System

Abstract

A system is provided.SOLUTION: A system comprising means for collecting position information (latitude, longitude, and altitude) every hour from a plurality of base stations installed all over the country, means for comparing the collected position information with previous position information and calculating horizontal and vertical displacement of the ground, means for inputting the calculated displacement data of the ground to a generative AI model and analyzing correlation with past earthquake data to predict an earthquake, means for notifying a self-governing body or a weather forecast service provider of a prediction result, and means for storing the prediction result in a database and archiving it in an analyzable form later.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] Earthquake forecasting remains extremely difficult even today, with many prediction systems currently lacking in accuracy. Therefore, in order to minimize damage from earthquakes, it is essential to develop a more accurate earthquake forecasting system. However, current earthquake prediction technology lacks the ability to measure minute ground shifts over a wide area with high accuracy in real time and to make immediate earthquake predictions based on that data. This has made real-time earthquake prediction a challenge. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country and a means for calculating horizontal and vertical ground displacement by comparing the collected location information with previous location information. The system also includes a means for inputting the calculated ground displacement data into a generative AI model and analyzing the correlation with past earthquake data to forecast earthquakes. The forecast results are notified to local governments and weather forecast service providers. The system also includes a means for storing the forecast results in a database and archiving them in a format that can be analyzed later. This significantly improves the accuracy of earthquake forecasts and minimizes damage caused by earthquakes. Furthermore, the system also includes filtering and data cleaning means for removing outliers and noise, and a feedback means for comparing actual earthquake occurrence conditions with the forecast results and improving the generative AI model, thereby achieving continuous improvement in accuracy.

[0006] A "base station" is a fixed transceiver that is part of a wireless communication network and is used to communicate with terminals.

[0007] "Location information" is data that represents the latitude, longitude, and altitude of a certain point, and is used to identify the geographic location of that point.

[0008] "Ground displacement" means any change in topography observed at the surface or underground of the Earth, and includes both horizontal and vertical displacements.

[0009] A "generative AI model" is a model designed using artificial intelligence technology that can learn patterns and correlations from large amounts of data and make predictions and classifications for new data.

[0010] "Earthquake forecast" refers to predicting the likelihood of an earthquake occurring at a certain time and place in the future, and providing the results of that prediction to the public.

[0011] "Municipality" refers to a local public body, an organization that provides administrative services to local residents.

[0012] "Weather forecasting service provider" means a company or organization that specializes in forecasting and providing information about weather phenomena.

[0013] A "database" is a system for systematically storing, managing, and searching large amounts of information.

[0014] "Filtering" refers to the process of removing unnecessary parts from data, which is done to remove noise and outliers.

[0015] "Data cleaning" is the process of correcting or removing erroneous, missing, or incomplete information from collected data.

[0016] "Archiving" refers to the long-term storage of important data and information in a form that can be accessed at a later date.

[0017] "Feedback measures" refer to methods and techniques for evaluating the results of a system or process and making improvements based on that evaluation. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0019] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0021] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0022] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0023] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0024] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0026] [First embodiment]

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

[0028] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0029] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0031] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0035] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0038] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0039] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses an AI model to forecast earthquakes.

[0040] 1. Location information collection from base stations

[0041] The server first collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed across the country. This makes it possible to obtain basic data for understanding ground movement in real time. For example, at midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. This location information is then stored in a database.

[0042] 2. Comparing location information and calculating ground displacement

[0043] The server compares the latest location information collected with previous location information and calculates the horizontal and vertical deviation of the ground. This comparison allows the server to determine how much the location of each base station has changed. For example, if the location of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), it is confirmed that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0044] 3. Noise removal and data cleaning

[0045] The server then performs filtering and data cleaning to remove outliers and noise from the calculated data, allowing for more accurate earthquake forecasts. For example, data containing obvious outliers or noise is ignored, and the location information is re-acquired.

[0046] 4. Earthquake Forecasting Using Generative AI Models

[0047] The server inputs the cleaned ground displacement data into a generative AI model and analyzes correlations with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur.

[0048] 5. Notification of prediction results

[0049] The server then notifies local governments and weather forecast service providers of the generated forecasts. This notification is done via email or API request, and a detailed report of the forecast results is also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly notified to local governments so that they can prepare countermeasures.

[0050] 6. Data Storage and Archiving

[0051] The server stores the prediction results and location information in a database and archives them in a format that can be analyzed later. This data will be used to improve future prediction models and develop new earthquake forecasting technologies. For example, monthly earthquake prediction results will be accumulated to help improve the accuracy of long-term earthquake predictions.

[0052] In this way, by providing a system that handles everything from collecting location information from base stations to data analysis, earthquake forecasting, notification of results, and data storage, it becomes possible to minimize damage caused by earthquakes. The system as a whole also achieves continuous improvement in accuracy, enabling more accurate earthquake forecasts over a wider area.

[0053] The processing flow will be explained below.

[0054] Step 1:

[0055] The server collects location information (latitude, longitude, and altitude) every hour from base stations across the country. The server sends a request to the API endpoint of each base station to obtain the location information. At this stage, the collected location information is temporarily stored in memory.

[0056] Step 2:

[0057] The server stores the collected location information in a database. When storing, it assigns a timestamp to the location information of each base station and stores it in a specific table. This allows for efficient comparison of past location information with the latest location information.

[0058] Step 3:

[0059] The server compares the latest saved location information with previous location information to calculate the horizontal and vertical deviation of the ground. For example, it calculates the positional fluctuation of each base station over the past hour and outputs the horizontal and vertical deviation as specific numerical values.

[0060] Step 4:

[0061] The server processes the calculated deviation data through filtering and data cleaning. It detects noise and outliers and removes or corrects them. For example, if an abnormally large fluctuation is detected, it will be removed or the data will be collected again.

[0062] Step 5:

[0063] The server inputs the cleaned ground displacement data into the generative AI model. It combines this with past earthquake data and executes earthquake forecasts using the AI ​​model. For example, if a similar displacement pattern appears in the past as a precursor to an earthquake, the AI ​​model can use this to predict the probability of an earthquake occurring.

[0064] Step 6:

[0065] The server evaluates the risk of earthquakes based on the prediction results and notifies local governments and weather forecast service providers. Notifications are sent via email or API requests, providing detailed forecast reports. For example, a warning could be sent to local governments stating that there is a high risk of earthquakes in the Tokyo area.

[0066] Step 7:

[0067] The server stores and archives the prediction results and analysis data in a database, making them available for future analysis and model improvement. The stored data is time-stamped and stored in a format that allows for easy comparison with past data.

[0068] Step 8:

[0069] The server compares the actual earthquake occurrences with the predicted results to evaluate the accuracy of the generative AI model. Based on the evaluation results, it updates the model's training dataset and continuously improves the accuracy of the AI ​​model through feedback means. For example, it checks whether past predictions matched actual earthquake occurrences and incorporates that information into the next training dataset.

[0070] Example 1

[0071] 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."

[0072] Earthquakes are difficult to predict, and accurate and prompt earthquake forecasts are necessary to minimize the damage caused by their occurrence. However, conventional methods have limitations in the accuracy and speed of earthquake forecasts, and many challenges remain. In particular, it is technically difficult to monitor ground movements across the country in real time and make highly accurate predictions based on that information. Therefore, there is a demand for a system that can achieve more accurate, real-time earthquake forecasts and provide prompt notifications.

[0073] 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.

[0074] In this invention, the server includes means for collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country, means for comparing the collected location information with previous location information and calculating horizontal and vertical displacement of the ground, means for inputting the calculated ground displacement data into a generative model and analyzing correlations with past earthquake data to forecast earthquakes, means for notifying administrative agencies and weather forecast providers of the results of the forecast, and means for storing the results of the forecast in an information accumulation device and archiving them in a form that can be analyzed later. This makes it possible to monitor ground movements throughout the country in real time, realize highly accurate earthquake forecasts, and provide information promptly.

[0075] A "base station" is a piece of equipment installed throughout the country that is used to obtain location information (latitude, longitude, and altitude) within a specific area.

[0076] "Location information" is data of latitude, longitude, and altitude acquired by a base station, and is information indicating a location at a specific point in time.

[0077] The "server" is a central computing device that processes location information collected from base stations and performs earthquake forecasting.

[0078] A "generative model" is an algorithm or machine learning model used to analyze correlations with past data and make earthquake forecasts.

[0079] "Filtering" is the process of removing outliers and noise from collected data.

[0080] "Data cleaning" is a process carried out to improve the quality of data by correcting or removing inaccurate or invalid data.

[0081] An "information accumulation device" is a database or other storage device that stores prediction results and location information and archives them in a form that can be analyzed later.

[0082] "Administrative agencies" are public institutions such as local governments and government agencies.

[0083] A "weather forecast provider" is a company or organization that provides weather information.

[0084] "Ground displacement" refers to the change in latitude, longitude, and altitude of a particular point over time, including horizontal and vertical variations.

[0085] "Real-time" refers to near-instantaneous processing or reaction, meaning minimal time delay.

[0086] MODE FOR CARRYING OUT THE INVENTION

[0087] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses a generative AI model to forecast earthquakes. Specifically, the following hardware and software are used to process and calculate the data.

[0088] Hardware and Software

[0089] Server: A central computing device for data processing, collecting location information, comparing and analyzing data, generating earthquake forecasts, sending notifications, and storing data.

[0090] Base station: A facility installed throughout the country to obtain location information (latitude, longitude, altitude) within a specific area.

[0091] Database: An information collection device that stores collected location information and prediction results and archives them as needed.

[0092] Generative model: A machine learning model built using Python's TensorFlow, which performs correlation analysis with past earthquake data.

[0093] Program processing

[0094] The server collects location information every hour from base stations installed throughout the country. This information is used as basic data to understand ground movement in real time. For example, at midnight, the server sends an API request to base stations nationwide and receives the latest location information from each base station. The received location information is stored in a database.

[0095] The server then compares the latest location information collected with the previous location information and calculates the horizontal and vertical deviation of the ground. For example, if base station A's location changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), it determines that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0096] The server performs filtering and data cleaning to remove outliers and noise from the calculated data. This allows earthquake forecasts to be made based on more accurate data. For example, data containing obvious outliers or noise is ignored, and the location information is retrieved again.

[0097] The server inputs the cleaned ground displacement data into a generative AI model and analyzes correlations with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur.

[0098] The forecast results are then sent to local governments and weather forecasting service providers. This notification is sent via email or API request, and detailed reports are also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly sent to local governments, allowing them to prepare countermeasures.

[0099] The server then stores the prediction results and location information in a database, archiving them for future analysis. This data will be used to improve future prediction models and develop new earthquake forecasting technologies. For example, monthly earthquake prediction results can be stored in a database to help improve the accuracy of long-term earthquake predictions.

[0100] Examples of specific examples and prompts

[0101] Specific examples

[0102] 1. Location information collection: The location information obtained by the server from the base station is (35.6895, 139.6917, 30).

[0103] 2. Compare and calculate: If the server detects that the same base station's location information has changed to (35.6896, 139.6918, 29.9) one hour later, the server will determine that there has been a shift of 0.0001 degrees horizontally and -0.1 meters vertically.

[0104] 3. Data cleaning: The server detects abnormal outliers and noise, and either removes them or re-acquires them to clean up the data.

[0105] 4. AI forecasting: The server inputs the cleaned-up deviation data into a generative AI model, which then compares it with past data to predict the likelihood of an earthquake.

[0106] 5. Notification: The server notifies the local government by email based on the prediction results and provides information in real time via a specific API.

[0107] 6. Data storage: The server stores all results and location data in a database for further analysis and model improvement.

[0108] Prompt Sentence Examples

[0109] "Please explain the system that uses location information data collected from base stations around the country to analyze correlations with past earthquake data and predict earthquakes. Please explain in detail what hardware and software is required, including specific steps. Also, please explain how the prediction results will be notified and how the data will be stored."

[0110] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0111] Step 1:

[0112] Input: Latest location information (latitude, longitude, altitude) from multiple base stations installed throughout the country

[0113] Processing: The server first collects location information every hour from multiple base stations installed across the country. The server then sends an API request to each base station, and each base station responds by sending back its location information.

[0114] Output: Retrieved location information

[0115] Specific operation: At midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. For example, it obtains the location information (35.0116, 135.7681, 45) from the base station in Kyoto and stores it in the database.

[0116] Step 2:

[0117] Input: Latest saved location and previous location

[0118] Processing: The server compares the latest collected position information with previous position information and calculates the horizontal and vertical deviation of the ground.

[0119] Output: Calculated ground displacement data

[0120] What it does: The server compares the latest location with the previous location stored in the database. For example, if the previous location is (35.0116, 135.7681, 45) and the latest location is (35.0117, 135.7682, 44.9), it calculates a horizontal offset of 0.0001 degrees and a vertical offset of -0.1 meters.

[0121] Step 3:

[0122] Input: Calculated ground displacement data

[0123] Processing: The server performs filtering and data cleaning to remove outliers and noise from the calculated data.

[0124] Output: Cleaned up ground displacement data

[0125] What it does: The server uses a specific algorithm to filter out abnormal data and noise. For example, if the location information suddenly deviates significantly, the data will be considered as noise. Such abnormal values ​​will be filtered out, and the server will process the location information again if necessary.

[0126] Step 4:

[0127] Input: Cleaned ground displacement data

[0128] Processing: The server inputs the cleaned-up ground displacement data into a generative AI model, analyzes the correlation with past earthquake data, and makes earthquake forecasts.

[0129] Output: Earthquake forecast results

[0130] How it works: The server inputs clean ground displacement data into a generative AI model. The generative AI model uses an AI model (e.g., Python's TensorFlow) to perform correlation analysis with past earthquake data and predict the likelihood of an earthquake. If a specific displacement pattern is similar to past earthquakes, it determines that there is a high probability of an earthquake occurring at that location.

[0131] Step 5:

[0132] Input: Earthquake forecast results

[0133] Processing: The server notifies the local government and weather forecast service providers of the earthquake forecast results.

[0134] Output: Notified earthquake forecast information

[0135] Specific operation: The server sends the generated forecast results to local governments and weather forecast service providers via API requests or email. For example, if a high risk of an earthquake is predicted in the Tokyo area, that information will be immediately sent via email with a detailed report attached.

[0136] Step 6:

[0137] Input: Prediction results and location data

[0138] Processing: The server stores the prediction results and location data in a database and archives them in a form that can be analyzed later.

[0139] Output: Stored and archived data

[0140] How it works: The server stores all prediction results and location data. This data is then used for later analysis and to improve the generative AI model. For example, monthly earthquake prediction results are accumulated to help improve the accuracy of long-term earthquake predictions.

[0141] Through these processing steps, a system will be created that can monitor ground movements across the country in real time, achieve highly accurate earthquake forecasts, and provide information quickly.

[0142] (Application example 1)

[0143] 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."

[0144] Rapid and accurate earthquake forecasts are necessary to minimize damage caused by earthquakes. However, existing earthquake forecasting systems require time to collect and analyze information, making it difficult to make real-time predictions and notifications. Furthermore, they often fail to provide users with prompt notifications or specific evacuation guidance, resulting in inadequate emergency response. Furthermore, technology to effectively remove outliers and noise from analytical data has not yet been fully established, so there is a need to improve the accuracy of earthquake forecasts.

[0145] 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.

[0146] In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical ground displacement; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to forecast earthquakes; means for notifying local governments and weather forecast service providers of the forecast results; means for storing the forecast results in a database and archiving them in a format that can be analyzed later; means for notifying users of the earthquake forecast results in real time via their smartphones or smart glasses; and means for providing safe evacuation routes in the event of an earthquake. This enables accurate real-time earthquake forecasts, enabling users to be notified promptly and provided with specific evacuation guidance. Furthermore, the accuracy of earthquake forecasts can be improved by removing outliers and noise from the analysis data.

[0147] A "base station" is a device installed to conduct wireless communication within a specific area, and is the infrastructure for collecting location information.

[0148] "Location information" means data relating to the latitude, longitude, and altitude of a particular point.

[0149] "Ground displacement" refers to numerical data that indicates the degree to which the ground has moved or deformed horizontally and vertically.

[0150] A "generative AI model" is a mathematical algorithm or model that uses artificial intelligence techniques to learn specific patterns and make predictions or classifications.

[0151] "Earthquake forecast" means providing information about the possibility of future earthquakes and the predicted timing and location of such occurrences.

[0152] "Real-time" is a term that indicates that the time between the collection of information, its analysis, and the notification of the results is extremely short and almost simultaneous.

[0153] "Noise" refers to irregular or abnormal values ​​contained in data, which hinder accurate analysis.

[0154] "Data cleaning" is the process of removing noise and outliers from collected data and preparing it for analysis.

[0155] An "evacuation route" refers to the paths and procedures that have been set up in advance to allow safe evacuation in the event of a disaster such as an earthquake.

[0156] A "smartphone" is a multi-functional mobile phone terminal that can connect to the Internet.

[0157] "Smart glasses" are eyeglass-type devices with built-in displays that directly display visual information, and can connect to the Internet and use applications.

[0158] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses a generative AI model to forecast earthquakes. This system notifies users of earthquake forecast results in real time via smartphones or smart glasses, and also has the function of providing safe evacuation routes in the event of an earthquake.

[0159] Hardware and software used

[0160] The system is implemented using the following hardware and software:

[0161] Hardware: Cloud servers, smartphones, smart glasses, wireless base stations

[0162] Software: Database management systems (e.g., AWS RDS), machine learning frameworks (e.g., TensorFlow, Keras), communication APIs, notification systems (e.g., Firebase Cloud Messaging)

[0163] Collecting location information and calculating ground displacement

[0164] The server first collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed across the country. This makes it possible to obtain basic data for understanding ground movement in real time. For example, at midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. This location information is then stored in a database.

[0165] The server then compares the latest location information collected with previous location information to calculate the horizontal and vertical deviations of the ground. This comparison allows the server to determine how much the location of each base station has changed. For example, if base station A's location changes from latitude 35.6895 degrees, longitude 139.6917 degrees, and altitude 30 meters to latitude 35.6896 degrees, longitude 139.6918 degrees, and altitude 29.8 meters, it is determined that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0166] Noise removal and data cleaning

[0167] The server then performs filtering and data cleaning to remove outliers and noise from the calculated data, enabling earthquake forecasts to be made based on more accurate data. For example, data containing obvious outliers or noise is ignored, and the location information is re-acquired.

[0168] Earthquake forecasting using generative AI models

[0169] The server inputs the cleaned ground displacement data into a generative AI model and analyzes the correlation with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur. As a concrete example, the following data is input into the prompt:

[0170] "latitude_diff:0.0001, longitude_diff:0.0001, altitude_diff:-0.2"

[0171] Notification of forecast results and evacuation routes

[0172] The server then notifies local governments and weather forecast service providers of the generated forecasts. This notification is done via email or API request, and a detailed report of the forecast results is also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly notified to local governments so that they can prepare countermeasures.

[0173] The server also notifies users of earthquake forecast results in real time via their smartphones or smart glasses. Users can receive earthquake forecast information for their region through the application and receive alerts in the event of an emergency. Information on evacuation routes is also provided, allowing users to take appropriate action to evacuate safely.

[0174] In this way, by providing a system that handles everything from collecting location information from base stations to data analysis, earthquake forecasting, notification of results, and data storage, it becomes possible to minimize damage caused by earthquakes. The system as a whole also achieves continuous improvement in accuracy, enabling more accurate earthquake forecasts over a wider area.

[0175] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0176] Step 1:

[0177] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed nationwide. This input data includes the current location information of each base station. The server sends a request to the API of each base station, receives the latest location information data returned, and stores it in a database.

[0178] Step 2:

[0179] The server compares the latest collected location information with previously saved location information to calculate the horizontal and vertical deviation of the ground. This comparison calculates the difference between the current location information received as input and the previous location information, and generates position fluctuation data for each base station as output. For example, if the position of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), the output shows that a deviation of 0.0001 degrees occurred horizontally and -0.2 meters vertically.

[0180] Step 3:

[0181] The server performs filtering and data cleaning to remove outliers and noise from the calculated ground displacement data. The input for this step is the positional fluctuation data of each base station. The server filters this data to remove noise and outliers and outputs clean ground displacement data.

