System

Abstract

A system is provided.SOLUTION: A system including means for acquiring a current position from a user terminal, means for transmitting the acquired position information to a server, means for acquiring weather data and disaster data by the server, means for integrating the position information and the disaster data and inputting the integrated data to a generation model, means for estimating safety of a user by the generation model, means for registering and notifying safety information of the user based on an estimation result, means for uploading an image or a moving image posted by the user to the server, means for analyzing the uploaded data by an analysis model, and means for transmitting information based on an analysis result and making a rescue request.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] A major issue during disasters is the disruption of truly necessary rescue efforts due to the spread of false rescue requests and rumours. Furthermore, push-type support often results in the delay of relief efforts, as necessary supplies do not reach affected areas. The present invention aims to solve these issues and provide accurate and prompt rescue efforts and appropriate relief. [Means for solving the problem]

[0005] The present invention is a system including the following means: a means for acquiring a current location from a user terminal, a means for transmitting the acquired location information to a server, a means for the server to acquire weather data and disaster data, a means for integrating the location information and disaster data and inputting the data into a generative model, a means for the generative model to estimate the user's safety, a means for registering and notifying the user's safety information based on the estimation results, a means for uploading images and videos posted by the user to a server, a means for analyzing the uploaded data using an analytical model, and a means for transmitting information and requesting rescue based on the analysis results. This makes it possible to eliminate false information and rumors and quickly provide truly needed rescue and support to disaster-stricken areas.

[0006] "User terminal" refers to an electronic device that allows users to acquire and transmit location information and upload various data in a disaster situation.

[0007] "Server" refers to a central system that receives, stores, and analyzes data sent from user terminals, and also acquires external data and transmits information.

[0008] "Location information" is data indicating the current location of a user terminal, and is obtained using technology such as GPS.

[0009] "Disaster data" refers to information about natural disasters such as earthquakes and typhoons obtained from organizations such as the Japan Meteorological Agency and earthquake early warning systems.

[0010] A "generative model" refers to an artificial intelligence model that integrates and analyzes a user's location information and disaster data to estimate the user's safety.

[0011] "Safety information" is information indicating whether a user is safe or unsafe, and refers to data obtained based on the estimation results of a generative model.

[0012] "Images and videos" refers to visual data that users take and record on their devices and upload to a server to capture the situation in the disaster area.

[0013] "Analysis model" refers to an artificial intelligence model that analyzes the contents of uploaded images and videos and assesses the extent of the damage and the need for assistance.

[0014] "Information dissemination" refers to the act of communicating analysis results and other reliable information to official social media accounts and related organizations.

[0015] "Rescue request" refers to the act of urging emergency assistance from an appropriate rescue organization based on the user's safety information and the extent of the damage. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] This invention provides a disaster prevention AI platform that enables rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes them using a generative model to estimate the user's safety, and automatically carries out necessary rescue and support.

[0038] Explaining program processing in natural language

[0039] Data collection

[0040] 1. Obtain location information

[0041] Device: The user's smartphone periodically obtains its current location using its GPS function and sends that location information to the server.

[0042] 2. Disaster information acquisition

[0043] Server: Real-time disaster information is obtained from the Japan Meteorological Agency and earthquake early warning APIs. This information is stored in a database on the server and linked to the user's location information.

[0044] Data analysis

[0045] 1. Data Integration

[0046] Server: Integrates the acquired user location information and disaster information to create a dataset to be input into the generative model.

[0047] 2. Safety Estimation

[0048] Generative model: Estimates the user's safety based on the user's location information, the scale of the disaster, etc. The generative model uses machine learning algorithms to perform real-time analysis and determine the user's safety status.

[0049] Safety registration and rescue request

[0050] 1. Safety assessment

[0051] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[0052] 2. Safety Notification

[0053] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[0054] 3. Request for help

[0055] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[0056] Situation Analysis

[0057] 1. Media Upload

[0058] Users can take photos and videos of disaster areas and evacuation centers and upload them to the server.

[0059] 2. Media Analysis

[0060] Server: Receives images and videos uploaded by users and passes them to the analysis model.

[0061] Analysis model: Uses machine learning to analyze damage and people's movements from images and videos, and generates detailed reports.

[0062] Information dissemination

[0063] 1. Information aggregation

[0064] Server: Integrates analytical results with data from other reliable sources.

[0065] 2. Information dissemination

[0066] Server: Disseminates reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and prompt support information.

[0067] Specific examples

[0068] Example 1: Rescue operations after an earthquake

[0069] 1. Device: User A is in an area close to the epicenter, so he obtains his location information using GPS and sends it to the server.

[0070] 2. Server: Integrates location information with earthquake information obtained from the Japan Meteorological Agency.

[0071] 3. Server: Input the integrated data into the generative model and estimate the safety of user A.

[0072] 4. Generative model: Determines that user A is in a dangerous situation and returns the result to the server.

[0073] 5. Server: Since the user is judged to be in danger, safety information is sent to User A's emergency contact.

[0074] 6. Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue agencies.

[0075] Example 2: Shelter situation report

[0076] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads it to the server.

[0077] 2. Server: Passes the received photos to the analysis model.

[0078] 3. Analysis model: Analyzes crowded evacuation shelters from images and generates a report.

[0079] 4. Server: Based on the analysis report, the system posts information about the congestion situation at evacuation centers and the supplies needed to the official social media accounts.

[0080] 5. Server: Requests relief supplies based on reliable information.

[0081] This will help eliminate false information and rumors and quickly provide the rescue and assistance that is truly needed in affected areas.

[0082] The processing flow will be explained below.

[0083] Data collection

[0084] Step 1:

[0085] Device: The user's smartphone uses the GPS function to obtain the current location.

[0086] Step 2:

[0087] Device: Sends the acquired location information to the server.

[0088] Step 3:

[0089] Server: Obtains real-time disaster information from the Japan Meteorological Agency and earthquake early warning APIs.

[0090] Step 4:

[0091] Server: The acquired disaster information is stored in a database and linked to the user's location information.

[0092] Data analysis

[0093] Step 5:

[0094] Server: Integrates user location information and disaster information to create a dataset.

[0095] Step 6:

[0096] Server: Inputs the integrated data into the generative model.

[0097] Step 7:

[0098] Generative model: Estimates the user's safety based on the user's location information and the scale of the disaster.

[0099] Step 8:

[0100] Generative model: Returns the inference results to the server.

[0101] Safety registration and rescue request

[0102] Step 9:

[0103] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[0104] Step 10:

[0105] Server: If a user is deemed to be at risk, a notification is automatically sent to pre-registered emergency contacts.

[0106] Step 11:

[0107] Server: If necessary, initiates an emergency rescue request to the appropriate rescue agency.

[0108] Situation Analysis

[0109] Step 12:

[0110] User: Take photos and videos of disaster areas and evacuation centers and upload them to the server.

[0111] Step 13:

[0112] Server: Receives uploaded image and video data.

[0113] Step 14:

[0114] Server: Passes the received data to the analysis model.

[0115] Step 15:

[0116] Analysis model: Analyzes damage conditions and people's movements from images and videos, and generates detailed reports.

[0117] Information dissemination

[0118] Step 16:

[0119] Server: Integrates analytical results with data from other reliable sources.

[0120] Step 17:

[0121] Server: Disseminates reliable information to official social media accounts and related organizations.

[0122] Step 18:

[0123] Server: If necessary, requests for relief supplies and the current situation at evacuation centers will be sent out.

[0124] Example 1

[0125] 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."

[0126] Conventional disaster prevention systems only collect user location information and disaster information separately, making it difficult to integrate these information to estimate the user's safety in real time and provide prompt and accurate rescue and support. Furthermore, the lack of proper aggregation of disaster information and the dissemination of reliable information has led to the spread of false information, causing confusion in the affected areas. A new system is needed to solve these problems.

[0127] 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.

[0128] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation result, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, and means for the server to transmit information and request rescue based on the analysis result to official social media accounts and related organizations, thereby enabling rapid and accurate rescue and support when a disaster occurs.

[0129] A "user terminal" refers to an electronic device that can be carried by a user, specifically a device equipped with a GPS function such as a smartphone or tablet.

[0130] "Current location" refers to information about the user's current location obtained by the user's device using the GPS function.

[0131] A "server" is a high-performance computer system connected to a network, which receives data from multiple users and devices and processes and stores it.

[0132] "Weather data" refers to weather-related data provided by the Japan Meteorological Agency or weather services, specifically information such as temperature, wind speed, and precipitation.

[0133] "Disaster data" refers to information about natural disasters such as earthquakes, typhoons, and floods, and specifically includes the date and time of occurrence, location, and extent of impact.

[0134] A "generative model" refers to a computer program that uses machine learning algorithms to estimate a user's safety.

[0135] "Safety" refers to whether the user is currently safe or in imminent danger.

[0136] "Notification" refers to alert information sent from the system to users and emergency contacts, specifically by email or SMS.

[0137] "Images and videos" refers to visual data captured by users and uploaded to the system.

[0138] An "analysis model" refers to a program that uses machine learning algorithms to analyze uploaded images and videos and assess the extent of damage and people's movements.

[0139] "Official social media account" refers to a highly reliable social media account that the system uses to disseminate information.

[0140] "Related organizations" refer to official organizations that carry out disaster prevention and rescue operations, including fire departments, police departments, and medical institutions.

[0141] "Request for Assistance" means a request sent to relevant authorities to carry out rescue operations required in the event of a disaster.

[0142] This invention is a disaster prevention AI platform that enables rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes them using a generative AI model to estimate the user's safety, and automatically carries out necessary rescue and support.

[0143] Hardware and software used

[0144] User device: A portable electronic device with a GPS function, such as a smartphone or tablet, is used. This device can acquire the user's current location and send it to the server.

[0145] Server: A high-performance computer system is used to receive, process, and store data from multiple users and devices. It also collects real-time weather and disaster data from the Japan Meteorological Agency and earthquake early warning APIs.

[0146] Generative model: Using an AI model that uses a machine learning algorithm, the safety of a user is estimated in real time based on the user's location information and disaster information.

[0147] Analysis Model: Using machine learning, the model analyzes damage from images and videos and generates a detailed report based on media files uploaded by users.

[0148] Program processing

[0149] 1. Location information acquisition: The user device uses the built-in GPS function to acquire the user's current location at specific intervals, encrypts the data, and sends it to the server.

[0150] 2. Disaster information acquisition: The server acquires disaster information in real time from the Japan Meteorological Agency and earthquake early warning APIs and stores this information in a database.

[0151] 3. Data integration: The server integrates user location information and disaster information to create a dataset to be input into the generative AI model.

[0152] 4. Safety Prediction: The generative model estimates the user's safety based on the integrated data. For example, it determines whether the user is in danger based on the scale of the disaster, the user's location, and past data.

[0153] 5. Safety determination: The server determines whether the user is "safe" or "dangerous" based on the inference results returned by the generative model.

[0154] 6. Safety Notification: If a user is deemed to be in danger, the server automatically sends a notification to the user's emergency contacts, including the user's location and status.

[0155] 7. Rescue Request: If the user is deemed to be in danger, the server will issue an emergency rescue request to the appropriate rescue agency. This request will include the user's detailed location information and the disaster situation.

[0156] 8. Media Upload: Users can take photos and videos of disaster areas and evacuation centers with their smartphones and upload the media files to the server.

[0157] 9. Media Analysis: The server receives images and videos sent by users, inputs them into an analytical model, and generates a detailed report assessing the damage situation and people's movements.

[0158] 10. Information aggregation: The server integrates the analysis results with data from other reliable sources to provide comprehensive information aggregation.

[0159] 11. Information Dissemination: The server will disseminate information based on reliable information and analysis results to official social media accounts and related organizations, thereby eliminating false information and providing accurate and prompt support information.

[0160] Examples of specific examples and prompts

[0161] Example 1: Rescue operations after an earthquake

[0162] 1. Device: User A is in an area close to the epicenter, so his smartphone uses the GPS function to obtain location information every five minutes.

[0163] 2. Server: The server obtains information on the epicenter from the Japan Meteorological Agency API and stores the data in a database in real time.

[0164] 3. Server: Integrates user A's location information and disaster information about the epicenter, and creates a dataset to be input into the generative AI model.

[0165] 4. Generative model: Based on the integrated data, determine that user A is in a dangerous situation.

[0166] 5. Server: Based on the results returned from the generative model, User A is determined to be “dangerous.”

[0167] 6. Server: Notifies User A's emergency contacts of their location and safety status. For example, send an SMS message saying, "User A is near the epicenter and in a dangerous situation."

[0168] 7. Server: Sends a similar emergency rescue request to nearby rescue organizations.

[0169] Example 2: Reporting on the situation at an evacuation shelter

[0170] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads the photo to the server.

[0171] 2. Server: The received photos are input into the analysis model to analyze the congestion situation at the evacuation center.

[0172] 3. Analytical model: Determines the level of congestion at evacuation centers from images and generates a detailed report.

[0173] 4. Server: Stores the analysis results in a database and integrates them with data from other sources.

[0174] 5. Server: Posts information about the congestion situation at evacuation centers and necessary supplies to the official social media account. Also, conveys the information to appropriate organizations to provide prompt assistance.

[0175] Prompt Sentence Examples

[0176] "User A is near the epicenter of an earthquake, so his location information is obtained using GPS and sent to a server. Please explain a program that uses a generative model to estimate the user's safety and status, and sends a notification to emergency contacts based on the results."

[0177] This system is designed to enable rapid and accurate rescue and support in the event of a disaster, and can ensure the safety of users through multi-layered data analysis and information dissemination.

[0178] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0179] Step 1:

[0180] Location information acquisition

[0181] Device: The user device (smartphone) periodically acquires its current location using its built-in GPS function. The acquired location information is sent to the server as latitude and longitude data.

[0182] Input: Location information from the GPS sensor

[0183] Output: Latitude and longitude data of the current location sent to the server

[0184] Step 2:

[0185] Disaster information acquisition

[0186] Server: The server collects disaster information in real time from the Japan Meteorological Agency and earthquake early warning APIs. This information provides information such as the type of disaster, the time of occurrence, scale, and extent of damage.

[0187] Input: Disaster information from the Japan Meteorological Agency and Earthquake Early Warning API

[0188] Output: Disaster information stored on the server (type, time of occurrence, scale, extent of damage, etc.)

[0189] Step 3:

[0190] Data Integration

[0191] Server: The server combines the user's location information and disaster information to create a dataset to input into the generative model. Specifically, it matches the user's latitude and longitude with disaster information in that area.

[0192] Input: User location information and disaster information

[0193] Output: A unified dataset to feed into the generative model

[0194] Step 4:

[0195] Estimated safety

[0196] Generative model: The generative model estimates the user's safety based on the integrated dataset. For example, if the user's location is close to the epicenter or if the area is hit by strong shaking, the model estimates that the user is in danger.

[0197] Input: Integrated dataset

[0198] Output: User's safety status (safe or unsafe)

[0199] Step 5:

[0200] Safety assessment

[0201] Server: Based on the safety estimation results obtained from the generative model, the server determines whether the user is safe or at risk. If the user is determined to be at risk, the server proceeds to the next step.

[0202] Input: Safety estimation results from the generative model

[0203] Output: User safety assessment result (safe or dangerous)

[0204] Step 6:

[0205] Safety notification

[0206] Server: If the user is deemed to be in danger, the server automatically sends a notification to pre-registered emergency contacts, including the user's current location and the reason for the danger.

[0207] Input: User safety assessment results, emergency contact information

[0208] Output: Safety notification sent to emergency contacts (including location and details of the situation)

[0209] Step 7:

[0210] Rescue request

[0211] Server: If the user is determined to be in danger, the server sends an emergency rescue request to the appropriate rescue agency, including the user's location and the disaster situation.

[0212] Input: User safety assessment results, location information, disaster information

[0213] Output: Emergency rescue request sent to rescue agencies, including user's detailed location and disaster situation.

[0214] Step 8:

[0215] Media Upload

[0216] User: The user takes photos of the disaster area or the situation at the evacuation center with their smartphone and uploads the images and videos to the server.

[0217] Input: Images and videos taken by the user

[0218] Output: Media files uploaded to the server

[0219] Step 9:

[0220] Media Analysis

[0221] Server: The server passes the received images and videos to the analysis model, which analyzes the damage situation and people's movements. The analysis model uses machine learning algorithms to generate a detailed report.

[0222] Input: Images and videos uploaded to the server

[0223] Output: Report of damage status based on analytical model

[0224] Step 10:

[0225] Information aggregation

[0226] Server: The server integrates the analysis results with data from other reliable sources to aggregate comprehensive information, allowing for a comprehensive understanding of the situation in the affected areas.

[0227] Input: Analysis model reports, data from other sources

[0228] Output: Aggregated comprehensive information

[0229] Step 11:

[0230] Information dissemination

[0231] Server: Based on reliable information, the server will disseminate important information to official social media accounts and relevant organizations, thereby eliminating false information and providing prompt assistance.

[0232] Input: Aggregated comprehensive information

[0233] Output: Information posted on official social media accounts and related organizations

[0234] In this way, each step works in coordination, enabling rapid and accurate rescue and support in the event of a disaster.

[0235] (Application example 1)

[0236] 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."

[0237] When natural disasters occur, it is difficult to quickly and accurately collect and analyze information, which often leads to delays in rescue and assistance to victims. Furthermore, normal services may be suspended during disasters, and the disruption of services such as food delivery can have a significant impact on the lives of victims. Furthermore, insufficient means exist for confirming the safety of victims, notifying emergency contacts, and requesting assistance, hindering rapid and efficient rescue efforts.

[0238] 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.

[0239] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, means for transmitting information and requesting rescue based on the analysis results, and means for suspending or resuming specific services depending on the user's safety status during a disaster. This enables rapid and accurate information collection and analysis, accelerating rescue and support for disaster victims and reducing the risk of service suspension during a disaster.

[0240] A "user terminal" is a device such as a mobile terminal or smartphone that is owned by a user and has the function of acquiring location information.

[0241] "Current location" is information that is detected in real time by the user terminal and indicates the user's current geographical location.

[0242] A "server" is a computer system that receives, processes, and stores data over a network and works in conjunction with other devices.

[0243] "Weather data" refers to information about weather, temperature, precipitation, and other meteorological information provided by meteorological agencies.

[0244] "Disaster data" refers to information about natural disasters such as earthquakes, floods, and typhoons, and includes the date, time, location, and scale of the disaster.

[0245] A "generative model" is a data model that uses algorithms based on machine learning or deep learning to analyze input data and make predictions or inferences.

[0246] "Safety" refers to the status of the user indicating whether they are safe or in danger in a disaster situation.

[0247] The "estimated result" is information indicating the safety status of the user obtained as a result of analysis by the generative model.

[0248] "Safety information" is data related to the user's safety status, and is notified to emergency contacts and relevant organizations.

[0249] "Means for notifying" refers to the method or technology for sending messages or alerts to user terminals or emergency contacts.

[0250] "Images and videos" are visual media data that users take and upload to a server.

[0251] An "analysis model" is an algorithm or data model that uses machine learning and artificial intelligence technologies to analyze images and videos and extract useful information.

[0252] A "request for assistance" is a request made to relevant agencies to request emergency assistance or rescue operations.

[0253] "Means for suspending or resuming services" refers to methods or techniques for temporarily suspending specific services during a disaster and resuming them after the situation has improved.

[0254] This invention is a disaster prevention support system consisting of a user device, a server, and a generative model. Specifically, the user device acquires its current location using its GPS function and transmits it to the server. The server collects meteorological and disaster data in real time and integrates it with the user's location information. To analyze the shared data, the generative model uses a machine learning algorithm to estimate the user's safety.

[0255] The server does the following:

[0256] 1. Obtaining and sending location information:

[0257] User devices such as smartphones periodically use their GPS to obtain their current location and send it to a server. The Geopy library is used to obtain accurate location information.

[0258] 2. Meteorological and disaster data collection:

[0259] The server obtains real-time information using weather data and disaster data APIs, and stores the current weather conditions and disaster occurrence information in a database.

[0260] 3. Data integration and generative model use:

[0261] The user's location information and acquired disaster data are combined and input into a generative model. The generative model uses a machine learning algorithm to estimate the user's safety status. The generative model uses frameworks such as TENSORFLOW (registered trademark) and PyTorch.

[0262] 4. Safety confirmation and notification:

[0263] Based on the inference results from the generative model, the system registers the user's safety information and notifies emergency contacts. If the user's safety is judged to be in danger, an emergency notification is automatically sent.

[0264] 5. Emergency relief request:

[0265] The server automatically sends an appropriate rescue request based on the analysis results, which includes detailed location information and current situation of the victim.

[0266] 6. Suspending and Resuming Services:

[0267] In the event of a disaster, services such as food delivery will be temporarily suspended and resumed as soon as the safety of users is confirmed. This is done by notifying service providers using a REST API.

[0268] The hardware and software used includes:

[0269] Hardware: User device (smartphone), server

[0270] Software: Geopy, machine learning frameworks (TensorFlow, PyTorch), disaster information API requests, REST API

[0271] Consider the following scenario:

[0272] Example 1: Suspension of food delivery during an earthquake

[0273] 1. Device: If the user is in an earthquake-prone area, the smartphone's GPS acquires location information and sends it to the server.

[0274] 2. Server: Collects earthquake information, integrates it with the acquired location information, and inputs it into the generative model.

[0275] 3. Generative model: If the user is judged to be at risk, the server sends a stop notification to the delivery service.

[0276] 4. Server: Additionally, notify the user's emergency contacts and request appropriate assistance.

[0277] Example 2: Resuming food delivery after safety confirmation

[0278] 1. Device: After the user moves to a safe location, the device acquires location information again and sends it to the server.

[0279] 2. Server: Based on location information and the latest disaster information, the generative model confirms the user's safety.

[0280] 3. Generative model: If the user is deemed safe, the server sends a resume notification to the delivery service.

[0281] Example prompt sentence:

[0282] "Please create a program that determines the availability of food delivery services based on the user's location information and disaster information, and notifies them when the service resumes after confirming safety. Please also include the user's emergency contact information and API endpoints for suspending and resuming the service."

[0283] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0284] Step 1:

[0285] The user terminal acquires the current location and transmits the location information to the server.

