step 7:

[0339] The emotion engine uses a camera and microphone to recognize the user's emotional state. It analyzes facial expressions and voice to generate a customized message if the user is feeling anxious or nervous. The input is the user's image and voice data, and the output is the emotion analysis result and a reassuring message based on that.

[0340] Step 8:

[0341] The device displays a customized message based on the user's emotional state based on the analysis results of the emotion engine. The input is the emotion analysis result, and the output is the message displayed on the device. This allows users to use the system with peace of mind.

[0342] Step 9:

[0343] Users can check evacuation shelter information and disaster preparedness information through their devices. The input is evacuation shelter information obtained from a map service, and the output is evacuation information and preparedness information displayed on the device. Based on this, users can take specific evacuation actions.

[0344] In this way, the entire system works seamlessly, allowing users to obtain the latest disaster and evacuation information in real time. Furthermore, by providing customized messages according to the user's emotional state, users can feel at ease and take appropriate action.

[0345] 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.

[0346] 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.

[0347] 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.

[0348] [Second embodiment]

[0349] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0350] 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.

[0351] 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).

[0352] 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.

[0353] 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.

[0354] 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).

[0355] 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.

[0356] 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.

[0357] 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.

[0358] 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.

[0359] 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.

[0360] 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."

[0361] This section describes an embodiment of the present invention. This system is composed of a server, a terminal, and a user component, and by these components working together, it is possible to collect, analyze, and predict natural disaster data in real time and visualize it through a map service.

[0362] Server Roles

[0363] The server first collects various natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. The collected data is then stored in a database and organized in an appropriate format.

[0364] The server then uses a generative AI model to train this data. The model is trained using past disaster data so that it can predict the probability of disaster occurrence with high accuracy. This training process requires preprocessing the data, such as normalization and missing data completion.

[0365] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. These prediction results are then stored in the database again for use in the map service.

[0366] The server then sends the predicted disaster probability data to the map service's API, which plots disaster risk information for each region on a map. The marks on the map use different colors and symbols to visualize the level of disaster risk.

[0367] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0368] Device Role

[0369] The device is the interface through which users access the map service. When a user accesses the map service via a web browser or mobile app, the device displays the latest disaster risk information and evacuation shelter information.

[0370] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0371] Marks on the map displayed on the device can be clicked to view details, and route guidance to evacuation shelters is also provided to support specific evacuation actions.

[0372] User Roles

[0373] Users can use the map service to check disaster risks and evacuation shelter information for their area, and if necessary, prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0374] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0375] Specific examples

[0376] As a specific example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0377] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0378] Users access the map service and check the warning symbols for the Tokyo area. When users click on the symbol, information about the nearest evacuation shelter is displayed, retrieved from the server. At the same time, specific preparation information is also provided, such as a list of items needed in an emergency evacuation bag and how to check on the safety of family members.

[0379] This will allow users to prepare for appropriate evacuation actions and evacuate quickly and safely in the event of an earthquake. This system is a powerful tool for promoting quick and appropriate responses before a disaster occurs.

[0380] The processing flow will be explained below.

[0381] Step 1:

[0382] Server: Periodically sends requests to APIs for various natural disaster data (earthquakes, fires, tsunamis, etc.) to collect new data. The collected data is formatted and stored in a database.

[0383] Step 2:

[0384] Server: Extracts past natural disaster data from the database and preprocesses it as a training dataset for the generative AI model. Preprocessing includes normalizing the data and filling in missing data.

[0385] Step 3:

[0386] Server: The preprocessed dataset is used to train a generative AI model, for example, by training the model using a machine learning framework (such as TensorFlow or PyTorch).

[0387] Step 4:

[0388] Server: Using the generative AI model that has completed training, new disaster data is input in real time to predict the probability of disaster occurrence. The prediction results are stored in a database.

[0389] Step 5:

[0390] Server: Connects to the Yahoo Maps API and reflects predicted disaster probability data on the map. Specifically, areas with high disaster risk are marked with warning marks or color-coded.

[0391] Step 6:

[0392] Device: When a user accesses Yahoo! Maps using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on a map.

[0393] Step 7:

[0394] Device: When a user clicks on a warning mark on the map, evacuation shelter information and disaster preparedness advice retrieved from the server are displayed, such as the location of the nearest evacuation shelter and a list of emergency supplies to take with you.

[0395] Step 8:

[0396] Server: Collects the latest evacuation shelter information from databases of local governments and government agencies and stores it in a database. This information is also reflected in the map service and made available to users.

[0397] Step 9:

[0398] User: Make necessary preparations based on the displayed disaster risk information and evacuation shelter information. For example, prepare an emergency kit and check evacuation routes with family members.

[0399] Step 10:

[0400] Server: Collects various feedback and log data to help improve our services. Analyzes user access history and click data to identify new features and service improvements.

[0401] Example 1

[0402] 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."

[0403] In modern society, the frequency and scale of damage caused by natural disasters continue to increase. Conventional disaster information systems have difficulty collecting real-time data and making highly accurate predictions, which means users are unable to obtain the information they need to take prompt and appropriate action. In addition, updates to evacuation shelter information and disaster preparation information are often delayed, preventing users from acting based on the latest information. Therefore, there is a need to build a system that can quickly and accurately predict disaster risks and provide users with useful information.

[0404] 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.

[0405] In this invention, the server includes means for collecting natural disaster data in real time, means for storing the collected natural disaster data in a database and organizing it while maintaining consistency, means for performing data preprocessing using the collected natural disaster data, means for training a generative AI model using the preprocessed data, means for predicting disaster occurrence probabilities in real time using the generative AI model, means for re-storing the predicted disaster occurrence probabilities in the database, means for reflecting the predicted disaster occurrence probabilities in a map service, means for collecting evacuation shelter information from local governments and government agencies and storing it in the database, means for providing evacuation shelter information and disaster preparation information to users, and terminal interface means for users to check the evacuation shelter information and disaster occurrence probability information. This allows users to quickly and accurately obtain predicted disaster risk information and take appropriate evacuation actions and preparations.

[0406] A "server" is a computer system that has the ability to provide various services and data to other computers over a network.

[0407] "Natural disaster data" refers to data that includes information on natural phenomena such as earthquakes, tsunamis, and fires.

[0408] A "database" is a system for efficiently storing, retrieving, and updating collected and organized data.

[0409] "Data preprocessing" is the process of converting data into a format suitable for analysis and learning models, and completing and normalizing missing values.

[0410] A "generative AI model" is an artificial intelligence model that learns from collected data and makes predictions about new data.

[0411] "Disaster occurrence probability" is a predicted value that indicates the probability that a specific natural disaster will occur within a certain period of time.

[0412] A "map service" is a web service or application that visually displays geographic information and various data.

[0413] "Evacuation shelter information" is information about places where people can evacuate in the event of a disaster.

[0414] A "terminal interface" is an operation screen for a device or application that allows a user to access a system and obtain information.

[0415] "Local governments and government agencies" refers to local government and national public institutions, which are organizations that provide various public services and information.

[0416] The following describes in detail an embodiment of the present invention. This system is composed of a server, a terminal, and a user component, and by these components working together, it is possible to collect, analyze, and predict natural disaster data in real time, and visualize it through a map service.

[0417] Server Roles

[0418] The server first collects natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. For example, it uses the Japan Meteorological Agency's API to obtain earthquake information. Requests are sent at regular intervals, and JSON-formatted data is returned as a response.

[0419] The server then stores the collected data in a database, which can be structured for consistency and organized by time and location, for example, using MongoDB.

[0420] The server then performs data preprocessing, including normalizing the data and imputing missing values. Normalization aligns the scale of the data, allowing for efficient training of generative AI models. For example, Pandas is used to normalize the "magnitude" scale from 0 to 1.

[0421] The server then uses the data to train a generative AI model. To train the model using data from past disasters, the server uses a machine learning framework such as TensorFlow to build a long short-term memory (LSTM) network and train the model on earthquake data.

[0422] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. The prediction results are then stored in the database again, ready for use with the map service.

[0423] The server then sends the prediction results to the API of the map service, which plots disaster risk information for each region on a map, with high-risk areas displayed in red and low-risk areas in green.

[0424] In addition, the server collects evacuation shelter information from databases of local governments and government agencies, stores the latest evacuation shelter information in the database, and sends this information to the map service so that users can check it.

[0425] Device Role

[0426] The device acts as an interface for using the map service. When a user accesses the map service via a web browser or mobile app, the device displays the latest disaster risk information and evacuation shelter information.

[0427] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0428] The marks displayed on the device's map can be clicked to view detailed information. Route guidance to evacuation shelters is also provided, supporting specific evacuation actions.

[0429] User Roles

[0430] Users can use the map service to check disaster risk information and evacuation shelter information for their area. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0431] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake occurring, they can check to see if they have prepared an emergency kit and how to contact their family members. They can also check the route to the nearest evacuation shelter in advance, allowing them to act quickly and calmly in the event of an emergency.

[0432] Prompt Sentence Examples

[0433] Example prompts to input to a generative AI model:

[0434] "Using the past earthquake data below, please predict the probability of an earthquake occurring within the next 24 hours. This data is limited to the Tokyo area."

[0435] This system is a powerful tool for promoting rapid and appropriate responses before a disaster occurs, and by each component working together, users can obtain the information they need in a timely manner and act safely.

[0436] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0437] Step 1:

[0438] The server collects natural disaster data. Specifically, it periodically sends requests to an API that provides data on earthquakes, fires, tsunamis, etc. In this process, the input is the API endpoint URL, and the output is natural disaster data in JSON format. For example, it sends a request to "https: / / api.kishou.go.jp / earthquake" and receives the returned data.

[0439] Step 2:

[0440] The server stores the collected natural disaster data in a database. The input is the collected data in JSON format, and the output is the organized data stored in the database. Specifically, it uses MongoDB to insert earthquake data into the "earthquake_data" collection, which contains fields such as "timestamp," "location," and "magnitude."

[0441] Step 3:

[0442] The server performs data preprocessing. The input is natural disaster data retrieved from the database, and the output is the preprocessed data. Specifically, it uses Pandas to normalize the data and convert the values ​​of the "magnitude" field into the range of 0 to 1. It also imputes missing values ​​with appropriate values ​​if any.

[0443] Step 4:

[0444] The server trains a generative AI model using the preprocessed data. The input is the preprocessed data, and the output is a trained generative AI model. Using TensorFlow, an LSTM network is constructed and the model is trained using past earthquake data. Specifically, the LSTM model is trained using earthquake data from the past five years.

[0445] Step 5:

[0446] The server uses a trained generative AI model to predict the probability of a disaster. The input is newly collected natural disaster data, and the output is the predicted probability of a disaster. New earthquake data is input into the model, and the probability of an earthquake occurring within the next 24 hours is calculated.

[0447] Step 6:

[0448] The server stores the prediction results in the database again. The input is the predicted probability of a disaster occurring, and the output is the prediction result stored in the database. Specifically, the server inserts the predicted earthquake occurrence probability into the "predictions" collection.

[0449] Step 7:

[0450] The server sends the prediction results to the map service's API. The input is the prediction results obtained from the database, and the output is the disaster risk information reflected in the map service. Using the Google Maps API, disaster risk data for each region is plotted on a map. Areas with high risk are displayed in red, and areas with low risk are displayed in green.

[0451] Step 8:

[0452] The terminal is an interface for using map services. The input is the user's search query, and the output is disaster risk information and evacuation shelter information displayed on the map. Specifically, when a user searches for a specific area, disaster information and the probability of disaster occurrence related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0453] Step 9:

[0454] Users use map services to check disaster risk information and evacuation shelter information for the area in which they live. The input is various operations performed by the user (for example, searching for an area or clicking a mark), and the output is related disaster information and evacuation shelter information. Specifically, when a user clicks on a warning mark on the map, information on the nearest evacuation shelter and preparation information such as a "list of items necessary for an emergency evacuation bag" and "how to check on the safety of family members" is displayed.

[0455] (Application example 1)

[0456] 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."

[0457] In recent years, the increasing number of natural disasters has created a need for real-time collection, analysis, and prediction of disaster information. However, existing systems do not provide sufficient, fast, and accurate information when a disaster occurs, which often delays evacuation and preparation. Furthermore, the provision of specific evacuation instructions and disaster prevention guidelines to users is insufficient, making it difficult to respond to disasters effectively.

[0458] 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.

[0459] In this invention, the server includes means for collecting natural disaster data in real time, means for training a generative AI model using the collected natural disaster data, means for predicting the probability of a disaster occurring using the generative AI model, means for reflecting the predicted probability of a disaster occurring in a map service, means for providing evacuation shelter information and disaster preparation information, means for providing disaster notifications in real time on the terminal, means for displaying route guidance to evacuation shelters on the terminal, and means for presenting a disaster action guide and an emergency kit list on the terminal. This allows users to obtain quick and accurate disaster information and provides specific evacuation instructions and disaster prevention guidelines, enabling appropriate disaster response.

[0460] "Means for collecting natural disaster data in real time" refers to technological means for instantly obtaining information on natural disasters such as earthquakes, fires, and tsunamis.

[0461] "Means for training a generative AI model using collected natural disaster data" refers to technical means for training a generative AI model based on acquired natural disaster data.

[0462] "Means for predicting the probability of disaster occurrence using a generative AI model" refers to a technical means for utilizing a trained generative AI model to quantify and predict the possibility of a disaster occurring.

[0463] "Means for reflecting the predicted probability of disaster occurrence on the map service" refers to the technical means for displaying the probability of disaster occurrence predicted by the generative AI model on the map service.

[0464] "Means for providing information on evacuation shelters and disaster preparation information" refers to technical means for providing users with information on the locations of evacuation shelters and the equipment needed in the event of a disaster.

[0465] "Means for providing real-time disaster notifications on devices" refers to technical means for sending users immediate notifications about disaster occurrences through devices such as smartphones or smart glasses.

[0466] "Means for displaying route guidance to a shelter on a device" refers to a technical means for displaying specific directions for a user to safely reach a shelter on a device such as a smartphone or smart glasses.

[0467] "Means for presenting disaster action guidelines and emergency bag lists on devices" refers to technological means for displaying specific actions to be taken and lists of necessary items to bring in the event of a disaster on devices such as smartphones or smart glasses.

[0468] An embodiment of the present invention will now be described. This system is comprised of a server, a terminal, and a user component, which work together to enable a rapid and appropriate response to natural disasters.

[0469] Server Roles

[0470] The server first collects natural disaster data in real time. This data is obtained through APIs related to disasters such as earthquakes, fires, and tsunamis. The collected data is then stored in a database and organized in an appropriate format. For example, this process involves using the Python requests library to obtain data from external APIs and storing the data in MySQL or PostgreSQL.

