system

Abstract

Provide a system. [Solution] means for obtaining weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; a means for storing the analysis results in a database and updating the database in real time; means for notifying the user when certain weather conditions are predicted based on the user's settings; A means for a user to refer to detailed weather forecast information through a terminal; A system including:

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] Current weather forecasting systems can have difficulty making short-term predictions or detailed forecasts for specific regions. Furthermore, many current systems lack regular data updates, making it impossible to provide the latest forecasts. This means they are unable to respond immediately to changing weather conditions in real time. Furthermore, there is a need for improvement in the lack of notifications that address specific user needs (such as thunderstorms, typhoons, and pollen counts). [Means for solving the problem]

[0005] The present invention provides a means for acquiring weather-related data from an external database and preprocessing the data to fill in missing values ​​and remove noise. It also provides a system that includes a means for analyzing weather forecasts using a machine learning or deep learning model based on the preprocessed data. The analysis results are stored in a database and updated in real time. The system notifies users when specific weather conditions are predicted based on their settings and allows them to view detailed weather forecast information via their device. This allows users to receive the latest forecast information in real time and quickly take appropriate measures to deal with specific weather conditions.

[0006] Below are definitions of important terms contained in the claims.

[0007] An "external database" is an external data storage system used to store weather-related data.

[0008] "Weather-related data" refers to information necessary for weather forecasting, including temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.

[0009] "Preprocessing" is a process for normalizing the acquired weather-related data, filling in missing values, and removing noise.

[0010] "Missing value imputation" is a process of filling in missing data with other data or estimated values.

[0011] "Noise removal" is a process that removes unnecessary information from data.

[0012] A "machine learning model" refers to an algorithm that learns from collected data and makes predictions and classifications.

[0013] A "deep learning model" refers to an algorithm that uses a neural network with multiple neuron layers to learn the characteristics of data and perform complex predictions and classifications.

[0014] "Analyzing weather forecasts" refers to using machine learning or deep learning models to predict future weather conditions from collected data.

[0015] A database is a system that stores data in a structured format and allows for fast search and retrieval.

[0016] "Real-time updating" refers to the process of instantly updating data and information and keeping it up to date.

[0017] "Specific weather conditions" refer to specific weather conditions that the user is interested in, such as thunderstorms, typhoons, pollen counts, and the like.

[0018] "Notification" refers to sending specific information to a user as an alert or message.

[0019] "Terminal" refers to a device (smartphone, PC, tablet, etc.) that a user uses to display and manipulate information.

[0020] "Weather forecast information" refers to forecast results created based on meteorological data, and includes detailed information such as temperature, probability of precipitation, wind speed, and wind direction. [Brief explanation of the drawings]

[0021] [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

[0022] 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.

[0023] First, the terms used in the following description will be explained.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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."

[0029] [First embodiment]

[0030] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0031] 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.

[0032] 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).

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] 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.

[0039] 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.

[0040] 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.

[0041] 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."

[0042] The following describes in detail an embodiment of the present invention. The present invention provides a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to complement missing values, and removes noise. Furthermore, the system uses the preprocessed data to analyze weather forecasts using a machine learning or deep learning model, and stores the analysis results in a database that is updated in real time. Based on user settings, the system notifies users when specific weather conditions are predicted and allows users to view detailed weather forecast information via their device.

[0043] Data collection

[0044] The server retrieves weather-related data from an external database, including information on temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. This data is collected periodically using an API.

[0045] Data Preprocessing

[0046] The server preprocesses the acquired weather-related data. This includes data normalization, missing value completion, and noise removal. For example, if there are gaps in the data obtained from a particular sensor, the server uses surrounding data to complete the gaps. It also applies a noise removal filter to improve the quality of the data.

[0047] Data analysis

[0048] Using the preprocessed data, the server analyzes the weather forecast using machine learning or deep learning models. For short-term predictions, it uses LSTM (recurrent neural networks), while for long-term predictions, it uses ensemble learning or physical models. This enables detailed predictions of temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, typhoon paths, and more.

[0049] Database storage and real-time updates

[0050] The analysis results are stored in a database by the server. This database is designed to be updated in real time and frequently re-runs the predictive model, ensuring that the latest forecast data is always available to users.

[0051] Notification function

[0052] The device will send notifications to users when certain weather conditions are predicted based on their settings, such as when a thunderstorm is predicted in a specific area based on their settings for "thunderstorm warnings."

[0053] Information reference

[0054] Users can open the application on their devices to view detailed weather forecast information, including the latest weather data, weather radar, typhoon tracks, pollen count information, and other details. For example, if a user wants to check the weather forecast for their current location and destination before commuting, they can simply open the application and the latest information will be displayed.

[0055] Specific examples

[0056] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database. After data preprocessing, it analyzes it using an LSTM model and predicts a thunderstorm in one hour. This prediction result is saved in the database and updated in real time. Next, the device sends a notification based on the "thunderstorm warning" set by the user, and the user opens the application to check detailed weather forecast information. As a result, the user can change their plans and take appropriate measures against the thunderstorm.

[0057] In this way, the weather forecast system of the present invention obtains weather-related data from an external database and provides comprehensive functions for preprocessing, analysis, storage, notification, and information reference, thereby providing users with the latest and most accurate weather forecast information in real time.

[0058] The processing flow will be explained below.

[0059] Step 1:

[0060] The server retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database through a process that periodically fetches the data via an API.

[0061] Step 2:

[0062] The weather-related data acquired by the server is preprocessed, which includes processes such as data normalization, missing value imputation, and noise removal. For example, missing data imputation uses an algorithm that calculates an estimate using surrounding data points.

[0063] Step 3:

[0064] The server then uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. It uses LSTM (long short-term memory) networks for short-term predictions and ensemble learning models for long-term predictions. This generates detailed forecasts such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon paths.

[0065] Step 4:

[0066] The server stores the analysis results in a database that is designed to be updated in real time, with the predictive model being re-run on new data at regular intervals (e.g., every 30 minutes).

[0067] Step 5:

[0068] The server generates a notification when certain weather conditions are predicted based on the user's settings. For example, a notification is generated when "thunderstorms are predicted in Tokyo within the next hour."

[0069] Step 6:

[0070] The device receives notifications sent from the server, which are customized based on the user's settings. For example, if the user requests thunderstorm notifications, the device immediately provides that information to the user.

[0071] Step 7:

[0072] Users can open the application on their devices to view detailed weather forecast information, including the latest weather forecast, weather radar, typhoon tracks, pollen count information, etc., which they can use to decide on appropriate actions.

[0073] Step 8:

[0074] The server periodically collects the latest data in real time, re-runs the prediction model to generate the latest prediction results, and stores them in the database again, so that the latest prediction information is always available to users.

[0075] In this way, each step works in cooperation with the other steps to provide users with highly accurate weather forecasts in real time.

[0076] Example 1

[0077] 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."

[0078] Conventional weather forecasting systems often lack accuracy and real-time response in the entire process, from collecting weather data to analyzing and notifying users. They also face difficulties in accessing the detailed weather information users require, and lack the ability to quickly notify users of specific weather conditions. In particular, inadequate handling of missing values ​​and noise reduces data quality and negatively impacts forecast accuracy. Furthermore, there is a need for greater system efficiency and smoother data updates.

[0079] 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.

[0080] In this invention, the server includes: means for acquiring weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; means for saving the analysis results in a database and updating them in real time; means for notifying the user when specific weather conditions are predicted based on user settings; means for the user to refer to detailed weather forecast information via a terminal; means for using an API to collect the data and preprocessing the collected data by filling in missing values ​​and applying a noise removal filter; means for using a recurrent neural network and an ensemble learning model to analyze the weather forecast; means for using a relational database management system to save and update the data in the database in real time; and means for using a cloud messaging service for the notification means. This enables efficient and highly accurate data collection, analysis, storage, notification, and reference.

[0081] An "external database" is a database that provides weather-related data and is accessible via the Internet or the like.

[0082] "API" is an abbreviation for Application Programming Interface, a means of exchanging data and functions between different software systems.

[0083] "Weather-related data" refers to all weather-related information, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0084] "Preprocessing" refers to processes to process the acquired data into a state suitable for analysis, such as normalizing the data, filling in missing values, and removing noise.

[0085] "Missing value imputation" is the process of filling in missing values ​​in a dataset with surrounding data, estimated values, etc.

[0086] "Noise removal" is a process for removing irregular fluctuations and outliers from data to improve data quality.

[0087] A "machine learning model" is an algorithm or method that learns patterns in data and makes predictions or classifications about future data.

[0088] A "deep learning model" is a highly accurate algorithm that uses a multi-layer neural network to extract data features and perform predictions and classifications.

[0089] A "recurrent neural network" is a special neural network for handling time series data, and is a model that retains past information and makes predictions.

[0090] "Ensemble learning" is a technique that combines multiple machine learning models to achieve higher predictive accuracy than individual models.

[0091] A "relational database management system" is a software system for storing, updating, searching, and managing data, and manages structured data in tabular format.

[0092] A "cloud messaging service" is a service for sending notification messages over the Internet, and is a means of providing real-time notifications to mobile devices and the like.

[0093] "User settings" refers to individual setting information such as notification conditions and reference conditions customized by the user on the system.

[0094] "Terminal" refers to an electronic device, such as a smartphone, tablet, or PC, that a user uses to access the system.

[0095] "Real-time updates" means that the data in the system is always kept up to date and updates are reflected almost immediately.

[0096] MODE FOR CARRYING OUT THE INVENTION

[0097] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention is a weather forecast system in which a server, a terminal, and a user are used in cooperation with each other.

[0098] Data collection

[0099] The server retrieves weather-related data from an external database. Specifically, it collects data via an API. The Python requests library is used to retrieve the data, and weather-related data includes temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. For example, the OpenWeatherMap API is used to retrieve this data.

[0100] Data Preprocessing

[0101] The server preprocesses the acquired weather-related data. This includes data normalization, missing value imputation, and noise removal. It uses Python's Pandas library to impute missing values ​​with surrounding data and applies SciPy's noise removal filter. This improves the quality of the data and prepares it for analysis.

[0102] Data analysis

[0103] The server uses preprocessed data to analyze weather forecasts using machine learning and deep learning models. It uses recurrent neural networks (LSTM) for short-term predictions and ensemble learning models for long-term predictions. It builds predictive models using libraries such as TENSORFLOW (registered trademark) and Keras to analyze temperature, precipitation probability, wind speed, wind direction, and more.

[0104] Database storage and real-time updates

[0105] The server stores the analysis results in a database and updates them in real time. The database uses a relational database management system such as PostgreSQL. The SQLAlchemy library is used to efficiently store and update data. This ensures that the latest prediction data is always available to users.

[0106] Notification function

[0107] The device sends notifications when certain weather conditions are predicted based on the user's settings. For example, if the user has set a "thunderstorm alert," it sends a push notification using a cloud messaging service. For this purpose, it uses Firebase Cloud Messaging (FCM).

[0108] Information reference

[0109] Users can access detailed weather forecast information on their devices. Through the application UI, they can view the latest weather data, weather radar, typhoon tracks, pollen count information, and more. For example, an application developed with React Native and Flutter (registered trademark) allows users to easily check the weather forecast for their current location and destination before commuting.

[0110] Specific examples

[0111] For example, on a certain day in Tokyo, the server retrieves weather data using the OpenWeatherMap API. After preprocessing the data, an LSTM model is used to predict thunderstorms one hour in the future. This prediction result is stored in a PostgreSQL database and updated in real time. Next, the device sends a notification via FCM based on the "thunderstorm warning" set by the user. The user opens the application created with React Native and checks detailed weather forecast information. As a result, the user can change their plans and take appropriate measures against the thunderstorm.

[0112] Prompt Sentence Examples

[0113] "Get the latest weather data for Tokyo and predict the chance of thunderstorms in the next hour."

[0114] In this way, by linking the server, terminal, and user, we provide a system that efficiently and accurately accomplishes everything from data collection to analysis, storage, notification, and reference.

[0115] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0116] Step 1: Data collection

[0117] The server retrieves weather-related data from an external database. Specifically, it sends a request to the API and receives data in JSON format, including temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. The input is the endpoint URL and API key as an API request, and the output is weather-related data in JSON format. This data collection occurs periodically, and the next step is taken whenever new data is retrieved.

[0118] Step 2: Data Preprocessing

[0119] The server preprocesses the weather-related data in JSON format. First, it converts the data into a data frame using the Pandas library. Next, it imputes missing values ​​based on neighboring data points and applies a SciPy denoising filter to improve the data quality. The input is weather data in JSON format, and the output is a preprocessed data frame.

[0120] Step 3: Data analysis

[0121] The server uses the preprocessed data to analyze weather forecasts using machine learning and deep learning models. First, it builds LSTM models and ensemble learning models using TensorFlow and Keras. Next, it inputs the preprocessed data into these models to obtain forecast results such as temperature, precipitation probability, wind speed, and wind direction. The input is the preprocessed data frame, and the output is the analyzed weather forecast results.

[0122] Step 4: Save to database and update in real time

[0123] The server saves the analysis results in a database and updates them in real time. First, it connects to a PostgreSQL database using SQLAlchemy. Then it writes the analysis results to a database table and updates existing records if any. The input is the analyzed weather forecast results, and the output is the latest weather forecast data stored in the database.

[0124] Step 5: Notifications

[0125] The device sends a notification when certain weather conditions are predicted based on the user's settings. First, a notification message is created using Firebase Cloud Messaging (FCM). Then, the notification is sent at the appropriate time based on the conditions set by the user. The input is the user's settings and the analyzed weather forecast results, and the output is the notification message sent to the device.

[0126] Step 6: Information Reference

[0127] The user opens the application on their device to view detailed weather forecast information. The application UI displays the latest weather data, weather radar, typhoon tracks, and pollen count information. The input is the latest weather forecast data stored in the database, and the output is detailed weather forecast information displayed on the user's screen.

[0128] To give a concrete example, on a certain day in Tokyo, the server obtains weather data using the OpenWeatherMap API (Step 1). Next, the data is preprocessed to fill in missing values ​​and remove noise (Step 2). After that, the weather forecast is analyzed using an LSTM model (Step 3), and the analysis results are saved and updated in a PostgreSQL database (Step 4). The device sends a notification via FCM based on the "thunderstorm warning" set by the user (Step 5). Upon receiving this notification, the user opens the application and checks detailed weather forecast information (Step 6). This allows the user to change their plans and take appropriate measures against the thunderstorm.

[0129] (Application example 1)

[0130] 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."

[0131] Existing weather forecasting systems have technical limitations in the preprocessing and analysis of weather data, and are particularly lacking in providing suggestions that have a concrete impact on users' lives and business activities. Furthermore, there are no effective systems for demand forecasting or optimizing delivery plans for food delivery services. As a result, users are unable to adequately respond to the inconveniences and risks caused by weather conditions, and business efficiency is reduced.

[0132] 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.

[0133] In this invention, the server includes: means for acquiring weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; means for saving the analysis results in a database and updating them in real time; means for notifying the user when specific weather conditions are predicted based on the user's settings; means for the user to refer to detailed weather forecast information via a terminal; and means for making specific suggestions that will influence the user's behavior based on the weather forecast information. This makes it possible to propose specific impacts on daily life and business activities based on weather conditions. Furthermore, demand forecasting and optimization of delivery plans can be realized in food delivery services, improving user and business efficiency.

[0134] An "external database" is a system that exists on the Internet or in a cloud environment and accumulates and provides weather-related data.

[0135] "Weather-related data" refers to data that indicates weather conditions such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion status.

[0136] "Preprocessing" refers to the process of performing processes on the acquired weather-related data, such as normalizing the data, filling in missing values, and removing noise.

[0137] A "machine learning model" is an algorithm that learns patterns and rules based on data and makes predictions and classifications for new data.

[0138] A "deep learning model" is an algorithm that uses a multi-layer neural network to automatically learn the characteristics of complex data and perform advanced predictions and classifications.

[0139] "Real-time updates" means that information in a database or system is kept up-to-date immediately whenever new data is added.

[0140] A "terminal" is a device such as a smartphone, tablet, or PC that allows a user to access the Internet and obtain information.

[0141] The "particular weather conditions" are weather phenomena that the user has set in advance as being of interest or concern, such as thunderstorms, typhoons, and heavy snow.

[0142] "Notification" is a feature that sends alerts when certain weather conditions are predicted based on conditions set by the user in advance.

[0143] "Specific suggestions to influence user behavior" refers to providing specific advice to users on their daily lives and business activities based on weather conditions, such as encouraging them to change the timing of going out or ordering food delivery.

[0144] "Demand forecasting" refers to predicting the number of orders for food delivery services in advance based on weather conditions and past data.

[0145] "Optimal delivery planning" refers to optimizing delivery routes and times based on predicted demand in order to provide efficient food delivery services.

[0146] The present invention provides a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to fill in missing values, and removes noise. The system then uses the preprocessed data to analyze weather forecasts using machine learning or deep learning models, and stores the analysis results in a database that is updated in real time. Based on user settings, the system notifies users when specific weather conditions are predicted and allows them to access detailed weather forecast information via their device.

[0147] Data collection

[0148] The server retrieves weather-related data from external databases. This data includes information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. This data is collected periodically using APIs, such as the OpenWeatherMap API and the Japan Meteorological Agency API.

[0149] Data Preprocessing

[0150] The server preprocesses the acquired weather-related data. This preprocessing includes data normalization, missing value completion, and noise removal. The Python pandas library is used here. For example, if there are gaps in the data obtained from a particular sensor, the gaps are filled in using surrounding data, and a noise removal filter is applied to improve the quality of the data.

[0151] Data analysis

[0152] Using the preprocessed data, the server analyzes the weather forecast using machine learning or deep learning models. For short-term predictions, it uses a recurrent neural network (LSTM) model using TensorFlow or PyTorch, while for long-term predictions, it uses an ensemble learning model. This enables detailed predictions such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon paths.

[0153] Database storage and real-time updates

[0154] The analysis results are stored in a database by the server. This database is designed to be updated in real time using Firebase Realtime Database or Amazon RDS, and the predictive model is re-executed frequently, ensuring that the latest forecast data is always provided to users.

[0155] Notification function

[0156] The device will send notifications when certain weather conditions are predicted based on the user's settings, such as "thunderstorm warnings" and "pollen alerts," using Firebase Cloud Messaging (FMC).

[0157] Information references and specific suggestions

[0158] Users can open the application on their devices to view detailed weather forecast information, including the latest weather data, weather radar, typhoon tracks, pollen count information, and other details. Based on the weather forecast information, the application will also provide specific suggestions to influence users' behavior, such as recommending when to go out or ordering food delivery.

[0159] Specific examples

[0160] For example, on a given day in Tokyo, the server retrieves the latest weather data from the weather database. After data preprocessing, it runs an LSTM model to predict a thunderstorm in one hour. This prediction result is saved in the database and updated in real time. The device then sends a notification based on the user's "thunderstorm warning" setting, and the user opens an application to check detailed weather forecast information. As a result, the user can change their plans, order food delivery, and take appropriate precautions against the thunderstorm.

[0161] Examples of specific prompts include the following:

[0162] "Predict the increase in delivery requests in Tokyo next week based on the amount of precipitation and average wind speed."

[0163] In this way, the weather forecast system of the present invention acquires weather-related data from an external database and provides comprehensive functions for preprocessing, analysis, storage, notification, information reference, and specific suggestions, thereby providing users with the latest and most accurate weather forecast information and action suggestions in real time.

[0164] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0165] Step 1:

[0166] Data collection

[0167] The server retrieves weather-related data from an external database. Specifically, it periodically requests information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count using the OpenWeatherMap API and the Japan Meteorological Agency API, and receives JSON-formatted data in response. The input data is raw weather data retrieved from the API, and the output data is this raw weather data.

[0168] Step 2:

[0169] Data Preprocessing

[0170] The server performs data preprocessing using the acquired weather-related data. Specifically, it normalizes the data using Python's pandas library, imputes missing values ​​using SimpleImputer, and removes noise using a noise removal filter. The input data is raw weather data, and the output data is preprocessed weather data that has been normalized, imputes missing values, and removed noise.

[0171] Step 3:

[0172] Data analysis

[0173] Using the preprocessed data, the server analyzes the weather forecast using a machine learning or deep learning model. The preprocessed data is input into an LSTM model or ensemble learning model built using TensorFlow or PyTorch, and weather forecasts such as future temperature, precipitation probability, wind speed, wind direction, thunderstorm predictions, and typhoon paths are output. The input data is preprocessed weather data, and the output data is analyzed weather forecast data.