[0286] Input: Current location obtained from the smartphone's GPS.

[0287] Data processing: Extract latitude and longitude information using the Geopy library.

[0288] Output: Location information including latitude and longitude of the user.

[0289] Step 2:

[0290] The server obtains weather and disaster data in real time.

[0291] Input: API requests for weather data and disaster information.

[0292] Data processing: Parse the data obtained from the API and extract the necessary information (weather, earthquake occurrence information, etc.).

[0293] Output: Datasets such as current weather, disaster occurrence information, etc.

[0294] Step 3:

[0295] The server integrates the user's location information and disaster data and inputs them into the generative model.

[0296] Input: User location information, weather data, disaster data.

[0297] Data processing: Creating a unified dataset and converting it into an input format for the generative model.

[0298] Output: The synthesized data that is input into the generative model.

[0299] Step 4:

[0300] The generative model estimates the user's safety.

[0301] Input: Integrated data.

[0302] Data calculation: Using machine learning algorithms, the user's safety status is estimated in real time.

[0303] Output: User safety information (safe / dangerous).

[0304] Step 5:

[0305] Based on the estimation results, the server notifies the user's safety information to emergency contacts.

[0306] Input: Estimation results (safety information), user's emergency contact data.

[0307] Data processing: Construct the necessary notification information (estimation results, user location information).

[0308] Output: Notify emergency contacts (send message).

[0309] Step 6:

[0310] The server will call for help if necessary.

[0311] Input: Estimation results, user location information, disaster data.

[0312] Data processing: Package the necessary information to request assistance.

[0313] Output: Request (send message) to appropriate rescue agency.

[0314] Step 7:

[0315] Servers stop or restart food delivery services in the event of a disaster.

[0316] Input: User safety information, service status information.

[0317] Data calculation: Decisions on suspending or resuming services based on safety information.

[0318] Output: Suspend or resume notification (API request) to the food delivery service.

[0319] These steps will ensure prompt relief efforts and proper management of services in the event of a disaster.

[0320] 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.

[0321] This invention provides a disaster prevention AI platform for providing rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes it using a generative model to estimate the user's safety, and automatically carries out necessary rescue and support. In addition, by adding an emotion engine that recognizes the user's emotions, it is possible to grasp the situation of victims in more detail and take appropriate measures.

[0322] Explaining program processing in natural language

[0323] Data collection

[0324] 1. Obtain location information

[0325] Device: The user's smartphone periodically obtains its current location using its GPS function and sends that location information to the server.

[0326] 2. Disaster information acquisition

[0327] Server: Real-time disaster information is obtained from the Japan Meteorological Agency and earthquake early warning APIs. This information is stored in a database on the server and linked to the user's location information.

[0328] Data analysis

[0329] 1. Data Integration

[0330] Server: Integrates user location information and disaster information to create a dataset to be input into the generative model.

[0331] 2. Safety Estimation

[0332] Generative model: Estimates the user's safety based on the user's location information, the scale of the disaster, etc. The generative model uses machine learning algorithms to perform real-time analysis and determine the user's safety status.

[0333] Safety registration and rescue request

[0334] 1. Safety assessment

[0335] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[0336] 2. Safety Notification

[0337] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[0338] 3. Request for help

[0339] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[0340] emotion recognition

[0341] 1. Media Upload

[0342] Users can take photos and videos of disaster areas and evacuation centers and upload them to the server.

[0343] 2. Emotion analysis

[0344] Server: Receives uploaded image and video data and passes it to the emotion engine.

[0345] Emotion Engine: Analyzes user emotions from images and videos and generates detailed emotional states.

[0346] 3. Emotion information notification

[0347] Server: Notifies the user's emotional state along with the estimated safety status to emergency contacts.

[0348] Situation Analysis

[0349] 1. Media Analysis

[0350] Server: The analysis results from the emotion engine are combined and passed to the analysis model.

[0351] Analysis model: Reassess the damage situation and people's movements from images and videos and generate detailed reports.

[0352] Information dissemination

[0353] 1. Information aggregation

[0354] Server: Integrates analytical results with data from other reliable sources.

[0355] 2. Information dissemination

[0356] Server: Disseminates reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and prompt support information.

[0357] Specific examples

[0358] Example 1: Rescue operations after an earthquake

[0359] 1. Device: User A is in an area close to the epicenter, so he obtains his location information using GPS and sends it to the server.

[0360] 2. Server: Integrates the acquired location information with real-time earthquake information.

[0361] 3. Server: Input the integrated data into the generative model and estimate the safety of user A.

[0362] 4. Generative model: Infers that user A is in a dangerous situation and returns the result to the server.

[0363] 5. Server: Since the user is judged to be in danger, safety information is sent to User A's emergency contact.

[0364] 6. Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue agencies.

[0365] Example 2: Identifying evacuees with a sense of sadness

[0366] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads it to the server.

[0367] 2. Server: Passes the received photo to the emotion engine.

[0368] 3. Emotion engine: Analyzes user B's sadness from the image and recognizes his emotional state.

[0369] 4. Server: Based on the emotion analysis results, notify the emergency contact of User B’s emotional state and the evacuation shelter situation.

[0370] 5. Server: Based on the analysis report, accurate information is posted on the official social media accounts and requests for relief supplies as needed.

[0371] By incorporating emotion recognition, it becomes possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[0372] The processing flow will be explained below.

[0373] Data collection

[0374] Step 1:

[0375] Device: The user's smartphone periodically obtains its current location using its GPS function.

[0376] Step 2:

[0377] Device: Sends the acquired location information to the server.

[0378] Step 3:

[0379] Server: Obtains real-time disaster information from the Japan Meteorological Agency and earthquake early warning APIs.

[0380] Step 4:

[0381] Server: The acquired disaster information is stored in a database and linked to the user's location information.

[0382] Data analysis

[0383] Step 5:

[0384] Server: Integrates user location information and disaster information to create a dataset.

[0385] Step 6:

[0386] Server: Inputs the integrated data into the generative model.

[0387] Step 7:

[0388] Generative model: Estimates the user's safety based on the user's location information and the scale of the disaster.

[0389] Step 8:

[0390] Generative model: Returns the inference results to the server.

[0391] Safety registration and rescue request

[0392] Step 9:

[0393] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[0394] Step 10:

[0395] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[0396] Step 11:

[0397] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[0398] emotion recognition

[0399] Step 12:

[0400] User: Take photos and videos of disaster areas and evacuation centers and upload them to the server.

[0401] Step 13:

[0402] Server: Receives uploaded image and video data.

[0403] Step 14:

[0404] Server: Passes the received data to the emotion engine.

[0405] Step 15:

[0406] Emotion engine: Analyzes user emotions from images and videos and generates emotional states.

[0407] Step 16:

[0408] Server: Integrates the analysis results from the emotion engine and the safety estimation results to generate detailed user information.

[0409] Information dissemination

[0410] Step 17:

[0411] Server: Based on the consolidated user information, sends notifications to the appropriate emergency contacts.

[0412] Step 18:

[0413] Server: Integrates analytical results with data from other reliable sources.

[0414] Step 19:

[0415] Server: Disseminates reliable information to official social media accounts and related organizations.

[0416] Step 20:

[0417] Server: If necessary, requests for relief supplies and the current situation at evacuation centers will be sent out.

[0418] Specific examples

[0419] Example 1: Rescue operations after an earthquake

[0420] Step 1:

[0421] Device: User A is in an area close to the epicenter, so his location information is obtained using GPS.

[0422] Step 2:

[0423] Device: Sends the acquired location information to the server.

[0424] Step 3:

[0425] Server: Obtains real-time earthquake information from the Japan Meteorological Agency and Earthquake Early Warning API.

[0426] Step 4:

[0427] Server: Integrates the acquired earthquake information and location information.

[0428] Step 5:

[0429] Server: Inputs the integrated data into the generative model.

[0430] Step 6:

[0431] Generative model: Estimates the safety of user A and returns the result to the server.

[0432] Step 7:

[0433] Server: Based on the safety estimation result, it is determined that User A is in a dangerous situation.

[0434] Step 8:

[0435] Server: Sends safety information to User A's emergency contacts. This notification includes location information and dangerous situations.

[0436] Step 9:

[0437] Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue organizations.

[0438] Example 2: Shelter situation report and emotion recognition

[0439] Step 1:

[0440] User: User B takes a photo of the situation at the evacuation shelter with his smartphone.

[0441] Step 2:

[0442] User: Uploads captured images and videos to the server.

[0443] Step 3:

[0444] Server: Receives uploaded image and video data.

[0445] Step 4:

[0446] Server: Passes the received data to the emotion engine.

[0447] Step 5:

[0448] Emotion engine: Analyzes user B's emotions from images and videos and generates their emotional state.

[0449] Step 6:

[0450] Server: Integrates the analysis results from the emotion engine with other data.

[0451] Step 7:

[0452] Server: Sends a notification to emergency contacts based on the integrated detailed information of User B.

[0453] Step 8:

[0454] Server: Posts the results of the emotion analysis and the situation at the evacuation center to the official social media account.

[0455] Step 9:

[0456] Server: If necessary, request relief supplies for evacuation centers.

[0457] By incorporating emotion recognition, it is possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[0458] Example 2

[0459] 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."

[0460] While conventional disaster prevention systems can collect user location information and disaster information, they have difficulty integrating this information to estimate the user's safety in real time and recognize emotions. Furthermore, they lack the ability to respond quickly when a user is deemed to be in danger, and they lack support that takes into account the emotional state of the victim, limiting the efficiency and effectiveness of rescue operations.

[0461] 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.

[0462] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data with an emotion engine, means for notifying emergency contacts based on the analysis results of the emotion engine, means for the analytical model to reassess the damage situation from the images and videos, and means for transmitting information and requesting rescue based on the analysis results. This enables fast and accurate rescue and support by combining user safety estimation and emotion recognition.

[0463] "User terminal" refers to an electronic device that is owned by a user and is capable of acquiring and transmitting location information and media data.

[0464] "Current location" refers to real-time geographical location information obtained by the user terminal using GPS.

[0465] "Server" refers to a central processing unit that receives information from user terminals and manages databases and performs data analysis.

[0466] "Weather data" refers to real-time weather information provided by the Japan Meteorological Agency and related organizations.

[0467] "Disaster data" refers to real-time damage status and disaster-related information obtained when disasters such as earthquakes and floods occur.

[0468] A "generative model" refers to a machine learning algorithm for estimating a user's safety based on their location and disaster information.

[0469] "Safety information" refers to information about the safety of a user estimated by a generative model.

[0470] "Notification" refers to the action of sending information about a user's safety or other important information to relevant people or organizations.

[0471] "Images and videos" refers to visual media data that users take and upload to a server.

[0472] "Emotion engine" refers to software that analyzes a user's emotional state from uploaded images and videos.

[0473] "Emergency Contacts" means the people or organizations that a User has registered to be notified in the event of an emergency.

[0474] "Analysis model" refers to a machine learning algorithm for reassessing the damage situation from uploaded media data.

[0475] "Requesting help" refers to an action of requesting help from an appropriate rescue organization when the user is determined to be in a dangerous situation.

[0476] "Information dissemination" refers to the action of providing reliable information based on the analysis results to related institutions and official social media accounts.

[0477] This invention relates to a disaster prevention AI platform for providing prompt and accurate rescue and support in the event of a disaster. This system acquires location information from the user's device, and the server integrates this information with disaster information and inputs it into an analysis model to estimate the user's safety. Furthermore, an emotion engine performs emotion analysis from images and videos uploaded by the user, allowing for a detailed understanding of the damage situation.

[0478] Data collection

[0479] Location information acquisition

[0480] The user's device periodically acquires the user's current location using the built-in GPS function and sends that information to the server. This location information is captured every 30 seconds using the location information service in the user's smartphone and uploaded to the server via HTTPS communication.

[0481] Disaster information acquisition

[0482] The server obtains real-time disaster information using the Japan Meteorological Agency and disaster alert APIs. This information is sent to the API at regular intervals, and the obtained data is parsed in JSON format and stored in a database.

[0483] Data analysis

[0484] Data Integration

[0485] The server combines user location information and disaster information to create a dataset to input into the generative AI model. This dataset retrieves the necessary information from a database on the server and is synchronized based on timestamps.

[0486] Estimated safety

[0487] The generative AI model estimates the user's safety based on their location and the scale of the disaster. The generative model uses a machine learning algorithm to perform real-time analysis and determine the user's safety status. This allows for a quick assessment of whether the user is "safe" or "at risk."

[0488] Safety registration and rescue request

[0489] Safety assessment

[0490] The server determines whether the user is safe or unsafe based on the estimation results of the generative model and records the result in a database.

[0491] Safety notification

[0492] If a user is deemed to be in danger, the server automatically sends a notification to emergency contacts via SMS or email, containing the user's current location and details of the disaster.

[0493] Rescue request

[0494] If necessary, the server will send an emergency call to the appropriate rescue agency, including detailed location and status, via the rescue agency's API endpoint.

[0495] emotion recognition

[0496] Media Upload

[0497] Users take photos of disaster areas and evacuation shelters with their smartphones and upload the media files to the server, which sends the files via HTTPS.

[0498] Emotion analysis

[0499] The server passes the uploaded images and videos to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the images and videos and returns the analysis results to the server. Based on this analysis result, the user's emotional state is stored in a database.

[0500] Emotion information notification

[0501] Based on the results of the emotion analysis, the server notifies emergency contacts of the user's emotional state and safety information.

[0502] Situation Analysis

[0503] Media Analysis

[0504] The server uses the emotion engine's analysis results to reassess the damage situation in more detail using an analytical model. The analytical model reanalyzes the detailed situation in the affected areas based on the data it receives and generates a report.

[0505] Information dissemination

[0506] Information aggregation

[0507] The server combines the analysis results with data from other reliable sources to create a single information set.

[0508] Information dissemination

[0509] The server sends reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and timely support. It posts information to official accounts using social media APIs and sends notifications to related organizations via email and dedicated APIs.

[0510] Specific examples

[0511] Rescue operations following an earthquake

[0512] Because User A is in an area close to the epicenter, his location information is acquired via GPS and sent to the server. The server combines the acquired location information with real-time earthquake information and inputs the data into a generative model to estimate User A's safety. The generative model estimates that User A is in danger and returns the result to the server. Since the server has determined that User A is in danger, it sends safety information to User A's emergency contacts and issues an emergency rescue request to nearby rescue organizations.

[0513] Confirmation of evacuees with a sense of despair

[0514] User B takes a photo of the evacuation shelter situation with their smartphone and uploads it to the server. The server passes the received photo to an emotion engine, which analyzes the image to determine User B's sense of despair and recognizes their emotional state. Based on the emotion analysis results, the server notifies emergency contacts of User B's emotional state and the evacuation shelter situation, and also posts accurate information to official social media accounts and requests relief supplies if necessary.

[0515] By incorporating emotion recognition, it is possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[0516] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0517] Step 1:

[0518] Location information acquisition

[0519] The user's terminal periodically obtains its current location using the GPS function.

[0520] Input: GPS satellite signal

[0521] Data processing: Obtaining the longitude and latitude information of the current location

[0522] Output: Acquired location data

[0523] Specific operation: The location information service in the smartphone runs, periodically capturing the current location, encrypting that information, and sending it to the server.

[0524] Step 2:

[0525] Location information transmission

[0526] The user's terminal transmits the acquired location information to the server.

[0527] Input: Location data

[0528] Data Computation: Location Encryption

[0529] Output: Encrypted location data

[0530] Specific operation: The device's transmission module is activated and uploads encrypted location information to the server via HTTPS communication.

[0531] Step 3:

[0532] Disaster information acquisition

[0533] The server obtains real-time disaster information from the Japan Meteorological Agency and disaster alert APIs.

[0534] Input: API request

[0535] Data processing: JSON parsing of acquired disaster information

[0536] Output: Disaster information data

[0537] Specific operation: The server periodically sends requests to the API endpoint, obtains the returned disaster information, and stores it in a database.

[0538] Step 4:

[0539] Data Integration

[0540] The server integrates location information and disaster information to create a dataset.

[0541] Input: User location information and disaster information data

[0542] Data Computation: Data Synchronization and Integration

[0543] Output: Unified dataset

[0544] Specific operation: The server retrieves location information and disaster information from the database and creates an integrated dataset based on that data.

[0545] Step 5:

[0546] Estimated safety

[0547] The server inputs the integrated data set into a generative AI model to estimate the user's safety.

[0548] Input: Integrated dataset

[0549] Data calculation: Estimation of user safety (application of machine learning algorithms)

[0550] Output: Safety estimation result

[0551] Specific operation: A dataset is input into the generative AI model, which uses a neural network to estimate the user's safety and status and returns the results to the server.

[0552] Step 6:

[0553] Safety assessment

[0554] The server determines whether the user is safe or dangerous based on the estimation results of the generative model.

[0555] Input: Safety estimation result

[0556] Data calculation: Determining safety status based on conditions

[0557] Output: Safety status (safe or dangerous)

[0558] Specific operation: The server analyzes the estimation results and determines the user's safety status according to the defined conditions.

[0559] Step 7:

[0560] Safety notification

[0561] The server sends a notification to emergency contacts if the user is determined to be at risk.

[0562] Input: Safety status, user details

[0563] Data operation: generating notification messages

[0564] Output: Notification message

[0565] Specific operation: The server sends a notification message to emergency contacts via SMS or email gateway.

[0566] Step 8:

[0567] Rescue request

[0568] The server will send an emergency rescue request to a rescue organization if necessary.

[0569] Input: User's detailed location information, disaster situation

[0570] Data calculation: Generating a help-call message

[0571] Output: Help request message

[0572] What happens: The server sends a request to the rescue agency's API endpoint, issuing an emergency rescue request including the user's details.

[0573] Step 9:

[0574] Media Upload

[0575] Users upload photos and videos of the disaster area to the server.

[0576] Input: Image and video data

[0577] Data Processing: Sending Media Files

[0578] Output: Media data stored on the server

[0579] Specific operation: A user takes a media file with their smartphone camera and uploads it to a server via HTTPS.

[0580] Step 10:

[0581] Emotion analysis

[0582] The server passes the uploaded media files to the emotion engine for emotion analysis.

[0583] Input: Media files

[0584] Data calculation: Emotional state analysis (application of image analysis algorithms)

[0585] Output: Emotion analysis results

[0586] Specific operation: The server sends media files to the emotion engine, which analyzes the user's emotional state from images and videos and returns the results to the server.

[0587] Step 11:

[0588] Emotion information notification

[0589] The server notifies emergency contacts of the user's emotional state and safety information based on the emotion analysis results.

[0590] Input: Emotion analysis results, safety information

[0591] Data operation: generating notification messages

[0592] Output: Notification message

[0593] Specific operation: The server generates a notification message containing the results of the sentiment analysis and sends it to emergency contacts via SMS or email.

[0594] Step 12:

[0595] Media Analysis

[0596] Based on the analysis results of the emotion engine, the server reevaluates the damage situation in more detail using an analytical model.

[0597] Input: Sentiment analysis results and other media data

[0598] Data calculation: Reassessment of damage situation (application of machine learning algorithms)

[0599] Output: Damage report

[0600] Specific operation: The server inputs data into the analytical model and generates a report to reassess the damage situation.

[0601] Step 13:

[0602] Information aggregation

[0603] The server integrates the analysis results with data from other reliable sources.

[0604] Input: Analysis results, data from other sources

[0605] Data Calculation: Data Integration

[0606] Output: Integrated information set

[0607] Specific operation: The server merges information obtained from multiple data sources and performs consistency checks.

[0608] Step 14:

[0609] Information dissemination

[0610] The server will send out reliable information to official social media accounts and related organizations.

[0611] Input: Integrated information set

[0612] Data operations: generating information messages

[0613] Output: Published information message

[0614] Specific operations: Post information to official accounts using SNS APIs, and send notifications to relevant organizations via email or dedicated APIs.

[0615] (Application example 2)

[0616] 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."

[0617] Currently, there are limited systems that can provide rapid and accurate rescue and assistance when disasters occur. In particular, there are no systems that can grasp the emotional state of victims and respond appropriately. Furthermore, technology to effectively utilize autonomous vehicles to carry out rapid rescue operations has not yet been fully developed. An efficient system to solve these issues is needed.

[0618] 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.

[0619] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for acquiring weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, means for transmitting information and requesting rescue based on the analysis results, means for acquiring the current location of the autonomous vehicle and calculating an optimal route, and means for recognizing the emotions of disaster victims and taking action based on their emotional state. This enables rapid and accurate rescue operations and appropriate responses based on the emotional state of disaster victims.

[0620] A "user terminal" is an electronic device owned by a user that is used to obtain current location information.

[0621] The "current location" is the geographical location information of the user terminal.

[0622] "Means" are methods or technical elements for performing a specific function.

[0623] A "server" is a computer system that can receive and process data sent from a user terminal.

[0624] "Weather data" is information relating to weather conditions and changes.

[0625] "Disaster data" refers to information related to disasters such as earthquakes, floods, and fires.

[0626] A "generative model" is an algorithm or machine learning model that analyzes acquired data and makes certain predictions or inferences.

[0627] "Safety" refers to the status of whether the user is in a safe state or in a dangerous state.

[0628] "Safety information" is information relating to the safety status of the user.

[0629] "Notification" is the act of transmitting specific information to other terminals or systems.

[0630] "Images and videos" are still images and videos taken by the user.

[0631] An "analytical model" is an algorithm or data processing method for analyzing collected data and extracting specific information.

[0632] A "request for rescue" is an act of requesting the support necessary to carry out rescue operations.

[0633] An "autonomous vehicle" is a vehicle that can be driven automatically without human operation.

[0634] A "route" is a path from a specific point to a destination.

[0635] "Calculation" means performing the necessary calculations to obtain the results.

[0636] An "emotion" is a psychological state such as joy, sadness, or fear felt by a user.

[0637] "Responding" means acting in a way that is adapted to a particular situation.

[0638] The system for realizing the present invention comprises the following elements.

[0639] 1. Data Collection

[0640] The user device periodically acquires its current location using its GPS function and sends that location information to the server. It also acquires media data (images and videos) using its camera and microphone to recognize the user's emotions, and uploads them to the server.