[0471] The server then uses this data to train a generative AI model. The model is trained using past disaster data to accurately predict the probability of a disaster. This training process involves preprocessing, such as normalizing the data and filling in missing data.

[0472] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. The prediction results are then stored in a database and sent to the map service's API, which visualizes disaster risk on a map.

[0473] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service API so that users can check it.

[0474] Device Role

[0475] The device is the interface through which users use the map service. When users access the map service using a web browser or mobile app, the latest disaster risk information and evacuation shelter information are displayed. For example, when a user searches for their local area, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information such as the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies is also provided.

[0476] The map displayed on the device can be clicked to view details. Route guidance to evacuation shelters is also provided, supporting specific evacuation actions. Real-time notifications are sent via smartphones, smart glasses, and other devices in the event of a disaster such as an earthquake or fire.

[0477] User Roles

[0478] Users can use the map service to check disaster risks and evacuation shelter information in their area. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors. For example, if a user checks their area on the map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. In addition, by checking the route to the nearest evacuation shelter in advance, they can act quickly and safely in the event of an emergency.

[0479] Specific examples

[0480] As a specific example, when a user in Tokyo receives a notification informing them of an increased earthquake risk, the message "Earthquake risk is increasing in the Tokyo area. Please check for the latest information" is displayed on the user's smartphone. Route guidance to an evacuation shelter, a list of emergency kits, and a prompt such as "List of items necessary for an emergency kit" are also provided.

[0481] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0482] Step 1:

[0483] The server collects natural disaster data in real time from an external API. The server periodically sends requests to the API endpoint to get the latest disaster information such as earthquakes, fires, tsunamis, etc. In this step, the input is the data obtained from the API, and the output is the raw natural disaster data.

[0484] Step 2:

[0485] The server stores the collected natural disaster data in a database. The server performs preprocessing to standardize the data format and fill in missing data. This inputs normalized natural disaster data into the database. The input is raw natural disaster data, and the output is organized database entries.

[0486] Step 3:

[0487] The server trains the generative AI model using the compiled natural disaster data. The server supplies past disaster data to the generative AI model and trains the disaster occurrence prediction algorithm. The input is past disaster data, and the output is the trained generative AI model.

[0488] Step 4:

[0489] The server inputs new disaster data into the AI ​​model in real time to predict the probability of a disaster occurring. The model evaluates disaster risk based on the latest data and generates the results. The input is the latest disaster data, and the output is a predicted disaster probability.

[0490] Step 5:

[0491] The server reflects the predicted disaster occurrence probability on the map service. The server sends this information to the map service API, which provides data for visualizing disaster risk on a map. The input is the prediction result, and the output is disaster risk information plotted on the map service.

[0492] Step 6:

[0493] The server collects evacuation shelter information from databases of local governments and government agencies and stores it in a database. The server retrieves the latest evacuation shelter information and prepares it for users to access. The input is the latest information on evacuation shelters, and the output is a compiled evacuation shelter database entry.

[0494] Step 7:

[0495] When a user accesses the map service, the device displays the latest disaster risk information and evacuation shelter information. When a user searches for their area, disaster data and evacuation shelter information for that area are displayed on the screen. The input is the user's search query, and the output is disaster information and evacuation shelter information displayed on the device's display.

[0496] Step 8:

[0497] The device provides users with real-time disaster notifications. The server pushes disaster information to the device so that users can check it immediately. The input is newly collected disaster data, and the output is notifications on the device.

[0498] Step 9:

[0499] The device displays route guidance to the nearest evacuation shelter. When the user clicks on the evacuation shelter marking, the device displays the route using GPS and map service APIs. The input is the user's current location and the location information of the evacuation shelter, and the output is the route guidance displayed on the device.

[0500] Step 10:

[0501] The device presents a list of action guidelines and emergency kits for emergencies. When the user checks the disaster information, a specific action plan and a list of necessary items are displayed. The input is predicted disaster risk information, and the output is the display of the action guidelines and list.

[0502] 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.

[0503] An embodiment of the present invention will now be described. The system of the present invention is composed of a server, a terminal, and a user component, and by combining it with an emotion engine that recognizes the user's emotions, it is possible to collect, analyze, and predict natural disaster data in real time and visualize it through a map service. This system not only provides disaster information and evacuation shelter information, but also provides customized information and messages according to the user's emotional state.

[0504] Server Roles

[0505] The server first collects various natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. The collected data is then stored in a database and organized in an appropriate format.

[0506] The server then uses a generative AI model to train this data. The model is trained using past disaster data so that it can predict the probability of disaster occurrence with high accuracy. This training process requires preprocessing the data, such as normalization and missing data completion.

[0507] New data is input in real time using the generative AI model to predict the probability of a disaster occurring. The prediction results are then stored in a database for use in the map service.

[0508] The server then sends the predicted disaster probability data to the map service's API, which plots disaster risk information for each region on a map. Marks on the map use different colors and symbols to visualize the level of disaster risk.

[0509] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0510] The role of the emotional engine

[0511] The emotion engine recognizes the user's emotional state and provides customized messages and advice based on that information. When the user browses disaster information, the emotion engine collects and analyzes emotional data via the camera and microphone. Using the results of this analysis, if the user is feeling anxious or tense, it displays a reassuring message.

[0512] Device Role

[0513] The device is the interface through which users use the map service. When a user accesses the map service using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. Furthermore, information based on the user's emotional state is also displayed based on the analysis results of the emotion engine.

[0514] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0515] Marks on the map displayed on the device can be clicked to view details, and route guidance to evacuation shelters is also provided to support specific evacuation actions.

[0516] User Roles

[0517] Users can use the map service to check disaster risks and evacuation shelter information for their area. They can also receive information customized based on their emotional state using the emotion engine. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0518] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0519] Specific examples

[0520] As a specific example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0521] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0522] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice through the camera and microphone to recognize their current emotional state. If the user feels anxious or nervous, it displays a reassuring message such as, "Please remain calm. There is a shelter nearby. We will show you the route to the nearest shelter."

[0523] Users access the map service and check the warning mark for the Tokyo area. When the user clicks on the warning mark, information about the nearest evacuation shelter retrieved from the server is displayed. At the same time, specific preparation information such as a list of items needed in an emergency evacuation bag and how to check on the safety of family members is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0524] In this way, by combining the emotion engine, support tailored to the user's emotional state can be provided, enabling more effective disaster response. This system is a powerful tool for encouraging prompt and appropriate responses before a disaster occurs.

[0525] The processing flow will be explained below.

[0526] Step 1:

[0527] Server: Periodically sends requests to APIs for various natural disaster data (earthquakes, fires, tsunamis, etc.) to collect new data. The collected data is formatted and stored in a database.

[0528] Step 2:

[0529] Server: Extracts past natural disaster data from the database and preprocesses it as a training dataset for the generative AI model. Preprocessing includes normalizing the data and filling in missing data.

[0530] Step 3:

[0531] Server: The preprocessed dataset is used to train a generative AI model, for example, by training the model using a machine learning framework (such as TensorFlow or PyTorch).

[0532] Step 4:

[0533] Server: Using the generative AI model that has completed training, new disaster data is input in real time to predict the probability of disaster occurrence. The prediction results are stored in a database.

[0534] Step 5:

[0535] Server: Connects to the Yahoo Maps API and reflects predicted disaster probability data on the map. Specifically, areas with high disaster risk are marked with warning marks or color-coded.

[0536] Step 6:

[0537] Device: When a user accesses Yahoo! Maps using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on a map.

[0538] Step 7:

[0539] Device: When a user clicks on a warning mark on the map, evacuation shelter information and disaster preparedness advice retrieved from the server are displayed, such as the location of the nearest evacuation shelter and a list of emergency supplies to take with you.

[0540] Step 8:

[0541] Server: Collects the latest evacuation shelter information from databases of local governments and government agencies and stores it in a database. This information is also reflected in the map service and made available to users.

[0542] Step 9:

[0543] Emotion engine: When a user uses the map service, the system collects emotional data from the user via the camera and microphone. For example, it uses facial recognition and voice analysis to assess whether the user is feeling anxious or nervous.

[0544] Step 10:

[0545] Emotion engine: Analyzes collected emotional data and generates customized messages and advice based on the user's emotional state. For example, if the user is feeling anxious, the engine generates a message such as, "Please remain calm. Information about the nearest evacuation shelter will be displayed."

[0546] Step 11:

[0547] Terminal: Displays customized messages generated by the emotion engine to the user, providing messages that reassure the user and specific guidelines for action.

[0548] Step 12:

[0549] User: Make necessary preparations based on the displayed disaster risk information and evacuation shelter information. For example, prepare an emergency kit and check evacuation routes with family members.

[0550] Step 13:

[0551] Server: Collects various feedback and log data to help improve our services. Analyzes user access history and click data to identify new features and service improvements.

[0552] Example 2

[0553] 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."

[0554] Natural disasters have a significant impact on human lives and property, so providing fast and accurate information is essential. However, existing disaster information systems lack the ability to collect and analyze information in real time, and they do not provide customized information based on the user's emotional state. This makes it difficult for users to take appropriate action, and anxiety and confusion are likely to arise.

[0555] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0556] In this invention, the server includes means for collecting natural disaster data in real time, means for storing the collected natural disaster data in a database, means for preprocessing the data and training a generative AI model, means for predicting the probability of a disaster occurring using the generative AI model, means for reflecting the predicted probability of a disaster occurring in a map service, means for providing evacuation shelter information and disaster preparation information, and means for recognizing a user's emotional state and displaying a customized message according to that state. This allows users to receive accurate natural disaster information in real time, and also to receive appropriate guidelines for action and reassurance messages according to their emotional state.

[0557] "Means for collecting natural disaster data in real time" refers to a system or device for obtaining information on natural disasters such as earthquakes, fires, and tsunamis in a timely manner from external data sources.

[0558] "Means for storing collected natural disaster data in a database" refers to a database system and related technologies for efficiently managing and storing acquired natural disaster data and making it accessible later.

[0559] "Data preprocessing and means for training generative AI models" refers to techniques for properly preparing data and converting it into the format required for an AI model to learn, and then using AI techniques to train the model.

[0560] "Means for predicting the probability of disaster occurrence using a generative AI model" refers to technologies and systems that use a generative AI model to calculate the likelihood of future disaster occurrence based on collected natural disaster data.

[0561] "Means for reflecting the predicted probability of disaster occurrence in map services" refers to technology and systems that display the probability information of disaster occurrence predicted by an AI model on a visual map, making it easier for users to understand visually.

[0562] "Means for providing evacuation shelter information and disaster preparation information" refers to technologies and systems that provide users with specific information about where to evacuate and necessary preparations when a disaster occurs, before, during, or after the disaster.

[0563] "Means for recognizing the user's emotional state and displaying a customized message according to that state" refers to technology that analyzes the user's facial and voice data to determine their emotional state, and then displays individual reassuring messages or advice based on the results.

[0564] MODE FOR CARRYING OUT THE INVENTION

[0565] The system of the present invention is composed of a server, a terminal, and a user component, and by combining it with an emotion engine, it collects, analyzes, and predicts natural disaster data in real time, and visualizes it through a map service. A specific embodiment of this system is described in detail below.

[0566] Server Roles

[0567] The server first collects real-time data on natural disasters such as earthquakes, fires, and tsunamis. Specifically, the server periodically sends requests to external APIs (e.g., earthquake information APIs, weather information APIs) to obtain the latest disaster data. To do this, it uses a program library such as Python's requests library. The obtained data is received in JSON format or similar and stored in a database (e.g., MySQL, MongoDB).

[0568] The server then trains the generative AI model using this data. Data preprocessing involves normalizing the data and imputing missing data. This is done using data analysis libraries such as scikit-learn and pandas. The preprocessed data is then used to train the generative AI model using an AI framework such as TensorFlow or PyTorch.

[0569] Using a trained generative AI model, new data is input in real time to predict the probability of a disaster occurring. The prediction results are then stored in a database and sent to the map service's API (e.g., Google Maps API). Disaster risk information for each region is then plotted on a map. Marks on the map use different colors and symbols to visualize the level of disaster risk.

[0570] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0571] The role of the emotional engine

[0572] The emotion engine recognizes the user's emotional state and provides customized messages and advice based on that information. When the user browses disaster information, emotional data is collected via the camera and microphone and analyzed. This analysis uses facial expression recognition technology (e.g., Amazon Rekognition) and voice analysis technology (e.g., Google Cloud Speech-to-Text). Using the analysis results, if the user is feeling anxious or tense, a reassuring message is displayed.

[0573] Device Role

[0574] The device functions as an interface for users to use the map service. When a user accesses the map service using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. The device displays the map and information using web technologies such as HTML, CSS, and JavaScript. It also displays information according to the user's emotional state based on the analysis results of an emotion engine.

[0575] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided. Marks on the map can be clicked to view details. Route guidance to the evacuation shelter is also provided to support specific evacuation actions.

[0576] User Roles

[0577] Users can use the map service to check disaster risks and evacuation shelter information for their area. They can also receive information customized based on their emotional state using the emotion engine. This allows them to prepare for evacuation as needed and take appropriate measures in cooperation with their family and neighbors.

[0578] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0579] Specific examples

[0580] For example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0581] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0582] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice via the camera and microphone to recognize their current emotional state. If the user feels anxious or nervous, it displays reassuring messages such as "Please stay calm. There is a shelter nearby" or "We will show you the route to the nearest shelter."

[0583] Users access the map service and check the warning symbols for the Tokyo area. When the user clicks on the symbol, information about the nearest evacuation shelter is displayed, retrieved from the server. At the same time, specific information on preparations, such as a list of items needed in an emergency bag and how to check on the safety of family members, is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0584] Prompt Sentence Examples

[0585] Here are some example prompts to input to a generative AI model:

[0586] "Predict the probability of an earthquake occurring in the next 12 hours."

[0587] "Get the latest shelter information and display it to the user."

[0588] "Analyze the user's emotional state and display appropriate messages."

[0589] As described above, the present invention not only collects, predicts, and displays disaster information, but also supports appropriate behavior in emergencies by providing customized messages that take into account the user's emotional state.

[0590] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0591] Step 1:

[0592] The server sends requests to an external natural disaster data API to retrieve natural disaster data in real time. Specifically, it uses the Python requests library to send a GET request to the API. It uses the API's endpoint URL as input and obtains the retrieved data in JSON format as output. This data includes disaster information such as earthquakes, fires, and tsunamis.

[0593] Step 2:

[0594] The server stores the acquired natural disaster data in a database. Specifically, it uses MySQL or MongoDB to store the data. The acquired JSON data is used as input, and disaster data is saved in the database as output. The database includes information such as the type of disaster, time of occurrence, location, and magnitude.

