System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting AI from a real-time sensor in real time; means for performing anomaly detection based on the collected information; means for predicting a future problem or failure based on the anomaly detection result; means for automatically generating a maintenance plan for the predicted problem; and means for notifying the generated maintenance plan.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Current payment systems are primarily maintained manually, and typically only respond after a problem or failure occurs. This approach does not prevent problems from occurring, and there is a risk of the system's reliability decreasing. Furthermore, responses to problems tend to be delayed when they occur, which could have a significant impact on users. To solve these issues, it is necessary to predict problems and failures in advance and carry out appropriate maintenance. [Means for solving the problem]

[0005] In order to improve the reliability of the payment system, the present invention provides the following means.

[0006] First, a means for collecting data from AI sensors in real time is provided. Next, a means for detecting anomalies based on the collected data is provided. By providing a means for predicting future problems and failures based on the results of this anomaly detection, a means for automatically generating maintenance plans for predicted problems is provided. Finally, the system includes a means for notifying the user of the generated maintenance plan.

[0007] Furthermore, the system further includes a means for analyzing past data to identify patterns, a means for updating the prediction model, and a means for saving the maintenance progress status, thereby improving the accuracy and reliability of the prediction.The system also includes a means for encouraging users to perform preventive maintenance based on the generated maintenance plan, and a means for improving the prediction model by feeding back the results of the preventive maintenance, thereby optimizing the maintenance process throughout the system and realizing a highly reliable payment system.

[0008] An "AI sensor" is a sensor device that collects data in real time and performs analysis based on AI algorithms, such as detecting anomalies.

[0009] "Real-time data" is data that represents ongoing processes or states and is collected immediately.

[0010] "Anomaly detection" is the process of analyzing collected data to identify unusual conditions or patterns that deviate from the norm.

[0011] A "predictive model" is a mathematical or statistical model used to predict future conditions or events based on past and current data.

[0012] A "maintenance plan" is a plan that defines the schedule and specific content of preventive maintenance work to be carried out with the aim of improving system reliability.

[0013] "User notification" is a means by which a system or terminal communicates important information or warnings to the user.

[0014] "Data analysis" is the process of processing collected data and extracting useful information and patterns.

[0015] "Pattern identification" is the process of discovering specific recurrences or trends by analyzing past data.

[0016] "Feedback" refers to returning the actual work performed and its results to the system, providing information that can be used for future improvements and optimization.

[0017] "Preventive maintenance" is the practice of carrying out planned maintenance work in advance to prevent anticipated problems or failures in a system. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0026] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0039] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, and automatically generates appropriate maintenance plans, thereby improving the reliability of payment systems.

[0040] Program processing

[0041] Data collection

[0042] The server receives real-time data from AI sensors, such as the number of transactions, response time, CPU usage, and memory usage, which are continuously collected.

[0043] The server stores the collected data in a database and backs it up as appropriate.

[0044] Data analysis

[0045] The server analyzes the data received in real time, removes noise, and then runs the analyzed data through an AI algorithm to detect anomalies.

[0046] If an anomaly is detected, the server logs it and generates an alert if necessary.

[0047] Update the forecast model

[0048] The server uses the data to update the forecasting model, taking into account current trends and seasonal fluctuations.

[0049] The server retrains the machine learning algorithm to improve prediction accuracy.

[0050] Preventive maintenance plan development

[0051] The server assesses future risks based on predictive models and automatically generates appropriate preventative maintenance plans based on the assessment results.

[0052] The terminal notifies the user of the generated preventive maintenance plan and informs the user when it should be carried out.

[0053] Maintenance

[0054] Users receive notifications from their devices and can perform preventative maintenance, such as rebooting the system, applying patches, or replacing hardware.

[0055] The terminal reports the progress of the performed maintenance to the server.

[0056] Feedback and learning from results

[0057] The server analyzes the results of the maintenance performed by the user and feeds the data back into the predictive model.

[0058] The server uses the feedback results to further improve the predictive model, which further optimizes the next preventive maintenance plan.

[0059] Specific examples

[0060] Case 1: Real-time anomaly detection

[0061] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[0062] The server compares this with past data and determines that this sudden increase is abnormal.

[0063] The terminal notifies the user of the abnormality and prompts them to implement system load balancing and traffic restrictions.

[0064] Users are notified and can take corrective action to maintain stable system operation.

[0065] Case 2: Predictive maintenance planning

[0066] The server uses past data to discover trends in the failure of specific components.

[0067] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[0068] The user is notified and replaces the part to prevent a breakdown.

[0069] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[0070] In this way, the system of the present invention utilizes AI sensors and data analysis to achieve preventive maintenance and improved reliability of the payment system.

[0071] The processing flow will be explained below.

[0072] Step 1:

[0073] The server receives real-time data from the AI ​​sensor. Specifically, it accesses the sensor through an API and collects data such as the number of transactions, response time, CPU usage, and memory usage. This data is periodically stored in a database.

[0074] Step 2:

[0075] The server pre-processes the collected data and removes noise, specifically by using filtering algorithms to correct and remove outliers and missing data, which improves the reliability of the data.

[0076] Step 3:

[0077] The server runs anomaly detection algorithms on the pre-processed data, for example comparing it with past normal data patterns to determine whether the current data is anomalous, and generates an alert if an anomaly is detected.

[0078] Step 4:

[0079] The server uses the anomaly detection results to predict future problems and outages. Specifically, it uses machine learning models to build predictive models from past data and apply them to real-time data to predict potential problems.

[0080] Step 5:

[0081] The server automatically generates maintenance plans for predicted problems, creating maintenance schedules that suggest specific actions, such as replacing specific parts or rebooting the system.

[0082] Step 6:

[0083] The terminal notifies the user of the maintenance plan received from the server. The notification includes the content, timing, and implementation method of the required maintenance. The user receives the notification and prepares for the maintenance.

[0084] Step 7:

[0085] Based on notifications from the terminal, the user performs preventive maintenance work, such as rebooting the system, applying patches, and replacing hardware. Once the work is completed, the results are reported to the terminal.

[0086] Step 8:

[0087] The terminal reports the progress and results of the maintenance work performed by the user to the server, including the work performed, the date and time of the work, and the results.

[0088] Step 9:

[0089] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[0090] Step 10:

[0091] The server uses the feedback results to further optimize predictive models and maintenance plans, thereby preventing future problems and continuously improving system reliability.

[0092] Example 1

[0093] 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."

[0094] Conventional systems lacked real-time data collection and anomaly detection, resulting in numerous limitations in the accuracy of predictive models and the automatic generation of maintenance plans. In particular, they lacked noise removal, consideration of trend fluctuations, and feedback on the status of preventive maintenance, which ultimately hindered efforts to improve the reliability of payment systems. Furthermore, inappropriate notification timing and alert generation for preventive maintenance plans hindered efficient maintenance.

[0095] 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.

[0096] In this invention, the server includes means for collecting data from the AI ​​sensor in real time, means for removing noise and detecting anomalies, means for recording the anomaly detection results in a log and generating alerts, means for analyzing past data and identifying patterns, means for updating the prediction model to take into account the latest trends and seasonal fluctuations, means for saving the maintenance progress status and analyzing the results, and means for feeding back the analysis results to improve the prediction model, thereby enabling real-time anomaly detection, improved prediction accuracy, and automatic generation and notification of efficient preventive maintenance plans.

[0097] An "AI sensor" is a sensor that uses machine learning algorithms to collect data from the environment in real time and analyze that data.

[0098] "Data collection" is the process in which the server acquires data from the AI ​​sensor in real time and stores it in a database or similar.

[0099] "Denoising" is the process of applying a filtering algorithm to remove unwanted noise from collected data.

[0100] Anomaly detection is the process of analyzing collected and denoised data to find patterns or signs that are out of the ordinary.

[0101] "Alert generation" is the process of creating an alert to notify a user when an anomaly is detected.

[0102] A "predictive model" is a machine learning algorithm or statistical model that uses historical data to forecast future movements and trends.

[0103] "Pattern analysis" is the process of analyzing historical data to find recurring rules or characteristics.

[0104] "Trend consideration" is the process of incorporating recent trends and seasonal fluctuations when updating a forecasting model.

[0105] "Maintenance progress" is information that records and tracks the implementation status of maintenance work such as preventive maintenance and repairs.

[0106] "Feedback" is the process of analyzing maintenance results and using that information to improve predictive models.

[0107] A "preventive maintenance plan" is a plan that automatically generates schedules and procedures for maintenance to be carried out in advance to prevent abnormalities and breakdowns.

[0108] "Notification" is the process by which users are informed of generated preventative maintenance plans and alerts.

[0109] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, and automatically generates appropriate preventive maintenance plans.

[0110] The server first has a means to obtain data such as the number of transactions, response time, CPU usage, and memory usage from the AI ​​sensor in real time. Specifically, the server periodically sends API requests to collect the latest data from the sensor. The collected data is stored in a database such as MySQL or PostgreSQL and is backed up as appropriate.

[0111] Next, the server analyzes the collected data and removes noise. This is done using algorithms such as moving average filters and low-pass filters. The noise-removed data is then analyzed using an AI algorithm to detect anomalies. For example, an LSTM (Long Short-Term Memory) anomaly detection algorithm is used. If an anomaly is detected, the server records the information in a log and notifies the user by sending an email or generating an alert via WebSocket.

[0112] The server then updates the predictive model based on abnormal data. This is done using machine learning libraries such as TensorFlow and PyTorch. The predictive model is updated to take into account the latest trends and seasonal fluctuations. Based on this updated predictive model, the server evaluates future risks and automatically generates appropriate preventive maintenance plans. For example, it uses an automatic scheduling algorithm to determine the timing of maintenance. The generated preventive maintenance plan is then notified to the user via their terminal, informing them of the timing of implementation.

[0113] During the preventive maintenance phase, users receive notifications from their devices and perform preventive maintenance such as rebooting the system, applying patches, and replacing hardware. The devices then report the progress of the maintenance to the server, which then analyzes the results of the user's maintenance and feeds the data back into the predictive model. Based on this feedback, the predictive model is further improved, and the next preventive maintenance plan is generated more precisely.

[0114] Specific examples

[0115] Case 1: Real-time anomaly detection

[0116] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[0117] The server compares this with past data and determines that this sudden increase is abnormal.

[0118] The terminal notifies the user of the abnormality and prompts them to implement system load balancing and traffic restrictions.

[0119] Users are notified and can take corrective action to maintain stable system operation.

[0120] Case 2: Predictive maintenance planning

[0121] The server uses past data to discover trends in the failure of specific components.

[0122] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[0123] The user is notified and replaces the part to prevent a breakdown.

[0124] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[0125] Prompt Sentence Examples

[0126] Prompt example 1: Real-time anomaly detection

[0127] Create a program that detects sudden spikes in the number of transactions obtained from the AI ​​sensor and compares the situation with past data to determine whether it is an abnormality. If an abnormality is detected, notify the user and include content that encourages load balancing or traffic restrictions.

[0128] Example prompt 2: Predictive maintenance plan

[0129] Please generate a program that analyzes the failure trends of specific parts based on past data and notifies the user when the next failure is predicted. Also, please include a program that suggests parts replacement in advance to the user and provides feedback on the results to improve the prediction model.

[0130] As described above, this system utilizes AI sensors and data analysis to achieve preventive maintenance and improve the reliability of the payment system.