[0641] 2. Disaster Information Collection

[0642] The server retrieves real-time weather and disaster data from the Japan Meteorological Agency and earthquake early warning APIs, stores this information in a database, and integrates the location information and disaster data to create a dataset to be input into the generative model.

[0643] 3. Analysis using generative models

[0644] The generative model stored on the server estimates the user's safety in real time based on the user's location information and disaster data. The generative model uses a machine learning algorithm to perform real-time analysis and determine safety information.

[0645] 4. Notification of safety information and rescue requests

[0646] Based on the predictions returned by the generative model, the server determines whether the user is safe or in danger, and sends a notification to emergency contacts as necessary. It also obtains the current location of the autonomous vehicle, calculates the optimal route, and dispatches it to the disaster area.

[0647] 5. Media Data Analysis and Emotion Recognition

[0648] Uploaded image and video data is analyzed by an analysis model on the server. The emotion recognition engine analyzes the user's emotions from the images and videos and generates a detailed emotional state. Based on the analysis results, the server notifies emergency contacts of the user's emotional state and situation.

[0649] 6. Relief operations management

[0650] The autonomous vehicle will then follow the calculated optimal route to the disaster area and carry out rescue operations. During the rescue operations, further media data will be acquired and analyzed, and the server will issue a rescue request if necessary.

[0651] Hardware and software used

[0652] GPS module: Obtains the current location information of the user device.

[0653] Camera and microphone: Capture images and videos to recognize user emotions.

[0654] Server: Collects, integrates, analyzes, and notifies data.

[0655] Japan Meteorological Agency API and Earthquake Early Warning API: Obtain real-time disaster data.

[0656] Generative model: Estimates safety using machine learning algorithms.

[0657] Emotion recognition engine: Analyzes emotions from images and videos.

[0658] Specific examples

[0659] If the user is in an area affected by an earthquake, the user's device will use GPS to obtain location information and send it to the server. The server will then collect disaster data via an earthquake early warning API and use a generative model to estimate the user's safety. Based on the estimation results, a notification will be sent to emergency contacts and an autonomous vehicle will be dispatched along the optimal route. After arriving, the camera will analyze the user's facial expressions and provide any necessary psychological care.

[0660] Prompt Sentence Examples

[0661] "Please design a system that obtains detailed location information of users who are currently near the epicenter and need to evacuate, and sets the optimal rescue route. Also, after arriving, please use a camera to analyze the emotions of the victims and provide the necessary support."

[0662] This will enable quick and accurate rescue operations and appropriate responses based on the emotional state of the victims.

[0663] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0664] Step 1:

[0665] Obtaining location information of user devices

[0666] The user terminal periodically obtains its current location using a GPS module.

[0667] Input: Location data from the GPS module

[0668] Output: Current location of the user

[0669] Specific operation: The user device obtains its current location from GPS every few seconds and stores the data in memory.

[0670] Step 2:

[0671] Sending location information

[0672] The user terminal transmits the acquired location information to the server.

[0673] Input: current location of the user

[0674] Output: Location information sent to the server

[0675] Specific operation: The user device sends location information to the server's API endpoint as an HTTP request.

[0676] Step 3:

[0677] Obtaining disaster information

[0678] The server obtains weather and disaster data from the Japan Meteorological Agency API and Earthquake Early Warning API.

[0679] Input: API request

[0680] Output: Real-time disaster data

[0681] Specific operation: The server periodically sends requests to the API endpoint to obtain the latest disaster information and stores it in the database.

[0682] Step 4:

[0683] Data integration

[0684] The server integrates the acquired location information and disaster data to create a dataset to be input into the generative model.

[0685] Input: User's current location information, disaster data

[0686] Output: The dataset to feed into the generative model

[0687] Specific operation: The server combines location information and disaster data and converts them into a format that can be analyzed by the generative model.

[0688] Step 5:

[0689] Safety estimation using generative models

[0690] The generative model estimates the user's safety based on the user's location information and disaster data.

[0691] Input: Integrated dataset

[0692] Output: Safety estimation result

[0693] Specific operation: The generative model processes the integrated dataset and estimates the user's safety in real time.

[0694] Step 6:

[0695] Registration and notification of safety information

[0696] Based on the estimation results, the server registers the user's safety information and notifies emergency contacts.

[0697] Input: Safety estimation result

[0698] Output: Safety information sent to emergency contacts

[0699] Specific operation: The server stores the safety information in a database and notifies emergency contacts via email or SMS.

[0700] Step 7:

[0701] Uploading media data

[0702] Users take photos and videos of disaster areas and evacuation shelters and upload them to the server.

[0703] Input: Image and video data

[0704] Output: Media data stored on the server

[0705] Specific operation: The user uses the device's camera to capture media data and sends it to the server through the application.

[0706] Step 8:

[0707] Media Data Analysis

[0708] The server analyzes the uploaded image and video data using an analytical model.

[0709] Input: Media data

[0710] Output: Analysis results

[0711] Specific operation: The server uses an emotion recognition engine to analyze the media data and recognize the emotional state of the victim.

[0712] Step 9:

[0713] Autonomous vehicle routing

[0714] The server obtains the current location of the autonomous vehicle and calculates the optimal route.

[0715] Input: current vehicle location, disaster information

[0716] Output: Optimal route

[0717] Specific operation: The server obtains GPS data from the autonomous vehicle and calculates the optimal rescue route based on disaster information.

[0718] Step 10:

[0719] Carrying out rescue operations

[0720] The autonomous vehicle will follow the calculated optimal route to the disaster area and carry out rescue operations.

[0721] Input: Optimal Route

[0722] Output: Rescue operations in disaster areas

[0723] Specific operation: The autonomous vehicle uses the navigation system to drive along the specified route and begins rescue operations upon arrival.

[0724] Step 11:

[0725] Emotional information notification

[0726] Based on the emotion analysis results, the server notifies the user's emotional state and situation to the emergency contact.

[0727] Input: Analysis results

[0728] Output: Emotion information sent to emergency contacts

[0729] Specific operation: The server notifies the emergency contact of the analysis results via email or SMS.

[0730] This will enable quick and accurate rescue operations and appropriate responses based on the emotional state of the victims.

[0731] 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.

[0732] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0733] 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.

[0734] [Second embodiment]

[0735] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0736] 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.

[0737] 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).

[0738] 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.

[0739] 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.

[0740] 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).

[0741] 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.

[0742] 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.

[0743] 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.

[0744] 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.

[0745] In the smart glasses 214, 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.

[0746] 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."

[0747] This invention provides a disaster prevention AI platform that enables rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes them using a generative model to estimate the user's safety, and automatically carries out necessary rescue and support.

[0748] Explaining program processing in natural language

[0749] Data collection

[0750] 1. Obtain location information

[0751] Device: The user's smartphone periodically obtains its current location using its GPS function and sends that location information to the server.

[0752] 2. Disaster information acquisition

[0753] Server: Real-time disaster information is obtained from the Japan Meteorological Agency and earthquake early warning APIs. This information is stored in a database on the server and linked to the user's location information.

[0754] Data analysis

[0755] 1. Data Integration

[0756] Server: Integrates the acquired user location information and disaster information to create a dataset to be input into the generative model.

[0757] 2. Safety Estimation

[0758] Generative model: Estimates the user's safety based on the user's location information, the scale of the disaster, etc. The generative model uses machine learning algorithms to perform real-time analysis and determine the user's safety status.

[0759] Safety registration and rescue request

[0760] 1. Safety assessment

[0761] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[0762] 2. Safety Notification

[0763] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[0764] 3. Request for help

[0765] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[0766] Situation Analysis

[0767] 1. Media Upload

[0768] Users can take photos and videos of disaster areas and evacuation centers and upload them to the server.

[0769] 2. Media Analysis

[0770] Server: Receives images and videos uploaded by users and passes them to the analysis model.

[0771] Analysis model: Uses machine learning to analyze damage and people's movements from images and videos, and generates detailed reports.

[0772] Information dissemination

[0773] 1. Information aggregation

[0774] Server: Integrates analytical results with data from other reliable sources.

[0775] 2. Information dissemination

[0776] Server: Disseminates reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and prompt support information.

[0777] Specific examples

[0778] Example 1: Rescue operations after an earthquake

[0779] 1. Device: User A is in an area close to the epicenter, so he obtains his location information using GPS and sends it to the server.

[0780] 2. Server: Integrates location information with earthquake information obtained from the Japan Meteorological Agency.

[0781] 3. Server: Input the integrated data into the generative model and estimate the safety of user A.

[0782] 4. Generative model: Determines that user A is in a dangerous situation and returns the result to the server.

[0783] 5. Server: Since the user is judged to be in danger, safety information is sent to User A's emergency contact.

[0784] 6. Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue agencies.

[0785] Example 2: Shelter situation report

[0786] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads it to the server.

[0787] 2. Server: Passes the received photos to the analysis model.

[0788] 3. Analysis model: Analyzes crowded evacuation shelters from images and generates a report.

[0789] 4. Server: Based on the analysis report, the system posts information about the congestion situation at evacuation centers and the supplies needed to the official social media accounts.

[0790] 5. Server: Requests relief supplies based on reliable information.

[0791] This will help eliminate false information and rumors and quickly provide the rescue and assistance that is truly needed in affected areas.

[0792] The processing flow will be explained below.

[0793] Data collection

[0794] Step 1:

[0795] Device: The user's smartphone uses the GPS function to obtain the current location.

[0796] Step 2:

[0797] Device: Sends the acquired location information to the server.

[0798] Step 3:

[0799] Server: Obtains real-time disaster information from the Japan Meteorological Agency and earthquake early warning APIs.

[0800] Step 4:

[0801] Server: The acquired disaster information is stored in a database and linked to the user's location information.

[0802] Data analysis

[0803] Step 5:

[0804] Server: Integrates user location information and disaster information to create a dataset.

[0805] Step 6:

[0806] Server: Inputs the integrated data into the generative model.

[0807] Step 7:

[0808] Generative model: Estimates the user's safety based on the user's location information and the scale of the disaster.

[0809] Step 8:

[0810] Generative model: Returns the inference results to the server.

[0811] Safety registration and rescue request

[0812] Step 9:

[0813] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[0814] Step 10:

[0815] Server: If a user is deemed to be at risk, a notification is automatically sent to pre-registered emergency contacts.

[0816] Step 11:

[0817] Server: If necessary, initiates an emergency rescue request to the appropriate rescue agency.

[0818] Situation Analysis

[0819] Step 12:

[0820] User: Take photos and videos of disaster areas and evacuation centers and upload them to the server.

[0821] Step 13:

[0822] Server: Receives uploaded image and video data.

[0823] Step 14:

[0824] Server: Passes the received data to the analysis model.

[0825] Step 15:

[0826] Analysis model: Analyzes damage conditions and people's movements from images and videos, and generates detailed reports.

[0827] Information dissemination

[0828] Step 16:

[0829] Server: Integrates analytical results with data from other reliable sources.

[0830] Step 17:

[0831] Server: Disseminates reliable information to official social media accounts and related organizations.

[0832] Step 18:

[0833] Server: If necessary, requests for relief supplies and the current situation at evacuation centers will be sent out.

[0834] Example 1

[0835] 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."

[0836] Conventional disaster prevention systems only collect user location information and disaster information separately, making it difficult to integrate these information to estimate the user's safety in real time and provide prompt and accurate rescue and support. Furthermore, the lack of proper aggregation of disaster information and the dissemination of reliable information has led to the spread of false information, causing confusion in the affected areas. A new system is needed to solve these problems.

[0837] 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.

[0838] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation result, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, and means for the server to transmit information and request rescue based on the analysis result to official social media accounts and related organizations, thereby enabling rapid and accurate rescue and support when a disaster occurs.

[0839] A "user terminal" refers to an electronic device that can be carried by a user, specifically a device equipped with a GPS function such as a smartphone or tablet.

[0840] "Current location" refers to information about the user's current location obtained by the user's device using the GPS function.

[0841] A "server" is a high-performance computer system connected to a network, which receives data from multiple users and devices and processes and stores it.

[0842] "Weather data" refers to weather-related data provided by the Japan Meteorological Agency or weather services, specifically information such as temperature, wind speed, and precipitation.

[0843] "Disaster data" refers to information about natural disasters such as earthquakes, typhoons, and floods, and specifically includes the date and time of occurrence, location, and extent of impact.

[0844] A "generative model" refers to a computer program that uses machine learning algorithms to estimate a user's safety.

[0845] "Safety" refers to whether the user is currently safe or in imminent danger.

[0846] "Notification" refers to alert information sent from the system to users and emergency contacts, specifically by email or SMS.

[0847] "Images and videos" refers to visual data captured by users and uploaded to the system.

[0848] An "analysis model" refers to a program that uses machine learning algorithms to analyze uploaded images and videos and assess the extent of damage and people's movements.

[0849] "Official social media account" refers to a highly reliable social media account that the system uses to disseminate information.

[0850] "Related organizations" refer to official organizations that carry out disaster prevention and rescue operations, including fire departments, police departments, and medical institutions.

[0851] "Request for Assistance" means a request sent to relevant authorities to carry out rescue operations required in the event of a disaster.

[0852] This invention is a disaster prevention AI platform that enables rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes them using a generative AI model to estimate the user's safety, and automatically carries out necessary rescue and support.

[0853] Hardware and software used

[0854] User device: A portable electronic device with a GPS function, such as a smartphone or tablet, is used. This device can acquire the user's current location and send it to the server.

[0855] Server: A high-performance computer system is used to receive, process, and store data from multiple users and devices. It also collects real-time weather and disaster data from the Japan Meteorological Agency and earthquake early warning APIs.

[0856] Generative model: Using an AI model that uses a machine learning algorithm, the safety of a user is estimated in real time based on the user's location information and disaster information.

[0857] Analysis Model: Using machine learning, the model analyzes damage from images and videos and generates a detailed report based on media files uploaded by users.

[0858] Program processing

[0859] 1. Location information acquisition: The user device uses the built-in GPS function to acquire the user's current location at specific intervals, encrypts the data, and sends it to the server.

[0860] 2. Disaster information acquisition: The server acquires disaster information in real time from the Japan Meteorological Agency and earthquake early warning APIs and stores this information in a database.

[0861] 3. Data integration: The server integrates user location information and disaster information to create a dataset to be input into the generative AI model.

[0862] 4. Safety Prediction: The generative model estimates the user's safety based on the integrated data. For example, it determines whether the user is in danger based on the scale of the disaster, the user's location, and past data.

[0863] 5. Safety determination: The server determines whether the user is "safe" or "dangerous" based on the inference results returned by the generative model.

[0864] 6. Safety Notification: If a user is deemed to be in danger, the server automatically sends a notification to the user's emergency contacts, including the user's location and status.

[0865] 7. Rescue Request: If the user is deemed to be in danger, the server will issue an emergency rescue request to the appropriate rescue agency. This request will include the user's detailed location information and the disaster situation.

[0866] 8. Media Upload: Users can take photos and videos of disaster areas and evacuation centers with their smartphones and upload the media files to the server.

[0867] 9. Media Analysis: The server receives images and videos sent by users, inputs them into an analytical model, and generates a detailed report assessing the damage situation and people's movements.

[0868] 10. Information aggregation: The server integrates the analysis results with data from other reliable sources to provide comprehensive information aggregation.

[0869] 11. Information Dissemination: The server will disseminate information based on reliable information and analysis results to official social media accounts and related organizations, thereby eliminating false information and providing accurate and prompt support information.

[0870] Examples of specific examples and prompts

[0871] Example 1: Rescue operations after an earthquake

[0872] 1. Device: User A is in an area close to the epicenter, so his smartphone uses the GPS function to obtain location information every five minutes.

[0873] 2. Server: The server obtains information on the epicenter from the Japan Meteorological Agency API and stores the data in a database in real time.

[0874] 3. Server: Integrates user A's location information and disaster information about the epicenter, and creates a dataset to be input into the generative AI model.

[0875] 4. Generative model: Based on the integrated data, determine that user A is in a dangerous situation.

[0876] 5. Server: Based on the results returned from the generative model, User A is determined to be “dangerous.”

[0877] 6. Server: Notifies User A's emergency contacts of their location and safety status. For example, send an SMS message saying, "User A is near the epicenter and in a dangerous situation."

[0878] 7. Server: Sends a similar emergency rescue request to nearby rescue organizations.

[0879] Example 2: Reporting on the situation at an evacuation shelter

[0880] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads the photo to the server.

[0881] 2. Server: The received photos are input into the analysis model to analyze the congestion situation at the evacuation center.

[0882] 3. Analytical model: Determines the level of congestion at evacuation centers from images and generates a detailed report.

[0883] 4. Server: Stores the analysis results in a database and integrates them with data from other sources.

[0884] 5. Server: Posts information about the congestion situation at evacuation centers and necessary supplies to the official social media account. Also, conveys the information to appropriate organizations to provide prompt assistance.

[0885] Prompt Sentence Examples

[0886] "User A is near the epicenter of an earthquake, so his location information is obtained using GPS and sent to a server. Please explain a program that uses a generative model to estimate the user's safety and status, and sends a notification to emergency contacts based on the results."

[0887] This system is designed to enable rapid and accurate rescue and support in the event of a disaster, and can ensure the safety of users through multi-layered data analysis and information dissemination.

[0888] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0889] Step 1:

[0890] Location information acquisition

[0891] Device: The user device (smartphone) periodically acquires its current location using its built-in GPS function. The acquired location information is sent to the server as latitude and longitude data.

[0892] Input: Location information from the GPS sensor

[0893] Output: Latitude and longitude data of the current location sent to the server

[0894] Step 2:

[0895] Disaster information acquisition

[0896] Server: The server collects disaster information in real time from the Japan Meteorological Agency and earthquake early warning APIs. This information provides information such as the type of disaster, the time of occurrence, scale, and extent of damage.

[0897] Input: Disaster information from the Japan Meteorological Agency and Earthquake Early Warning API

[0898] Output: Disaster information stored on the server (type, time of occurrence, scale, extent of damage, etc.)

[0899] Step 3:

[0900] Data Integration

[0901] Server: The server combines the user's location information and disaster information to create a dataset to input into the generative model. Specifically, it matches the user's latitude and longitude with disaster information in that area.

[0902] Input: User location information and disaster information

[0903] Output: A unified dataset to feed into the generative model

[0904] Step 4:

[0905] Estimated safety

[0906] Generative model: The generative model estimates the user's safety based on the integrated dataset. For example, if the user's location is close to the epicenter or if the area is hit by strong shaking, the model estimates that the user is in danger.

[0907] Input: Integrated dataset

[0908] Output: User's safety status (safe or unsafe)

[0909] Step 5:

[0910] Safety assessment

[0911] Server: Based on the safety estimation results obtained from the generative model, the server determines whether the user is safe or at risk. If the user is determined to be at risk, the server proceeds to the next step.

[0912] Input: Safety estimation results from the generative model

[0913] Output: User safety assessment result (safe or dangerous)

[0914] Step 6:

[0915] Safety notification

[0916] Server: If the user is deemed to be in danger, the server automatically sends a notification to pre-registered emergency contacts, including the user's current location and the reason for the danger.

[0917] Input: User safety assessment results, emergency contact information

[0918] Output: Safety notification sent to emergency contacts (including location and details of the situation)

[0919] Step 7:

[0920] Rescue request

[0921] Server: If the user is determined to be in danger, the server sends an emergency rescue request to the appropriate rescue agency, including the user's location and the disaster situation.

[0922] Input: User safety assessment results, location information, disaster information

[0923] Output: Emergency rescue request sent to rescue agencies, including user's detailed location and disaster situation.

[0924] Step 8:

[0925] Media Upload

[0926] User: The user takes photos of the disaster area or the situation at the evacuation center with their smartphone and uploads the images and videos to the server.

[0927] Input: Images and videos taken by the user

[0928] Output: Media files uploaded to the server

[0929] Step 9:

[0930] Media Analysis

[0931] Server: The server passes the received images and videos to the analysis model, which analyzes the damage situation and people's movements. The analysis model uses machine learning algorithms to generate a detailed report.

[0932] Input: Images and videos uploaded to the server

[0933] Output: Report of damage status based on analytical model

[0934] Step 10:

[0935] Information aggregation

[0936] Server: The server integrates the analysis results with data from other reliable sources to aggregate comprehensive information, allowing for a comprehensive understanding of the situation in the affected areas.

[0937] Input: Analysis model reports, data from other sources

[0938] Output: Aggregated comprehensive information

[0939] Step 11:

[0940] Information dissemination

[0941] Server: Based on reliable information, the server will disseminate important information to official social media accounts and relevant organizations, thereby eliminating false information and providing prompt assistance.

[0942] Input: Aggregated comprehensive information

[0943] Output: Information posted on official social media accounts and related organizations

[0944] In this way, each step works in coordination, enabling rapid and accurate rescue and support in the event of a disaster.

[0945] (Application example 1)

[0946] 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."

[0947] When natural disasters occur, it is difficult to quickly and accurately collect and analyze information, which often leads to delays in rescue and assistance to victims. Furthermore, normal services may be suspended during disasters, and the disruption of services such as food delivery can have a significant impact on the lives of victims. Furthermore, insufficient means exist for confirming the safety of victims, notifying emergency contacts, and requesting assistance, hindering rapid and efficient rescue efforts.

[0948] 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.

[0949] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, means for transmitting information and requesting rescue based on the analysis results, and means for suspending or resuming specific services depending on the user's safety status during a disaster. This enables rapid and accurate information collection and analysis, accelerating rescue and support for disaster victims and reducing the risk of service suspension during a disaster.

[0950] A "user terminal" is a device such as a mobile terminal or smartphone that is owned by a user and has the function of acquiring location information.

[0951] "Current location" is information that is detected in real time by the user terminal and indicates the user's current geographical location.

[0952] A "server" is a computer system that receives, processes, and stores data over a network and works in conjunction with other devices.

[0953] "Weather data" refers to information about weather, temperature, precipitation, and other meteorological information provided by meteorological agencies.

[0954] "Disaster data" refers to information about natural disasters such as earthquakes, floods, and typhoons, and includes the date, time, location, and scale of the disaster.