[0182] Step 4:

[0183] The server inputs the clean ground displacement data into the generative AI model and analyzes the correlation with past earthquake data to forecast earthquakes. In this step, filtered ground displacement data is used as input. The server uses the generative AI model to compare and analyze this data with past earthquake patterns to predict the possibility of an earthquake occurring. The output is the earthquake forecast result. As a concrete example, enter the following prompt sentence: "latitude_diff: 0.0001, longitude_diff: 0.0001, altitude_diff: -0.2"

[0184] Step 5:

[0185] The server notifies the generated forecast results to local governments and weather forecast service providers. The input for this step is the earthquake forecast results obtained in step 4. Based on these results, the server notifies relevant organizations of the details of the forecast results via email or API request. For example, if the risk of an earthquake occurring in the Tokyo area increases, the server can quickly notify the local government of this information so that they can prepare countermeasures.

[0186] Step 6:

[0187] The server stores the prediction results in a database and archives them for future analysis and improvement of the AI ​​model. The inputs for this step are the prediction results from step 4 and the notification results generated in step 5. By storing these data in the database, they can be used later for analysis and to improve the accuracy of the model.

[0188] Step 7:

[0189] The server notifies the user of the earthquake forecast results in real time via their smartphone or smart glasses. The input of this step is the earthquake forecast results obtained in step 4. The server uses the notification system to immediately send an alert to the smartphone or smart glasses, informing the user of the risk of an earthquake. In addition, it also provides safe evacuation route guidance as an output.

[0190] 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.

[0191] The present invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses an AI model to forecast earthquakes.In addition, by combining this with an emotion engine that recognizes the user's emotions, the content of earthquake forecast notifications can be optimized and the user's reactions analyzed.

[0192] 1. Location information collection from base stations

[0193] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed nationwide. The server sends a request to the API endpoint of each base station to obtain the location information. At this stage, the collected location information is temporarily stored in memory.

[0194] 2. Saving and comparing location information

[0195] The server stores the collected location information in a database and compares it with past location information, thereby calculating the horizontal and vertical deviation of the ground. For example, if the location of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8) by comparing the previous location information stored in the database with the new location information, it can be determined that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0196] 3. Noise removal and data cleaning

[0197] The server filters and cleans the calculated deviation data to remove or correct noise and outliers, for example, ignoring obviously abnormal or noisy data and trying to obtain the correct data again.

[0198] 4. Earthquake Forecasting Using Generative AI Models

[0199] The server inputs the cleaned ground displacement data into a generative AI model, which then combines it with past earthquake data to make predictions. The AI ​​model detects specific displacement patterns and, if they are similar to past earthquake precursors, predicts the probability of an earthquake occurring.

[0200] 5. User Emotion Recognition by Emotion Engine

[0201] An emotion engine is installed on the device (user's smartphone or computer) and recognizes the user's emotion by analyzing the user's facial expression and voice data when receiving the earthquake forecast result. For example, if the user's facial expression upon receiving the notification is recognized as "surprise" or "anxiety," that information is sent to the server.

[0202] 6. Utilizing Emotional Data

[0203] The server uses the emotion data obtained from the emotion engine to optimize the content of the earthquake forecast notification. For example, if the user is feeling "anxious," the server will provide the notification content in kind language and with additional reassuring information.

[0204] 7. Notification of prediction results

[0205] The server then sends the optimized forecast results to the local government and users' devices via email or app notification, providing a detailed forecast report. User responses and other information are used to make the next forecast.

[0206] 8. Data Storage and Archiving

[0207] The server stores and archives the analysis data, including prediction results and user sentiment data, in a database, which will be used as important information for future analysis and model improvement.

[0208] 9. Model Improvement through Feedback Methods

[0209] The server compares the actual earthquake occurrence situation with the predicted results to evaluate the accuracy of the generated AI model. It also evaluates user emotional data and updates the learning dataset of the AI ​​model through feedback means to continuously improve accuracy.

[0210] In this way, a system that combines an emotion engine enables flexible earthquake forecast notifications that correspond to the user's emotional state, and can also contribute to improving the accuracy of the model.

[0211] The processing flow will be explained below.

[0212] Step 1:

[0213] The server collects location information (latitude, longitude, and altitude) every hour from base stations across the country. The server sends a request to the API endpoint of each base station to obtain the current location information. The location information obtained at this stage is temporarily stored in memory.

[0214] Step 2:

[0215] The server stores the collected location information in a database, adding a timestamp to the information and storing it in a database table organized by date and time, making it easy to compare previous and current location information.

[0216] Step 3:

[0217] The server compares the latest stored location information with previous location information and calculates the horizontal and vertical deviation of the ground. Specifically, it calculates the change in latitude, longitude, and altitude for each base station to derive the horizontal and vertical deviation values.

[0218] Step 4:

[0219] The server filters and cleans the calculated deviation data, detecting and removing outliers and noise. For example, if an abnormally large fluctuation is detected, the server will either ignore the data or try collecting it again.

[0220] Step 5:

[0221] The server inputs the cleaned ground displacement data into a generative AI model. The generative AI model analyzes correlations with past earthquake data and predicts the probability of an earthquake. For example, if a specific displacement pattern is similar to that observed before a past earthquake, it determines that there is a high probability of an earthquake occurring in that area.

[0222] Step 6:

[0223] The server temporarily stores the earthquake forecast results obtained from the generative AI model and then notifies local governments and weather forecast service providers via email or API requests, providing detailed forecast reports.

[0224] Step 7:

[0225] The device notifies the user of the earthquake forecast results and simultaneously runs an emotion engine to collect the user's emotional data. Specifically, when the user receives the notification, the device uses a camera and microphone to collect facial and voice data and analyzes their emotions.

[0226] Step 8:

[0227] The server optimizes the notification content based on the user's emotional data obtained from the emotion engine. For example, if the user is feeling "anxiety" or "fear," the notification content will include additional reassurance and supportive messages to alleviate the user's anxiety.

[0228] Step 9:

[0229] The server then sends the optimized prediction results and notification content back to the local government or user's device. Notifications are sent immediately via email or app notification, and users receive detailed prediction reports and emotionally sensitive messages.

[0230] Step 10:

[0231] The server stores the prediction results and user emotion data in a database and archives them in a format that can be analyzed later. This data will be used for future analysis and model improvement.

[0232] Step 11:

[0233] The server compares the actual earthquake occurrence situation with the predicted results to evaluate the accuracy of the generated AI model. It also evaluates user emotional data and updates the AI ​​model's learning dataset with this information as a means of feedback, thereby continuously improving the model's accuracy.

[0234] Through the above steps, a system incorporating an emotion engine will be able to provide flexible earthquake forecast notifications that take into account the user's emotions, and will also contribute to improving the accuracy of the model.

[0235] Example 2

[0236] 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."

[0237] Conventional earthquake forecasting systems were capable of making predictions based on ground displacement data and past earthquake data, but the notification of the prediction results was uniform and did not take into account the individual emotions of users. This could cause users to feel anxious or confused. Furthermore, to improve the accuracy of the prediction model, it was necessary to take into account user feedback and emotional data, but this had not been realized. Furthermore, data cleaning to remove outliers and noise was insufficient, which could lead to a decrease in prediction accuracy.

[0238] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical ground displacement; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to perform earthquake forecasting; means for recognizing the user's emotions using an emotion engine in the user terminal and reflecting the data in generating optimized notification content; means for notifying local governments, weather forecast service providers, and users of the prediction results; and means for storing the prediction results in a database and archiving them in a form that can be analyzed later. This makes it possible to provide flexible notification content that takes into account the emotions of individual users and improve the accuracy of the prediction model based on feedback.

[0239] A "base station" is a communication facility installed throughout the country that provides location information (latitude, longitude, and altitude).

[0240] "Location information" is data relating to geographic latitude, longitude, and altitude.

[0241] "Horizontal and vertical deviation" is data indicating the amount of horizontal and vertical displacement relative to a specific position.

[0242] A "generative AI model" is an artificial intelligence model that analyzes ground displacement data and past earthquake data to carry out earthquake forecasting.

[0243] An "emotion engine" is software or hardware that analyzes a user's facial expressions and voice data to recognize emotions.

[0244] "Data cleaning" is a process for removing outliers and noise and shaping collected data accurately and appropriately.

[0245] The "feedback method" is a process for comparing actual earthquake occurrence conditions with predicted results to evaluate and improve the accuracy of the generative AI model.

[0246] "Notification content" is information for informing users and local governments of earthquake forecast results.

[0247] "Means for collecting" refers to the process and devices for obtaining location information from the base station to the server.

[0248] "Means for comparison" refers to the processes and algorithms for comparing collected location information with past location information and calculating ground displacement.

[0249] A "database" is a digital storage system for storing location information, prediction results, emotional data, etc.

[0250] "Archiving measures" are processes and systems for storing data for long periods of time in an analyzable form.

[0251] This invention relates to a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that location information, and uses a generative AI model to forecast earthquakes.Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system optimizes the content of earthquake forecast notifications and analyzes the user's reactions.

[0252] First, the server collects hourly location information (latitude, longitude, and altitude) from multiple base stations installed nationwide. Specifically, the server sends an HTTP request to the API endpoint of each base station to obtain latitude, longitude, and altitude information. At this stage, the collected location information is temporarily stored in memory.

[0253] The server then stores the collected location information in a database and compares it with past location information to calculate the horizontal and vertical deviations of each base station's location. For example, the server compares the previous location information with the newly acquired location information to specifically calculate the horizontal and vertical deviations.

[0254] The server then filters the calculated deviation data to remove noise and outliers, improving the accuracy of the data. Any obviously outliers are ignored and a re-acquisition attempt is made.

[0255] The server then inputs the cleaned ground displacement data into a generative AI model, which analyzes correlations with past earthquake data to predict earthquakes. The AI ​​model detects specific displacement patterns and, if they resemble past earthquake precursors, predicts the probability of an earthquake occurring. A specific prompt is, "Predict the likelihood of the next earthquake based on earthquake data from the past year and current ground displacement data."

[0256] The device (user's smartphone or computer) is equipped with an emotion engine that analyzes the user's facial expression and voice data when receiving the earthquake forecast result to recognize their emotion. For example, if the user's facial expression upon receiving the notification is recognized as "surprise" or "anxiety," that information is sent to the server.

[0257] Next, the server uses the emotion data obtained from the emotion engine to optimize the content of the earthquake forecast notification. For example, if the user is feeling "anxious," the server will provide the notification content in a gentler tone and with additional reassuring information. As a specific example, it will generate a notification that reads, "The possibility of an earthquake is increasing in your area. There is no need to feel anxious. Please check your disaster prevention measures."

[0258] The server then sends the optimized forecast results to the local government and the user's device via email or app notification, and provides a detailed forecast report. User responses are also used to make the next forecast.

[0259] Finally, the server stores the prediction results and user emotion data in a database and archives them in an analyzable format. This provides valuable information for future analysis and model improvement. Furthermore, the server compares the prediction results with actual earthquake occurrence conditions, providing a feedback mechanism for evaluating the accuracy of the generative AI model, allowing for continuous improvement of the model.

[0260] In this way, the present invention enables flexible earthquake forecast notifications that correspond to the user's emotional state through a system that combines an emotion engine, and can also contribute to improving the accuracy of the model.

[0261] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0262] Step 1:

[0263] The server collects hourly location information (latitude, longitude, altitude) from multiple base stations installed nationwide. The API endpoint of each base station is used as input. Specifically, the server sends an HTTP GET request to the API endpoint and temporarily stores the acquired location data (latitude, longitude, altitude) in memory. The output of this step is the collected location data.

[0264] Step 2:

[0265] The server stores the collected location information in a database and compares it with past location information. The current location data and past location information stored in the database are used as input. Specifically, the server inserts the collected location information into an SQL database and extracts past location information. Next, the server compares the new and old location information and calculates the horizontal and vertical displacement of the ground. The output of this step is the calculated horizontal and vertical displacement data.

[0266] Step 3:

[0267] The server filters the calculated deviation data to remove noise and outliers. The deviation data obtained in step 2 is used as input. Specifically, the server detects anomalous values ​​above a certain threshold and performs a data cleaning process to remove or correct them. The output of this step is the cleaned deviation data.

[0268] Step 4:

[0269] The server inputs the cleaned ground displacement data into the generative AI model and combines it with past earthquake data to make earthquake forecasts. The cleaned displacement data and past earthquake data are used as inputs. Specifically, the server formats this data and converts it into a format suitable for the generative AI model. The server then uses the prompt, "Predict the likelihood of the next earthquake based on the earthquake data from the past year and the current ground displacement data," to have the AI ​​model make a prediction. The output of this step is the probability of an earthquake occurring and the predicted results.

[0270] Step 5:

[0271] The device (user's smartphone or computer) implements an emotion engine and receives a notification of the earthquake forecast result. The prediction result obtained in step 4 is used as input. Specifically, when the device receives the notification, it activates the camera and microphone to collect and analyze the user's facial expression and voice data to recognize the emotion. The output of this step is the user's emotion data.

[0272] Step 6:

[0273] The server optimizes the notification content of the earthquake forecast results using the emotion data obtained from the emotion engine. The emotion data and the earthquake forecast results are used as inputs. Specifically, the server runs an algorithm that uses the user's emotion data to generate optimal notification content. For example, if the user is feeling "anxious," the notification content will be provided with gentle language and reassuring information. The output of this step is the optimized notification content.

[0274] Step 7:

[0275] The server notifies the local government and user devices of the optimized prediction results. The optimized notification content and the prediction results are used as input. Specifically, the server references the local government and user contact information and selects the appropriate notification method (e.g., email, SMS, app notification, etc.). The output of this step is the sent notification.

[0276] Step 8:

[0277] The server stores the prediction results and user emotion data in a database and archives them in an analyzable format. The prediction results and emotion data are used as input. Specifically, the server inserts these data into the database and periodically moves them to archive storage for backup. The output of this step is the stored and archived data.

[0278] Step 9:

[0279] The server compares the actual earthquake occurrence situation with the predicted results, evaluates the accuracy of the generated AI model, and provides feedback. Actual earthquake occurrence data and predicted results are used as input. Specifically, the server collects actual earthquake occurrence data and compares it with the predicted results to evaluate the accuracy of the model. Emotion data is also included in the evaluation, and the AI ​​model's training dataset is updated through feedback means, allowing for continuous improvement of the model. The output of this step is an updated AI model and a new training dataset.

[0280] (Application example 2)

[0281] 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."

[0282] Current earthquake forecasting systems aim to improve prediction accuracy, but lack consideration for the reliability of the forecast and the user's psychological state. In particular, no systems exist that take into account how earthquake forecasts actually affect users. This raises concerns that users may feel unnecessary anxiety or surprise. Therefore, it is necessary to provide a system that can accurately and quickly notify users of earthquake forecasts while taking into account their emotions.

[0283] 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.

[0284] In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical displacement of the ground; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to make earthquake forecasts; means for recognizing the user's emotions using voice and image data and optimizing the notification content based on the emotion data; and means for storing the prediction results and emotion data in a database and archiving them in a form that can be analyzed later. This enables earthquake forecast notifications that take the user's psychological state into consideration, making it possible to provide fast and accurate earthquake forecasts while reducing the user's anxiety.

[0285] A "base station" is a device installed in a communications network to provide location information.

[0286] "Location information" is data including latitude, longitude, and altitude, and is information for identifying a geographical location.

[0287] "Ground displacement" is part of the crustal movement that occurs due to the effects of earthquakes and other events, and refers to changes in position in the horizontal and vertical directions.

[0288] A "generative AI model" is an artificial intelligence model used to predict earthquake occurrences using past earthquake data and other input data.

[0289] A "database" is a system for systematically storing and managing data.

[0290] "Emotion recognition" is a technology that analyzes a user's voice and image data to recognize the user's emotional state (for example, anxiety or surprise).

[0291] "Optimizing notification content" refers to adjusting the content of a notification message depending on the emotional state of the user receiving the notification.

[0292] "Archiving" means storing data for the long term so that it can be analyzed or referenced later.

[0293] "Noise reduction" is the process of removing outliers and irrelevant values ​​in data processing.

[0294] "Feedback measures" are methods for comparing actual earthquake occurrence conditions with predicted results to improve the accuracy of the generative AI model.

[0295] This paper explains a system that collects location information (latitude, longitude, altitude) every hour from multiple base stations installed throughout the country and calculates ground displacement based on this information. The core part of the system is executed by the server, terminals, and users.

[0296] The server first collects location information from base stations across the country using API endpoints. The technology used includes the Python requests library. The collected data is temporarily stored in memory, and then the location information is stored in a database, such as Databricks.

[0297] The server then compares the data with past position information to calculate the horizontal and vertical displacement of the ground. This analysis is performed using the GeoAnalytics library. The analyzed data is then filtered to remove noise and data cleaning, and outliers are removed. This process is also performed using the GeoAnalytics library.

[0298] The cleaned ground displacement data is then input into a generative AI model. This model uses machine learning libraries such as TensorFlow to analyze correlations with past earthquake data and generate earthquake forecasts. The results of this forecast are then sent to local governments and weather forecast service providers via a server.

[0299] The device recognizes emotions using the user's voice and image data. It uses the smartphone's camera and microphone to analyze the user's emotions using the OpenCV and Pyaudio libraries. The analysis results are sent to the server, which then optimizes the notification content based on the emotion. The NotificationEngine library is used for this purpose.

[0300] Furthermore, these prediction results and emotion data are stored in a database. This storage allows for post-analysis of the prediction results and serves as a feedback tool to help improve the AI ​​model. Data obtained from the emotion engine is also archived, with the aim of improving the accuracy of the model in the future.

[0301] To maximize the effectiveness of this system, the following specific example is given. For example, if the earthquake prediction result is highly likely and the user shows a surprised expression, the notification content can be customized as follows: "Notice. An earthquake of magnitude 6 has been predicted. However, please remain calm and be prepared to evacuate quickly. Necessary support information will also be provided."

[0302] The following is an example of a prompt sentence to input to the generative AI model.

[0303] You received a high probability earthquake forecast. What reassuring notification message would be appropriate for a user who recognized a surprised expression?

[0304] In this way, a system is constructed that takes into account the user's psychological state and provides accurate and prompt earthquake forecast notifications.

[0305] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0306] Step 1:

[0307] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations across the country. It sends a request to the API endpoint of each base station to obtain location information. The collected location information is temporarily stored in memory.

[0308] Input: Location information from base station

[0309] Output: Raw position information stored in memory

[0310] Step 2:

[0311] The server stores the collected location information in a database, then compares it with past location information to calculate horizontal and vertical displacements of the ground, thereby identifying fluctuations in location information.

[0312] Input: Raw location information stored in memory, past location information

[0313] Output: Calculated ground displacement data

[0314] Step 3:

[0315] The server filters and cleans the calculated ground displacement data, removing or correcting noise and outliers to produce a clean dataset.

[0316] Input: Calculated ground displacement data

[0317] Output: Clean ground displacement data

[0318] Step 4:

[0319] The server inputs clean ground displacement data into a generative AI model, which analyzes correlations with past earthquake data to forecast earthquakes. The generative AI model is trained using TensorFlow.

[0320] Input: Clean ground displacement data

[0321] Output: Earthquake forecast results

[0322] Step 5:

[0323] The device collects the user's voice and image data and uses them to recognize emotions. It uses OpenCV and Pyaudio libraries to analyze the user's facial expressions and voice to identify emotions.

[0324] Input: User's voice and image data

[0325] Output: Recognized emotion data

[0326] Step 6:

[0327] The server optimizes the notification content of earthquake forecast results based on emotion data. It uses the NotificationEngine library to generate customized notifications according to the emotion. For example, if the user is feeling "anxious," the server will provide the notification content with kind language and additional reassurance information.

[0328] Input: Earthquake forecast results, recognized emotion data

[0329] Output: Optimized notification content

[0330] Step 7:

[0331] The server then sends the earthquake forecast results and optimized notification content to local governments and users' devices via email and app notifications.

[0332] Input: Optimized notification content

[0333] Output: Notification to local government and user devices

[0334] Step 8:

[0335] The server stores and archives the prediction results and user emotion data in a database, which will be used as important information for future analysis and model improvement.