[0131] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0132] Step 1: Data Acquisition

[0133] The server receives data from the AI ​​sensors in real time. Specifically, the server periodically sends API requests to collect data from the sensors, such as response time, CPU usage, memory usage, and number of transactions.

[0134] Input: Data from AI sensors

[0135] Output: Real-time data captured

[0136] Step 2: Save data

[0137] The server stores the collected data in a database and performs regular backups. Specifically, data is inserted using SQL queries using MySQL or PostgreSQL, and the data is saved periodically using a backup script.

[0138] Input: Real-time data captured

[0139] Output: Data stored in the database

[0140] Step 3: Noise reduction

[0141] The server removes noise from the collected data by applying a moving average filter or low-pass filter to remove unnecessary data.

[0142] Input: Data stored in the database

[0143] Output: Denoised data

[0144] Step 4: Anomaly detection

[0145] The server then feeds the denoised data into an AI algorithm to detect anomalies, for example using a Long Short-Term Memory (LSTM) algorithm to detect anomalous patterns.

[0146] Input: Denoised data

[0147] Output: Detected anomalies

[0148] Step 5: Alert Generation

[0149] The server logs any detected anomalies and generates alerts if necessary, which are sent to users in real-time via email and / or WebSocket notifications.

[0150] Input: Detected anomaly

[0151] Output: Log entries, alert notifications

[0152] Step 6: Update the predictive model

[0153] The server uses anomalies and historical data to update the predictive model, specifically by retraining the machine learning algorithm using TensorFlow or PyTorch.

[0154] Input: Anomaly data, historical data

[0155] Output: Updated forecast model

[0156] Step 7: Consider the trend

[0157] The server modifies the forecast model to take into account recent trends and seasonal fluctuations, specifically including semi-annual and annual data in the forecast model.

[0158] Input: Updated forecast model

[0159] Output: A revised forecast model

[0160] Step 8: Risk assessment and maintenance plan generation

[0161] The server evaluates future risks based on the modified predictive model, automatically generates appropriate preventive maintenance plans, and determines maintenance timing using an automatic scheduling algorithm.

[0162] Input: Modified forecast model

[0163] Output: Risk assessment results, preventive maintenance plan

[0164] Step 9: Notification

[0165] The device will notify the user of the generated preventive maintenance plan and notify them when it is time to implement it, using a mobile app or desktop notification.

[0166] Input: Preventive Maintenance Plan

[0167] Output: User notification

[0168] Step 10: Perform preventative maintenance

[0169] Users receive notifications from their devices and can then perform preventative maintenance, such as rebooting the system, applying patches, or replacing hardware.

[0170] Input: Notification

[0171] Output: Preventive maintenance performed

[0172] Step 11: Progress Report

[0173] The terminal reports the progress of the performed maintenance to the server. For example, after the maintenance is completed, the terminal presses a confirmation button to send a completion status to the server.

[0174] Input: Preventive maintenance performed

[0175] Output: progress report

[0176] Step 12: Maintenance results analysis and feedback

[0177] The server analyzes the results of the maintenance performed by the user and feeds that data back into the predictive model. Specifically, it compares performance data before and after the maintenance to evaluate the effectiveness, and adds the analysis results to the dataset for retraining.

[0178] Input: Progress report

[0179] Output: Feedback data

[0180] Step 13: Model Improvement

[0181] The server improves the predictive model based on the feedback results, and ultimately the improved predictive model generates a more accurate next preventive maintenance plan.

[0182] Input: Feedback data

[0183] Output: An improved predictive model

[0184] (Application example 1)

[0185] 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."

[0186] Maintaining system reliability and stability is essential for modern electronic payment services. However, conventional monitoring systems are slow to detect anomalies and are unable to quickly generate appropriate maintenance plans. Furthermore, there are limited means of properly notifying users about the timing and methods of preventive maintenance, making it difficult to completely eliminate the risk of breakdowns and service outages.

[0187] 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.

[0188] In this invention, the server includes means for collecting data from an AI sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the results of the anomaly detection, means for automatically generating a maintenance plan for the predicted problems, means for notifying the user of the generated maintenance plan, and means for notifying the user of the anomaly and presenting a maintenance procedure via a smart device. This makes it possible to significantly improve the reliability and stability of payment services by detecting system anomalies early and immediately implementing appropriate preventive maintenance.

[0189] An "AI sensor" is a sensor equipped with artificial intelligence that monitors the status of the environment and system in real time and collects data.

[0190] "Data collection" is the process of compiling information from a system or environment using sensors or other data acquisition methods and storing it for analysis and processing.

[0191] "Anomaly detection" is the process of identifying unusual patterns and values ​​based on collected data and finding signs of system abnormalities or problems.

[0192] "Predicting future problems and failures" is a method of estimating possible future problems and system failures based on past data and current anomaly detection results, and taking measures in advance.

[0193] "Automatic generation of maintenance plans" is the process of automatically formulating specific maintenance procedures and schedules for predicted problems or failures.

[0194] "Notification" is an information transmission means for informing the user of abnormalities and the results of the maintenance plan generation.

[0195] A "smart device" is a mobile terminal or wearable device that has the ability to connect to the Internet and is a device that provides information to users and receives notifications.

[0196] "Feedback" is the process of recording the results and effectiveness of maintenance and preventative care to help improve future predictive models and optimize the system.

[0197] A "machine learning algorithm" is an algorithm that automatically learns from data, finds patterns and relationships, and makes predictions and classifications.

[0198] "Preventive maintenance" refers to maintenance activities carried out in advance to prevent system failures.

[0199] The system for implementing this invention consists of three main components: an AI sensor, a server, and a user terminal (smart device).

[0200] Data collection

[0201] The server collects data in real time from AI sensors, such as the number of transactions, response time, CPU usage, and memory consumption of the payment system, and has built-in functionality to obtain data via sensors and APIs.

[0202] Data analysis

[0203] The server performs anomaly detection based on the collected data. To do this, the first step is to remove noise from the data. After the clean data is generated, anomalies are detected using machine learning algorithms (e.g., RandomForestClassifier). When an anomaly is detected, the information is saved as a log and an alert is generated if necessary.

[0204] Generate future forecasts and maintenance plans

[0205] The server predicts potential future problems and failures based on the anomaly detection results. By analyzing past data and continually updating the prediction model, trends and seasonal fluctuations are taken into account. For predicted problems, the server automatically generates an appropriate maintenance plan.

[0206] Notifications and Feedback

[0207] The generated maintenance plan is sent to the user's smart device (such as a smartphone or tablet). The user can receive this notification and carry out preventive maintenance on the system. Once the maintenance is complete, the results are fed back to the server and used to improve the accuracy of the predictive model.

[0208] Specific examples

[0209] For example, if a server analyzes real-time data and detects a sudden increase in transaction response time, it will recognize this as an anomaly. Based on this information, the server can predict future load increases or system failures and send notifications to smart devices such as "System load is increasing. Please perform maintenance." Users can receive these notifications and perform maintenance to maintain system stability.

[0210] An example of a prompt is as follows:

[0211] "Please implement a function that analyzes transaction data from payment systems, detects anomalies, and automatically generates appropriate maintenance plans."

[0212] In this way, the present invention builds a system that improves the reliability and stability of electronic payment services by linking AI sensors, servers, and smart devices.

[0213] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0214] Step 1: Data collection

[0215] The server collects data in real time from the AI ​​sensors, including the number of transactions, response time, CPU usage, memory usage, etc. The data is retrieved through an API and sent to the server, providing the raw data needed to accurately monitor the system's operating status.

[0216] Step 2: Save data

[0217] The server stores the collected data in a database for subsequent analysis and to update the predictive model. During this process, data is backed up regularly to ensure data reliability and integrity.

[0218] Step 3: Data analysis

[0219] The server analyzes the data and removes noise. Outliers and missing values ​​are filtered out during the data cleaning process. After that, anomalies are detected using a machine learning algorithm (e.g., RandomForestClassifier) ​​based on the clean data. The clean data is given as input, and the output is the presence or absence of anomalies.

[0220] Step 4: Log anomalies and generate alerts

[0221] If an abnormality is detected, the server records the details in a log and generates an alert if necessary. The log stores detailed information about the abnormality and the time of occurrence. The alert is used for notification in the next step.

[0222] Step 5: Auto-generate maintenance plans

[0223] The server predicts future problems and failures based on the anomaly detection results. This involves statistical trend analysis that takes into account past data and current conditions. Detailed maintenance work to address predicted problems is automatically planned. The maintenance plan is generated as a result of calculations using the predictive model.

[0224] Step 6: Notification

[0225] The server notifies the user of the generated maintenance plan via their smart device (smartphone or tablet), where a push notification is sent via the application to prompt the user to take action. The notification includes specific maintenance procedures and the timing of their implementation.

[0226] Step 7: User performs preventative maintenance

[0227] Users receive notifications and can then perform preventative maintenance as instructed, such as rebooting the system or applying software patches, with the actions taken directly fed back to the next step.

[0228] Step 8: Feedback

[0229] Users provide feedback on the results of their maintenance work to the server, which then uses this feedback to retrain the predictive model, improving the accuracy of future anomaly detection and maintenance planning.

[0230] Through the above steps, the present invention provides a system that improves the reliability and stability of electronic payment services.

[0231] 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.

[0232] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, automatically generates appropriate maintenance plans, and provides more personalized notifications to users by combining it with an emotion engine that recognizes user emotions. This not only improves the reliability of the payment system, but also increases user satisfaction.

[0233] Program processing

[0234] Data collection

[0235] The server receives real-time data from the AI ​​sensors, such as the number of transactions, response time, CPU usage, memory usage, etc. This data is stored in a database and backed up regularly.

[0236] Data analysis

[0237] The server pre-processes the collected data to remove noise and improve the reliability of the data by using filtering algorithms to correct and remove outliers and missing data.

[0238] The server then runs the pre-processed data through AI algorithms to detect anomalies, for example by comparing the current data with normal data patterns from the past to determine whether they are abnormal. If an anomaly is detected, an alert is generated.

[0239] Update the forecast model

[0240] The server updates the forecasting model based on the data, retraining the machine learning model to take into account current trends and seasonal fluctuations, improving forecast accuracy.

[0241] Preventive maintenance plan development

[0242] The server evaluates future risks based on predictive models, automatically generates a maintenance plan that proposes appropriate preventive maintenance work, and defines specific actions.

[0243] The terminal notifies the user of the generated preventive maintenance plan and notifies them of the timing of its implementation. This notification includes the specific content, timing, and method of the maintenance.

[0244] emotion recognition

[0245] Before sending a notification, the device uses an emotion engine to recognize the user's emotions, for example, by reading emotions from facial expressions, voice tone, and text.

[0246] The device personalizes notification messages based on the recognized emotion: if the user has a negative emotion, it provides a more polite and supportive message, and if the user has a positive emotion, it adds a message of thanks or encouragement.

[0247] Maintenance

[0248] Based on notifications from the terminal, the user performs preventive maintenance work, such as rebooting the system, applying patches, or replacing hardware. Once the work is completed, the results are reported to the terminal.

[0249] Feedback and learning from results

[0250] The terminal reports the progress and results of the maintenance work performed by the user to the server, including the work performed, the date and time of the work, and the results.

[0251] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[0252] Specific examples

[0253] Case 1: Real-time anomaly detection and emotion recognition

[0254] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[0255] The server compares this spike with historical data and determines that it is abnormal, and an alert is generated.