[0595] Step 3:

[0596] The server preprocesses the data stored in the database. Specifically, it uses scikit-learn and pandas to normalize the data and impute missing data. It uses raw data from the database as input and obtains preprocessed data as output. This puts the data in a format suitable for training AI models.

[0597] Step 4:

[0598] The server uses the preprocessed data to train a generative AI model, specifically using TensorFlow or PyTorch. It uses the preprocessed data and labeled data as input and obtains a trained AI model as output. This AI model is trained to be used to predict the probability of disaster occurrence.

[0599] Step 5:

[0600] The server inputs new natural disaster data into the generative AI model to predict the probability of disaster occurrence. It uses real-time data as input and obtains the predicted probability of disaster occurrence as output. To do this, the AI ​​model performs predictive calculations to calculate the probability of the next possible disaster.

[0601] Step 6:

[0602] The server sends the predicted disaster probability to a map service API, which plots disaster risk information for each region on a map. It uses disaster probability data as input and obtains a visual plot on the map service as output. It places marks on the map using APIs such as Google Maps to display the risk level.

[0603] Step 7:

[0604] The server retrieves evacuation shelter information from local government and government agency databases and stores it in a database. It uses the endpoint URL of the external database as input and stores the retrieved evacuation shelter information as output in an internal database. This includes information such as the latest evacuation shelter locations and capacity.

[0605] Step 8:

[0606] The emotion engine collects emotion data from the camera and microphone when the user browses disaster information. It uses the user's facial expression and voice data as input and obtains emotion analysis results as output. This is done using facial expression recognition software (e.g., Amazon Rekognition) and voice analysis software (e.g., Google Cloud Speech-to-Text).

[0607] Step 9:

[0608] Based on the analysis results, the emotion engine generates and displays a reassuring message if the user feels anxious or tense. It uses the emotion analysis results as input and gets a customized reassuring message as output, which helps the user relax and take appropriate action.

[0609] Step 10:

[0610] The device displays disaster risk information and evacuation shelter information to the user. It uses disaster risk data and evacuation shelter information obtained from the server as input and displays them on a map service or information screen as output. Information is provided to the user in real time via a web browser or mobile app.

[0611] (Application example 2)

[0612] 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."

[0613] In recent years, the frequency of natural disasters has increased, making countermeasures an urgent need. Meanwhile, with the widespread adoption of autonomous vehicles, it is necessary to ensure the safe operation of vehicles and the peace of mind of users during disasters. However, conventional systems have struggled to efficiently collect and analyze disaster information in real time and provide specific disaster prevention measures. Furthermore, they have been unable to provide customized messages based on the user's emotional state, making it difficult to encourage appropriate responses when users feel anxious.

[0614] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0615] In this invention, the server includes means for collecting natural disaster data in real time, means for training a generative AI model using the collected natural disaster data, means for predicting the probability of disaster occurrence using the generative AI model, means for reflecting the predicted probability of disaster occurrence in a map service, means for providing evacuation shelter information and disaster preparation information, means for recognizing a user's emotional state and providing a customized message based thereon, and means for optimizing a safe driving route in the event of a natural disaster. This allows a user to obtain the latest disaster information in real time, ensure a safe driving route, and receive a customized message according to their emotional state, enabling them to use autonomous vehicles with peace of mind.

[0616] "Real-time" means instantly processing and reflecting events and information occurring at that time.

[0617] "Natural disaster data" refers to data that includes information on natural disasters such as earthquakes, tsunamis, and fires.

[0618] A "generative AI model" is a type of artificial intelligence that uses machine learning algorithms to learn from past data and make predictions and classifications based on new data.

[0619] "Probability of disaster occurrence" is an index that indicates the degree of possibility of a disaster occurring in a specific region or under specific conditions.

[0620] A "map service" is a service that provides visual geographic information and displays location and route information.

[0621] "Evacuation shelter information" is information about safe places to evacuate to in the event of a disaster.

[0622] "Disaster preparedness information" refers to information on specific methods of response and items to prepare in the event of a disaster.

[0623] "Emotional state" refers to the emotions and mental state that a user is feeling at that time.

[0624] A "customized message" is information or a message that is optimized to a user's individual needs and emotional state.

[0625] A "safe route during a natural disaster" is a route that allows safe travel while avoiding danger even in the event of a natural disaster.

[0626] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

[0627] System Overview

[0628] This system, which consists of a server, terminals, and users, collects natural disaster information in real time and provides users with appropriate countermeasures.The system also uses a generative AI model to predict the probability of disaster occurrence and provides customized messages according to the user's emotional state, supporting the safe operation of autonomous vehicles.

[0629] Server Roles

[0630] The server collects natural disaster data in real time. Specifically, it periodically sends requests to APIs provided by data providers for earthquakes, fires, tsunamis, etc. to obtain new data. The obtained data is stored in a database and organized appropriately.

[0631] The server then uses a generative AI model to train this data and predict the probability of a disaster occurring. During this training process, data preprocessing is performed, including normalization and missing data completion. The prediction results are then stored in the database again and sent to the map service's API. This allows the disaster risk for each region to be visualized on a map. Marks on the map use different colors and symbols to indicate the level of disaster risk.

[0632] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies and updates it in real time, which is also sent to the map service for users to view.

[0633] The role of the emotional engine

[0634] The emotion engine recognizes the user's emotional state and provides customized messages based on that. When a user browses disaster information, emotional data is collected and analyzed via the camera and microphone. Based on the analysis results, if the user is feeling anxious or tense, a reassuring message is displayed.

[0635] Device Role

[0636] The device is the interface through which users use the map service. When a user accesses the map service through a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. Furthermore, information corresponding to the user's emotional state is also displayed based on the analysis results of the emotion engine.

[0637] Disaster information and occurrence probability related to the area searched by the user are displayed on a map. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided. Marks on the map can be clicked to check details, and route guidance to the evacuation shelter is also provided.

[0638] User Roles

[0639] Users can use this system to check disaster risks and evacuation shelter information for their area. They can also receive customized information based on their emotional state through the emotion engine, which allows them to make appropriate evacuation preparations and take appropriate evacuation actions.

[0640] Specific examples

[0641] For example, if a prediction is made that there is a high probability of a major earthquake occurring in Tokyo, the server will train the generative AI model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and send this prediction result to the map service API, which will display a red warning mark in the Tokyo area.

[0642] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice via the camera and microphone to recognize their current emotional state. If the user is feeling anxious or nervous, it displays a reassuring message such as, "Please remain calm. There is a shelter nearby. We will show you the route to the nearest shelter."

[0643] Users access the map service and check the warning symbols for the Tokyo area. Clicking on the symbol displays information about the nearest evacuation shelter. At the same time, specific information on preparations, such as a list of items needed in an emergency bag and how to check on the safety of family members, is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0644] Example prompt sentence:

[0645] If an earthquake is predicted to occur in the Tokyo area, the system will display routes to the nearest evacuation shelter, check the user's emotional state, and display reassuring messages if they feel anxious. Meanwhile, the system will safely optimize the routes of autonomous vehicles.

[0646] As described above, this system obtains disaster information in real time and provides customized messages according to the user's emotional state, thereby creating an environment in which users can use self-driving vehicles with peace of mind.

[0647] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0648] Step 1:

[0649] The server collects natural disaster data in real time from an external API. It sends API requests and stores the acquired data in a database. The input is the natural disaster data acquired from the API, and the output is the organized data stored in the database. The API requests are programmed to be made periodically.

[0650] Step 2:

[0651] The server trains the generative AI model using natural disaster data stored in a database. Data preprocessing involves normalization and missing data completion. The input is past disaster data obtained from the database, and the output is the trained generative AI model. This makes it possible to predict the probability of disaster occurrence with high accuracy.

[0652] Step 3:

[0653] The server uses the trained generative AI model to input new data in real time and predict the probability of a disaster. The input is the latest natural disaster data, and the output is data indicating the probability of a disaster occurring. The results are then stored in a database and sent to the map service's API.

[0654] Step 4:

[0655] The server visualizes the disaster risk for each region on a map based on the data sent to the map service's API. Marks on the map use different colors and symbols to indicate the level of disaster risk. The input is disaster probability data, and the output is visually plotted map data.

[0656] Step 5:

[0657] The server retrieves the latest evacuation shelter information from local government and government agency databases and stores it in the database. The input is the evacuation shelter information retrieved from the external database, and the output is the updated evacuation shelter data. This information is also sent to the map service so that users can view it.

[0658] Step 6:

[0659] The device accesses the map service through a web browser or mobile app to obtain disaster risk information for the user's current location or a specified area. The input is the user's location information and data from the map service API, and the output is a disaster risk map displayed on the device.

[0660] Step 7:

[0661] The emotion engine uses a camera and microphone to recognize the user's emotional state. It analyzes facial expressions and voice to generate a customized message if the user is feeling anxious or nervous. The input is the user's image and voice data, and the output is the emotion analysis result and a reassuring message based on that.

[0662] Step 8:

[0663] The device displays a customized message based on the user's emotional state based on the analysis results of the emotion engine. The input is the emotion analysis result, and the output is the message displayed on the device. This allows users to use the system with peace of mind.

[0664] Step 9:

[0665] Users can check evacuation shelter information and disaster preparedness information through their devices. The input is evacuation shelter information obtained from a map service, and the output is evacuation information and preparedness information displayed on the device. Based on this, users can take specific evacuation actions.

[0666] In this way, the entire system works seamlessly, allowing users to obtain the latest disaster and evacuation information in real time. Furthermore, by providing customized messages according to the user's emotional state, users can feel at ease and take appropriate action.

[0667] 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.

[0668] 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.

[0669] 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.

[0670] [Third embodiment]

[0671] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0672] 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.

[0673] 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).

[0674] 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.

[0675] 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.

[0676] 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).

[0677] 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.

[0678] 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.

[0679] 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.

[0680] 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.

[0681] 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.

[0682] 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."

[0683] This section describes an embodiment of the present invention. This system is composed of a server, a terminal, and a user component, and by these components working together, it is possible to collect, analyze, and predict natural disaster data in real time and visualize it through a map service.

[0684] Server Roles

[0685] The server first collects various natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. The collected data is then stored in a database and organized in an appropriate format.

[0686] The server then uses a generative AI model to train this data. The model is trained using past disaster data so that it can predict the probability of disaster occurrence with high accuracy. This training process requires preprocessing the data, such as normalization and missing data completion.

[0687] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. These prediction results are then stored in the database again for use in the map service.

[0688] The server then sends the predicted disaster probability data to the map service's API, which plots disaster risk information for each region on a map. The marks on the map use different colors and symbols to visualize the level of disaster risk.

[0689] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0690] Device Role

[0691] The device is the interface through which users access the map service. When a user accesses the map service via a web browser or mobile app, the device displays the latest disaster risk information and evacuation shelter information.

[0692] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0693] Marks on the map displayed on the device can be clicked to view details, and route guidance to evacuation shelters is also provided to support specific evacuation actions.

[0694] User Roles

[0695] Users can use the map service to check disaster risks and evacuation shelter information for their area, and if necessary, prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0696] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0697] Specific examples

[0698] As a specific example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0699] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0700] Users access the map service and check the warning symbols for the Tokyo area. When users click on the symbol, information about the nearest evacuation shelter is displayed, retrieved from the server. At the same time, specific preparation information is also provided, such as a list of items needed in an emergency evacuation bag and how to check on the safety of family members.

[0701] This will allow users to prepare for appropriate evacuation actions and evacuate quickly and safely in the event of an earthquake. This system is a powerful tool for promoting quick and appropriate responses before a disaster occurs.

[0702] The processing flow will be explained below.

[0703] Step 1:

[0704] Server: Periodically sends requests to APIs for various natural disaster data (earthquakes, fires, tsunamis, etc.) to collect new data. The collected data is formatted and stored in a database.

[0705] Step 2:

[0706] Server: Extracts past natural disaster data from the database and preprocesses it as a training dataset for the generative AI model. Preprocessing includes normalizing the data and filling in missing data.

[0707] Step 3:

[0708] Server: The preprocessed dataset is used to train a generative AI model, for example, by training the model using a machine learning framework (such as TensorFlow or PyTorch).

[0709] Step 4:

[0710] Server: Using the generative AI model that has completed training, new disaster data is input in real time to predict the probability of disaster occurrence. The prediction results are stored in a database.

[0711] Step 5:

[0712] Server: Connects to the Yahoo Maps API and reflects predicted disaster probability data on the map. Specifically, areas with high disaster risk are marked with warning marks or color-coded.

[0713] Step 6:

[0714] Device: When a user accesses Yahoo! Maps using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on a map.

[0715] Step 7:

[0716] Device: When a user clicks on a warning mark on the map, evacuation shelter information and disaster preparedness advice retrieved from the server are displayed, such as the location of the nearest evacuation shelter and a list of emergency supplies to take with you.

[0717] Step 8:

[0718] Server: Collects the latest evacuation shelter information from databases of local governments and government agencies and stores it in a database. This information is also reflected in the map service and made available to users.

[0719] Step 9:

[0720] User: Make necessary preparations based on the displayed disaster risk information and evacuation shelter information. For example, prepare an emergency kit and check evacuation routes with family members.

[0721] Step 10:

[0722] Server: Collects various feedback and log data to help improve our services. Analyzes user access history and click data to identify new features and service improvements.

[0723] Example 1

[0724] 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."

[0725] In modern society, the frequency and scale of damage caused by natural disasters continue to increase. Conventional disaster information systems have difficulty collecting real-time data and making highly accurate predictions, which means users are unable to obtain the information they need to take prompt and appropriate action. In addition, updates to evacuation shelter information and disaster preparation information are often delayed, preventing users from acting based on the latest information. Therefore, there is a need to build a system that can quickly and accurately predict disaster risks and provide users with useful information.

[0726] 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.

[0727] In this invention, the server includes means for collecting natural disaster data in real time, means for storing the collected natural disaster data in a database and organizing it while maintaining consistency, means for performing data preprocessing using the collected natural disaster data, means for training a generative AI model using the preprocessed data, means for predicting disaster occurrence probabilities in real time using the generative AI model, means for re-storing the predicted disaster occurrence probabilities in the database, means for reflecting the predicted disaster occurrence probabilities in a map service, means for collecting evacuation shelter information from local governments and government agencies and storing it in the database, means for providing evacuation shelter information and disaster preparation information to users, and terminal interface means for users to check the evacuation shelter information and disaster occurrence probability information. This allows users to quickly and accurately obtain predicted disaster risk information and take appropriate evacuation actions and preparations.

[0728] A "server" is a computer system that has the ability to provide various services and data to other computers over a network.