[0336] Input: Prediction results, user emotion data

[0337] Output: Analysis data stored in a database

[0338] Step 9:

[0339] The server compares the actual earthquake occurrence situation with the predicted results, and updates the learning dataset of the AI ​​model through a feedback means to improve the generated AI model, thereby continuously improving its accuracy.

[0340] Input: Actual earthquake occurrence situation, predicted results

[0341] Output: An improved generative AI model

[0342] 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.

[0343] 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.

[0344] 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.

[0345] [Second embodiment]

[0346] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0347] 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.

[0348] 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).

[0349] 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.

[0350] 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.

[0351] 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).

[0352] 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.

[0353] 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.

[0354] 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.

[0355] 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.

[0356] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0357] 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."

[0358] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses an AI model to forecast earthquakes.

[0359] 1. Location information collection from base stations

[0360] The server first collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed across the country. This makes it possible to obtain basic data for understanding ground movement in real time. For example, at midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. This location information is then stored in a database.

[0361] 2. Comparing location information and calculating ground displacement

[0362] The server compares the latest location information collected with previous location information and calculates the horizontal and vertical deviation of the ground. This comparison allows the server to determine how much the location of each base station has changed. For example, if the location of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), it is confirmed that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0363] 3. Noise removal and data cleaning

[0364] The server then performs filtering and data cleaning to remove outliers and noise from the calculated data, allowing for more accurate earthquake forecasts. For example, data containing obvious outliers or noise is ignored, and the location information is re-acquired.

[0365] 4. Earthquake Forecasting Using Generative AI Models

[0366] The server inputs the cleaned ground displacement data into a generative AI model and analyzes correlations with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur.

[0367] 5. Notification of prediction results

[0368] The server then notifies local governments and weather forecast service providers of the generated forecasts. This notification is done via email or API request, and a detailed report of the forecast results is also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly notified to local governments so that they can prepare countermeasures.

[0369] 6. Data Storage and Archiving

[0370] The server stores the prediction results and location information in a database and archives them in a format that can be analyzed later. This data will be used to improve future prediction models and develop new earthquake forecasting technologies. For example, monthly earthquake prediction results will be accumulated to help improve the accuracy of long-term earthquake predictions.

[0371] In this way, by providing a system that handles everything from collecting location information from base stations to data analysis, earthquake forecasting, notification of results, and data storage, it becomes possible to minimize damage caused by earthquakes. The system as a whole also achieves continuous improvement in accuracy, enabling more accurate earthquake forecasts over a wider area.

[0372] The processing flow will be explained below.

[0373] Step 1:

[0374] The server collects location information (latitude, longitude, and altitude) every hour from base stations across the country. The server sends a request to the API endpoint of each base station to obtain the location information. At this stage, the collected location information is temporarily stored in memory.

[0375] Step 2:

[0376] The server stores the collected location information in a database. When storing, it assigns a timestamp to the location information of each base station and stores it in a specific table. This allows for efficient comparison of past location information with the latest location information.

[0377] Step 3:

[0378] The server compares the latest saved location information with previous location information to calculate the horizontal and vertical deviation of the ground. For example, it calculates the positional fluctuation of each base station over the past hour and outputs the horizontal and vertical deviation as specific numerical values.

[0379] Step 4:

[0380] The server processes the calculated deviation data through filtering and data cleaning. It detects noise and outliers and removes or corrects them. For example, if an abnormally large fluctuation is detected, it will be removed or the data will be collected again.

[0381] Step 5:

[0382] The server inputs the cleaned ground displacement data into the generative AI model. It combines this with past earthquake data and executes earthquake forecasts using the AI ​​model. For example, if a similar displacement pattern appears in the past as a precursor to an earthquake, the AI ​​model can use this to predict the probability of an earthquake occurring.

[0383] Step 6:

[0384] The server evaluates the risk of earthquakes based on the prediction results and notifies local governments and weather forecast service providers. Notifications are sent via email or API requests, providing detailed forecast reports. For example, a warning could be sent to local governments stating that there is a high risk of earthquakes in the Tokyo area.

[0385] Step 7:

[0386] The server stores and archives the prediction results and analysis data in a database, making them available for future analysis and model improvement. The stored data is time-stamped and stored in a format that allows for easy comparison with past data.

[0387] Step 8:

[0388] The server compares the actual earthquake occurrences with the predicted results to evaluate the accuracy of the generative AI model. Based on the evaluation results, it updates the model's training dataset and continuously improves the accuracy of the AI ​​model through feedback means. For example, it checks whether past predictions matched actual earthquake occurrences and incorporates that information into the next training dataset.

[0389] Example 1

[0390] 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."

[0391] Earthquakes are difficult to predict, and accurate and prompt earthquake forecasts are necessary to minimize the damage caused by their occurrence. However, conventional methods have limitations in the accuracy and speed of earthquake forecasts, and many challenges remain. In particular, it is technically difficult to monitor ground movements across the country in real time and make highly accurate predictions based on that information. Therefore, there is a demand for a system that can achieve more accurate, real-time earthquake forecasts and provide prompt notifications.

[0392] 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.

[0393] In this invention, the server includes means for collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country, means for comparing the collected location information with previous location information and calculating horizontal and vertical displacement of the ground, means for inputting the calculated ground displacement data into a generative model and analyzing correlations with past earthquake data to forecast earthquakes, means for notifying administrative agencies and weather forecast providers of the results of the forecast, and means for storing the results of the forecast in an information accumulation device and archiving them in a form that can be analyzed later. This makes it possible to monitor ground movements throughout the country in real time, realize highly accurate earthquake forecasts, and provide information promptly.

[0394] A "base station" is a piece of equipment installed throughout the country that is used to obtain location information (latitude, longitude, and altitude) within a specific area.

[0395] "Location information" is data of latitude, longitude, and altitude acquired by a base station, and is information indicating a location at a specific point in time.

[0396] The "server" is a central computing device that processes location information collected from base stations and performs earthquake forecasting.

[0397] A "generative model" is an algorithm or machine learning model used to analyze correlations with past data and make earthquake forecasts.

[0398] "Filtering" is the process of removing outliers and noise from collected data.

[0399] "Data cleaning" is a process carried out to improve the quality of data by correcting or removing inaccurate or invalid data.

[0400] An "information accumulation device" is a database or other storage device that stores prediction results and location information and archives them in a form that can be analyzed later.

[0401] "Administrative agencies" are public institutions such as local governments and government agencies.

[0402] A "weather forecast provider" is a company or organization that provides weather information.

[0403] "Ground displacement" refers to the change in latitude, longitude, and altitude of a particular point over time, including horizontal and vertical variations.

[0404] "Real-time" refers to near-instantaneous processing or reaction, meaning minimal time delay.

[0405] MODE FOR CARRYING OUT THE INVENTION

[0406] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses a generative AI model to forecast earthquakes. Specifically, the following hardware and software are used to process and calculate the data.

[0407] Hardware and Software

[0408] Server: A central computing device for data processing, collecting location information, comparing and analyzing data, generating earthquake forecasts, sending notifications, and storing data.

[0409] Base station: A facility installed throughout the country to obtain location information (latitude, longitude, altitude) within a specific area.

[0410] Database: An information collection device that stores collected location information and prediction results and archives them as needed.

[0411] Generative model: A machine learning model built using Python's TensorFlow, which performs correlation analysis with past earthquake data.

[0412] Program processing

[0413] The server collects location information every hour from base stations installed throughout the country. This information is used as basic data to understand ground movement in real time. For example, at midnight, the server sends an API request to base stations nationwide and receives the latest location information from each base station. The received location information is stored in a database.

[0414] The server then compares the latest location information collected with the previous location information and calculates the horizontal and vertical deviation of the ground. For example, if base station A's location changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), it determines that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0415] The server performs filtering and data cleaning to remove outliers and noise from the calculated data. This allows earthquake forecasts to be made based on more accurate data. For example, data containing obvious outliers or noise is ignored, and the location information is retrieved again.

[0416] The server inputs the cleaned ground displacement data into a generative AI model and analyzes correlations with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur.

[0417] The forecast results are then sent to local governments and weather forecasting service providers. This notification is sent via email or API request, and detailed reports are also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly sent to local governments, allowing them to prepare countermeasures.

[0418] The server then stores the prediction results and location information in a database, archiving them for future analysis. This data will be used to improve future prediction models and develop new earthquake forecasting technologies. For example, monthly earthquake prediction results can be stored in a database to help improve the accuracy of long-term earthquake predictions.

[0419] Examples of specific examples and prompts

[0420] Specific examples

[0421] 1. Location information collection: The location information obtained by the server from the base station is (35.6895, 139.6917, 30).

[0422] 2. Compare and calculate: If the server detects that the same base station's location information has changed to (35.6896, 139.6918, 29.9) one hour later, the server will determine that there has been a shift of 0.0001 degrees horizontally and -0.1 meters vertically.

[0423] 3. Data cleaning: The server detects abnormal outliers and noise, and either removes them or re-acquires them to clean up the data.

[0424] 4. AI forecasting: The server inputs the cleaned-up deviation data into a generative AI model, which then compares it with past data to predict the likelihood of an earthquake.

[0425] 5. Notification: The server notifies the local government by email based on the prediction results and provides information in real time via a specific API.

[0426] 6. Data storage: The server stores all results and location data in a database for further analysis and model improvement.

[0427] Prompt Sentence Examples

[0428] "Please explain the system that uses location information data collected from base stations around the country to analyze correlations with past earthquake data and predict earthquakes. Please explain in detail what hardware and software is required, including specific steps. Also, please explain how the prediction results will be notified and how the data will be stored."

[0429] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0430] Step 1:

[0431] Input: Latest location information (latitude, longitude, altitude) from multiple base stations installed throughout the country

[0432] Processing: The server first collects location information every hour from multiple base stations installed across the country. The server then sends an API request to each base station, and each base station responds by sending back its location information.

[0433] Output: Retrieved location information

[0434] Specific operation: At midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. For example, it obtains the location information (35.0116, 135.7681, 45) from the base station in Kyoto and stores it in the database.

[0435] Step 2:

[0436] Input: Latest saved location and previous location

[0437] Processing: The server compares the latest collected position information with previous position information and calculates the horizontal and vertical deviation of the ground.

[0438] Output: Calculated ground displacement data

[0439] What it does: The server compares the latest location with the previous location stored in the database. For example, if the previous location is (35.0116, 135.7681, 45) and the latest location is (35.0117, 135.7682, 44.9), it calculates a horizontal offset of 0.0001 degrees and a vertical offset of -0.1 meters.

[0440] Step 3:

[0441] Input: Calculated ground displacement data

[0442] Processing: The server performs filtering and data cleaning to remove outliers and noise from the calculated data.

[0443] Output: Cleaned up ground displacement data

[0444] What it does: The server uses a specific algorithm to filter out abnormal data and noise. For example, if the location information suddenly deviates significantly, the data will be considered as noise. Such abnormal values ​​will be filtered out, and the server will process the location information again if necessary.

[0445] Step 4:

[0446] Input: Cleaned ground displacement data

[0447] Processing: The server inputs the cleaned-up ground displacement data into a generative AI model, analyzes the correlation with past earthquake data, and makes earthquake forecasts.

[0448] Output: Earthquake forecast results

[0449] How it works: The server inputs clean ground displacement data into a generative AI model. The generative AI model uses an AI model (e.g., Python's TensorFlow) to perform correlation analysis with past earthquake data and predict the likelihood of an earthquake. If a specific displacement pattern is similar to past earthquakes, it determines that there is a high probability of an earthquake occurring at that location.

[0450] Step 5:

[0451] Input: Earthquake forecast results

[0452] Processing: The server notifies the local government and weather forecast service providers of the earthquake forecast results.

[0453] Output: Notified earthquake forecast information

[0454] Specific operation: The server sends the generated forecast results to local governments and weather forecast service providers via API requests or email. For example, if a high risk of an earthquake is predicted in the Tokyo area, that information will be immediately sent via email with a detailed report attached.

[0455] Step 6:

[0456] Input: Prediction results and location data

[0457] Processing: The server stores the prediction results and location data in a database and archives them in a form that can be analyzed later.

[0458] Output: Stored and archived data

[0459] How it works: The server stores all prediction results and location data. This data is then used for later analysis and to improve the generative AI model. For example, monthly earthquake prediction results are accumulated to help improve the accuracy of long-term earthquake predictions.

[0460] Through these processing steps, a system will be created that can monitor ground movements across the country in real time, achieve highly accurate earthquake forecasts, and provide information quickly.

[0461] (Application example 1)

[0462] 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."

[0463] Rapid and accurate earthquake forecasts are necessary to minimize damage caused by earthquakes. However, existing earthquake forecasting systems require time to collect and analyze information, making it difficult to make real-time predictions and notifications. Furthermore, they often fail to provide users with prompt notifications or specific evacuation guidance, resulting in inadequate emergency response. Furthermore, technology to effectively remove outliers and noise from analytical data has not yet been fully established, so there is a need to improve the accuracy of earthquake forecasts.

[0464] 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.

[0465] In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical ground displacement; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to forecast earthquakes; means for notifying local governments and weather forecast service providers of the forecast results; means for storing the forecast results in a database and archiving them in a format that can be analyzed later; means for notifying users of the earthquake forecast results in real time via their smartphones or smart glasses; and means for providing safe evacuation routes in the event of an earthquake. This enables accurate real-time earthquake forecasts, enabling users to be notified promptly and provided with specific evacuation guidance. Furthermore, the accuracy of earthquake forecasts can be improved by removing outliers and noise from the analysis data.

[0466] A "base station" is a device installed to conduct wireless communication within a specific area, and is the infrastructure for collecting location information.

[0467] "Location information" means data relating to the latitude, longitude, and altitude of a particular point.

[0468] "Ground displacement" refers to numerical data that indicates the degree to which the ground has moved or deformed horizontally and vertically.

[0469] A "generative AI model" is a mathematical algorithm or model that uses artificial intelligence techniques to learn specific patterns and make predictions or classifications.

[0470] "Earthquake forecast" means providing information about the possibility of future earthquakes and the predicted timing and location of such occurrences.

[0471] "Real-time" is a term that indicates that the time between the collection of information, its analysis, and the notification of the results is extremely short and almost simultaneous.

[0472] "Noise" refers to irregular or abnormal values ​​contained in data, which hinder accurate analysis.

[0473] "Data cleaning" is the process of removing noise and outliers from collected data and preparing it for analysis.

[0474] An "evacuation route" refers to the paths and procedures that have been set up in advance to allow safe evacuation in the event of a disaster such as an earthquake.

[0475] A "smartphone" is a multi-functional mobile phone terminal that can connect to the Internet.

[0476] "Smart glasses" are eyeglass-type devices with built-in displays that directly display visual information, and can connect to the Internet and use applications.

[0477] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses a generative AI model to forecast earthquakes. This system notifies users of earthquake forecast results in real time via smartphones or smart glasses, and also has the function of providing safe evacuation routes in the event of an earthquake.

[0478] Hardware and software used

[0479] The system is implemented using the following hardware and software:

[0480] Hardware: Cloud servers, smartphones, smart glasses, wireless base stations

[0481] Software: Database management systems (e.g., AWS RDS), machine learning frameworks (e.g., TensorFlow, Keras), communication APIs, notification systems (e.g., Firebase Cloud Messaging)

[0482] Collecting location information and calculating ground displacement

[0483] The server first collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed across the country. This makes it possible to obtain basic data for understanding ground movement in real time. For example, at midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. This location information is then stored in a database.

[0484] The server then compares the latest location information collected with previous location information to calculate the horizontal and vertical deviations of the ground. This comparison allows the server to determine how much the location of each base station has changed. For example, if base station A's location changes from latitude 35.6895 degrees, longitude 139.6917 degrees, and altitude 30 meters to latitude 35.6896 degrees, longitude 139.6918 degrees, and altitude 29.8 meters, it is determined that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0485] Noise removal and data cleaning

[0486] The server then performs filtering and data cleaning to remove outliers and noise from the calculated data, enabling earthquake forecasts to be made based on more accurate data. For example, data containing obvious outliers or noise is ignored, and the location information is re-acquired.

[0487] Earthquake forecasting using generative AI models

[0488] The server inputs the cleaned ground displacement data into a generative AI model and analyzes the correlation with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur. As a concrete example, the following data is input into the prompt:

[0489] "latitude_diff:0.0001, longitude_diff:0.0001, altitude_diff:-0.2"

[0490] Notification of forecast results and evacuation routes

[0491] The server then notifies local governments and weather forecast service providers of the generated forecasts. This notification is done via email or API request, and a detailed report of the forecast results is also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly notified to local governments so that they can prepare countermeasures.

[0492] The server also notifies users of earthquake forecast results in real time via their smartphones or smart glasses. Users can receive earthquake forecast information for their region through the application and receive alerts in the event of an emergency. Information on evacuation routes is also provided, allowing users to take appropriate action to evacuate safely.

[0493] In this way, by providing a system that handles everything from collecting location information from base stations to data analysis, earthquake forecasting, notification of results, and data storage, it becomes possible to minimize damage caused by earthquakes. The system as a whole also achieves continuous improvement in accuracy, enabling more accurate earthquake forecasts over a wider area.

[0494] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0495] Step 1:

[0496] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed nationwide. This input data includes the current location information of each base station. The server sends a request to the API of each base station, receives the latest location information data returned, and stores it in a database.

[0497] Step 2:

[0498] The server compares the latest collected location information with previously saved location information to calculate the horizontal and vertical deviation of the ground. This comparison calculates the difference between the current location information received as input and the previous location information, and generates position fluctuation data for each base station as output. For example, if the position of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), the output shows that a deviation of 0.0001 degrees occurred horizontally and -0.2 meters vertically.

[0499] Step 3:

[0500] The server performs filtering and data cleaning to remove outliers and noise from the calculated ground displacement data. The input for this step is the positional fluctuation data of each base station. The server filters this data to remove noise and outliers and outputs clean ground displacement data.

[0501] Step 4:

[0502] The server inputs the clean ground displacement data into the generative AI model and analyzes the correlation with past earthquake data to forecast earthquakes. In this step, filtered ground displacement data is used as input. The server uses the generative AI model to compare and analyze this data with past earthquake patterns to predict the possibility of an earthquake occurring. The output is the earthquake forecast result. As a concrete example, enter the following prompt sentence: "latitude_diff: 0.0001, longitude_diff: 0.0001, altitude_diff: -0.2"

[0503] Step 5:

[0504] The server notifies the generated forecast results to local governments and weather forecast service providers. The input for this step is the earthquake forecast results obtained in step 4. Based on these results, the server notifies relevant organizations of the details of the forecast results via email or API request. For example, if the risk of an earthquake occurring in the Tokyo area increases, the server can quickly notify the local government of this information so that they can prepare countermeasures.

[0505] Step 6:

[0506] The server stores the prediction results in a database and archives them for future analysis and improvement of the AI ​​model. The inputs for this step are the prediction results from step 4 and the notification results generated in step 5. By storing these data in the database, they can be used later for analysis and to improve the accuracy of the model.

[0507] Step 7:

[0508] The server notifies the user of the earthquake forecast results in real time via their smartphone or smart glasses. The input of this step is the earthquake forecast results obtained in step 4. The server uses the notification system to immediately send an alert to the smartphone or smart glasses, informing the user of the risk of an earthquake. In addition, it also provides safe evacuation route guidance as an output.

[0509] 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.

[0510] The present invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses an AI model to forecast earthquakes.In addition, by combining this with an emotion engine that recognizes the user's emotions, the content of earthquake forecast notifications can be optimized and the user's reactions analyzed.

[0511] 1. Location information collection from base stations

[0512] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed nationwide. The server sends a request to the API endpoint of each base station to obtain the location information. At this stage, the collected location information is temporarily stored in memory.

[0513] 2. Saving and comparing location information

[0514] The server stores the collected location information in a database and compares it with past location information, thereby calculating the horizontal and vertical deviation of the ground. For example, if the location of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8) by comparing the previous location information stored in the database with the new location information, it can be determined that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0515] 3. Noise removal and data cleaning

[0516] The server filters and cleans the calculated deviation data to remove or correct noise and outliers, for example, ignoring obviously abnormal or noisy data and trying to obtain the correct data again.