[0256] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. For example, if the user is feeling anxious, the notification message will include a message such as, "The system load is increasing rapidly, but please rest assured that we are taking measures to address this."

[0257] Users are notified and can take corrective action to maintain stable system operation.

[0258] Case 2: Predictive maintenance planning and emotion recognition

[0259] The server uses past data to discover trends in the failure of specific components.

[0260] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[0261] The device recognizes the user's emotions before sending a notification, and if the user is expressing positive emotions, it will include a thank-you message such as, "Thank you for your continued cooperation. Please replace parts in advance to ensure safe operation."

[0262] The user is notified and replaces the part to prevent a breakdown.

[0263] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[0264] As described above, the system of the present invention utilizes AI sensors and data analysis, and further combines it with an emotion engine to achieve preventative maintenance and improved reliability of the payment system, while also improving user satisfaction.

[0265] The processing flow will be explained below.

[0266] Step 1:

[0267] The server obtains data from the AI ​​sensors in real time. Specifically, it polls the sensors via API to collect data such as the number of transactions, response time, CPU usage, and memory usage. This data is recorded in a database and is backed up to always reflect the latest status.

[0268] Step 2:

[0269] The server pre-processes the collected data, applying filtering algorithms to correct or remove outliers and missing data, a process that improves the accuracy of the data used for analysis.

[0270] Step 3:

[0271] The server performs anomaly detection based on pre-processed data. It uses AI algorithms to compare the current data with past normal data patterns and determine whether it is anomalous. If an anomaly is detected, it is logged and an alert is generated if necessary.

[0272] Step 4:

[0273] The server predicts future problems and outages based on the anomaly detection results. It uses machine learning models to build predictive models from past data and apply them to real-time data. This prediction is used to assess the risk of system problems occurring.

[0274] Step 5:

[0275] The server automatically generates maintenance plans for predicted problems, creating a maintenance schedule with specific preventative maintenance actions such as part replacement needs and system reboots.

[0276] Step 6:

[0277] Before notifying the user of the maintenance plan received from the server, the device uses an emotion engine to recognize the user's emotions. Emotion recognition is performed through the camera, microphone, text analysis, etc.

[0278] Step 7:

[0279] The device customizes the maintenance notification message based on the user's recognized emotions. For example, if the user is expressing negative emotions, the notification message will include a thoughtful message such as "Don't worry, our technical team is working on it." On the other hand, if the user is expressing positive emotions, the notification message will include words of gratitude and encouragement.

[0280] Step 8:

[0281] Users can perform preventative maintenance tasks based on customized notifications from their devices, such as rebooting the system, applying patches, or replacing hardware, and the results are reported back to the device once the task is complete.

[0282] Step 9:

[0283] The terminal reports the progress and results of the maintenance work performed by the user to the server, including details of the work performed, the date and time of the work, and the results.

[0284] Step 10:

[0285] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[0286] Step 11:

[0287] The server uses the feedback results to further optimize predictive models and maintenance plans, thereby preventing future problems and continuously improving system reliability.

[0288] Example 2

[0289] 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."

[0290] In conventional systems, it is common to detect anomalies and perform preventive maintenance based on the results. However, more efficient and personalized notifications are required to further improve system reliability and user satisfaction. Furthermore, conventional systems have difficulty in providing notifications that take user emotions into account, which can lead to user dissatisfaction. Therefore, there is a need for a system that can detect anomalies in real time, automatically generate appropriate maintenance plans, and personalize notifications based on user emotions.

[0291] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting data from an AI sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the anomaly detection results, means for automatically generating a maintenance plan for the predicted problems, means for notifying the user of the generated maintenance plan, and means for recognizing the user's emotions and personalizing the notification message. This not only enables the system reliability to be improved by detecting anomalies in real time and automatically generating a preventive maintenance plan, but also improves user satisfaction by personalized notifications based on emotion recognition.

[0292] An "AI sensor" is a sensor that uses artificial intelligence technology to collect and analyze data in real time.

[0293] "Real-time" refers to data and information being processed and acquired the moment it is generated.

[0294] "Data collection means" means means for collecting data through sensors or other measuring devices.

[0295] An "anomaly detection means" is a means for detecting abnormal conditions or patterns based on collected data.

[0296] The "prediction means" is a means for predicting future problems or failures based on the anomaly detection results.

[0297] The "automatic maintenance plan generation means" is a means for automatically generating a maintenance plan for a predicted problem.

[0298] The "notification means" is a means for notifying the user of the generated maintenance plan.

[0299] "Emotion recognition means" refers to a means for recognizing and analyzing the user's emotions.

[0300] "Personalization" means customizing content and formats to suit individual users.

[0301] "Past data analysis means" is a means for analyzing data collected in the past and identifying patterns.

[0302] The "prediction model update means" is a means for updating the prediction model to the latest state based on the analyzed data.

[0303] The "maintenance progress storage means" is a means for storing the progress and results of maintenance work.

[0304] The "maintenance result feedback means" is a means for collecting the results of the performed maintenance and reflecting them in the prediction model.

[0305] The "alert generation means" is a means for generating a warning message when an abnormality is detected.

[0306] A "system" is the entire device or network that combines and operates the above means.

[0307] This invention is a system that detects abnormalities in real time and automatically generates appropriate maintenance plans by recognizing the user's emotions and personalizing notification messages.

[0308] The server uses AI sensors to collect data in real time, including the number of transactions, response time, CPU usage, and memory usage, which is then stored in a database and backed up regularly.

[0309] The collected data is pre-processed using a filtering algorithm. The server corrects and removes outliers and missing data to remove noise, improving the reliability of the data. The pre-processed data is compared with past normal data patterns. The server performs anomaly detection using a machine learning model (e.g., TensorFlow model) and generates an alert if an anomaly is detected.

[0310] Based on the anomaly detection results, the server predicts future problems and outages. The predictive model is updated with the latest data and retrained to take into account trends and seasonal fluctuations, improving the accuracy of the predictions.

[0311] The server automatically generates a maintenance plan that proposes preventive maintenance work for predicted problems. The generated maintenance plan includes the content and timing of specific maintenance work (e.g., system restart, part replacement, etc.). The terminal notifies the user of this maintenance plan and informs them of the timing of its implementation.

[0312] Before sending a notification, the device uses an emotion engine to recognize the user's emotions. It uses an emotion recognition engine (e.g., OpenCV + deep learning model) that analyzes emotions from facial expressions, voice tone, and text. Based on the recognized emotion, the notification message is personalized. For example, if the user has negative emotions, a polite message of support will be added to the notification text. If the emotion is positive, a message of thanks or encouragement will be added.

[0313] Based on notifications from the terminal, the user performs preventive maintenance tasks, such as rebooting the system, applying patches, and replacing hardware. Once the task is complete, the user reports the results to the terminal.

[0314] The terminal reports the progress and results of the maintenance work performed by the user to the server. The report includes the work performed, the date and time, and the results. The server analyzes the reported maintenance results and feeds the data back into the prediction model. The fed-back data is used to improve the accuracy of the next prediction.

[0315] As a specific example, a server may analyze data obtained in real time from an AI sensor to detect a sudden increase in the number of transactions. If an abnormality is detected by comparing it with past data, an alert is generated, and the device uses an emotion engine to recognize emotions from the user's facial expressions and voice. Users who express anxiety may receive a thoughtful notification such as, "The system load is increasing rapidly, but please rest assured that we are taking measures." Users can then take measures upon receiving the notification and maintain stable system operation.

[0316] An example of a prompt sentence to be input to the generative AI model is, "Analyze the data obtained from the AI ​​sensor and detect anomalies. If an anomaly is detected, automatically generate an appropriate maintenance plan and recognize the user's emotions to personalize the notification message."

[0317] By doing all of this, the system can improve system reliability and user satisfaction by utilizing AI sensors and data analysis to detect anomalies in real time, automatically generating appropriate maintenance plans, and combining an emotion engine to provide personalized notifications to users.

[0318] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0319] Step 1: Data collection

[0320] The server receives data from the AI ​​sensors in real time, including the number of transactions, response time, CPU usage, memory usage, etc. The input is the data from the AI ​​sensors, and the output is the data stored in the database.

[0321] Specific behavior:

[0322] The server receives data packets sent from the AI ​​sensor.

[0323] The received data is written to a database and the data is backed up to storage at regular intervals.

[0324] Step 2: Data Preprocessing

[0325] The server pre-processes the retrieved data: the input is the raw data retrieved from the database, and the output is the denoised data.

[0326] Specific behavior:

[0327] The server runs filtering algorithms to detect and correct outliers and missing data in the data.

[0328] Remove noise from the data set, for example by using a low-pass filter.

[0329] Step 3: Anomaly detection

[0330] The server uses the pre-processed data to perform anomaly detection, where the input is the pre-processed data and the output is the anomaly detection results and alert messages.

[0331] Specific behavior:

[0332] The server loads a machine learning model (e.g., a TensorFlow model) that has been trained on past normal data.

[0333] Newly acquired data is input into the model and a prediction is made as to whether it is an anomaly.

[0334] If an abnormality is detected, an alert message is generated and sent to the system administrator.

[0335] Step 4: Update the predictive model

[0336] The server updates the predictive model based on the latest data. The input is past data and current data, and the output is the updated predictive model.

[0337] Specific behavior:

[0338] The server extracts the last six months' worth of data from the database.

[0339] Use the extracted data to retrain machine learning models.

[0340] Save the retrained model for the next anomaly detection.

[0341] Step 5: Auto-generate maintenance plans

[0342] The server generates a maintenance plan based on the predictive model: the input is the updated predictive model and the anomaly detection results, and the output is a maintenance plan that suggests specific preventive maintenance actions.

[0343] Specific behavior:

[0344] The server uses a predictive model to calculate when the next maintenance is required.

[0345] Generate a list of specific maintenance tasks (e.g., part replacement, system reboot, etc.).

[0346] Generate notifications to send maintenance plans to users.

[0347] Step 6: Emotion Recognition and Notification Personalization

[0348] The device uses an emotion engine to recognize the user's emotions before sending a notification. The input is the user's facial expression, voice, and text information, and the output is a personalized notification message.

[0349] Specific behavior:

[0350] The device uses a camera and microphone to capture the user's facial expressions and voice.

[0351] Use an emotion recognition engine (e.g. OpenCV + deep learning model) to recognize user emotions.

[0352] If the user expresses negative emotions, add a "polite support message" to the notification text. If the user expresses positive emotions, add a message of thanks or encouragement.

[0353] Step 7: Perform maintenance

[0354] The user performs preventive maintenance work based on notifications from the terminal. The input is the maintenance plan notification from the terminal, and the output is the results of the performed maintenance.

[0355] Specific behavior:

[0356] The user follows the instructions on the terminal screen to perform maintenance tasks such as restarting the system and replacing parts.

[0357] After the work is completed, the results of the work (e.g., part replacement date and time, whether there were any problems) are entered into the input form on the terminal.

[0358] This data is transmitted from the terminal to the server.

[0359] Step 8: Feedback and learning from results

[0360] The terminal sends the maintenance results reported by the user to the server, which analyzes the received data and feeds it back into the predictive model. The input is the maintenance results reported by the user, and the output is an improved predictive model.

[0361] Specific behavior:

[0362] The terminal calls an API that sends the user's input data to the server.

[0363] The server stores the received data in a database and uses it the next time the predictive model is trained.