[0174] Step 4:

[0175] Database storage and real-time updates

[0176] The server stores the analysis results in Firebase Realtime Database or Amazon RDS and updates them in real time. Specifically, the analyzed weather forecast data is written to these databases and set up to be automatically updated each time new data is added. The input data is the analyzed weather forecast data, and the output data is the latest weather forecast data stored in the database.

[0177] Step 5:

[0178] Notification function

[0179] The device sends notifications when certain weather conditions are predicted based on the user's settings. Specifically, it uses Firebase Cloud Messaging (FCM) to instantly send notifications about predicted weather conditions based on the user's settings for "thunderstorm warnings" and "pollen alerts." The input data is the analyzed weather forecast data retrieved from the database and the user's settings, and the output data is the notification message sent over the Internet.

[0180] Step 6:

[0181] Information references and specific suggestions

[0182] Users can access the application through their devices to view detailed weather forecast information. Based on the weather forecast information, the application also provides specific suggestions, such as when to go out or encouraging people to order food delivery. The input data is the weather forecast information displayed on the device and data related to the user's suggested actions, and the output data is the specific suggested actions provided to the user.

[0183] Step 7:

[0184] Food delivery demand forecasting and delivery plan optimization

[0185] The server predicts demand for food delivery services based on weather conditions and past data, and then formulates optimal delivery plans based on the predicted demand. This involves using a machine learning model to predict demand and then running an algorithm to optimize delivery routes and times based on the prediction results. The input data is preprocessed weather data and past delivery records, and the output data is the demand forecast and optimal delivery plan information.

[0186] 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.

[0187] The following describes in detail an embodiment of the present invention. The present invention combines a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to fill in missing values, and removes noise, with an emotion engine that recognizes user emotions. The system updates the weather forecast in real time based on the analysis results and notifies the user when specific weather conditions are predicted based on the user's settings. The system also allows the user to view detailed weather forecast information via a terminal.

[0188] Data collection

[0189] The server periodically retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database. This data is fetched via an API.

[0190] Data Preprocessing

[0191] The server preprocesses the collected weather-related data, including normalizing the data, filling in missing values, and removing noise. Specifically, it fills in missing data with surrounding data points and applies a noise removal filter to improve data quality.

[0192] Data analysis

[0193] The server uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. LSTM (long short-term memory) networks are used for short-term predictions, while ensemble learning models are used for long-term predictions. This results in a detailed weather forecast (temperature, probability of precipitation, wind speed, wind direction, thunderstorm forecast, and typhoon path).

[0194] Database storage and real-time updates

[0195] The analysis results are stored in a database by the server and updated in real time. New data is collected at regular intervals (e.g., every 30 minutes), and the predictive model is re-run to generate the latest forecast results.

[0196] Emotion analysis using an emotion engine

[0197] The server uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state from their voice, facial expressions, text input, etc. The analyzed emotional state is stored in a database and reflected in the content and format of the weather forecast notification.

[0198] User Notification Settings

[0199] The device will send notifications that take into account the user's emotional state based on specific weather conditions (thunderstorms, typhoons, pollen counts, etc.) set by the user. For example, if the user has set up a thunderstorm notification, the device will send a notification with more detailed information and follow-up advice depending on the user's stress level detected by the emotion engine.

[0200] Information reference

[0201] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, and other details. The priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0202] Specific examples

[0203] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database, and after preprocessing, analyzes it using an LSTM model, predicting a thunderstorm in one hour. The analysis results are stored in the database and updated in real time. The server then analyzes the user's stress level using an emotion engine and creates a notification based on that state. The device receives this notification and sends it to the user with detailed advice such as, "Thunderstorms are predicted. Avoid sudden outings or bring an umbrella." The user can then change their plans based on this notification and take appropriate precautions against the thunderstorm.

[0204] In this way, the system of the present invention acquires weather-related data from an external database, and in addition to preprocessing, analyzing, storing, notifying, and referencing the information, it uses an emotion engine to provide personalized weather information based on the user's emotional state.

[0205] The processing flow will be explained below.

[0206] Step 1:

[0207] The server retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database by fetching the data periodically using an API.

[0208] Step 2:

[0209] The weather-related data acquired by the server is preprocessed. Specifically, data normalization, missing value completion, and noise removal are performed. For example, if there are missing parts in the acquired data, the parts are completed using surrounding data points.

[0210] Also, a noise reduction filter is applied to improve the quality of the data.

[0211] Step 3:

[0212] The server then uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. It uses LSTM (long short-term memory) networks for short-term predictions and ensemble learning models for long-term predictions. This generates detailed forecasts such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon tracks.

[0213] Step 4:

[0214] The server stores the analysis results in a database that is designed to be updated in real time, collecting new data at regular intervals (e.g., every 30 minutes) and re-running the predictive model to generate the latest forecasts.

[0215] Step 5:

[0216] The server uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state from their voice, facial expressions, text input, etc. The analyzed emotional state is stored in a database.

[0217] Step 6:

[0218] The server adjusts the content and format of notifications based on user settings when certain weather conditions (e.g., thunderstorms or typhoons) are predicted. For example, if the emotion engine determines that the user is feeling stressed, it creates a notification with follow-up advice appropriate to that state.

[0219] Step 7:

[0220] The device receives notifications sent from the server. These notifications are customized based on the user's settings and emotional state. For example, if a user has set up "thunderstorm notifications" and a stressful state is detected, the device receives a notification with specific advice such as "A thunderstorm is predicted within 30 minutes. Avoid sudden outings and bring an umbrella if necessary."

[0221] Step 8:

[0222] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, and pollen count information. For example, before commuting, users can check the weather forecast for their current location and destination to determine the best course of action.

[0223] Step 9:

[0224] The server periodically collects the latest data in real time, re-runs the prediction model to generate the latest prediction results, and stores them in the database again, thereby always providing the latest prediction information to users.

[0225] In this way, each step works in conjunction with the other steps to provide a personalized, real-time, and highly accurate weather forecast based on the user's emotional state.

[0226] Example 2

[0227] 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."

[0228] The appropriate collection and prediction of weather-related data is important for daily life and industrial activities. However, improving the accuracy of weather forecasts and providing information tailored to individual users' situations presents many challenges. In particular, existing systems are unable to adequately address issues such as missing or noisy data, real-time updates, and generating notifications that take into account the user's emotional state.

[0229] The identification processing 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 acquiring weather-related data from an external database, means for preprocessing the weather-related data to complement missing values ​​and remove noise, means for analyzing a weather forecast using a machine learning model or a deep learning model using the preprocessed data, means for saving the analysis results in a database and updating them in real time, means for recognizing and analyzing a user's emotional state, means for providing notifications based on specific weather conditions and the user's emotional state based on user settings, and means for the user to refer to detailed weather forecast information via a terminal. This allows a series of processes, such as collection, preprocessing, analysis, storage, and notification of weather-related data, to be efficiently executed, making it possible to provide personalized weather forecast information that takes the user's emotional state into consideration.

[0230] An "external database" is a collection of data that resides outside the system and stores and provides weather-related data.

[0231] "Weather-related data" refers to data relating to weather and meteorological conditions, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0232] "Preprocessing" refers to a series of processes performed on acquired weather-related data, such as data normalization, missing value completion, and noise removal.

[0233] "Missing value imputation" is the process of filling in missing data points in a dataset using surrounding data.

[0234] "Noise removal" refers to the process of filtering out abnormal values ​​and unnecessary data from data.

[0235] A "machine learning model" is an algorithm for predicting future data based on past data, and primarily uses statistical methods.

[0236] A "deep learning model" is an algorithm that uses neural networks to analyze data in a multi-layered manner and make advanced predictions.

[0237] "Real-time updating" refers to the process of collecting data at regular intervals and periodically replacing the analysis results with the latest ones.

[0238] "Emotional state" refers to the psychological and emotional state of the user as analyzed from their voice, facial expressions, text input, etc.

[0239] "Notifications" are messages or alerts that provide information about specific weather conditions based on the user's preferences and emotional state.

[0240] A "terminal" is a device (e.g., smartphone, tablet, computer) that a user uses to access and operate the system.

[0241] "Detailed weather forecast information" is weather forecast data that goes beyond simple weather forecasts and includes additional information such as weather radar, typhoon tracks, and pollen dispersion information.

[0242] The present invention is a weather forecasting system that retrieves weather-related data from an external database, preprocesses it to fill in missing values, and removes noise. The system also incorporates an emotion engine that recognizes the user's emotions. The system updates the weather forecast in real time based on the analysis results and notifies the user when specific weather conditions are predicted based on the user's settings. The system also allows the user to view detailed weather forecast information via their device.

[0243] The server periodically retrieves weather-related data (e.g., temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database. This data is fetched via an API. For example, the data could be retrieved using the OpenWeatherMap API.

[0244] The server preprocesses the collected weather-related data. Specific preprocessing steps include normalizing the data, filling in missing values, and removing noise. For example, the Pandas library is used to fill in NaN values ​​with approximate values, and the Scipy library is used to apply a low-pass filter to smooth the data.

[0245] The server uses the preprocessed data to analyze weather forecasts using machine learning or deep learning models. It uses a long short-term memory (LSTM) network for short-term predictions and an ensemble learning model for long-term predictions. Specific implementations could use frameworks such as TensorFlow or PyTorch.

[0246] The analysis results are stored in a database by the server and updated in real time. New data is collected at regular intervals (for example, every 30 minutes), and the predictive model is re-executed to generate the latest prediction results. A relational database such as MySQL (registered trademark) or PostgreSQL can be used as the database.

[0247] The server uses an emotion engine to recognize the user's emotions. This engine analyzes the user's emotional state based on voice, facial expressions, and text input. For example, it can capture facial expressions using OpenCV and analyze them with a deep learning model, or convert voice data into text using a speech recognition API such as IBM Watson (registered trademark), and then perform emotion analysis.

[0248] The device sends notifications based on specific weather conditions (e.g., thunderstorms, typhoons, pollen counts, etc.) set by the user, taking into account the user's emotional state. A real-time notification function is implemented using JavaScript (registered trademark) and the notification is sent in a way that reflects the user's emotional state. For example, if the user is feeling stressed, a detailed and kind notification will be sent.

[0249] Users can access detailed weather forecast information by opening the app on their devices. The app displays the latest forecast data, weather radar, typhoon tracks, pollen count information, etc. Furthermore, the priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0250] Specific examples

[0251] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database, and after preprocessing, analyzes it using an LSTM model, predicting a thunderstorm in one hour. The analysis results are stored in the database and updated in real time. The server then analyzes the user's stress level using an emotion engine and creates a notification based on that state. The device receives this notification and sends it to the user with detailed advice such as, "Thunderstorms are predicted. Avoid sudden outings or bring an umbrella." The user can then change their plans based on this notification and take appropriate precautions against the thunderstorm.

[0252] Examples of prompt statements

[0253] "Describe a scenario where a server retrieves the latest weather data from an external database and predicts a thunderstorm in the next hour in Tokyo on a certain day. The emotion engine detects the user's stress state and sends a notification with detailed advice."

[0254] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0255] Step 1: Data collection

[0256] The server periodically retrieves weather-related data from an external database. Specifically, the server sends an API request and receives a JSON-formatted response from the external database. This input data includes temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. As output, this weather-related data is obtained and passed to the next preprocessing step.

[0257] Example of how it works:

[0258] The server sends an API request every hour.

[0259] Receive the API response (weather data JSON) and save it in local storage.

[0260] Step 2: Data Preprocessing

[0261] The server preprocesses the collected weather-related data. This preprocessing involves normalizing the data, filling in missing values, and removing noise. The input is the raw data collected in step 1, and the output is preprocessed, high-quality data.

[0262] Example of how it works:

[0263] Use the Pandas library to read the JSON data and convert it to a dataframe.

[0264] To impute missing values, fill NaN values ​​in the data frame with the average of the preceding and following data points.

[0265] Apply a low-pass filter using the Scipy library to remove noise.

[0266] Step 3: Data analysis

[0267] The server uses the preprocessed data to analyze the weather forecast with a machine learning or deep learning model, where the input is the preprocessed data from step 2 and the output is the weather forecast data.

[0268] Example of how it works:

[0269] Use TensorFlow or PyTorch to load the LSTM model and run predictions using the preprocessed data as input.

[0270] Short-term predictions (1 to 24 hours) are performed using LSTM, and long-term predictions (1 week) are performed using an ensemble learning model.

[0271] Evaluate the analysis results and record performance metrics.

[0272] Step 4: Save to database and update in real time

[0273] The analysis results are stored in a database by the server and updated at regular intervals. The input is the analysis result from step 3, and the output is the prediction results stored in the database.

[0274] Example of how it works:

[0275] Connect to a MySQL or PostgreSQL database and insert the analysis results into a table.

[0276] Every 30 minutes, start with a new data collection and rerun the preprocessing and analysis.

[0277] Delete the old prediction results in the database and update with the new results.

[0278] Step 5: Emotion analysis using the emotion engine

[0279] The server uses an emotion engine to recognize the user's emotions. The input is the user's voice, facial expression, text input, etc., and the output is the user's emotional state.

[0280] Example of how it works:

[0281] Facial expressions are captured from a webcam using OpenCV and analyzed using a deep learning model.

[0282] Voice data is converted into text through a voice recognition API and sentiment analysis is performed.

[0283] Text analysis algorithms are used to recognize emotions from messages and comments.

[0284] Step 6: User Notification Settings

[0285] The device sends notifications based on specific weather conditions set by the user, taking into account their emotional state. The input is the data from steps 4 and 5, and the output is the notification sent to the user.

[0286] Example of how it works:

[0287] Check the user's settings and generate notifications when certain weather conditions (such as thunderstorms, typhoons, and pollen counts) are predicted.

[0288] Send real-time notifications to your device using JavaScript.

[0289] If the user is feeling stressed, a detailed notification containing kind words is generated.

[0290] Step 7: Information Reference

[0291] The user opens the application on the device and views detailed weather forecast information. The input is the forecast result stored in the database in step 4, and the output is the weather forecast information displayed on the device.

[0292] Example of how it works:

[0293] Open a mobile or web application.

[0294] The app screen displays the latest weather forecast, weather radar, typhoon path, pollen information, and more.

[0295] The information display format is automatically adjusted according to the user's emotional state.

[0296] (Application example 2)

[0297] 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."

[0298] In today's world, highly accurate weather forecasts have become important information in many fields. However, simply providing weather information makes it difficult to present users with measures appropriate to their individual needs and situations. In particular, physical stores require inventory management that takes weather conditions into account and appropriate product recommendations to customers. Furthermore, services that do not take into account user emotions make it difficult to improve customer satisfaction.

[0299] Therefore, the objective of the present invention is to provide a system that combines weather data and a user's emotional state to optimize inventory management and promotion notifications in physical stores and meet individual needs.

[0300] 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.

[0301] In this invention, the server includes means for acquiring weather-related data from an external database, means for preprocessing the weather-related data to fill in missing values ​​and remove noise, means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data, means for storing the analysis results in a database and updating them in real time, means for notifying the user when specific weather conditions are predicted based on user settings, means for the user to refer to detailed weather forecast information via a terminal, means for performing demand forecasts based on the weather data and optimizing inventory in physical stores, and means for sending personalized promotional notifications according to the user's emotional state. This enables optimal inventory management in response to weather conditions and individual promotional notifications that take the user's emotional state into consideration.

[0302] "Weather-related data" refers to data relating to weather and meteorological conditions, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0303] "Preprocessing" refers to the process of preparing data in an appropriate form before data analysis, such as normalizing data, filling in missing values, and removing noise.

[0304] A "machine learning model" is a model that learns patterns and regularities based on data and makes predictions and classifications.

[0305] A "deep learning model" is a model that uses a multi-layer neural network that can learn complex data patterns.

[0306] "Weather forecast" is information that predicts future weather conditions based on acquired weather data.

[0307] "Real-time updates" means obtaining new data and updating the analysis results based on that data at any time to provide the latest information.

[0308] The "notification means" is a means for notifying the user of information when a specific condition is met.

[0309] A "brick and mortar store" is a retail or service establishment that exists in a physical location.

[0310] "Inventory management" refers to the proper management and adjustment of the quantity of goods and resources, and maintaining an optimal inventory level according to demand.

[0311] A "promotional notice" is a notice to inform users of information about a particular product or service as part of an advertisement or promotion.

[0312] "Emotional state" refers to the user's emotional state, such as stress, joy, or sadness.

[0313] "Personalized" means providing information and services tailored to the specific needs and circumstances of each individual user.

[0314] This system acquires weather-related data from an external database, performs missing value imputation and noise removal, and then generates highly accurate weather forecasts using machine learning and deep learning models. It also analyzes the user's emotional state and provides personalized notifications and information based on that.

[0315] Data collection and preprocessing

[0316] The server periodically retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database (e.g., a weather API). This data is fetched via the API. The server then preprocesses the collected weather data, imputes missing values, and removes noise. Specifically, it uses Scikit-learn's SimpleImputer to impute missing data and StandardScaler to normalize the data.

[0317] Weather forecast analysis

[0318] Using the preprocessed weather data, the server analyzes the weather forecast using machine learning models (e.g., LSTM models) or deep learning models. LSTM (long short-term memory) networks are used for short-term predictions, while ensemble learning models are used for long-term predictions. This results in a detailed weather forecast (temperature, probability of precipitation, wind speed, wind direction, thunderstorm forecast, and typhoon path). The analysis results are stored in a database and updated in real time.

[0319] Sentiment Analysis and Notifications

[0320] The server implements an emotion analysis engine to analyze the user's emotional state. This engine analyzes the user's voice, facial expressions, text input, etc. to determine the user's emotional state. For example, it analyzes text input using an emotion recognition library such as Emojify.

[0321] Based on the results of sentiment analysis, notifications are sent when certain weather conditions are predicted. For example, if a thunderstorm is predicted and the user is feeling stressed, a personalized notification with detailed advice is sent. The notification may include specific advice such as, "Thunderstorms are predicted. Avoid sudden outings and, if necessary, check out stress-relieving products."

[0322] Inventory management and promotion notifications

[0323] The server uses weather data to predict demand and optimize inventory in physical stores. For example, if rain is predicted, it increases the stock of umbrellas and raincoats. It also sends personalized promotional notifications based on the user's emotional state. For example, if it determines that the user needs relaxation, it notifies them of promotions for relaxation-related products.

[0324] Information reference

[0325] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, and other details. The priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0326] Specific examples

[0327] For example, if a thunderstorm is predicted for Tokyo on a certain day and the user is feeling stressed, the following notification will be displayed: "Thunderstorms are predicted. Avoid sudden outings and, if necessary, check for products to relieve stress."

[0328] Prompt Sentence Examples

[0329] "Based on data obtained from a weather data API, impute and normalize missing values ​​and use an LSTM model to predict the weather. Next, analyze the user's emotional state from input and generate a promotional notification for relaxation-related products if the user is feeling stressed when a thunderstorm is predicted."

[0330] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0331] Step 1:

[0332] The server retrieves weather-related data from an external database through an API, which contains information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. The input is the API endpoint, and the output is the weather-related data.

[0333] Step 2:

[0334] The server preprocesses the acquired weather-related data. Specifically, it uses SimpleImputer to impute missing values ​​and StandardScaler to normalize the data. This improves the quality of the data and makes it suitable for analysis. The input is the acquired weather-related data, and the output is the preprocessed data.

[0335] Step 3:

[0336] The server uses the preprocessed data to analyze the weather forecast using a machine learning or deep learning model. It uses an LSTM model to perform short-term predictions and an ensemble learning model to perform long-term predictions. The input is the preprocessed data, and the output is the weather forecast results.

[0337] Step 4:

[0338] The server stores the weather forecast results in a database and updates it in real time. Every 30 minutes, it collects new data and re-runs the model to generate the latest forecast results. The input is the weather forecast results, and the output is the updated database.

[0339] Step 5:

[0340] The server uses an emotion analysis engine to analyze the user's emotional state. It determines the emotional state from the user's voice, facial expression, text input, etc. When analyzing text input using tools such as Emojify, the input is the text from the user and the output is the user's emotional state.

[0341] Step 6:

[0342] The server notifies users when certain weather conditions are predicted based on the emotion analysis results and weather forecast results. Specifically, if a thunderstorm is predicted and the user is feeling stressed, the server generates a personalized notification with detailed advice. The input is the weather forecast result and the user's emotional state, and the output is the notification content.

[0343] Step 7:

[0344] The server predicts demand based on weather data and optimizes inventory in physical stores. If rain is predicted, it will increase the stock of umbrellas and raincoats, and perform optimal inventory management. The input is the weather forecast result, and the output is inventory management instructions.