[0729] "Natural disaster data" refers to data that includes information on natural phenomena such as earthquakes, tsunamis, and fires.

[0730] A "database" is a system for efficiently storing, retrieving, and updating collected and organized data.

[0731] "Data preprocessing" is the process of converting data into a format suitable for analysis and learning models, and completing and normalizing missing values.

[0732] A "generative AI model" is an artificial intelligence model that learns from collected data and makes predictions about new data.

[0733] "Disaster occurrence probability" is a predicted value that indicates the probability that a specific natural disaster will occur within a certain period of time.

[0734] A "map service" is a web service or application that visually displays geographic information and various data.

[0735] "Evacuation shelter information" is information about places where people can evacuate in the event of a disaster.

[0736] A "terminal interface" is an operation screen for a device or application that allows a user to access a system and obtain information.

[0737] "Local governments and government agencies" refers to local government and national public institutions, which are organizations that provide various public services and information.

[0738] The following describes in detail an embodiment of the present invention. This system is composed of a server, a terminal, and a user component, and by these components working together, it is possible to collect, analyze, and predict natural disaster data in real time, and visualize it through a map service.

[0739] Server Roles

[0740] The server first collects natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. For example, it uses the Japan Meteorological Agency's API to obtain earthquake information. Requests are sent at regular intervals, and JSON-formatted data is returned as a response.

[0741] The server then stores the collected data in a database, which can be structured for consistency and organized by time and location, for example, using MongoDB.

[0742] The server then performs data preprocessing, including normalizing the data and imputing missing values. Normalization aligns the scale of the data, allowing for efficient training of generative AI models. For example, Pandas is used to normalize the "magnitude" scale from 0 to 1.

[0743] The server then uses the data to train a generative AI model. To train the model using data from past disasters, the server uses a machine learning framework such as TensorFlow to build a long short-term memory (LSTM) network and train the model on earthquake data.

[0744] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. The prediction results are then stored in the database again, ready for use with the map service.

[0745] The server then sends the prediction results to the API of the map service, which plots disaster risk information for each region on a map, with high-risk areas displayed in red and low-risk areas in green.

[0746] In addition, the server collects evacuation shelter information from databases of local governments and government agencies, stores the latest evacuation shelter information in the database, and sends this information to the map service so that users can check it.

[0747] Device Role

[0748] The device acts as an interface for using the map service. When a user accesses the map service via a web browser or mobile app, the device displays the latest disaster risk information and evacuation shelter information.

[0749] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0750] The marks displayed on the device's map can be clicked to view detailed information. Route guidance to evacuation shelters is also provided, supporting specific evacuation actions.

[0751] User Roles

[0752] Users can use the map service to check disaster risk information and evacuation shelter information for their area. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0753] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake occurring, they can check to see if they have prepared an emergency kit and how to contact their family members. They can also check the route to the nearest evacuation shelter in advance, allowing them to act quickly and calmly in the event of an emergency.

[0754] Prompt Sentence Examples

[0755] Example prompts to input to a generative AI model:

[0756] "Using the past earthquake data below, please predict the probability of an earthquake occurring within the next 24 hours. This data is limited to the Tokyo area."

[0757] This system is a powerful tool for promoting rapid and appropriate responses before a disaster occurs, and by each component working together, users can obtain the information they need in a timely manner and act safely.

[0758] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0759] Step 1:

[0760] The server collects natural disaster data. Specifically, it periodically sends requests to an API that provides data on earthquakes, fires, tsunamis, etc. In this process, the input is the API endpoint URL, and the output is natural disaster data in JSON format. For example, it sends a request to "https: / / api.kishou.go.jp / earthquake" and receives the returned data.

[0761] Step 2:

[0762] The server stores the collected natural disaster data in a database. The input is the collected data in JSON format, and the output is the organized data stored in the database. Specifically, it uses MongoDB to insert earthquake data into the "earthquake_data" collection, which contains fields such as "timestamp," "location," and "magnitude."

[0763] Step 3:

[0764] The server performs data preprocessing. The input is natural disaster data retrieved from the database, and the output is the preprocessed data. Specifically, it uses Pandas to normalize the data and convert the values ​​of the "magnitude" field into the range of 0 to 1. It also imputes missing values ​​with appropriate values ​​if any.

[0765] Step 4:

[0766] The server trains a generative AI model using the preprocessed data. The input is the preprocessed data, and the output is a trained generative AI model. Using TensorFlow, an LSTM network is constructed and the model is trained using past earthquake data. Specifically, the LSTM model is trained using earthquake data from the past five years.

[0767] Step 5:

[0768] The server uses a trained generative AI model to predict the probability of a disaster. The input is newly collected natural disaster data, and the output is the predicted probability of a disaster. New earthquake data is input into the model, and the probability of an earthquake occurring within the next 24 hours is calculated.

[0769] Step 6:

[0770] The server stores the prediction results in the database again. The input is the predicted probability of a disaster occurring, and the output is the prediction result stored in the database. Specifically, the server inserts the predicted earthquake occurrence probability into the "predictions" collection.

[0771] Step 7:

[0772] The server sends the prediction results to the map service's API. The input is the prediction results obtained from the database, and the output is the disaster risk information reflected in the map service. Using the Google Maps API, disaster risk data for each region is plotted on a map. Areas with high risk are displayed in red, and areas with low risk are displayed in green.

[0773] Step 8:

[0774] The terminal is an interface for using map services. The input is the user's search query, and the output is disaster risk information and evacuation shelter information displayed on the map. Specifically, when a user searches for a specific area, disaster information and the probability of disaster occurrence related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0775] Step 9:

[0776] Users use map services to check disaster risk information and evacuation shelter information for the area in which they live. The input is various operations performed by the user (for example, searching for an area or clicking a mark), and the output is related disaster information and evacuation shelter information. Specifically, when a user clicks on a warning mark on the map, information on the nearest evacuation shelter and preparation information such as a "list of items necessary for an emergency evacuation bag" and "how to check on the safety of family members" is displayed.

[0777] (Application example 1)

[0778] 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."

[0779] In recent years, the increasing number of natural disasters has created a need for real-time collection, analysis, and prediction of disaster information. However, existing systems do not provide sufficient, fast, and accurate information when a disaster occurs, which often delays evacuation and preparation. Furthermore, the provision of specific evacuation instructions and disaster prevention guidelines to users is insufficient, making it difficult to respond to disasters effectively.

[0780] 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.

[0781] In this invention, the server includes means for collecting natural disaster data in real time, means for training a generative AI model using the collected natural disaster data, means for predicting the probability of a disaster occurring using the generative AI model, means for reflecting the predicted probability of a disaster occurring in a map service, means for providing evacuation shelter information and disaster preparation information, means for providing disaster notifications in real time on the terminal, means for displaying route guidance to evacuation shelters on the terminal, and means for presenting a disaster action guide and an emergency kit list on the terminal. This allows users to obtain quick and accurate disaster information and provides specific evacuation instructions and disaster prevention guidelines, enabling appropriate disaster response.

[0782] "Means for collecting natural disaster data in real time" refers to technological means for instantly obtaining information on natural disasters such as earthquakes, fires, and tsunamis.

[0783] "Means for training a generative AI model using collected natural disaster data" refers to technical means for training a generative AI model based on acquired natural disaster data.

[0784] "Means for predicting the probability of disaster occurrence using a generative AI model" refers to a technical means for utilizing a trained generative AI model to quantify and predict the possibility of a disaster occurring.

[0785] "Means for reflecting the predicted probability of disaster occurrence on the map service" refers to the technical means for displaying the probability of disaster occurrence predicted by the generative AI model on the map service.

[0786] "Means for providing information on evacuation shelters and disaster preparation information" refers to technical means for providing users with information on the locations of evacuation shelters and the equipment needed in the event of a disaster.

[0787] "Means for providing real-time disaster notifications on devices" refers to technical means for sending users immediate notifications about disaster occurrences through devices such as smartphones or smart glasses.

[0788] "Means for displaying route guidance to a shelter on a device" refers to a technical means for displaying specific directions for a user to safely reach a shelter on a device such as a smartphone or smart glasses.

[0789] "Means for presenting disaster action guidelines and emergency bag lists on devices" refers to technological means for displaying specific actions to be taken and lists of necessary items to bring in the event of a disaster on devices such as smartphones or smart glasses.

[0790] An embodiment of the present invention will now be described. This system is comprised of a server, a terminal, and a user component, which work together to enable a rapid and appropriate response to natural disasters.

[0791] Server Roles

[0792] The server first collects natural disaster data in real time. This data is obtained through APIs related to disasters such as earthquakes, fires, and tsunamis. The collected data is then stored in a database and organized in an appropriate format. For example, this process involves using the Python requests library to obtain data from external APIs and storing the data in MySQL or PostgreSQL.

[0793] The server then uses this data to train a generative AI model. The model is trained using past disaster data to accurately predict the probability of a disaster. This training process involves preprocessing, such as normalizing the data and filling in missing data.

[0794] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. The prediction results are then stored in a database and sent to the map service's API, which visualizes disaster risk on a map.

[0795] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service API so that users can check it.

[0796] Device Role

[0797] The device is the interface through which users use the map service. When users access the map service using a web browser or mobile app, the latest disaster risk information and evacuation shelter information are displayed. For example, when a user searches for their local area, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information such as the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies is also provided.

[0798] The map displayed on the device can be clicked to view details. Route guidance to evacuation shelters is also provided, supporting specific evacuation actions. Real-time notifications are sent via smartphones, smart glasses, and other devices in the event of a disaster such as an earthquake or fire.

[0799] User Roles

[0800] Users can use the map service to check disaster risks and evacuation shelter information in their area. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors. For example, if a user checks their area on the map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. In addition, by checking the route to the nearest evacuation shelter in advance, they can act quickly and safely in the event of an emergency.

[0801] Specific examples

[0802] As a specific example, when a user in Tokyo receives a notification informing them of an increased earthquake risk, the message "Earthquake risk is increasing in the Tokyo area. Please check for the latest information" is displayed on the user's smartphone. Route guidance to an evacuation shelter, a list of emergency kits, and a prompt such as "List of items necessary for an emergency kit" are also provided.

[0803] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0804] Step 1:

[0805] The server collects natural disaster data in real time from an external API. The server periodically sends requests to the API endpoint to get the latest disaster information such as earthquakes, fires, tsunamis, etc. In this step, the input is the data obtained from the API, and the output is the raw natural disaster data.

[0806] Step 2:

[0807] The server stores the collected natural disaster data in a database. The server performs preprocessing to standardize the data format and fill in missing data. This inputs normalized natural disaster data into the database. The input is raw natural disaster data, and the output is organized database entries.

[0808] Step 3:

[0809] The server trains the generative AI model using the compiled natural disaster data. The server supplies past disaster data to the generative AI model and trains the disaster occurrence prediction algorithm. The input is past disaster data, and the output is the trained generative AI model.

[0810] Step 4:

[0811] The server inputs new disaster data into the AI ​​model in real time to predict the probability of a disaster occurring. The model evaluates disaster risk based on the latest data and generates the results. The input is the latest disaster data, and the output is a predicted disaster probability.

[0812] Step 5:

[0813] The server reflects the predicted disaster occurrence probability on the map service. The server sends this information to the map service API, which provides data for visualizing disaster risk on a map. The input is the prediction result, and the output is disaster risk information plotted on the map service.

[0814] Step 6:

[0815] The server collects evacuation shelter information from databases of local governments and government agencies and stores it in a database. The server retrieves the latest evacuation shelter information and prepares it for users to access. The input is the latest information on evacuation shelters, and the output is a compiled evacuation shelter database entry.

[0816] Step 7:

[0817] When a user accesses the map service, the device displays the latest disaster risk information and evacuation shelter information. When a user searches for their area, disaster data and evacuation shelter information for that area are displayed on the screen. The input is the user's search query, and the output is disaster information and evacuation shelter information displayed on the device's display.

[0818] Step 8:

[0819] The device provides users with real-time disaster notifications. The server pushes disaster information to the device so that users can check it immediately. The input is newly collected disaster data, and the output is notifications on the device.

[0820] Step 9:

[0821] The device displays route guidance to the nearest evacuation shelter. When the user clicks on the evacuation shelter marking, the device displays the route using GPS and map service APIs. The input is the user's current location and the location information of the evacuation shelter, and the output is the route guidance displayed on the device.

[0822] Step 10:

[0823] The device presents a list of action guidelines and emergency kits for emergencies. When the user checks the disaster information, a specific action plan and a list of necessary items are displayed. The input is predicted disaster risk information, and the output is the display of the action guidelines and list.

[0824] 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.

[0825] An embodiment of the present invention will now be described. The system of the present invention is composed of a server, a terminal, and a user component, and by combining it with an emotion engine that recognizes the user's emotions, it is possible to collect, analyze, and predict natural disaster data in real time and visualize it through a map service. This system not only provides disaster information and evacuation shelter information, but also provides customized information and messages according to the user's emotional state.

[0826] Server Roles

[0827] The server first collects various natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. The collected data is then stored in a database and organized in an appropriate format.

[0828] The server then uses a generative AI model to train this data. The model is trained using past disaster data so that it can predict the probability of disaster occurrence with high accuracy. This training process requires preprocessing the data, such as normalization and missing data completion.

[0829] New data is input in real time using the generative AI model to predict the probability of a disaster occurring. The prediction results are then stored in a database for use in the map service.

[0830] The server then sends the predicted disaster probability data to the map service's API, which plots disaster risk information for each region on a map. Marks on the map use different colors and symbols to visualize the level of disaster risk.

[0831] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0832] The role of the emotional engine

[0833] The emotion engine recognizes the user's emotional state and provides customized messages and advice based on that information. When the user browses disaster information, the emotion engine collects and analyzes emotional data via the camera and microphone. Using the results of this analysis, if the user is feeling anxious or tense, it displays a reassuring message.

[0834] Device Role

[0835] The device is the interface through which users use the map service. When a user accesses the map service using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. Furthermore, information based on the user's emotional state is also displayed based on the analysis results of the emotion engine.

[0836] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[0837] Marks on the map displayed on the device can be clicked to view details, and route guidance to evacuation shelters is also provided to support specific evacuation actions.

[0838] User Roles

[0839] Users can use the map service to check disaster risks and evacuation shelter information for their area. They can also receive information customized based on their emotional state using the emotion engine. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[0840] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0841] Specific examples

[0842] As a specific example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0843] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0844] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice through the camera and microphone to recognize their current emotional state. If the user feels anxious or nervous, it displays a reassuring message such as, "Please remain calm. There is a shelter nearby. We will show you the route to the nearest shelter."