[0955] A "generative model" is a data model that uses algorithms based on machine learning or deep learning to analyze input data and make predictions or inferences.

[0956] "Safety" refers to the status of the user indicating whether they are safe or in danger in a disaster situation.

[0957] The "estimated result" is information indicating the safety status of the user obtained as a result of analysis by the generative model.

[0958] "Safety information" is data related to the user's safety status, and is notified to emergency contacts and relevant organizations.

[0959] "Means for notifying" refers to the method or technology for sending messages or alerts to user terminals or emergency contacts.

[0960] "Images and videos" are visual media data that users take and upload to a server.

[0961] An "analysis model" is an algorithm or data model that uses machine learning and artificial intelligence technologies to analyze images and videos and extract useful information.

[0962] A "request for assistance" is a request made to relevant agencies to request emergency assistance or rescue operations.

[0963] "Means for suspending or resuming services" refers to methods or techniques for temporarily suspending specific services during a disaster and resuming them after the situation has improved.

[0964] This invention is a disaster prevention support system consisting of a user device, a server, and a generative model. Specifically, the user device acquires its current location using its GPS function and transmits it to the server. The server collects meteorological and disaster data in real time and integrates it with the user's location information. To analyze the shared data, the generative model uses a machine learning algorithm to estimate the user's safety.

[0965] The server does the following:

[0966] 1. Obtaining and sending location information:

[0967] User devices such as smartphones periodically use their GPS to obtain their current location and send it to a server. The Geopy library is used to obtain accurate location information.

[0968] 2. Meteorological and disaster data collection:

[0969] The server obtains real-time information using weather data and disaster data APIs, and stores the current weather conditions and disaster occurrence information in a database.

[0970] 3. Data integration and generative model use:

[0971] The user's location information and acquired disaster data are combined and input into a generative model, which uses machine learning algorithms to estimate the user's safety status. The generative model uses frameworks such as TensorFlow and PyTorch.

[0972] 4. Safety confirmation and notification:

[0973] Based on the inference results from the generative model, the system registers the user's safety information and notifies emergency contacts. If the user's safety is judged to be in danger, an emergency notification is automatically sent.

[0974] 5. Emergency relief request:

[0975] The server automatically sends an appropriate rescue request based on the analysis results, which includes detailed location information and current situation of the victim.

[0976] 6. Suspending and Resuming Services:

[0977] In the event of a disaster, services such as food delivery will be temporarily suspended and resumed as soon as the safety of users is confirmed. This is done by notifying service providers using a REST API.

[0978] The hardware and software used includes:

[0979] Hardware: User device (smartphone), server

[0980] Software: Geopy, machine learning frameworks (TensorFlow, PyTorch), disaster information API requests, REST API

[0981] Consider the following scenario:

[0982] Example 1: Suspension of food delivery during an earthquake

[0983] 1. Device: If the user is in an earthquake-prone area, the smartphone's GPS acquires location information and sends it to the server.

[0984] 2. Server: Collects earthquake information, integrates it with the acquired location information, and inputs it into the generative model.

[0985] 3. Generative model: If the user is judged to be at risk, the server sends a stop notification to the delivery service.

[0986] 4. Server: Additionally, notify the user's emergency contacts and request appropriate assistance.

[0987] Example 2: Resuming food delivery after safety confirmation

[0988] 1. Device: After the user moves to a safe location, the device acquires location information again and sends it to the server.

[0989] 2. Server: Based on location information and the latest disaster information, the generative model confirms the user's safety.

[0990] 3. Generative model: If the user is deemed safe, the server sends a resume notification to the delivery service.

[0991] Example prompt sentence:

[0992] "Please create a program that determines the availability of food delivery services based on the user's location information and disaster information, and notifies them when the service resumes after confirming safety. Please also include the user's emergency contact information and API endpoints for suspending and resuming the service."

[0993] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0994] Step 1:

[0995] The user terminal acquires the current location and transmits the location information to the server.

[0996] Input: Current location obtained from the smartphone's GPS.

[0997] Data processing: Extract latitude and longitude information using the Geopy library.

[0998] Output: Location information including latitude and longitude of the user.

[0999] Step 2:

[1000] The server obtains weather and disaster data in real time.

[1001] Input: API requests for weather data and disaster information.

[1002] Data processing: Parse the data obtained from the API and extract the necessary information (weather, earthquake occurrence information, etc.).

[1003] Output: Datasets such as current weather, disaster occurrence information, etc.

[1004] Step 3:

[1005] The server integrates the user's location information and disaster data and inputs them into the generative model.

[1006] Input: User location information, weather data, disaster data.

[1007] Data processing: Creating a unified dataset and converting it into an input format for the generative model.

[1008] Output: The synthesized data that is input into the generative model.

[1009] Step 4:

[1010] The generative model estimates the user's safety.

[1011] Input: Integrated data.

[1012] Data calculation: Using machine learning algorithms, the user's safety status is estimated in real time.

[1013] Output: User safety information (safe / dangerous).

[1014] Step 5:

[1015] Based on the estimation results, the server notifies the user's safety information to emergency contacts.

[1016] Input: Estimation results (safety information), user's emergency contact data.

[1017] Data processing: Construct the necessary notification information (estimation results, user location information).

[1018] Output: Notify emergency contacts (send message).

[1019] Step 6:

[1020] The server will call for help if necessary.

[1021] Input: Estimation results, user location information, disaster data.

[1022] Data processing: Package the necessary information to request assistance.

[1023] Output: Request (send message) to appropriate rescue agency.

[1024] Step 7:

[1025] Servers stop or restart food delivery services in the event of a disaster.

[1026] Input: User safety information, service status information.

[1027] Data calculation: Decisions on suspending or resuming services based on safety information.

[1028] Output: Suspend or resume notification (API request) to the food delivery service.

[1029] These steps will ensure prompt relief efforts and proper management of services in the event of a disaster.

[1030] 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.

[1031] This invention provides a disaster prevention AI platform for providing rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes it using a generative model to estimate the user's safety, and automatically carries out necessary rescue and support. In addition, by adding an emotion engine that recognizes the user's emotions, it is possible to grasp the situation of victims in more detail and take appropriate measures.

[1032] Explaining program processing in natural language

[1033] Data collection

[1034] 1. Obtain location information

[1035] Device: The user's smartphone periodically obtains its current location using its GPS function and sends that location information to the server.

[1036] 2. Disaster information acquisition

[1037] Server: Real-time disaster information is obtained from the Japan Meteorological Agency and earthquake early warning APIs. This information is stored in a database on the server and linked to the user's location information.

[1038] Data analysis

[1039] 1. Data Integration

[1040] Server: Integrates user location information and disaster information to create a dataset to be input into the generative model.

[1041] 2. Safety Estimation

[1042] Generative model: Estimates the user's safety based on the user's location information, the scale of the disaster, etc. The generative model uses machine learning algorithms to perform real-time analysis and determine the user's safety status.

[1043] Safety registration and rescue request

[1044] 1. Safety assessment

[1045] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[1046] 2. Safety Notification

[1047] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[1048] 3. Request for help

[1049] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[1050] emotion recognition

[1051] 1. Media Upload

[1052] Users can take photos and videos of disaster areas and evacuation centers and upload them to the server.

[1053] 2. Emotion analysis

[1054] Server: Receives uploaded image and video data and passes it to the emotion engine.

[1055] Emotion Engine: Analyzes user emotions from images and videos and generates detailed emotional states.

[1056] 3. Emotion information notification

[1057] Server: Notifies the user's emotional state along with the estimated safety status to emergency contacts.

[1058] Situation Analysis

[1059] 1. Media Analysis

[1060] Server: The analysis results from the emotion engine are combined and passed to the analysis model.

[1061] Analysis model: Reassess the damage situation and people's movements from images and videos and generate detailed reports.

[1062] Information dissemination

[1063] 1. Information aggregation

[1064] Server: Integrates analytical results with data from other reliable sources.

[1065] 2. Information dissemination

[1066] Server: Disseminates reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and prompt support information.

[1067] Specific examples

[1068] Example 1: Rescue operations after an earthquake

[1069] 1. Device: User A is in an area close to the epicenter, so he obtains his location information using GPS and sends it to the server.

[1070] 2. Server: Integrates the acquired location information with real-time earthquake information.

[1071] 3. Server: Input the integrated data into the generative model and estimate the safety of user A.

[1072] 4. Generative model: Infers that user A is in a dangerous situation and returns the result to the server.

[1073] 5. Server: Since the user is judged to be in danger, safety information is sent to User A's emergency contact.

[1074] 6. Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue agencies.

[1075] Example 2: Identifying evacuees with a sense of sadness

[1076] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads it to the server.

[1077] 2. Server: Passes the received photo to the emotion engine.

[1078] 3. Emotion engine: Analyzes user B's sadness from the image and recognizes his emotional state.

[1079] 4. Server: Based on the emotion analysis results, notify the emergency contact of User B’s emotional state and the evacuation shelter situation.

[1080] 5. Server: Based on the analysis report, accurate information is posted on the official social media accounts and requests for relief supplies as needed.

[1081] By incorporating emotion recognition, it becomes possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[1082] The processing flow will be explained below.

[1083] Data collection

[1084] Step 1:

[1085] Device: The user's smartphone periodically obtains its current location using its GPS function.

[1086] Step 2:

[1087] Device: Sends the acquired location information to the server.

[1088] Step 3:

[1089] Server: Obtains real-time disaster information from the Japan Meteorological Agency and earthquake early warning APIs.

[1090] Step 4:

[1091] Server: The acquired disaster information is stored in a database and linked to the user's location information.

[1092] Data analysis

[1093] Step 5:

[1094] Server: Integrates user location information and disaster information to create a dataset.

[1095] Step 6:

[1096] Server: Inputs the integrated data into the generative model.

[1097] Step 7:

[1098] Generative model: Estimates the user's safety based on the user's location information and the scale of the disaster.

[1099] Step 8:

[1100] Generative model: Returns the inference results to the server.

[1101] Safety registration and rescue request

[1102] Step 9:

[1103] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[1104] Step 10:

[1105] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[1106] Step 11:

[1107] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[1108] emotion recognition

[1109] Step 12:

[1110] User: Take photos and videos of disaster areas and evacuation centers and upload them to the server.

[1111] Step 13:

[1112] Server: Receives uploaded image and video data.

[1113] Step 14:

[1114] Server: Passes the received data to the emotion engine.

[1115] Step 15:

[1116] Emotion engine: Analyzes user emotions from images and videos and generates emotional states.

[1117] Step 16:

[1118] Server: Integrates the analysis results from the emotion engine and the safety estimation results to generate detailed user information.

[1119] Information dissemination

[1120] Step 17:

[1121] Server: Based on the consolidated user information, sends notifications to the appropriate emergency contacts.

[1122] Step 18:

[1123] Server: Integrates analytical results with data from other reliable sources.

[1124] Step 19:

[1125] Server: Disseminates reliable information to official social media accounts and related organizations.

[1126] Step 20:

[1127] Server: If necessary, requests for relief supplies and the current situation at evacuation centers will be sent out.

[1128] Specific examples

[1129] Example 1: Rescue operations after an earthquake

[1130] Step 1:

[1131] Device: User A is in an area close to the epicenter, so his location information is obtained using GPS.

[1132] Step 2:

[1133] Device: Sends the acquired location information to the server.

[1134] Step 3:

[1135] Server: Obtains real-time earthquake information from the Japan Meteorological Agency and Earthquake Early Warning API.

[1136] Step 4:

[1137] Server: Integrates the acquired earthquake information and location information.

[1138] Step 5:

[1139] Server: Inputs the integrated data into the generative model.

[1140] Step 6:

[1141] Generative model: Estimates the safety of user A and returns the result to the server.

[1142] Step 7:

[1143] Server: Based on the safety estimation result, it is determined that User A is in a dangerous situation.

[1144] Step 8:

[1145] Server: Sends safety information to User A's emergency contacts. This notification includes location information and dangerous situations.

[1146] Step 9:

[1147] Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue organizations.

[1148] Example 2: Shelter situation report and emotion recognition

[1149] Step 1:

[1150] User: User B takes a photo of the situation at the evacuation shelter with his smartphone.

[1151] Step 2:

[1152] User: Uploads captured images and videos to the server.

[1153] Step 3:

[1154] Server: Receives uploaded image and video data.

[1155] Step 4:

[1156] Server: Passes the received data to the emotion engine.

[1157] Step 5:

[1158] Emotion engine: Analyzes user B's emotions from images and videos and generates their emotional state.

[1159] Step 6:

[1160] Server: Integrates the analysis results from the emotion engine with other data.

[1161] Step 7:

[1162] Server: Sends a notification to emergency contacts based on the integrated detailed information of User B.

[1163] Step 8:

[1164] Server: Posts the results of the emotion analysis and the situation at the evacuation center to the official social media account.

[1165] Step 9:

[1166] Server: If necessary, request relief supplies for evacuation centers.

[1167] By incorporating emotion recognition, it is possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[1168] Example 2

[1169] 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."

[1170] While conventional disaster prevention systems can collect user location information and disaster information, they have difficulty integrating this information to estimate the user's safety in real time and recognize emotions. Furthermore, they lack the ability to respond quickly when a user is deemed to be in danger, and they lack support that takes into account the emotional state of the victim, limiting the efficiency and effectiveness of rescue operations.

[1171] 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.

[1172] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data with an emotion engine, means for notifying emergency contacts based on the analysis results of the emotion engine, means for the analytical model to reassess the damage situation from the images and videos, and means for transmitting information and requesting rescue based on the analysis results. This enables fast and accurate rescue and support by combining user safety estimation and emotion recognition.

[1173] "User terminal" refers to an electronic device that is owned by a user and is capable of acquiring and transmitting location information and media data.

[1174] "Current location" refers to real-time geographical location information obtained by the user terminal using GPS.

[1175] "Server" refers to a central processing unit that receives information from user terminals and manages databases and performs data analysis.

[1176] "Weather data" refers to real-time weather information provided by the Japan Meteorological Agency and related organizations.

[1177] "Disaster data" refers to real-time damage status and disaster-related information obtained when disasters such as earthquakes and floods occur.

[1178] A "generative model" refers to a machine learning algorithm for estimating a user's safety based on their location and disaster information.

[1179] "Safety information" refers to information about the safety of a user estimated by a generative model.

[1180] "Notification" refers to the action of sending information about a user's safety or other important information to relevant people or organizations.

[1181] "Images and videos" refers to visual media data that users take and upload to a server.

[1182] "Emotion engine" refers to software that analyzes a user's emotional state from uploaded images and videos.

[1183] "Emergency Contacts" means the people or organizations that a User has registered to be notified in the event of an emergency.

[1184] "Analysis model" refers to a machine learning algorithm for reassessing the damage situation from uploaded media data.

[1185] "Requesting help" refers to an action of requesting help from an appropriate rescue organization when the user is determined to be in a dangerous situation.

[1186] "Information dissemination" refers to the action of providing reliable information based on the analysis results to related institutions and official social media accounts.

[1187] This invention relates to a disaster prevention AI platform for providing prompt and accurate rescue and support in the event of a disaster. This system acquires location information from the user's device, and the server integrates this information with disaster information and inputs it into an analysis model to estimate the user's safety. Furthermore, an emotion engine performs emotion analysis from images and videos uploaded by the user, allowing for a detailed understanding of the damage situation.

[1188] Data collection

[1189] Location information acquisition

[1190] The user's device periodically acquires the user's current location using the built-in GPS function and sends that information to the server. This location information is captured every 30 seconds using the location information service in the user's smartphone and uploaded to the server via HTTPS communication.

[1191] Disaster information acquisition

[1192] The server obtains real-time disaster information using the Japan Meteorological Agency and disaster alert APIs. This information is sent to the API at regular intervals, and the obtained data is parsed in JSON format and stored in a database.

[1193] Data analysis

[1194] Data Integration

[1195] The server combines user location information and disaster information to create a dataset to input into the generative AI model. This dataset retrieves the necessary information from a database on the server and is synchronized based on timestamps.

[1196] Estimated safety

[1197] The generative AI model estimates the user's safety based on their location and the scale of the disaster. The generative model uses a machine learning algorithm to perform real-time analysis and determine the user's safety status. This allows for a quick assessment of whether the user is "safe" or "at risk."

[1198] Safety registration and rescue request

[1199] Safety assessment

[1200] The server determines whether the user is safe or unsafe based on the estimation results of the generative model and records the result in a database.

[1201] Safety notification

[1202] If a user is deemed to be in danger, the server automatically sends a notification to emergency contacts via SMS or email, containing the user's current location and details of the disaster.

[1203] Rescue request

[1204] If necessary, the server will send an emergency call to the appropriate rescue agency, including detailed location and status, via the rescue agency's API endpoint.

[1205] emotion recognition

[1206] Media Upload

[1207] Users take photos of disaster areas and evacuation shelters with their smartphones and upload the media files to the server, which sends the files via HTTPS.

[1208] Emotion analysis

[1209] The server passes the uploaded images and videos to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the images and videos and returns the analysis results to the server. Based on this analysis result, the user's emotional state is stored in a database.

[1210] Emotion information notification

[1211] Based on the results of the emotion analysis, the server notifies emergency contacts of the user's emotional state and safety information.

[1212] Situation Analysis

[1213] Media Analysis

[1214] The server uses the emotion engine's analysis results to reassess the damage situation in more detail using an analytical model. The analytical model reanalyzes the detailed situation in the affected areas based on the data it receives and generates a report.

[1215] Information dissemination

[1216] Information aggregation

[1217] The server combines the analysis results with data from other reliable sources to create a single information set.

[1218] Information dissemination

[1219] The server sends reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and timely support. It posts information to official accounts using social media APIs and sends notifications to related organizations via email and dedicated APIs.

[1220] Specific examples

[1221] Rescue operations following an earthquake

[1222] Because User A is in an area close to the epicenter, his location information is acquired via GPS and sent to the server. The server combines the acquired location information with real-time earthquake information and inputs the data into a generative model to estimate User A's safety. The generative model estimates that User A is in danger and returns the result to the server. Since the server has determined that User A is in danger, it sends safety information to User A's emergency contacts and issues an emergency rescue request to nearby rescue organizations.

[1223] Confirmation of evacuees with a sense of despair

[1224] User B takes a photo of the evacuation shelter situation with their smartphone and uploads it to the server. The server passes the received photo to an emotion engine, which analyzes the image to determine User B's sense of despair and recognizes their emotional state. Based on the emotion analysis results, the server notifies emergency contacts of User B's emotional state and the evacuation shelter situation, and also posts accurate information to official social media accounts and requests relief supplies if necessary.

[1225] By incorporating emotion recognition, it is possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[1226] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1227] Step 1:

[1228] Location information acquisition

[1229] The user's terminal periodically obtains its current location using the GPS function.

[1230] Input: GPS satellite signal

[1231] Data processing: Obtaining the longitude and latitude information of the current location

[1232] Output: Acquired location data

[1233] Specific operation: The location information service in the smartphone runs, periodically capturing the current location, encrypting that information, and sending it to the server.

[1234] Step 2:

[1235] Location information transmission

[1236] The user's terminal transmits the acquired location information to the server.

[1237] Input: Location data

[1238] Data Computation: Location Encryption

[1239] Output: Encrypted location data

[1240] Specific operation: The device's transmission module is activated and uploads encrypted location information to the server via HTTPS communication.

[1241] Step 3:

[1242] Disaster information acquisition

[1243] The server obtains real-time disaster information from the Japan Meteorological Agency and disaster alert APIs.

[1244] Input: API request

[1245] Data processing: JSON parsing of acquired disaster information

[1246] Output: Disaster information data

[1247] Specific operation: The server periodically sends requests to the API endpoint, obtains the returned disaster information, and stores it in a database.

[1248] Step 4:

[1249] Data Integration

[1250] The server integrates location information and disaster information to create a dataset.

[1251] Input: User location information and disaster information data

[1252] Data Computation: Data Synchronization and Integration

[1253] Output: Unified dataset

[1254] Specific operation: The server retrieves location information and disaster information from the database and creates an integrated dataset based on that data.

[1255] Step 5:

[1256] Estimated safety

[1257] The server inputs the integrated data set into a generative AI model to estimate the user's safety.

[1258] Input: Integrated dataset

[1259] Data calculation: Estimation of user safety (application of machine learning algorithms)

[1260] Output: Safety estimation result

[1261] Specific operation: A dataset is input into the generative AI model, which uses a neural network to estimate the user's safety and status and returns the results to the server.

[1262] Step 6:

[1263] Safety assessment

[1264] The server determines whether the user is safe or dangerous based on the estimation results of the generative model.

[1265] Input: Safety estimation result

[1266] Data calculation: Determining safety status based on conditions

[1267] Output: Safety status (safe or dangerous)

[1268] Specific operation: The server analyzes the estimation results and determines the user's safety status according to the defined conditions.

[1269] Step 7:

[1270] Safety notification

[1271] The server sends a notification to emergency contacts if the user is determined to be at risk.

[1272] Input: Safety status, user details

[1273] Data operation: generating notification messages

[1274] Output: Notification message

[1275] Specific operation: The server sends a notification message to emergency contacts via SMS or email gateway.

[1276] Step 8:

[1277] Rescue request

[1278] The server will send an emergency rescue request to a rescue organization if necessary.

[1279] Input: User's detailed location information, disaster situation

[1280] Data calculation: Generating a help-call message

[1281] Output: Help request message

[1282] What happens: The server sends a request to the rescue agency's API endpoint, issuing an emergency rescue request including the user's details.

[1283] Step 9:

[1284] Media Upload

[1285] Users upload photos and videos of the disaster area to the server.

[1286] Input: Image and video data

[1287] Data Processing: Sending Media Files

[1288] Output: Media data stored on the server

[1289] Specific operation: A user takes a media file with their smartphone camera and uploads it to a server via HTTPS.

[1290] Step 10:

[1291] Emotion analysis

[1292] The server passes the uploaded media files to the emotion engine for emotion analysis.

[1293] Input: Media files

[1294] Data calculation: Emotional state analysis (application of image analysis algorithms)

[1295] Output: Emotion analysis results

[1296] Specific operation: The server sends media files to the emotion engine, which analyzes the user's emotional state from images and videos and returns the results to the server.