[0345] Step 8:

[0346] The server sends personalized promotion notifications according to the user's emotional state. For example, if the server determines that the user needs relaxation, it notifies the user of promotions for relaxation-related products. The input is the user's emotional state, and the output is the promotion notification.

[0347] Step 9:

[0348] Users can open the application on their devices and view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, etc. The input is launching the application, and the output is the displayed weather forecast information.

[0349] 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.

[0350] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search<url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0351] 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.

[0352] [Second embodiment]

[0353] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0354] 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.

[0355] 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).

[0356] 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.

[0357] 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.

[0358] 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).

[0359] 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. 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.

[0360] 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.

[0361] 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.

[0362] 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.

[0363] 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.

[0364] 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."

[0365] The following describes in detail an embodiment of the present invention. The present invention provides a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to complement missing values, and removes noise. Furthermore, the system uses the preprocessed data to analyze weather forecasts using a machine learning or deep learning model, and stores the analysis results in a database that is updated in real time. Based on user settings, the system notifies users when specific weather conditions are predicted and allows users to view detailed weather forecast information via their device.

[0366] Data collection

[0367] The server retrieves weather-related data from an external database, including information on temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. This data is collected periodically using an API.

[0368] Data Preprocessing

[0369] The server preprocesses the acquired weather-related data. This includes data normalization, missing value completion, and noise removal. For example, if there are gaps in the data obtained from a particular sensor, the server uses surrounding data to complete the gaps. It also applies a noise removal filter to improve the quality of the data.

[0370] Data analysis

[0371] Using the preprocessed data, the server analyzes the weather forecast using machine learning or deep learning models. For short-term predictions, it uses LSTM (recurrent neural networks), while for long-term predictions, it uses ensemble learning or physical models. This enables detailed predictions of temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, typhoon paths, and more.

[0372] Database storage and real-time updates

[0373] The analysis results are stored in a database by the server. This database is designed to be updated in real time and frequently re-runs the predictive model, ensuring that the latest forecast data is always available to users.

[0374] Notification function

[0375] The device will send notifications to users when certain weather conditions are predicted based on their settings, such as when a thunderstorm is predicted in a specific area based on their settings for "thunderstorm warnings."

[0376] Information reference

[0377] Users can open the application on their devices to view detailed weather forecast information, including the latest weather data, weather radar, typhoon tracks, pollen count information, and other details. For example, if a user wants to check the weather forecast for their current location and destination before commuting, they can simply open the application and the latest information will be displayed.

[0378] Specific examples

[0379] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database. After data preprocessing, it analyzes it using an LSTM model and predicts a thunderstorm in one hour. This prediction result is saved in the database and updated in real time. Next, the device sends a notification based on the "thunderstorm warning" set by the user, and the user opens the application to check detailed weather forecast information. As a result, the user can change their plans and take appropriate measures against the thunderstorm.

[0380] In this way, the weather forecast system of the present invention obtains weather-related data from an external database and provides comprehensive functions for preprocessing, analysis, storage, notification, and information reference, thereby providing users with the latest and most accurate weather forecast information in real time.

[0381] The processing flow will be explained below.

[0382] Step 1:

[0383] The server retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database through a process that periodically fetches the data via an API.

[0384] Step 2:

[0385] The weather-related data acquired by the server is preprocessed, which includes processes such as data normalization, missing value imputation, and noise removal. For example, missing data imputation uses an algorithm that calculates an estimate using surrounding data points.

[0386] Step 3:

[0387] The server then uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. It uses LSTM (long short-term memory) networks for short-term predictions and ensemble learning models for long-term predictions. This generates detailed forecasts such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon paths.

[0388] Step 4:

[0389] The server stores the analysis results in a database that is designed to be updated in real time, with the predictive model being re-run on new data at regular intervals (e.g., every 30 minutes).

[0390] Step 5:

[0391] The server generates a notification when certain weather conditions are predicted based on the user's settings. For example, a notification is generated when "thunderstorms are predicted in Tokyo within the next hour."

[0392] Step 6:

[0393] The device receives notifications sent from the server, which are customized based on the user's settings. For example, if the user requests thunderstorm notifications, the device immediately provides that information to the user.

[0394] Step 7:

[0395] Users can open the application on their devices to view detailed weather forecast information, including the latest weather forecast, weather radar, typhoon tracks, pollen count information, etc., which they can use to decide on appropriate actions.

[0396] Step 8:

[0397] The server periodically collects the latest data in real time, re-runs the prediction model to generate the latest prediction results, and stores them in the database again, so that the latest prediction information is always available to users.

[0398] In this way, each step works in cooperation with the other steps to provide users with highly accurate weather forecasts in real time.

[0399] Example 1

[0400] 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."

[0401] Conventional weather forecasting systems often lack accuracy and real-time response in the entire process, from collecting weather data to analyzing and notifying users. They also face difficulties in accessing the detailed weather information users require, and lack the ability to quickly notify users of specific weather conditions. In particular, inadequate handling of missing values ​​and noise reduces data quality and negatively impacts forecast accuracy. Furthermore, there is a need for greater system efficiency and smoother data updates.

[0402] 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.

[0403] In this invention, the server includes: means for acquiring weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; means for saving the analysis results in a database and updating them in real time; means for notifying the user when specific weather conditions are predicted based on user settings; means for the user to refer to detailed weather forecast information via a terminal; means for using an API to collect the data and preprocessing the collected data by filling in missing values ​​and applying a noise removal filter; means for using a recurrent neural network and an ensemble learning model to analyze the weather forecast; means for using a relational database management system to save and update the data in the database in real time; and means for using a cloud messaging service for the notification means. This enables efficient and highly accurate data collection, analysis, storage, notification, and reference.

[0404] An "external database" is a database that provides weather-related data and is accessible via the Internet or the like.

[0405] "API" is an abbreviation for Application Programming Interface, a means of exchanging data and functions between different software systems.

[0406] "Weather-related data" refers to all weather-related information, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0407] "Preprocessing" refers to processes to process the acquired data into a state suitable for analysis, such as normalizing the data, filling in missing values, and removing noise.

[0408] "Missing value imputation" is the process of filling in missing values ​​in a dataset with surrounding data, estimated values, etc.

[0409] "Noise removal" is a process for removing irregular fluctuations and outliers from data to improve data quality.

[0410] A "machine learning model" is an algorithm or method that learns patterns in data and makes predictions or classifications about future data.

[0411] A "deep learning model" is a highly accurate algorithm that uses a multi-layer neural network to extract data features and perform predictions and classifications.

[0412] A "recurrent neural network" is a special neural network for handling time series data, and is a model that retains past information and makes predictions.

[0413] "Ensemble learning" is a technique that combines multiple machine learning models to achieve higher predictive accuracy than individual models.

[0414] A "relational database management system" is a software system for storing, updating, searching, and managing data, and manages structured data in tabular format.

[0415] A "cloud messaging service" is a service for sending notification messages over the Internet, and is a means of providing real-time notifications to mobile devices and the like.

[0416] "User settings" refers to individual setting information such as notification conditions and reference conditions customized by the user on the system.

[0417] "Terminal" refers to an electronic device, such as a smartphone, tablet, or PC, that a user uses to access the system.

[0418] "Real-time updates" means that the data in the system is always kept up to date and updates are reflected almost immediately.

[0419] MODE FOR CARRYING OUT THE INVENTION

[0420] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention is a weather forecast system in which a server, a terminal, and a user are used in cooperation with each other.

[0421] Data collection

[0422] The server retrieves weather-related data from an external database. Specifically, it collects data via an API. The Python requests library is used to retrieve the data, and weather-related data includes temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. For example, the OpenWeatherMap API is used to retrieve this data.

[0423] Data Preprocessing

[0424] The server preprocesses the acquired weather-related data. This includes data normalization, missing value imputation, and noise removal. It uses Python's Pandas library to impute missing values ​​with surrounding data and applies SciPy's noise removal filter. This improves the quality of the data and prepares it for analysis.

[0425] Data analysis

[0426] The server uses preprocessed data to analyze weather forecasts using machine learning and deep learning models. It uses recurrent neural networks (LSTM) for short-term predictions and ensemble learning models for long-term predictions. It uses libraries such as TensorFlow and Keras to build predictive models and analyze temperature, precipitation probability, wind speed, wind direction, and more.

[0427] Database storage and real-time updates

[0428] The server stores the analysis results in a database and updates them in real time. The database uses a relational database management system such as PostgreSQL. The SQLAlchemy library is used to efficiently store and update data. This ensures that the latest prediction data is always available to users.

[0429] Notification function

[0430] The device sends notifications when certain weather conditions are predicted based on the user's settings. For example, if the user has set a "thunderstorm alert," it sends a push notification using a cloud messaging service. For this purpose, it uses Firebase Cloud Messaging (FCM).

[0431] Information reference

[0432] Users can access detailed weather forecast information on their devices. Through the application UI, they can view the latest weather data, weather radar, typhoon tracks, pollen count information, and more. For example, an application developed with React Native and Flutter allows users to easily check the weather forecast for their current location and destination before commuting to work.

[0433] Specific examples

[0434] For example, on a certain day in Tokyo, the server retrieves weather data using the OpenWeatherMap API. After preprocessing the data, an LSTM model is used to predict thunderstorms one hour in the future. This prediction result is stored in a PostgreSQL database and updated in real time. Next, the device sends a notification via FCM based on the "thunderstorm warning" set by the user. The user opens the application created with React Native and checks detailed weather forecast information. As a result, the user can change their plans and take appropriate measures against the thunderstorm.

[0435] Prompt Sentence Examples

[0436] "Get the latest weather data for Tokyo and predict the chance of thunderstorms in the next hour."

[0437] In this way, by linking the server, terminal, and user, we provide a system that efficiently and accurately accomplishes everything from data collection to analysis, storage, notification, and reference.

[0438] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0439] Step 1: Data collection

[0440] The server retrieves weather-related data from an external database. Specifically, it sends a request to the API and receives data in JSON format, including temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. The input is the endpoint URL and API key as an API request, and the output is weather-related data in JSON format. This data collection occurs periodically, and the next step is taken whenever new data is retrieved.

[0441] Step 2: Data Preprocessing

[0442] The server preprocesses the weather-related data in JSON format. First, it converts the data into a data frame using the Pandas library. Next, it imputes missing values ​​based on neighboring data points and applies a SciPy denoising filter to improve the data quality. The input is weather data in JSON format, and the output is a preprocessed data frame.

[0443] Step 3: Data analysis

[0444] The server uses the preprocessed data to analyze weather forecasts using machine learning and deep learning models. First, it builds LSTM models and ensemble learning models using TensorFlow and Keras. Next, it inputs the preprocessed data into these models to obtain forecast results such as temperature, precipitation probability, wind speed, and wind direction. The input is the preprocessed data frame, and the output is the analyzed weather forecast results.

[0445] Step 4: Save to database and update in real time

[0446] The server saves the analysis results in a database and updates them in real time. First, it connects to a PostgreSQL database using SQLAlchemy. Then it writes the analysis results to a database table and updates existing records if any. The input is the analyzed weather forecast results, and the output is the latest weather forecast data stored in the database.

[0447] Step 5: Notifications

[0448] The device sends a notification when certain weather conditions are predicted based on the user's settings. First, a notification message is created using Firebase Cloud Messaging (FCM). Then, the notification is sent at the appropriate time based on the conditions set by the user. The input is the user's settings and the analyzed weather forecast results, and the output is the notification message sent to the device.

[0449] Step 6: Information Reference

[0450] The user opens the application on their device to view detailed weather forecast information. The application UI displays the latest weather data, weather radar, typhoon tracks, and pollen count information. The input is the latest weather forecast data stored in the database, and the output is detailed weather forecast information displayed on the user's screen.

[0451] To give a concrete example, on a certain day in Tokyo, the server obtains weather data using the OpenWeatherMap API (Step 1). Next, the data is preprocessed to fill in missing values ​​and remove noise (Step 2). After that, the weather forecast is analyzed using an LSTM model (Step 3), and the analysis results are saved and updated in a PostgreSQL database (Step 4). The device sends a notification via FCM based on the "thunderstorm warning" set by the user (Step 5). Upon receiving this notification, the user opens the application and checks detailed weather forecast information (Step 6). This allows the user to change their plans and take appropriate measures against the thunderstorm.

[0452] (Application example 1)

[0453] 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."

[0454] Existing weather forecasting systems have technical limitations in the preprocessing and analysis of weather data, and are particularly lacking in providing suggestions that have a concrete impact on users' lives and business activities. Furthermore, there are no effective systems for demand forecasting or optimizing delivery plans for food delivery services. As a result, users are unable to adequately respond to the inconveniences and risks caused by weather conditions, and business efficiency is reduced.

[0455] 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.

[0456] In this invention, the server includes: means for acquiring weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; means for saving the analysis results in a database and updating them in real time; means for notifying the user when specific weather conditions are predicted based on the user's settings; means for the user to refer to detailed weather forecast information via a terminal; and means for making specific suggestions that will influence the user's behavior based on the weather forecast information. This makes it possible to propose specific impacts on daily life and business activities based on weather conditions. Furthermore, demand forecasting and optimization of delivery plans can be realized in food delivery services, improving user and business efficiency.

[0457] An "external database" is a system that exists on the Internet or in a cloud environment and accumulates and provides weather-related data.

[0458] "Weather-related data" refers to data that indicates weather conditions such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion status.

[0459] "Preprocessing" refers to the process of performing processes on the acquired weather-related data, such as normalizing the data, filling in missing values, and removing noise.

[0460] A "machine learning model" is an algorithm that learns patterns and rules based on data and makes predictions and classifications for new data.

[0461] A "deep learning model" is an algorithm that uses a multi-layer neural network to automatically learn the characteristics of complex data and perform advanced predictions and classifications.

[0462] "Real-time updates" means that information in a database or system is kept up-to-date immediately whenever new data is added.

[0463] A "terminal" is a device such as a smartphone, tablet, or PC that allows a user to access the Internet and obtain information.

[0464] The "particular weather conditions" are weather phenomena that the user has set in advance as being of interest or concern, such as thunderstorms, typhoons, and heavy snow.

[0465] "Notification" is a feature that sends alerts when certain weather conditions are predicted based on conditions set by the user in advance.

[0466] "Specific suggestions to influence user behavior" refers to providing specific advice to users on their daily lives and business activities based on weather conditions, such as encouraging them to change the timing of going out or ordering food delivery.

[0467] "Demand forecasting" refers to predicting the number of orders for food delivery services in advance based on weather conditions and past data.

[0468] "Optimal delivery planning" refers to optimizing delivery routes and times based on predicted demand in order to provide efficient food delivery services.

[0469] The present invention provides a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to fill in missing values, and removes noise. The system then uses the preprocessed data to analyze weather forecasts using machine learning or deep learning models, and stores the analysis results in a database that is updated in real time. Based on user settings, the system notifies users when specific weather conditions are predicted and allows them to access detailed weather forecast information via their device.

[0470] Data collection

[0471] The server retrieves weather-related data from external databases. This data includes information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. This data is collected periodically using APIs, such as the OpenWeatherMap API and the Japan Meteorological Agency API.

[0472] Data Preprocessing

[0473] The server preprocesses the acquired weather-related data. This preprocessing includes data normalization, missing value completion, and noise removal. The Python pandas library is used here. For example, if there are gaps in the data obtained from a particular sensor, the gaps are filled in using surrounding data, and a noise removal filter is applied to improve the quality of the data.

[0474] Data analysis

[0475] Using the preprocessed data, the server analyzes the weather forecast using machine learning or deep learning models. For short-term predictions, it uses a recurrent neural network (LSTM) model using TensorFlow or PyTorch, while for long-term predictions, it uses an ensemble learning model. This enables detailed predictions such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon paths.

[0476] Database storage and real-time updates

[0477] The analysis results are stored in a database by the server. This database is designed to be updated in real time using Firebase Realtime Database or Amazon RDS, and the predictive model is re-executed frequently, ensuring that the latest forecast data is always provided to users.

[0478] Notification function

[0479] The device will send notifications when certain weather conditions are predicted based on the user's settings, such as "thunderstorm warnings" and "pollen alerts," using Firebase Cloud Messaging (FMC).

[0480] Information references and specific suggestions

[0481] Users can open the application on their devices to view detailed weather forecast information, including the latest weather data, weather radar, typhoon tracks, pollen count information, and other details. Based on the weather forecast information, the application will also provide specific suggestions to influence users' behavior, such as recommending when to go out or ordering food delivery.

[0482] Specific examples

[0483] For example, on a given day in Tokyo, the server retrieves the latest weather data from the weather database. After data preprocessing, it runs an LSTM model to predict a thunderstorm in one hour. This prediction result is saved in the database and updated in real time. The device then sends a notification based on the user's "thunderstorm warning" setting, and the user opens an application to check detailed weather forecast information. As a result, the user can change their plans, order food delivery, and take appropriate precautions against the thunderstorm.

[0484] Examples of specific prompts include the following:

[0485] "Predict the increase in delivery requests in Tokyo next week based on the amount of precipitation and average wind speed."

[0486] In this way, the weather forecast system of the present invention acquires weather-related data from an external database and provides comprehensive functions for preprocessing, analysis, storage, notification, information reference, and specific suggestions, thereby providing users with the latest and most accurate weather forecast information and action suggestions in real time.

[0487] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0488] Step 1:

[0489] Data collection

[0490] The server retrieves weather-related data from an external database. Specifically, it periodically requests information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count using the OpenWeatherMap API and the Japan Meteorological Agency API, and receives JSON-formatted data in response. The input data is raw weather data retrieved from the API, and the output data is this raw weather data.

[0491] Step 2:

[0492] Data Preprocessing

[0493] The server performs data preprocessing using the acquired weather-related data. Specifically, it normalizes the data using Python's pandas library, imputes missing values ​​using SimpleImputer, and removes noise using a noise removal filter. The input data is raw weather data, and the output data is preprocessed weather data that has been normalized, imputes missing values, and removed noise.

[0494] Step 3:

[0495] Data analysis

[0496] Using the preprocessed data, the server analyzes the weather forecast using a machine learning or deep learning model. The preprocessed data is input into an LSTM model or ensemble learning model built using TensorFlow or PyTorch, and weather forecasts such as future temperature, precipitation probability, wind speed, wind direction, thunderstorm predictions, and typhoon paths are output. The input data is preprocessed weather data, and the output data is analyzed weather forecast data.

[0497] Step 4:

[0498] Database storage and real-time updates

[0499] The server stores the analysis results in Firebase Realtime Database or Amazon RDS and updates them in real time. Specifically, the analyzed weather forecast data is written to these databases and set up to be automatically updated each time new data is added. The input data is the analyzed weather forecast data, and the output data is the latest weather forecast data stored in the database.

[0500] Step 5:

[0501] Notification function

[0502] The device sends notifications when certain weather conditions are predicted based on the user's settings. Specifically, it uses Firebase Cloud Messaging (FCM) to instantly send notifications about predicted weather conditions based on the user's settings for "thunderstorm warnings" and "pollen alerts." The input data is the analyzed weather forecast data retrieved from the database and the user's settings, and the output data is the notification message sent over the Internet.

[0503] Step 6:

[0504] Information references and specific suggestions

[0505] Users can access the application through their devices to view detailed weather forecast information. Based on the weather forecast information, the application also provides specific suggestions, such as when to go out or encouraging people to order food delivery. The input data is the weather forecast information displayed on the device and data related to the user's suggested actions, and the output data is the specific suggested actions provided to the user.

[0506] Step 7:

[0507] Food delivery demand forecasting and delivery plan optimization

[0508] The server predicts demand for food delivery services based on weather conditions and past data, and then formulates optimal delivery plans based on the predicted demand. This involves using a machine learning model to predict demand and then running an algorithm to optimize delivery routes and times based on the prediction results. The input data is preprocessed weather data and past delivery records, and the output data is the demand forecast and optimal delivery plan information.

[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 following describes in detail an embodiment of the present invention. The present invention combines a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to fill in missing values, and removes noise, with an emotion engine that recognizes user emotions. The system updates the weather forecast in real time based on the analysis results and notifies the user when specific weather conditions are predicted based on the user's settings. The system also allows the user to view detailed weather forecast information via a terminal.

[0511] Data collection

[0512] The server periodically retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database. This data is fetched via an API.

[0513] Data Preprocessing

[0514] The server preprocesses the collected weather-related data, including normalizing the data, filling in missing values, and removing noise. Specifically, it fills in missing data with surrounding data points and applies a noise removal filter to improve data quality.

[0515] Data analysis

[0516] The server uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. LSTM (long short-term memory) networks are used for short-term predictions, while ensemble learning models are used for long-term predictions. This results in a detailed weather forecast (temperature, probability of precipitation, wind speed, wind direction, thunderstorm forecast, and typhoon path).