[0517] 4. Earthquake Forecasting Using Generative AI Models

[0518] The server inputs the cleaned ground displacement data into a generative AI model, which then combines it with past earthquake data to make predictions. The AI ​​model detects specific displacement patterns and, if they are similar to past earthquake precursors, predicts the probability of an earthquake occurring.

[0519] 5. User Emotion Recognition by Emotion Engine

[0520] An emotion engine is installed on the device (user's smartphone or computer) and recognizes the user's emotion by analyzing the user's facial expression and voice data when receiving the earthquake forecast result. For example, if the user's facial expression upon receiving the notification is recognized as "surprise" or "anxiety," that information is sent to the server.

[0521] 6. Utilizing Emotional Data

[0522] The server uses the emotion data obtained from the emotion engine to optimize the content of the earthquake forecast notification. For example, if the user is feeling "anxious," the server will provide the notification content in kind language and with additional reassuring information.

[0523] 7. Notification of prediction results

[0524] The server then sends the optimized forecast results to the local government and users' devices via email or app notification, providing a detailed forecast report. User responses and other information are used to make the next forecast.

[0525] 8. Data Storage and Archiving

[0526] The server stores and archives the analysis data, including prediction results and user sentiment data, in a database, which will be used as important information for future analysis and model improvement.

[0527] 9. Model Improvement through Feedback Methods

[0528] The server compares the actual earthquake occurrence situation with the predicted results to evaluate the accuracy of the generated AI model. It also evaluates user emotional data and updates the learning dataset of the AI ​​model through feedback means to continuously improve accuracy.

[0529] In this way, a system that combines an emotion engine enables flexible earthquake forecast notifications that correspond to the user's emotional state, and can also contribute to improving the accuracy of the model.

[0530] The processing flow will be explained below.

[0531] Step 1:

[0532] The server collects location information (latitude, longitude, and altitude) every hour from base stations across the country. The server sends a request to the API endpoint of each base station to obtain the current location information. The location information obtained at this stage is temporarily stored in memory.

[0533] Step 2:

[0534] The server stores the collected location information in a database, adding a timestamp to the information and storing it in a database table organized by date and time, making it easy to compare previous and current location information.

[0535] Step 3:

[0536] The server compares the latest stored location information with previous location information and calculates the horizontal and vertical deviation of the ground. Specifically, it calculates the change in latitude, longitude, and altitude for each base station to derive the horizontal and vertical deviation values.

[0537] Step 4:

[0538] The server filters and cleans the calculated deviation data, detecting and removing outliers and noise. For example, if an abnormally large fluctuation is detected, the server will either ignore the data or try collecting it again.

[0539] Step 5:

[0540] The server inputs the cleaned ground displacement data into a generative AI model. The generative AI model analyzes correlations with past earthquake data and predicts the probability of an earthquake. For example, if a specific displacement pattern is similar to that observed before a past earthquake, it determines that there is a high probability of an earthquake occurring in that area.

[0541] Step 6:

[0542] The server temporarily stores the earthquake forecast results obtained from the generative AI model and then notifies local governments and weather forecast service providers via email or API requests, providing detailed forecast reports.

[0543] Step 7:

[0544] The device notifies the user of the earthquake forecast results and simultaneously runs an emotion engine to collect the user's emotional data. Specifically, when the user receives the notification, the device uses a camera and microphone to collect facial and voice data and analyzes their emotions.

[0545] Step 8:

[0546] The server optimizes the notification content based on the user's emotional data obtained from the emotion engine. For example, if the user is feeling "anxiety" or "fear," the notification content will include additional reassurance and supportive messages to alleviate the user's anxiety.

[0547] Step 9:

[0548] The server then sends the optimized prediction results and notification content back to the local government or user's device. Notifications are sent immediately via email or app notification, and users receive detailed prediction reports and emotionally sensitive messages.

[0549] Step 10:

[0550] The server stores the prediction results and user emotion data in a database and archives them in a format that can be analyzed later. This data will be used for future analysis and model improvement.

[0551] Step 11:

[0552] The server compares the actual earthquake occurrence situation with the predicted results to evaluate the accuracy of the generated AI model. It also evaluates user emotional data and updates the AI ​​model's learning dataset with this information as a means of feedback, thereby continuously improving the model's accuracy.

[0553] Through the above steps, a system incorporating an emotion engine will be able to provide flexible earthquake forecast notifications that take into account the user's emotions, and will also contribute to improving the accuracy of the model.

[0554] Example 2

[0555] 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."

[0556] Conventional earthquake forecasting systems were capable of making predictions based on ground displacement data and past earthquake data, but the notification of the prediction results was uniform and did not take into account the individual emotions of users. This could cause users to feel anxious or confused. Furthermore, to improve the accuracy of the prediction model, it was necessary to take into account user feedback and emotional data, but this had not been realized. Furthermore, data cleaning to remove outliers and noise was insufficient, which could lead to a decrease in prediction accuracy.

[0557] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical ground displacement; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to perform earthquake forecasting; means for recognizing the user's emotions using an emotion engine in the user terminal and reflecting the data in generating optimized notification content; means for notifying local governments, weather forecast service providers, and users of the prediction results; and means for storing the prediction results in a database and archiving them in a form that can be analyzed later. This makes it possible to provide flexible notification content that takes into account the emotions of individual users and improve the accuracy of the prediction model based on feedback.

[0558] A "base station" is a communication facility installed throughout the country that provides location information (latitude, longitude, and altitude).

[0559] "Location information" is data relating to geographic latitude, longitude, and altitude.

[0560] "Horizontal and vertical deviation" is data indicating the amount of horizontal and vertical displacement relative to a specific position.

[0561] A "generative AI model" is an artificial intelligence model that analyzes ground displacement data and past earthquake data to carry out earthquake forecasting.

[0562] An "emotion engine" is software or hardware that analyzes a user's facial expressions and voice data to recognize emotions.

[0563] "Data cleaning" is a process for removing outliers and noise and shaping collected data accurately and appropriately.

[0564] The "feedback method" is a process for comparing actual earthquake occurrence conditions with predicted results to evaluate and improve the accuracy of the generative AI model.

[0565] "Notification content" is information for informing users and local governments of earthquake forecast results.

[0566] "Means for collecting" refers to the process and devices for obtaining location information from the base station to the server.

[0567] "Means for comparison" refers to the processes and algorithms for comparing collected location information with past location information and calculating ground displacement.

[0568] A "database" is a digital storage system for storing location information, prediction results, emotional data, etc.

[0569] "Archiving measures" are processes and systems for storing data for long periods of time in an analyzable form.

[0570] This invention relates to a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that location information, and uses a generative AI model to forecast earthquakes.Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system optimizes the content of earthquake forecast notifications and analyzes the user's reactions.

[0571] First, the server collects hourly location information (latitude, longitude, and altitude) from multiple base stations installed nationwide. Specifically, the server sends an HTTP request to the API endpoint of each base station to obtain latitude, longitude, and altitude information. At this stage, the collected location information is temporarily stored in memory.

[0572] The server then stores the collected location information in a database and compares it with past location information to calculate the horizontal and vertical deviations of each base station's location. For example, the server compares the previous location information with the newly acquired location information to specifically calculate the horizontal and vertical deviations.

[0573] The server then filters the calculated deviation data to remove noise and outliers, improving the accuracy of the data. Any obviously outliers are ignored and a re-acquisition attempt is made.

[0574] The server then inputs the cleaned ground displacement data into a generative AI model, which analyzes correlations with past earthquake data to predict earthquakes. The AI ​​model detects specific displacement patterns and, if they resemble past earthquake precursors, predicts the probability of an earthquake occurring. A specific prompt is, "Predict the likelihood of the next earthquake based on earthquake data from the past year and current ground displacement data."

[0575] The device (user's smartphone or computer) is equipped with an emotion engine that analyzes the user's facial expression and voice data when receiving the earthquake forecast result to recognize their emotion. For example, if the user's facial expression upon receiving the notification is recognized as "surprise" or "anxiety," that information is sent to the server.

[0576] Next, the server uses the emotion data obtained from the emotion engine to optimize the content of the earthquake forecast notification. For example, if the user is feeling "anxious," the server will provide the notification content in a gentler tone and with additional reassuring information. As a specific example, it will generate a notification that reads, "The possibility of an earthquake is increasing in your area. There is no need to feel anxious. Please check your disaster prevention measures."

[0577] The server then sends the optimized forecast results to the local government and the user's device via email or app notification, and provides a detailed forecast report. User responses are also used to make the next forecast.

[0578] Finally, the server stores the prediction results and user emotion data in a database and archives them in an analyzable format. This provides valuable information for future analysis and model improvement. Furthermore, the server compares the prediction results with actual earthquake occurrence conditions, providing a feedback mechanism for evaluating the accuracy of the generative AI model, allowing for continuous improvement of the model.

[0579] In this way, the present invention enables flexible earthquake forecast notifications that correspond to the user's emotional state through a system that combines an emotion engine, and can also contribute to improving the accuracy of the model.

[0580] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0581] Step 1:

[0582] The server collects hourly location information (latitude, longitude, altitude) from multiple base stations installed nationwide. The API endpoint of each base station is used as input. Specifically, the server sends an HTTP GET request to the API endpoint and temporarily stores the acquired location data (latitude, longitude, altitude) in memory. The output of this step is the collected location data.

[0583] Step 2:

[0584] The server stores the collected location information in a database and compares it with past location information. The current location data and past location information stored in the database are used as input. Specifically, the server inserts the collected location information into an SQL database and extracts past location information. Next, the server compares the new and old location information and calculates the horizontal and vertical displacement of the ground. The output of this step is the calculated horizontal and vertical displacement data.

[0585] Step 3:

[0586] The server filters the calculated deviation data to remove noise and outliers. The deviation data obtained in step 2 is used as input. Specifically, the server detects anomalous values ​​above a certain threshold and performs a data cleaning process to remove or correct them. The output of this step is the cleaned deviation data.

[0587] Step 4:

[0588] The server inputs the cleaned ground displacement data into the generative AI model and combines it with past earthquake data to make earthquake forecasts. The cleaned displacement data and past earthquake data are used as inputs. Specifically, the server formats this data and converts it into a format suitable for the generative AI model. The server then uses the prompt, "Predict the likelihood of the next earthquake based on the earthquake data from the past year and the current ground displacement data," to have the AI ​​model make a prediction. The output of this step is the probability of an earthquake occurring and the predicted results.

[0589] Step 5:

[0590] The device (user's smartphone or computer) implements an emotion engine and receives a notification of the earthquake forecast result. The prediction result obtained in step 4 is used as input. Specifically, when the device receives the notification, it activates the camera and microphone to collect and analyze the user's facial expression and voice data to recognize the emotion. The output of this step is the user's emotion data.

[0591] Step 6:

[0592] The server optimizes the notification content of the earthquake forecast results using the emotion data obtained from the emotion engine. The emotion data and the earthquake forecast results are used as inputs. Specifically, the server runs an algorithm that uses the user's emotion data to generate optimal notification content. For example, if the user is feeling "anxious," the notification content will be provided with gentle language and reassuring information. The output of this step is the optimized notification content.

[0593] Step 7:

[0594] The server notifies the local government and user devices of the optimized prediction results. The optimized notification content and the prediction results are used as input. Specifically, the server references the local government and user contact information and selects the appropriate notification method (e.g., email, SMS, app notification, etc.). The output of this step is the sent notification.

[0595] Step 8:

[0596] The server stores the prediction results and user emotion data in a database and archives them in an analyzable format. The prediction results and emotion data are used as input. Specifically, the server inserts these data into the database and periodically moves them to archive storage for backup. The output of this step is the stored and archived data.

[0597] Step 9:

[0598] The server compares the actual earthquake occurrence situation with the predicted results, evaluates the accuracy of the generated AI model, and provides feedback. Actual earthquake occurrence data and predicted results are used as input. Specifically, the server collects actual earthquake occurrence data and compares it with the predicted results to evaluate the accuracy of the model. Emotion data is also included in the evaluation, and the AI ​​model's training dataset is updated through feedback means, allowing for continuous improvement of the model. The output of this step is an updated AI model and a new training dataset.

[0599] (Application example 2)

[0600] 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."

[0601] Current earthquake forecasting systems aim to improve prediction accuracy, but lack consideration for the reliability of the forecast and the user's psychological state. In particular, no systems exist that take into account how earthquake forecasts actually affect users. This raises concerns that users may feel unnecessary anxiety or surprise. Therefore, it is necessary to provide a system that can accurately and quickly notify users of earthquake forecasts while taking into account their emotions.

[0602] 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.

[0603] In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical displacement of the ground; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to make earthquake forecasts; means for recognizing the user's emotions using voice and image data and optimizing the notification content based on the emotion data; and means for storing the prediction results and emotion data in a database and archiving them in a form that can be analyzed later. This enables earthquake forecast notifications that take the user's psychological state into consideration, making it possible to provide fast and accurate earthquake forecasts while reducing the user's anxiety.

[0604] A "base station" is a device installed in a communications network to provide location information.

[0605] "Location information" is data including latitude, longitude, and altitude, and is information for identifying a geographical location.

[0606] "Ground displacement" is part of the crustal movement that occurs due to the effects of earthquakes and other events, and refers to changes in position in the horizontal and vertical directions.

[0607] A "generative AI model" is an artificial intelligence model used to predict earthquake occurrences using past earthquake data and other input data.

[0608] A "database" is a system for systematically storing and managing data.

[0609] "Emotion recognition" is a technology that analyzes a user's voice and image data to recognize the user's emotional state (for example, anxiety or surprise).

[0610] "Optimizing notification content" refers to adjusting the content of a notification message depending on the emotional state of the user receiving the notification.

[0611] "Archiving" means storing data for the long term so that it can be analyzed or referenced later.

[0612] "Noise reduction" is the process of removing outliers and irrelevant values ​​in data processing.

[0613] "Feedback measures" are methods for comparing actual earthquake occurrence conditions with predicted results to improve the accuracy of the generative AI model.

[0614] This paper explains a system that collects location information (latitude, longitude, altitude) every hour from multiple base stations installed throughout the country and calculates ground displacement based on this information. The core part of the system is executed by the server, terminals, and users.

[0615] The server first collects location information from base stations across the country using API endpoints. The technology used includes the Python requests library. The collected data is temporarily stored in memory, and then the location information is stored in a database, such as Databricks.

[0616] The server then compares the data with past position information to calculate the horizontal and vertical displacement of the ground. This analysis is performed using the GeoAnalytics library. The analyzed data is then filtered to remove noise and data cleaning, and outliers are removed. This process is also performed using the GeoAnalytics library.

[0617] The cleaned ground displacement data is then input into a generative AI model. This model uses machine learning libraries such as TensorFlow to analyze correlations with past earthquake data and generate earthquake forecasts. The results of this forecast are then sent to local governments and weather forecast service providers via a server.

[0618] The device recognizes emotions using the user's voice and image data. It uses the smartphone's camera and microphone to analyze the user's emotions using the OpenCV and Pyaudio libraries. The analysis results are sent to the server, which then optimizes the notification content based on the emotion. The NotificationEngine library is used for this purpose.

[0619] Furthermore, these prediction results and emotion data are stored in a database. This storage allows for post-analysis of the prediction results and serves as a feedback tool to help improve the AI ​​model. Data obtained from the emotion engine is also archived, with the aim of improving the accuracy of the model in the future.

[0620] To maximize the effectiveness of this system, the following specific example is given. For example, if the earthquake prediction result is highly likely and the user shows a surprised expression, the notification content can be customized as follows: "Notice. An earthquake of magnitude 6 has been predicted. However, please remain calm and be prepared to evacuate quickly. Necessary support information will also be provided."

[0621] The following is an example of a prompt sentence to input to the generative AI model.

[0622] You received a high probability earthquake forecast. What reassuring notification message would be appropriate for a user who recognized a surprised expression?

[0623] In this way, a system is constructed that takes into account the user's psychological state and provides accurate and prompt earthquake forecast notifications.

[0624] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0625] Step 1:

[0626] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations across the country. It sends a request to the API endpoint of each base station to obtain location information. The collected location information is temporarily stored in memory.

[0627] Input: Location information from base station

[0628] Output: Raw position information stored in memory

[0629] Step 2:

[0630] The server stores the collected location information in a database, then compares it with past location information to calculate horizontal and vertical displacements of the ground, thereby identifying fluctuations in location information.

[0631] Input: Raw location information stored in memory, past location information

[0632] Output: Calculated ground displacement data

[0633] Step 3:

[0634] The server filters and cleans the calculated ground displacement data, removing or correcting noise and outliers to produce a clean dataset.

[0635] Input: Calculated ground displacement data

[0636] Output: Clean ground displacement data

[0637] Step 4:

[0638] The server inputs clean ground displacement data into a generative AI model, which analyzes correlations with past earthquake data to forecast earthquakes. The generative AI model is trained using TensorFlow.

[0639] Input: Clean ground displacement data

[0640] Output: Earthquake forecast results

[0641] Step 5:

[0642] The device collects the user's voice and image data and uses them to recognize emotions. It uses OpenCV and Pyaudio libraries to analyze the user's facial expressions and voice to identify emotions.

[0643] Input: User's voice and image data

[0644] Output: Recognized emotion data

[0645] Step 6:

[0646] The server optimizes the notification content of earthquake forecast results based on emotion data. It uses the NotificationEngine library to generate customized notifications according to the emotion. For example, if the user is feeling "anxious," the server will provide the notification content with kind language and additional reassurance information.

[0647] Input: Earthquake forecast results, recognized emotion data

[0648] Output: Optimized notification content

[0649] Step 7:

[0650] The server then sends the earthquake forecast results and optimized notification content to local governments and users' devices via email and app notifications.

[0651] Input: Optimized notification content

[0652] Output: Notification to local government and user devices

[0653] Step 8:

[0654] The server stores and archives the prediction results and user emotion data in a database, which will be used as important information for future analysis and model improvement.

[0655] Input: Prediction results, user emotion data

[0656] Output: Analysis data stored in a database

[0657] Step 9:

[0658] The server compares the actual earthquake occurrence situation with the predicted results, and updates the learning dataset of the AI ​​model through a feedback means to improve the generated AI model, thereby continuously improving its accuracy.

[0659] Input: Actual earthquake occurrence situation, predicted results

[0660] Output: An improved generative AI model

[0661] 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.

[0662] 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.

[0663] 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.

[0664] [Third embodiment]

[0665] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0666] 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.

[0667] 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).

[0668] 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.

[0669] 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.

[0670] 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).

[0671] 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.

[0672] 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.

[0673] 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.

[0674] 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.

[0675] 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.

[0676] 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."

[0677] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses an AI model to forecast earthquakes.

[0678] 1. Location information collection from base stations

[0679] The server first collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed across the country. This makes it possible to obtain basic data for understanding ground movement in real time. For example, at midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. This location information is then stored in a database.

[0680] 2. Comparing location information and calculating ground displacement

[0681] The server compares the latest location information collected with previous location information and calculates the horizontal and vertical deviation of the ground. This comparison allows the server to determine how much the location of each base station has changed. For example, if the location of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), it is confirmed that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0682] 3. Noise removal and data cleaning

[0683] The server then performs filtering and data cleaning to remove outliers and noise from the calculated data, allowing for more accurate earthquake forecasts. For example, data containing obvious outliers or noise is ignored, and the location information is re-acquired.

[0684] 4. Earthquake Forecasting Using Generative AI Models

[0685] The server inputs the cleaned ground displacement data into a generative AI model and analyzes correlations with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur.

[0686] 5. Notification of prediction results

[0687] The server then notifies local governments and weather forecast service providers of the generated forecasts. This notification is done via email or API request, and a detailed report of the forecast results is also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly notified to local governments so that they can prepare countermeasures.

[0688] 6. Data Storage and Archiving

[0689] The server stores the prediction results and location information in a database and archives them in a format that can be analyzed later. This data will be used to improve future prediction models and develop new earthquake forecasting technologies. For example, monthly earthquake prediction results will be accumulated to help improve the accuracy of long-term earthquake predictions.