[0364] Retrain the predictive model based on the results of the maintenance performed to improve the model's accuracy.

[0365] Through these steps, the system can detect anomalies in real time, develop preventive maintenance plans, and provide personalized notifications based on user sentiment.

[0366] (Application example 2)

[0367] 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."

[0368] In current autonomous vehicles, system anomaly detection and maintenance planning are important. However, automated notifications without considering passenger emotions can cause anxiety and dissatisfaction. This can result in lower user satisfaction and a loss of trust in the system. Furthermore, conventional systems lack the ability to recognize emotions in real time and generate personalized notifications, resulting in a lack of consideration for passengers.

[0369] 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.

[0370] In this invention, the server includes means for collecting data from the AI ​​sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the anomaly detection results, means for automatically generating a maintenance plan for the predicted problems, means for notifying the generated maintenance plan, means for recognizing passenger emotions, and means for personalizing notification messages based on the recognized emotions. This enables not only real-time anomaly detection and predictive maintenance, but also personalized notifications according to passenger emotions.

[0371] An "AI sensor" is a device that uses artificial intelligence technology to measure and collect information on the environment and equipment status in real time.

[0372] "Real-time" is a concept that refers to processing and analysis being carried out immediately at the moment data is generated.

[0373] "Data collection" is the process of obtaining information from devices such as AI sensors and incorporating it into the system.

[0374] "Anomaly detection" is the process of identifying unusual behaviors or conditions based on collected data.

[0375] "Predicting future problems and outages" is the process of assessing the likelihood of future risks and outages based on past data.

[0376] A "maintenance plan" is a schedule of maintenance work required to keep equipment and systems in good condition.

[0377] "Automatically generating" means that the system generates plans and data according to certain rules and algorithms without user intervention.

[0378] "Notifying" is the act of conveying information or warnings to a user.

[0379] "Recognizing passenger emotions" is the process of determining the user's emotional state through facial and voice analysis.

[0380] "Personalization" refers to tailoring the content of services and messages provided to each user based on their characteristics and preferences.

[0381] The system that realizes this invention collects data in real time from AI sensors, detects anomalies, and automatically generates predictive maintenance plans. It also has the ability to recognize passenger emotions and provide personalized notifications based on those emotions. The specific implementation method for this system is described below.

[0382] First, the server receives real-time data from the AI ​​sensors. This data includes information about each vehicle component and its operating status, such as temperature, vibration, transaction count, and response time. The data is stored in a database and regularly backed up to ensure reliability.

[0383] The server then pre-processes the collected data. Specifically, it uses filtering algorithms to remove noise and correct or remove outliers and missing data, improving the reliability of the data. The pre-processed data is then passed through an AI algorithm for anomaly detection. It compares the current data with past normal data patterns to determine whether it is abnormal, and generates an alert if an anomaly is detected.

[0384] The server then updates the predictive model based on the data. It retrains the machine learning model, taking into account the latest trends and seasonal fluctuations, improving prediction accuracy. This automatically generates a maintenance plan that suggests appropriate preventive maintenance work. The generated maintenance plan includes specific actions and their implementation timing, and is notified to the user via their device.

[0385] Before sending a notification, the device uses an emotion engine to recognize the user's emotions. Emotion recognition involves analyzing camera footage and audio captured by the microphone to read emotions from facial expressions and voice tone. The device then personalizes the notification message based on the recognized emotion. For example, if the user has negative emotions, it will provide a more polite, supportive message, and if the user has positive emotions, it will include a message of thanks or encouragement.

[0386] As a concrete example, consider the case where an autonomous vehicle detects an abnormality while driving. The server analyzes data obtained from sensors and detects the abnormality. If the emotion engine recognizes the passenger's anxiety, the terminal will notify the passenger, "The system has detected an abnormality, but please rest assured that we are taking measures."

[0387] Additionally, if a part needs to be replaced in advance based on a predictive maintenance plan, the server will notify the passenger when the next breakdown is predicted and suggest replacing the part in advance. If the emotion engine recognizes a passenger's positive emotion, it will notify them with a thank-you message saying, "Thank you for your pleasant ride. We will carry out maintenance in advance."

[0388] Examples of prompts include the following:

[0389] Write a Python program for an anomaly detection and maintenance scheduler for an autonomous vehicle, as well as a system that recognizes passenger emotions and provides personalized notifications. Use a machine learning model for anomaly detection and a pre-trained emotion recognition model for emotion recognition. Notify passengers with appropriate messages based on the results.

[0390] This invention combines AI sensors and an emotion recognition engine to enable real-time anomaly detection and predictive maintenance in autonomous vehicles, as well as personalized notifications for passengers, thereby improving passenger safety and satisfaction.

[0391] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0392] Step 1:

[0393] The server collects data in real time from the AI ​​sensors. The input includes data describing the status of each vehicle component and system (e.g., temperature, vibration, number of transactions, response time). The server continuously stores this data in a database, which is backed up periodically. The output is the stored real-time data.

[0394] Step 2:

[0395] The server pre-processes the collected data. This process uses a filtering algorithm to remove noise from the data. The input includes the raw data collected in step 1. Data processing involves correcting and removing outliers and missing data, converting it into reliable data. The output is pre-processed, clean data.

[0396] Step 3:

[0397] The server uses the pre-processed data to detect anomalies. The input includes the cleaned data from step 2. The server uses a machine learning model (anomaly detection model) to compare the current data with historical normal data patterns. Data calculations identify outliers and generate alerts if anomalies are detected. The output is the anomaly detection results and an alert if required.

[0398] Step 4:

[0399] The server predicts future problems and failures based on the anomaly detection results. The input includes the anomaly detection results obtained in step 3. The server uses a predictive model (machine learning algorithm) to evaluate future risks based on past data. Data calculation involves trend analysis of abnormal data. The output is the evaluation result of the predicted problems and failures.

[0400] Step 5:

[0401] The server automatically generates a maintenance plan for the predicted problems. The input includes the prediction data obtained in step 4. The server uses a generative AI model to automatically generate appropriate preventive maintenance work and its schedule. Data calculations determine the ideal maintenance time and the required work content. The output is an automatically generated maintenance plan.

[0402] Step 6:

[0403] The terminal recognizes passengers' emotions before notifying the user of the generated maintenance plan. Inputs include camera footage and audio data. The terminal uses an emotion recognition engine (a pre-trained emotion recognition model) to read emotions from facial expressions and voice tone. Emotion analysis is performed as data processing. The output is the recognized emotion data.

[0404] Step 7:

[0405] The device personalizes the notification message based on the recognized emotion. The input includes the emotion data obtained in step 6 and the maintenance plan generated in step 5. The device uses the emotion data to customize the maintenance notification message. Data processing involves adding corresponding wording according to the user's emotion. The output is a personalized notification message.

[0406] Step 8:

[0407] The user performs preventive maintenance based on the personalized notification sent from the terminal. The input includes the personalized notification message created in step 7. The user performs preventive maintenance tasks such as rebooting the system, applying patches, or replacing parts according to the notification. The output is the result of the performed maintenance task.

[0408] Step 9:

[0409] The terminal reports the results of the maintenance performed by the user to the server. The input includes the results of the maintenance work performed in step 8. The terminal sends the report content to the server, and the results are saved. Data processing involves aggregating and organizing the reported data. The output is the maintenance result data saved on the server.

[0410] Step 10:

[0411] The server analyzes the reported maintenance results and feeds them back into the next predictive model. The input includes the maintenance result data saved in step 9. The server analyzes the data to adjust and retrain the predictive model. Data calculations include improving the model to improve prediction accuracy. The output is a predictive model with improved accuracy.

[0412] 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.

[0413] 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.

[0414] 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.

[0415] [Second embodiment]

[0416] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0417] 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.

[0418] 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).

[0419] 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.

[0420] 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.

[0421] 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).

[0422] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0423] 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.

[0424] 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.

[0425] 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.

[0426] 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.

[0427] 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."

[0428] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, and automatically generates appropriate maintenance plans, thereby improving the reliability of payment systems.

[0429] Program processing

[0430] Data collection

[0431] The server receives real-time data from AI sensors, such as the number of transactions, response time, CPU usage, and memory usage, which are continuously collected.

[0432] The server stores the collected data in a database and backs it up as appropriate.

[0433] Data analysis

[0434] The server analyzes the data received in real time, removes noise, and then runs the analyzed data through an AI algorithm to detect anomalies.

[0435] If an anomaly is detected, the server logs it and generates an alert if necessary.

[0436] Update the forecast model

[0437] The server uses the data to update the forecasting model, taking into account current trends and seasonal fluctuations.

[0438] The server retrains the machine learning algorithm to improve prediction accuracy.

[0439] Preventive maintenance plan development

[0440] The server assesses future risks based on predictive models and automatically generates appropriate preventative maintenance plans based on the assessment results.

[0441] The terminal notifies the user of the generated preventive maintenance plan and informs the user when it should be carried out.

[0442] Maintenance

[0443] Users receive notifications from their devices and can perform preventative maintenance, such as rebooting the system, applying patches, or replacing hardware.

[0444] The terminal reports the progress of the performed maintenance to the server.

[0445] Feedback and learning from results

[0446] The server analyzes the results of the maintenance performed by the user and feeds the data back into the predictive model.

[0447] The server uses the feedback results to further improve the predictive model, which further optimizes the next preventive maintenance plan.

[0448] Specific examples

[0449] Case 1: Real-time anomaly detection

[0450] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[0451] The server compares this with past data and determines that this sudden increase is abnormal.

[0452] The terminal notifies the user of the abnormality and prompts them to implement system load balancing and traffic restrictions.

[0453] Users are notified and can take corrective action to maintain stable system operation.

[0454] Case 2: Predictive maintenance planning

[0455] The server uses past data to discover trends in the failure of specific components.

[0456] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[0457] The user is notified and replaces the part to prevent a breakdown.

[0458] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[0459] In this way, the system of the present invention utilizes AI sensors and data analysis to achieve preventive maintenance and improved reliability of the payment system.

[0460] The processing flow will be explained below.

[0461] Step 1:

[0462] The server receives real-time data from the AI ​​sensor. Specifically, it accesses the sensor through an API and collects data such as the number of transactions, response time, CPU usage, and memory usage. This data is periodically stored in a database.

[0463] Step 2:

[0464] The server pre-processes the collected data and removes noise, specifically by using filtering algorithms to correct and remove outliers and missing data, which improves the reliability of the data.

[0465] Step 3:

[0466] The server runs anomaly detection algorithms on the pre-processed data, for example comparing it with past normal data patterns to determine whether the current data is anomalous, and generates an alert if an anomaly is detected.

[0467] Step 4:

[0468] The server uses the anomaly detection results to predict future problems and outages. Specifically, it uses machine learning models to build predictive models from past data and apply them to real-time data to predict potential problems.

[0469] Step 5:

[0470] The server automatically generates maintenance plans for predicted problems, creating maintenance schedules that suggest specific actions, such as replacing specific parts or rebooting the system.

[0471] Step 6:

[0472] The terminal notifies the user of the maintenance plan received from the server. The notification includes the content, timing, and implementation method of the required maintenance. The user receives the notification and prepares for the maintenance.

[0473] Step 7:

[0474] Based on notifications from the terminal, the user performs preventive maintenance work, such as rebooting the system, applying patches, and replacing hardware. Once the work is completed, the results are reported to the terminal.

[0475] Step 8:

[0476] The terminal reports the progress and results of the maintenance work performed by the user to the server, including the work performed, the date and time of the work, and the results.