[0517] Database storage and real-time updates

[0518] The analysis results are stored in a database by the server and updated in real time. New data is collected at regular intervals (e.g., every 30 minutes), and the predictive model is re-run to generate the latest forecast results.

[0519] Emotion analysis using an emotion engine

[0520] The server uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state from their voice, facial expressions, text input, etc. The analyzed emotional state is stored in a database and reflected in the content and format of the weather forecast notification.

[0521] User Notification Settings

[0522] The device will send notifications that take into account the user's emotional state based on specific weather conditions (thunderstorms, typhoons, pollen counts, etc.) set by the user. For example, if the user has set up a thunderstorm notification, the device will send a notification with more detailed information and follow-up advice depending on the user's stress level detected by the emotion engine.

[0523] Information reference

[0524] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, and other details. The priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0525] Specific examples

[0526] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database, and after preprocessing, analyzes it using an LSTM model, predicting a thunderstorm in one hour. The analysis results are stored in the database and updated in real time. The server then analyzes the user's stress level using an emotion engine and creates a notification based on that state. The device receives this notification and sends it to the user with detailed advice such as, "Thunderstorms are predicted. Avoid sudden outings or bring an umbrella." The user can then change their plans based on this notification and take appropriate precautions against the thunderstorm.

[0527] In this way, the system of the present invention acquires weather-related data from an external database, and in addition to preprocessing, analyzing, storing, notifying, and referencing the information, it uses an emotion engine to provide personalized weather information based on the user's emotional state.

[0528] The processing flow will be explained below.

[0529] Step 1:

[0530] The server retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database by fetching the data periodically using an API.

[0531] Step 2:

[0532] The weather-related data acquired by the server is preprocessed. Specifically, data normalization, missing value completion, and noise removal are performed. For example, if there are missing parts in the acquired data, the parts are completed using surrounding data points.

[0533] Also, a noise reduction filter is applied to improve the quality of the data.

[0534] Step 3:

[0535] The server then uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. It uses LSTM (long short-term memory) networks for short-term predictions and ensemble learning models for long-term predictions. This generates detailed forecasts such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon tracks.

[0536] Step 4:

[0537] The server stores the analysis results in a database that is designed to be updated in real time, collecting new data at regular intervals (e.g., every 30 minutes) and re-running the predictive model to generate the latest forecasts.

[0538] Step 5:

[0539] The server uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state from their voice, facial expressions, text input, etc. The analyzed emotional state is stored in a database.

[0540] Step 6:

[0541] The server adjusts the content and format of notifications based on user settings when certain weather conditions (e.g., thunderstorms or typhoons) are predicted. For example, if the emotion engine determines that the user is feeling stressed, it creates a notification with follow-up advice appropriate to that state.

[0542] Step 7:

[0543] The device receives notifications sent from the server. These notifications are customized based on the user's settings and emotional state. For example, if a user has set up "thunderstorm notifications" and a stressful state is detected, the device receives a notification with specific advice such as "A thunderstorm is predicted within 30 minutes. Avoid sudden outings and bring an umbrella if necessary."

[0544] Step 8:

[0545] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, and pollen count information. For example, before commuting, users can check the weather forecast for their current location and destination to determine the best course of action.

[0546] Step 9:

[0547] The server periodically collects the latest data in real time, re-runs the prediction model to generate the latest prediction results, and stores them in the database again, thereby always providing the latest prediction information to users.

[0548] In this way, each step works in conjunction with the other steps to provide a personalized, real-time, and highly accurate weather forecast based on the user's emotional state.

[0549] Example 2

[0550] 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."

[0551] The appropriate collection and prediction of weather-related data is important for daily life and industrial activities. However, improving the accuracy of weather forecasts and providing information tailored to individual users' situations presents many challenges. In particular, existing systems are unable to adequately address issues such as missing or noisy data, real-time updates, and generating notifications that take into account the user's emotional state.

[0552] The identification processing 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 acquiring weather-related data from an external database, means for preprocessing the weather-related data to complement missing values ​​and remove noise, means for analyzing a weather forecast using a machine learning model or a deep learning model using the preprocessed data, means for saving the analysis results in a database and updating them in real time, means for recognizing and analyzing a user's emotional state, means for providing notifications based on specific weather conditions and the user's emotional state based on user settings, and means for the user to refer to detailed weather forecast information via a terminal. This allows a series of processes, such as collection, preprocessing, analysis, storage, and notification of weather-related data, to be efficiently executed, making it possible to provide personalized weather forecast information that takes the user's emotional state into consideration.

[0553] An "external database" is a collection of data that resides outside the system and stores and provides weather-related data.

[0554] "Weather-related data" refers to data relating to weather and meteorological conditions, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0555] "Preprocessing" refers to a series of processes performed on acquired weather-related data, such as data normalization, missing value completion, and noise removal.

[0556] "Missing value imputation" is the process of filling in missing data points in a dataset using surrounding data.

[0557] "Noise removal" refers to the process of filtering out abnormal values ​​and unnecessary data from data.

[0558] A "machine learning model" is an algorithm for predicting future data based on past data, and primarily uses statistical methods.

[0559] A "deep learning model" is an algorithm that uses neural networks to analyze data in a multi-layered manner and make advanced predictions.

[0560] "Real-time updating" refers to the process of collecting data at regular intervals and periodically replacing the analysis results with the latest ones.

[0561] "Emotional state" refers to the psychological and emotional state of the user as analyzed from their voice, facial expressions, text input, etc.

[0562] "Notifications" are messages or alerts that provide information about specific weather conditions based on the user's preferences and emotional state.

[0563] A "terminal" is a device (e.g., smartphone, tablet, computer) that a user uses to access and operate the system.

[0564] "Detailed weather forecast information" is weather forecast data that goes beyond simple weather forecasts and includes additional information such as weather radar, typhoon tracks, and pollen dispersion information.

[0565] The present invention is a weather forecasting system that retrieves weather-related data from an external database, preprocesses it to fill in missing values, and removes noise. The system also incorporates an emotion engine that recognizes the user's emotions. The system updates the weather forecast in real time based on the analysis results and notifies the user when specific weather conditions are predicted based on the user's settings. The system also allows the user to view detailed weather forecast information via their device.

[0566] The server periodically retrieves weather-related data (e.g., temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database. This data is fetched via an API. For example, the data could be retrieved using the OpenWeatherMap API.

[0567] The server preprocesses the collected weather-related data. Specific preprocessing steps include normalizing the data, filling in missing values, and removing noise. For example, the Pandas library is used to fill in NaN values ​​with approximate values, and the Scipy library is used to apply a low-pass filter to smooth the data.

[0568] The server uses the preprocessed data to analyze weather forecasts using machine learning or deep learning models. It uses a long short-term memory (LSTM) network for short-term predictions and an ensemble learning model for long-term predictions. Specific implementations could use frameworks such as TensorFlow or PyTorch.

[0569] The analysis results are stored in a database by the server and updated in real time. New data is collected at regular intervals (for example, every 30 minutes), and the predictive model is re-run to generate the latest prediction results. The database can be a relational database such as MySQL or PostgreSQL.

[0570] The server uses an emotion engine to recognize the user's emotions. This engine analyzes the user's emotional state based on voice, facial expressions, and text input. For example, facial expressions can be captured using OpenCV and analyzed using a deep learning model, or voice data can be converted into text using a speech recognition API such as IBM Watson, and emotion analysis can then be performed.

[0571] The device sends notifications that take into account the user's emotional state based on specific weather conditions set by the user (e.g., thunderstorms, typhoons, pollen counts, etc.). A real-time notification function is implemented using JavaScript, etc., and notifications are sent with wording that reflects the user's emotional state. For example, if the user is feeling stressed, a detailed and kind notification will be sent.

[0572] Users can access detailed weather forecast information by opening the app on their devices. The app displays the latest forecast data, weather radar, typhoon tracks, pollen count information, etc. Furthermore, the priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0573] Specific examples

[0574] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database, and after preprocessing, analyzes it using an LSTM model, predicting a thunderstorm in one hour. The analysis results are stored in the database and updated in real time. The server then analyzes the user's stress level using an emotion engine and creates a notification based on that state. The device receives this notification and sends it to the user with detailed advice such as, "Thunderstorms are predicted. Avoid sudden outings or bring an umbrella." The user can then change their plans based on this notification and take appropriate precautions against the thunderstorm.

[0575] Examples of prompt statements

[0576] "Describe a scenario where a server retrieves the latest weather data from an external database and predicts a thunderstorm in the next hour in Tokyo on a certain day. The emotion engine detects the user's stress state and sends a notification with detailed advice."

[0577] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0578] Step 1: Data collection

[0579] The server periodically retrieves weather-related data from an external database. Specifically, the server sends an API request and receives a JSON-formatted response from the external database. This input data includes temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. As output, this weather-related data is obtained and passed to the next preprocessing step.

[0580] Example of how it works:

[0581] The server sends an API request every hour.

[0582] Receive the API response (weather data JSON) and save it in local storage.

[0583] Step 2: Data Preprocessing

[0584] The server preprocesses the collected weather-related data. This preprocessing involves normalizing the data, filling in missing values, and removing noise. The input is the raw data collected in step 1, and the output is preprocessed, high-quality data.

[0585] Example of how it works:

[0586] Use the Pandas library to read the JSON data and convert it to a dataframe.

[0587] To impute missing values, fill NaN values ​​in the data frame with the average of the preceding and following data points.

[0588] Apply a low-pass filter using the Scipy library to remove noise.

[0589] Step 3: Data analysis

[0590] The server uses the preprocessed data to analyze the weather forecast with a machine learning or deep learning model, where the input is the preprocessed data from step 2 and the output is the weather forecast data.

[0591] Example of how it works:

[0592] Use TensorFlow or PyTorch to load the LSTM model and run predictions using the preprocessed data as input.

[0593] Short-term predictions (1 to 24 hours) are performed using LSTM, and long-term predictions (1 week) are performed using an ensemble learning model.

[0594] Evaluate the analysis results and record performance metrics.

[0595] Step 4: Save to database and update in real time

[0596] The analysis results are stored in a database by the server and updated at regular intervals. The input is the analysis result from step 3, and the output is the prediction results stored in the database.

[0597] Example of how it works:

[0598] Connect to a MySQL or PostgreSQL database and insert the analysis results into a table.

[0599] Every 30 minutes, start with a new data collection and rerun the preprocessing and analysis.

[0600] Delete the old prediction results in the database and update with the new results.

[0601] Step 5: Emotion analysis using the emotion engine

[0602] The server uses an emotion engine to recognize the user's emotions. The input is the user's voice, facial expression, text input, etc., and the output is the user's emotional state.

[0603] Example of how it works:

[0604] Facial expressions are captured from a webcam using OpenCV and analyzed using a deep learning model.

[0605] Voice data is converted into text through a voice recognition API and sentiment analysis is performed.

[0606] Text analysis algorithms are used to recognize emotions from messages and comments.

[0607] Step 6: User Notification Settings

[0608] The device sends notifications based on specific weather conditions set by the user, taking into account their emotional state. The input is the data from steps 4 and 5, and the output is the notification sent to the user.

[0609] Example of how it works:

[0610] Check the user's settings and generate notifications when certain weather conditions (such as thunderstorms, typhoons, and pollen counts) are predicted.

[0611] Send real-time notifications to your device using JavaScript.

[0612] If the user is feeling stressed, a detailed notification containing kind words is generated.

[0613] Step 7: Information Reference

[0614] The user opens the application on the device and views detailed weather forecast information. The input is the forecast result stored in the database in step 4, and the output is the weather forecast information displayed on the device.

[0615] Example of how it works:

[0616] Open a mobile or web application.

[0617] The app screen displays the latest weather forecast, weather radar, typhoon path, pollen information, and more.

[0618] The information display format is automatically adjusted according to the user's emotional state.

[0619] (Application example 2)

[0620] 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."

[0621] In today's world, highly accurate weather forecasts have become important information in many fields. However, simply providing weather information makes it difficult to present users with measures appropriate to their individual needs and situations. In particular, physical stores require inventory management that takes weather conditions into account and appropriate product recommendations to customers. Furthermore, services that do not take into account user emotions make it difficult to improve customer satisfaction.

[0622] Therefore, the objective of the present invention is to provide a system that combines weather data and a user's emotional state to optimize inventory management and promotion notifications in physical stores and meet individual needs.

[0623] 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.

[0624] In this invention, the server includes means for acquiring weather-related data from an external database, means for preprocessing the weather-related data to fill in missing values ​​and remove noise, means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data, means for storing the analysis results in a database and updating them in real time, means for notifying the user when specific weather conditions are predicted based on user settings, means for the user to refer to detailed weather forecast information via a terminal, means for performing demand forecasts based on the weather data and optimizing inventory in physical stores, and means for sending personalized promotional notifications according to the user's emotional state. This enables optimal inventory management in response to weather conditions and individual promotional notifications that take the user's emotional state into consideration.

[0625] "Weather-related data" refers to data relating to weather and meteorological conditions, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0626] "Preprocessing" refers to the process of preparing data in an appropriate form before data analysis, such as normalizing data, filling in missing values, and removing noise.

[0627] A "machine learning model" is a model that learns patterns and regularities based on data and makes predictions and classifications.

[0628] A "deep learning model" is a model that uses a multi-layer neural network that can learn complex data patterns.

[0629] "Weather forecast" is information that predicts future weather conditions based on acquired weather data.

[0630] "Real-time updates" means obtaining new data and updating the analysis results based on that data at any time to provide the latest information.

[0631] The "notification means" is a means for notifying the user of information when a specific condition is met.

[0632] A "brick and mortar store" is a retail or service establishment that exists in a physical location.

[0633] "Inventory management" refers to the proper management and adjustment of the quantity of goods and resources, and maintaining an optimal inventory level according to demand.

[0634] A "promotional notice" is a notice to inform users of information about a particular product or service as part of an advertisement or promotion.

[0635] "Emotional state" refers to the user's emotional state, such as stress, joy, or sadness.

[0636] "Personalized" means providing information and services tailored to the specific needs and circumstances of each individual user.

[0637] This system acquires weather-related data from an external database, performs missing value imputation and noise removal, and then generates highly accurate weather forecasts using machine learning and deep learning models. It also analyzes the user's emotional state and provides personalized notifications and information based on that.

[0638] Data collection and preprocessing

[0639] The server periodically retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database (e.g., a weather API). This data is fetched via the API. The server then preprocesses the collected weather data, imputes missing values, and removes noise. Specifically, it uses Scikit-learn's SimpleImputer to impute missing data and StandardScaler to normalize the data.

[0640] Weather forecast analysis

[0641] Using the preprocessed weather data, the server analyzes the weather forecast using machine learning models (e.g., LSTM models) or deep learning models. LSTM (long short-term memory) networks are used for short-term predictions, while ensemble learning models are used for long-term predictions. This results in a detailed weather forecast (temperature, probability of precipitation, wind speed, wind direction, thunderstorm forecast, and typhoon path). The analysis results are stored in a database and updated in real time.

[0642] Sentiment Analysis and Notifications

[0643] The server implements an emotion analysis engine to analyze the user's emotional state. This engine analyzes the user's voice, facial expressions, text input, etc. to determine the user's emotional state. For example, it analyzes text input using an emotion recognition library such as Emojify.

[0644] Based on the results of sentiment analysis, notifications are sent when certain weather conditions are predicted. For example, if a thunderstorm is predicted and the user is feeling stressed, a personalized notification with detailed advice is sent. The notification may include specific advice such as, "Thunderstorms are predicted. Avoid sudden outings and, if necessary, check out stress-relieving products."

[0645] Inventory management and promotion notifications

[0646] The server uses weather data to predict demand and optimize inventory in physical stores. For example, if rain is predicted, it increases the stock of umbrellas and raincoats. It also sends personalized promotional notifications based on the user's emotional state. For example, if it determines that the user needs relaxation, it notifies them of promotions for relaxation-related products.

[0647] Information reference

[0648] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, and other details. The priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0649] Specific examples

[0650] For example, if a thunderstorm is predicted for Tokyo on a certain day and the user is feeling stressed, the following notification will be displayed: "Thunderstorms are predicted. Avoid sudden outings and, if necessary, check for products to relieve stress."

[0651] Prompt Sentence Examples

[0652] "Based on data obtained from a weather data API, impute and normalize missing values ​​and use an LSTM model to predict the weather. Next, analyze the user's emotional state from input and generate a promotional notification for relaxation-related products if the user is feeling stressed when a thunderstorm is predicted."

[0653] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0654] Step 1:

[0655] The server retrieves weather-related data from an external database through an API, which contains information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. The input is the API endpoint, and the output is the weather-related data.

[0656] Step 2:

[0657] The server preprocesses the acquired weather-related data. Specifically, it uses SimpleImputer to impute missing values ​​and StandardScaler to normalize the data. This improves the quality of the data and makes it suitable for analysis. The input is the acquired weather-related data, and the output is the preprocessed data.

[0658] Step 3:

[0659] The server uses the preprocessed data to analyze the weather forecast using a machine learning or deep learning model. It uses an LSTM model to perform short-term predictions and an ensemble learning model to perform long-term predictions. The input is the preprocessed data, and the output is the weather forecast results.

[0660] Step 4:

[0661] The server stores the weather forecast results in a database and updates it in real time. Every 30 minutes, it collects new data and re-runs the model to generate the latest forecast results. The input is the weather forecast results, and the output is the updated database.

[0662] Step 5:

[0663] The server uses an emotion analysis engine to analyze the user's emotional state. It determines the emotional state from the user's voice, facial expression, text input, etc. When analyzing text input using tools such as Emojify, the input is the text from the user and the output is the user's emotional state.

[0664] Step 6:

[0665] The server notifies users when certain weather conditions are predicted based on the emotion analysis results and weather forecast results. Specifically, if a thunderstorm is predicted and the user is feeling stressed, the server generates a personalized notification with detailed advice. The input is the weather forecast result and the user's emotional state, and the output is the notification content.

[0666] Step 7:

[0667] The server predicts demand based on weather data and optimizes inventory in physical stores. If rain is predicted, it will increase the stock of umbrellas and raincoats, and perform optimal inventory management. The input is the weather forecast result, and the output is inventory management instructions.

[0668] Step 8:

[0669] The server sends personalized promotion notifications according to the user's emotional state. For example, if the server determines that the user needs relaxation, it notifies the user of promotions for relaxation-related products. The input is the user's emotional state, and the output is the promotion notification.

[0670] Step 9:

[0671] Users can open the application on their devices and view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, etc. The input is launching the application, and the output is the displayed weather forecast information.

[0672] 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.

[0673] 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.

[0674] 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.

[0675] [Third embodiment]

[0676] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0677] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0678] 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).

[0679] 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.

[0680] 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.

[0681] 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).

[0682] 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. 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.

[0683] 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.

[0684] 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.

[0685] 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.

[0686] 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.

[0687] 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."

[0688] The following describes in detail an embodiment of the present invention. The present invention provides a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to complement missing values, and removes noise. Furthermore, the system uses the preprocessed data to analyze weather forecasts using a machine learning or deep learning model, and stores the analysis results in a database that is updated in real time. Based on user settings, the system notifies users when specific weather conditions are predicted and allows users to view detailed weather forecast information via their device.

[0689] Data collection

[0690] The server retrieves weather-related data from an external database, including information on temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. This data is collected periodically using an API.

[0691] Data Preprocessing

[0692] The server preprocesses the acquired weather-related data. This includes data normalization, missing value completion, and noise removal. For example, if there are gaps in the data obtained from a particular sensor, the server uses surrounding data to complete the gaps. It also applies a noise removal filter to improve the quality of the data.

[0693] Data analysis

[0694] Using the preprocessed data, the server analyzes the weather forecast using machine learning or deep learning models. For short-term predictions, it uses LSTM (recurrent neural networks), while for long-term predictions, it uses ensemble learning or physical models. This enables detailed predictions of temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, typhoon paths, and more.

[0695] Database storage and real-time updates

[0696] The analysis results are stored in a database by the server. This database is designed to be updated in real time and frequently re-runs the predictive model, ensuring that the latest forecast data is always available to users.

[0697] Notification function

[0698] The device will send notifications to users when certain weather conditions are predicted based on their settings, such as when a thunderstorm is predicted in a specific area based on their settings for "thunderstorm warnings."

[0699] Information reference

[0700] Users can open the application on their devices to view detailed weather forecast information, including the latest weather data, weather radar, typhoon tracks, pollen count information, and other details. For example, if a user wants to check the weather forecast for their current location and destination before commuting, they can simply open the application and the latest information will be displayed.