[1297] Step 11:

[1298] Emotion information notification

[1299] The server notifies emergency contacts of the user's emotional state and safety information based on the emotion analysis results.

[1300] Input: Emotion analysis results, safety information

[1301] Data operation: generating notification messages

[1302] Output: Notification message

[1303] Specific operation: The server generates a notification message containing the results of the sentiment analysis and sends it to emergency contacts via SMS or email.

[1304] Step 12:

[1305] Media Analysis

[1306] Based on the analysis results of the emotion engine, the server reevaluates the damage situation in more detail using an analytical model.

[1307] Input: Sentiment analysis results and other media data

[1308] Data calculation: Reassessment of damage situation (application of machine learning algorithms)

[1309] Output: Damage report

[1310] Specific operation: The server inputs data into the analytical model and generates a report to reassess the damage situation.

[1311] Step 13:

[1312] Information aggregation

[1313] The server integrates the analysis results with data from other reliable sources.

[1314] Input: Analysis results, data from other sources

[1315] Data Calculation: Data Integration

[1316] Output: Integrated information set

[1317] Specific operation: The server merges information obtained from multiple data sources and performs consistency checks.

[1318] Step 14:

[1319] Information dissemination

[1320] The server will send out reliable information to official social media accounts and related organizations.

[1321] Input: Integrated information set

[1322] Data operations: generating information messages

[1323] Output: Published information message

[1324] Specific operations: Post information to official accounts using SNS APIs, and send notifications to relevant organizations via email or dedicated APIs.

[1325] (Application example 2)

[1326] 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."

[1327] Currently, there are limited systems that can provide rapid and accurate rescue and assistance when disasters occur. In particular, there are no systems that can grasp the emotional state of victims and respond appropriately. Furthermore, technology to effectively utilize autonomous vehicles to carry out rapid rescue operations has not yet been fully developed. An efficient system to solve these issues is needed.

[1328] 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.

[1329] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for acquiring weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, means for transmitting information and requesting rescue based on the analysis results, means for acquiring the current location of the autonomous vehicle and calculating an optimal route, and means for recognizing the emotions of disaster victims and taking action based on their emotional state. This enables rapid and accurate rescue operations and appropriate responses based on the emotional state of disaster victims.

[1330] A "user terminal" is an electronic device owned by a user that is used to obtain current location information.

[1331] The "current location" is the geographical location information of the user terminal.

[1332] "Means" are methods or technical elements for performing a specific function.

[1333] A "server" is a computer system that can receive and process data sent from a user terminal.

[1334] "Weather data" is information relating to weather conditions and changes.

[1335] "Disaster data" refers to information related to disasters such as earthquakes, floods, and fires.

[1336] A "generative model" is an algorithm or machine learning model that analyzes acquired data and makes certain predictions or inferences.

[1337] "Safety" refers to the status of whether the user is in a safe state or in a dangerous state.

[1338] "Safety information" is information relating to the safety status of the user.

[1339] "Notification" is the act of transmitting specific information to other terminals or systems.

[1340] "Images and videos" are still images and videos taken by the user.

[1341] An "analytical model" is an algorithm or data processing method for analyzing collected data and extracting specific information.

[1342] A "request for rescue" is an act of requesting the support necessary to carry out rescue operations.

[1343] An "autonomous vehicle" is a vehicle that can be driven automatically without human operation.

[1344] A "route" is a path from a specific point to a destination.

[1345] "Calculation" means performing the necessary calculations to obtain the results.

[1346] An "emotion" is a psychological state such as joy, sadness, or fear felt by a user.

[1347] "Responding" means acting in a way that is adapted to a particular situation.

[1348] The system for realizing the present invention comprises the following elements.

[1349] 1. Data Collection

[1350] The user device periodically acquires its current location using its GPS function and sends that location information to the server. It also acquires media data (images and videos) using its camera and microphone to recognize the user's emotions, and uploads them to the server.

[1351] 2. Disaster Information Collection

[1352] The server retrieves real-time weather and disaster data from the Japan Meteorological Agency and earthquake early warning APIs, stores this information in a database, and integrates the location information and disaster data to create a dataset to be input into the generative model.

[1353] 3. Analysis using generative models

[1354] The generative model stored on the server estimates the user's safety in real time based on the user's location information and disaster data. The generative model uses a machine learning algorithm to perform real-time analysis and determine safety information.

[1355] 4. Notification of safety information and rescue requests

[1356] Based on the predictions returned by the generative model, the server determines whether the user is safe or in danger, and sends a notification to emergency contacts as necessary. It also obtains the current location of the autonomous vehicle, calculates the optimal route, and dispatches it to the disaster area.

[1357] 5. Media Data Analysis and Emotion Recognition

[1358] Uploaded image and video data is analyzed by an analysis model on the server. The emotion recognition engine analyzes the user's emotions from the images and videos and generates a detailed emotional state. Based on the analysis results, the server notifies emergency contacts of the user's emotional state and situation.

[1359] 6. Relief operations management

[1360] The autonomous vehicle will then follow the calculated optimal route to the disaster area and carry out rescue operations. During the rescue operations, further media data will be acquired and analyzed, and the server will issue a rescue request if necessary.

[1361] Hardware and software used

[1362] GPS module: Obtains the current location information of the user device.

[1363] Camera and microphone: Capture images and videos to recognize user emotions.

[1364] Server: Collects, integrates, analyzes, and notifies data.

[1365] Japan Meteorological Agency API and Earthquake Early Warning API: Obtain real-time disaster data.

[1366] Generative model: Estimates safety using machine learning algorithms.

[1367] Emotion recognition engine: Analyzes emotions from images and videos.

[1368] Specific examples

[1369] If the user is in an area affected by an earthquake, the user's device will use GPS to obtain location information and send it to the server. The server will then collect disaster data via an earthquake early warning API and use a generative model to estimate the user's safety. Based on the estimation results, a notification will be sent to emergency contacts and an autonomous vehicle will be dispatched along the optimal route. After arriving, the camera will analyze the user's facial expressions and provide any necessary psychological care.

[1370] Prompt Sentence Examples

[1371] "Please design a system that obtains detailed location information of users who are currently near the epicenter and need to evacuate, and sets the optimal rescue route. Also, after arriving, please use a camera to analyze the emotions of the victims and provide the necessary support."

[1372] This will enable quick and accurate rescue operations and appropriate responses based on the emotional state of the victims.

[1373] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1374] Step 1:

[1375] Obtaining location information of user devices

[1376] The user terminal periodically obtains its current location using a GPS module.

[1377] Input: Location data from the GPS module

[1378] Output: Current location of the user

[1379] Specific operation: The user device obtains its current location from GPS every few seconds and stores the data in memory.

[1380] Step 2:

[1381] Sending location information

[1382] The user terminal transmits the acquired location information to the server.

[1383] Input: current location of the user

[1384] Output: Location information sent to the server

[1385] Specific operation: The user device sends location information to the server's API endpoint as an HTTP request.

[1386] Step 3:

[1387] Obtaining disaster information

[1388] The server obtains weather and disaster data from the Japan Meteorological Agency API and Earthquake Early Warning API.

[1389] Input: API request

[1390] Output: Real-time disaster data

[1391] Specific operation: The server periodically sends requests to the API endpoint to obtain the latest disaster information and stores it in the database.

[1392] Step 4:

[1393] Data integration

[1394] The server integrates the acquired location information and disaster data to create a dataset to be input into the generative model.

[1395] Input: User's current location information, disaster data

[1396] Output: The dataset to feed into the generative model

[1397] Specific operation: The server combines location information and disaster data and converts them into a format that can be analyzed by the generative model.

[1398] Step 5:

[1399] Safety estimation using generative models

[1400] The generative model estimates the user's safety based on the user's location information and disaster data.

[1401] Input: Integrated dataset

[1402] Output: Safety estimation result

[1403] Specific operation: The generative model processes the integrated dataset and estimates the user's safety in real time.

[1404] Step 6:

[1405] Registration and notification of safety information

[1406] Based on the estimation results, the server registers the user's safety information and notifies emergency contacts.

[1407] Input: Safety estimation result

[1408] Output: Safety information sent to emergency contacts

[1409] Specific operation: The server stores the safety information in a database and notifies emergency contacts via email or SMS.

[1410] Step 7:

[1411] Uploading media data

[1412] Users take photos and videos of disaster areas and evacuation shelters and upload them to the server.

[1413] Input: Image and video data

[1414] Output: Media data stored on the server

[1415] Specific operation: The user uses the device's camera to capture media data and sends it to the server through the application.

[1416] Step 8:

[1417] Media Data Analysis

[1418] The server analyzes the uploaded image and video data using an analytical model.

[1419] Input: Media data

[1420] Output: Analysis results

[1421] Specific operation: The server uses an emotion recognition engine to analyze the media data and recognize the emotional state of the victim.

[1422] Step 9:

[1423] Autonomous vehicle routing

[1424] The server obtains the current location of the autonomous vehicle and calculates the optimal route.

[1425] Input: current vehicle location, disaster information

[1426] Output: Optimal route

[1427] Specific operation: The server obtains GPS data from the autonomous vehicle and calculates the optimal rescue route based on disaster information.

[1428] Step 10:

[1429] Carrying out rescue operations

[1430] The autonomous vehicle will follow the calculated optimal route to the disaster area and carry out rescue operations.

[1431] Input: Optimal Route

[1432] Output: Rescue operations in disaster areas

[1433] Specific operation: The autonomous vehicle uses the navigation system to drive along the specified route and begins rescue operations upon arrival.

[1434] Step 11:

[1435] Emotional information notification

[1436] Based on the emotion analysis results, the server notifies the user's emotional state and situation to the emergency contact.

[1437] Input: Analysis results

[1438] Output: Emotion information sent to emergency contacts

[1439] Specific operation: The server notifies the emergency contact of the analysis results via email or SMS.

[1440] This will enable quick and accurate rescue operations and appropriate responses based on the emotional state of the victims.

[1441] 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.

[1442] 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.

[1443] 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.

[1444] [Third embodiment]

[1445] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1446] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1447] 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).

[1448] 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.

[1449] 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.

[1450] 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).

[1451] 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.

[1452] 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.

[1453] 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.

[1454] 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.

[1455] 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.

[1456] 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."

[1457] This invention provides a disaster prevention AI platform that enables rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes them using a generative model to estimate the user's safety, and automatically carries out necessary rescue and support.

[1458] Explaining program processing in natural language

[1459] Data collection

[1460] 1. Obtain location information

[1461] Device: The user's smartphone periodically obtains its current location using its GPS function and sends that location information to the server.

[1462] 2. Disaster information acquisition

[1463] Server: Real-time disaster information is obtained from the Japan Meteorological Agency and earthquake early warning APIs. This information is stored in a database on the server and linked to the user's location information.

[1464] Data analysis

[1465] 1. Data Integration

[1466] Server: Integrates the acquired user location information and disaster information to create a dataset to be input into the generative model.

[1467] 2. Safety Estimation

[1468] Generative model: Estimates the user's safety based on the user's location information, the scale of the disaster, etc. The generative model uses machine learning algorithms to perform real-time analysis and determine the user's safety status.

[1469] Safety registration and rescue request

[1470] 1. Safety assessment

[1471] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[1472] 2. Safety Notification

[1473] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[1474] 3. Request for help

[1475] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[1476] Situation Analysis

[1477] 1. Media Upload

[1478] Users can take photos and videos of disaster areas and evacuation centers and upload them to the server.

[1479] 2. Media Analysis

[1480] Server: Receives images and videos uploaded by users and passes them to the analysis model.

[1481] Analysis model: Uses machine learning to analyze damage and people's movements from images and videos, and generates detailed reports.

[1482] Information dissemination

[1483] 1. Information aggregation

[1484] Server: Integrates analytical results with data from other reliable sources.

[1485] 2. Information dissemination

[1486] Server: Disseminates reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and prompt support information.

[1487] Specific examples

[1488] Example 1: Rescue operations after an earthquake

[1489] 1. Device: User A is in an area close to the epicenter, so he obtains his location information using GPS and sends it to the server.

[1490] 2. Server: Integrates location information with earthquake information obtained from the Japan Meteorological Agency.

[1491] 3. Server: Input the integrated data into the generative model and estimate the safety of user A.

[1492] 4. Generative model: Determines that user A is in a dangerous situation and returns the result to the server.

[1493] 5. Server: Since the user is judged to be in danger, safety information is sent to User A's emergency contact.

[1494] 6. Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue agencies.

[1495] Example 2: Shelter situation report

[1496] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads it to the server.

[1497] 2. Server: Passes the received photos to the analysis model.

[1498] 3. Analysis model: Analyzes crowded evacuation shelters from images and generates a report.

[1499] 4. Server: Based on the analysis report, the system posts information about the congestion situation at evacuation centers and the supplies needed to the official social media accounts.

[1500] 5. Server: Requests relief supplies based on reliable information.

[1501] This will help eliminate false information and rumors and quickly provide the rescue and assistance that is truly needed in affected areas.

[1502] The processing flow will be explained below.

[1503] Data collection

[1504] Step 1:

[1505] Device: The user's smartphone uses the GPS function to obtain the current location.

[1506] Step 2:

[1507] Device: Sends the acquired location information to the server.

[1508] Step 3:

[1509] Server: Obtains real-time disaster information from the Japan Meteorological Agency and earthquake early warning APIs.

[1510] Step 4:

[1511] Server: The acquired disaster information is stored in a database and linked to the user's location information.

[1512] Data analysis

[1513] Step 5:

[1514] Server: Integrates user location information and disaster information to create a dataset.

[1515] Step 6:

[1516] Server: Inputs the integrated data into the generative model.

[1517] Step 7:

[1518] Generative model: Estimates the user's safety based on the user's location information and the scale of the disaster.

[1519] Step 8:

[1520] Generative model: Returns the inference results to the server.

[1521] Safety registration and rescue request

[1522] Step 9:

[1523] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[1524] Step 10:

[1525] Server: If a user is deemed to be at risk, a notification is automatically sent to pre-registered emergency contacts.

[1526] Step 11:

[1527] Server: If necessary, initiates an emergency rescue request to the appropriate rescue agency.

[1528] Situation Analysis

[1529] Step 12:

[1530] User: Take photos and videos of disaster areas and evacuation centers and upload them to the server.

[1531] Step 13:

[1532] Server: Receives uploaded image and video data.

[1533] Step 14:

[1534] Server: Passes the received data to the analysis model.

[1535] Step 15:

[1536] Analysis model: Analyzes damage conditions and people's movements from images and videos, and generates detailed reports.

[1537] Information dissemination

[1538] Step 16:

[1539] Server: Integrates analytical results with data from other reliable sources.

[1540] Step 17:

[1541] Server: Disseminates reliable information to official social media accounts and related organizations.

[1542] Step 18:

[1543] Server: If necessary, requests for relief supplies and the current situation at evacuation centers will be sent out.

[1544] Example 1

[1545] 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."

[1546] Conventional disaster prevention systems only collect user location information and disaster information separately, making it difficult to integrate these information to estimate the user's safety in real time and provide prompt and accurate rescue and support. Furthermore, the lack of proper aggregation of disaster information and the dissemination of reliable information has led to the spread of false information, causing confusion in the affected areas. A new system is needed to solve these problems.

[1547] 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.

[1548] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation result, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, and means for the server to transmit information and request rescue based on the analysis result to official social media accounts and related organizations, thereby enabling rapid and accurate rescue and support when a disaster occurs.

[1549] A "user terminal" refers to an electronic device that can be carried by a user, specifically a device equipped with a GPS function such as a smartphone or tablet.

[1550] "Current location" refers to information about the user's current location obtained by the user's device using the GPS function.

[1551] A "server" is a high-performance computer system connected to a network, which receives data from multiple users and devices and processes and stores it.

[1552] "Weather data" refers to weather-related data provided by the Japan Meteorological Agency or weather services, specifically information such as temperature, wind speed, and precipitation.

[1553] "Disaster data" refers to information about natural disasters such as earthquakes, typhoons, and floods, and specifically includes the date and time of occurrence, location, and extent of impact.

[1554] A "generative model" refers to a computer program that uses machine learning algorithms to estimate a user's safety.

[1555] "Safety" refers to whether the user is currently safe or in imminent danger.

[1556] "Notification" refers to alert information sent from the system to users and emergency contacts, specifically by email or SMS.

[1557] "Images and videos" refers to visual data captured by users and uploaded to the system.

[1558] An "analysis model" refers to a program that uses machine learning algorithms to analyze uploaded images and videos and assess the extent of damage and people's movements.

[1559] "Official social media account" refers to a highly reliable social media account that the system uses to disseminate information.

[1560] "Related organizations" refer to official organizations that carry out disaster prevention and rescue operations, including fire departments, police departments, and medical institutions.

[1561] "Request for Assistance" means a request sent to relevant authorities to carry out rescue operations required in the event of a disaster.

[1562] This invention is a disaster prevention AI platform that enables rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes them using a generative AI model to estimate the user's safety, and automatically carries out necessary rescue and support.

[1563] Hardware and software used

[1564] User device: A portable electronic device with a GPS function, such as a smartphone or tablet, is used. This device can acquire the user's current location and send it to the server.

[1565] Server: A high-performance computer system is used to receive, process, and store data from multiple users and devices. It also collects real-time weather and disaster data from the Japan Meteorological Agency and earthquake early warning APIs.

[1566] Generative model: Using an AI model that uses a machine learning algorithm, the safety of a user is estimated in real time based on the user's location information and disaster information.

[1567] Analysis Model: Using machine learning, the model analyzes damage from images and videos and generates a detailed report based on media files uploaded by users.

[1568] Program processing

[1569] 1. Location information acquisition: The user device uses the built-in GPS function to acquire the user's current location at specific intervals, encrypts the data, and sends it to the server.

[1570] 2. Disaster information acquisition: The server acquires disaster information in real time from the Japan Meteorological Agency and earthquake early warning APIs and stores this information in a database.

[1571] 3. Data integration: The server integrates user location information and disaster information to create a dataset to be input into the generative AI model.

[1572] 4. Safety Prediction: The generative model estimates the user's safety based on the integrated data. For example, it determines whether the user is in danger based on the scale of the disaster, the user's location, and past data.

[1573] 5. Safety determination: The server determines whether the user is "safe" or "dangerous" based on the inference results returned by the generative model.

[1574] 6. Safety Notification: If a user is deemed to be in danger, the server automatically sends a notification to the user's emergency contacts, including the user's location and status.

[1575] 7. Rescue Request: If the user is deemed to be in danger, the server will issue an emergency rescue request to the appropriate rescue agency. This request will include the user's detailed location information and the disaster situation.

[1576] 8. Media Upload: Users can take photos and videos of disaster areas and evacuation centers with their smartphones and upload the media files to the server.

[1577] 9. Media Analysis: The server receives images and videos sent by users, inputs them into an analytical model, and generates a detailed report assessing the damage situation and people's movements.

[1578] 10. Information aggregation: The server integrates the analysis results with data from other reliable sources to provide comprehensive information aggregation.

[1579] 11. Information Dissemination: The server will disseminate information based on reliable information and analysis results to official social media accounts and related organizations, thereby eliminating false information and providing accurate and prompt support information.

[1580] Examples of specific examples and prompts

[1581] Example 1: Rescue operations after an earthquake

[1582] 1. Device: User A is in an area close to the epicenter, so his smartphone uses the GPS function to obtain location information every five minutes.

[1583] 2. Server: The server obtains information on the epicenter from the Japan Meteorological Agency API and stores the data in a database in real time.

[1584] 3. Server: Integrates user A's location information and disaster information about the epicenter, and creates a dataset to be input into the generative AI model.

[1585] 4. Generative model: Based on the integrated data, determine that user A is in a dangerous situation.

[1586] 5. Server: Based on the results returned from the generative model, User A is determined to be “dangerous.”

[1587] 6. Server: Notifies User A's emergency contacts of their location and safety status. For example, send an SMS message saying, "User A is near the epicenter and in a dangerous situation."

[1588] 7. Server: Sends a similar emergency rescue request to nearby rescue organizations.

[1589] Example 2: Reporting on the situation at an evacuation shelter

[1590] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads the photo to the server.

[1591] 2. Server: The received photos are input into the analysis model to analyze the congestion situation at the evacuation center.

[1592] 3. Analytical model: Determines the level of congestion at evacuation centers from images and generates a detailed report.

[1593] 4. Server: Stores the analysis results in a database and integrates them with data from other sources.

[1594] 5. Server: Posts information about the congestion situation at evacuation centers and necessary supplies to the official social media account. Also, conveys the information to appropriate organizations to provide prompt assistance.

[1595] Prompt Sentence Examples

[1596] "User A is near the epicenter of an earthquake, so his location information is obtained using GPS and sent to a server. Please explain a program that uses a generative model to estimate the user's safety and status, and sends a notification to emergency contacts based on the results."

[1597] This system is designed to enable rapid and accurate rescue and support in the event of a disaster, and can ensure the safety of users through multi-layered data analysis and information dissemination.

[1598] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1599] Step 1:

[1600] Location information acquisition

[1601] Device: The user device (smartphone) periodically acquires its current location using its built-in GPS function. The acquired location information is sent to the server as latitude and longitude data.

[1602] Input: Location information from the GPS sensor

[1603] Output: Latitude and longitude data of the current location sent to the server

[1604] Step 2:

[1605] Disaster information acquisition

[1606] Server: The server collects disaster information in real time from the Japan Meteorological Agency and earthquake early warning APIs. This information provides information such as the type of disaster, the time of occurrence, scale, and extent of damage.

[1607] Input: Disaster information from the Japan Meteorological Agency and Earthquake Early Warning API

[1608] Output: Disaster information stored on the server (type, time of occurrence, scale, extent of damage, etc.)

[1609] Step 3:

[1610] Data Integration

[1611] Server: The server combines the user's location information and disaster information to create a dataset to input into the generative model. Specifically, it matches the user's latitude and longitude with disaster information in that area.

[1612] Input: User location information and disaster information

[1613] Output: A unified dataset to feed into the generative model

[1614] Step 4:

[1615] Estimated safety

[1616] Generative model: The generative model estimates the user's safety based on the integrated dataset. For example, if the user's location is close to the epicenter or if the area is hit by strong shaking, the model estimates that the user is in danger.