[0477] Step 9:

[0478] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[0479] Step 10:

[0480] The server uses the feedback results to further optimize predictive models and maintenance plans, thereby preventing future problems and continuously improving system reliability.

[0481] Example 1

[0482] 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."

[0483] Conventional systems lacked real-time data collection and anomaly detection, resulting in numerous limitations in the accuracy of predictive models and the automatic generation of maintenance plans. In particular, they lacked noise removal, consideration of trend fluctuations, and feedback on the status of preventive maintenance, which ultimately hindered efforts to improve the reliability of payment systems. Furthermore, inappropriate notification timing and alert generation for preventive maintenance plans hindered efficient maintenance.

[0484] 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.

[0485] In this invention, the server includes means for collecting data from the AI ​​sensor in real time, means for removing noise and detecting anomalies, means for recording the anomaly detection results in a log and generating alerts, means for analyzing past data and identifying patterns, means for updating the prediction model to take into account the latest trends and seasonal fluctuations, means for saving the maintenance progress status and analyzing the results, and means for feeding back the analysis results to improve the prediction model, thereby enabling real-time anomaly detection, improved prediction accuracy, and automatic generation and notification of efficient preventive maintenance plans.

[0486] An "AI sensor" is a sensor that uses machine learning algorithms to collect data from the environment in real time and analyze that data.

[0487] "Data collection" is the process in which the server acquires data from the AI ​​sensor in real time and stores it in a database or similar.

[0488] "Denoising" is the process of applying a filtering algorithm to remove unwanted noise from collected data.

[0489] Anomaly detection is the process of analyzing collected and denoised data to find patterns or signs that are out of the ordinary.

[0490] "Alert generation" is the process of creating an alert to notify a user when an anomaly is detected.

[0491] A "predictive model" is a machine learning algorithm or statistical model that uses historical data to forecast future movements and trends.

[0492] "Pattern analysis" is the process of analyzing historical data to find recurring rules or characteristics.

[0493] "Trend consideration" is the process of incorporating recent trends and seasonal fluctuations when updating a forecasting model.

[0494] "Maintenance progress" is information that records and tracks the implementation status of maintenance work such as preventive maintenance and repairs.

[0495] "Feedback" is the process of analyzing maintenance results and using that information to improve predictive models.

[0496] A "preventive maintenance plan" is a plan that automatically generates schedules and procedures for maintenance to be carried out in advance to prevent abnormalities and breakdowns.

[0497] "Notification" is the process by which users are informed of generated preventative maintenance plans and alerts.

[0498] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, and automatically generates appropriate preventive maintenance plans.

[0499] The server first has a means to obtain data such as the number of transactions, response time, CPU usage, and memory usage from the AI ​​sensor in real time. Specifically, the server periodically sends API requests to collect the latest data from the sensor. The collected data is stored in a database such as MySQL or PostgreSQL and is backed up as appropriate.

[0500] Next, the server analyzes the collected data and removes noise. This is done using algorithms such as moving average filters and low-pass filters. The noise-removed data is then analyzed using an AI algorithm to detect anomalies. For example, an LSTM (Long Short-Term Memory) anomaly detection algorithm is used. If an anomaly is detected, the server records the information in a log and notifies the user by sending an email or generating an alert via WebSocket.

[0501] The server then updates the predictive model based on abnormal data. This is done using machine learning libraries such as TensorFlow and PyTorch. The predictive model is updated to take into account the latest trends and seasonal fluctuations. Based on this updated predictive model, the server evaluates future risks and automatically generates appropriate preventive maintenance plans. For example, it uses an automatic scheduling algorithm to determine the timing of maintenance. The generated preventive maintenance plan is then notified to the user via their terminal, informing them of the timing of implementation.

[0502] During the preventive maintenance phase, users receive notifications from their devices and perform preventive maintenance such as rebooting the system, applying patches, and replacing hardware. The devices then report the progress of the maintenance to the server, which then analyzes the results of the user's maintenance and feeds the data back into the predictive model. Based on this feedback, the predictive model is further improved, and the next preventive maintenance plan is generated more precisely.

[0503] Specific examples

[0504] Case 1: Real-time anomaly detection

[0505] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[0506] The server compares this with past data and determines that this sudden increase is abnormal.

[0507] The terminal notifies the user of the abnormality and prompts them to implement system load balancing and traffic restrictions.

[0508] Users are notified and can take corrective action to maintain stable system operation.

[0509] Case 2: Predictive maintenance planning

[0510] The server uses past data to discover trends in the failure of specific components.

[0511] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[0512] The user is notified and replaces the part to prevent a breakdown.

[0513] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[0514] Prompt Sentence Examples

[0515] Prompt example 1: Real-time anomaly detection

[0516] Create a program that detects sudden spikes in the number of transactions obtained from the AI ​​sensor and compares the situation with past data to determine whether it is an abnormality. If an abnormality is detected, notify the user and include content that encourages load balancing or traffic restrictions.

[0517] Example prompt 2: Predictive maintenance plan

[0518] Please generate a program that analyzes the failure trends of specific parts based on past data and notifies the user when the next failure is predicted. Also, please include a program that suggests parts replacement in advance to the user and provides feedback on the results to improve the prediction model.

[0519] As described above, this system utilizes AI sensors and data analysis to achieve preventive maintenance and improve the reliability of the payment system.

[0520] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0521] Step 1: Data Acquisition

[0522] The server receives data from the AI ​​sensors in real time. Specifically, the server periodically sends API requests to collect data from the sensors, such as response time, CPU usage, memory usage, and number of transactions.

[0523] Input: Data from AI sensors

[0524] Output: Real-time data captured

[0525] Step 2: Save data

[0526] The server stores the collected data in a database and performs regular backups. Specifically, data is inserted using SQL queries using MySQL or PostgreSQL, and the data is saved periodically using a backup script.

[0527] Input: Real-time data captured

[0528] Output: Data stored in the database

[0529] Step 3: Noise reduction

[0530] The server removes noise from the collected data by applying a moving average filter or low-pass filter to remove unnecessary data.

[0531] Input: Data stored in the database

[0532] Output: Denoised data

[0533] Step 4: Anomaly detection

[0534] The server then feeds the denoised data into an AI algorithm to detect anomalies, for example using a Long Short-Term Memory (LSTM) algorithm to detect anomalous patterns.

[0535] Input: Denoised data

[0536] Output: Detected anomalies

[0537] Step 5: Alert Generation

[0538] The server logs any detected anomalies and generates alerts if necessary, which are sent to users in real-time via email and / or WebSocket notifications.

[0539] Input: Detected anomaly

[0540] Output: Log entries, alert notifications

[0541] Step 6: Update the predictive model

[0542] The server uses anomalies and historical data to update the predictive model, specifically by retraining the machine learning algorithm using TensorFlow or PyTorch.

[0543] Input: Anomaly data, historical data

[0544] Output: Updated forecast model

[0545] Step 7: Consider the trend

[0546] The server modifies the forecast model to take into account recent trends and seasonal fluctuations, specifically including semi-annual and annual data in the forecast model.

[0547] Input: Updated forecast model

[0548] Output: A revised forecast model

[0549] Step 8: Risk assessment and maintenance plan generation

[0550] The server evaluates future risks based on the modified predictive model, automatically generates appropriate preventive maintenance plans, and determines maintenance timing using an automatic scheduling algorithm.

[0551] Input: Modified forecast model

[0552] Output: Risk assessment results, preventive maintenance plan

[0553] Step 9: Notification

[0554] The device will notify the user of the generated preventive maintenance plan and notify them when it is time to implement it, using a mobile app or desktop notification.

[0555] Input: Preventive Maintenance Plan

[0556] Output: User notification

[0557] Step 10: Perform preventative maintenance

[0558] Users receive notifications from their devices and can then perform preventative maintenance, such as rebooting the system, applying patches, or replacing hardware.

[0559] Input: Notification

[0560] Output: Preventive maintenance performed

[0561] Step 11: Progress Report

[0562] The terminal reports the progress of the performed maintenance to the server. For example, after the maintenance is completed, the terminal presses a confirmation button to send a completion status to the server.

[0563] Input: Preventive maintenance performed

[0564] Output: progress report

[0565] Step 12: Maintenance results analysis and feedback

[0566] The server analyzes the results of the maintenance performed by the user and feeds that data back into the predictive model. Specifically, it compares performance data before and after the maintenance to evaluate the effectiveness, and adds the analysis results to the dataset for retraining.

[0567] Input: Progress report

[0568] Output: Feedback data

[0569] Step 13: Model Improvement

[0570] The server improves the predictive model based on the feedback results, and ultimately the improved predictive model generates a more accurate next preventive maintenance plan.

[0571] Input: Feedback data

[0572] Output: An improved predictive model

[0573] (Application example 1)

[0574] 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."

[0575] Maintaining system reliability and stability is essential for modern electronic payment services. However, conventional monitoring systems are slow to detect anomalies and are unable to quickly generate appropriate maintenance plans. Furthermore, there are limited means of properly notifying users about the timing and methods of preventive maintenance, making it difficult to completely eliminate the risk of breakdowns and service outages.

[0576] 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.

[0577] In this invention, the server includes means for collecting data from an AI sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the results of the anomaly detection, means for automatically generating a maintenance plan for the predicted problems, means for notifying the user of the generated maintenance plan, and means for notifying the user of the anomaly and presenting a maintenance procedure via a smart device. This makes it possible to significantly improve the reliability and stability of payment services by detecting system anomalies early and immediately implementing appropriate preventive maintenance.

[0578] An "AI sensor" is a sensor equipped with artificial intelligence that monitors the status of the environment and system in real time and collects data.

[0579] "Data collection" is the process of compiling information from a system or environment using sensors or other data acquisition methods and storing it for analysis and processing.

[0580] "Anomaly detection" is the process of identifying unusual patterns and values ​​based on collected data and finding signs of system abnormalities or problems.

[0581] "Predicting future problems and failures" is a method of estimating possible future problems and system failures based on past data and current anomaly detection results, and taking measures in advance.

[0582] "Automatic generation of maintenance plans" is the process of automatically formulating specific maintenance procedures and schedules for predicted problems or failures.

[0583] "Notification" is an information transmission means for informing the user of abnormalities and the results of the maintenance plan generation.

[0584] A "smart device" is a mobile terminal or wearable device that has the ability to connect to the Internet and is a device that provides information to users and receives notifications.

[0585] "Feedback" is the process of recording the results and effectiveness of maintenance and preventative care to help improve future predictive models and optimize the system.

[0586] A "machine learning algorithm" is an algorithm that automatically learns from data, finds patterns and relationships, and makes predictions and classifications.

[0587] "Preventive maintenance" refers to maintenance activities carried out in advance to prevent system failures.

[0588] The system for implementing this invention consists of three main components: an AI sensor, a server, and a user terminal (smart device).

[0589] Data collection

[0590] The server collects data in real time from AI sensors, such as the number of transactions, response time, CPU usage, and memory consumption of the payment system, and has built-in functionality to obtain data via sensors and APIs.

[0591] Data analysis

[0592] The server performs anomaly detection based on the collected data. To do this, the first step is to remove noise from the data. After the clean data is generated, anomalies are detected using machine learning algorithms (e.g., RandomForestClassifier). When an anomaly is detected, the information is saved as a log and an alert is generated if necessary.