[0701] Specific examples

[0702] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database. After data preprocessing, it analyzes it using an LSTM model and predicts a thunderstorm in one hour. This prediction result is saved in the database and updated in real time. Next, the device sends a notification based on the "thunderstorm warning" set by the user, and the user opens the application to check detailed weather forecast information. As a result, the user can change their plans and take appropriate measures against the thunderstorm.

[0703] In this way, the weather forecast system of the present invention obtains weather-related data from an external database and provides comprehensive functions for preprocessing, analysis, storage, notification, and information reference, thereby providing users with the latest and most accurate weather forecast information in real time.

[0704] The processing flow will be explained below.

[0705] Step 1:

[0706] The server retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database through a process that periodically fetches the data via an API.

[0707] Step 2:

[0708] The weather-related data acquired by the server is preprocessed, which includes processes such as data normalization, missing value imputation, and noise removal. For example, missing data imputation uses an algorithm that calculates an estimate using surrounding data points.

[0709] Step 3:

[0710] The server then uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. It uses LSTM (long short-term memory) networks for short-term predictions and ensemble learning models for long-term predictions. This generates detailed forecasts such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon paths.

[0711] Step 4:

[0712] The server stores the analysis results in a database that is designed to be updated in real time, with the predictive model being re-run on new data at regular intervals (e.g., every 30 minutes).

[0713] Step 5:

[0714] The server generates a notification when certain weather conditions are predicted based on the user's settings. For example, a notification is generated when "thunderstorms are predicted in Tokyo within the next hour."

[0715] Step 6:

[0716] The device receives notifications sent from the server, which are customized based on the user's settings. For example, if the user requests thunderstorm notifications, the device immediately provides that information to the user.

[0717] Step 7:

[0718] Users can open the application on their devices to view detailed weather forecast information, including the latest weather forecast, weather radar, typhoon tracks, pollen count information, etc., which they can use to decide on appropriate actions.

[0719] Step 8:

[0720] The server periodically collects the latest data in real time, re-runs the prediction model to generate the latest prediction results, and stores them in the database again, so that the latest prediction information is always available to users.

[0721] In this way, each step works in cooperation with the other steps to provide users with highly accurate weather forecasts in real time.

[0722] Example 1

[0723] 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."

[0724] Conventional weather forecasting systems often lack accuracy and real-time response in the entire process, from collecting weather data to analyzing and notifying users. They also face difficulties in accessing the detailed weather information users require, and lack the ability to quickly notify users of specific weather conditions. In particular, inadequate handling of missing values ​​and noise reduces data quality and negatively impacts forecast accuracy. Furthermore, there is a need for greater system efficiency and smoother data updates.

[0725] 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.

[0726] In this invention, the server includes: means for acquiring weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; means for saving the analysis results in a database and updating them in real time; means for notifying the user when specific weather conditions are predicted based on user settings; means for the user to refer to detailed weather forecast information via a terminal; means for using an API to collect the data and preprocessing the collected data by filling in missing values ​​and applying a noise removal filter; means for using a recurrent neural network and an ensemble learning model to analyze the weather forecast; means for using a relational database management system to save and update the data in the database in real time; and means for using a cloud messaging service for the notification means. This enables efficient and highly accurate data collection, analysis, storage, notification, and reference.

[0727] An "external database" is a database that provides weather-related data and is accessible via the Internet or the like.

[0728] "API" is an abbreviation for Application Programming Interface, a means of exchanging data and functions between different software systems.

[0729] "Weather-related data" refers to all weather-related information, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0730] "Preprocessing" refers to processes to process the acquired data into a state suitable for analysis, such as normalizing the data, filling in missing values, and removing noise.

[0731] "Missing value imputation" is the process of filling in missing values ​​in a dataset with surrounding data, estimated values, etc.

[0732] "Noise removal" is a process for removing irregular fluctuations and outliers from data to improve data quality.

[0733] A "machine learning model" is an algorithm or method that learns patterns in data and makes predictions or classifications about future data.

[0734] A "deep learning model" is a highly accurate algorithm that uses a multi-layer neural network to extract data features and perform predictions and classifications.

[0735] A "recurrent neural network" is a special neural network for handling time series data, and is a model that retains past information and makes predictions.

[0736] "Ensemble learning" is a technique that combines multiple machine learning models to achieve higher predictive accuracy than individual models.

[0737] A "relational database management system" is a software system for storing, updating, searching, and managing data, and manages structured data in tabular format.

[0738] A "cloud messaging service" is a service for sending notification messages over the Internet, and is a means of providing real-time notifications to mobile devices and the like.

[0739] "User settings" refers to individual setting information such as notification conditions and reference conditions customized by the user on the system.

[0740] "Terminal" refers to an electronic device, such as a smartphone, tablet, or PC, that a user uses to access the system.

[0741] "Real-time updates" means that the data in the system is always kept up to date and updates are reflected almost immediately.

[0742] MODE FOR CARRYING OUT THE INVENTION

[0743] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention is a weather forecast system in which a server, a terminal, and a user are used in cooperation with each other.

[0744] Data collection

[0745] The server retrieves weather-related data from an external database. Specifically, it collects data via an API. The Python requests library is used to retrieve the data, and weather-related data includes temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. For example, the OpenWeatherMap API is used to retrieve this data.

[0746] Data Preprocessing

[0747] The server preprocesses the acquired weather-related data. This includes data normalization, missing value imputation, and noise removal. It uses Python's Pandas library to impute missing values ​​with surrounding data and applies SciPy's noise removal filter. This improves the quality of the data and prepares it for analysis.

[0748] Data analysis

[0749] The server uses preprocessed data to analyze weather forecasts using machine learning and deep learning models. It uses recurrent neural networks (LSTM) for short-term predictions and ensemble learning models for long-term predictions. It uses libraries such as TensorFlow and Keras to build predictive models and analyze temperature, precipitation probability, wind speed, wind direction, and more.

[0750] Database storage and real-time updates

[0751] The server stores the analysis results in a database and updates them in real time. The database uses a relational database management system such as PostgreSQL. The SQLAlchemy library is used to efficiently store and update data. This ensures that the latest prediction data is always available to users.

[0752] Notification function

[0753] The device sends notifications when certain weather conditions are predicted based on the user's settings. For example, if the user has set a "thunderstorm alert," it sends a push notification using a cloud messaging service. For this purpose, it uses Firebase Cloud Messaging (FCM).

[0754] Information reference

[0755] Users can access detailed weather forecast information on their devices. Through the application UI, they can view the latest weather data, weather radar, typhoon tracks, pollen count information, and more. For example, an application developed with React Native and Flutter allows users to easily check the weather forecast for their current location and destination before commuting to work.

[0756] Specific examples

[0757] For example, on a certain day in Tokyo, the server retrieves weather data using the OpenWeatherMap API. After preprocessing the data, an LSTM model is used to predict thunderstorms one hour in the future. This prediction result is stored in a PostgreSQL database and updated in real time. Next, the device sends a notification via FCM based on the "thunderstorm warning" set by the user. The user opens the application created with React Native and checks detailed weather forecast information. As a result, the user can change their plans and take appropriate measures against the thunderstorm.

[0758] Prompt Sentence Examples

[0759] "Get the latest weather data for Tokyo and predict the chance of thunderstorms in the next hour."

[0760] In this way, by linking the server, terminal, and user, we provide a system that efficiently and accurately accomplishes everything from data collection to analysis, storage, notification, and reference.

[0761] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0762] Step 1: Data collection

[0763] The server retrieves weather-related data from an external database. Specifically, it sends a request to the API and receives data in JSON format, including temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. The input is the endpoint URL and API key as an API request, and the output is weather-related data in JSON format. This data collection occurs periodically, and the next step is taken whenever new data is retrieved.

[0764] Step 2: Data Preprocessing

[0765] The server preprocesses the weather-related data in JSON format. First, it converts the data into a data frame using the Pandas library. Next, it imputes missing values ​​based on neighboring data points and applies a SciPy denoising filter to improve the data quality. The input is weather data in JSON format, and the output is a preprocessed data frame.

[0766] Step 3: Data analysis

[0767] The server uses the preprocessed data to analyze weather forecasts using machine learning and deep learning models. First, it builds LSTM models and ensemble learning models using TensorFlow and Keras. Next, it inputs the preprocessed data into these models to obtain forecast results such as temperature, precipitation probability, wind speed, and wind direction. The input is the preprocessed data frame, and the output is the analyzed weather forecast results.

[0768] Step 4: Save to database and update in real time

[0769] The server saves the analysis results in a database and updates them in real time. First, it connects to a PostgreSQL database using SQLAlchemy. Then it writes the analysis results to a database table and updates existing records if any. The input is the analyzed weather forecast results, and the output is the latest weather forecast data stored in the database.

[0770] Step 5: Notifications

[0771] The device sends a notification when certain weather conditions are predicted based on the user's settings. First, a notification message is created using Firebase Cloud Messaging (FCM). Then, the notification is sent at the appropriate time based on the conditions set by the user. The input is the user's settings and the analyzed weather forecast results, and the output is the notification message sent to the device.

[0772] Step 6: Information Reference

[0773] The user opens the application on their device to view detailed weather forecast information. The application UI displays the latest weather data, weather radar, typhoon tracks, and pollen count information. The input is the latest weather forecast data stored in the database, and the output is detailed weather forecast information displayed on the user's screen.

[0774] To give a concrete example, on a certain day in Tokyo, the server obtains weather data using the OpenWeatherMap API (Step 1). Next, the data is preprocessed to fill in missing values ​​and remove noise (Step 2). After that, the weather forecast is analyzed using an LSTM model (Step 3), and the analysis results are saved and updated in a PostgreSQL database (Step 4). The device sends a notification via FCM based on the "thunderstorm warning" set by the user (Step 5). Upon receiving this notification, the user opens the application and checks detailed weather forecast information (Step 6). This allows the user to change their plans and take appropriate measures against the thunderstorm.

[0775] (Application example 1)

[0776] 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."

[0777] Existing weather forecasting systems have technical limitations in the preprocessing and analysis of weather data, and are particularly lacking in providing suggestions that have a concrete impact on users' lives and business activities. Furthermore, there are no effective systems for demand forecasting or optimizing delivery plans for food delivery services. As a result, users are unable to adequately respond to the inconveniences and risks caused by weather conditions, and business efficiency is reduced.

[0778] 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.

[0779] In this invention, the server includes: means for acquiring weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; means for saving the analysis results in a database and updating them in real time; means for notifying the user when specific weather conditions are predicted based on the user's settings; means for the user to refer to detailed weather forecast information via a terminal; and means for making specific suggestions that will influence the user's behavior based on the weather forecast information. This makes it possible to propose specific impacts on daily life and business activities based on weather conditions. Furthermore, demand forecasting and optimization of delivery plans can be realized in food delivery services, improving user and business efficiency.

[0780] An "external database" is a system that exists on the Internet or in a cloud environment and accumulates and provides weather-related data.

[0781] "Weather-related data" refers to data that indicates weather conditions such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion status.

[0782] "Preprocessing" refers to the process of performing processes on the acquired weather-related data, such as normalizing the data, filling in missing values, and removing noise.

[0783] A "machine learning model" is an algorithm that learns patterns and rules based on data and makes predictions and classifications for new data.

[0784] A "deep learning model" is an algorithm that uses a multi-layer neural network to automatically learn the characteristics of complex data and perform advanced predictions and classifications.

[0785] "Real-time updates" means that information in a database or system is kept up-to-date immediately whenever new data is added.

[0786] A "terminal" is a device such as a smartphone, tablet, or PC that allows a user to access the Internet and obtain information.

[0787] The "particular weather conditions" are weather phenomena that the user has set in advance as being of interest or concern, such as thunderstorms, typhoons, and heavy snow.

[0788] "Notification" is a feature that sends alerts when certain weather conditions are predicted based on conditions set by the user in advance.

[0789] "Specific suggestions to influence user behavior" refers to providing specific advice to users on their daily lives and business activities based on weather conditions, such as encouraging them to change the timing of going out or ordering food delivery.

[0790] "Demand forecasting" refers to predicting the number of orders for food delivery services in advance based on weather conditions and past data.

[0791] "Optimal delivery planning" refers to optimizing delivery routes and times based on predicted demand in order to provide efficient food delivery services.

[0792] The present invention provides a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to fill in missing values, and removes noise. The system then uses the preprocessed data to analyze weather forecasts using machine learning or deep learning models, and stores the analysis results in a database that is updated in real time. Based on user settings, the system notifies users when specific weather conditions are predicted and allows them to access detailed weather forecast information via their device.

[0793] Data collection

[0794] The server retrieves weather-related data from external databases. This data includes information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. This data is collected periodically using APIs, such as the OpenWeatherMap API and the Japan Meteorological Agency API.

[0795] Data Preprocessing

[0796] The server preprocesses the acquired weather-related data. This preprocessing includes data normalization, missing value completion, and noise removal. The Python pandas library is used here. For example, if there are gaps in the data obtained from a particular sensor, the gaps are filled in using surrounding data, and a noise removal filter is applied to improve the quality of the data.

[0797] Data analysis

[0798] Using the preprocessed data, the server analyzes the weather forecast using machine learning or deep learning models. For short-term predictions, it uses a recurrent neural network (LSTM) model using TensorFlow or PyTorch, while for long-term predictions, it uses an ensemble learning model. This enables detailed predictions such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon paths.

[0799] Database storage and real-time updates

[0800] The analysis results are stored in a database by the server. This database is designed to be updated in real time using Firebase Realtime Database or Amazon RDS, and the predictive model is re-executed frequently, ensuring that the latest forecast data is always provided to users.

[0801] Notification function

[0802] The device will send notifications when certain weather conditions are predicted based on the user's settings, such as "thunderstorm warnings" and "pollen alerts," using Firebase Cloud Messaging (FMC).

[0803] Information references and specific suggestions

[0804] Users can open the application on their devices to view detailed weather forecast information, including the latest weather data, weather radar, typhoon tracks, pollen count information, and other details. Based on the weather forecast information, the application will also provide specific suggestions to influence users' behavior, such as recommending when to go out or ordering food delivery.

[0805] Specific examples

[0806] For example, on a given day in Tokyo, the server retrieves the latest weather data from the weather database. After data preprocessing, it runs an LSTM model to predict a thunderstorm in one hour. This prediction result is saved in the database and updated in real time. The device then sends a notification based on the user's "thunderstorm warning" setting, and the user opens an application to check detailed weather forecast information. As a result, the user can change their plans, order food delivery, and take appropriate precautions against the thunderstorm.

[0807] Examples of specific prompts include the following:

[0808] "Predict the increase in delivery requests in Tokyo next week based on the amount of precipitation and average wind speed."

[0809] In this way, the weather forecast system of the present invention acquires weather-related data from an external database and provides comprehensive functions for preprocessing, analysis, storage, notification, information reference, and specific suggestions, thereby providing users with the latest and most accurate weather forecast information and action suggestions in real time.

[0810] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0811] Step 1:

[0812] Data collection

[0813] The server retrieves weather-related data from an external database. Specifically, it periodically requests information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count using the OpenWeatherMap API and the Japan Meteorological Agency API, and receives JSON-formatted data in response. The input data is raw weather data retrieved from the API, and the output data is this raw weather data.

[0814] Step 2:

[0815] Data Preprocessing

[0816] The server performs data preprocessing using the acquired weather-related data. Specifically, it normalizes the data using Python's pandas library, imputes missing values ​​using SimpleImputer, and removes noise using a noise removal filter. The input data is raw weather data, and the output data is preprocessed weather data that has been normalized, imputes missing values, and removed noise.

[0817] Step 3:

[0818] Data analysis

[0819] Using the preprocessed data, the server analyzes the weather forecast using a machine learning or deep learning model. The preprocessed data is input into an LSTM model or ensemble learning model built using TensorFlow or PyTorch, and weather forecasts such as future temperature, precipitation probability, wind speed, wind direction, thunderstorm predictions, and typhoon paths are output. The input data is preprocessed weather data, and the output data is analyzed weather forecast data.

[0820] Step 4:

[0821] Database storage and real-time updates

[0822] The server stores the analysis results in Firebase Realtime Database or Amazon RDS and updates them in real time. Specifically, the analyzed weather forecast data is written to these databases and set up to be automatically updated each time new data is added. The input data is the analyzed weather forecast data, and the output data is the latest weather forecast data stored in the database.

[0823] Step 5:

[0824] Notification function

[0825] The device sends notifications when certain weather conditions are predicted based on the user's settings. Specifically, it uses Firebase Cloud Messaging (FCM) to instantly send notifications about predicted weather conditions based on the user's settings for "thunderstorm warnings" and "pollen alerts." The input data is the analyzed weather forecast data retrieved from the database and the user's settings, and the output data is the notification message sent over the Internet.

[0826] Step 6:

[0827] Information references and specific suggestions

[0828] Users can access the application through their devices to view detailed weather forecast information. Based on the weather forecast information, the application also provides specific suggestions, such as when to go out or encouraging people to order food delivery. The input data is the weather forecast information displayed on the device and data related to the user's suggested actions, and the output data is the specific suggested actions provided to the user.

[0829] Step 7:

[0830] Food delivery demand forecasting and delivery plan optimization

[0831] The server predicts demand for food delivery services based on weather conditions and past data, and then formulates optimal delivery plans based on the predicted demand. This involves using a machine learning model to predict demand and then running an algorithm to optimize delivery routes and times based on the prediction results. The input data is preprocessed weather data and past delivery records, and the output data is the demand forecast and optimal delivery plan information.

[0832] 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.

[0833] The following describes in detail an embodiment of the present invention. The present invention combines a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to fill in missing values, and removes noise, with an emotion engine that recognizes user emotions. The system updates the weather forecast in real time based on the analysis results and notifies the user when specific weather conditions are predicted based on the user's settings. The system also allows the user to view detailed weather forecast information via a terminal.

[0834] Data collection

[0835] The server periodically retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database. This data is fetched via an API.

[0836] Data Preprocessing

[0837] The server preprocesses the collected weather-related data, including normalizing the data, filling in missing values, and removing noise. Specifically, it fills in missing data with surrounding data points and applies a noise removal filter to improve data quality.

[0838] Data analysis

[0839] The server uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. LSTM (long short-term memory) networks are used for short-term predictions, while ensemble learning models are used for long-term predictions. This results in a detailed weather forecast (temperature, probability of precipitation, wind speed, wind direction, thunderstorm forecast, and typhoon path).

[0840] Database storage and real-time updates

[0841] The analysis results are stored in a database by the server and updated in real time. New data is collected at regular intervals (e.g., every 30 minutes), and the predictive model is re-run to generate the latest forecast results.

[0842] Emotion analysis using an emotion engine

[0843] The server uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state from their voice, facial expressions, text input, etc. The analyzed emotional state is stored in a database and reflected in the content and format of the weather forecast notification.

[0844] User Notification Settings

[0845] The device will send notifications that take into account the user's emotional state based on specific weather conditions (thunderstorms, typhoons, pollen counts, etc.) set by the user. For example, if the user has set up a thunderstorm notification, the device will send a notification with more detailed information and follow-up advice depending on the user's stress level detected by the emotion engine.

[0846] Information reference

[0847] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, and other details. The priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0848] Specific examples

[0849] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database, and after preprocessing, analyzes it using an LSTM model, predicting a thunderstorm in one hour. The analysis results are stored in the database and updated in real time. The server then analyzes the user's stress level using an emotion engine and creates a notification based on that state. The device receives this notification and sends it to the user with detailed advice such as, "Thunderstorms are predicted. Avoid sudden outings or bring an umbrella." The user can then change their plans based on this notification and take appropriate precautions against the thunderstorm.

[0850] In this way, the system of the present invention acquires weather-related data from an external database, and in addition to preprocessing, analyzing, storing, notifying, and referencing the information, it uses an emotion engine to provide personalized weather information based on the user's emotional state.

[0851] The processing flow will be explained below.

[0852] Step 1:

[0853] The server retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database by fetching the data periodically using an API.

[0854] Step 2:

[0855] The weather-related data acquired by the server is preprocessed. Specifically, data normalization, missing value completion, and noise removal are performed. For example, if there are missing parts in the acquired data, the parts are completed using surrounding data points.

[0856] Also, a noise reduction filter is applied to improve the quality of the data.

[0857] Step 3:

[0858] The server then uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. It uses LSTM (long short-term memory) networks for short-term predictions and ensemble learning models for long-term predictions. This generates detailed forecasts such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon tracks.

[0859] Step 4:

[0860] The server stores the analysis results in a database that is designed to be updated in real time, collecting new data at regular intervals (e.g., every 30 minutes) and re-running the predictive model to generate the latest forecasts.