[0690] In this way, by providing a system that handles everything from collecting location information from base stations to data analysis, earthquake forecasting, notification of results, and data storage, it becomes possible to minimize damage caused by earthquakes. The system as a whole also achieves continuous improvement in accuracy, enabling more accurate earthquake forecasts over a wider area.

[0691] The processing flow will be explained below.

[0692] Step 1:

[0693] The server collects location information (latitude, longitude, and altitude) every hour from base stations across the country. The server sends a request to the API endpoint of each base station to obtain the location information. At this stage, the collected location information is temporarily stored in memory.

[0694] Step 2:

[0695] The server stores the collected location information in a database. When storing, it assigns a timestamp to the location information of each base station and stores it in a specific table. This allows for efficient comparison of past location information with the latest location information.

[0696] Step 3:

[0697] The server compares the latest saved location information with previous location information to calculate the horizontal and vertical deviation of the ground. For example, it calculates the positional fluctuation of each base station over the past hour and outputs the horizontal and vertical deviation as specific numerical values.

[0698] Step 4:

[0699] The server processes the calculated deviation data through filtering and data cleaning. It detects noise and outliers and removes or corrects them. For example, if an abnormally large fluctuation is detected, it will be removed or the data will be collected again.

[0700] Step 5:

[0701] The server inputs the cleaned ground displacement data into the generative AI model. It combines this with past earthquake data and executes earthquake forecasts using the AI ​​model. For example, if a similar displacement pattern appears in the past as a precursor to an earthquake, the AI ​​model can use this to predict the probability of an earthquake occurring.

[0702] Step 6:

[0703] The server evaluates the risk of earthquakes based on the prediction results and notifies local governments and weather forecast service providers. Notifications are sent via email or API requests, providing detailed forecast reports. For example, a warning could be sent to local governments stating that there is a high risk of earthquakes in the Tokyo area.

[0704] Step 7:

[0705] The server stores and archives the prediction results and analysis data in a database, making them available for future analysis and model improvement. The stored data is time-stamped and stored in a format that allows for easy comparison with past data.

[0706] Step 8:

[0707] The server compares the actual earthquake occurrences with the predicted results to evaluate the accuracy of the generative AI model. Based on the evaluation results, it updates the model's training dataset and continuously improves the accuracy of the AI ​​model through feedback means. For example, it checks whether past predictions matched actual earthquake occurrences and incorporates that information into the next training dataset.

[0708] Example 1

[0709] 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."

[0710] Earthquakes are difficult to predict, and accurate and prompt earthquake forecasts are necessary to minimize the damage caused by their occurrence. However, conventional methods have limitations in the accuracy and speed of earthquake forecasts, and many challenges remain. In particular, it is technically difficult to monitor ground movements across the country in real time and make highly accurate predictions based on that information. Therefore, there is a demand for a system that can achieve more accurate, real-time earthquake forecasts and provide prompt notifications.

[0711] 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.

[0712] In this invention, the server includes means for collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country, means for comparing the collected location information with previous location information and calculating horizontal and vertical displacement of the ground, means for inputting the calculated ground displacement data into a generative model and analyzing correlations with past earthquake data to forecast earthquakes, means for notifying administrative agencies and weather forecast providers of the results of the forecast, and means for storing the results of the forecast in an information accumulation device and archiving them in a form that can be analyzed later. This makes it possible to monitor ground movements throughout the country in real time, realize highly accurate earthquake forecasts, and provide information promptly.

[0713] A "base station" is a piece of equipment installed throughout the country that is used to obtain location information (latitude, longitude, and altitude) within a specific area.

[0714] "Location information" is data of latitude, longitude, and altitude acquired by a base station, and is information indicating a location at a specific point in time.

[0715] The "server" is a central computing device that processes location information collected from base stations and performs earthquake forecasting.

[0716] A "generative model" is an algorithm or machine learning model used to analyze correlations with past data and make earthquake forecasts.

[0717] "Filtering" is the process of removing outliers and noise from collected data.

[0718] "Data cleaning" is a process carried out to improve the quality of data by correcting or removing inaccurate or invalid data.

[0719] An "information accumulation device" is a database or other storage device that stores prediction results and location information and archives them in a form that can be analyzed later.

[0720] "Administrative agencies" are public institutions such as local governments and government agencies.

[0721] A "weather forecast provider" is a company or organization that provides weather information.

[0722] "Ground displacement" refers to the change in latitude, longitude, and altitude of a particular point over time, including horizontal and vertical variations.

[0723] "Real-time" refers to near-instantaneous processing or reaction, meaning minimal time delay.

[0724] MODE FOR CARRYING OUT THE INVENTION

[0725] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses a generative AI model to forecast earthquakes. Specifically, the following hardware and software are used to process and calculate the data.

[0726] Hardware and Software

[0727] Server: A central computing device for data processing, collecting location information, comparing and analyzing data, generating earthquake forecasts, sending notifications, and storing data.

[0728] Base station: A facility installed throughout the country to obtain location information (latitude, longitude, altitude) within a specific area.

[0729] Database: An information collection device that stores collected location information and prediction results and archives them as needed.

[0730] Generative model: A machine learning model built using Python's TensorFlow, which performs correlation analysis with past earthquake data.

[0731] Program processing

[0732] The server collects location information every hour from base stations installed throughout the country. This information is used as basic data to understand ground movement in real time. For example, at midnight, the server sends an API request to base stations nationwide and receives the latest location information from each base station. The received location information is stored in a database.

[0733] The server then compares the latest location information collected with the previous location information and calculates the horizontal and vertical deviation of the ground. For example, if base station A's location changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), it determines that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0734] The server performs filtering and data cleaning to remove outliers and noise from the calculated data. This allows earthquake forecasts to be made based on more accurate data. For example, data containing obvious outliers or noise is ignored, and the location information is retrieved again.

[0735] The server inputs the cleaned ground displacement data into a generative AI model and analyzes correlations with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur.

[0736] The forecast results are then sent to local governments and weather forecasting service providers. This notification is sent via email or API request, and detailed reports are also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly sent to local governments, allowing them to prepare countermeasures.

[0737] The server then stores the prediction results and location information in a database, archiving them for future analysis. This data will be used to improve future prediction models and develop new earthquake forecasting technologies. For example, monthly earthquake prediction results can be stored in a database to help improve the accuracy of long-term earthquake predictions.

[0738] Examples of specific examples and prompts

[0739] Specific examples

[0740] 1. Location information collection: The location information obtained by the server from the base station is (35.6895, 139.6917, 30).

[0741] 2. Compare and calculate: If the server detects that the same base station's location information has changed to (35.6896, 139.6918, 29.9) one hour later, the server will determine that there has been a shift of 0.0001 degrees horizontally and -0.1 meters vertically.

[0742] 3. Data cleaning: The server detects abnormal outliers and noise, and either removes them or re-acquires them to clean up the data.

[0743] 4. AI forecasting: The server inputs the cleaned-up deviation data into a generative AI model, which then compares it with past data to predict the likelihood of an earthquake.

[0744] 5. Notification: The server notifies the local government by email based on the prediction results and provides information in real time via a specific API.

[0745] 6. Data storage: The server stores all results and location data in a database for further analysis and model improvement.

[0746] Prompt Sentence Examples

[0747] "Please explain the system that uses location information data collected from base stations around the country to analyze correlations with past earthquake data and predict earthquakes. Please explain in detail what hardware and software is required, including specific steps. Also, please explain how the prediction results will be notified and how the data will be stored."

[0748] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0749] Step 1:

[0750] Input: Latest location information (latitude, longitude, altitude) from multiple base stations installed throughout the country

[0751] Processing: The server first collects location information every hour from multiple base stations installed across the country. The server then sends an API request to each base station, and each base station responds by sending back its location information.

[0752] Output: Retrieved location information

[0753] Specific operation: At midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. For example, it obtains the location information (35.0116, 135.7681, 45) from the base station in Kyoto and stores it in the database.

[0754] Step 2:

[0755] Input: Latest saved location and previous location

[0756] Processing: The server compares the latest collected position information with previous position information and calculates the horizontal and vertical deviation of the ground.

[0757] Output: Calculated ground displacement data

[0758] What it does: The server compares the latest location with the previous location stored in the database. For example, if the previous location is (35.0116, 135.7681, 45) and the latest location is (35.0117, 135.7682, 44.9), it calculates a horizontal offset of 0.0001 degrees and a vertical offset of -0.1 meters.

[0759] Step 3:

[0760] Input: Calculated ground displacement data

[0761] Processing: The server performs filtering and data cleaning to remove outliers and noise from the calculated data.

[0762] Output: Cleaned up ground displacement data

[0763] What it does: The server uses a specific algorithm to filter out abnormal data and noise. For example, if the location information suddenly deviates significantly, the data will be considered as noise. Such abnormal values ​​will be filtered out, and the server will process the location information again if necessary.

[0764] Step 4:

[0765] Input: Cleaned ground displacement data

[0766] Processing: The server inputs the cleaned-up ground displacement data into a generative AI model, analyzes the correlation with past earthquake data, and makes earthquake forecasts.

[0767] Output: Earthquake forecast results

[0768] How it works: The server inputs clean ground displacement data into a generative AI model. The generative AI model uses an AI model (e.g., Python's TensorFlow) to perform correlation analysis with past earthquake data and predict the likelihood of an earthquake. If a specific displacement pattern is similar to past earthquakes, it determines that there is a high probability of an earthquake occurring at that location.

[0769] Step 5:

[0770] Input: Earthquake forecast results

[0771] Processing: The server notifies the local government and weather forecast service providers of the earthquake forecast results.

[0772] Output: Notified earthquake forecast information

[0773] Specific operation: The server sends the generated forecast results to local governments and weather forecast service providers via API requests or email. For example, if a high risk of an earthquake is predicted in the Tokyo area, that information will be immediately sent via email with a detailed report attached.

[0774] Step 6:

[0775] Input: Prediction results and location data

[0776] Processing: The server stores the prediction results and location data in a database and archives them in a form that can be analyzed later.

[0777] Output: Stored and archived data

[0778] How it works: The server stores all prediction results and location data. This data is then used for later analysis and to improve the generative AI model. For example, monthly earthquake prediction results are accumulated to help improve the accuracy of long-term earthquake predictions.

[0779] Through these processing steps, a system will be created that can monitor ground movements across the country in real time, achieve highly accurate earthquake forecasts, and provide information quickly.

[0780] (Application example 1)

[0781] 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."

[0782] Rapid and accurate earthquake forecasts are necessary to minimize damage caused by earthquakes. However, existing earthquake forecasting systems require time to collect and analyze information, making it difficult to make real-time predictions and notifications. Furthermore, they often fail to provide users with prompt notifications or specific evacuation guidance, resulting in inadequate emergency response. Furthermore, technology to effectively remove outliers and noise from analytical data has not yet been fully established, so there is a need to improve the accuracy of earthquake forecasts.

[0783] 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.

[0784] In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical ground displacement; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to forecast earthquakes; means for notifying local governments and weather forecast service providers of the forecast results; means for storing the forecast results in a database and archiving them in a format that can be analyzed later; means for notifying users of the earthquake forecast results in real time via their smartphones or smart glasses; and means for providing safe evacuation routes in the event of an earthquake. This enables accurate real-time earthquake forecasts, enabling users to be notified promptly and provided with specific evacuation guidance. Furthermore, the accuracy of earthquake forecasts can be improved by removing outliers and noise from the analysis data.

[0785] A "base station" is a device installed to conduct wireless communication within a specific area, and is the infrastructure for collecting location information.

[0786] "Location information" means data relating to the latitude, longitude, and altitude of a particular point.

[0787] "Ground displacement" refers to numerical data that indicates the degree to which the ground has moved or deformed horizontally and vertically.

[0788] A "generative AI model" is a mathematical algorithm or model that uses artificial intelligence techniques to learn specific patterns and make predictions or classifications.

[0789] "Earthquake forecast" means providing information about the possibility of future earthquakes and the predicted timing and location of such occurrences.

[0790] "Real-time" is a term that indicates that the time between the collection of information, its analysis, and the notification of the results is extremely short and almost simultaneous.

[0791] "Noise" refers to irregular or abnormal values ​​contained in data, which hinder accurate analysis.

[0792] "Data cleaning" is the process of removing noise and outliers from collected data and preparing it for analysis.

[0793] An "evacuation route" refers to the paths and procedures that have been set up in advance to allow safe evacuation in the event of a disaster such as an earthquake.

[0794] A "smartphone" is a multi-functional mobile phone terminal that can connect to the Internet.

[0795] "Smart glasses" are eyeglass-type devices with built-in displays that directly display visual information, and can connect to the Internet and use applications.

[0796] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses a generative AI model to forecast earthquakes. This system notifies users of earthquake forecast results in real time via smartphones or smart glasses, and also has the function of providing safe evacuation routes in the event of an earthquake.

[0797] Hardware and software used

[0798] The system is implemented using the following hardware and software:

[0799] Hardware: Cloud servers, smartphones, smart glasses, wireless base stations

[0800] Software: Database management systems (e.g., AWS RDS), machine learning frameworks (e.g., TensorFlow, Keras), communication APIs, notification systems (e.g., Firebase Cloud Messaging)

[0801] Collecting location information and calculating ground displacement

[0802] The server first collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed across the country. This makes it possible to obtain basic data for understanding ground movement in real time. For example, at midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. This location information is then stored in a database.

[0803] The server then compares the latest location information collected with previous location information to calculate the horizontal and vertical deviations of the ground. This comparison allows the server to determine how much the location of each base station has changed. For example, if base station A's location changes from latitude 35.6895 degrees, longitude 139.6917 degrees, and altitude 30 meters to latitude 35.6896 degrees, longitude 139.6918 degrees, and altitude 29.8 meters, it is determined that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0804] Noise removal and data cleaning

[0805] The server then performs filtering and data cleaning to remove outliers and noise from the calculated data, enabling earthquake forecasts to be made based on more accurate data. For example, data containing obvious outliers or noise is ignored, and the location information is re-acquired.

[0806] Earthquake forecasting using generative AI models

[0807] The server inputs the cleaned ground displacement data into a generative AI model and analyzes the correlation with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur. As a concrete example, the following data is input into the prompt:

[0808] "latitude_diff:0.0001, longitude_diff:0.0001, altitude_diff:-0.2"

[0809] Notification of forecast results and evacuation routes

[0810] The server then notifies local governments and weather forecast service providers of the generated forecasts. This notification is done via email or API request, and a detailed report of the forecast results is also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly notified to local governments so that they can prepare countermeasures.

[0811] The server also notifies users of earthquake forecast results in real time via their smartphones or smart glasses. Users can receive earthquake forecast information for their region through the application and receive alerts in the event of an emergency. Information on evacuation routes is also provided, allowing users to take appropriate action to evacuate safely.

[0812] In this way, by providing a system that handles everything from collecting location information from base stations to data analysis, earthquake forecasting, notification of results, and data storage, it becomes possible to minimize damage caused by earthquakes. The system as a whole also achieves continuous improvement in accuracy, enabling more accurate earthquake forecasts over a wider area.

[0813] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0814] Step 1:

[0815] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed nationwide. This input data includes the current location information of each base station. The server sends a request to the API of each base station, receives the latest location information data returned, and stores it in a database.

[0816] Step 2:

[0817] The server compares the latest collected location information with previously saved location information to calculate the horizontal and vertical deviation of the ground. This comparison calculates the difference between the current location information received as input and the previous location information, and generates position fluctuation data for each base station as output. For example, if the position of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), the output shows that a deviation of 0.0001 degrees occurred horizontally and -0.2 meters vertically.

[0818] Step 3:

[0819] The server performs filtering and data cleaning to remove outliers and noise from the calculated ground displacement data. The input for this step is the positional fluctuation data of each base station. The server filters this data to remove noise and outliers and outputs clean ground displacement data.

[0820] Step 4:

[0821] The server inputs the clean ground displacement data into the generative AI model and analyzes the correlation with past earthquake data to forecast earthquakes. In this step, filtered ground displacement data is used as input. The server uses the generative AI model to compare and analyze this data with past earthquake patterns to predict the possibility of an earthquake occurring. The output is the earthquake forecast result. As a concrete example, enter the following prompt sentence: "latitude_diff: 0.0001, longitude_diff: 0.0001, altitude_diff: -0.2"

[0822] Step 5:

[0823] The server notifies the generated forecast results to local governments and weather forecast service providers. The input for this step is the earthquake forecast results obtained in step 4. Based on these results, the server notifies relevant organizations of the details of the forecast results via email or API request. For example, if the risk of an earthquake occurring in the Tokyo area increases, the server can quickly notify the local government of this information so that they can prepare countermeasures.

[0824] Step 6:

[0825] The server stores the prediction results in a database and archives them for future analysis and improvement of the AI ​​model. The inputs for this step are the prediction results from step 4 and the notification results generated in step 5. By storing these data in the database, they can be used later for analysis and to improve the accuracy of the model.

[0826] Step 7:

[0827] The server notifies the user of the earthquake forecast results in real time via their smartphone or smart glasses. The input of this step is the earthquake forecast results obtained in step 4. The server uses the notification system to immediately send an alert to the smartphone or smart glasses, informing the user of the risk of an earthquake. In addition, it also provides safe evacuation route guidance as an output.

[0828] 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.

[0829] The present invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses an AI model to forecast earthquakes.In addition, by combining this with an emotion engine that recognizes the user's emotions, the content of earthquake forecast notifications can be optimized and the user's reactions analyzed.

[0830] 1. Location information collection from base stations

[0831] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed nationwide. The server sends a request to the API endpoint of each base station to obtain the location information. At this stage, the collected location information is temporarily stored in memory.

[0832] 2. Saving and comparing location information

[0833] The server stores the collected location information in a database and compares it with past location information, thereby calculating the horizontal and vertical deviation of the ground. For example, if the location of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8) by comparing the previous location information stored in the database with the new location information, it can be determined that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[0834] 3. Noise removal and data cleaning

[0835] The server filters and cleans the calculated deviation data to remove or correct noise and outliers, for example, ignoring obviously abnormal or noisy data and trying to obtain the correct data again.

[0836] 4. Earthquake Forecasting Using Generative AI Models

[0837] The server inputs the cleaned ground displacement data into a generative AI model, which then combines it with past earthquake data to make predictions. The AI ​​model detects specific displacement patterns and, if they are similar to past earthquake precursors, predicts the probability of an earthquake occurring.

[0838] 5. User Emotion Recognition by Emotion Engine

[0839] An emotion engine is installed on the device (user's smartphone or computer) and recognizes the user's emotion by analyzing the user's facial expression and voice data when receiving the earthquake forecast result. For example, if the user's facial expression upon receiving the notification is recognized as "surprise" or "anxiety," that information is sent to the server.

[0840] 6. Utilizing Emotional Data

[0841] The server uses the emotion data obtained from the emotion engine to optimize the content of the earthquake forecast notification. For example, if the user is feeling "anxious," the server will provide the notification content in kind language and with additional reassuring information.

[0842] 7. Notification of prediction results

[0843] The server then sends the optimized forecast results to the local government and users' devices via email or app notification, providing a detailed forecast report. User responses and other information are used to make the next forecast.

[0844] 8. Data Storage and Archiving

[0845] The server stores and archives the analysis data, including prediction results and user sentiment data, in a database, which will be used as important information for future analysis and model improvement.

[0846] 9. Model Improvement through Feedback Methods

[0847] The server compares the actual earthquake occurrence situation with the predicted results to evaluate the accuracy of the generated AI model. It also evaluates user emotional data and updates the learning dataset of the AI ​​model through feedback means to continuously improve accuracy.

[0848] In this way, a system that combines an emotion engine enables flexible earthquake forecast notifications that correspond to the user's emotional state, and can also contribute to improving the accuracy of the model.

[0849] The processing flow will be explained below.

[0850] Step 1:

[0851] The server collects location information (latitude, longitude, and altitude) every hour from base stations across the country. The server sends a request to the API endpoint of each base station to obtain the current location information. The location information obtained at this stage is temporarily stored in memory.

[0852] Step 2:

[0853] The server stores the collected location information in a database, adding a timestamp to the information and storing it in a database table organized by date and time, making it easy to compare previous and current location information.