[0845] Users access the map service and check the warning mark for the Tokyo area. When the user clicks on the warning mark, information about the nearest evacuation shelter retrieved from the server is displayed. At the same time, specific preparation information such as a list of items needed in an emergency evacuation bag and how to check on the safety of family members is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0846] In this way, by combining the emotion engine, support tailored to the user's emotional state can be provided, enabling more effective disaster response. This system is a powerful tool for encouraging prompt and appropriate responses before a disaster occurs.

[0847] The processing flow will be explained below.

[0848] Step 1:

[0849] Server: Periodically sends requests to APIs for various natural disaster data (earthquakes, fires, tsunamis, etc.) to collect new data. The collected data is formatted and stored in a database.

[0850] Step 2:

[0851] Server: Extracts past natural disaster data from the database and preprocesses it as a training dataset for the generative AI model. Preprocessing includes normalizing the data and filling in missing data.

[0852] Step 3:

[0853] Server: The preprocessed dataset is used to train a generative AI model, for example, by training the model using a machine learning framework (such as TensorFlow or PyTorch).

[0854] Step 4:

[0855] Server: Using the generative AI model that has completed training, new disaster data is input in real time to predict the probability of disaster occurrence. The prediction results are stored in a database.

[0856] Step 5:

[0857] Server: Connects to the Yahoo Maps API and reflects predicted disaster probability data on the map. Specifically, areas with high disaster risk are marked with warning marks or color-coded.

[0858] Step 6:

[0859] Device: When a user accesses Yahoo! Maps using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on a map.

[0860] Step 7:

[0861] Device: When a user clicks on a warning mark on the map, evacuation shelter information and disaster preparedness advice retrieved from the server are displayed, such as the location of the nearest evacuation shelter and a list of emergency supplies to take with you.

[0862] Step 8:

[0863] Server: Collects the latest evacuation shelter information from databases of local governments and government agencies and stores it in a database. This information is also reflected in the map service and made available to users.

[0864] Step 9:

[0865] Emotion engine: When a user uses the map service, the system collects emotional data from the user via the camera and microphone. For example, it uses facial recognition and voice analysis to assess whether the user is feeling anxious or nervous.

[0866] Step 10:

[0867] Emotion engine: Analyzes collected emotional data and generates customized messages and advice based on the user's emotional state. For example, if the user is feeling anxious, the engine generates a message such as, "Please remain calm. Information about the nearest evacuation shelter will be displayed."

[0868] Step 11:

[0869] Terminal: Displays customized messages generated by the emotion engine to the user, providing messages that reassure the user and specific guidelines for action.

[0870] Step 12:

[0871] User: Make necessary preparations based on the displayed disaster risk information and evacuation shelter information. For example, prepare an emergency kit and check evacuation routes with family members.

[0872] Step 13:

[0873] Server: Collects various feedback and log data to help improve our services. Analyzes user access history and click data to identify new features and service improvements.

[0874] Example 2

[0875] 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."

[0876] Natural disasters have a significant impact on human lives and property, so providing fast and accurate information is essential. However, existing disaster information systems lack the ability to collect and analyze information in real time, and they do not provide customized information based on the user's emotional state. This makes it difficult for users to take appropriate action, and anxiety and confusion are likely to arise.

[0877] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0878] In this invention, the server includes means for collecting natural disaster data in real time, means for storing the collected natural disaster data in a database, means for preprocessing the data and training a generative AI model, means for predicting the probability of a disaster occurring using the generative AI model, means for reflecting the predicted probability of a disaster occurring in a map service, means for providing evacuation shelter information and disaster preparation information, and means for recognizing a user's emotional state and displaying a customized message according to that state. This allows users to receive accurate natural disaster information in real time, and also to receive appropriate guidelines for action and reassurance messages according to their emotional state.

[0879] "Means for collecting natural disaster data in real time" refers to a system or device for obtaining information on natural disasters such as earthquakes, fires, and tsunamis in a timely manner from external data sources.

[0880] "Means for storing collected natural disaster data in a database" refers to a database system and related technologies for efficiently managing and storing acquired natural disaster data and making it accessible later.

[0881] "Data preprocessing and means for training generative AI models" refers to techniques for properly preparing data and converting it into the format required for an AI model to learn, and then using AI techniques to train the model.

[0882] "Means for predicting the probability of disaster occurrence using a generative AI model" refers to technologies and systems that use a generative AI model to calculate the likelihood of future disaster occurrence based on collected natural disaster data.

[0883] "Means for reflecting the predicted probability of disaster occurrence in map services" refers to technology and systems that display the probability information of disaster occurrence predicted by an AI model on a visual map, making it easier for users to understand visually.

[0884] "Means for providing evacuation shelter information and disaster preparation information" refers to technologies and systems that provide users with specific information about where to evacuate and necessary preparations when a disaster occurs, before, during, or after the disaster.

[0885] "Means for recognizing the user's emotional state and displaying a customized message according to that state" refers to technology that analyzes the user's facial and voice data to determine their emotional state, and then displays individual reassuring messages or advice based on the results.

[0886] MODE FOR CARRYING OUT THE INVENTION

[0887] The system of the present invention is composed of a server, a terminal, and a user component, and by combining it with an emotion engine, it collects, analyzes, and predicts natural disaster data in real time, and visualizes it through a map service. A specific embodiment of this system is described in detail below.

[0888] Server Roles

[0889] The server first collects real-time data on natural disasters such as earthquakes, fires, and tsunamis. Specifically, the server periodically sends requests to external APIs (e.g., earthquake information APIs, weather information APIs) to obtain the latest disaster data. To do this, it uses a program library such as Python's requests library. The obtained data is received in JSON format or similar and stored in a database (e.g., MySQL, MongoDB).

[0890] The server then trains the generative AI model using this data. Data preprocessing involves normalizing the data and imputing missing data. This is done using data analysis libraries such as scikit-learn and pandas. The preprocessed data is then used to train the generative AI model using an AI framework such as TensorFlow or PyTorch.

[0891] Using a trained generative AI model, new data is input in real time to predict the probability of a disaster occurring. The prediction results are then stored in a database and sent to the map service's API (e.g., Google Maps API). Disaster risk information for each region is then plotted on a map. Marks on the map use different colors and symbols to visualize the level of disaster risk.

[0892] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[0893] The role of the emotional engine

[0894] The emotion engine recognizes the user's emotional state and provides customized messages and advice based on that information. When the user browses disaster information, emotional data is collected via the camera and microphone and analyzed. This analysis uses facial expression recognition technology (e.g., Amazon Rekognition) and voice analysis technology (e.g., Google Cloud Speech-to-Text). Using the analysis results, if the user is feeling anxious or tense, a reassuring message is displayed.

[0895] Device Role

[0896] The device functions as an interface for users to use the map service. When a user accesses the map service using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. The device displays the map and information using web technologies such as HTML, CSS, and JavaScript. It also displays information according to the user's emotional state based on the analysis results of an emotion engine.

[0897] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided. Marks on the map can be clicked to view details. Route guidance to the evacuation shelter is also provided to support specific evacuation actions.

[0898] User Roles

[0899] Users can use the map service to check disaster risks and evacuation shelter information for their area. They can also receive information customized based on their emotional state using the emotion engine. This allows them to prepare for evacuation as needed and take appropriate measures in cooperation with their family and neighbors.

[0900] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[0901] Specific examples

[0902] For example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[0903] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[0904] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice via the camera and microphone to recognize their current emotional state. If the user feels anxious or nervous, it displays reassuring messages such as "Please stay calm. There is a shelter nearby" or "We will show you the route to the nearest shelter."

[0905] Users access the map service and check the warning symbols for the Tokyo area. When the user clicks on the symbol, information about the nearest evacuation shelter is displayed, retrieved from the server. At the same time, specific information on preparations, such as a list of items needed in an emergency bag and how to check on the safety of family members, is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0906] Prompt Sentence Examples

[0907] Here are some example prompts to input to a generative AI model:

[0908] "Predict the probability of an earthquake occurring in the next 12 hours."

[0909] "Get the latest shelter information and display it to the user."

[0910] "Analyze the user's emotional state and display appropriate messages."

[0911] As described above, the present invention not only collects, predicts, and displays disaster information, but also supports appropriate behavior in emergencies by providing customized messages that take into account the user's emotional state.

[0912] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0913] Step 1:

[0914] The server sends requests to an external natural disaster data API to retrieve natural disaster data in real time. Specifically, it uses the Python requests library to send a GET request to the API. It uses the API's endpoint URL as input and obtains the retrieved data in JSON format as output. This data includes disaster information such as earthquakes, fires, and tsunamis.

[0915] Step 2:

[0916] The server stores the acquired natural disaster data in a database. Specifically, it uses MySQL or MongoDB to store the data. The acquired JSON data is used as input, and disaster data is saved in the database as output. The database includes information such as the type of disaster, time of occurrence, location, and magnitude.

[0917] Step 3:

[0918] The server preprocesses the data stored in the database. Specifically, it uses scikit-learn and pandas to normalize the data and impute missing data. It uses raw data from the database as input and obtains preprocessed data as output. This puts the data in a format suitable for training AI models.

[0919] Step 4:

[0920] The server uses the preprocessed data to train a generative AI model, specifically using TensorFlow or PyTorch. It uses the preprocessed data and labeled data as input and obtains a trained AI model as output. This AI model is trained to be used to predict the probability of disaster occurrence.

[0921] Step 5:

[0922] The server inputs new natural disaster data into the generative AI model to predict the probability of disaster occurrence. It uses real-time data as input and obtains the predicted probability of disaster occurrence as output. To do this, the AI ​​model performs predictive calculations to calculate the probability of the next possible disaster.

[0923] Step 6:

[0924] The server sends the predicted disaster probability to a map service API, which plots disaster risk information for each region on a map. It uses disaster probability data as input and obtains a visual plot on the map service as output. It places marks on the map using APIs such as Google Maps to display the risk level.

[0925] Step 7:

[0926] The server retrieves evacuation shelter information from local government and government agency databases and stores it in a database. It uses the endpoint URL of the external database as input and stores the retrieved evacuation shelter information as output in an internal database. This includes information such as the latest evacuation shelter locations and capacity.

[0927] Step 8:

[0928] The emotion engine collects emotion data from the camera and microphone when the user browses disaster information. It uses the user's facial expression and voice data as input and obtains emotion analysis results as output. This is done using facial expression recognition software (e.g., Amazon Rekognition) and voice analysis software (e.g., Google Cloud Speech-to-Text).

[0929] Step 9:

[0930] Based on the analysis results, the emotion engine generates and displays a reassuring message if the user feels anxious or tense. It uses the emotion analysis results as input and gets a customized reassuring message as output, which helps the user relax and take appropriate action.

[0931] Step 10:

[0932] The device displays disaster risk information and evacuation shelter information to the user. It uses disaster risk data and evacuation shelter information obtained from the server as input and displays them on a map service or information screen as output. Information is provided to the user in real time via a web browser or mobile app.

[0933] (Application example 2)

[0934] 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."

[0935] In recent years, the frequency of natural disasters has increased, making countermeasures an urgent need. Meanwhile, with the widespread adoption of autonomous vehicles, it is necessary to ensure the safe operation of vehicles and the peace of mind of users during disasters. However, conventional systems have struggled to efficiently collect and analyze disaster information in real time and provide specific disaster prevention measures. Furthermore, they have been unable to provide customized messages based on the user's emotional state, making it difficult to encourage appropriate responses when users feel anxious.

[0936] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0937] In this invention, the server includes means for collecting natural disaster data in real time, means for training a generative AI model using the collected natural disaster data, means for predicting the probability of disaster occurrence using the generative AI model, means for reflecting the predicted probability of disaster occurrence in a map service, means for providing evacuation shelter information and disaster preparation information, means for recognizing a user's emotional state and providing a customized message based thereon, and means for optimizing a safe driving route in the event of a natural disaster. This allows a user to obtain the latest disaster information in real time, ensure a safe driving route, and receive a customized message according to their emotional state, enabling them to use autonomous vehicles with peace of mind.

[0938] "Real-time" means instantly processing and reflecting events and information occurring at that time.

[0939] "Natural disaster data" refers to data that includes information on natural disasters such as earthquakes, tsunamis, and fires.

[0940] A "generative AI model" is a type of artificial intelligence that uses machine learning algorithms to learn from past data and make predictions and classifications based on new data.

[0941] "Probability of disaster occurrence" is an index that indicates the degree of possibility of a disaster occurring in a specific region or under specific conditions.

[0942] A "map service" is a service that provides visual geographic information and displays location and route information.

[0943] "Evacuation shelter information" is information about safe places to evacuate to in the event of a disaster.

[0944] "Disaster preparedness information" refers to information on specific methods of response and items to prepare in the event of a disaster.

[0945] "Emotional state" refers to the emotions and mental state that a user is feeling at that time.

[0946] A "customized message" is information or a message that is optimized to a user's individual needs and emotional state.

[0947] A "safe route during a natural disaster" is a route that allows safe travel while avoiding danger even in the event of a natural disaster.

[0948] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

[0949] System Overview

[0950] This system, which consists of a server, terminals, and users, collects natural disaster information in real time and provides users with appropriate countermeasures.The system also uses a generative AI model to predict the probability of disaster occurrence and provides customized messages according to the user's emotional state, supporting the safe operation of autonomous vehicles.

[0951] Server Roles

[0952] The server collects natural disaster data in real time. Specifically, it periodically sends requests to APIs provided by data providers for earthquakes, fires, tsunamis, etc. to obtain new data. The obtained data is stored in a database and organized appropriately.

[0953] The server then uses a generative AI model to train this data and predict the probability of a disaster occurring. During this training process, data preprocessing is performed, including normalization and missing data completion. The prediction results are then stored in the database again and sent to the map service's API. This allows the disaster risk for each region to be visualized on a map. Marks on the map use different colors and symbols to indicate the level of disaster risk.

[0954] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies and updates it in real time, which is also sent to the map service for users to view.

[0955] The role of the emotional engine

[0956] The emotion engine recognizes the user's emotional state and provides customized messages based on that. When a user browses disaster information, emotional data is collected and analyzed via the camera and microphone. Based on the analysis results, if the user is feeling anxious or tense, a reassuring message is displayed.

[0957] Device Role

[0958] The device is the interface through which users use the map service. When a user accesses the map service through a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. Furthermore, information corresponding to the user's emotional state is also displayed based on the analysis results of the emotion engine.

[0959] Disaster information and occurrence probability related to the area searched by the user are displayed on a map. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided. Marks on the map can be clicked to check details, and route guidance to the evacuation shelter is also provided.

[0960] User Roles

[0961] Users can use this system to check disaster risks and evacuation shelter information for their area. They can also receive customized information based on their emotional state through the emotion engine, which allows them to make appropriate evacuation preparations and take appropriate evacuation actions.