[0861] Step 5:

[0862] The server uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state from their voice, facial expressions, text input, etc. The analyzed emotional state is stored in a database.

[0863] Step 6:

[0864] The server adjusts the content and format of notifications based on user settings when certain weather conditions (e.g., thunderstorms or typhoons) are predicted. For example, if the emotion engine determines that the user is feeling stressed, it creates a notification with follow-up advice appropriate to that state.

[0865] Step 7:

[0866] The device receives notifications sent from the server. These notifications are customized based on the user's settings and emotional state. For example, if a user has set up "thunderstorm notifications" and a stressful state is detected, the device receives a notification with specific advice such as "A thunderstorm is predicted within 30 minutes. Avoid sudden outings and bring an umbrella if necessary."

[0867] Step 8:

[0868] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, and pollen count information. For example, before commuting, users can check the weather forecast for their current location and destination to determine the best course of action.

[0869] Step 9:

[0870] The server periodically collects the latest data in real time, re-runs the prediction model to generate the latest prediction results, and stores them in the database again, thereby always providing the latest prediction information to users.

[0871] In this way, each step works in conjunction with the other steps to provide a personalized, real-time, and highly accurate weather forecast based on the user's emotional state.

[0872] Example 2

[0873] 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."

[0874] The appropriate collection and prediction of weather-related data is important for daily life and industrial activities. However, improving the accuracy of weather forecasts and providing information tailored to individual users' situations presents many challenges. In particular, existing systems are unable to adequately address issues such as missing or noisy data, real-time updates, and generating notifications that take into account the user's emotional state.

[0875] The identification processing 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 acquiring weather-related data from an external database, means for preprocessing the weather-related data to complement missing values ​​and remove noise, means for analyzing a weather forecast using a machine learning model or a deep learning model using the preprocessed data, means for saving the analysis results in a database and updating them in real time, means for recognizing and analyzing a user's emotional state, means for providing notifications based on specific weather conditions and the user's emotional state based on user settings, and means for the user to refer to detailed weather forecast information via a terminal. This allows a series of processes, such as collection, preprocessing, analysis, storage, and notification of weather-related data, to be efficiently executed, making it possible to provide personalized weather forecast information that takes the user's emotional state into consideration.

[0876] An "external database" is a collection of data that resides outside the system and stores and provides weather-related data.

[0877] "Weather-related data" refers to data relating to weather and meteorological conditions, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0878] "Preprocessing" refers to a series of processes performed on acquired weather-related data, such as data normalization, missing value completion, and noise removal.

[0879] "Missing value imputation" is the process of filling in missing data points in a dataset using surrounding data.

[0880] "Noise removal" refers to the process of filtering out abnormal values ​​and unnecessary data from data.

[0881] A "machine learning model" is an algorithm for predicting future data based on past data, and primarily uses statistical methods.

[0882] A "deep learning model" is an algorithm that uses neural networks to analyze data in a multi-layered manner and make advanced predictions.

[0883] "Real-time updating" refers to the process of collecting data at regular intervals and periodically replacing the analysis results with the latest ones.

[0884] "Emotional state" refers to the psychological and emotional state of the user as analyzed from their voice, facial expressions, text input, etc.

[0885] "Notifications" are messages or alerts that provide information about specific weather conditions based on the user's preferences and emotional state.

[0886] A "terminal" is a device (e.g., smartphone, tablet, computer) that a user uses to access and operate the system.

[0887] "Detailed weather forecast information" is weather forecast data that goes beyond simple weather forecasts and includes additional information such as weather radar, typhoon tracks, and pollen dispersion information.

[0888] The present invention is a weather forecasting system that retrieves weather-related data from an external database, preprocesses it to fill in missing values, and removes noise. The system also incorporates an emotion engine that recognizes the user's emotions. The system updates the weather forecast in real time based on the analysis results and notifies the user when specific weather conditions are predicted based on the user's settings. The system also allows the user to view detailed weather forecast information via their device.

[0889] The server periodically retrieves weather-related data (e.g., temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database. This data is fetched via an API. For example, the data could be retrieved using the OpenWeatherMap API.

[0890] The server preprocesses the collected weather-related data. Specific preprocessing steps include normalizing the data, filling in missing values, and removing noise. For example, the Pandas library is used to fill in NaN values ​​with approximate values, and the Scipy library is used to apply a low-pass filter to smooth the data.

[0891] The server uses the preprocessed data to analyze weather forecasts using machine learning or deep learning models. It uses a long short-term memory (LSTM) network for short-term predictions and an ensemble learning model for long-term predictions. Specific implementations could use frameworks such as TensorFlow or PyTorch.

[0892] The analysis results are stored in a database by the server and updated in real time. New data is collected at regular intervals (for example, every 30 minutes), and the predictive model is re-run to generate the latest prediction results. The database can be a relational database such as MySQL or PostgreSQL.

[0893] The server uses an emotion engine to recognize the user's emotions. This engine analyzes the user's emotional state based on voice, facial expressions, and text input. For example, facial expressions can be captured using OpenCV and analyzed using a deep learning model, or voice data can be converted into text using a speech recognition API such as IBM Watson, and emotion analysis can then be performed.

[0894] The device sends notifications that take into account the user's emotional state based on specific weather conditions set by the user (e.g., thunderstorms, typhoons, pollen counts, etc.). A real-time notification function is implemented using JavaScript, etc., and notifications are sent with wording that reflects the user's emotional state. For example, if the user is feeling stressed, a detailed and kind notification will be sent.

[0895] Users can access detailed weather forecast information by opening the app on their devices. The app displays the latest forecast data, weather radar, typhoon tracks, pollen count information, etc. Furthermore, the priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0896] Specific examples

[0897] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database, and after preprocessing, analyzes it using an LSTM model, predicting a thunderstorm in one hour. The analysis results are stored in the database and updated in real time. The server then analyzes the user's stress level using an emotion engine and creates a notification based on that state. The device receives this notification and sends it to the user with detailed advice such as, "Thunderstorms are predicted. Avoid sudden outings or bring an umbrella." The user can then change their plans based on this notification and take appropriate precautions against the thunderstorm.

[0898] Examples of prompt statements

[0899] "Describe a scenario where a server retrieves the latest weather data from an external database and predicts a thunderstorm in the next hour in Tokyo on a certain day. The emotion engine detects the user's stress state and sends a notification with detailed advice."

[0900] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0901] Step 1: Data collection

[0902] The server periodically retrieves weather-related data from an external database. Specifically, the server sends an API request and receives a JSON-formatted response from the external database. This input data includes temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. As output, this weather-related data is obtained and passed to the next preprocessing step.

[0903] Example of how it works:

[0904] The server sends an API request every hour.

[0905] Receive the API response (weather data JSON) and save it in local storage.

[0906] Step 2: Data Preprocessing

[0907] The server preprocesses the collected weather-related data. This preprocessing involves normalizing the data, filling in missing values, and removing noise. The input is the raw data collected in step 1, and the output is preprocessed, high-quality data.

[0908] Example of how it works:

[0909] Use the Pandas library to read the JSON data and convert it to a dataframe.

[0910] To impute missing values, fill NaN values ​​in the data frame with the average of the preceding and following data points.

[0911] Apply a low-pass filter using the Scipy library to remove noise.

[0912] Step 3: Data analysis

[0913] The server uses the preprocessed data to analyze the weather forecast with a machine learning or deep learning model, where the input is the preprocessed data from step 2 and the output is the weather forecast data.

[0914] Example of how it works:

[0915] Use TensorFlow or PyTorch to load the LSTM model and run predictions using the preprocessed data as input.

[0916] Short-term predictions (1 to 24 hours) are performed using LSTM, and long-term predictions (1 week) are performed using an ensemble learning model.

[0917] Evaluate the analysis results and record performance metrics.

[0918] Step 4: Save to database and update in real time

[0919] The analysis results are stored in a database by the server and updated at regular intervals. The input is the analysis result from step 3, and the output is the prediction results stored in the database.

[0920] Example of how it works:

[0921] Connect to a MySQL or PostgreSQL database and insert the analysis results into a table.

[0922] Every 30 minutes, start with a new data collection and rerun the preprocessing and analysis.

[0923] Delete the old prediction results in the database and update with the new results.

[0924] Step 5: Emotion analysis using the emotion engine

[0925] The server uses an emotion engine to recognize the user's emotions. The input is the user's voice, facial expression, text input, etc., and the output is the user's emotional state.

[0926] Example of how it works:

[0927] Facial expressions are captured from a webcam using OpenCV and analyzed using a deep learning model.

[0928] Voice data is converted into text through a voice recognition API and sentiment analysis is performed.

[0929] Text analysis algorithms are used to recognize emotions from messages and comments.

[0930] Step 6: User Notification Settings

[0931] The device sends notifications based on specific weather conditions set by the user, taking into account their emotional state. The input is the data from steps 4 and 5, and the output is the notification sent to the user.

[0932] Example of how it works:

[0933] Check the user's settings and generate notifications when certain weather conditions (such as thunderstorms, typhoons, and pollen counts) are predicted.

[0934] Send real-time notifications to your device using JavaScript.

[0935] If the user is feeling stressed, a detailed notification containing kind words is generated.

[0936] Step 7: Information Reference

[0937] The user opens the application on the device and views detailed weather forecast information. The input is the forecast result stored in the database in step 4, and the output is the weather forecast information displayed on the device.

[0938] Example of how it works:

[0939] Open a mobile or web application.

[0940] The app screen displays the latest weather forecast, weather radar, typhoon path, pollen information, and more.

[0941] The information display format is automatically adjusted according to the user's emotional state.

[0942] (Application example 2)

[0943] 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."

[0944] In today's world, highly accurate weather forecasts have become important information in many fields. However, simply providing weather information makes it difficult to present users with measures appropriate to their individual needs and situations. In particular, physical stores require inventory management that takes weather conditions into account and appropriate product recommendations to customers. Furthermore, services that do not take into account user emotions make it difficult to improve customer satisfaction.

[0945] Therefore, the objective of the present invention is to provide a system that combines weather data and a user's emotional state to optimize inventory management and promotion notifications in physical stores and meet individual needs.

[0946] 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.

[0947] In this invention, the server includes means for acquiring weather-related data from an external database, means for preprocessing the weather-related data to fill in missing values ​​and remove noise, means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data, means for storing the analysis results in a database and updating them in real time, means for notifying the user when specific weather conditions are predicted based on user settings, means for the user to refer to detailed weather forecast information via a terminal, means for performing demand forecasts based on the weather data and optimizing inventory in physical stores, and means for sending personalized promotional notifications according to the user's emotional state. This enables optimal inventory management in response to weather conditions and individual promotional notifications that take the user's emotional state into consideration.

[0948] "Weather-related data" refers to data relating to weather and meteorological conditions, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[0949] "Preprocessing" refers to the process of preparing data in an appropriate form before data analysis, such as normalizing data, filling in missing values, and removing noise.

[0950] A "machine learning model" is a model that learns patterns and regularities based on data and makes predictions and classifications.

[0951] A "deep learning model" is a model that uses a multi-layer neural network that can learn complex data patterns.

[0952] "Weather forecast" is information that predicts future weather conditions based on acquired weather data.

[0953] "Real-time updates" means obtaining new data and updating the analysis results based on that data at any time to provide the latest information.

[0954] The "notification means" is a means for notifying the user of information when a specific condition is met.

[0955] A "brick and mortar store" is a retail or service establishment that exists in a physical location.

[0956] "Inventory management" refers to the proper management and adjustment of the quantity of goods and resources, and maintaining an optimal inventory level according to demand.

[0957] A "promotional notice" is a notice to inform users of information about a particular product or service as part of an advertisement or promotion.

[0958] "Emotional state" refers to the user's emotional state, such as stress, joy, or sadness.

[0959] "Personalized" means providing information and services tailored to the specific needs and circumstances of each individual user.

[0960] This system acquires weather-related data from an external database, performs missing value imputation and noise removal, and then generates highly accurate weather forecasts using machine learning and deep learning models. It also analyzes the user's emotional state and provides personalized notifications and information based on that.

[0961] Data collection and preprocessing

[0962] The server periodically retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database (e.g., a weather API). This data is fetched via the API. The server then preprocesses the collected weather data, imputes missing values, and removes noise. Specifically, it uses Scikit-learn's SimpleImputer to impute missing data and StandardScaler to normalize the data.

[0963] Weather forecast analysis

[0964] Using the preprocessed weather data, the server analyzes the weather forecast using machine learning models (e.g., LSTM models) or deep learning models. LSTM (long short-term memory) networks are used for short-term predictions, while ensemble learning models are used for long-term predictions. This results in a detailed weather forecast (temperature, probability of precipitation, wind speed, wind direction, thunderstorm forecast, and typhoon path). The analysis results are stored in a database and updated in real time.

[0965] Sentiment Analysis and Notifications

[0966] The server implements an emotion analysis engine to analyze the user's emotional state. This engine analyzes the user's voice, facial expressions, text input, etc. to determine the user's emotional state. For example, it analyzes text input using an emotion recognition library such as Emojify.

[0967] Based on the results of sentiment analysis, notifications are sent when certain weather conditions are predicted. For example, if a thunderstorm is predicted and the user is feeling stressed, a personalized notification with detailed advice is sent. The notification may include specific advice such as, "Thunderstorms are predicted. Avoid sudden outings and, if necessary, check out stress-relieving products."

[0968] Inventory management and promotion notifications

[0969] The server uses weather data to predict demand and optimize inventory in physical stores. For example, if rain is predicted, it increases the stock of umbrellas and raincoats. It also sends personalized promotional notifications based on the user's emotional state. For example, if it determines that the user needs relaxation, it notifies them of promotions for relaxation-related products.

[0970] Information reference

[0971] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, and other details. The priority and format of the displayed information is automatically optimized based on the user's emotional state.

[0972] Specific examples

[0973] For example, if a thunderstorm is predicted for Tokyo on a certain day and the user is feeling stressed, the following notification will be displayed: "Thunderstorms are predicted. Avoid sudden outings and, if necessary, check for products to relieve stress."

[0974] Prompt Sentence Examples

[0975] "Based on data obtained from a weather data API, impute and normalize missing values ​​and use an LSTM model to predict the weather. Next, analyze the user's emotional state from input and generate a promotional notification for relaxation-related products if the user is feeling stressed when a thunderstorm is predicted."

[0976] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0977] Step 1:

[0978] The server retrieves weather-related data from an external database through an API, which contains information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. The input is the API endpoint, and the output is the weather-related data.

[0979] Step 2:

[0980] The server preprocesses the acquired weather-related data. Specifically, it uses SimpleImputer to impute missing values ​​and StandardScaler to normalize the data. This improves the quality of the data and makes it suitable for analysis. The input is the acquired weather-related data, and the output is the preprocessed data.

[0981] Step 3:

[0982] The server uses the preprocessed data to analyze the weather forecast using a machine learning or deep learning model. It uses an LSTM model to perform short-term predictions and an ensemble learning model to perform long-term predictions. The input is the preprocessed data, and the output is the weather forecast results.

[0983] Step 4:

[0984] The server stores the weather forecast results in a database and updates it in real time. Every 30 minutes, it collects new data and re-runs the model to generate the latest forecast results. The input is the weather forecast results, and the output is the updated database.

[0985] Step 5:

[0986] The server uses an emotion analysis engine to analyze the user's emotional state. It determines the emotional state from the user's voice, facial expression, text input, etc. When analyzing text input using tools such as Emojify, the input is the text from the user and the output is the user's emotional state.

[0987] Step 6:

[0988] The server notifies users when certain weather conditions are predicted based on the emotion analysis results and weather forecast results. Specifically, if a thunderstorm is predicted and the user is feeling stressed, the server generates a personalized notification with detailed advice. The input is the weather forecast result and the user's emotional state, and the output is the notification content.

[0989] Step 7:

[0990] The server predicts demand based on weather data and optimizes inventory in physical stores. If rain is predicted, it will increase the stock of umbrellas and raincoats, and perform optimal inventory management. The input is the weather forecast result, and the output is inventory management instructions.

[0991] Step 8:

[0992] The server sends personalized promotion notifications according to the user's emotional state. For example, if the server determines that the user needs relaxation, it notifies the user of promotions for relaxation-related products. The input is the user's emotional state, and the output is the promotion notification.

[0993] Step 9:

[0994] Users can open the application on their devices and view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, etc. The input is launching the application, and the output is the displayed weather forecast information.

[0995] 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.

[0996] 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.

[0997] 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.

[0998] [Fourth embodiment]

[0999] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1000] 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.

[1001] 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).

[1002] 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.

[1003] 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.

[1004] 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).

[1005] 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. 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.

[1006] 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.

[1007] 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.

[1008] 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.

[1009] 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.

[1010] 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.

[1011] 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."

[1012] The following describes in detail an embodiment of the present invention. The present invention provides a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to complement missing values, and removes noise. Furthermore, the system uses the preprocessed data to analyze weather forecasts using a machine learning or deep learning model, and stores the analysis results in a database that is updated in real time. Based on user settings, the system notifies users when specific weather conditions are predicted and allows users to view detailed weather forecast information via their device.

[1013] Data collection

[1014] The server retrieves weather-related data from an external database, including information on temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. This data is collected periodically using an API.

[1015] Data Preprocessing

[1016] The server preprocesses the acquired weather-related data. This includes data normalization, missing value completion, and noise removal. For example, if there are gaps in the data obtained from a particular sensor, the server uses surrounding data to complete the gaps. It also applies a noise removal filter to improve the quality of the data.

[1017] Data analysis

[1018] Using the preprocessed data, the server analyzes the weather forecast using machine learning or deep learning models. For short-term predictions, it uses LSTM (recurrent neural networks), while for long-term predictions, it uses ensemble learning or physical models. This enables detailed predictions of temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, typhoon paths, and more.

[1019] Database storage and real-time updates

[1020] The analysis results are stored in a database by the server. This database is designed to be updated in real time and frequently re-runs the predictive model, ensuring that the latest forecast data is always available to users.

[1021] Notification function

[1022] The device will send notifications to users when certain weather conditions are predicted based on their settings, such as when a thunderstorm is predicted in a specific area based on their settings for "thunderstorm warnings."

[1023] Information reference

[1024] Users can open the application on their devices to view detailed weather forecast information, including the latest weather data, weather radar, typhoon tracks, pollen count information, and other details. For example, if a user wants to check the weather forecast for their current location and destination before commuting, they can simply open the application and the latest information will be displayed.

[1025] Specific examples

[1026] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database. After data preprocessing, it analyzes it using an LSTM model and predicts a thunderstorm in one hour. This prediction result is saved in the database and updated in real time. Next, the device sends a notification based on the "thunderstorm warning" set by the user, and the user opens the application to check detailed weather forecast information. As a result, the user can change their plans and take appropriate measures against the thunderstorm.

[1027] In this way, the weather forecast system of the present invention obtains weather-related data from an external database and provides comprehensive functions for preprocessing, analysis, storage, notification, and information reference, thereby providing users with the latest and most accurate weather forecast information in real time.

[1028] The processing flow will be explained below.

[1029] Step 1:

[1030] The server retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database through a process that periodically fetches the data via an API.

[1031] Step 2:

[1032] The weather-related data acquired by the server is preprocessed, which includes processes such as data normalization, missing value imputation, and noise removal. For example, missing data imputation uses an algorithm that calculates an estimate using surrounding data points.

[1033] Step 3:

[1034] The server then uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. It uses LSTM (long short-term memory) networks for short-term predictions and ensemble learning models for long-term predictions. This generates detailed forecasts such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon paths.

[1035] Step 4:

[1036] The server stores the analysis results in a database that is designed to be updated in real time, with the predictive model being re-run on new data at regular intervals (e.g., every 30 minutes).

[1037] Step 5:

[1038] The server generates a notification when certain weather conditions are predicted based on the user's settings. For example, a notification is generated when "thunderstorms are predicted in Tokyo within the next hour."

[1039] Step 6:

[1040] The device receives notifications sent from the server, which are customized based on the user's settings. For example, if the user requests thunderstorm notifications, the device immediately provides that information to the user.

[1041] Step 7:

[1042] Users can open the application on their devices to view detailed weather forecast information, including the latest weather forecast, weather radar, typhoon tracks, pollen count information, etc., which they can use to decide on appropriate actions.

[1043] Step 8:

[1044] The server periodically collects the latest data in real time, re-runs the prediction model to generate the latest prediction results, and stores them in the database again, so that the latest prediction information is always available to users.

[1045] In this way, each step works in cooperation with the other steps to provide users with highly accurate weather forecasts in real time.