[0593] Generate future forecasts and maintenance plans

[0594] The server predicts potential future problems and failures based on the anomaly detection results. By analyzing past data and continually updating the prediction model, trends and seasonal fluctuations are taken into account. For predicted problems, the server automatically generates an appropriate maintenance plan.

[0595] Notifications and Feedback

[0596] The generated maintenance plan is sent to the user's smart device (such as a smartphone or tablet). The user can receive this notification and carry out preventive maintenance on the system. Once the maintenance is complete, the results are fed back to the server and used to improve the accuracy of the predictive model.

[0597] Specific examples

[0598] For example, if a server analyzes real-time data and detects a sudden increase in transaction response time, it will recognize this as an anomaly. Based on this information, the server can predict future load increases or system failures and send notifications to smart devices such as "System load is increasing. Please perform maintenance." Users can receive these notifications and perform maintenance to maintain system stability.

[0599] An example of a prompt is as follows:

[0600] "Please implement a function that analyzes transaction data from payment systems, detects anomalies, and automatically generates appropriate maintenance plans."

[0601] In this way, the present invention builds a system that improves the reliability and stability of electronic payment services by linking AI sensors, servers, and smart devices.

[0602] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0603] Step 1: Data collection

[0604] The server collects data in real time from the AI ​​sensors, including the number of transactions, response time, CPU usage, memory usage, etc. The data is retrieved through an API and sent to the server, providing the raw data needed to accurately monitor the system's operating status.

[0605] Step 2: Save data

[0606] The server stores the collected data in a database for subsequent analysis and to update the predictive model. During this process, data is backed up regularly to ensure data reliability and integrity.

[0607] Step 3: Data analysis

[0608] The server analyzes the data and removes noise. Outliers and missing values ​​are filtered out during the data cleaning process. After that, anomalies are detected using a machine learning algorithm (e.g., RandomForestClassifier) ​​based on the clean data. The clean data is given as input, and the output is the presence or absence of anomalies.

[0609] Step 4: Log anomalies and generate alerts

[0610] If an abnormality is detected, the server records the details in a log and generates an alert if necessary. The log stores detailed information about the abnormality and the time of occurrence. The alert is used for notification in the next step.

[0611] Step 5: Auto-generate maintenance plans

[0612] The server predicts future problems and failures based on the anomaly detection results. This involves statistical trend analysis that takes into account past data and current conditions. Detailed maintenance work to address predicted problems is automatically planned. The maintenance plan is generated as a result of calculations using the predictive model.

[0613] Step 6: Notification

[0614] The server notifies the user of the generated maintenance plan via their smart device (smartphone or tablet), where a push notification is sent via the application to prompt the user to take action. The notification includes specific maintenance procedures and the timing of their implementation.

[0615] Step 7: User performs preventative maintenance

[0616] Users receive notifications and can then perform preventative maintenance as instructed, such as rebooting the system or applying software patches, with the actions taken directly fed back to the next step.

[0617] Step 8: Feedback

[0618] Users provide feedback on the results of their maintenance work to the server, which then uses this feedback to retrain the predictive model, improving the accuracy of future anomaly detection and maintenance planning.

[0619] Through the above steps, the present invention provides a system that improves the reliability and stability of electronic payment services.

[0620] 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.

[0621] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, automatically generates appropriate maintenance plans, and provides more personalized notifications to users by combining it with an emotion engine that recognizes user emotions. This not only improves the reliability of the payment system, but also increases user satisfaction.

[0622] Program processing

[0623] Data collection

[0624] The server receives real-time data from the AI ​​sensors, such as the number of transactions, response time, CPU usage, memory usage, etc. This data is stored in a database and backed up regularly.

[0625] Data analysis

[0626] The server pre-processes the collected data to remove noise and improve the reliability of the data by using filtering algorithms to correct and remove outliers and missing data.

[0627] The server then runs the pre-processed data through AI algorithms to detect anomalies, for example by comparing the current data with normal data patterns from the past to determine whether they are abnormal. If an anomaly is detected, an alert is generated.

[0628] Update the forecast model

[0629] The server updates the forecasting model based on the data, retraining the machine learning model to take into account current trends and seasonal fluctuations, improving forecast accuracy.

[0630] Preventive maintenance plan development

[0631] The server evaluates future risks based on predictive models, automatically generates a maintenance plan that proposes appropriate preventive maintenance work, and defines specific actions.

[0632] The terminal notifies the user of the generated preventive maintenance plan and notifies them of the timing of its implementation. This notification includes the specific content, timing, and method of the maintenance.

[0633] emotion recognition

[0634] Before sending a notification, the device uses an emotion engine to recognize the user's emotions, for example, by reading emotions from facial expressions, voice tone, and text.

[0635] The device personalizes notification messages based on the recognized emotion: if the user has a negative emotion, it provides a more polite and supportive message, and if the user has a positive emotion, it adds a message of thanks or encouragement.

[0636] Maintenance

[0637] Based on notifications from the terminal, the user performs preventive maintenance work, such as rebooting the system, applying patches, or replacing hardware. Once the work is completed, the results are reported to the terminal.

[0638] Feedback and learning from results

[0639] The terminal reports the progress and results of the maintenance work performed by the user to the server, including the work performed, the date and time of the work, and the results.

[0640] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[0641] Specific examples

[0642] Case 1: Real-time anomaly detection and emotion recognition

[0643] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[0644] The server compares this spike with historical data and determines that it is abnormal, and an alert is generated.

[0645] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. For example, if the user is feeling anxious, the notification message will include a message such as, "The system load is increasing rapidly, but please rest assured that we are taking measures to address this."

[0646] Users are notified and can take corrective action to maintain stable system operation.

[0647] Case 2: Predictive maintenance planning and emotion recognition

[0648] The server uses past data to discover trends in the failure of specific components.

[0649] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[0650] The device recognizes the user's emotions before sending a notification, and if the user is expressing positive emotions, it will include a thank-you message such as, "Thank you for your continued cooperation. Please replace parts in advance to ensure safe operation."

[0651] The user is notified and replaces the part to prevent a breakdown.

[0652] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[0653] As described above, the system of the present invention utilizes AI sensors and data analysis, and further combines it with an emotion engine to achieve preventative maintenance and improved reliability of the payment system, while also improving user satisfaction.

[0654] The processing flow will be explained below.

[0655] Step 1:

[0656] The server obtains data from the AI ​​sensors in real time. Specifically, it polls the sensors via API to collect data such as the number of transactions, response time, CPU usage, and memory usage. This data is recorded in a database and is backed up to always reflect the latest status.

[0657] Step 2:

[0658] The server pre-processes the collected data, applying filtering algorithms to correct or remove outliers and missing data, a process that improves the accuracy of the data used for analysis.

[0659] Step 3:

[0660] The server performs anomaly detection based on pre-processed data. It uses AI algorithms to compare the current data with past normal data patterns and determine whether it is anomalous. If an anomaly is detected, it is logged and an alert is generated if necessary.

[0661] Step 4:

[0662] The server predicts future problems and outages based on the anomaly detection results. It uses machine learning models to build predictive models from past data and apply them to real-time data. This prediction is used to assess the risk of system problems occurring.

[0663] Step 5:

[0664] The server automatically generates maintenance plans for predicted problems, creating a maintenance schedule with specific preventative maintenance actions such as part replacement needs and system reboots.

[0665] Step 6:

[0666] Before notifying the user of the maintenance plan received from the server, the device uses an emotion engine to recognize the user's emotions. Emotion recognition is performed through the camera, microphone, text analysis, etc.

[0667] Step 7:

[0668] The device customizes the maintenance notification message based on the user's recognized emotions. For example, if the user is expressing negative emotions, the notification message will include a thoughtful message such as "Don't worry, our technical team is working on it." On the other hand, if the user is expressing positive emotions, the notification message will include words of gratitude and encouragement.

[0669] Step 8:

[0670] Users can perform preventative maintenance tasks based on customized notifications from their devices, such as rebooting the system, applying patches, or replacing hardware, and the results are reported back to the device once the task is complete.

[0671] Step 9:

[0672] The terminal reports the progress and results of the maintenance work performed by the user to the server, including details of the work performed, the date and time of the work, and the results.

[0673] Step 10:

[0674] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[0675] Step 11:

[0676] The server uses the feedback results to further optimize predictive models and maintenance plans, thereby preventing future problems and continuously improving system reliability.

[0677] Example 2

[0678] 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."

[0679] In conventional systems, it is common to detect anomalies and perform preventive maintenance based on the results. However, more efficient and personalized notifications are required to further improve system reliability and user satisfaction. Furthermore, conventional systems have difficulty in providing notifications that take user emotions into account, which can lead to user dissatisfaction. Therefore, there is a need for a system that can detect anomalies in real time, automatically generate appropriate maintenance plans, and personalize notifications based on user emotions.

[0680] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting data from an AI sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the anomaly detection results, means for automatically generating a maintenance plan for the predicted problems, means for notifying the user of the generated maintenance plan, and means for recognizing the user's emotions and personalizing the notification message. This not only enables the system reliability to be improved by detecting anomalies in real time and automatically generating a preventive maintenance plan, but also improves user satisfaction by personalized notifications based on emotion recognition.

[0681] An "AI sensor" is a sensor that uses artificial intelligence technology to collect and analyze data in real time.

[0682] "Real-time" refers to data and information being processed and acquired the moment it is generated.

[0683] "Data collection means" means means for collecting data through sensors or other measuring devices.

[0684] An "anomaly detection means" is a means for detecting abnormal conditions or patterns based on collected data.

[0685] The "prediction means" is a means for predicting future problems or failures based on the anomaly detection results.

[0686] The "automatic maintenance plan generation means" is a means for automatically generating a maintenance plan for a predicted problem.

[0687] The "notification means" is a means for notifying the user of the generated maintenance plan.

[0688] "Emotion recognition means" refers to a means for recognizing and analyzing the user's emotions.

[0689] "Personalization" means customizing content and formats to suit individual users.

[0690] "Past data analysis means" is a means for analyzing data collected in the past and identifying patterns.

[0691] The "prediction model update means" is a means for updating the prediction model to the latest state based on the analyzed data.

[0692] The "maintenance progress storage means" is a means for storing the progress and results of maintenance work.

[0693] The "maintenance result feedback means" is a means for collecting the results of the performed maintenance and reflecting them in the prediction model.

[0694] The "alert generation means" is a means for generating a warning message when an abnormality is detected.

[0695] A "system" is the entire device or network that combines and operates the above means.

[0696] This invention is a system that detects abnormalities in real time and automatically generates appropriate maintenance plans by recognizing the user's emotions and personalizing notification messages.

[0697] The server uses AI sensors to collect data in real time, including the number of transactions, response time, CPU usage, and memory usage, which is then stored in a database and backed up regularly.

[0698] The collected data is pre-processed using a filtering algorithm. The server corrects and removes outliers and missing data to remove noise, improving the reliability of the data. The pre-processed data is compared with past normal data patterns. The server performs anomaly detection using a machine learning model (e.g., TensorFlow model) and generates an alert if an anomaly is detected.

[0699] Based on the anomaly detection results, the server predicts future problems and outages. The predictive model is updated with the latest data and retrained to take into account trends and seasonal fluctuations, improving the accuracy of the predictions.

[0700] The server automatically generates a maintenance plan that proposes preventive maintenance work for predicted problems. The generated maintenance plan includes the content and timing of specific maintenance work (e.g., system restart, part replacement, etc.). The terminal notifies the user of this maintenance plan and informs them of the timing of its implementation.