[0854] Step 3:

[0855] The server compares the latest stored location information with previous location information and calculates the horizontal and vertical deviation of the ground. Specifically, it calculates the change in latitude, longitude, and altitude for each base station to derive the horizontal and vertical deviation values.

[0856] Step 4:

[0857] The server filters and cleans the calculated deviation data, detecting and removing outliers and noise. For example, if an abnormally large fluctuation is detected, the server will either ignore the data or try collecting it again.

[0858] Step 5:

[0859] The server inputs the cleaned ground displacement data into a generative AI model. The generative AI model analyzes correlations with past earthquake data and predicts the probability of an earthquake. For example, if a specific displacement pattern is similar to that observed before a past earthquake, it determines that there is a high probability of an earthquake occurring in that area.

[0860] Step 6:

[0861] The server temporarily stores the earthquake forecast results obtained from the generative AI model and then notifies local governments and weather forecast service providers via email or API requests, providing detailed forecast reports.

[0862] Step 7:

[0863] The device notifies the user of the earthquake forecast results and simultaneously runs an emotion engine to collect the user's emotional data. Specifically, when the user receives the notification, the device uses a camera and microphone to collect facial and voice data and analyzes their emotions.

[0864] Step 8:

[0865] The server optimizes the notification content based on the user's emotional data obtained from the emotion engine. For example, if the user is feeling "anxiety" or "fear," the notification content will include additional reassurance and supportive messages to alleviate the user's anxiety.

[0866] Step 9:

[0867] The server then sends the optimized prediction results and notification content back to the local government or user's device. Notifications are sent immediately via email or app notification, and users receive detailed prediction reports and emotionally sensitive messages.

[0868] Step 10:

[0869] The server stores the prediction results and user emotion data in a database and archives them in a format that can be analyzed later. This data will be used for future analysis and model improvement.

[0870] Step 11:

[0871] The server compares the actual earthquake occurrence situation with the predicted results to evaluate the accuracy of the generated AI model. It also evaluates user emotional data and updates the AI ​​model's learning dataset with this information as a means of feedback, thereby continuously improving the model's accuracy.

[0872] Through the above steps, a system incorporating an emotion engine will be able to provide flexible earthquake forecast notifications that take into account the user's emotions, and will also contribute to improving the accuracy of the model.

[0873] Example 2

[0874] 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."

[0875] Conventional earthquake forecasting systems were capable of making predictions based on ground displacement data and past earthquake data, but the notification of the prediction results was uniform and did not take into account the individual emotions of users. This could cause users to feel anxious or confused. Furthermore, to improve the accuracy of the prediction model, it was necessary to take into account user feedback and emotional data, but this had not been realized. Furthermore, data cleaning to remove outliers and noise was insufficient, which could lead to a decrease in prediction accuracy.

[0876] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical ground displacement; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to perform earthquake forecasting; means for recognizing the user's emotions using an emotion engine in the user terminal and reflecting the data in generating optimized notification content; means for notifying local governments, weather forecast service providers, and users of the prediction results; and means for storing the prediction results in a database and archiving them in a form that can be analyzed later. This makes it possible to provide flexible notification content that takes into account the emotions of individual users and improve the accuracy of the prediction model based on feedback.

[0877] A "base station" is a communication facility installed throughout the country that provides location information (latitude, longitude, and altitude).

[0878] "Location information" is data relating to geographic latitude, longitude, and altitude.

[0879] "Horizontal and vertical deviation" is data indicating the amount of horizontal and vertical displacement relative to a specific position.

[0880] A "generative AI model" is an artificial intelligence model that analyzes ground displacement data and past earthquake data to carry out earthquake forecasting.

[0881] An "emotion engine" is software or hardware that analyzes a user's facial expressions and voice data to recognize emotions.

[0882] "Data cleaning" is a process for removing outliers and noise and shaping collected data accurately and appropriately.

[0883] The "feedback method" is a process for comparing actual earthquake occurrence conditions with predicted results to evaluate and improve the accuracy of the generative AI model.

[0884] "Notification content" is information for informing users and local governments of earthquake forecast results.

[0885] "Means for collecting" refers to the process and devices for obtaining location information from the base station to the server.

[0886] "Means for comparison" refers to the processes and algorithms for comparing collected location information with past location information and calculating ground displacement.

[0887] A "database" is a digital storage system for storing location information, prediction results, emotional data, etc.

[0888] "Archiving measures" are processes and systems for storing data for long periods of time in an analyzable form.

[0889] This invention relates to a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that location information, and uses a generative AI model to forecast earthquakes.Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system optimizes the content of earthquake forecast notifications and analyzes the user's reactions.

[0890] First, the server collects hourly location information (latitude, longitude, and altitude) from multiple base stations installed nationwide. Specifically, the server sends an HTTP request to the API endpoint of each base station to obtain latitude, longitude, and altitude information. At this stage, the collected location information is temporarily stored in memory.

[0891] The server then stores the collected location information in a database and compares it with past location information to calculate the horizontal and vertical deviations of each base station's location. For example, the server compares the previous location information with the newly acquired location information to specifically calculate the horizontal and vertical deviations.

[0892] The server then filters the calculated deviation data to remove noise and outliers, improving the accuracy of the data. Any obviously outliers are ignored and a re-acquisition attempt is made.

[0893] The server then inputs the cleaned ground displacement data into a generative AI model, which analyzes correlations with past earthquake data to predict earthquakes. The AI ​​model detects specific displacement patterns and, if they resemble past earthquake precursors, predicts the probability of an earthquake occurring. A specific prompt is, "Predict the likelihood of the next earthquake based on earthquake data from the past year and current ground displacement data."

[0894] The device (user's smartphone or computer) is equipped with an emotion engine that analyzes the user's facial expression and voice data when receiving the earthquake forecast result to recognize their emotion. For example, if the user's facial expression upon receiving the notification is recognized as "surprise" or "anxiety," that information is sent to the server.

[0895] Next, the server uses the emotion data obtained from the emotion engine to optimize the content of the earthquake forecast notification. For example, if the user is feeling "anxious," the server will provide the notification content in a gentler tone and with additional reassuring information. As a specific example, it will generate a notification that reads, "The possibility of an earthquake is increasing in your area. There is no need to feel anxious. Please check your disaster prevention measures."

[0896] The server then sends the optimized forecast results to the local government and the user's device via email or app notification, and provides a detailed forecast report. User responses are also used to make the next forecast.

[0897] Finally, the server stores the prediction results and user emotion data in a database and archives them in an analyzable format. This provides valuable information for future analysis and model improvement. Furthermore, the server compares the prediction results with actual earthquake occurrence conditions, providing a feedback mechanism for evaluating the accuracy of the generative AI model, allowing for continuous improvement of the model.

[0898] In this way, the present invention enables flexible earthquake forecast notifications that correspond to the user's emotional state through a system that combines an emotion engine, and can also contribute to improving the accuracy of the model.

[0899] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0900] Step 1:

[0901] The server collects hourly location information (latitude, longitude, altitude) from multiple base stations installed nationwide. The API endpoint of each base station is used as input. Specifically, the server sends an HTTP GET request to the API endpoint and temporarily stores the acquired location data (latitude, longitude, altitude) in memory. The output of this step is the collected location data.

[0902] Step 2:

[0903] The server stores the collected location information in a database and compares it with past location information. The current location data and past location information stored in the database are used as input. Specifically, the server inserts the collected location information into an SQL database and extracts past location information. Next, the server compares the new and old location information and calculates the horizontal and vertical displacement of the ground. The output of this step is the calculated horizontal and vertical displacement data.

[0904] Step 3:

[0905] The server filters the calculated deviation data to remove noise and outliers. The deviation data obtained in step 2 is used as input. Specifically, the server detects anomalous values ​​above a certain threshold and performs a data cleaning process to remove or correct them. The output of this step is the cleaned deviation data.

[0906] Step 4:

[0907] The server inputs the cleaned ground displacement data into the generative AI model and combines it with past earthquake data to make earthquake forecasts. The cleaned displacement data and past earthquake data are used as inputs. Specifically, the server formats this data and converts it into a format suitable for the generative AI model. The server then uses the prompt, "Predict the likelihood of the next earthquake based on the earthquake data from the past year and the current ground displacement data," to have the AI ​​model make a prediction. The output of this step is the probability of an earthquake occurring and the predicted results.

[0908] Step 5:

[0909] The device (user's smartphone or computer) implements an emotion engine and receives a notification of the earthquake forecast result. The prediction result obtained in step 4 is used as input. Specifically, when the device receives the notification, it activates the camera and microphone to collect and analyze the user's facial expression and voice data to recognize the emotion. The output of this step is the user's emotion data.

[0910] Step 6:

[0911] The server optimizes the notification content of the earthquake forecast results using the emotion data obtained from the emotion engine. The emotion data and the earthquake forecast results are used as inputs. Specifically, the server runs an algorithm that uses the user's emotion data to generate optimal notification content. For example, if the user is feeling "anxious," the notification content will be provided with gentle language and reassuring information. The output of this step is the optimized notification content.

[0912] Step 7:

[0913] The server notifies the local government and user devices of the optimized prediction results. The optimized notification content and the prediction results are used as input. Specifically, the server references the local government and user contact information and selects the appropriate notification method (e.g., email, SMS, app notification, etc.). The output of this step is the sent notification.

[0914] Step 8:

[0915] The server stores the prediction results and user emotion data in a database and archives them in an analyzable format. The prediction results and emotion data are used as input. Specifically, the server inserts these data into the database and periodically moves them to archive storage for backup. The output of this step is the stored and archived data.

[0916] Step 9:

[0917] The server compares the actual earthquake occurrence situation with the predicted results, evaluates the accuracy of the generated AI model, and provides feedback. Actual earthquake occurrence data and predicted results are used as input. Specifically, the server collects actual earthquake occurrence data and compares it with the predicted results to evaluate the accuracy of the model. Emotion data is also included in the evaluation, and the AI ​​model's training dataset is updated through feedback means, allowing for continuous improvement of the model. The output of this step is an updated AI model and a new training dataset.

[0918] (Application example 2)

[0919] 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."

[0920] Current earthquake forecasting systems aim to improve prediction accuracy, but lack consideration for the reliability of the forecast and the user's psychological state. In particular, no systems exist that take into account how earthquake forecasts actually affect users. This raises concerns that users may feel unnecessary anxiety or surprise. Therefore, it is necessary to provide a system that can accurately and quickly notify users of earthquake forecasts while taking into account their emotions.

[0921] 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.

[0922] In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical displacement of the ground; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to make earthquake forecasts; means for recognizing the user's emotions using voice and image data and optimizing the notification content based on the emotion data; and means for storing the prediction results and emotion data in a database and archiving them in a form that can be analyzed later. This enables earthquake forecast notifications that take the user's psychological state into consideration, making it possible to provide fast and accurate earthquake forecasts while reducing the user's anxiety.

[0923] A "base station" is a device installed in a communications network to provide location information.

[0924] "Location information" is data including latitude, longitude, and altitude, and is information for identifying a geographical location.

[0925] "Ground displacement" is part of the crustal movement that occurs due to the effects of earthquakes and other events, and refers to changes in position in the horizontal and vertical directions.

[0926] A "generative AI model" is an artificial intelligence model used to predict earthquake occurrences using past earthquake data and other input data.

[0927] A "database" is a system for systematically storing and managing data.

[0928] "Emotion recognition" is a technology that analyzes a user's voice and image data to recognize the user's emotional state (for example, anxiety or surprise).

[0929] "Optimizing notification content" refers to adjusting the content of a notification message depending on the emotional state of the user receiving the notification.

[0930] "Archiving" means storing data for the long term so that it can be analyzed or referenced later.

[0931] "Noise reduction" is the process of removing outliers and irrelevant values ​​in data processing.

[0932] "Feedback measures" are methods for comparing actual earthquake occurrence conditions with predicted results to improve the accuracy of the generative AI model.

[0933] This paper explains a system that collects location information (latitude, longitude, altitude) every hour from multiple base stations installed throughout the country and calculates ground displacement based on this information. The core part of the system is executed by the server, terminals, and users.

[0934] The server first collects location information from base stations across the country using API endpoints. The technology used includes the Python requests library. The collected data is temporarily stored in memory, and then the location information is stored in a database, such as Databricks.

[0935] The server then compares the data with past position information to calculate the horizontal and vertical displacement of the ground. This analysis is performed using the GeoAnalytics library. The analyzed data is then filtered to remove noise and data cleaning, and outliers are removed. This process is also performed using the GeoAnalytics library.

[0936] The cleaned ground displacement data is then input into a generative AI model. This model uses machine learning libraries such as TensorFlow to analyze correlations with past earthquake data and generate earthquake forecasts. The results of this forecast are then sent to local governments and weather forecast service providers via a server.

[0937] The device recognizes emotions using the user's voice and image data. It uses the smartphone's camera and microphone to analyze the user's emotions using the OpenCV and Pyaudio libraries. The analysis results are sent to the server, which then optimizes the notification content based on the emotion. The NotificationEngine library is used for this purpose.

[0938] Furthermore, these prediction results and emotion data are stored in a database. This storage allows for post-analysis of the prediction results and serves as a feedback tool to help improve the AI ​​model. Data obtained from the emotion engine is also archived, with the aim of improving the accuracy of the model in the future.

[0939] To maximize the effectiveness of this system, the following specific example is given. For example, if the earthquake prediction result is highly likely and the user shows a surprised expression, the notification content can be customized as follows: "Notice. An earthquake of magnitude 6 has been predicted. However, please remain calm and be prepared to evacuate quickly. Necessary support information will also be provided."

[0940] The following is an example of a prompt sentence to input to the generative AI model.

[0941] You received a high probability earthquake forecast. What reassuring notification message would be appropriate for a user who recognized a surprised expression?

[0942] In this way, a system is constructed that takes into account the user's psychological state and provides accurate and prompt earthquake forecast notifications.

[0943] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0944] Step 1:

[0945] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations across the country. It sends a request to the API endpoint of each base station to obtain location information. The collected location information is temporarily stored in memory.

[0946] Input: Location information from base station

[0947] Output: Raw position information stored in memory

[0948] Step 2:

[0949] The server stores the collected location information in a database, then compares it with past location information to calculate horizontal and vertical displacements of the ground, thereby identifying fluctuations in location information.

[0950] Input: Raw location information stored in memory, past location information

[0951] Output: Calculated ground displacement data

[0952] Step 3:

[0953] The server filters and cleans the calculated ground displacement data, removing or correcting noise and outliers to produce a clean dataset.

[0954] Input: Calculated ground displacement data

[0955] Output: Clean ground displacement data

[0956] Step 4:

[0957] The server inputs clean ground displacement data into a generative AI model, which analyzes correlations with past earthquake data to forecast earthquakes. The generative AI model is trained using TensorFlow.

[0958] Input: Clean ground displacement data

[0959] Output: Earthquake forecast results

[0960] Step 5:

[0961] The device collects the user's voice and image data and uses them to recognize emotions. It uses OpenCV and Pyaudio libraries to analyze the user's facial expressions and voice to identify emotions.

[0962] Input: User's voice and image data

[0963] Output: Recognized emotion data

[0964] Step 6:

[0965] The server optimizes the notification content of earthquake forecast results based on emotion data. It uses the NotificationEngine library to generate customized notifications according to the emotion. For example, if the user is feeling "anxious," the server will provide the notification content with kind language and additional reassurance information.

[0966] Input: Earthquake forecast results, recognized emotion data

[0967] Output: Optimized notification content

[0968] Step 7:

[0969] The server then sends the earthquake forecast results and optimized notification content to local governments and users' devices via email and app notifications.

[0970] Input: Optimized notification content

[0971] Output: Notification to local government and user devices

[0972] Step 8:

[0973] The server stores and archives the prediction results and user emotion data in a database, which will be used as important information for future analysis and model improvement.

[0974] Input: Prediction results, user emotion data

[0975] Output: Analysis data stored in a database

[0976] Step 9:

[0977] The server compares the actual earthquake occurrence situation with the predicted results, and updates the learning dataset of the AI ​​model through a feedback means to improve the generated AI model, thereby continuously improving its accuracy.

[0978] Input: Actual earthquake occurrence situation, predicted results

[0979] Output: An improved generative AI model

[0980] 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.

[0981] 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.

[0982] 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.

[0983] [Fourth embodiment]

[0984] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0985] 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.

[0986] 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).

[0987] 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.

[0988] 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.

[0989] 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).

[0990] 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.

[0991] 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.

[0992] 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.

[0993] 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.

[0994] 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.

[0995] 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.

[0996] 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."

[0997] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses an AI model to forecast earthquakes.

[0998] 1. Location information collection from base stations

[0999] The server first collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed across the country. This makes it possible to obtain basic data for understanding ground movement in real time. For example, at midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. This location information is then stored in a database.

[1000] 2. Comparing location information and calculating ground displacement

[1001] The server compares the latest location information collected with previous location information and calculates the horizontal and vertical deviation of the ground. This comparison allows the server to determine how much the location of each base station has changed. For example, if the location of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), it is confirmed that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[1002] 3. Noise removal and data cleaning

[1003] The server then performs filtering and data cleaning to remove outliers and noise from the calculated data, allowing for more accurate earthquake forecasts. For example, data containing obvious outliers or noise is ignored, and the location information is re-acquired.

[1004] 4. Earthquake Forecasting Using Generative AI Models

[1005] The server inputs the cleaned ground displacement data into a generative AI model and analyzes correlations with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur.

[1006] 5. Notification of prediction results

[1007] The server then notifies local governments and weather forecast service providers of the generated forecasts. This notification is done via email or API request, and a detailed report of the forecast results is also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly notified to local governments so that they can prepare countermeasures.

[1008] 6. Data Storage and Archiving

[1009] The server stores the prediction results and location information in a database and archives them in a format that can be analyzed later. This data will be used to improve future prediction models and develop new earthquake forecasting technologies. For example, monthly earthquake prediction results will be accumulated to help improve the accuracy of long-term earthquake predictions.

[1010] In this way, by providing a system that handles everything from collecting location information from base stations to data analysis, earthquake forecasting, notification of results, and data storage, it becomes possible to minimize damage caused by earthquakes. The system as a whole also achieves continuous improvement in accuracy, enabling more accurate earthquake forecasts over a wider area.

[1011] The processing flow will be explained below.

[1012] Step 1:

[1013] The server collects location information (latitude, longitude, and altitude) every hour from base stations across the country. The server sends a request to the API endpoint of each base station to obtain the location information. At this stage, the collected location information is temporarily stored in memory.

[1014] Step 2:

[1015] The server stores the collected location information in a database. When storing, it assigns a timestamp to the location information of each base station and stores it in a specific table. This allows for efficient comparison of past location information with the latest location information.

[1016] Step 3:

[1017] The server compares the latest saved location information with previous location information to calculate the horizontal and vertical deviation of the ground. For example, it calculates the positional fluctuation of each base station over the past hour and outputs the horizontal and vertical deviation as specific numerical values.

[1018] Step 4:

[1019] The server processes the calculated deviation data through filtering and data cleaning. It detects noise and outliers and removes or corrects them. For example, if an abnormally large fluctuation is detected, it will be removed or the data will be collected again.

[1020] Step 5:

[1021] The server inputs the cleaned ground displacement data into the generative AI model. It combines this with past earthquake data and executes earthquake forecasts using the AI ​​model. For example, if a similar displacement pattern appears in the past as a precursor to an earthquake, the AI ​​model can use this to predict the probability of an earthquake occurring.

[1022] Step 6:

[1023] The server evaluates the risk of earthquakes based on the prediction results and notifies local governments and weather forecast service providers. Notifications are sent via email or API requests, providing detailed forecast reports. For example, a warning could be sent to local governments stating that there is a high risk of earthquakes in the Tokyo area.

[1024] Step 7:

[1025] The server stores and archives the prediction results and analysis data in a database, making them available for future analysis and model improvement. The stored data is time-stamped and stored in a format that allows for easy comparison with past data.

[1026] Step 8:

[1027] The server compares the actual earthquake occurrences with the predicted results to evaluate the accuracy of the generative AI model. Based on the evaluation results, it updates the model's training dataset and continuously improves the accuracy of the AI ​​model through feedback means. For example, it checks whether past predictions matched actual earthquake occurrences and incorporates that information into the next training dataset.