[1046] Example 1

[1047] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1048] Conventional weather forecasting systems often lack accuracy and real-time response in the entire process, from collecting weather data to analyzing and notifying users. They also face difficulties in accessing the detailed weather information users require, and lack the ability to quickly notify users of specific weather conditions. In particular, inadequate handling of missing values ​​and noise reduces data quality and negatively impacts forecast accuracy. Furthermore, there is a need for greater system efficiency and smoother data updates.

[1049] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1050] In this invention, the server includes: means for acquiring weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; means for saving the analysis results in a database and updating them in real time; means for notifying the user when specific weather conditions are predicted based on user settings; means for the user to refer to detailed weather forecast information via a terminal; means for using an API to collect the data and preprocessing the collected data by filling in missing values ​​and applying a noise removal filter; means for using a recurrent neural network and an ensemble learning model to analyze the weather forecast; means for using a relational database management system to save and update the data in the database in real time; and means for using a cloud messaging service for the notification means. This enables efficient and highly accurate data collection, analysis, storage, notification, and reference.

[1051] An "external database" is a database that provides weather-related data and is accessible via the Internet or the like.

[1052] "API" is an abbreviation for Application Programming Interface, a means of exchanging data and functions between different software systems.

[1053] "Weather-related data" refers to all weather-related information, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[1054] "Preprocessing" refers to processes to process the acquired data into a state suitable for analysis, such as normalizing the data, filling in missing values, and removing noise.

[1055] "Missing value imputation" is the process of filling in missing values ​​in a dataset with surrounding data, estimated values, etc.

[1056] "Noise removal" is a process for removing irregular fluctuations and outliers from data to improve data quality.

[1057] A "machine learning model" is an algorithm or method that learns patterns in data and makes predictions or classifications about future data.

[1058] A "deep learning model" is a highly accurate algorithm that uses a multi-layer neural network to extract data features and perform predictions and classifications.

[1059] A "recurrent neural network" is a special neural network for handling time series data, and is a model that retains past information and makes predictions.

[1060] "Ensemble learning" is a technique that combines multiple machine learning models to achieve higher predictive accuracy than individual models.

[1061] A "relational database management system" is a software system for storing, updating, searching, and managing data, and manages structured data in tabular format.

[1062] A "cloud messaging service" is a service for sending notification messages over the Internet, and is a means of providing real-time notifications to mobile devices and the like.

[1063] "User settings" refers to individual setting information such as notification conditions and reference conditions customized by the user on the system.

[1064] "Terminal" refers to an electronic device, such as a smartphone, tablet, or PC, that a user uses to access the system.

[1065] "Real-time updates" means that the data in the system is always kept up to date and updates are reflected almost immediately.

[1066] MODE FOR CARRYING OUT THE INVENTION

[1067] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention is a weather forecast system in which a server, a terminal, and a user are used in cooperation with each other.

[1068] Data collection

[1069] The server retrieves weather-related data from an external database. Specifically, it collects data via an API. The Python requests library is used to retrieve the data, and weather-related data includes temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. For example, the OpenWeatherMap API is used to retrieve this data.

[1070] Data Preprocessing

[1071] The server preprocesses the acquired weather-related data. This includes data normalization, missing value imputation, and noise removal. It uses Python's Pandas library to impute missing values ​​with surrounding data and applies SciPy's noise removal filter. This improves the quality of the data and prepares it for analysis.

[1072] Data analysis

[1073] The server uses preprocessed data to analyze weather forecasts using machine learning and deep learning models. It uses recurrent neural networks (LSTM) for short-term predictions and ensemble learning models for long-term predictions. It uses libraries such as TensorFlow and Keras to build predictive models and analyze temperature, precipitation probability, wind speed, wind direction, and more.

[1074] Database storage and real-time updates

[1075] The server stores the analysis results in a database and updates them in real time. The database uses a relational database management system such as PostgreSQL. The SQLAlchemy library is used to efficiently store and update data. This ensures that the latest prediction data is always available to users.

[1076] Notification function

[1077] The device sends notifications when certain weather conditions are predicted based on the user's settings. For example, if the user has set a "thunderstorm alert," it sends a push notification using a cloud messaging service. For this purpose, it uses Firebase Cloud Messaging (FCM).

[1078] Information reference

[1079] Users can access detailed weather forecast information on their devices. Through the application UI, they can view the latest weather data, weather radar, typhoon tracks, pollen count information, and more. For example, an application developed with React Native and Flutter allows users to easily check the weather forecast for their current location and destination before commuting to work.

[1080] Specific examples

[1081] For example, on a certain day in Tokyo, the server retrieves weather data using the OpenWeatherMap API. After preprocessing the data, an LSTM model is used to predict thunderstorms one hour in the future. This prediction result is stored in a PostgreSQL database and updated in real time. Next, the device sends a notification via FCM based on the "thunderstorm warning" set by the user. The user opens the application created with React Native and checks detailed weather forecast information. As a result, the user can change their plans and take appropriate measures against the thunderstorm.

[1082] Prompt Sentence Examples

[1083] "Get the latest weather data for Tokyo and predict the chance of thunderstorms in the next hour."

[1084] In this way, by linking the server, terminal, and user, we provide a system that efficiently and accurately accomplishes everything from data collection to analysis, storage, notification, and reference.

[1085] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1086] Step 1: Data collection

[1087] The server retrieves weather-related data from an external database. Specifically, it sends a request to the API and receives data in JSON format, including temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. The input is the endpoint URL and API key as an API request, and the output is weather-related data in JSON format. This data collection occurs periodically, and the next step is taken whenever new data is retrieved.

[1088] Step 2: Data Preprocessing

[1089] The server preprocesses the weather-related data in JSON format. First, it converts the data into a data frame using the Pandas library. Next, it imputes missing values ​​based on neighboring data points and applies a SciPy denoising filter to improve the data quality. The input is weather data in JSON format, and the output is a preprocessed data frame.

[1090] Step 3: Data analysis

[1091] The server uses the preprocessed data to analyze weather forecasts using machine learning and deep learning models. First, it builds LSTM models and ensemble learning models using TensorFlow and Keras. Next, it inputs the preprocessed data into these models to obtain forecast results such as temperature, precipitation probability, wind speed, and wind direction. The input is the preprocessed data frame, and the output is the analyzed weather forecast results.

[1092] Step 4: Save to database and update in real time

[1093] The server saves the analysis results in a database and updates them in real time. First, it connects to a PostgreSQL database using SQLAlchemy. Then it writes the analysis results to a database table and updates existing records if any. The input is the analyzed weather forecast results, and the output is the latest weather forecast data stored in the database.

[1094] Step 5: Notifications

[1095] The device sends a notification when certain weather conditions are predicted based on the user's settings. First, a notification message is created using Firebase Cloud Messaging (FCM). Then, the notification is sent at the appropriate time based on the conditions set by the user. The input is the user's settings and the analyzed weather forecast results, and the output is the notification message sent to the device.

[1096] Step 6: Information Reference

[1097] The user opens the application on their device to view detailed weather forecast information. The application UI displays the latest weather data, weather radar, typhoon tracks, and pollen count information. The input is the latest weather forecast data stored in the database, and the output is detailed weather forecast information displayed on the user's screen.

[1098] To give a concrete example, on a certain day in Tokyo, the server obtains weather data using the OpenWeatherMap API (Step 1). Next, the data is preprocessed to fill in missing values ​​and remove noise (Step 2). After that, the weather forecast is analyzed using an LSTM model (Step 3), and the analysis results are saved and updated in a PostgreSQL database (Step 4). The device sends a notification via FCM based on the "thunderstorm warning" set by the user (Step 5). Upon receiving this notification, the user opens the application and checks detailed weather forecast information (Step 6). This allows the user to change their plans and take appropriate measures against the thunderstorm.

[1099] (Application example 1)

[1100] 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."

[1101] Existing weather forecasting systems have technical limitations in the preprocessing and analysis of weather data, and are particularly lacking in providing suggestions that have a concrete impact on users' lives and business activities. Furthermore, there are no effective systems for demand forecasting or optimizing delivery plans for food delivery services. As a result, users are unable to adequately respond to the inconveniences and risks caused by weather conditions, and business efficiency is reduced.

[1102] 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.

[1103] In this invention, the server includes: means for acquiring weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; means for saving the analysis results in a database and updating them in real time; means for notifying the user when specific weather conditions are predicted based on the user's settings; means for the user to refer to detailed weather forecast information via a terminal; and means for making specific suggestions that will influence the user's behavior based on the weather forecast information. This makes it possible to propose specific impacts on daily life and business activities based on weather conditions. Furthermore, demand forecasting and optimization of delivery plans can be realized in food delivery services, improving user and business efficiency.

[1104] An "external database" is a system that exists on the Internet or in a cloud environment and accumulates and provides weather-related data.

[1105] "Weather-related data" refers to data that indicates weather conditions such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion status.

[1106] "Preprocessing" refers to the process of performing processes on the acquired weather-related data, such as normalizing the data, filling in missing values, and removing noise.

[1107] A "machine learning model" is an algorithm that learns patterns and rules based on data and makes predictions and classifications for new data.

[1108] A "deep learning model" is an algorithm that uses a multi-layer neural network to automatically learn the characteristics of complex data and perform advanced predictions and classifications.

[1109] "Real-time updates" means that information in a database or system is kept up-to-date immediately whenever new data is added.

[1110] A "terminal" is a device such as a smartphone, tablet, or PC that allows a user to access the Internet and obtain information.

[1111] The "particular weather conditions" are weather phenomena that the user has set in advance as being of interest or concern, such as thunderstorms, typhoons, and heavy snow.

[1112] "Notification" is a feature that sends alerts when certain weather conditions are predicted based on conditions set by the user in advance.

[1113] "Specific suggestions to influence user behavior" refers to providing specific advice to users on their daily lives and business activities based on weather conditions, such as encouraging them to change the timing of going out or ordering food delivery.

[1114] "Demand forecasting" refers to predicting the number of orders for food delivery services in advance based on weather conditions and past data.

[1115] "Optimal delivery planning" refers to optimizing delivery routes and times based on predicted demand in order to provide efficient food delivery services.

[1116] The present invention provides a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to fill in missing values, and removes noise. The system then uses the preprocessed data to analyze weather forecasts using machine learning or deep learning models, and stores the analysis results in a database that is updated in real time. Based on user settings, the system notifies users when specific weather conditions are predicted and allows them to access detailed weather forecast information via their device.

[1117] Data collection

[1118] The server retrieves weather-related data from external databases. This data includes information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count. This data is collected periodically using APIs, such as the OpenWeatherMap API and the Japan Meteorological Agency API.

[1119] Data Preprocessing

[1120] The server preprocesses the acquired weather-related data. This preprocessing includes data normalization, missing value completion, and noise removal. The Python pandas library is used here. For example, if there are gaps in the data obtained from a particular sensor, the gaps are filled in using surrounding data, and a noise removal filter is applied to improve the quality of the data.

[1121] Data analysis

[1122] Using the preprocessed data, the server analyzes the weather forecast using machine learning or deep learning models. For short-term predictions, it uses a recurrent neural network (LSTM) model using TensorFlow or PyTorch, while for long-term predictions, it uses an ensemble learning model. This enables detailed predictions such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon paths.

[1123] Database storage and real-time updates

[1124] The analysis results are stored in a database by the server. This database is designed to be updated in real time using Firebase Realtime Database or Amazon RDS, and the predictive model is re-executed frequently, ensuring that the latest forecast data is always provided to users.

[1125] Notification function

[1126] The device will send notifications when certain weather conditions are predicted based on the user's settings, such as "thunderstorm warnings" and "pollen alerts," using Firebase Cloud Messaging (FMC).

[1127] Information references and specific suggestions

[1128] Users can open the application on their devices to view detailed weather forecast information, including the latest weather data, weather radar, typhoon tracks, pollen count information, and other details. Based on the weather forecast information, the application will also provide specific suggestions to influence users' behavior, such as recommending when to go out or ordering food delivery.

[1129] Specific examples

[1130] For example, on a given day in Tokyo, the server retrieves the latest weather data from the weather database. After data preprocessing, it runs an LSTM model to predict a thunderstorm in one hour. This prediction result is saved in the database and updated in real time. The device then sends a notification based on the user's "thunderstorm warning" setting, and the user opens an application to check detailed weather forecast information. As a result, the user can change their plans, order food delivery, and take appropriate precautions against the thunderstorm.

[1131] Examples of specific prompts include the following:

[1132] "Predict the increase in delivery requests in Tokyo next week based on the amount of precipitation and average wind speed."

[1133] In this way, the weather forecast system of the present invention acquires weather-related data from an external database and provides comprehensive functions for preprocessing, analysis, storage, notification, information reference, and specific suggestions, thereby providing users with the latest and most accurate weather forecast information and action suggestions in real time.

[1134] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1135] Step 1:

[1136] Data collection

[1137] The server retrieves weather-related data from an external database. Specifically, it periodically requests information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen count using the OpenWeatherMap API and the Japan Meteorological Agency API, and receives JSON-formatted data in response. The input data is raw weather data retrieved from the API, and the output data is this raw weather data.

[1138] Step 2:

[1139] Data Preprocessing

[1140] The server performs data preprocessing using the acquired weather-related data. Specifically, it normalizes the data using Python's pandas library, imputes missing values ​​using SimpleImputer, and removes noise using a noise removal filter. The input data is raw weather data, and the output data is preprocessed weather data that has been normalized, imputes missing values, and removed noise.

[1141] Step 3:

[1142] Data analysis

[1143] Using the preprocessed data, the server analyzes the weather forecast using a machine learning or deep learning model. The preprocessed data is input into an LSTM model or ensemble learning model built using TensorFlow or PyTorch, and weather forecasts such as future temperature, precipitation probability, wind speed, wind direction, thunderstorm predictions, and typhoon paths are output. The input data is preprocessed weather data, and the output data is analyzed weather forecast data.

[1144] Step 4:

[1145] Database storage and real-time updates

[1146] The server stores the analysis results in Firebase Realtime Database or Amazon RDS and updates them in real time. Specifically, the analyzed weather forecast data is written to these databases and set up to be automatically updated each time new data is added. The input data is the analyzed weather forecast data, and the output data is the latest weather forecast data stored in the database.

[1147] Step 5:

[1148] Notification function

[1149] The device sends notifications when certain weather conditions are predicted based on the user's settings. Specifically, it uses Firebase Cloud Messaging (FCM) to instantly send notifications about predicted weather conditions based on the user's settings for "thunderstorm warnings" and "pollen alerts." The input data is the analyzed weather forecast data retrieved from the database and the user's settings, and the output data is the notification message sent over the Internet.

[1150] Step 6:

[1151] Information references and specific suggestions

[1152] Users can access the application through their devices to view detailed weather forecast information. Based on the weather forecast information, the application also provides specific suggestions, such as when to go out or encouraging people to order food delivery. The input data is the weather forecast information displayed on the device and data related to the user's suggested actions, and the output data is the specific suggested actions provided to the user.

[1153] Step 7:

[1154] Food delivery demand forecasting and delivery plan optimization

[1155] The server predicts demand for food delivery services based on weather conditions and past data, and then formulates optimal delivery plans based on the predicted demand. This involves using a machine learning model to predict demand and then running an algorithm to optimize delivery routes and times based on the prediction results. The input data is preprocessed weather data and past delivery records, and the output data is the demand forecast and optimal delivery plan information.

[1156] 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.

[1157] The following describes in detail an embodiment of the present invention. The present invention combines a weather forecasting system that acquires weather-related data from an external database, preprocesses the data to fill in missing values, and removes noise, with an emotion engine that recognizes user emotions. The system updates the weather forecast in real time based on the analysis results and notifies the user when specific weather conditions are predicted based on the user's settings. The system also allows the user to view detailed weather forecast information via a terminal.

[1158] Data collection

[1159] The server periodically retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database. This data is fetched via an API.

[1160] Data Preprocessing

[1161] The server preprocesses the collected weather-related data, including normalizing the data, filling in missing values, and removing noise. Specifically, it fills in missing data with surrounding data points and applies a noise removal filter to improve data quality.

[1162] Data analysis

[1163] The server uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. LSTM (long short-term memory) networks are used for short-term predictions, while ensemble learning models are used for long-term predictions. This results in a detailed weather forecast (temperature, probability of precipitation, wind speed, wind direction, thunderstorm forecast, and typhoon path).

[1164] Database storage and real-time updates

[1165] The analysis results are stored in a database by the server and updated in real time. New data is collected at regular intervals (e.g., every 30 minutes), and the predictive model is re-run to generate the latest forecast results.

[1166] Emotion analysis using an emotion engine

[1167] The server uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state from their voice, facial expressions, text input, etc. The analyzed emotional state is stored in a database and reflected in the content and format of the weather forecast notification.

[1168] User Notification Settings

[1169] The device will send notifications that take into account the user's emotional state based on specific weather conditions (thunderstorms, typhoons, pollen counts, etc.) set by the user. For example, if the user has set up a thunderstorm notification, the device will send a notification with more detailed information and follow-up advice depending on the user's stress level detected by the emotion engine.

[1170] Information reference

[1171] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, and other details. The priority and format of the displayed information is automatically optimized based on the user's emotional state.

[1172] Specific examples

[1173] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database, and after preprocessing, analyzes it using an LSTM model, predicting a thunderstorm in one hour. The analysis results are stored in the database and updated in real time. The server then analyzes the user's stress level using an emotion engine and creates a notification based on that state. The device receives this notification and sends it to the user with detailed advice such as, "Thunderstorms are predicted. Avoid sudden outings or bring an umbrella." The user can then change their plans based on this notification and take appropriate precautions against the thunderstorm.

[1174] In this way, the system of the present invention acquires weather-related data from an external database, and in addition to preprocessing, analyzing, storing, notifying, and referencing the information, it uses an emotion engine to provide personalized weather information based on the user's emotional state.

[1175] The processing flow will be explained below.

[1176] Step 1:

[1177] The server retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database by fetching the data periodically using an API.

[1178] Step 2:

[1179] The weather-related data acquired by the server is preprocessed. Specifically, data normalization, missing value completion, and noise removal are performed. For example, if there are missing parts in the acquired data, the parts are completed using surrounding data points.

[1180] Also, a noise reduction filter is applied to improve the quality of the data.

[1181] Step 3:

[1182] The server then uses the preprocessed data to analyze the weather forecast using machine learning or deep learning models. It uses LSTM (long short-term memory) networks for short-term predictions and ensemble learning models for long-term predictions. This generates detailed forecasts such as temperature, precipitation probability, wind speed, wind direction, thunderstorm forecasts, and typhoon tracks.

[1183] Step 4:

[1184] The server stores the analysis results in a database that is designed to be updated in real time, collecting new data at regular intervals (e.g., every 30 minutes) and re-running the predictive model to generate the latest forecasts.

[1185] Step 5:

[1186] The server uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state from their voice, facial expressions, text input, etc. The analyzed emotional state is stored in a database.

[1187] Step 6:

[1188] The server adjusts the content and format of notifications based on user settings when certain weather conditions (e.g., thunderstorms or typhoons) are predicted. For example, if the emotion engine determines that the user is feeling stressed, it creates a notification with follow-up advice appropriate to that state.

[1189] Step 7:

[1190] The device receives notifications sent from the server. These notifications are customized based on the user's settings and emotional state. For example, if a user has set up "thunderstorm notifications" and a stressful state is detected, the device receives a notification with specific advice such as "A thunderstorm is predicted within 30 minutes. Avoid sudden outings and bring an umbrella if necessary."

[1191] Step 8:

[1192] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, and pollen count information. For example, before commuting, users can check the weather forecast for their current location and destination to determine the best course of action.

[1193] Step 9:

[1194] The server periodically collects the latest data in real time, re-runs the prediction model to generate the latest prediction results, and stores them in the database again, thereby always providing the latest prediction information to users.

[1195] In this way, each step works in conjunction with the other steps to provide a personalized, real-time, and highly accurate weather forecast based on the user's emotional state.

[1196] Example 2

[1197] 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."

[1198] The appropriate collection and prediction of weather-related data is important for daily life and industrial activities. However, improving the accuracy of weather forecasts and providing information tailored to individual users' situations presents many challenges. In particular, existing systems are unable to adequately address issues such as missing or noisy data, real-time updates, and generating notifications that take into account the user's emotional state.