[0701] Before sending a notification, the device uses an emotion engine to recognize the user's emotions. It uses an emotion recognition engine (e.g., OpenCV + deep learning model) that analyzes emotions from facial expressions, voice tone, and text. Based on the recognized emotion, the notification message is personalized. For example, if the user has negative emotions, a polite message of support will be added to the notification text. If the emotion is positive, a message of thanks or encouragement will be added.

[0702] Based on notifications from the terminal, the user performs preventive maintenance tasks, such as rebooting the system, applying patches, and replacing hardware. Once the task is complete, the user reports the results to the terminal.

[0703] The terminal reports the progress and results of the maintenance work performed by the user to the server. The report includes the work performed, the date and time, and the results. The server analyzes the reported maintenance results and feeds the data back into the prediction model. The fed-back data is used to improve the accuracy of the next prediction.

[0704] As a specific example, a server may analyze data obtained in real time from an AI sensor to detect a sudden increase in the number of transactions. If an abnormality is detected by comparing it with past data, an alert is generated, and the device uses an emotion engine to recognize emotions from the user's facial expressions and voice. Users who express anxiety may receive a thoughtful notification such as, "The system load is increasing rapidly, but please rest assured that we are taking measures." Users can then take measures upon receiving the notification and maintain stable system operation.

[0705] An example of a prompt sentence to be input to the generative AI model is, "Analyze the data obtained from the AI ​​sensor and detect anomalies. If an anomaly is detected, automatically generate an appropriate maintenance plan and recognize the user's emotions to personalize the notification message."

[0706] By doing all of this, the system can improve system reliability and user satisfaction by utilizing AI sensors and data analysis to detect anomalies in real time, automatically generating appropriate maintenance plans, and combining an emotion engine to provide personalized notifications to users.

[0707] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0708] Step 1: Data collection

[0709] The server receives data from the AI ​​sensors in real time, including the number of transactions, response time, CPU usage, memory usage, etc. The input is the data from the AI ​​sensors, and the output is the data stored in the database.

[0710] Specific behavior:

[0711] The server receives data packets sent from the AI ​​sensor.

[0712] The received data is written to a database and the data is backed up to storage at regular intervals.

[0713] Step 2: Data Preprocessing

[0714] The server pre-processes the retrieved data: the input is the raw data retrieved from the database, and the output is the denoised data.

[0715] Specific behavior:

[0716] The server runs filtering algorithms to detect and correct outliers and missing data in the data.

[0717] Remove noise from the data set, for example by using a low-pass filter.

[0718] Step 3: Anomaly detection

[0719] The server uses the pre-processed data to perform anomaly detection, where the input is the pre-processed data and the output is the anomaly detection results and alert messages.

[0720] Specific behavior:

[0721] The server loads a machine learning model (e.g., a TensorFlow model) that has been trained on past normal data.

[0722] Newly acquired data is input into the model and a prediction is made as to whether it is an anomaly.

[0723] If an abnormality is detected, an alert message is generated and sent to the system administrator.

[0724] Step 4: Update the predictive model

[0725] The server updates the predictive model based on the latest data. The input is past data and current data, and the output is the updated predictive model.

[0726] Specific behavior:

[0727] The server extracts the last six months' worth of data from the database.

[0728] Use the extracted data to retrain machine learning models.

[0729] Save the retrained model for the next anomaly detection.

[0730] Step 5: Auto-generate maintenance plans

[0731] The server generates a maintenance plan based on the predictive model: the input is the updated predictive model and the anomaly detection results, and the output is a maintenance plan that suggests specific preventive maintenance actions.

[0732] Specific behavior:

[0733] The server uses a predictive model to calculate when the next maintenance is required.

[0734] Generate a list of specific maintenance tasks (e.g., part replacement, system reboot, etc.).

[0735] Generate notifications to send maintenance plans to users.

[0736] Step 6: Emotion Recognition and Notification Personalization

[0737] The device uses an emotion engine to recognize the user's emotions before sending a notification. The input is the user's facial expression, voice, and text information, and the output is a personalized notification message.

[0738] Specific behavior:

[0739] The device uses a camera and microphone to capture the user's facial expressions and voice.

[0740] Use an emotion recognition engine (e.g. OpenCV + deep learning model) to recognize user emotions.

[0741] If the user expresses negative emotions, add a "polite support message" to the notification text. If the user expresses positive emotions, add a message of thanks or encouragement.

[0742] Step 7: Perform maintenance

[0743] The user performs preventive maintenance work based on notifications from the terminal. The input is the maintenance plan notification from the terminal, and the output is the results of the performed maintenance.

[0744] Specific behavior:

[0745] The user follows the instructions on the terminal screen to perform maintenance tasks such as restarting the system and replacing parts.

[0746] After the work is completed, the results of the work (e.g., part replacement date and time, whether there were any problems) are entered into the input form on the terminal.

[0747] This data is transmitted from the terminal to the server.

[0748] Step 8: Feedback and learning from results

[0749] The terminal sends the maintenance results reported by the user to the server, which analyzes the received data and feeds it back into the predictive model. The input is the maintenance results reported by the user, and the output is an improved predictive model.

[0750] Specific behavior:

[0751] The terminal calls an API that sends the user's input data to the server.

[0752] The server stores the received data in a database and uses it the next time the predictive model is trained.

[0753] Retrain the predictive model based on the results of the maintenance performed to improve the model's accuracy.

[0754] Through these steps, the system can detect anomalies in real time, develop preventive maintenance plans, and provide personalized notifications based on user sentiment.

[0755] (Application example 2)

[0756] 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."

[0757] In current autonomous vehicles, system anomaly detection and maintenance planning are important. However, automated notifications without considering passenger emotions can cause anxiety and dissatisfaction. This can result in lower user satisfaction and a loss of trust in the system. Furthermore, conventional systems lack the ability to recognize emotions in real time and generate personalized notifications, resulting in a lack of consideration for passengers.

[0758] 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.

[0759] In this invention, the server includes means for collecting data from the AI ​​sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the anomaly detection results, means for automatically generating a maintenance plan for the predicted problems, means for notifying the generated maintenance plan, means for recognizing passenger emotions, and means for personalizing notification messages based on the recognized emotions. This enables not only real-time anomaly detection and predictive maintenance, but also personalized notifications according to passenger emotions.

[0760] An "AI sensor" is a device that uses artificial intelligence technology to measure and collect information on the environment and equipment status in real time.

[0761] "Real-time" is a concept that refers to processing and analysis being carried out immediately at the moment data is generated.

[0762] "Data collection" is the process of obtaining information from devices such as AI sensors and incorporating it into the system.

[0763] "Anomaly detection" is the process of identifying unusual behaviors or conditions based on collected data.

[0764] "Predicting future problems and outages" is the process of assessing the likelihood of future risks and outages based on past data.

[0765] A "maintenance plan" is a schedule of maintenance work required to keep equipment and systems in good condition.

[0766] "Automatically generating" means that the system generates plans and data according to certain rules and algorithms without user intervention.

[0767] "Notifying" is the act of conveying information or warnings to a user.

[0768] "Recognizing passenger emotions" is the process of determining the user's emotional state through facial and voice analysis.

[0769] "Personalization" refers to tailoring the content of services and messages provided to each user based on their characteristics and preferences.

[0770] The system that realizes this invention collects data in real time from AI sensors, detects anomalies, and automatically generates predictive maintenance plans. It also has the ability to recognize passenger emotions and provide personalized notifications based on those emotions. The specific implementation method for this system is described below.

[0771] First, the server receives real-time data from the AI ​​sensors. This data includes information about each vehicle component and its operating status, such as temperature, vibration, transaction count, and response time. The data is stored in a database and regularly backed up to ensure reliability.

[0772] The server then pre-processes the collected data. Specifically, it uses filtering algorithms to remove noise and correct or remove outliers and missing data, improving the reliability of the data. The pre-processed data is then passed through an AI algorithm for anomaly detection. It compares the current data with past normal data patterns to determine whether it is abnormal, and generates an alert if an anomaly is detected.

[0773] The server then updates the predictive model based on the data. It retrains the machine learning model, taking into account the latest trends and seasonal fluctuations, improving prediction accuracy. This automatically generates a maintenance plan that suggests appropriate preventive maintenance work. The generated maintenance plan includes specific actions and their implementation timing, and is notified to the user via their device.

[0774] Before sending a notification, the device uses an emotion engine to recognize the user's emotions. Emotion recognition involves analyzing camera footage and audio captured by the microphone to read emotions from facial expressions and voice tone. The device then personalizes the notification message based on the recognized emotion. For example, if the user has negative emotions, it will provide a more polite, supportive message, and if the user has positive emotions, it will include a message of thanks or encouragement.

[0775] As a concrete example, consider the case where an autonomous vehicle detects an abnormality while driving. The server analyzes data obtained from sensors and detects the abnormality. If the emotion engine recognizes the passenger's anxiety, the terminal will notify the passenger, "The system has detected an abnormality, but please rest assured that we are taking measures."

[0776] Additionally, if a part needs to be replaced in advance based on a predictive maintenance plan, the server will notify the passenger when the next breakdown is predicted and suggest replacing the part in advance. If the emotion engine recognizes a passenger's positive emotion, it will notify them with a thank-you message saying, "Thank you for your pleasant ride. We will carry out maintenance in advance."

[0777] Examples of prompts include the following:

[0778] Write a Python program for an anomaly detection and maintenance scheduler for an autonomous vehicle, as well as a system that recognizes passenger emotions and provides personalized notifications. Use a machine learning model for anomaly detection and a pre-trained emotion recognition model for emotion recognition. Notify passengers with appropriate messages based on the results.

[0779] This invention combines AI sensors and an emotion recognition engine to enable real-time anomaly detection and predictive maintenance in autonomous vehicles, as well as personalized notifications for passengers, thereby improving passenger safety and satisfaction.

[0780] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0781] Step 1:

[0782] The server collects data in real time from the AI ​​sensors. The input includes data describing the status of each vehicle component and system (e.g., temperature, vibration, number of transactions, response time). The server continuously stores this data in a database, which is backed up periodically. The output is the stored real-time data.

[0783] Step 2:

[0784] The server pre-processes the collected data. This process uses a filtering algorithm to remove noise from the data. The input includes the raw data collected in step 1. Data processing involves correcting and removing outliers and missing data, converting it into reliable data. The output is pre-processed, clean data.

[0785] Step 3:

[0786] The server uses the pre-processed data to detect anomalies. The input includes the cleaned data from step 2. The server uses a machine learning model (anomaly detection model) to compare the current data with historical normal data patterns. Data calculations identify outliers and generate alerts if anomalies are detected. The output is the anomaly detection results and an alert if required.

[0787] Step 4:

[0788] The server predicts future problems and failures based on the anomaly detection results. The input includes the anomaly detection results obtained in step 3. The server uses a predictive model (machine learning algorithm) to evaluate future risks based on past data. Data calculation involves trend analysis of abnormal data. The output is the evaluation result of the predicted problems and failures.

[0789] Step 5:

[0790] The server automatically generates a maintenance plan for the predicted problems. The input includes the prediction data obtained in step 4. The server uses a generative AI model to automatically generate appropriate preventive maintenance work and its schedule. Data calculations determine the ideal maintenance time and the required work content. The output is an automatically generated maintenance plan.