[0962] Specific examples

[0963] For example, if a prediction is made that there is a high probability of a major earthquake occurring in Tokyo, the server will train the generative AI model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and send this prediction result to the map service API, which will display a red warning mark in the Tokyo area.

[0964] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice via the camera and microphone to recognize their current emotional state. If the user is feeling anxious or nervous, it displays a reassuring message such as, "Please remain calm. There is a shelter nearby. We will show you the route to the nearest shelter."

[0965] Users access the map service and check the warning symbols for the Tokyo area. Clicking on the symbol displays information about the nearest evacuation shelter. At the same time, specific information on preparations, such as a list of items needed in an emergency bag and how to check on the safety of family members, is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[0966] Example prompt sentence:

[0967] If an earthquake is predicted to occur in the Tokyo area, the system will display routes to the nearest evacuation shelter, check the user's emotional state, and display reassuring messages if they feel anxious. Meanwhile, the system will safely optimize the routes of autonomous vehicles.

[0968] As described above, this system obtains disaster information in real time and provides customized messages according to the user's emotional state, thereby creating an environment in which users can use self-driving vehicles with peace of mind.

[0969] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0970] Step 1:

[0971] The server collects natural disaster data in real time from an external API. It sends API requests and stores the acquired data in a database. The input is the natural disaster data acquired from the API, and the output is the organized data stored in the database. The API requests are programmed to be made periodically.

[0972] Step 2:

[0973] The server trains the generative AI model using natural disaster data stored in a database. Data preprocessing involves normalization and missing data completion. The input is past disaster data obtained from the database, and the output is the trained generative AI model. This makes it possible to predict the probability of disaster occurrence with high accuracy.

[0974] Step 3:

[0975] The server uses the trained generative AI model to input new data in real time and predict the probability of a disaster. The input is the latest natural disaster data, and the output is data indicating the probability of a disaster occurring. The results are then stored in a database and sent to the map service's API.

[0976] Step 4:

[0977] The server visualizes the disaster risk for each region on a map based on the data sent to the map service's API. Marks on the map use different colors and symbols to indicate the level of disaster risk. The input is disaster probability data, and the output is visually plotted map data.

[0978] Step 5:

[0979] The server retrieves the latest evacuation shelter information from local government and government agency databases and stores it in the database. The input is the evacuation shelter information retrieved from the external database, and the output is the updated evacuation shelter data. This information is also sent to the map service so that users can view it.

[0980] Step 6:

[0981] The device accesses the map service through a web browser or mobile app to obtain disaster risk information for the user's current location or a specified area. The input is the user's location information and data from the map service API, and the output is a disaster risk map displayed on the device.

[0982] Step 7:

[0983] The emotion engine uses a camera and microphone to recognize the user's emotional state. It analyzes facial expressions and voice to generate a customized message if the user is feeling anxious or nervous. The input is the user's image and voice data, and the output is the emotion analysis result and a reassuring message based on that.

[0984] Step 8:

[0985] The device displays a customized message based on the user's emotional state based on the analysis results of the emotion engine. The input is the emotion analysis result, and the output is the message displayed on the device. This allows users to use the system with peace of mind.

[0986] Step 9:

[0987] Users can check evacuation shelter information and disaster preparedness information through their devices. The input is evacuation shelter information obtained from a map service, and the output is evacuation information and preparedness information displayed on the device. Based on this, users can take specific evacuation actions.

[0988] In this way, the entire system works seamlessly, allowing users to obtain the latest disaster and evacuation information in real time. Furthermore, by providing customized messages according to the user's emotional state, users can feel at ease and take appropriate action.

[0989] 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.

[0990] 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.

[0991] 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.

[0992] [Fourth embodiment]

[0993] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0994] 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.

[0995] 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).

[0996] 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.

[0997] 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.

[0998] 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).

[0999] 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.

[1000] 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.

[1001] 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.

[1002] 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.

[1003] 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.

[1004] 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.

[1005] 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."

[1006] This section describes an embodiment of the present invention. This system is composed of a server, a terminal, and a user component, and by these components working together, it is possible to collect, analyze, and predict natural disaster data in real time and visualize it through a map service.

[1007] Server Roles

[1008] The server first collects various natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. The collected data is then stored in a database and organized in an appropriate format.

[1009] The server then uses a generative AI model to train this data. The model is trained using past disaster data so that it can predict the probability of disaster occurrence with high accuracy. This training process requires preprocessing the data, such as normalization and missing data completion.

[1010] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. These prediction results are then stored in the database again for use in the map service.

[1011] The server then sends the predicted disaster probability data to the map service's API, which plots disaster risk information for each region on a map. The marks on the map use different colors and symbols to visualize the level of disaster risk.

[1012] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[1013] Device Role

[1014] The device is the interface through which users access the map service. When a user accesses the map service via a web browser or mobile app, the device displays the latest disaster risk information and evacuation shelter information.

[1015] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[1016] Marks on the map displayed on the device can be clicked to view details, and route guidance to evacuation shelters is also provided to support specific evacuation actions.

[1017] User Roles

[1018] Users can use the map service to check disaster risks and evacuation shelter information for their area, and if necessary, prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[1019] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[1020] Specific examples

[1021] As a specific example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[1022] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[1023] Users access the map service and check the warning symbols for the Tokyo area. When users click on the symbol, information about the nearest evacuation shelter is displayed, retrieved from the server. At the same time, specific preparation information is also provided, such as a list of items needed in an emergency evacuation bag and how to check on the safety of family members.

[1024] This will allow users to prepare for appropriate evacuation actions and evacuate quickly and safely in the event of an earthquake. This system is a powerful tool for promoting quick and appropriate responses before a disaster occurs.

[1025] The processing flow will be explained below.

[1026] Step 1:

[1027] Server: Periodically sends requests to APIs for various natural disaster data (earthquakes, fires, tsunamis, etc.) to collect new data. The collected data is formatted and stored in a database.

[1028] Step 2:

[1029] Server: Extracts past natural disaster data from the database and preprocesses it as a training dataset for the generative AI model. Preprocessing includes normalizing the data and filling in missing data.

[1030] Step 3:

[1031] Server: The preprocessed dataset is used to train a generative AI model, for example, by training the model using a machine learning framework (such as TensorFlow or PyTorch).

[1032] Step 4:

[1033] Server: Using the generative AI model that has completed training, new disaster data is input in real time to predict the probability of disaster occurrence. The prediction results are stored in a database.

[1034] Step 5:

[1035] Server: Connects to the Yahoo Maps API and reflects predicted disaster probability data on the map. Specifically, areas with high disaster risk are marked with warning marks or color-coded.

[1036] Step 6:

[1037] Device: When a user accesses Yahoo! Maps using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on a map.

[1038] Step 7:

[1039] Device: When a user clicks on a warning mark on the map, evacuation shelter information and disaster preparedness advice retrieved from the server are displayed, such as the location of the nearest evacuation shelter and a list of emergency supplies to take with you.

[1040] Step 8:

[1041] Server: Collects the latest evacuation shelter information from databases of local governments and government agencies and stores it in a database. This information is also reflected in the map service and made available to users.

[1042] Step 9:

[1043] User: Make necessary preparations based on the displayed disaster risk information and evacuation shelter information. For example, prepare an emergency kit and check evacuation routes with family members.

[1044] Step 10:

[1045] Server: Collects various feedback and log data to help improve our services. Analyzes user access history and click data to identify new features and service improvements.

[1046] Example 1

[1047] 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."

[1048] In modern society, the frequency and scale of damage caused by natural disasters continue to increase. Conventional disaster information systems have difficulty collecting real-time data and making highly accurate predictions, which means users are unable to obtain the information they need to take prompt and appropriate action. In addition, updates to evacuation shelter information and disaster preparation information are often delayed, preventing users from acting based on the latest information. Therefore, there is a need to build a system that can quickly and accurately predict disaster risks and provide users with useful information.

[1049] 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.

[1050] In this invention, the server includes means for collecting natural disaster data in real time, means for storing the collected natural disaster data in a database and organizing it while maintaining consistency, means for performing data preprocessing using the collected natural disaster data, means for training a generative AI model using the preprocessed data, means for predicting disaster occurrence probabilities in real time using the generative AI model, means for re-storing the predicted disaster occurrence probabilities in the database, means for reflecting the predicted disaster occurrence probabilities in a map service, means for collecting evacuation shelter information from local governments and government agencies and storing it in the database, means for providing evacuation shelter information and disaster preparation information to users, and terminal interface means for users to check the evacuation shelter information and disaster occurrence probability information. This allows users to quickly and accurately obtain predicted disaster risk information and take appropriate evacuation actions and preparations.

[1051] A "server" is a computer system that has the ability to provide various services and data to other computers over a network.

[1052] "Natural disaster data" refers to data that includes information on natural phenomena such as earthquakes, tsunamis, and fires.

[1053] A "database" is a system for efficiently storing, retrieving, and updating collected and organized data.

[1054] "Data preprocessing" is the process of converting data into a format suitable for analysis and learning models, and completing and normalizing missing values.

[1055] A "generative AI model" is an artificial intelligence model that learns from collected data and makes predictions about new data.

[1056] "Disaster occurrence probability" is a predicted value that indicates the probability that a specific natural disaster will occur within a certain period of time.

[1057] A "map service" is a web service or application that visually displays geographic information and various data.

[1058] "Evacuation shelter information" is information about places where people can evacuate in the event of a disaster.

[1059] A "terminal interface" is an operation screen for a device or application that allows a user to access a system and obtain information.

[1060] "Local governments and government agencies" refers to local government and national public institutions, which are organizations that provide various public services and information.

[1061] The following describes in detail an embodiment of the present invention. This system is composed of a server, a terminal, and a user component, and by these components working together, it is possible to collect, analyze, and predict natural disaster data in real time, and visualize it through a map service.

[1062] Server Roles

[1063] The server first collects natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. For example, it uses the Japan Meteorological Agency's API to obtain earthquake information. Requests are sent at regular intervals, and JSON-formatted data is returned as a response.

[1064] The server then stores the collected data in a database, which can be structured for consistency and organized by time and location, for example, using MongoDB.

[1065] The server then performs data preprocessing, including normalizing the data and imputing missing values. Normalization aligns the scale of the data, allowing for efficient training of generative AI models. For example, Pandas is used to normalize the "magnitude" scale from 0 to 1.

[1066] The server then uses the data to train a generative AI model. To train the model using data from past disasters, the server uses a machine learning framework such as TensorFlow to build a long short-term memory (LSTM) network and train the model on earthquake data.

[1067] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. The prediction results are then stored in the database again, ready for use with the map service.

[1068] The server then sends the prediction results to the API of the map service, which plots disaster risk information for each region on a map, with high-risk areas displayed in red and low-risk areas in green.

[1069] In addition, the server collects evacuation shelter information from databases of local governments and government agencies, stores the latest evacuation shelter information in the database, and sends this information to the map service so that users can check it.

[1070] Device Role

[1071] The device acts as an interface for using the map service. When a user accesses the map service via a web browser or mobile app, the device displays the latest disaster risk information and evacuation shelter information.

[1072] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[1073] The marks displayed on the device's map can be clicked to view detailed information. Route guidance to evacuation shelters is also provided, supporting specific evacuation actions.

[1074] User Roles

[1075] Users can use the map service to check disaster risk information and evacuation shelter information for their area. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[1076] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake occurring, they can check to see if they have prepared an emergency kit and how to contact their family members. They can also check the route to the nearest evacuation shelter in advance, allowing them to act quickly and calmly in the event of an emergency.

[1077] Prompt Sentence Examples

[1078] Example prompts to input to a generative AI model:

[1079] "Using the past earthquake data below, please predict the probability of an earthquake occurring within the next 24 hours. This data is limited to the Tokyo area."

[1080] This system is a powerful tool for promoting rapid and appropriate responses before a disaster occurs, and by each component working together, users can obtain the information they need in a timely manner and act safely.

[1081] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1082] Step 1:

[1083] The server collects natural disaster data. Specifically, it periodically sends requests to an API that provides data on earthquakes, fires, tsunamis, etc. In this process, the input is the API endpoint URL, and the output is natural disaster data in JSON format. For example, it sends a request to "https: / / api.kishou.go.jp / earthquake" and receives the returned data.

[1084] Step 2:

[1085] The server stores the collected natural disaster data in a database. The input is the collected data in JSON format, and the output is the organized data stored in the database. Specifically, it uses MongoDB to insert earthquake data into the "earthquake_data" collection, which contains fields such as "timestamp," "location," and "magnitude."

[1086] Step 3:

[1087] The server performs data preprocessing. The input is natural disaster data retrieved from the database, and the output is the preprocessed data. Specifically, it uses Pandas to normalize the data and convert the values ​​of the "magnitude" field into the range of 0 to 1. It also imputes missing values ​​with appropriate values ​​if any.

[1088] Step 4:

[1089] The server trains a generative AI model using the preprocessed data. The input is the preprocessed data, and the output is a trained generative AI model. Using TensorFlow, an LSTM network is constructed and the model is trained using past earthquake data. Specifically, the LSTM model is trained using earthquake data from the past five years.

[1090] Step 5:

[1091] The server uses a trained generative AI model to predict the probability of a disaster. The input is newly collected natural disaster data, and the output is the predicted probability of a disaster. New earthquake data is input into the model, and the probability of an earthquake occurring within the next 24 hours is calculated.

[1092] Step 6:

[1093] The server stores the prediction results in the database again. The input is the predicted probability of a disaster occurring, and the output is the prediction result stored in the database. Specifically, the server inserts the predicted earthquake occurrence probability into the "predictions" collection.

[1094] Step 7:

[1095] The server sends the prediction results to the map service's API. The input is the prediction results obtained from the database, and the output is the disaster risk information reflected in the map service. Using the Google Maps API, disaster risk data for each region is plotted on a map. Areas with high risk are displayed in red, and areas with low risk are displayed in green.

[1096] Step 8:

[1097] The terminal is an interface for using map services. The input is the user's search query, and the output is disaster risk information and evacuation shelter information displayed on the map. Specifically, when a user searches for a specific area, disaster information and the probability of disaster occurrence related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[1098] Step 9:

[1099] Users use map services to check disaster risk information and evacuation shelter information for the area in which they live. The input is various operations performed by the user (for example, searching for an area or clicking a mark), and the output is related disaster information and evacuation shelter information. Specifically, when a user clicks on a warning mark on the map, information on the nearest evacuation shelter and preparation information such as a "list of items necessary for an emergency evacuation bag" and "how to check on the safety of family members" is displayed.

[1100] (Application example 1)

[1101] 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."