[1617] Input: Integrated dataset

[1618] Output: User's safety status (safe or unsafe)

[1619] Step 5:

[1620] Safety assessment

[1621] Server: Based on the safety estimation results obtained from the generative model, the server determines whether the user is safe or at risk. If the user is determined to be at risk, the server proceeds to the next step.

[1622] Input: Safety estimation results from the generative model

[1623] Output: User safety assessment result (safe or dangerous)

[1624] Step 6:

[1625] Safety notification

[1626] Server: If the user is deemed to be in danger, the server automatically sends a notification to pre-registered emergency contacts, including the user's current location and the reason for the danger.

[1627] Input: User safety assessment results, emergency contact information

[1628] Output: Safety notification sent to emergency contacts (including location and details of the situation)

[1629] Step 7:

[1630] Rescue request

[1631] Server: If the user is determined to be in danger, the server sends an emergency rescue request to the appropriate rescue agency, including the user's location and the disaster situation.

[1632] Input: User safety assessment results, location information, disaster information

[1633] Output: Emergency rescue request sent to rescue agencies, including user's detailed location and disaster situation.

[1634] Step 8:

[1635] Media Upload

[1636] User: The user takes photos of the disaster area or the situation at the evacuation center with their smartphone and uploads the images and videos to the server.

[1637] Input: Images and videos taken by the user

[1638] Output: Media files uploaded to the server

[1639] Step 9:

[1640] Media Analysis

[1641] Server: The server passes the received images and videos to the analysis model, which analyzes the damage situation and people's movements. The analysis model uses machine learning algorithms to generate a detailed report.

[1642] Input: Images and videos uploaded to the server

[1643] Output: Report of damage status based on analytical model

[1644] Step 10:

[1645] Information aggregation

[1646] Server: The server integrates the analysis results with data from other reliable sources to aggregate comprehensive information, allowing for a comprehensive understanding of the situation in the affected areas.

[1647] Input: Analysis model reports, data from other sources

[1648] Output: Aggregated comprehensive information

[1649] Step 11:

[1650] Information dissemination

[1651] Server: Based on reliable information, the server will disseminate important information to official social media accounts and relevant organizations, thereby eliminating false information and providing prompt assistance.

[1652] Input: Aggregated comprehensive information

[1653] Output: Information posted on official social media accounts and related organizations

[1654] In this way, each step works in coordination, enabling rapid and accurate rescue and support in the event of a disaster.

[1655] (Application example 1)

[1656] 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."

[1657] When natural disasters occur, it is difficult to quickly and accurately collect and analyze information, which often leads to delays in rescue and assistance to victims. Furthermore, normal services may be suspended during disasters, and the disruption of services such as food delivery can have a significant impact on the lives of victims. Furthermore, insufficient means exist for confirming the safety of victims, notifying emergency contacts, and requesting assistance, hindering rapid and efficient rescue efforts.

[1658] 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.

[1659] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, means for transmitting information and requesting rescue based on the analysis results, and means for suspending or resuming specific services depending on the user's safety status during a disaster. This enables rapid and accurate information collection and analysis, accelerating rescue and support for disaster victims and reducing the risk of service suspension during a disaster.

[1660] A "user terminal" is a device such as a mobile terminal or smartphone that is owned by a user and has the function of acquiring location information.

[1661] "Current location" is information that is detected in real time by the user terminal and indicates the user's current geographical location.

[1662] A "server" is a computer system that receives, processes, and stores data over a network and works in conjunction with other devices.

[1663] "Weather data" refers to information about weather, temperature, precipitation, and other meteorological information provided by meteorological agencies.

[1664] "Disaster data" refers to information about natural disasters such as earthquakes, floods, and typhoons, and includes the date, time, location, and scale of the disaster.

[1665] A "generative model" is a data model that uses algorithms based on machine learning or deep learning to analyze input data and make predictions or inferences.

[1666] "Safety" refers to the status of the user indicating whether they are safe or in danger in a disaster situation.

[1667] The "estimated result" is information indicating the safety status of the user obtained as a result of analysis by the generative model.

[1668] "Safety information" is data related to the user's safety status, and is notified to emergency contacts and relevant organizations.

[1669] "Means for notifying" refers to the method or technology for sending messages or alerts to user terminals or emergency contacts.

[1670] "Images and videos" are visual media data that users take and upload to a server.

[1671] An "analysis model" is an algorithm or data model that uses machine learning and artificial intelligence technologies to analyze images and videos and extract useful information.

[1672] A "request for assistance" is a request made to relevant agencies to request emergency assistance or rescue operations.

[1673] "Means for suspending or resuming services" refers to methods or techniques for temporarily suspending specific services during a disaster and resuming them after the situation has improved.

[1674] This invention is a disaster prevention support system consisting of a user device, a server, and a generative model. Specifically, the user device acquires its current location using its GPS function and transmits it to the server. The server collects meteorological and disaster data in real time and integrates it with the user's location information. To analyze the shared data, the generative model uses a machine learning algorithm to estimate the user's safety.

[1675] The server does the following:

[1676] 1. Obtaining and sending location information:

[1677] User devices such as smartphones periodically use their GPS to obtain their current location and send it to a server. The Geopy library is used to obtain accurate location information.

[1678] 2. Meteorological and disaster data collection:

[1679] The server obtains real-time information using weather data and disaster data APIs, and stores the current weather conditions and disaster occurrence information in a database.

[1680] 3. Data integration and generative model use:

[1681] The user's location information and acquired disaster data are combined and input into a generative model, which uses machine learning algorithms to estimate the user's safety status. The generative model uses frameworks such as TensorFlow and PyTorch.

[1682] 4. Safety confirmation and notification:

[1683] Based on the inference results from the generative model, the system registers the user's safety information and notifies emergency contacts. If the user's safety is judged to be in danger, an emergency notification is automatically sent.

[1684] 5. Emergency relief request:

[1685] The server automatically sends an appropriate rescue request based on the analysis results, which includes detailed location information and current situation of the victim.

[1686] 6. Suspending and Resuming Services:

[1687] In the event of a disaster, services such as food delivery will be temporarily suspended and resumed as soon as the safety of users is confirmed. This is done by notifying service providers using a REST API.

[1688] The hardware and software used includes:

[1689] Hardware: User device (smartphone), server

[1690] Software: Geopy, machine learning frameworks (TensorFlow, PyTorch), disaster information API requests, REST API

[1691] Consider the following scenario:

[1692] Example 1: Suspension of food delivery during an earthquake

[1693] 1. Device: If the user is in an earthquake-prone area, the smartphone's GPS acquires location information and sends it to the server.

[1694] 2. Server: Collects earthquake information, integrates it with the acquired location information, and inputs it into the generative model.

[1695] 3. Generative model: If the user is judged to be at risk, the server sends a stop notification to the delivery service.

[1696] 4. Server: Additionally, notify the user's emergency contacts and request appropriate assistance.

[1697] Example 2: Resuming food delivery after safety confirmation

[1698] 1. Device: After the user moves to a safe location, the device acquires location information again and sends it to the server.

[1699] 2. Server: Based on location information and the latest disaster information, the generative model confirms the user's safety.

[1700] 3. Generative model: If the user is deemed safe, the server sends a resume notification to the delivery service.

[1701] Example prompt sentence:

[1702] "Please create a program that determines the availability of food delivery services based on the user's location information and disaster information, and notifies them when the service resumes after confirming safety. Please also include the user's emergency contact information and API endpoints for suspending and resuming the service."

[1703] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1704] Step 1:

[1705] The user terminal acquires the current location and transmits the location information to the server.

[1706] Input: Current location obtained from the smartphone's GPS.

[1707] Data processing: Extract latitude and longitude information using the Geopy library.

[1708] Output: Location information including latitude and longitude of the user.

[1709] Step 2:

[1710] The server obtains weather and disaster data in real time.

[1711] Input: API requests for weather data and disaster information.

[1712] Data processing: Parse the data obtained from the API and extract the necessary information (weather, earthquake occurrence information, etc.).

[1713] Output: Datasets such as current weather, disaster occurrence information, etc.

[1714] Step 3:

[1715] The server integrates the user's location information and disaster data and inputs them into the generative model.

[1716] Input: User location information, weather data, disaster data.

[1717] Data processing: Creating a unified dataset and converting it into an input format for the generative model.

[1718] Output: The synthesized data that is input into the generative model.

[1719] Step 4:

[1720] The generative model estimates the user's safety.

[1721] Input: Integrated data.

[1722] Data calculation: Using machine learning algorithms, the user's safety status is estimated in real time.

[1723] Output: User safety information (safe / dangerous).

[1724] Step 5:

[1725] Based on the estimation results, the server notifies the user's safety information to emergency contacts.

[1726] Input: Estimation results (safety information), user's emergency contact data.

[1727] Data processing: Construct the necessary notification information (estimation results, user location information).

[1728] Output: Notify emergency contacts (send message).

[1729] Step 6:

[1730] The server will call for help if necessary.

[1731] Input: Estimation results, user location information, disaster data.

[1732] Data processing: Package the necessary information to request assistance.

[1733] Output: Request (send message) to appropriate rescue agency.

[1734] Step 7:

[1735] Servers stop or restart food delivery services in the event of a disaster.

[1736] Input: User safety information, service status information.

[1737] Data calculation: Decisions on suspending or resuming services based on safety information.

[1738] Output: Suspend or resume notification (API request) to the food delivery service.

[1739] These steps will ensure prompt relief efforts and proper management of services in the event of a disaster.

[1740] 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.

[1741] This invention provides a disaster prevention AI platform for providing rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes it using a generative model to estimate the user's safety, and automatically carries out necessary rescue and support. In addition, by adding an emotion engine that recognizes the user's emotions, it is possible to grasp the situation of victims in more detail and take appropriate measures.

[1742] Explaining program processing in natural language

[1743] Data collection

[1744] 1. Obtain location information

[1745] Device: The user's smartphone periodically obtains its current location using its GPS function and sends that location information to the server.

[1746] 2. Disaster information acquisition

[1747] Server: Real-time disaster information is obtained from the Japan Meteorological Agency and earthquake early warning APIs. This information is stored in a database on the server and linked to the user's location information.

[1748] Data analysis

[1749] 1. Data Integration

[1750] Server: Integrates user location information and disaster information to create a dataset to be input into the generative model.

[1751] 2. Safety Estimation

[1752] Generative model: Estimates the user's safety based on the user's location information, the scale of the disaster, etc. The generative model uses machine learning algorithms to perform real-time analysis and determine the user's safety status.

[1753] Safety registration and rescue request

[1754] 1. Safety assessment

[1755] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[1756] 2. Safety Notification

[1757] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[1758] 3. Request for help

[1759] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[1760] emotion recognition

[1761] 1. Media Upload

[1762] Users can take photos and videos of disaster areas and evacuation centers and upload them to the server.

[1763] 2. Emotion analysis

[1764] Server: Receives uploaded image and video data and passes it to the emotion engine.

[1765] Emotion Engine: Analyzes user emotions from images and videos and generates detailed emotional states.

[1766] 3. Emotion information notification

[1767] Server: Notifies the user's emotional state along with the estimated safety status to emergency contacts.

[1768] Situation Analysis

[1769] 1. Media Analysis

[1770] Server: The analysis results from the emotion engine are combined and passed to the analysis model.

[1771] Analysis model: Reassess the damage situation and people's movements from images and videos and generate detailed reports.

[1772] Information dissemination

[1773] 1. Information aggregation

[1774] Server: Integrates analytical results with data from other reliable sources.

[1775] 2. Information dissemination

[1776] Server: Disseminates reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and prompt support information.

[1777] Specific examples

[1778] Example 1: Rescue operations after an earthquake

[1779] 1. Device: User A is in an area close to the epicenter, so he obtains his location information using GPS and sends it to the server.

[1780] 2. Server: Integrates the acquired location information with real-time earthquake information.

[1781] 3. Server: Input the integrated data into the generative model and estimate the safety of user A.

[1782] 4. Generative model: Infers that user A is in a dangerous situation and returns the result to the server.

[1783] 5. Server: Since the user is judged to be in danger, safety information is sent to User A's emergency contact.

[1784] 6. Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue agencies.

[1785] Example 2: Identifying evacuees with a sense of sadness

[1786] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads it to the server.

[1787] 2. Server: Passes the received photo to the emotion engine.

[1788] 3. Emotion engine: Analyzes user B's sadness from the image and recognizes his emotional state.

[1789] 4. Server: Based on the emotion analysis results, notify the emergency contact of User B’s emotional state and the evacuation shelter situation.

[1790] 5. Server: Based on the analysis report, accurate information is posted on the official social media accounts and requests for relief supplies as needed.

[1791] By incorporating emotion recognition, it becomes possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[1792] The processing flow will be explained below.

[1793] Data collection

[1794] Step 1:

[1795] Device: The user's smartphone periodically obtains its current location using its GPS function.

[1796] Step 2:

[1797] Device: Sends the acquired location information to the server.

[1798] Step 3:

[1799] Server: Obtains real-time disaster information from the Japan Meteorological Agency and earthquake early warning APIs.

[1800] Step 4:

[1801] Server: The acquired disaster information is stored in a database and linked to the user's location information.

[1802] Data analysis

[1803] Step 5:

[1804] Server: Integrates user location information and disaster information to create a dataset.

[1805] Step 6:

[1806] Server: Inputs the integrated data into the generative model.

[1807] Step 7:

[1808] Generative model: Estimates the user's safety based on the user's location information and the scale of the disaster.

[1809] Step 8:

[1810] Generative model: Returns the inference results to the server.

[1811] Safety registration and rescue request

[1812] Step 9:

[1813] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[1814] Step 10:

[1815] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[1816] Step 11:

[1817] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[1818] emotion recognition

[1819] Step 12:

[1820] User: Take photos and videos of disaster areas and evacuation centers and upload them to the server.

[1821] Step 13:

[1822] Server: Receives uploaded image and video data.

[1823] Step 14:

[1824] Server: Passes the received data to the emotion engine.

[1825] Step 15:

[1826] Emotion engine: Analyzes user emotions from images and videos and generates emotional states.

[1827] Step 16:

[1828] Server: Integrates the analysis results from the emotion engine and the safety estimation results to generate detailed user information.

[1829] Information dissemination

[1830] Step 17:

[1831] Server: Based on the consolidated user information, sends notifications to the appropriate emergency contacts.

[1832] Step 18:

[1833] Server: Integrates analytical results with data from other reliable sources.

[1834] Step 19:

[1835] Server: Disseminates reliable information to official social media accounts and related organizations.

[1836] Step 20:

[1837] Server: If necessary, requests for relief supplies and the current situation at evacuation centers will be sent out.

[1838] Specific examples

[1839] Example 1: Rescue operations after an earthquake

[1840] Step 1:

[1841] Device: User A is in an area close to the epicenter, so his location information is obtained using GPS.

[1842] Step 2:

[1843] Device: Sends the acquired location information to the server.

[1844] Step 3:

[1845] Server: Obtains real-time earthquake information from the Japan Meteorological Agency and Earthquake Early Warning API.

[1846] Step 4:

[1847] Server: Integrates the acquired earthquake information and location information.

[1848] Step 5:

[1849] Server: Inputs the integrated data into the generative model.

[1850] Step 6:

[1851] Generative model: Estimates the safety of user A and returns the result to the server.

[1852] Step 7:

[1853] Server: Based on the safety estimation result, it is determined that User A is in a dangerous situation.

[1854] Step 8:

[1855] Server: Sends safety information to User A's emergency contacts. This notification includes location information and dangerous situations.

[1856] Step 9:

[1857] Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue organizations.

[1858] Example 2: Shelter situation report and emotion recognition

[1859] Step 1:

[1860] User: User B takes a photo of the situation at the evacuation shelter with his smartphone.

[1861] Step 2:

[1862] User: Uploads captured images and videos to the server.

[1863] Step 3:

[1864] Server: Receives uploaded image and video data.

[1865] Step 4:

[1866] Server: Passes the received data to the emotion engine.

[1867] Step 5:

[1868] Emotion engine: Analyzes user B's emotions from images and videos and generates their emotional state.

[1869] Step 6:

[1870] Server: Integrates the analysis results from the emotion engine with other data.

[1871] Step 7:

[1872] Server: Sends a notification to emergency contacts based on the integrated detailed information of User B.

[1873] Step 8:

[1874] Server: Posts the results of the emotion analysis and the situation at the evacuation center to the official social media account.

[1875] Step 9:

[1876] Server: If necessary, request relief supplies for evacuation centers.

[1877] By incorporating emotion recognition, it is possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[1878] Example 2

[1879] 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."

[1880] While conventional disaster prevention systems can collect user location information and disaster information, they have difficulty integrating this information to estimate the user's safety in real time and recognize emotions. Furthermore, they lack the ability to respond quickly when a user is deemed to be in danger, and they lack support that takes into account the emotional state of the victim, limiting the efficiency and effectiveness of rescue operations.

[1881] 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.

[1882] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data with an emotion engine, means for notifying emergency contacts based on the analysis results of the emotion engine, means for the analytical model to reassess the damage situation from the images and videos, and means for transmitting information and requesting rescue based on the analysis results. This enables fast and accurate rescue and support by combining user safety estimation and emotion recognition.

[1883] "User terminal" refers to an electronic device that is owned by a user and is capable of acquiring and transmitting location information and media data.

[1884] "Current location" refers to real-time geographical location information obtained by the user terminal using GPS.

[1885] "Server" refers to a central processing unit that receives information from user terminals and manages databases and performs data analysis.

[1886] "Weather data" refers to real-time weather information provided by the Japan Meteorological Agency and related organizations.

[1887] "Disaster data" refers to real-time damage status and disaster-related information obtained when disasters such as earthquakes and floods occur.

[1888] A "generative model" refers to a machine learning algorithm for estimating a user's safety based on their location and disaster information.

[1889] "Safety information" refers to information about the safety of a user estimated by a generative model.

[1890] "Notification" refers to the action of sending information about a user's safety or other important information to relevant people or organizations.

[1891] "Images and videos" refers to visual media data that users take and upload to a server.

[1892] "Emotion engine" refers to software that analyzes a user's emotional state from uploaded images and videos.

[1893] "Emergency Contacts" means the people or organizations that a User has registered to be notified in the event of an emergency.

[1894] "Analysis model" refers to a machine learning algorithm for reassessing the damage situation from uploaded media data.

[1895] "Requesting help" refers to an action of requesting help from an appropriate rescue organization when the user is determined to be in a dangerous situation.

[1896] "Information dissemination" refers to the action of providing reliable information based on the analysis results to related institutions and official social media accounts.

[1897] This invention relates to a disaster prevention AI platform for providing prompt and accurate rescue and support in the event of a disaster. This system acquires location information from the user's device, and the server integrates this information with disaster information and inputs it into an analysis model to estimate the user's safety. Furthermore, an emotion engine performs emotion analysis from images and videos uploaded by the user, allowing for a detailed understanding of the damage situation.

[1898] Data collection

[1899] Location information acquisition

[1900] The user's device periodically acquires the user's current location using the built-in GPS function and sends that information to the server. This location information is captured every 30 seconds using the location information service in the user's smartphone and uploaded to the server via HTTPS communication.

[1901] Disaster information acquisition

[1902] The server obtains real-time disaster information using the Japan Meteorological Agency and disaster alert APIs. This information is sent to the API at regular intervals, and the obtained data is parsed in JSON format and stored in a database.

[1903] Data analysis

[1904] Data Integration

[1905] The server combines user location information and disaster information to create a dataset to input into the generative AI model. This dataset retrieves the necessary information from a database on the server and is synchronized based on timestamps.

[1906] Estimated safety

[1907] The generative AI model estimates the user's safety based on their location and the scale of the disaster. The generative model uses a machine learning algorithm to perform real-time analysis and determine the user's safety status. This allows for a quick assessment of whether the user is "safe" or "at risk."

[1908] Safety registration and rescue request

[1909] Safety assessment

[1910] The server determines whether the user is safe or unsafe based on the estimation results of the generative model and records the result in a database.

[1911] Safety notification

[1912] If a user is deemed to be in danger, the server automatically sends a notification to emergency contacts via SMS or email, containing the user's current location and details of the disaster.

[1913] Rescue request

[1914] If necessary, the server will send an emergency call to the appropriate rescue agency, including detailed location and status, via the rescue agency's API endpoint.

[1915] emotion recognition

[1916] Media Upload

[1917] Users take photos of disaster areas and evacuation shelters with their smartphones and upload the media files to the server, which sends the files via HTTPS.

[1918] Emotion analysis

[1919] The server passes the uploaded images and videos to the emotion engine for emotion analysis. The emotion engine identifies the user's emotional state from the images and videos and returns the analysis results to the server. Based on this analysis result, the user's emotional state is stored in a database.

[1920] Emotion information notification

[1921] Based on the results of the emotion analysis, the server notifies emergency contacts of the user's emotional state and safety information.

[1922] Situation Analysis

[1923] Media Analysis

[1924] The server uses the emotion engine's analysis results to reassess the damage situation in more detail using an analytical model. The analytical model reanalyzes the detailed situation in the affected areas based on the data it receives and generates a report.

[1925] Information dissemination

[1926] Information aggregation

[1927] The server combines the analysis results with data from other reliable sources to create a single information set.

[1928] Information dissemination

[1929] The server sends reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and timely support. It posts information to official accounts using social media APIs and sends notifications to related organizations via email and dedicated APIs.

[1930] Specific examples

[1931] Rescue operations following an earthquake

[1932] Because User A is in an area close to the epicenter, his location information is acquired via GPS and sent to the server. The server combines the acquired location information with real-time earthquake information and inputs the data into a generative model to estimate User A's safety. The generative model estimates that User A is in danger and returns the result to the server. Since the server has determined that User A is in danger, it sends safety information to User A's emergency contacts and issues an emergency rescue request to nearby rescue organizations.

[1933] Confirmation of evacuees with a sense of despair

[1934] User B takes a photo of the evacuation shelter situation with their smartphone and uploads it to the server. The server passes the received photo to an emotion engine, which analyzes the image to determine User B's sense of despair and recognizes their emotional state. Based on the emotion analysis results, the server notifies emergency contacts of User B's emotional state and the evacuation shelter situation, and also posts accurate information to official social media accounts and requests relief supplies if necessary.