[1199] The identification processing 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 acquiring weather-related data from an external database, means for preprocessing the weather-related data to complement missing values ​​and remove noise, means for analyzing a weather forecast using a machine learning model or a deep learning model using the preprocessed data, means for saving the analysis results in a database and updating them in real time, means for recognizing and analyzing a user's emotional state, means for providing notifications based on specific weather conditions and the user's emotional state based on user settings, and means for the user to refer to detailed weather forecast information via a terminal. This allows a series of processes, such as collection, preprocessing, analysis, storage, and notification of weather-related data, to be efficiently executed, making it possible to provide personalized weather forecast information that takes the user's emotional state into consideration.

[1200] An "external database" is a collection of data that resides outside the system and stores and provides weather-related data.

[1201] "Weather-related data" refers to data relating to weather and meteorological conditions, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[1202] "Preprocessing" refers to a series of processes performed on acquired weather-related data, such as data normalization, missing value completion, and noise removal.

[1203] "Missing value imputation" is the process of filling in missing data points in a dataset using surrounding data.

[1204] "Noise removal" refers to the process of filtering out abnormal values ​​and unnecessary data from data.

[1205] A "machine learning model" is an algorithm for predicting future data based on past data, and primarily uses statistical methods.

[1206] A "deep learning model" is an algorithm that uses neural networks to analyze data in a multi-layered manner and make advanced predictions.

[1207] "Real-time updating" refers to the process of collecting data at regular intervals and periodically replacing the analysis results with the latest ones.

[1208] "Emotional state" refers to the psychological and emotional state of the user as analyzed from their voice, facial expressions, text input, etc.

[1209] "Notifications" are messages or alerts that provide information about specific weather conditions based on the user's preferences and emotional state.

[1210] A "terminal" is a device (e.g., smartphone, tablet, computer) that a user uses to access and operate the system.

[1211] "Detailed weather forecast information" is weather forecast data that goes beyond simple weather forecasts and includes additional information such as weather radar, typhoon tracks, and pollen dispersion information.

[1212] The present invention is a weather forecasting system that retrieves weather-related data from an external database, preprocesses it to fill in missing values, and removes noise. The system also incorporates an emotion engine that recognizes the user's emotions. The system updates the weather forecast in real time based on the analysis results and notifies the user when specific weather conditions are predicted based on the user's settings. The system also allows the user to view detailed weather forecast information via their device.

[1213] The server periodically retrieves weather-related data (e.g., temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database. This data is fetched via an API. For example, the data could be retrieved using the OpenWeatherMap API.

[1214] The server preprocesses the collected weather-related data. Specific preprocessing steps include normalizing the data, filling in missing values, and removing noise. For example, the Pandas library is used to fill in NaN values ​​with approximate values, and the Scipy library is used to apply a low-pass filter to smooth the data.

[1215] The server uses the preprocessed data to analyze weather forecasts using machine learning or deep learning models. It uses a long short-term memory (LSTM) network for short-term predictions and an ensemble learning model for long-term predictions. Specific implementations could use frameworks such as TensorFlow or PyTorch.

[1216] The analysis results are stored in a database by the server and updated in real time. New data is collected at regular intervals (for example, every 30 minutes), and the predictive model is re-run to generate the latest prediction results. The database can be a relational database such as MySQL or PostgreSQL.

[1217] The server uses an emotion engine to recognize the user's emotions. This engine analyzes the user's emotional state based on voice, facial expressions, and text input. For example, facial expressions can be captured using OpenCV and analyzed using a deep learning model, or voice data can be converted into text using a speech recognition API such as IBM Watson, and emotion analysis can then be performed.

[1218] The device sends notifications that take into account the user's emotional state based on specific weather conditions set by the user (e.g., thunderstorms, typhoons, pollen counts, etc.). A real-time notification function is implemented using JavaScript, etc., and notifications are sent with wording that reflects the user's emotional state. For example, if the user is feeling stressed, a detailed and kind notification will be sent.

[1219] Users can access detailed weather forecast information by opening the app on their devices. The app displays the latest forecast data, weather radar, typhoon tracks, pollen count information, etc. Furthermore, the priority and format of the displayed information is automatically optimized based on the user's emotional state.

[1220] Specific examples

[1221] For example, on a certain day in Tokyo, the server retrieves the latest data from the weather database, and after preprocessing, analyzes it using an LSTM model, predicting a thunderstorm in one hour. The analysis results are stored in the database and updated in real time. The server then analyzes the user's stress level using an emotion engine and creates a notification based on that state. The device receives this notification and sends it to the user with detailed advice such as, "Thunderstorms are predicted. Avoid sudden outings or bring an umbrella." The user can then change their plans based on this notification and take appropriate precautions against the thunderstorm.

[1222] Examples of prompt statements

[1223] "Describe a scenario where a server retrieves the latest weather data from an external database and predicts a thunderstorm in the next hour in Tokyo on a certain day. The emotion engine detects the user's stress state and sends a notification with detailed advice."

[1224] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1225] Step 1: Data collection

[1226] The server periodically retrieves weather-related data from an external database. Specifically, the server sends an API request and receives a JSON-formatted response from the external database. This input data includes temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. As output, this weather-related data is obtained and passed to the next preprocessing step.

[1227] Example of how it works:

[1228] The server sends an API request every hour.

[1229] Receive the API response (weather data JSON) and save it in local storage.

[1230] Step 2: Data Preprocessing

[1231] The server preprocesses the collected weather-related data. This preprocessing involves normalizing the data, filling in missing values, and removing noise. The input is the raw data collected in step 1, and the output is preprocessed, high-quality data.

[1232] Example of how it works:

[1233] Use the Pandas library to read the JSON data and convert it to a dataframe.

[1234] To impute missing values, fill NaN values ​​in the data frame with the average of the preceding and following data points.

[1235] Apply a low-pass filter using the Scipy library to remove noise.

[1236] Step 3: Data analysis

[1237] The server uses the preprocessed data to analyze the weather forecast with a machine learning or deep learning model, where the input is the preprocessed data from step 2 and the output is the weather forecast data.

[1238] Example of how it works:

[1239] Use TensorFlow or PyTorch to load the LSTM model and run predictions using the preprocessed data as input.

[1240] Short-term predictions (1 to 24 hours) are performed using LSTM, and long-term predictions (1 week) are performed using an ensemble learning model.

[1241] Evaluate the analysis results and record performance metrics.

[1242] Step 4: Save to database and update in real time

[1243] The analysis results are stored in a database by the server and updated at regular intervals. The input is the analysis result from step 3, and the output is the prediction results stored in the database.

[1244] Example of how it works:

[1245] Connect to a MySQL or PostgreSQL database and insert the analysis results into a table.

[1246] Every 30 minutes, start with a new data collection and rerun the preprocessing and analysis.

[1247] Delete the old prediction results in the database and update with the new results.

[1248] Step 5: Emotion analysis using the emotion engine

[1249] The server uses an emotion engine to recognize the user's emotions. The input is the user's voice, facial expression, text input, etc., and the output is the user's emotional state.

[1250] Example of how it works:

[1251] Facial expressions are captured from a webcam using OpenCV and analyzed using a deep learning model.

[1252] Voice data is converted into text through a voice recognition API and sentiment analysis is performed.

[1253] Text analysis algorithms are used to recognize emotions from messages and comments.

[1254] Step 6: User Notification Settings

[1255] The device sends notifications based on specific weather conditions set by the user, taking into account their emotional state. The input is the data from steps 4 and 5, and the output is the notification sent to the user.

[1256] Example of how it works:

[1257] Check the user's settings and generate notifications when certain weather conditions (such as thunderstorms, typhoons, and pollen counts) are predicted.

[1258] Send real-time notifications to your device using JavaScript.

[1259] If the user is feeling stressed, a detailed notification containing kind words is generated.

[1260] Step 7: Information Reference

[1261] The user opens the application on the device and views detailed weather forecast information. The input is the forecast result stored in the database in step 4, and the output is the weather forecast information displayed on the device.

[1262] Example of how it works:

[1263] Open a mobile or web application.

[1264] The app screen displays the latest weather forecast, weather radar, typhoon path, pollen information, and more.

[1265] The information display format is automatically adjusted according to the user's emotional state.

[1266] (Application example 2)

[1267] 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."

[1268] In today's world, highly accurate weather forecasts have become important information in many fields. However, simply providing weather information makes it difficult to present users with measures appropriate to their individual needs and situations. In particular, physical stores require inventory management that takes weather conditions into account and appropriate product recommendations to customers. Furthermore, services that do not take into account user emotions make it difficult to improve customer satisfaction.

[1269] Therefore, the objective of the present invention is to provide a system that combines weather data and a user's emotional state to optimize inventory management and promotion notifications in physical stores and meet individual needs.

[1270] 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.

[1271] In this invention, the server includes means for acquiring weather-related data from an external database, means for preprocessing the weather-related data to fill in missing values ​​and remove noise, means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data, means for storing the analysis results in a database and updating them in real time, means for notifying the user when specific weather conditions are predicted based on user settings, means for the user to refer to detailed weather forecast information via a terminal, means for performing demand forecasts based on the weather data and optimizing inventory in physical stores, and means for sending personalized promotional notifications according to the user's emotional state. This enables optimal inventory management in response to weather conditions and individual promotional notifications that take the user's emotional state into consideration.

[1272] "Weather-related data" refers to data relating to weather and meteorological conditions, such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, and pollen dispersion.

[1273] "Preprocessing" refers to the process of preparing data in an appropriate form before data analysis, such as normalizing data, filling in missing values, and removing noise.

[1274] A "machine learning model" is a model that learns patterns and regularities based on data and makes predictions and classifications.

[1275] A "deep learning model" is a model that uses a multi-layer neural network that can learn complex data patterns.

[1276] "Weather forecast" is information that predicts future weather conditions based on acquired weather data.

[1277] "Real-time updates" means obtaining new data and updating the analysis results based on that data at any time to provide the latest information.

[1278] The "notification means" is a means for notifying the user of information when a specific condition is met.

[1279] A "brick and mortar store" is a retail or service establishment that exists in a physical location.

[1280] "Inventory management" refers to the proper management and adjustment of the quantity of goods and resources, and maintaining an optimal inventory level according to demand.

[1281] A "promotional notice" is a notice to inform users of information about a particular product or service as part of an advertisement or promotion.

[1282] "Emotional state" refers to the user's emotional state, such as stress, joy, or sadness.

[1283] "Personalized" means providing information and services tailored to the specific needs and circumstances of each individual user.

[1284] This system acquires weather-related data from an external database, performs missing value imputation and noise removal, and then generates highly accurate weather forecasts using machine learning and deep learning models. It also analyzes the user's emotional state and provides personalized notifications and information based on that.

[1285] Data collection and preprocessing

[1286] The server periodically retrieves weather-related data (temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc.) from an external database (e.g., a weather API). This data is fetched via the API. The server then preprocesses the collected weather data, imputes missing values, and removes noise. Specifically, it uses Scikit-learn's SimpleImputer to impute missing data and StandardScaler to normalize the data.

[1287] Weather forecast analysis

[1288] Using the preprocessed weather data, the server analyzes the weather forecast using machine learning models (e.g., LSTM models) or deep learning models. LSTM (long short-term memory) networks are used for short-term predictions, while ensemble learning models are used for long-term predictions. This results in a detailed weather forecast (temperature, probability of precipitation, wind speed, wind direction, thunderstorm forecast, and typhoon path). The analysis results are stored in a database and updated in real time.

[1289] Sentiment Analysis and Notifications

[1290] The server implements an emotion analysis engine to analyze the user's emotional state. This engine analyzes the user's voice, facial expressions, text input, etc. to determine the user's emotional state. For example, it analyzes text input using an emotion recognition library such as Emojify.

[1291] Based on the results of sentiment analysis, notifications are sent when certain weather conditions are predicted. For example, if a thunderstorm is predicted and the user is feeling stressed, a personalized notification with detailed advice is sent. The notification may include specific advice such as, "Thunderstorms are predicted. Avoid sudden outings and, if necessary, check out stress-relieving products."

[1292] Inventory management and promotion notifications

[1293] The server uses weather data to predict demand and optimize inventory in physical stores. For example, if rain is predicted, it increases the stock of umbrellas and raincoats. It also sends personalized promotional notifications based on the user's emotional state. For example, if it determines that the user needs relaxation, it notifies them of promotions for relaxation-related products.

[1294] Information reference

[1295] Users can open the application on their devices to view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, and other details. The priority and format of the displayed information is automatically optimized based on the user's emotional state.

[1296] Specific examples

[1297] For example, if a thunderstorm is predicted for Tokyo on a certain day and the user is feeling stressed, the following notification will be displayed: "Thunderstorms are predicted. Avoid sudden outings and, if necessary, check for products to relieve stress."

[1298] Prompt Sentence Examples

[1299] "Based on data obtained from a weather data API, impute and normalize missing values ​​and use an LSTM model to predict the weather. Next, analyze the user's emotional state from input and generate a promotional notification for relaxation-related products if the user is feeling stressed when a thunderstorm is predicted."

[1300] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1301] Step 1:

[1302] The server retrieves weather-related data from an external database through an API, which contains information such as temperature, humidity, wind speed, wind direction, precipitation, cloud cover, pollen count, etc. The input is the API endpoint, and the output is the weather-related data.

[1303] Step 2:

[1304] The server preprocesses the acquired weather-related data. Specifically, it uses SimpleImputer to impute missing values ​​and StandardScaler to normalize the data. This improves the quality of the data and makes it suitable for analysis. The input is the acquired weather-related data, and the output is the preprocessed data.

[1305] Step 3:

[1306] The server uses the preprocessed data to analyze the weather forecast using a machine learning or deep learning model. It uses an LSTM model to perform short-term predictions and an ensemble learning model to perform long-term predictions. The input is the preprocessed data, and the output is the weather forecast results.

[1307] Step 4:

[1308] The server stores the weather forecast results in a database and updates it in real time. Every 30 minutes, it collects new data and re-runs the model to generate the latest forecast results. The input is the weather forecast results, and the output is the updated database.

[1309] Step 5:

[1310] The server uses an emotion analysis engine to analyze the user's emotional state. It determines the emotional state from the user's voice, facial expression, text input, etc. When analyzing text input using tools such as Emojify, the input is the text from the user and the output is the user's emotional state.

[1311] Step 6:

[1312] The server notifies users when certain weather conditions are predicted based on the emotion analysis results and weather forecast results. Specifically, if a thunderstorm is predicted and the user is feeling stressed, the server generates a personalized notification with detailed advice. The input is the weather forecast result and the user's emotional state, and the output is the notification content.

[1313] Step 7:

[1314] The server predicts demand based on weather data and optimizes inventory in physical stores. If rain is predicted, it will increase the stock of umbrellas and raincoats, and perform optimal inventory management. The input is the weather forecast result, and the output is inventory management instructions.

[1315] Step 8:

[1316] The server sends personalized promotion notifications according to the user's emotional state. For example, if the server determines that the user needs relaxation, it notifies the user of promotions for relaxation-related products. The input is the user's emotional state, and the output is the promotion notification.

[1317] Step 9:

[1318] Users can open the application on their devices and view detailed weather forecast information, including the latest forecast data, weather radar, typhoon tracks, pollen count information, etc. The input is launching the application, and the output is the displayed weather forecast information.

[1319] 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.

[1320] 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.

[1321] 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.

[1322] 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.

[1323] 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.

[1324] 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.

[1325] 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).

[1326] 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.

[1327] 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."

[1328] 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.

[1329] 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).

[1330] 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.

[1331] 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.

[1332] 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.

[1333] 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.

[1334] 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.

[1335] 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.

[1336] 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.

[1337] 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.

[1338] 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.

[1339] 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.

[1340] The following is further disclosed regarding the above embodiment.

[1341] (Claim 1)

[1342] means for obtaining weather-related data from an external database;

[1343] means for preprocessing the weather-related data to fill in missing values ​​and remove noise;

[1344] means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data;

[1345] a means for storing the analysis results in a database and updating the database in real time;

[1346] means for notifying the user when certain weather conditions are predicted based on the user's settings;

[1347] A means for a user to refer to detailed weather forecast information through a terminal;

[1348] A system including:

[1349] (Claim 2)

[1350] The system of claim 1 , further comprising: generating short-term and long-term weather forecasts based on the acquired weather-related data.

[1351] (Claim 3)

[1352] 2. The system according to claim 1, wherein the notification means based on the user's settings notifies the user of thunderstorms, typhoons, and pollen dispersion conditions in a specific area.

[1353] "Example 1"

[1354] (Claim 1)

[1355] means for obtaining weather-related data from an external database;

[1356] means for preprocessing the weather-related data to fill in missing values ​​and remove noise;

[1357] means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data;

[1358] a means for storing the analysis results in a database and updating the database in real time;

[1359] means for notifying the user when certain weather conditions are predicted based on the user's settings;

[1360] A means for a user to refer to detailed weather forecast information through a terminal;

[1361] The data collection is performed using an API, and the collected data is preprocessed by applying missing value completion and a noise removal filter;

[1362] means for using a recurrent neural network and an ensemble learning model to analyze the weather forecast;

[1363] means for using a relational database management system to store and update said data in a database in real time;

[1364] The notification means includes a means for using a cloud messaging service;

[1365] A system including:

[1366] (Claim 2)

[1367] The system of claim 1 , further comprising: generating short-term and long-term weather forecasts based on the acquired weather-related data.

[1368] (Claim 3)

[1369] 2. The system according to claim 1, wherein the notification means based on the user's settings notifies the user of thunderstorms, typhoons, and pollen dispersion conditions in a specific area.

[1370] "Application Example 1"

[1371] (Claim 1)

[1372] means for obtaining weather-related data from an external database;

[1373] means for preprocessing the weather-related data to fill in missing values ​​and remove noise;

[1374] means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data;

[1375] a means for storing the analysis results in a database and updating the database in real time;

[1376] means for notifying the user when certain weather conditions are predicted based on the user's settings;

[1377] A means for a user to refer to detailed weather forecast information through a terminal;

[1378] means for making specific suggestions that influence user behavior based on the weather forecast information;

[1379] A system including:

[1380] (Claim 2)

[1381] The system of claim 1 , further comprising: generating short-term and long-term weather forecasts based on the acquired weather-related data.

[1382] (Claim 3)

[1383] The system according to claim 1, wherein the notification means based on the user's settings predicts demand for food delivery services based on weather conditions and notifies the user of an optimal delivery plan.

[1384] "Example 2: Combining Emotion Engines"

[1385] (Claim 1)

[1386] means for obtaining weather-related data from an external database;

[1387] means for preprocessing the weather-related data to fill in missing values ​​and remove noise;

[1388] means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data;

[1389] a means for storing the analysis results in a database and updating the database in real time;

[1390] means for recognizing and analyzing the emotional state of a user;

[1391] means for providing notifications according to specific weather conditions and the user's emotional state based on user settings;

[1392] A means for a user to refer to detailed weather forecast information through a terminal;

[1393] A system including:

[1394] (Claim 2)

[1395] The system of claim 1 , further comprising: generating short-term and long-term weather forecasts based on the acquired weather-related data.

[1396] (Claim 3)

[1397] 2. The system according to claim 1, wherein the notification means based on the user's settings notifies the user of thunderstorms, typhoons, and pollen dispersion conditions in a specific area.

[1398] "Application example 2 when combining emotion engines"

[1399] (Claim 1)

[1400] means for obtaining weather-related data from an external database;

[1401] means for preprocessing the weather-related data to fill in missing values ​​and remove noise;

[1402] means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data;

[1403] a means for storing the analysis results in a database and updating the database in real time;

[1404] means for notifying the user when certain weather conditions are predicted based on the user's settings;

[1405] A means for a user to refer to detailed weather forecast information through a terminal;

[1406] A means of forecasting demand based on weather data and optimizing inventory in physical stores;

[1407] means for sending personalized promotional notifications according to the emotional state of the user;

[1408] A system including:

[1409] (Claim 2)

[1410] The system of claim 1 , further comprising: generating short-term and long-term weather forecasts based on the acquired weather-related data.

[1411] (Claim 3)

[1412] 2. The system according to claim 1, wherein the notification means based on the user's settings notifies the user of thunderstorms, typhoons, and pollen dispersion conditions in a specific area. [Explanation of symbols]

[1413] 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. means for obtaining weather-related data from an external database; means for preprocessing the weather-related data to fill in missing values ​​and remove noise; means for analyzing weather forecasts using a machine learning model or a deep learning model using the preprocessed data; a means for storing the analysis results in a database and updating the database in real time; means for notifying the user when certain weather conditions are predicted based on the user's settings; A means for a user to refer to detailed weather forecast information through a terminal; A system including:

2. The system of claim 1 , further comprising: generating short-term and long-term weather forecasts based on the acquired weather-related data.

3. The system according to claim 1, wherein the notification means based on the user's settings notifies the user of thunderstorms, typhoons, and pollen dispersion conditions in a specific area.

Images