[1102] In recent years, the increasing number of natural disasters has created a need for real-time collection, analysis, and prediction of disaster information. However, existing systems do not provide sufficient, fast, and accurate information when a disaster occurs, which often delays evacuation and preparation. Furthermore, the provision of specific evacuation instructions and disaster prevention guidelines to users is insufficient, making it difficult to respond to disasters effectively.

[1103] 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.

[1104] In this invention, the server includes means for collecting natural disaster data in real time, means for training a generative AI model using the collected natural disaster data, means for predicting the probability of a disaster occurring using the generative AI model, means for reflecting the predicted probability of a disaster occurring in a map service, means for providing evacuation shelter information and disaster preparation information, means for providing disaster notifications in real time on the terminal, means for displaying route guidance to evacuation shelters on the terminal, and means for presenting a disaster action guide and an emergency kit list on the terminal. This allows users to obtain quick and accurate disaster information and provides specific evacuation instructions and disaster prevention guidelines, enabling appropriate disaster response.

[1105] "Means for collecting natural disaster data in real time" refers to technological means for instantly obtaining information on natural disasters such as earthquakes, fires, and tsunamis.

[1106] "Means for training a generative AI model using collected natural disaster data" refers to technical means for training a generative AI model based on acquired natural disaster data.

[1107] "Means for predicting the probability of disaster occurrence using a generative AI model" refers to a technical means for utilizing a trained generative AI model to quantify and predict the possibility of a disaster occurring.

[1108] "Means for reflecting the predicted probability of disaster occurrence on the map service" refers to the technical means for displaying the probability of disaster occurrence predicted by the generative AI model on the map service.

[1109] "Means for providing information on evacuation shelters and disaster preparation information" refers to technical means for providing users with information on the locations of evacuation shelters and the equipment needed in the event of a disaster.

[1110] "Means for providing real-time disaster notifications on devices" refers to technical means for sending users immediate notifications about disaster occurrences through devices such as smartphones or smart glasses.

[1111] "Means for displaying route guidance to a shelter on a device" refers to a technical means for displaying specific directions for a user to safely reach a shelter on a device such as a smartphone or smart glasses.

[1112] "Means for presenting disaster action guidelines and emergency bag lists on devices" refers to technological means for displaying specific actions to be taken and lists of necessary items to bring in the event of a disaster on devices such as smartphones or smart glasses.

[1113] An embodiment of the present invention will now be described. This system is comprised of a server, a terminal, and a user component, which work together to enable a rapid and appropriate response to natural disasters.

[1114] Server Roles

[1115] The server first collects natural disaster data in real time. This data is obtained through APIs related to disasters such as earthquakes, fires, and tsunamis. The collected data is then stored in a database and organized in an appropriate format. For example, this process involves using the Python requests library to obtain data from external APIs and storing the data in MySQL or PostgreSQL.

[1116] The server then uses this data to train a generative AI model. The model is trained using past disaster data to accurately predict the probability of a disaster. This training process involves preprocessing, such as normalizing the data and filling in missing data.

[1117] Once trained, the generative AI model predicts the probability of a disaster occurring in real time whenever new data is input. The prediction results are then stored in a database and sent to the map service's API, which visualizes disaster risk on a map.

[1118] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest evacuation shelter information in the database, and sends this information to the map service API so that users can check it.

[1119] Device Role

[1120] The device is the interface through which users use the map service. When users access the map service using a web browser or mobile app, the latest disaster risk information and evacuation shelter information are displayed. For example, when a user searches for their local area, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information such as the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies is also provided.

[1121] The map displayed on the device can be clicked to view details. Route guidance to evacuation shelters is also provided, supporting specific evacuation actions. Real-time notifications are sent via smartphones, smart glasses, and other devices in the event of a disaster such as an earthquake or fire.

[1122] User Roles

[1123] Users can use the map service to check disaster risks and evacuation shelter information in their area. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors. For example, if a user checks their area on the map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. In addition, by checking the route to the nearest evacuation shelter in advance, they can act quickly and safely in the event of an emergency.

[1124] Specific examples

[1125] As a specific example, when a user in Tokyo receives a notification informing them of an increased earthquake risk, the message "Earthquake risk is increasing in the Tokyo area. Please check for the latest information" is displayed on the user's smartphone. Route guidance to an evacuation shelter, a list of emergency kits, and a prompt such as "List of items necessary for an emergency kit" are also provided.

[1126] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1127] Step 1:

[1128] The server collects natural disaster data in real time from an external API. The server periodically sends requests to the API endpoint to get the latest disaster information such as earthquakes, fires, tsunamis, etc. In this step, the input is the data obtained from the API, and the output is the raw natural disaster data.

[1129] Step 2:

[1130] The server stores the collected natural disaster data in a database. The server performs preprocessing to standardize the data format and fill in missing data. This inputs normalized natural disaster data into the database. The input is raw natural disaster data, and the output is organized database entries.

[1131] Step 3:

[1132] The server trains the generative AI model using the compiled natural disaster data. The server supplies past disaster data to the generative AI model and trains the disaster occurrence prediction algorithm. The input is past disaster data, and the output is the trained generative AI model.

[1133] Step 4:

[1134] The server inputs new disaster data into the AI ​​model in real time to predict the probability of a disaster occurring. The model evaluates disaster risk based on the latest data and generates the results. The input is the latest disaster data, and the output is a predicted disaster probability.

[1135] Step 5:

[1136] The server reflects the predicted disaster occurrence probability on the map service. The server sends this information to the map service API, which provides data for visualizing disaster risk on a map. The input is the prediction result, and the output is disaster risk information plotted on the map service.

[1137] Step 6:

[1138] The server collects evacuation shelter information from databases of local governments and government agencies and stores it in a database. The server retrieves the latest evacuation shelter information and prepares it for users to access. The input is the latest information on evacuation shelters, and the output is a compiled evacuation shelter database entry.

[1139] Step 7:

[1140] When a user accesses the map service, the device displays the latest disaster risk information and evacuation shelter information. When a user searches for their area, disaster data and evacuation shelter information for that area are displayed on the screen. The input is the user's search query, and the output is disaster information and evacuation shelter information displayed on the device's display.

[1141] Step 8:

[1142] The device provides users with real-time disaster notifications. The server pushes disaster information to the device so that users can check it immediately. The input is newly collected disaster data, and the output is notifications on the device.

[1143] Step 9:

[1144] The device displays route guidance to the nearest evacuation shelter. When the user clicks on the evacuation shelter marking, the device displays the route using GPS and map service APIs. The input is the user's current location and the location information of the evacuation shelter, and the output is the route guidance displayed on the device.

[1145] Step 10:

[1146] The device presents a list of action guidelines and emergency kits for emergencies. When the user checks the disaster information, a specific action plan and a list of necessary items are displayed. The input is predicted disaster risk information, and the output is the display of the action guidelines and list.

[1147] 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.

[1148] An embodiment of the present invention will now be described. The system of the present invention is composed of a server, a terminal, and a user component, and by combining it with an emotion engine that recognizes the user's emotions, it is possible to collect, analyze, and predict natural disaster data in real time and visualize it through a map service. This system not only provides disaster information and evacuation shelter information, but also provides customized information and messages according to the user's emotional state.

[1149] Server Roles

[1150] The server first collects various natural disaster data in real time. Specifically, it periodically sends requests to APIs that provide data on earthquakes, fires, tsunamis, etc. to obtain new data. The collected data is then stored in a database and organized in an appropriate format.

[1151] The server then uses a generative AI model to train this data. The model is trained using past disaster data so that it can predict the probability of disaster occurrence with high accuracy. This training process requires preprocessing the data, such as normalization and missing data completion.

[1152] New data is input in real time using the generative AI model to predict the probability of a disaster occurring. The prediction results are then stored in a database for use in the map service.

[1153] The server then sends the predicted disaster probability data to the map service's API, which plots disaster risk information for each region on a map. Marks on the map use different colors and symbols to visualize the level of disaster risk.

[1154] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[1155] The role of the emotional engine

[1156] The emotion engine recognizes the user's emotional state and provides customized messages and advice based on that information. When the user browses disaster information, the emotion engine collects and analyzes emotional data via the camera and microphone. Using the results of this analysis, if the user is feeling anxious or tense, it displays a reassuring message.

[1157] Device Role

[1158] The device is the interface through which users use the map service. When a user accesses the map service using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. Furthermore, information based on the user's emotional state is also displayed based on the analysis results of the emotion engine.

[1159] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided.

[1160] Marks on the map displayed on the device can be clicked to view details, and route guidance to evacuation shelters is also provided to support specific evacuation actions.

[1161] User Roles

[1162] Users can use the map service to check disaster risks and evacuation shelter information for their area. They can also receive information customized based on their emotional state using the emotion engine. If necessary, they can prepare for evacuation and take appropriate measures in cooperation with their family and neighbors.

[1163] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[1164] Specific examples

[1165] As a specific example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[1166] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[1167] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice through the camera and microphone to recognize their current emotional state. If the user feels anxious or nervous, it displays a reassuring message such as, "Please remain calm. There is a shelter nearby. We will show you the route to the nearest shelter."

[1168] Users access the map service and check the warning mark for the Tokyo area. When the user clicks on the warning mark, information about the nearest evacuation shelter retrieved from the server is displayed. At the same time, specific preparation information such as a list of items needed in an emergency evacuation bag and how to check on the safety of family members is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[1169] In this way, by combining the emotion engine, support tailored to the user's emotional state can be provided, enabling more effective disaster response. This system is a powerful tool for encouraging prompt and appropriate responses before a disaster occurs.

[1170] The processing flow will be explained below.

[1171] Step 1:

[1172] Server: Periodically sends requests to APIs for various natural disaster data (earthquakes, fires, tsunamis, etc.) to collect new data. The collected data is formatted and stored in a database.

[1173] Step 2:

[1174] Server: Extracts past natural disaster data from the database and preprocesses it as a training dataset for the generative AI model. Preprocessing includes normalizing the data and filling in missing data.

[1175] Step 3:

[1176] Server: The preprocessed dataset is used to train a generative AI model, for example, by training the model using a machine learning framework (such as TensorFlow or PyTorch).

[1177] Step 4:

[1178] Server: Using the generative AI model that has completed training, new disaster data is input in real time to predict the probability of disaster occurrence. The prediction results are stored in a database.

[1179] Step 5:

[1180] Server: Connects to the Yahoo Maps API and reflects predicted disaster probability data on the map. Specifically, areas with high disaster risk are marked with warning marks or color-coded.

[1181] Step 6:

[1182] Device: When a user accesses Yahoo! Maps using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on a map.

[1183] Step 7:

[1184] Device: When a user clicks on a warning mark on the map, evacuation shelter information and disaster preparedness advice retrieved from the server are displayed, such as the location of the nearest evacuation shelter and a list of emergency supplies to take with you.

[1185] Step 8:

[1186] Server: Collects the latest evacuation shelter information from databases of local governments and government agencies and stores it in a database. This information is also reflected in the map service and made available to users.

[1187] Step 9:

[1188] Emotion engine: When a user uses the map service, the system collects emotional data from the user via the camera and microphone. For example, it uses facial recognition and voice analysis to assess whether the user is feeling anxious or nervous.

[1189] Step 10:

[1190] Emotion engine: Analyzes collected emotional data and generates customized messages and advice based on the user's emotional state. For example, if the user is feeling anxious, the engine generates a message such as, "Please remain calm. Information about the nearest evacuation shelter will be displayed."

[1191] Step 11:

[1192] Terminal: Displays customized messages generated by the emotion engine to the user, providing messages that reassure the user and specific guidelines for action.

[1193] Step 12:

[1194] User: Make necessary preparations based on the displayed disaster risk information and evacuation shelter information. For example, prepare an emergency kit and check evacuation routes with family members.

[1195] Step 13:

[1196] Server: Collects various feedback and log data to help improve our services. Analyzes user access history and click data to identify new features and service improvements.

[1197] Example 2

[1198] 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."

[1199] Natural disasters have a significant impact on human lives and property, so providing fast and accurate information is essential. However, existing disaster information systems lack the ability to collect and analyze information in real time, and they do not provide customized information based on the user's emotional state. This makes it difficult for users to take appropriate action, and anxiety and confusion are likely to arise.

[1200] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1201] In this invention, the server includes means for collecting natural disaster data in real time, means for storing the collected natural disaster data in a database, means for preprocessing the data and training a generative AI model, means for predicting the probability of a disaster occurring using the generative AI model, means for reflecting the predicted probability of a disaster occurring in a map service, means for providing evacuation shelter information and disaster preparation information, and means for recognizing a user's emotional state and displaying a customized message according to that state. This allows users to receive accurate natural disaster information in real time, and also to receive appropriate guidelines for action and reassurance messages according to their emotional state.

[1202] "Means for collecting natural disaster data in real time" refers to a system or device for obtaining information on natural disasters such as earthquakes, fires, and tsunamis in a timely manner from external data sources.

[1203] "Means for storing collected natural disaster data in a database" refers to a database system and related technologies for efficiently managing and storing acquired natural disaster data and making it accessible later.

[1204] "Data preprocessing and means for training generative AI models" refers to techniques for properly preparing data and converting it into the format required for an AI model to learn, and then using AI techniques to train the model.

[1205] "Means for predicting the probability of disaster occurrence using a generative AI model" refers to technologies and systems that use a generative AI model to calculate the likelihood of future disaster occurrence based on collected natural disaster data.

[1206] "Means for reflecting the predicted probability of disaster occurrence in map services" refers to technology and systems that display the probability information of disaster occurrence predicted by an AI model on a visual map, making it easier for users to understand visually.

[1207] "Means for providing evacuation shelter information and disaster preparation information" refers to technologies and systems that provide users with specific information about where to evacuate and necessary preparations when a disaster occurs, before, during, or after the disaster.

[1208] "Means for recognizing the user's emotional state and displaying a customized message according to that state" refers to technology that analyzes the user's facial and voice data to determine their emotional state, and then displays individual reassuring messages or advice based on the results.

[1209] MODE FOR CARRYING OUT THE INVENTION

[1210] The system of the present invention is composed of server, terminal, and user components, and by combining it with an emotion engine, collects, analyzes, and predicts natural disaster data in real time, and visualizes it through a map service. A specific embodiment of this system is described in detail below.

[1211] Server Roles

[1212] The server first collects real-time data on natural disasters such as earthquakes, fires, and tsunamis. Specifically, the server periodically sends requests to external APIs (e.g., earthquake information APIs, weather information APIs) to obtain the latest disaster data. To do this, it uses a program library such as Python's requests library. The obtained data is received in JSON format or similar and stored in a database (e.g., MySQL, MongoDB).