[1935] By incorporating emotion recognition, it is possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[1936] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1937] Step 1:

[1938] Location information acquisition

[1939] The user's terminal periodically obtains its current location using the GPS function.

[1940] Input: GPS satellite signal

[1941] Data processing: Obtaining the longitude and latitude information of the current location

[1942] Output: Acquired location data

[1943] Specific operation: The location information service in the smartphone runs, periodically capturing the current location, encrypting that information, and sending it to the server.

[1944] Step 2:

[1945] Location information transmission

[1946] The user's terminal transmits the acquired location information to the server.

[1947] Input: Location data

[1948] Data Computation: Location Encryption

[1949] Output: Encrypted location data

[1950] Specific operation: The device's transmission module is activated and uploads encrypted location information to the server via HTTPS communication.

[1951] Step 3:

[1952] Disaster information acquisition

[1953] The server obtains real-time disaster information from the Japan Meteorological Agency and disaster alert APIs.

[1954] Input: API request

[1955] Data processing: JSON parsing of acquired disaster information

[1956] Output: Disaster information data

[1957] Specific operation: The server periodically sends requests to the API endpoint, obtains the returned disaster information, and stores it in a database.

[1958] Step 4:

[1959] Data Integration

[1960] The server integrates location information and disaster information to create a dataset.

[1961] Input: User location information and disaster information data

[1962] Data Computation: Data Synchronization and Integration

[1963] Output: Unified dataset

[1964] Specific operation: The server retrieves location information and disaster information from the database and creates an integrated dataset based on that data.

[1965] Step 5:

[1966] Estimated safety

[1967] The server inputs the integrated data set into a generative AI model to estimate the user's safety.

[1968] Input: Integrated dataset

[1969] Data calculation: Estimation of user safety (application of machine learning algorithms)

[1970] Output: Safety estimation result

[1971] Specific operation: A dataset is input into the generative AI model, which uses a neural network to estimate the user's safety and status and returns the results to the server.

[1972] Step 6:

[1973] Safety assessment

[1974] The server determines whether the user is safe or dangerous based on the estimation results of the generative model.

[1975] Input: Safety estimation result

[1976] Data calculation: Determining safety status based on conditions

[1977] Output: Safety status (safe or dangerous)

[1978] Specific operation: The server analyzes the estimation results and determines the user's safety status according to the defined conditions.

[1979] Step 7:

[1980] Safety notification

[1981] The server sends a notification to emergency contacts if the user is determined to be at risk.

[1982] Input: Safety status, user details

[1983] Data operation: generating notification messages

[1984] Output: Notification message

[1985] Specific operation: The server sends a notification message to emergency contacts via SMS or email gateway.

[1986] Step 8:

[1987] Rescue request

[1988] The server will send an emergency rescue request to a rescue organization if necessary.

[1989] Input: User's detailed location information, disaster situation

[1990] Data calculation: Generating a help-call message

[1991] Output: Help request message

[1992] What happens: The server sends a request to the rescue agency's API endpoint, issuing an emergency rescue request including the user's details.

[1993] Step 9:

[1994] Media Upload

[1995] Users upload photos and videos of the disaster area to the server.

[1996] Input: Image and video data

[1997] Data Processing: Sending Media Files

[1998] Output: Media data stored on the server

[1999] Specific operation: A user takes a media file with their smartphone camera and uploads it to a server via HTTPS.

[2000] Step 10:

[2001] Emotion analysis

[2002] The server passes the uploaded media files to the emotion engine for emotion analysis.

[2003] Input: Media files

[2004] Data calculation: Emotional state analysis (application of image analysis algorithms)

[2005] Output: Emotion analysis results

[2006] Specific operation: The server sends media files to the emotion engine, which analyzes the user's emotional state from images and videos and returns the results to the server.

[2007] Step 11:

[2008] Emotion information notification

[2009] The server notifies emergency contacts of the user's emotional state and safety information based on the emotion analysis results.

[2010] Input: Emotion analysis results, safety information

[2011] Data operation: generating notification messages

[2012] Output: Notification message

[2013] Specific operation: The server generates a notification message containing the results of the sentiment analysis and sends it to emergency contacts via SMS or email.

[2014] Step 12:

[2015] Media Analysis

[2016] Based on the analysis results of the emotion engine, the server reevaluates the damage situation in more detail using an analytical model.

[2017] Input: Sentiment analysis results and other media data

[2018] Data calculation: Reassessment of damage situation (application of machine learning algorithms)

[2019] Output: Damage report

[2020] Specific operation: The server inputs data into the analytical model and generates a report to reassess the damage situation.

[2021] Step 13:

[2022] Information aggregation

[2023] The server integrates the analysis results with data from other reliable sources.

[2024] Input: Analysis results, data from other sources

[2025] Data Calculation: Data Integration

[2026] Output: Integrated information set

[2027] Specific operation: The server merges information obtained from multiple data sources and performs consistency checks.

[2028] Step 14:

[2029] Information dissemination

[2030] The server will send out reliable information to official social media accounts and related organizations.

[2031] Input: Integrated information set

[2032] Data operations: generating information messages

[2033] Output: Published information message

[2034] Specific operations: Post information to official accounts using SNS APIs, and send notifications to relevant organizations via email or dedicated APIs.

[2035] (Application example 2)

[2036] 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."

[2037] Currently, there are limited systems that can provide rapid and accurate rescue and assistance when disasters occur. In particular, there are no systems that can grasp the emotional state of victims and respond appropriately. Furthermore, technology to effectively utilize autonomous vehicles to carry out rapid rescue operations has not yet been fully developed. An efficient system to solve these issues is needed.

[2038] 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.

[2039] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for acquiring weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, means for transmitting information and requesting rescue based on the analysis results, means for acquiring the current location of the autonomous vehicle and calculating an optimal route, and means for recognizing the emotions of disaster victims and taking action based on their emotional state. This enables rapid and accurate rescue operations and appropriate responses based on the emotional state of disaster victims.

[2040] A "user terminal" is an electronic device owned by a user that is used to obtain current location information.

[2041] The "current location" is the geographical location information of the user terminal.

[2042] "Means" are methods or technical elements for performing a specific function.

[2043] A "server" is a computer system that can receive and process data sent from a user terminal.

[2044] "Weather data" is information relating to weather conditions and changes.

[2045] "Disaster data" refers to information related to disasters such as earthquakes, floods, and fires.

[2046] A "generative model" is an algorithm or machine learning model that analyzes acquired data and makes certain predictions or inferences.

[2047] "Safety" refers to the status of whether the user is in a safe state or in a dangerous state.

[2048] "Safety information" is information relating to the safety status of the user.

[2049] "Notification" is the act of transmitting specific information to other terminals or systems.

[2050] "Images and videos" are still images and videos taken by the user.

[2051] An "analytical model" is an algorithm or data processing method for analyzing collected data and extracting specific information.

[2052] A "request for rescue" is an act of requesting the support necessary to carry out rescue operations.

[2053] An "autonomous vehicle" is a vehicle that can be driven automatically without human operation.

[2054] A "route" is a path from a specific point to a destination.

[2055] "Calculation" means performing the necessary calculations to obtain the results.

[2056] An "emotion" is a psychological state such as joy, sadness, or fear felt by a user.

[2057] "Responding" means acting in a way that is adapted to a particular situation.

[2058] The system for realizing the present invention comprises the following elements.

[2059] 1. Data Collection

[2060] The user device periodically acquires its current location using its GPS function and sends that location information to the server. It also acquires media data (images and videos) using its camera and microphone to recognize the user's emotions, and uploads them to the server.

[2061] 2. Disaster Information Collection

[2062] The server retrieves real-time weather and disaster data from the Japan Meteorological Agency and earthquake early warning APIs, stores this information in a database, and integrates the location information and disaster data to create a dataset to be input into the generative model.

[2063] 3. Analysis using generative models

[2064] The generative model stored on the server estimates the user's safety in real time based on the user's location information and disaster data. The generative model uses a machine learning algorithm to perform real-time analysis and determine safety information.

[2065] 4. Notification of safety information and rescue requests

[2066] Based on the predictions returned by the generative model, the server determines whether the user is safe or in danger, and sends a notification to emergency contacts as necessary. It also obtains the current location of the autonomous vehicle, calculates the optimal route, and dispatches it to the disaster area.

[2067] 5. Media Data Analysis and Emotion Recognition

[2068] Uploaded image and video data is analyzed by an analysis model on the server. The emotion recognition engine analyzes the user's emotions from the images and videos and generates a detailed emotional state. Based on the analysis results, the server notifies emergency contacts of the user's emotional state and situation.

[2069] 6. Relief operations management

[2070] The autonomous vehicle will then follow the calculated optimal route to the disaster area and carry out rescue operations. During the rescue operations, further media data will be acquired and analyzed, and the server will issue a rescue request if necessary.

[2071] Hardware and software used

[2072] GPS module: Obtains the current location information of the user device.

[2073] Camera and microphone: Capture images and videos to recognize user emotions.

[2074] Server: Collects, integrates, analyzes, and notifies data.

[2075] Japan Meteorological Agency API and Earthquake Early Warning API: Obtain real-time disaster data.

[2076] Generative model: Estimates safety using machine learning algorithms.

[2077] Emotion recognition engine: Analyzes emotions from images and videos.

[2078] Specific examples

[2079] If the user is in an area affected by an earthquake, the user's device will use GPS to obtain location information and send it to the server. The server will then collect disaster data via an earthquake early warning API and use a generative model to estimate the user's safety. Based on the estimation results, a notification will be sent to emergency contacts and an autonomous vehicle will be dispatched along the optimal route. After arriving, the camera will analyze the user's facial expressions and provide any necessary psychological care.

[2080] Prompt Sentence Examples

[2081] "Please design a system that obtains detailed location information of users who are currently near the epicenter and need to evacuate, and sets the optimal rescue route. Also, after arriving, please use a camera to analyze the emotions of the victims and provide the necessary support."

[2082] This will enable quick and accurate rescue operations and appropriate responses based on the emotional state of the victims.

[2083] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2084] Step 1:

[2085] Obtaining location information of user devices

[2086] The user terminal periodically obtains its current location using a GPS module.

[2087] Input: Location data from the GPS module

[2088] Output: Current location of the user

[2089] Specific operation: The user device obtains its current location from GPS every few seconds and stores the data in memory.

[2090] Step 2:

[2091] Sending location information

[2092] The user terminal transmits the acquired location information to the server.

[2093] Input: current location of the user

[2094] Output: Location information sent to the server

[2095] Specific operation: The user device sends location information to the server's API endpoint as an HTTP request.

[2096] Step 3:

[2097] Obtaining disaster information

[2098] The server obtains weather and disaster data from the Japan Meteorological Agency API and Earthquake Early Warning API.

[2099] Input: API request

[2100] Output: Real-time disaster data

[2101] Specific operation: The server periodically sends requests to the API endpoint to obtain the latest disaster information and stores it in the database.

[2102] Step 4:

[2103] Data integration

[2104] The server integrates the acquired location information and disaster data to create a dataset to be input into the generative model.

[2105] Input: User's current location information, disaster data

[2106] Output: The dataset to feed into the generative model

[2107] Specific operation: The server combines location information and disaster data and converts them into a format that can be analyzed by the generative model.

[2108] Step 5:

[2109] Safety estimation using generative models

[2110] The generative model estimates the user's safety based on the user's location information and disaster data.

[2111] Input: Integrated dataset

[2112] Output: Safety estimation result

[2113] Specific operation: The generative model processes the integrated dataset and estimates the user's safety in real time.

[2114] Step 6:

[2115] Registration and notification of safety information

[2116] Based on the estimation results, the server registers the user's safety information and notifies emergency contacts.

[2117] Input: Safety estimation result

[2118] Output: Safety information sent to emergency contacts

[2119] Specific operation: The server stores the safety information in a database and notifies emergency contacts via email or SMS.

[2120] Step 7:

[2121] Uploading media data

[2122] Users take photos and videos of disaster areas and evacuation shelters and upload them to the server.

[2123] Input: Image and video data

[2124] Output: Media data stored on the server

[2125] Specific operation: The user uses the device's camera to capture media data and sends it to the server through the application.

[2126] Step 8:

[2127] Media Data Analysis

[2128] The server analyzes the uploaded image and video data using an analytical model.

[2129] Input: Media data

[2130] Output: Analysis results

[2131] Specific operation: The server uses an emotion recognition engine to analyze the media data and recognize the emotional state of the victim.

[2132] Step 9:

[2133] Autonomous vehicle routing

[2134] The server obtains the current location of the autonomous vehicle and calculates the optimal route.

[2135] Input: current vehicle location, disaster information

[2136] Output: Optimal route

[2137] Specific operation: The server obtains GPS data from the autonomous vehicle and calculates the optimal rescue route based on disaster information.

[2138] Step 10:

[2139] Carrying out rescue operations

[2140] The autonomous vehicle will follow the calculated optimal route to the disaster area and carry out rescue operations.

[2141] Input: Optimal Route

[2142] Output: Rescue operations in disaster areas

[2143] Specific operation: The autonomous vehicle uses the navigation system to drive along the specified route and begins rescue operations upon arrival.

[2144] Step 11:

[2145] Emotional information notification

[2146] Based on the emotion analysis results, the server notifies the user's emotional state and situation to the emergency contact.

[2147] Input: Analysis results

[2148] Output: Emotion information sent to emergency contacts

[2149] Specific operation: The server notifies the emergency contact of the analysis results via email or SMS.

[2150] This will enable quick and accurate rescue operations and appropriate responses based on the emotional state of the victims.

[2151] 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.

[2152] 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.

[2153] 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.

[2154] [Fourth embodiment]

[2155] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[2156] 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.

[2157] 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).

[2158] 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.

[2159] 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.

[2160] 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).

[2161] 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.

[2162] 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.

[2163] 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.

[2164] 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.

[2165] 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.

[2166] 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.

[2167] 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."

[2168] This invention provides a disaster prevention AI platform that enables rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes them using a generative model to estimate the user's safety, and automatically carries out necessary rescue and support.

[2169] Explaining program processing in natural language

[2170] Data collection

[2171] 1. Obtain location information

[2172] Device: The user's smartphone periodically obtains its current location using its GPS function and sends that location information to the server.

[2173] 2. Disaster information acquisition

[2174] Server: Real-time disaster information is obtained from the Japan Meteorological Agency and earthquake early warning APIs. This information is stored in a database on the server and linked to the user's location information.

[2175] Data analysis

[2176] 1. Data Integration

[2177] Server: Integrates the acquired user location information and disaster information to create a dataset to be input into the generative model.

[2178] 2. Safety Estimation

[2179] Generative model: Estimates the user's safety based on the user's location information, the scale of the disaster, etc. The generative model uses machine learning algorithms to perform real-time analysis and determine the user's safety status.

[2180] Safety registration and rescue request

[2181] 1. Safety assessment

[2182] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[2183] 2. Safety Notification

[2184] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[2185] 3. Request for help

[2186] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[2187] Situation Analysis

[2188] 1. Media Upload

[2189] Users can take photos and videos of disaster areas and evacuation centers and upload them to the server.

[2190] 2. Media Analysis

[2191] Server: Receives images and videos uploaded by users and passes them to the analysis model.

[2192] Analysis model: Uses machine learning to analyze damage and people's movements from images and videos, and generates detailed reports.

[2193] Information dissemination

[2194] 1. Information aggregation

[2195] Server: Integrates analytical results with data from other reliable sources.

[2196] 2. Information dissemination

[2197] Server: Disseminates reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and prompt support information.

[2198] Specific examples

[2199] Example 1: Rescue operations after an earthquake

[2200] 1. Device: User A is in an area close to the epicenter, so he obtains his location information using GPS and sends it to the server.

[2201] 2. Server: Integrates location information with earthquake information obtained from the Japan Meteorological Agency.

[2202] 3. Server: Input the integrated data into the generative model and estimate the safety of user A.

[2203] 4. Generative model: Determines that user A is in a dangerous situation and returns the result to the server.

[2204] 5. Server: Since the user is judged to be in danger, safety information is sent to User A's emergency contact.

[2205] 6. Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue agencies.

[2206] Example 2: Shelter situation report

[2207] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads it to the server.

[2208] 2. Server: Passes the received photos to the analysis model.

[2209] 3. Analysis model: Analyzes crowded evacuation shelters from images and generates a report.

[2210] 4. Server: Based on the analysis report, the system posts information about the congestion situation at evacuation centers and the supplies needed to the official social media accounts.

[2211] 5. Server: Requests relief supplies based on reliable information.

[2212] This will help eliminate false information and rumors and quickly provide the rescue and assistance that is truly needed in affected areas.

[2213] The processing flow will be explained below.

[2214] Data collection

[2215] Step 1:

[2216] Device: The user's smartphone uses the GPS function to obtain the current location.

[2217] Step 2:

[2218] Device: Sends the acquired location information to the server.

[2219] Step 3:

[2220] Server: Obtains real-time disaster information from the Japan Meteorological Agency and earthquake early warning APIs.

[2221] Step 4:

[2222] Server: The acquired disaster information is stored in a database and linked to the user's location information.

[2223] Data analysis

[2224] Step 5:

[2225] Server: Integrates user location information and disaster information to create a dataset.

[2226] Step 6:

[2227] Server: Inputs the integrated data into the generative model.

[2228] Step 7:

[2229] Generative model: Estimates the user's safety based on the user's location information and the scale of the disaster.

[2230] Step 8:

[2231] Generative model: Returns the inference results to the server.

[2232] Safety registration and rescue request

[2233] Step 9:

[2234] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[2235] Step 10:

[2236] Server: If a user is deemed to be at risk, a notification is automatically sent to pre-registered emergency contacts.

[2237] Step 11:

[2238] Server: If necessary, initiates an emergency rescue request to the appropriate rescue agency.

[2239] Situation Analysis

[2240] Step 12:

[2241] User: Take photos and videos of disaster areas and evacuation centers and upload them to the server.

[2242] Step 13:

[2243] Server: Receives uploaded image and video data.

[2244] Step 14:

[2245] Server: Passes the received data to the analysis model.

[2246] Step 15:

[2247] Analysis model: Analyzes damage conditions and people's movements from images and videos, and generates detailed reports.

[2248] Information dissemination

[2249] Step 16:

[2250] Server: Integrates analytical results with data from other reliable sources.

[2251] Step 17:

[2252] Server: Disseminates reliable information to official social media accounts and related organizations.

[2253] Step 18:

[2254] Server: If necessary, requests for relief supplies and the current situation at evacuation centers will be sent out.

[2255] Example 1

[2256] 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."

[2257] Conventional disaster prevention systems only collect user location information and disaster information separately, making it difficult to integrate these information to estimate the user's safety in real time and provide prompt and accurate rescue and support. Furthermore, the lack of proper aggregation of disaster information and the dissemination of reliable information has led to the spread of false information, causing confusion in the affected areas. A new system is needed to solve these problems.

[2258] 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.

[2259] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation result, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, and means for the server to transmit information and request rescue based on the analysis result to official social media accounts and related organizations, thereby enabling rapid and accurate rescue and support when a disaster occurs.

[2260] A "user terminal" refers to an electronic device that can be carried by a user, specifically a device equipped with a GPS function such as a smartphone or tablet.

[2261] "Current location" refers to information about the user's current location obtained by the user's device using the GPS function.

[2262] A "server" is a high-performance computer system connected to a network, which receives data from multiple users and devices and processes and stores it.

[2263] "Weather data" refers to weather-related data provided by the Japan Meteorological Agency or weather services, specifically information such as temperature, wind speed, and precipitation.

[2264] "Disaster data" refers to information about natural disasters such as earthquakes, typhoons, and floods, and specifically includes the date and time of occurrence, location, and extent of impact.

[2265] A "generative model" refers to a computer program that uses machine learning algorithms to estimate a user's safety.

[2266] "Safety" refers to whether the user is currently safe or in imminent danger.

[2267] "Notification" refers to alert information sent from the system to users and emergency contacts, specifically by email or SMS.

[2268] "Images and videos" refers to visual data captured by users and uploaded to the system.

[2269] An "analysis model" refers to a program that uses machine learning algorithms to analyze uploaded images and videos and assess the extent of damage and people's movements.

[2270] "Official social media account" refers to a highly reliable social media account that the system uses to disseminate information.

[2271] "Related organizations" refer to official organizations that carry out disaster prevention and rescue operations, including fire departments, police departments, and medical institutions.

[2272] "Request for Assistance" means a request sent to relevant authorities to carry out rescue operations required in the event of a disaster.

[2273] This invention is a disaster prevention AI platform that enables rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes them using a generative AI model to estimate the user's safety, and automatically carries out necessary rescue and support.

[2274] Hardware and software used

[2275] User device: A portable electronic device with a GPS function, such as a smartphone or tablet, is used. This device can acquire the user's current location and send it to the server.

[2276] Server: A high-performance computer system is used to receive, process, and store data from multiple users and devices. It also collects real-time weather and disaster data from the Japan Meteorological Agency and earthquake early warning APIs.

[2277] Generative model: Using an AI model that uses a machine learning algorithm, the safety of a user is estimated in real time based on the user's location information and disaster information.

[2278] Analysis Model: Using machine learning, the model analyzes damage from images and videos and generates a detailed report based on media files uploaded by users.

[2279] Program processing

[2280] 1. Location information acquisition: The user device uses the built-in GPS function to acquire the user's current location at specific intervals, encrypts the data, and sends it to the server.

[2281] 2. Disaster information acquisition: The server acquires disaster information in real time from the Japan Meteorological Agency and earthquake early warning APIs and stores this information in a database.

[2282] 3. Data integration: The server integrates user location information and disaster information to create a dataset to be input into the generative AI model.

[2283] 4. Safety Prediction: The generative model estimates the user's safety based on the integrated data. For example, it determines whether the user is in danger based on the scale of the disaster, the user's location, and past data.

[2284] 5. Safety determination: The server determines whether the user is "safe" or "dangerous" based on the inference results returned by the generative model.

[2285] 6. Safety Notification: If a user is deemed to be in danger, the server automatically sends a notification to the user's emergency contacts, including the user's location and status.