[1028] Example 1

[1029] 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."

[1030] Earthquakes are difficult to predict, and accurate and prompt earthquake forecasts are necessary to minimize the damage caused by their occurrence. However, conventional methods have limitations in the accuracy and speed of earthquake forecasts, and many challenges remain. In particular, it is technically difficult to monitor ground movements across the country in real time and make highly accurate predictions based on that information. Therefore, there is a demand for a system that can achieve more accurate, real-time earthquake forecasts and provide prompt notifications.

[1031] 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.

[1032] In this invention, the server includes means for collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country, means for comparing the collected location information with previous location information and calculating horizontal and vertical displacement of the ground, means for inputting the calculated ground displacement data into a generative model and analyzing correlations with past earthquake data to forecast earthquakes, means for notifying administrative agencies and weather forecast providers of the results of the forecast, and means for storing the results of the forecast in an information accumulation device and archiving them in a form that can be analyzed later. This makes it possible to monitor ground movements throughout the country in real time, realize highly accurate earthquake forecasts, and provide information promptly.

[1033] A "base station" is a piece of equipment installed throughout the country that is used to obtain location information (latitude, longitude, and altitude) within a specific area.

[1034] "Location information" is data of latitude, longitude, and altitude acquired by a base station, and is information indicating a location at a specific point in time.

[1035] The "server" is a central computing device that processes location information collected from base stations and performs earthquake forecasting.

[1036] A "generative model" is an algorithm or machine learning model used to analyze correlations with past data and make earthquake forecasts.

[1037] "Filtering" is the process of removing outliers and noise from collected data.

[1038] "Data cleaning" is a process carried out to improve the quality of data by correcting or removing inaccurate or invalid data.

[1039] An "information accumulation device" is a database or other storage device that stores prediction results and location information and archives them in a form that can be analyzed later.

[1040] "Administrative agencies" are public institutions such as local governments and government agencies.

[1041] A "weather forecast provider" is a company or organization that provides weather information.

[1042] "Ground displacement" refers to the change in latitude, longitude, and altitude of a particular point over time, including horizontal and vertical variations.

[1043] "Real-time" refers to near-instantaneous processing or reaction, meaning minimal time delay.

[1044] MODE FOR CARRYING OUT THE INVENTION

[1045] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses a generative AI model to forecast earthquakes. Specifically, the following hardware and software are used to process and calculate the data.

[1046] Hardware and Software

[1047] Server: A central computing device for data processing, collecting location information, comparing and analyzing data, generating earthquake forecasts, sending notifications, and storing data.

[1048] Base station: A facility installed throughout the country to obtain location information (latitude, longitude, altitude) within a specific area.

[1049] Database: An information collection device that stores collected location information and prediction results and archives them as needed.

[1050] Generative model: A machine learning model built using Python's TensorFlow, which performs correlation analysis with past earthquake data.

[1051] Program processing

[1052] The server collects location information every hour from base stations installed throughout the country. This information is used as basic data to understand ground movement in real time. For example, at midnight, the server sends an API request to base stations nationwide and receives the latest location information from each base station. The received location information is stored in a database.

[1053] The server then compares the latest location information collected with the previous location information and calculates the horizontal and vertical deviation of the ground. For example, if base station A's location changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), it determines that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[1054] The server performs filtering and data cleaning to remove outliers and noise from the calculated data. This allows earthquake forecasts to be made based on more accurate data. For example, data containing obvious outliers or noise is ignored, and the location information is retrieved again.

[1055] The server inputs the cleaned ground displacement data into a generative AI model and analyzes correlations with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur.

[1056] The forecast results are then sent to local governments and weather forecasting service providers. This notification is sent via email or API request, and detailed reports are also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly sent to local governments, allowing them to prepare countermeasures.

[1057] The server then stores the prediction results and location information in a database, archiving them for future analysis. This data will be used to improve future prediction models and develop new earthquake forecasting technologies. For example, monthly earthquake prediction results can be stored in a database to help improve the accuracy of long-term earthquake predictions.

[1058] Examples of specific examples and prompts

[1059] Specific examples

[1060] 1. Location information collection: The location information obtained by the server from the base station is (35.6895, 139.6917, 30).

[1061] 2. Compare and calculate: If the server detects that the same base station's location information has changed to (35.6896, 139.6918, 29.9) one hour later, the server will determine that there has been a shift of 0.0001 degrees horizontally and -0.1 meters vertically.

[1062] 3. Data cleaning: The server detects abnormal outliers and noise, and either removes them or re-acquires them to clean up the data.

[1063] 4. AI forecasting: The server inputs the cleaned-up deviation data into a generative AI model, which then compares it with past data to predict the likelihood of an earthquake.

[1064] 5. Notification: The server notifies the local government by email based on the prediction results and provides information in real time via a specific API.

[1065] 6. Data storage: The server stores all results and location data in a database for further analysis and model improvement.

[1066] Prompt Sentence Examples

[1067] "Please explain the system that uses location information data collected from base stations around the country to analyze correlations with past earthquake data and predict earthquakes. Please explain in detail what hardware and software is required, including specific steps. Also, please explain how the prediction results will be notified and how the data will be stored."

[1068] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1069] Step 1:

[1070] Input: Latest location information (latitude, longitude, altitude) from multiple base stations installed throughout the country

[1071] Processing: The server first collects location information every hour from multiple base stations installed across the country. The server then sends an API request to each base station, and each base station responds by sending back its location information.

[1072] Output: Retrieved location information

[1073] Specific operation: At midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. For example, it obtains the location information (35.0116, 135.7681, 45) from the base station in Kyoto and stores it in the database.

[1074] Step 2:

[1075] Input: Latest saved location and previous location

[1076] Processing: The server compares the latest collected position information with previous position information and calculates the horizontal and vertical deviation of the ground.

[1077] Output: Calculated ground displacement data

[1078] What it does: The server compares the latest location with the previous location stored in the database. For example, if the previous location is (35.0116, 135.7681, 45) and the latest location is (35.0117, 135.7682, 44.9), it calculates a horizontal offset of 0.0001 degrees and a vertical offset of -0.1 meters.

[1079] Step 3:

[1080] Input: Calculated ground displacement data

[1081] Processing: The server performs filtering and data cleaning to remove outliers and noise from the calculated data.

[1082] Output: Cleaned up ground displacement data

[1083] What it does: The server uses a specific algorithm to filter out abnormal data and noise. For example, if the location information suddenly deviates significantly, the data will be considered as noise. Such abnormal values ​​will be filtered out, and the server will process the location information again if necessary.

[1084] Step 4:

[1085] Input: Cleaned ground displacement data

[1086] Processing: The server inputs the cleaned-up ground displacement data into a generative AI model, analyzes the correlation with past earthquake data, and makes earthquake forecasts.

[1087] Output: Earthquake forecast results

[1088] How it works: The server inputs clean ground displacement data into a generative AI model. The generative AI model uses an AI model (e.g., Python's TensorFlow) to perform correlation analysis with past earthquake data and predict the likelihood of an earthquake. If a specific displacement pattern is similar to past earthquakes, it determines that there is a high probability of an earthquake occurring at that location.

[1089] Step 5:

[1090] Input: Earthquake forecast results

[1091] Processing: The server notifies the local government and weather forecast service providers of the earthquake forecast results.

[1092] Output: Notified earthquake forecast information

[1093] Specific operation: The server sends the generated forecast results to local governments and weather forecast service providers via API requests or email. For example, if a high risk of an earthquake is predicted in the Tokyo area, that information will be immediately sent via email with a detailed report attached.

[1094] Step 6:

[1095] Input: Prediction results and location data

[1096] Processing: The server stores the prediction results and location data in a database and archives them in a form that can be analyzed later.

[1097] Output: Stored and archived data

[1098] How it works: The server stores all prediction results and location data. This data is then used for later analysis and to improve the generative AI model. For example, monthly earthquake prediction results are accumulated to help improve the accuracy of long-term earthquake predictions.

[1099] Through these processing steps, a system will be created that can monitor ground movements across the country in real time, achieve highly accurate earthquake forecasts, and provide information quickly.

[1100] (Application example 1)

[1101] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1102] Rapid and accurate earthquake forecasts are necessary to minimize damage caused by earthquakes. However, existing earthquake forecasting systems require time to collect and analyze information, making it difficult to make real-time predictions and notifications. Furthermore, they often fail to provide users with prompt notifications or specific evacuation guidance, resulting in inadequate emergency response. Furthermore, technology to effectively remove outliers and noise from analytical data has not yet been fully established, so there is a need to improve the accuracy of earthquake forecasts.

[1103] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1104] In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical ground displacement; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to forecast earthquakes; means for notifying local governments and weather forecast service providers of the forecast results; means for storing the forecast results in a database and archiving them in a format that can be analyzed later; means for notifying users of the earthquake forecast results in real time via their smartphones or smart glasses; and means for providing safe evacuation routes in the event of an earthquake. This enables accurate real-time earthquake forecasts, enabling users to be notified promptly and provided with specific evacuation guidance. Furthermore, the accuracy of earthquake forecasts can be improved by removing outliers and noise from the analysis data.

[1105] A "base station" is a device installed to conduct wireless communication within a specific area, and is the infrastructure for collecting location information.

[1106] "Location information" means data relating to the latitude, longitude, and altitude of a particular point.

[1107] "Ground displacement" refers to numerical data that indicates the degree to which the ground has moved or deformed horizontally and vertically.

[1108] A "generative AI model" is a mathematical algorithm or model that uses artificial intelligence techniques to learn specific patterns and make predictions or classifications.

[1109] "Earthquake forecast" means providing information about the possibility of future earthquakes and the predicted timing and location of such occurrences.

[1110] "Real-time" is a term that indicates that the time between the collection of information, its analysis, and the notification of the results is extremely short and almost simultaneous.

[1111] "Noise" refers to irregular or abnormal values ​​contained in data, which hinder accurate analysis.

[1112] "Data cleaning" is the process of removing noise and outliers from collected data and preparing it for analysis.

[1113] An "evacuation route" refers to the paths and procedures that have been set up in advance to allow safe evacuation in the event of a disaster such as an earthquake.

[1114] A "smartphone" is a multi-functional mobile phone terminal that can connect to the Internet.

[1115] "Smart glasses" are eyeglass-type devices with built-in displays that directly display visual information, and can connect to the Internet and use applications.

[1116] This invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses a generative AI model to forecast earthquakes. This system notifies users of earthquake forecast results in real time via smartphones or smart glasses, and also has the function of providing safe evacuation routes in the event of an earthquake.

[1117] Hardware and software used

[1118] The system is implemented using the following hardware and software:

[1119] Hardware: Cloud servers, smartphones, smart glasses, wireless base stations

[1120] Software: Database management systems (e.g., AWS RDS), machine learning frameworks (e.g., TensorFlow, Keras), communication APIs, notification systems (e.g., Firebase Cloud Messaging)

[1121] Collecting location information and calculating ground displacement

[1122] The server first collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed across the country. This makes it possible to obtain basic data for understanding ground movement in real time. For example, at midnight, the server sends an API request to base stations across the country and receives the latest location information from each base station. This location information is then stored in a database.

[1123] The server then compares the latest location information collected with previous location information to calculate the horizontal and vertical deviations of the ground. This comparison allows the server to determine how much the location of each base station has changed. For example, if base station A's location changes from latitude 35.6895 degrees, longitude 139.6917 degrees, and altitude 30 meters to latitude 35.6896 degrees, longitude 139.6918 degrees, and altitude 29.8 meters, it is determined that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[1124] Noise removal and data cleaning

[1125] The server then performs filtering and data cleaning to remove outliers and noise from the calculated data, enabling earthquake forecasts to be made based on more accurate data. For example, data containing obvious outliers or noise is ignored, and the location information is re-acquired.

[1126] Earthquake forecasting using generative AI models

[1127] The server inputs the cleaned ground displacement data into a generative AI model and analyzes the correlation with past earthquake data to forecast earthquakes. For example, if the AI ​​model detects a specific displacement pattern and that pattern is similar to that seen before past earthquakes, it predicts that an earthquake is likely to occur. As a concrete example, the following data is input into the prompt:

[1128] "latitude_diff:0.0001, longitude_diff:0.0001, altitude_diff:-0.2"

[1129] Notification of forecast results and evacuation routes

[1130] The server then notifies local governments and weather forecast service providers of the generated forecasts. This notification is done via email or API request, and a detailed report of the forecast results is also provided. For example, if the risk of an earthquake in the Tokyo area increases, this information can be quickly notified to local governments so that they can prepare countermeasures.

[1131] The server also notifies users of earthquake forecast results in real time via their smartphones or smart glasses. Users can receive earthquake forecast information for their region through the application and receive alerts in the event of an emergency. Information on evacuation routes is also provided, allowing users to take appropriate action to evacuate safely.

[1132] In this way, by providing a system that handles everything from collecting location information from base stations to data analysis, earthquake forecasting, notification of results, and data storage, it becomes possible to minimize damage caused by earthquakes. The system as a whole also achieves continuous improvement in accuracy, enabling more accurate earthquake forecasts over a wider area.

[1133] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1134] Step 1:

[1135] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed nationwide. This input data includes the current location information of each base station. The server sends a request to the API of each base station, receives the latest location information data returned, and stores it in a database.

[1136] Step 2:

[1137] The server compares the latest collected location information with previously saved location information to calculate the horizontal and vertical deviation of the ground. This comparison calculates the difference between the current location information received as input and the previous location information, and generates position fluctuation data for each base station as output. For example, if the position of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8), the output shows that a deviation of 0.0001 degrees occurred horizontally and -0.2 meters vertically.

[1138] Step 3:

[1139] The server performs filtering and data cleaning to remove outliers and noise from the calculated ground displacement data. The input for this step is the positional fluctuation data of each base station. The server filters this data to remove noise and outliers and outputs clean ground displacement data.

[1140] Step 4:

[1141] The server inputs the clean ground displacement data into the generative AI model and analyzes the correlation with past earthquake data to forecast earthquakes. In this step, filtered ground displacement data is used as input. The server uses the generative AI model to compare and analyze this data with past earthquake patterns to predict the possibility of an earthquake occurring. The output is the earthquake forecast result. As a concrete example, enter the following prompt sentence: "latitude_diff: 0.0001, longitude_diff: 0.0001, altitude_diff: -0.2"

[1142] Step 5:

[1143] The server notifies the generated forecast results to local governments and weather forecast service providers. The input for this step is the earthquake forecast results obtained in step 4. Based on these results, the server notifies relevant organizations of the details of the forecast results via email or API request. For example, if the risk of an earthquake occurring in the Tokyo area increases, the server can quickly notify the local government of this information so that they can prepare countermeasures.

[1144] Step 6:

[1145] The server stores the prediction results in a database and archives them for future analysis and improvement of the AI ​​model. The inputs for this step are the prediction results from step 4 and the notification results generated in step 5. By storing these data in the database, they can be used later for analysis and to improve the accuracy of the model.

[1146] Step 7:

[1147] The server notifies the user of the earthquake forecast results in real time via their smartphone or smart glasses. The input of this step is the earthquake forecast results obtained in step 4. The server uses the notification system to immediately send an alert to the smartphone or smart glasses, informing the user of the risk of an earthquake. In addition, it also provides safe evacuation route guidance as an output.

[1148] 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.

[1149] The present invention is a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that data, and uses an AI model to forecast earthquakes.In addition, by combining this with an emotion engine that recognizes the user's emotions, the content of earthquake forecast notifications can be optimized and the user's reactions analyzed.

[1150] 1. Location information collection from base stations

[1151] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations installed nationwide. The server sends a request to the API endpoint of each base station to obtain the location information. At this stage, the collected location information is temporarily stored in memory.

[1152] 2. Saving and comparing location information

[1153] The server stores the collected location information in a database and compares it with past location information, thereby calculating the horizontal and vertical deviation of the ground. For example, if the location of base station A changes from (35.6895, 139.6917, 30) to (35.6896, 139.6918, 29.8) by comparing the previous location information stored in the database with the new location information, it can be determined that a deviation of 0.0001 degrees has occurred horizontally and -0.2 meters vertically.

[1154] 3. Noise removal and data cleaning

[1155] The server filters and cleans the calculated deviation data to remove or correct noise and outliers, for example, ignoring obviously abnormal or noisy data and trying to obtain the correct data again.

[1156] 4. Earthquake Forecasting Using Generative AI Models

[1157] The server inputs the cleaned ground displacement data into a generative AI model, which then combines it with past earthquake data to make predictions. The AI ​​model detects specific displacement patterns and, if they are similar to past earthquake precursors, predicts the probability of an earthquake occurring.

[1158] 5. User Emotion Recognition by Emotion Engine

[1159] An emotion engine is installed on the device (user's smartphone or computer) and recognizes the user's emotion by analyzing the user's facial expression and voice data when receiving the earthquake forecast result. For example, if the user's facial expression upon receiving the notification is recognized as "surprise" or "anxiety," that information is sent to the server.

[1160] 6. Utilizing Emotional Data

[1161] The server uses the emotion data obtained from the emotion engine to optimize the content of the earthquake forecast notification. For example, if the user is feeling "anxious," the server will provide the notification content in kind language and with additional reassuring information.

[1162] 7. Notification of prediction results

[1163] The server then sends the optimized forecast results to the local government and users' devices via email or app notification, providing a detailed forecast report. User responses and other information are used to make the next forecast.

[1164] 8. Data Storage and Archiving

[1165] The server stores and archives the analysis data, including prediction results and user sentiment data, in a database, which will be used as important information for future analysis and model improvement.

[1166] 9. Model Improvement through Feedback Methods

[1167] The server compares the actual earthquake occurrence situation with the predicted results to evaluate the accuracy of the generated AI model. It also evaluates user emotional data and updates the learning dataset of the AI ​​model through feedback means to continuously improve accuracy.

[1168] In this way, a system that combines an emotion engine enables flexible earthquake forecast notifications that correspond to the user's emotional state, and can also contribute to improving the accuracy of the model.

[1169] The processing flow will be explained below.

[1170] Step 1:

[1171] The server collects location information (latitude, longitude, and altitude) every hour from base stations across the country. The server sends a request to the API endpoint of each base station to obtain the current location information. The location information obtained at this stage is temporarily stored in memory.

[1172] Step 2:

[1173] The server stores the collected location information in a database, adding a timestamp to the information and storing it in a database table organized by date and time, making it easy to compare previous and current location information.

[1174] Step 3:

[1175] The server compares the latest stored location information with previous location information and calculates the horizontal and vertical deviation of the ground. Specifically, it calculates the change in latitude, longitude, and altitude for each base station to derive the horizontal and vertical deviation values.

[1176] Step 4:

[1177] The server filters and cleans the calculated deviation data, detecting and removing outliers and noise. For example, if an abnormally large fluctuation is detected, the server will either ignore the data or try collecting it again.

[1178] Step 5:

[1179] The server inputs the cleaned ground displacement data into a generative AI model. The generative AI model analyzes correlations with past earthquake data and predicts the probability of an earthquake. For example, if a specific displacement pattern is similar to that observed before a past earthquake, it determines that there is a high probability of an earthquake occurring in that area.

[1180] Step 6:

[1181] The server temporarily stores the earthquake forecast results obtained from the generative AI model and then notifies local governments and weather forecast service providers via email or API requests, providing detailed forecast reports.

[1182] Step 7:

[1183] The device notifies the user of the earthquake forecast results and simultaneously runs an emotion engine to collect the user's emotional data. Specifically, when the user receives the notification, the device uses a camera and microphone to collect facial and voice data and analyzes their emotions.

[1184] Step 8:

[1185] The server optimizes the notification content based on the user's emotional data obtained from the emotion engine. For example, if the user is feeling "anxiety" or "fear," the notification content will include additional reassurance and supportive messages to alleviate the user's anxiety.

[1186] Step 9:

[1187] The server then sends the optimized prediction results and notification content back to the local government or user's device. Notifications are sent immediately via email or app notification, and users receive detailed prediction reports and emotionally sensitive messages.

[1188] Step 10:

[1189] The server stores the prediction results and user emotion data in a database and archives them in a format that can be analyzed later. This data will be used for future analysis and model improvement.