[1213] The server then trains the generative AI model using this data. Data preprocessing involves normalizing the data and imputing missing data. This is done using data analysis libraries such as scikit-learn and pandas. The preprocessed data is then used to train the generative AI model using an AI framework such as TensorFlow or PyTorch.

[1214] Using a trained generative AI model, new data is input in real time to predict the probability of a disaster occurring. The prediction results are then stored in a database and sent to the map service's API (e.g., Google Maps API). Disaster risk information for each region is then plotted on a map. Marks on the map use different colors and symbols to visualize the level of disaster risk.

[1215] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies, stores the latest information in the database, and sends this information to the map service so that users can check it.

[1216] The role of the emotional engine

[1217] The emotion engine recognizes the user's emotional state and provides customized messages and advice based on that information. When the user browses disaster information, emotional data is collected via the camera and microphone and analyzed. This analysis uses facial expression recognition technology (e.g., Amazon Rekognition) and voice analysis technology (e.g., Google Cloud Speech-to-Text). Using the analysis results, if the user is feeling anxious or tense, a reassuring message is displayed.

[1218] Device Role

[1219] The device functions as an interface for users to use the map service. When a user accesses the map service using a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. The device displays the map and information using web technologies such as HTML, CSS, and JavaScript. It also displays information according to the user's emotional state based on the analysis results of an emotion engine.

[1220] For example, when a user searches for the area where they live, disaster information and the probability of a disaster occurring related to that area are displayed. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided. Marks on the map can be clicked to view details. Route guidance to the evacuation shelter is also provided to support specific evacuation actions.

[1221] User Roles

[1222] Users can use the map service to check disaster risks and evacuation shelter information for their area. They can also receive information customized based on their emotional state using the emotion engine. This allows them to prepare for evacuation as needed and take appropriate measures in cooperation with their family and neighbors.

[1223] For example, if a user checks their area on a map service and finds that there is a high probability of an earthquake, they can check to see if they have prepared an emergency kit and how to contact their family. Also, by checking the route to the nearest evacuation shelter in advance, they can act quickly and calmly in the event of an emergency.

[1224] Specific examples

[1225] For example, consider a case where it is predicted that there is a high possibility of a large earthquake occurring in Tokyo.

[1226] The server trains the AI ​​model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and sends the prediction results to the map service API, which displays a red warning mark in the Tokyo area.

[1227] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice via the camera and microphone to recognize their current emotional state. If the user feels anxious or nervous, it displays reassuring messages such as "Please stay calm. There is a shelter nearby" or "We will show you the route to the nearest shelter."

[1228] Users access the map service and check the warning symbols for the Tokyo area. When the user clicks on the symbol, information about the nearest evacuation shelter is displayed, retrieved from the server. At the same time, specific information on preparations, such as a list of items needed in an emergency bag and how to check on the safety of family members, is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[1229] Prompt Sentence Examples

[1230] Here are some example prompts to input to a generative AI model:

[1231] "Predict the probability of an earthquake occurring in the next 12 hours."

[1232] "Get the latest shelter information and display it to the user."

[1233] "Analyze the user's emotional state and display appropriate messages."

[1234] As described above, the present invention not only collects, predicts, and displays disaster information, but also supports appropriate behavior in emergencies by providing customized messages that take into account the user's emotional state.

[1235] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1236] Step 1:

[1237] The server sends requests to an external natural disaster data API to retrieve natural disaster data in real time. Specifically, it uses the Python requests library to send a GET request to the API. It uses the API's endpoint URL as input and obtains the retrieved data in JSON format as output. This data includes disaster information such as earthquakes, fires, and tsunamis.

[1238] Step 2:

[1239] The server stores the acquired natural disaster data in a database. Specifically, it uses MySQL or MongoDB to store the data. The acquired JSON data is used as input, and disaster data is saved in the database as output. The database includes information such as the type of disaster, time of occurrence, location, and magnitude.

[1240] Step 3:

[1241] The server preprocesses the data stored in the database. Specifically, it uses scikit-learn and pandas to normalize the data and impute missing data. It uses raw data from the database as input and obtains preprocessed data as output. This puts the data in a format suitable for training AI models.

[1242] Step 4:

[1243] The server uses the preprocessed data to train a generative AI model, specifically using TensorFlow or PyTorch. It uses the preprocessed data and labeled data as input and obtains a trained AI model as output. This AI model is trained to be used to predict the probability of disaster occurrence.

[1244] Step 5:

[1245] The server inputs new natural disaster data into the generative AI model to predict the probability of disaster occurrence. It uses real-time data as input and obtains the predicted probability of disaster occurrence as output. To do this, the AI ​​model performs predictive calculations to calculate the probability of the next possible disaster.

[1246] Step 6:

[1247] The server sends the predicted disaster probability to a map service API, which plots disaster risk information for each region on a map. It uses disaster probability data as input and obtains a visual plot on the map service as output. It places marks on the map using APIs such as Google Maps to display the risk level.

[1248] Step 7:

[1249] The server retrieves evacuation shelter information from local government and government agency databases and stores it in a database. It uses the endpoint URL of the external database as input and stores the retrieved evacuation shelter information as output in an internal database. This includes information such as the latest evacuation shelter locations and capacity.

[1250] Step 8:

[1251] The emotion engine collects emotion data from the camera and microphone when the user browses disaster information. It uses the user's facial expression and voice data as input and obtains emotion analysis results as output. This is done using facial expression recognition software (e.g., Amazon Rekognition) and voice analysis software (e.g., Google Cloud Speech-to-Text).

[1252] Step 9:

[1253] Based on the analysis results, the emotion engine generates and displays a reassuring message if the user feels anxious or tense. It uses the emotion analysis results as input and gets a customized reassuring message as output, which helps the user relax and take appropriate action.

[1254] Step 10:

[1255] The device displays disaster risk information and evacuation shelter information to the user. It uses disaster risk data and evacuation shelter information obtained from the server as input and displays them on a map service or information screen as output. Information is provided to the user in real time via a web browser or mobile app.

[1256] (Application example 2)

[1257] 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."

[1258] In recent years, the frequency of natural disasters has increased, making countermeasures an urgent need. Meanwhile, with the widespread adoption of autonomous vehicles, it is necessary to ensure the safe operation of vehicles and the peace of mind of users during disasters. However, conventional systems have struggled to efficiently collect and analyze disaster information in real time and provide specific disaster prevention measures. Furthermore, they have been unable to provide customized messages based on the user's emotional state, making it difficult to encourage appropriate responses when users feel anxious.

[1259] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1260] In this invention, the server includes means for collecting natural disaster data in real time, means for training a generative AI model using the collected natural disaster data, means for predicting the probability of disaster occurrence using the generative AI model, means for reflecting the predicted probability of disaster occurrence in a map service, means for providing evacuation shelter information and disaster preparation information, means for recognizing a user's emotional state and providing a customized message based thereon, and means for optimizing a safe driving route in the event of a natural disaster. This allows a user to obtain the latest disaster information in real time, ensure a safe driving route, and receive a customized message according to their emotional state, enabling them to use autonomous vehicles with peace of mind.

[1261] "Real-time" means instantly processing and reflecting events and information occurring at that time.

[1262] "Natural disaster data" refers to data that includes information on natural disasters such as earthquakes, tsunamis, and fires.

[1263] A "generative AI model" is a type of artificial intelligence that uses machine learning algorithms to learn from past data and make predictions and classifications based on new data.

[1264] "Probability of disaster occurrence" is an index that indicates the degree of possibility of a disaster occurring in a specific region or under specific conditions.

[1265] A "map service" is a service that provides visual geographic information and displays location and route information.

[1266] "Evacuation shelter information" is information about safe places to evacuate to in the event of a disaster.

[1267] "Disaster preparedness information" refers to information on specific methods of response and items to prepare in the event of a disaster.

[1268] "Emotional state" refers to the emotions and mental state that a user is feeling at that time.

[1269] A "customized message" is information or a message that is optimized to a user's individual needs and emotional state.

[1270] A "safe route during a natural disaster" is a route that allows safe travel while avoiding danger even in the event of a natural disaster.

[1271] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

[1272] System Overview

[1273] This system, which consists of a server, terminals, and users, collects natural disaster information in real time and provides users with appropriate countermeasures.The system also uses a generative AI model to predict the probability of disaster occurrence and provides customized messages according to the user's emotional state, supporting the safe operation of autonomous vehicles.

[1274] Server Roles

[1275] The server collects natural disaster data in real time. Specifically, it periodically sends requests to APIs provided by data providers for earthquakes, fires, tsunamis, etc. to obtain new data. The obtained data is stored in a database and organized appropriately.

[1276] The server then uses a generative AI model to train this data and predict the probability of a disaster occurring. During this training process, data preprocessing is performed, including normalization and missing data completion. The prediction results are then stored in the database again and sent to the map service's API. This allows the disaster risk for each region to be visualized on a map. Marks on the map use different colors and symbols to indicate the level of disaster risk.

[1277] In addition, the server retrieves evacuation shelter information from databases of local governments and government agencies and updates it in real time, which is also sent to the map service for users to view.

[1278] The role of the emotional engine

[1279] The emotion engine recognizes the user's emotional state and provides customized messages based on that. When a user browses disaster information, emotional data is collected and analyzed via the camera and microphone. Based on the analysis results, if the user is feeling anxious or tense, a reassuring message is displayed.

[1280] Device Role

[1281] The device is the interface through which users use the map service. When a user accesses the map service through a web browser or mobile app, disaster risk information for the current location or a specified area is obtained and displayed on the map. Furthermore, information corresponding to the user's emotional state is also displayed based on the analysis results of the emotion engine.

[1282] Disaster information and occurrence probability related to the area searched by the user are displayed on a map. At the same time, information on the nearest evacuation shelter, guidelines for action in the event of a disaster, and a list of disaster prevention supplies are also provided. Marks on the map can be clicked to check details, and route guidance to the evacuation shelter is also provided.

[1283] User Roles

[1284] Users can use this system to check disaster risks and evacuation shelter information for their area. They can also receive customized information based on their emotional state through the emotion engine, which allows them to make appropriate evacuation preparations and take appropriate evacuation actions.

[1285] Specific examples

[1286] For example, if a prediction is made that there is a high probability of a major earthquake occurring in Tokyo, the server will train the generative AI model using past earthquake data for the Tokyo area, predicting a high probability of an earthquake occurring within the next 24 hours, and send this prediction result to the map service API, which will display a red warning mark in the Tokyo area.

[1287] When a user browses the map service, the emotion engine analyzes the user's facial expressions and voice via the camera and microphone to recognize their current emotional state. If the user is feeling anxious or nervous, it displays a reassuring message such as, "Please remain calm. There is a shelter nearby. We will show you the route to the nearest shelter."

[1288] Users access the map service and check the warning symbols for the Tokyo area. Clicking on the symbol displays information about the nearest evacuation shelter. At the same time, specific information on preparations, such as a list of items needed in an emergency bag and how to check on the safety of family members, is also provided. Reassuring messages from the emotion engine also allow users to respond calmly.

[1289] Example prompt sentence:

[1290] If an earthquake is predicted to occur in the Tokyo area, the system will display routes to the nearest evacuation shelter, check the user's emotional state, and display reassuring messages if they feel anxious. Meanwhile, the system will safely optimize the routes of autonomous vehicles.

[1291] As described above, this system obtains disaster information in real time and provides customized messages according to the user's emotional state, thereby creating an environment in which users can use self-driving vehicles with peace of mind.

[1292] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1293] Step 1:

[1294] The server collects natural disaster data in real time from an external API. It sends API requests and stores the acquired data in a database. The input is the natural disaster data acquired from the API, and the output is the organized data stored in the database. The API requests are programmed to be made periodically.

[1295] Step 2:

[1296] The server trains the generative AI model using natural disaster data stored in a database. Data preprocessing involves normalization and missing data completion. The input is past disaster data obtained from the database, and the output is the trained generative AI model. This makes it possible to predict the probability of disaster occurrence with high accuracy.

[1297] Step 3:

[1298] The server uses the trained generative AI model to input new data in real time and predict the probability of a disaster. The input is the latest natural disaster data, and the output is data indicating the probability of a disaster occurring. The results are then stored in a database and sent to the map service's API.

[1299] Step 4:

[1300] The server visualizes the disaster risk for each region on a map based on the data sent to the map service's API. Marks on the map use different colors and symbols to indicate the level of disaster risk. The input is disaster probability data, and the output is visually plotted map data.

[1301] Step 5:

[1302] The server retrieves the latest evacuation shelter information from local government and government agency databases and stores it in the database. The input is the evacuation shelter information retrieved from the external database, and the output is the updated evacuation shelter data. This information is also sent to the map service so that users can view it.

[1303] Step 6:

[1304] The device accesses the map service through a web browser or mobile app to obtain disaster risk information for the user's current location or a specified area. The input is the user's location information and data from the map service API, and the output is a disaster risk map displayed on the device.

[1305] Step 7:

[1306] The emotion engine uses a camera and microphone to recognize the user's emotional state. It analyzes facial expressions and voice to generate a customized message if the user is feeling anxious or nervous. The input is the user's image and voice data, and the output is the emotion analysis result and a reassuring message based on that.

[1307] Step 8:

[1308] The device displays a customized message based on the user's emotional state based on the analysis results of the emotion engine. The input is the emotion analysis result, and the output is the message displayed on the device. This allows users to use the system with peace of mind.

[1309] Step 9:

[1310] Users can check evacuation shelter information and disaster preparedness information through their devices. The input is evacuation shelter information obtained from a map service, and the output is evacuation information and preparedness information displayed on the device. Based on this, users can take specific evacuation actions.

[1311] In this way, the entire system works seamlessly, allowing users to obtain the latest disaster and evacuation information in real time. Furthermore, by providing customized messages according to the user's emotional state, users can feel at ease and take appropriate action.

[1312] 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.

[1313] 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.

[1314] 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.

[1315] 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.

[1316] 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.

[1317] 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.

[1318] 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).

[1319] 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.

[1320] 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."

[1321] 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.

[1322] 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).

[1323] 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.

[1324] 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.

[1325] 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.

[1326] 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.

[1327] 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.

[1328] 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.

[1329] 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 h...

Claims

1. A means of collecting natural disaster data in real time; A means of training a generative AI model using collected natural disaster data; and A means for predicting the probability of disaster occurrence using a generative AI model; A means for reflecting the predicted probability of disaster occurrence in a map service; a means of providing evacuation shelter information and disaster preparedness information; A system including:

2. The system of claim 1 further comprising means for storing past natural disaster data in a database and creating a dataset required for the generative AI model to learn.

3. The system according to claim 1, further comprising means for collecting evacuation shelter information from databases of local governments and government agencies, and providing the user with the latest evacuation shelter information in real time.

Images