[2286] 7. Rescue Request: If the user is deemed to be in danger, the server will issue an emergency rescue request to the appropriate rescue agency. This request will include the user's detailed location information and the disaster situation.

[2287] 8. Media Upload: Users can take photos and videos of disaster areas and evacuation centers with their smartphones and upload the media files to the server.

[2288] 9. Media Analysis: The server receives images and videos sent by users, inputs them into an analytical model, and generates a detailed report assessing the damage situation and people's movements.

[2289] 10. Information aggregation: The server integrates the analysis results with data from other reliable sources to provide comprehensive information aggregation.

[2290] 11. Information Dissemination: The server will disseminate information based on reliable information and analysis results to official social media accounts and related organizations, thereby eliminating false information and providing accurate and prompt support information.

[2291] Examples of specific examples and prompts

[2292] Example 1: Rescue operations after an earthquake

[2293] 1. Device: User A is in an area close to the epicenter, so his smartphone uses the GPS function to obtain location information every five minutes.

[2294] 2. Server: The server obtains information on the epicenter from the Japan Meteorological Agency API and stores the data in a database in real time.

[2295] 3. Server: Integrates user A's location information and disaster information about the epicenter, and creates a dataset to be input into the generative AI model.

[2296] 4. Generative model: Based on the integrated data, determine that user A is in a dangerous situation.

[2297] 5. Server: Based on the results returned from the generative model, User A is determined to be “dangerous.”

[2298] 6. Server: Notifies User A's emergency contacts of their location and safety status. For example, send an SMS message saying, "User A is near the epicenter and in a dangerous situation."

[2299] 7. Server: Sends a similar emergency rescue request to nearby rescue organizations.

[2300] Example 2: Reporting on the situation at an evacuation shelter

[2301] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads the photo to the server.

[2302] 2. Server: The received photos are input into the analysis model to analyze the congestion situation at the evacuation center.

[2303] 3. Analytical model: Determines the level of congestion at evacuation centers from images and generates a detailed report.

[2304] 4. Server: Stores the analysis results in a database and integrates them with data from other sources.

[2305] 5. Server: Posts information about the congestion situation at evacuation centers and necessary supplies to the official social media account. Also, conveys the information to appropriate organizations to provide prompt assistance.

[2306] Prompt Sentence Examples

[2307] "User A is near the epicenter of an earthquake, so his location information is obtained using GPS and sent to a server. Please explain a program that uses a generative model to estimate the user's safety and status, and sends a notification to emergency contacts based on the results."

[2308] This system is designed to enable rapid and accurate rescue and support in the event of a disaster, and can ensure the safety of users through multi-layered data analysis and information dissemination.

[2309] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2310] Step 1:

[2311] Location information acquisition

[2312] Device: The user device (smartphone) periodically acquires its current location using its built-in GPS function. The acquired location information is sent to the server as latitude and longitude data.

[2313] Input: Location information from the GPS sensor

[2314] Output: Latitude and longitude data of the current location sent to the server

[2315] Step 2:

[2316] Disaster information acquisition

[2317] Server: The server collects disaster information in real time from the Japan Meteorological Agency and earthquake early warning APIs. This information provides information such as the type of disaster, the time of occurrence, scale, and extent of damage.

[2318] Input: Disaster information from the Japan Meteorological Agency and Earthquake Early Warning API

[2319] Output: Disaster information stored on the server (type, time of occurrence, scale, extent of damage, etc.)

[2320] Step 3:

[2321] Data Integration

[2322] Server: The server combines the user's location information and disaster information to create a dataset to input into the generative model. Specifically, it matches the user's latitude and longitude with disaster information in that area.

[2323] Input: User location information and disaster information

[2324] Output: A unified dataset to feed into the generative model

[2325] Step 4:

[2326] Estimated safety

[2327] Generative model: The generative model estimates the user's safety based on the integrated dataset. For example, if the user's location is close to the epicenter or if the area is hit by strong shaking, the model estimates that the user is in danger.

[2328] Input: Integrated dataset

[2329] Output: User's safety status (safe or unsafe)

[2330] Step 5:

[2331] Safety assessment

[2332] Server: Based on the safety estimation results obtained from the generative model, the server determines whether the user is safe or at risk. If the user is determined to be at risk, the server proceeds to the next step.

[2333] Input: Safety estimation results from the generative model

[2334] Output: User safety assessment result (safe or dangerous)

[2335] Step 6:

[2336] Safety notification

[2337] Server: If the user is deemed to be in danger, the server automatically sends a notification to pre-registered emergency contacts, including the user's current location and the reason for the danger.

[2338] Input: User safety assessment results, emergency contact information

[2339] Output: Safety notification sent to emergency contacts (including location and details of the situation)

[2340] Step 7:

[2341] Rescue request

[2342] Server: If the user is determined to be in danger, the server sends an emergency rescue request to the appropriate rescue agency, including the user's location and the disaster situation.

[2343] Input: User safety assessment results, location information, disaster information

[2344] Output: Emergency rescue request sent to rescue agencies, including user's detailed location and disaster situation.

[2345] Step 8:

[2346] Media Upload

[2347] User: The user takes photos of the disaster area or the situation at the evacuation center with their smartphone and uploads the images and videos to the server.

[2348] Input: Images and videos taken by the user

[2349] Output: Media files uploaded to the server

[2350] Step 9:

[2351] Media Analysis

[2352] Server: The server passes the received images and videos to the analysis model, which analyzes the damage situation and people's movements. The analysis model uses machine learning algorithms to generate a detailed report.

[2353] Input: Images and videos uploaded to the server

[2354] Output: Report of damage status based on analytical model

[2355] Step 10:

[2356] Information aggregation

[2357] Server: The server integrates the analysis results with data from other reliable sources to aggregate comprehensive information, allowing for a comprehensive understanding of the situation in the affected areas.

[2358] Input: Analysis model reports, data from other sources

[2359] Output: Aggregated comprehensive information

[2360] Step 11:

[2361] Information dissemination

[2362] Server: Based on reliable information, the server will disseminate important information to official social media accounts and relevant organizations, thereby eliminating false information and providing prompt assistance.

[2363] Input: Aggregated comprehensive information

[2364] Output: Information posted on official social media accounts and related organizations

[2365] In this way, each step works in coordination, enabling rapid and accurate rescue and support in the event of a disaster.

[2366] (Application example 1)

[2367] 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."

[2368] When natural disasters occur, it is difficult to quickly and accurately collect and analyze information, which often leads to delays in rescue and assistance to victims. Furthermore, normal services may be suspended during disasters, and the disruption of services such as food delivery can have a significant impact on the lives of victims. Furthermore, insufficient means exist for confirming the safety of victims, notifying emergency contacts, and requesting assistance, hindering rapid and efficient rescue efforts.

[2369] 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.

[2370] In this invention, the server includes means for acquiring a current location from a user terminal, means for transmitting the acquired location information to the server, means for the server to acquire weather data and disaster data, means for integrating the location information and the disaster data and inputting the information into a generative model, means for the generative model to estimate the user's safety, means for registering and notifying the user's safety information based on the estimation results, means for uploading images and videos posted by the user to the server, means for analyzing the uploaded data using an analytical model, means for transmitting information and requesting rescue based on the analysis results, and means for suspending or resuming specific services depending on the user's safety status during a disaster. This enables rapid and accurate information collection and analysis, accelerating rescue and support for disaster victims and reducing the risk of service suspension during a disaster.

[2371] A "user terminal" is a device such as a mobile terminal or smartphone that is owned by a user and has the function of acquiring location information.

[2372] "Current location" is information that is detected in real time by the user terminal and indicates the user's current geographical location.

[2373] A "server" is a computer system that receives, processes, and stores data over a network and works in conjunction with other devices.

[2374] "Weather data" refers to information about weather, temperature, precipitation, and other meteorological information provided by meteorological agencies.

[2375] "Disaster data" refers to information about natural disasters such as earthquakes, floods, and typhoons, and includes the date, time, location, and scale of the disaster.

[2376] A "generative model" is a data model that uses algorithms based on machine learning or deep learning to analyze input data and make predictions or inferences.

[2377] "Safety" refers to the status of the user indicating whether they are safe or in danger in a disaster situation.

[2378] The "estimated result" is information indicating the safety status of the user obtained as a result of analysis by the generative model.

[2379] "Safety information" is data related to the user's safety status, and is notified to emergency contacts and relevant organizations.

[2380] "Means for notifying" refers to the method or technology for sending messages or alerts to user terminals or emergency contacts.

[2381] "Images and videos" are visual media data that users take and upload to a server.

[2382] An "analysis model" is an algorithm or data model that uses machine learning and artificial intelligence technologies to analyze images and videos and extract useful information.

[2383] A "request for assistance" is a request made to relevant agencies to request emergency assistance or rescue operations.

[2384] "Means for suspending or resuming services" refers to methods or techniques for temporarily suspending specific services during a disaster and resuming them after the situation has improved.

[2385] This invention is a disaster prevention support system consisting of a user device, a server, and a generative model. Specifically, the user device acquires its current location using its GPS function and transmits it to the server. The server collects meteorological and disaster data in real time and integrates it with the user's location information. To analyze the shared data, the generative model uses a machine learning algorithm to estimate the user's safety.

[2386] The server does the following:

[2387] 1. Obtaining and sending location information:

[2388] User devices such as smartphones periodically use their GPS to obtain their current location and send it to a server. The Geopy library is used to obtain accurate location information.

[2389] 2. Meteorological and disaster data collection:

[2390] The server obtains real-time information using weather data and disaster data APIs, and stores the current weather conditions and disaster occurrence information in a database.

[2391] 3. Data integration and generative model use:

[2392] The user's location information and acquired disaster data are combined and input into a generative model, which uses machine learning algorithms to estimate the user's safety status. The generative model uses frameworks such as TensorFlow and PyTorch.

[2393] 4. Safety confirmation and notification:

[2394] Based on the inference results from the generative model, the system registers the user's safety information and notifies emergency contacts. If the user's safety is judged to be in danger, an emergency notification is automatically sent.

[2395] 5. Emergency relief request:

[2396] The server automatically sends an appropriate rescue request based on the analysis results, which includes detailed location information and current situation of the victim.

[2397] 6. Suspending and Resuming Services:

[2398] In the event of a disaster, services such as food delivery will be temporarily suspended and resumed as soon as the safety of users is confirmed. This is done by notifying service providers using a REST API.

[2399] The hardware and software used includes:

[2400] Hardware: User device (smartphone), server

[2401] Software: Geopy, machine learning frameworks (TensorFlow, PyTorch), disaster information API requests, REST API

[2402] Consider the following scenario:

[2403] Example 1: Suspension of food delivery during an earthquake

[2404] 1. Device: If the user is in an earthquake-prone area, the smartphone's GPS acquires location information and sends it to the server.

[2405] 2. Server: Collects earthquake information, integrates it with the acquired location information, and inputs it into the generative model.

[2406] 3. Generative model: If the user is judged to be at risk, the server sends a stop notification to the delivery service.

[2407] 4. Server: Additionally, notify the user's emergency contacts and request appropriate assistance.

[2408] Example 2: Resuming food delivery after safety confirmation

[2409] 1. Device: After the user moves to a safe location, the device acquires location information again and sends it to the server.

[2410] 2. Server: Based on location information and the latest disaster information, the generative model confirms the user's safety.

[2411] 3. Generative model: If the user is deemed safe, the server sends a resume notification to the delivery service.

[2412] Example prompt sentence:

[2413] "Please create a program that determines the availability of food delivery services based on the user's location information and disaster information, and notifies them when the service resumes after confirming safety. Please also include the user's emergency contact information and API endpoints for suspending and resuming the service."

[2414] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2415] Step 1:

[2416] The user terminal acquires the current location and transmits the location information to the server.

[2417] Input: Current location obtained from the smartphone's GPS.

[2418] Data processing: Extract latitude and longitude information using the Geopy library.

[2419] Output: Location information including latitude and longitude of the user.

[2420] Step 2:

[2421] The server obtains weather and disaster data in real time.

[2422] Input: API requests for weather data and disaster information.

[2423] Data processing: Parse the data obtained from the API and extract the necessary information (weather, earthquake occurrence information, etc.).

[2424] Output: Datasets such as current weather, disaster occurrence information, etc.

[2425] Step 3:

[2426] The server integrates the user's location information and disaster data and inputs them into the generative model.

[2427] Input: User location information, weather data, disaster data.

[2428] Data processing: Creating a unified dataset and converting it into an input format for the generative model.

[2429] Output: The synthesized data that is input into the generative model.

[2430] Step 4:

[2431] The generative model estimates the user's safety.

[2432] Input: Integrated data.

[2433] Data calculation: Using machine learning algorithms, the user's safety status is estimated in real time.

[2434] Output: User safety information (safe / dangerous).

[2435] Step 5:

[2436] Based on the estimation results, the server notifies the user's safety information to emergency contacts.

[2437] Input: Estimation results (safety information), user's emergency contact data.

[2438] Data processing: Construct the necessary notification information (estimation results, user location information).

[2439] Output: Notify emergency contacts (send message).

[2440] Step 6:

[2441] The server will call for help if necessary.

[2442] Input: Estimation results, user location information, disaster data.

[2443] Data processing: Package the necessary information to request assistance.

[2444] Output: Request (send message) to appropriate rescue agency.

[2445] Step 7:

[2446] Servers stop or restart food delivery services in the event of a disaster.

[2447] Input: User safety information, service status information.

[2448] Data calculation: Decisions on suspending or resuming services based on safety information.

[2449] Output: Suspend or resume notification (API request) to the food delivery service.

[2450] These steps will ensure prompt relief efforts and proper management of services in the event of a disaster.

[2451] 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.

[2452] This invention provides a disaster prevention AI platform for providing rapid and accurate rescue and support in the event of a disaster. This system collects user location information and disaster information, analyzes it using a generative model to estimate the user's safety, and automatically carries out necessary rescue and support. In addition, by adding an emotion engine that recognizes the user's emotions, it is possible to grasp the situation of victims in more detail and take appropriate measures.

[2453] Explaining program processing in natural language

[2454] Data collection

[2455] 1. Obtain location information

[2456] Device: The user's smartphone periodically obtains its current location using its GPS function and sends that location information to the server.

[2457] 2. Disaster information acquisition

[2458] Server: Real-time disaster information is obtained from the Japan Meteorological Agency and earthquake early warning APIs. This information is stored in a database on the server and linked to the user's location information.

[2459] Data analysis

[2460] 1. Data Integration

[2461] Server: Integrates user location information and disaster information to create a dataset to be input into the generative model.

[2462] 2. Safety Estimation

[2463] Generative model: Estimates the user's safety based on the user's location information, the scale of the disaster, etc. The generative model uses machine learning algorithms to perform real-time analysis and determine the user's safety status.

[2464] Safety registration and rescue request

[2465] 1. Safety assessment

[2466] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[2467] 2. Safety Notification

[2468] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[2469] 3. Request for help

[2470] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[2471] emotion recognition

[2472] 1. Media Upload

[2473] Users can take photos and videos of disaster areas and evacuation centers and upload them to the server.

[2474] 2. Emotion analysis

[2475] Server: Receives uploaded image and video data and passes it to the emotion engine.

[2476] Emotion Engine: Analyzes user emotions from images and videos and generates detailed emotional states.

[2477] 3. Emotion information notification

[2478] Server: Notifies the user's emotional state along with the estimated safety status to emergency contacts.

[2479] Situation Analysis

[2480] 1. Media Analysis

[2481] Server: The analysis results from the emotion engine are combined and passed to the analysis model.

[2482] Analysis model: Reassess the damage situation and people's movements from images and videos and generate detailed reports.

[2483] Information dissemination

[2484] 1. Information aggregation

[2485] Server: Integrates analytical results with data from other reliable sources.

[2486] 2. Information dissemination

[2487] Server: Disseminates reliable information to official social media accounts and related organizations, eliminating false information and providing accurate and prompt support information.

[2488] Specific examples

[2489] Example 1: Rescue operations after an earthquake

[2490] 1. Device: User A is in an area close to the epicenter, so he obtains his location information using GPS and sends it to the server.

[2491] 2. Server: Integrates the acquired location information with real-time earthquake information.

[2492] 3. Server: Input the integrated data into the generative model and estimate the safety of user A.

[2493] 4. Generative model: Infers that user A is in a dangerous situation and returns the result to the server.

[2494] 5. Server: Since the user is judged to be in danger, safety information is sent to User A's emergency contact.

[2495] 6. Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue agencies.

[2496] Example 2: Identifying evacuees with a sense of sadness

[2497] 1. User: User B takes a photo of the situation at the evacuation shelter with his smartphone and uploads it to the server.

[2498] 2. Server: Passes the received photo to the emotion engine.

[2499] 3. Emotion engine: Analyzes user B's sadness from the image and recognizes his emotional state.

[2500] 4. Server: Based on the emotion analysis results, notify the emergency contact of User B’s emotional state and the evacuation shelter situation.

[2501] 5. Server: Based on the analysis report, accurate information is posted on the official social media accounts and requests for relief supplies as needed.

[2502] By incorporating emotion recognition, it becomes possible to grasp the situation of disaster victims in more detail and take appropriate measures quickly.

[2503] The processing flow will be explained below.

[2504] Data collection

[2505] Step 1:

[2506] Device: The user's smartphone periodically obtains its current location using its GPS function.

[2507] Step 2:

[2508] Device: Sends the acquired location information to the server.

[2509] Step 3:

[2510] Server: Obtains real-time disaster information from the Japan Meteorological Agency and earthquake early warning APIs.

[2511] Step 4:

[2512] Server: The acquired disaster information is stored in a database and linked to the user's location information.

[2513] Data analysis

[2514] Step 5:

[2515] Server: Integrates user location information and disaster information to create a dataset.

[2516] Step 6:

[2517] Server: Inputs the integrated data into the generative model.

[2518] Step 7:

[2519] Generative model: Estimates the user's safety based on the user's location information and the scale of the disaster.

[2520] Step 8:

[2521] Generative model: Returns the inference results to the server.

[2522] Safety registration and rescue request

[2523] Step 9:

[2524] Server: Determines whether the user is safe or unsafe based on the inference results returned by the generative model.

[2525] Step 10:

[2526] Server: If a user is deemed to be in danger, a notification is automatically sent to pre-registered emergency contacts, including the user's location and status.

[2527] Step 11:

[2528] Server: If necessary, it issues an emergency rescue request to the appropriate rescue agency. The rescue request includes the user's detailed location information and the disaster situation.

[2529] emotion recognition

[2530] Step 12:

[2531] User: Take photos and videos of disaster areas and evacuation centers and upload them to the server.

[2532] Step 13:

[2533] Server: Receives uploaded image and video data.

[2534] Step 14:

[2535] Server: Passes the received data to the emotion engine.

[2536] Step 15:

[2537] Emotion engine: Analyzes user emotions from images and videos and generates emotional states.

[2538] Step 16:

[2539] Server: Integrates the analysis results from the emotion engine and the safety estimation results to generate detailed user information.

[2540] Information dissemination

[2541] Step 17:

[2542] Server: Based on the consolidated user information, sends notifications to the appropriate emergency contacts.

[2543] Step 18:

[2544] Server: Integrates analytical results with data from other reliable sources.

[2545] Step 19:

[2546] Server: Disseminates reliable information to official social media accounts and related organizations.

[2547] Step 20:

[2548] Server: If necessary, requests for relief supplies and the current situation at evacuation centers will be sent out.

[2549] Specific examples

[2550] Example 1: Rescue operations after an earthquake

[2551] Step 1:

[2552] Device: User A is in an area close to the epicenter, so his location information is obtained using GPS.

[2553] Step 2:

[2554] Device: Sends the acquired location information to the server.

[2555] Step 3:

[2556] Server: Obtains real-time earthquake information from the Japan Meteorological Agency and Earthquake Early Warning API.

[2557] Step 4:

[2558] Server: Integrates the acquired earthquake information and location information.

[2559] Step 5:

[2560] Server: Inputs the integrated data into the generative model.

[2561] Step 6:

[2562] Generative model: Estimates the safety of user A and returns the result to the server.

[2563] Step 7:

[2564] Server: Based on the safety estimation result, it is determined that User A is in a dangerous situation.

[2565] Step 8:

[2566] Server: Sends safety information to User A's emergency contacts. This notification includes location information and dangerous situations.

[2567] Step 9:

[2568] Server: Based on User A's detailed location information, issues an emergency rescue request to nearby rescue organizations.

[2569] Example 2: Shelter situation report and emotion recognition

[2570] Step 1:

[2571] User: User B takes a photo of the situation at the evacuation shelter with his smartphone.

[2572] Step 2:

[2573] User: Uploads captured images and videos to the server.

[2574] Step 3:

[2575] Server: Receives uploaded image and video data.

[2576] Step 4:

[2577] Server: Passes the received data to the emotion engine.

[2578] Step 5:

[2579] Emotion engine: Analyzes user B's emotions from images and videos and generates their emotional state.

[2580] Step 6:

[2581] Server: Integrates the analysis results from the emotion engine with other data.

[2582] Step 7:

[2583] Server: Sends a notification to emergency contacts based on the integrated detailed information of User B.

[2584] Step 8:

[2585] Server: Posts the results of the emotion analysis and the situation at the evacuation center to the official social media account.

[2586] Step 9:

[2587] Server: If necessary, request relief supplies for evacuation centers.

[2588] By incorporating emotion recognition, it is possible to grasp the situation of disaster victims in more detail and take appropriate measure...

Claims

1. means for acquiring a current location from a user terminal; means for transmitting the acquired location information to a server; A means for the server to acquire weather data and disaster data; A means for integrating location information and disaster data and inputting them into the generative model; A means for the generative model to estimate the user's safety; A means for registering and notifying the user's safety information based on the estimation result; A means for uploading images and videos posted by users to the server, means for analyzing the uploaded data with an analytical model; A system that includes a means to transmit information and request rescue based on the analysis results.

2. The system according to claim 1 , further comprising means for the generative model to estimate the safety of the user in real time based on the user's location information and disaster data.

3. The system of claim 1 , further comprising means for the user terminal to transmit the estimated result to an emergency contact.

Images