[1190] Step 11:

[1191] The server compares the actual earthquake occurrence situation with the predicted results to evaluate the accuracy of the generated AI model. It also evaluates user emotional data and updates the AI ​​model's learning dataset with this information as a means of feedback, thereby continuously improving the model's accuracy.

[1192] Through the above steps, a system incorporating an emotion engine will be able to provide flexible earthquake forecast notifications that take into account the user's emotions, and will also contribute to improving the accuracy of the model.

[1193] Example 2

[1194] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1195] Conventional earthquake forecasting systems were capable of making predictions based on ground displacement data and past earthquake data, but the notification of the prediction results was uniform and did not take into account the individual emotions of users. This could cause users to feel anxious or confused. Furthermore, to improve the accuracy of the prediction model, it was necessary to take into account user feedback and emotional data, but this had not been realized. Furthermore, data cleaning to remove outliers and noise was insufficient, which could lead to a decrease in prediction accuracy.

[1196] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical ground displacement; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to perform earthquake forecasting; means for recognizing the user's emotions using an emotion engine in the user terminal and reflecting the data in generating optimized notification content; means for notifying local governments, weather forecast service providers, and users of the prediction results; and means for storing the prediction results in a database and archiving them in a form that can be analyzed later. This makes it possible to provide flexible notification content that takes into account the emotions of individual users and improve the accuracy of the prediction model based on feedback.

[1197] A "base station" is a communication facility installed throughout the country that provides location information (latitude, longitude, and altitude).

[1198] "Location information" is data relating to geographic latitude, longitude, and altitude.

[1199] "Horizontal and vertical deviation" is data indicating the amount of horizontal and vertical displacement relative to a specific position.

[1200] A "generative AI model" is an artificial intelligence model that analyzes ground displacement data and past earthquake data to carry out earthquake forecasting.

[1201] An "emotion engine" is software or hardware that analyzes a user's facial expressions and voice data to recognize emotions.

[1202] "Data cleaning" is a process for removing outliers and noise and shaping collected data accurately and appropriately.

[1203] The "feedback method" is a process for comparing actual earthquake occurrence conditions with predicted results to evaluate and improve the accuracy of the generative AI model.

[1204] "Notification content" is information for informing users and local governments of earthquake forecast results.

[1205] "Means for collecting" refers to the process and devices for obtaining location information from the base station to the server.

[1206] "Means for comparison" refers to the processes and algorithms for comparing collected location information with past location information and calculating ground displacement.

[1207] A "database" is a digital storage system for storing location information, prediction results, emotional data, etc.

[1208] "Archiving measures" are processes and systems for storing data for long periods of time in an analyzable form.

[1209] This invention relates to a system that collects location information from multiple base stations installed throughout the country, calculates and analyzes ground displacement based on that location information, and uses a generative AI model to forecast earthquakes.Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system optimizes the content of earthquake forecast notifications and analyzes the user's reactions.

[1210] First, the server collects hourly location information (latitude, longitude, and altitude) from multiple base stations installed nationwide. Specifically, the server sends an HTTP request to the API endpoint of each base station to obtain latitude, longitude, and altitude information. At this stage, the collected location information is temporarily stored in memory.

[1211] The server then stores the collected location information in a database and compares it with past location information to calculate the horizontal and vertical deviations of each base station's location. For example, the server compares the previous location information with the newly acquired location information to specifically calculate the horizontal and vertical deviations.

[1212] The server then filters the calculated deviation data to remove noise and outliers, improving the accuracy of the data. Any obviously outliers are ignored and a re-acquisition attempt is made.

[1213] The server then inputs the cleaned ground displacement data into a generative AI model, which analyzes correlations with past earthquake data to predict earthquakes. The AI ​​model detects specific displacement patterns and, if they resemble past earthquake precursors, predicts the probability of an earthquake occurring. A specific prompt is, "Predict the likelihood of the next earthquake based on earthquake data from the past year and current ground displacement data."

[1214] The device (user's smartphone or computer) is equipped with an emotion engine that analyzes the user's facial expression and voice data when receiving the earthquake forecast result to recognize their emotion. For example, if the user's facial expression upon receiving the notification is recognized as "surprise" or "anxiety," that information is sent to the server.

[1215] Next, the server uses the emotion data obtained from the emotion engine to optimize the content of the earthquake forecast notification. For example, if the user is feeling "anxious," the server will provide the notification content in a gentler tone and with additional reassuring information. As a specific example, it will generate a notification that reads, "The possibility of an earthquake is increasing in your area. There is no need to feel anxious. Please check your disaster prevention measures."

[1216] The server then sends the optimized forecast results to the local government and the user's device via email or app notification, and provides a detailed forecast report. User responses are also used to make the next forecast.

[1217] Finally, the server stores the prediction results and user emotion data in a database and archives them in an analyzable format. This provides valuable information for future analysis and model improvement. Furthermore, the server compares the prediction results with actual earthquake occurrence conditions, providing a feedback mechanism for evaluating the accuracy of the generative AI model, allowing for continuous improvement of the model.

[1218] In this way, the present invention enables flexible earthquake forecast notifications that correspond to the user's emotional state through a system that combines an emotion engine, and can also contribute to improving the accuracy of the model.

[1219] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1220] Step 1:

[1221] The server collects hourly location information (latitude, longitude, altitude) from multiple base stations installed nationwide. The API endpoint of each base station is used as input. Specifically, the server sends an HTTP GET request to the API endpoint and temporarily stores the acquired location data (latitude, longitude, altitude) in memory. The output of this step is the collected location data.

[1222] Step 2:

[1223] The server stores the collected location information in a database and compares it with past location information. The current location data and past location information stored in the database are used as input. Specifically, the server inserts the collected location information into an SQL database and extracts past location information. Next, the server compares the new and old location information and calculates the horizontal and vertical displacement of the ground. The output of this step is the calculated horizontal and vertical displacement data.

[1224] Step 3:

[1225] The server filters the calculated deviation data to remove noise and outliers. The deviation data obtained in step 2 is used as input. Specifically, the server detects anomalous values ​​above a certain threshold and performs a data cleaning process to remove or correct them. The output of this step is the cleaned deviation data.

[1226] Step 4:

[1227] The server inputs the cleaned ground displacement data into the generative AI model and combines it with past earthquake data to make earthquake forecasts. The cleaned displacement data and past earthquake data are used as inputs. Specifically, the server formats this data and converts it into a format suitable for the generative AI model. The server then uses the prompt, "Predict the likelihood of the next earthquake based on the earthquake data from the past year and the current ground displacement data," to have the AI ​​model make a prediction. The output of this step is the probability of an earthquake occurring and the predicted results.

[1228] Step 5:

[1229] The device (user's smartphone or computer) implements an emotion engine and receives a notification of the earthquake forecast result. The prediction result obtained in step 4 is used as input. Specifically, when the device receives the notification, it activates the camera and microphone to collect and analyze the user's facial expression and voice data to recognize the emotion. The output of this step is the user's emotion data.

[1230] Step 6:

[1231] The server optimizes the notification content of the earthquake forecast results using the emotion data obtained from the emotion engine. The emotion data and the earthquake forecast results are used as inputs. Specifically, the server runs an algorithm that uses the user's emotion data to generate optimal notification content. For example, if the user is feeling "anxious," the notification content will be provided with gentle language and reassuring information. The output of this step is the optimized notification content.

[1232] Step 7:

[1233] The server notifies the local government and user devices of the optimized prediction results. The optimized notification content and the prediction results are used as input. Specifically, the server references the local government and user contact information and selects the appropriate notification method (e.g., email, SMS, app notification, etc.). The output of this step is the sent notification.

[1234] Step 8:

[1235] The server stores the prediction results and user emotion data in a database and archives them in an analyzable format. The prediction results and emotion data are used as input. Specifically, the server inserts these data into the database and periodically moves them to archive storage for backup. The output of this step is the stored and archived data.

[1236] Step 9:

[1237] The server compares the actual earthquake occurrence situation with the predicted results, evaluates the accuracy of the generated AI model, and provides feedback. Actual earthquake occurrence data and predicted results are used as input. Specifically, the server collects actual earthquake occurrence data and compares it with the predicted results to evaluate the accuracy of the model. Emotion data is also included in the evaluation, and the AI ​​model's training dataset is updated through feedback means, allowing for continuous improvement of the model. The output of this step is an updated AI model and a new training dataset.

[1238] (Application example 2)

[1239] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1240] Current earthquake forecasting systems aim to improve prediction accuracy, but lack consideration for the reliability of the forecast and the user's psychological state. In particular, no systems exist that take into account how earthquake forecasts actually affect users. This raises concerns that users may feel unnecessary anxiety or surprise. Therefore, it is necessary to provide a system that can accurately and quickly notify users of earthquake forecasts while taking into account their emotions.

[1241] 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.

[1242] In this invention, the server includes: means for collecting hourly location information (latitude, longitude, and altitude) from multiple base stations installed throughout the country; means for comparing the collected location information with previous location information and calculating horizontal and vertical displacement of the ground; means for inputting the calculated ground displacement data into a generative AI model and analyzing correlations with past earthquake data to make earthquake forecasts; means for recognizing the user's emotions using voice and image data and optimizing the notification content based on the emotion data; and means for storing the prediction results and emotion data in a database and archiving them in a form that can be analyzed later. This enables earthquake forecast notifications that take the user's psychological state into consideration, making it possible to provide fast and accurate earthquake forecasts while reducing the user's anxiety.

[1243] A "base station" is a device installed in a communications network to provide location information.

[1244] "Location information" is data including latitude, longitude, and altitude, and is information for identifying a geographical location.

[1245] "Ground displacement" is part of the crustal movement that occurs due to the effects of earthquakes and other events, and refers to changes in position in the horizontal and vertical directions.

[1246] A "generative AI model" is an artificial intelligence model used to predict earthquake occurrences using past earthquake data and other input data.

[1247] A "database" is a system for systematically storing and managing data.

[1248] "Emotion recognition" is a technology that analyzes a user's voice and image data to recognize the user's emotional state (for example, anxiety or surprise).

[1249] "Optimizing notification content" refers to adjusting the content of a notification message depending on the emotional state of the user receiving the notification.

[1250] "Archiving" means storing data for the long term so that it can be analyzed or referenced later.

[1251] "Noise reduction" is the process of removing outliers and irrelevant values ​​in data processing.

[1252] "Feedback measures" are methods for comparing actual earthquake occurrence conditions with predicted results to improve the accuracy of the generative AI model.

[1253] This paper explains a system that collects location information (latitude, longitude, altitude) every hour from multiple base stations installed throughout the country and calculates ground displacement based on this information. The core part of the system is executed by the server, terminals, and users.

[1254] The server first collects location information from base stations across the country using API endpoints. The technology used includes the Python requests library. The collected data is temporarily stored in memory, and then the location information is stored in a database, such as Databricks.

[1255] The server then compares the data with past position information to calculate the horizontal and vertical displacement of the ground. This analysis is performed using the GeoAnalytics library. The analyzed data is then filtered to remove noise and data cleaning, and outliers are removed. This process is also performed using the GeoAnalytics library.

[1256] The cleaned ground displacement data is then input into a generative AI model. This model uses machine learning libraries such as TensorFlow to analyze correlations with past earthquake data and generate earthquake forecasts. The results of this forecast are then sent to local governments and weather forecast service providers via a server.

[1257] The device recognizes emotions using the user's voice and image data. It uses the smartphone's camera and microphone to analyze the user's emotions using the OpenCV and Pyaudio libraries. The analysis results are sent to the server, which then optimizes the notification content based on the emotion. The NotificationEngine library is used for this purpose.

[1258] Furthermore, these prediction results and emotion data are stored in a database. This storage allows for post-analysis of the prediction results and serves as a feedback tool to help improve the AI ​​model. Data obtained from the emotion engine is also archived, with the aim of improving the accuracy of the model in the future.

[1259] To maximize the effectiveness of this system, the following specific example is given. For example, if the earthquake prediction result is highly likely and the user shows a surprised expression, the notification content can be customized as follows: "Notice. An earthquake of magnitude 6 has been predicted. However, please remain calm and be prepared to evacuate quickly. Necessary support information will also be provided."

[1260] The following is an example of a prompt sentence to input to the generative AI model.

[1261] You received a high probability earthquake forecast. What reassuring notification message would be appropriate for a user who recognized a surprised expression?

[1262] In this way, a system is constructed that takes into account the user's psychological state and provides accurate and prompt earthquake forecast notifications.

[1263] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1264] Step 1:

[1265] The server collects location information (latitude, longitude, and altitude) every hour from multiple base stations across the country. It sends a request to the API endpoint of each base station to obtain location information. The collected location information is temporarily stored in memory.

[1266] Input: Location information from base station

[1267] Output: Raw position information stored in memory

[1268] Step 2:

[1269] The server stores the collected location information in a database, then compares it with past location information to calculate horizontal and vertical displacements of the ground, thereby identifying fluctuations in location information.

[1270] Input: Raw location information stored in memory, past location information

[1271] Output: Calculated ground displacement data

[1272] Step 3:

[1273] The server filters and cleans the calculated ground displacement data, removing or correcting noise and outliers to produce a clean dataset.

[1274] Input: Calculated ground displacement data

[1275] Output: Clean ground displacement data

[1276] Step 4:

[1277] The server inputs clean ground displacement data into a generative AI model, which analyzes correlations with past earthquake data to forecast earthquakes. The generative AI model is trained using TensorFlow.

[1278] Input: Clean ground displacement data

[1279] Output: Earthquake forecast results

[1280] Step 5:

[1281] The device collects the user's voice and image data and uses them to recognize emotions. It uses OpenCV and Pyaudio libraries to analyze the user's facial expressions and voice to identify emotions.

[1282] Input: User's voice and image data

[1283] Output: Recognized emotion data

[1284] Step 6:

[1285] The server optimizes the notification content of earthquake forecast results based on emotion data. It uses the NotificationEngine library to generate customized notifications according to the emotion. For example, if the user is feeling "anxious," the server will provide the notification content with kind language and additional reassurance information.

[1286] Input: Earthquake forecast results, recognized emotion data

[1287] Output: Optimized notification content

[1288] Step 7:

[1289] The server then sends the earthquake forecast results and optimized notification content to local governments and users' devices via email and app notifications.

[1290] Input: Optimized notification content

[1291] Output: Notification to local government and user devices

[1292] Step 8:

[1293] The server stores and archives the prediction results and user emotion data in a database, which will be used as important information for future analysis and model improvement.

[1294] Input: Prediction results, user emotion data

[1295] Output: Analysis data stored in a database

[1296] Step 9:

[1297] The server compares the actual earthquake occurrence situation with the predicted results, and updates the learning dataset of the AI ​​model through a feedback means to improve the generated AI model, thereby continuously improving its accuracy.

[1298] Input: Actual earthquake occurrence situation, predicted results

[1299] Output: An improved generative AI model

[1300] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1301] 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.

[1302] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1303] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1304] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1305] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1306] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1307] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1308] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1309] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1310] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1311] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1312] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1313] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1314] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1315] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1316] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1317] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1318] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1319] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1320] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1321] The following is further disclosed regarding the above embodiment.

[1322] (Claim 1)

[1323] A means of collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country,

[1324] means for comparing the collected position information with previous position information and calculating horizontal and vertical displacements of the ground;

[1325] The calculated ground displacement data is input into a generative AI model, and correlations with past earthquake data are analyzed to forecast earthquakes.

[1326] A means of notifying local governments and weather forecast service providers of the forecast results;

[1327] means for storing the results of the prediction in a database and archiving them in a form that can be analyzed later;

[1328] A system including:

[1329] (Claim 2)

[1330] 10. The system of claim 1, further comprising filtering and data cleaning means for removing outliers and noise.

[1331] (Claim 3)

[1332] The system of claim 1 further comprising a feedback means for comparing the actual earthquake occurrence situation with the predicted results and improving the generated AI model.

[1333] "Example 1"

[1334] (Claim 1)

[1335] A means of collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country,

[1336] means for comparing the collected position information with previous position information and calculating horizontal and vertical displacements of the ground;

[1337] A means for inputting the calculated ground displacement data into a generation model and analyzing the correlation with past earthquake data to forecast earthquakes;

[1338] A means of notifying government agencies and weather forecast providers of forecast results;

[1339] a means for storing the results of the prediction in an information accumulation device and archiving them in a form that can be analyzed later;

[1340] A system including:

[1341] (Claim 2)

[1342] 10. The system of claim 1, further comprising filtering and data cleaning means for removing outliers and noise.

[1343] (Claim 3)

[1344] The system according to claim 1, further comprising a feedback means for comparing the actual earthquake occurrence situation with the predicted results and improving the generation model.

[1345] "Application Example 1"

[1346] (Claim 1)

[1347] A means of collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country,

[1348] means for comparing the collected position information with previous position information and calculating horizontal and vertical displacements of the ground;

[1349] The calculated ground displacement data is input into a generative AI model, and correlations with past earthquake data are analyzed to forecast earthquakes.

[1350] A means of notifying local governments and weather forecast service providers of the forecast results;

[1351] means for storing the results of the prediction in a database and archiving them in a form that can be analyzed later;

[1352] A means of notifying users of earthquake forecast results in real time via smartphones or smart glasses;

[1353] A means of providing safe evacuation routes in the event of an earthquake;

[1354] A system including:

[1355] (Claim 2)

[1356] 10. The system of claim 1, further comprising filtering and data cleaning means for removing outliers and noise.

[1357] (Claim 3)

[1358] The system of claim 1 further comprising a feedback means for comparing the actual earthquake occurrence situation with the predicted results and improving the generated AI model.

[1359] "Example 2: Combining Emotion Engines"

[1360] (Claim 1)

[1361] A means of collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country,

[1362] means for comparing the collected position information with previous position information and calculating horizontal and vertical displacements of the ground;

[1363] The calculated ground displacement data is input into a generative AI model, and correlations with past earthquake data are analyzed to forecast earthquakes.

[1364] A means for recognizing a user's emotion using an emotion engine of the user terminal and reflecting the data in generating optimized notification content;

[1365] A means for notifying local governments, weather forecast service providers, and users of the forecast results;

[1366] means for storing the results of the prediction in a database and archiving them in a form that can be analyzed later;

[1367] A system including:

[1368] (Claim 2)

[1369] 10. The system of claim 1, further comprising filtering and data cleaning means for removing outliers and noise.

[1370] (Claim 3)

[1371] The system of claim 1 further comprising means for comparing the actual earthquake occurrence situation with the predicted results, evaluating the accuracy of the generated AI model, and providing feedback.

[1372] "Application example 2 when combining emotion engines"

[1373] (Claim 1)

[1374] A means of collecting hourly location information (latitude, longitude, altitude) from multiple base stations installed throughout the country,

[1375] means for comparing the collected position information with previous position information and calculating horizontal and vertical displacements of the ground;

[1376] The calculated ground displacement data is input into a generative AI model, and correlations with past earthquake data are analyzed to forecast earthquakes.

[1377] A means of notifying local governments and weather forecast service providers of the forecast results;

[1378] means for recognizing a user's emotion using voice and image data and optimizing notification content based on the emotion data;

[1379] a means for storing the prediction results and emotion data in a database and archiving them in a form that can be analyzed later;

[1380] A system including:

[1381] (Claim 2)

[1382] 10. The system of claim 1, further comprising filtering and data cleaning means for removing outliers and noise.

[1383] (Claim 3)

[1384] The system of claim 1 further comprising a feedback means for comparing the actual earthquake occurrence situation with the predicted results and improving the generated AI model. [Explanation of symbols]

[1385] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for collecting location information every hour from a plurality of base stations installed throughout the country; means for comparing the collected position information with previous position information and calculating horizontal and vertical displacements of the ground; The calculated ground displacement data is input into a generative AI model, and correlations with past earthquake data are analyzed to forecast earthquakes. A means of notifying local governments and weather forecast service providers of the forecast results; means for storing the results of the prediction in a database and archiving them in a form that can be analyzed later; A system including:

2. 10. The system of claim 1, further comprising filtering and data cleaning means for removing outliers and noise.

3. The system according to claim 1, further comprising a feedback means for comparing the actual earthquake occurrence situation with the predicted results and improving the generated AI model.

Images