[0791] Step 6:

[0792] The terminal recognizes passengers' emotions before notifying the user of the generated maintenance plan. Inputs include camera footage and audio data. The terminal uses an emotion recognition engine (a pre-trained emotion recognition model) to read emotions from facial expressions and voice tone. Emotion analysis is performed as data processing. The output is the recognized emotion data.

[0793] Step 7:

[0794] The device personalizes the notification message based on the recognized emotion. The input includes the emotion data obtained in step 6 and the maintenance plan generated in step 5. The device uses the emotion data to customize the maintenance notification message. Data processing involves adding corresponding wording according to the user's emotion. The output is a personalized notification message.

[0795] Step 8:

[0796] The user performs preventive maintenance based on the personalized notification sent from the terminal. The input includes the personalized notification message created in step 7. The user performs preventive maintenance tasks such as rebooting the system, applying patches, or replacing parts according to the notification. The output is the result of the performed maintenance task.

[0797] Step 9:

[0798] The terminal reports the results of the maintenance performed by the user to the server. The input includes the results of the maintenance work performed in step 8. The terminal sends the report content to the server, and the results are saved. Data processing involves aggregating and organizing the reported data. The output is the maintenance result data saved on the server.

[0799] Step 10:

[0800] The server analyzes the reported maintenance results and feeds them back into the next predictive model. The input includes the maintenance result data saved in step 9. The server analyzes the data to adjust and retrain the predictive model. Data calculations include improving the model to improve prediction accuracy. The output is a predictive model with improved accuracy.

[0801] 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.

[0802] 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.

[0803] 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.

[0804] [Third embodiment]

[0805] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0806] 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.

[0807] 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).

[0808] 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.

[0809] 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.

[0810] 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).

[0811] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0812] 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.

[0813] 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.

[0814] 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.

[0815] 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.

[0816] 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."

[0817] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, and automatically generates appropriate maintenance plans, thereby improving the reliability of payment systems.

[0818] Program processing

[0819] Data collection

[0820] The server receives real-time data from AI sensors, such as the number of transactions, response time, CPU usage, and memory usage, which are continuously collected.

[0821] The server stores the collected data in a database and backs it up as appropriate.

[0822] Data analysis

[0823] The server analyzes the data received in real time, removes noise, and then runs the analyzed data through an AI algorithm to detect anomalies.

[0824] If an anomaly is detected, the server logs it and generates an alert if necessary.

[0825] Update the forecast model

[0826] The server uses the data to update the forecasting model, taking into account current trends and seasonal fluctuations.

[0827] The server retrains the machine learning algorithm to improve prediction accuracy.

[0828] Preventive maintenance plan development

[0829] The server assesses future risks based on predictive models and automatically generates appropriate preventative maintenance plans based on the assessment results.

[0830] The terminal notifies the user of the generated preventive maintenance plan and informs the user when it should be carried out.

[0831] Maintenance

[0832] Users receive notifications from their devices and can perform preventative maintenance, such as rebooting the system, applying patches, or replacing hardware.

[0833] The terminal reports the progress of the performed maintenance to the server.

[0834] Feedback and learning from results

[0835] The server analyzes the results of the maintenance performed by the user and feeds the data back into the predictive model.

[0836] The server uses the feedback results to further improve the predictive model, which further optimizes the next preventive maintenance plan.

[0837] Specific examples

[0838] Case 1: Real-time anomaly detection

[0839] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[0840] The server compares this with past data and determines that this sudden increase is abnormal.

[0841] The terminal notifies the user of the abnormality and prompts them to implement system load balancing and traffic restrictions.

[0842] Users are notified and can take corrective action to maintain stable system operation.

[0843] Case 2: Predictive maintenance planning

[0844] The server uses past data to discover trends in the failure of specific components.

[0845] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[0846] The user is notified and replaces the part to prevent a breakdown.

[0847] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[0848] In this way, the system of the present invention utilizes AI sensors and data analysis to achieve preventive maintenance and improved reliability of the payment system.

[0849] The processing flow will be explained below.

[0850] Step 1:

[0851] The server receives real-time data from the AI ​​sensor. Specifically, it accesses the sensor through an API and collects data such as the number of transactions, response time, CPU usage, and memory usage. This data is periodically stored in a database.

[0852] Step 2:

[0853] The server pre-processes the collected data and removes noise, specifically by using filtering algorithms to correct and remove outliers and missing data, which improves the reliability of the data.

[0854] Step 3:

[0855] The server runs anomaly detection algorithms on the pre-processed data, for example comparing it with past normal data patterns to determine whether the current data is anomalous, and generates an alert if an anomaly is detected.

[0856] Step 4:

[0857] The server uses the anomaly detection results to predict future problems and outages. Specifically, it uses machine learning models to build predictive models from past data and apply them to real-time data to predict potential problems.

[0858] Step 5:

[0859] The server automatically generates maintenance plans for predicted problems, creating maintenance schedules that suggest specific actions, such as replacing specific parts or rebooting the system.

[0860] Step 6:

[0861] The terminal notifies the user of the maintenance plan received from the server. The notification includes the content, timing, and implementation method of the required maintenance. The user receives the notification and prepares for the maintenance.

[0862] Step 7:

[0863] Based on notifications from the terminal, the user performs preventive maintenance work, such as rebooting the system, applying patches, and replacing hardware. Once the work is completed, the results are reported to the terminal.

[0864] Step 8:

[0865] The terminal reports the progress and results of the maintenance work performed by the user to the server, including the work performed, the date and time of the work, and the results.

[0866] Step 9:

[0867] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[0868] Step 10:

[0869] The server uses the feedback results to further optimize predictive models and maintenance plans, thereby preventing future problems and continuously improving system reliability.

[0870] Example 1

[0871] 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."

[0872] Conventional systems lacked real-time data collection and anomaly detection, resulting in numerous limitations in the accuracy of predictive models and the automatic generation of maintenance plans. In particular, they lacked noise removal, consideration of trend fluctuations, and feedback on the status of preventive maintenance, which ultimately hindered efforts to improve the reliability of payment systems. Furthermore, inappropriate notification timing and alert generation for preventive maintenance plans hindered efficient maintenance.

[0873] 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.

[0874] In this invention, the server includes means for collecting data from the AI ​​sensor in real time, means for removing noise and detecting anomalies, means for recording the anomaly detection results in a log and generating alerts, means for analyzing past data and identifying patterns, means for updating the prediction model to take into account the latest trends and seasonal fluctuations, means for saving the maintenance progress status and analyzing the results, and means for feeding back the analysis results to improve the prediction model, thereby enabling real-time anomaly detection, improved prediction accuracy, and automatic generation and notification of efficient preventive maintenance plans.

[0875] An "AI sensor" is a sensor that uses machine learning algorithms to collect data from the environment in real time and analyze that data.

[0876] "Data collection" is the process in which the server acquires data from the AI ​​sensor in real time and stores it in a database or similar.

[0877] "Denoising" is the process of applying a filtering algorithm to remove unwanted noise from collected data.

[0878] Anomaly detection is the process of analyzing collected and denoised data to find patterns or signs that are out of the ordinary.

[0879] "Alert generation" is the process of creating an alert to notify a user when an anomaly is detected.

[0880] A "predictive model" is a machine learning algorithm or statistical model that uses historical data to forecast future movements and trends.

[0881] "Pattern analysis" is the process of analyzing historical data to find recurring rules or characteristics.

[0882] "Trend consideration" is the process of incorporating recent trends and seasonal fluctuations when updating a forecasting model.

[0883] "Maintenance progress" is information that records and tracks the implementation status of maintenance work such as preventive maintenance and repairs.

[0884] "Feedback" is the process of analyzing maintenance results and using that information to improve predictive models.

[0885] A "preventive maintenance plan" is a plan that automatically generates schedules and procedures for maintenance to be carried out in advance to prevent abnormalities and breakdowns.

[0886] "Notification" is the process by which users are informed of generated preventative maintenance plans and alerts.

[0887] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, and automatically generates appropriate preventive maintenance plans.

[0888] The server first has a means to obtain data such as the number of transactions, response time, CPU usage, and memory usage from the AI ​​sensor in real time. Specifically, the server periodically sends API requests to collect the latest data from the sensor. The collected data is stored in a database such as MySQL or PostgreSQL and is backed up as appropriate.

[0889] Next, the server analyzes the collected data and removes noise. This is done using algorithms such as moving average filters and low-pass filters. The noise-removed data is then analyzed using an AI algorithm to detect anomalies. For example, an LSTM (Long Short-Term Memory) anomaly detection algorithm is used. If an anomaly is detected, the server records the information in a log and notifies the user by sending an email or generating an alert via WebSocket.

[0890] The server then updates the predictive model based on abnormal data. This is done using machine learning libraries such as TensorFlow and PyTorch. The predictive model is updated to take into account the latest trends and seasonal fluctuations. Based on this updated predictive model, the server evaluates future risks and automatically generates appropriate preventive maintenance plans. For example, it uses an automatic scheduling algorithm to determine the timing of maintenance. The generated preventive maintenance plan is then notified to the user via their terminal, informing them of the timing of implementation.

[0891] During the preventive maintenance phase, users receive notifications from their devices and perform preventive maintenance such as rebooting the system, applying patches, and replacing hardware. The devices then report the progress of the maintenance to the server, which then analyzes the results of the user's maintenance and feeds the data back into the predictive model. Based on this feedback, the predictive model is further improved, and the next preventive maintenance plan is generated more precisely.

[0892] Specific examples

[0893] Case 1: Real-time anomaly detection

[0894] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[0895] The server compares this with past data and determines that this sudden increase is abnormal.

[0896] The terminal notifies the user of the abnormality and prompts them to implement system load balancing and traffic restrictions.

[0897] Users are notified and can take corrective action to maintain stable system operation.

[0898] Case 2: Predictive maintenance planning

[0899] The server uses past data to discover trends in the failure of specific components.

[0900] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[0901] The user is notified and replaces the part to prevent a breakdown.

[0902] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[0903] Prompt Sentence Examples

[0904] Prompt example 1: Real-time anomaly detection

[0905] Create a program that detects sudden spikes in the number of transactions obtained from the AI ​​sensor and compares the situation with past data to determine whether it is an abnormality. If an abnormality is detected, notify the user and include content that encourages load balancing or traffic restrictions.

[0906] Example prompt 2: Predictive maintenance plan

[0907] Please generate a program that analyzes the failure trends of specific parts based on past data and notifies the user when the next failure is predicted. Also, please include a program that suggests parts replacement in advance to the user and provides feedback on the results to improve the prediction model.

[0908] As described above, this system utilizes AI sensors and data analysis to achieve preventive maintenance and improve the reliability of the payment system.

[0909] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0910] Step 1: Data Acquisition

[0911] The server receives data from the AI ​​sensors in real time. Specifically, the server periodically sends API requests to collect data from the sensors, such as response time, CPU usage, memory usage, and number of transactions.

[0912] Input: Data from AI sensors

[0913] Output: Real-time data captured

[0914] Step 2: Save data

[0915] The server stores the collected data in a database and performs regular backups. Specifically, data is inserted using SQL queries using MySQL or PostgreSQL, and the data is saved periodically using a backup script.

[0916] Input: Real-time data captured

[0917] Output: Data stored in the database

[0918] Step 3: Noise reduction

[0919] The server removes noise from the collected data by applying a moving average filter or low-pass filter to remove unnecessary data.

[0920] Input: Data stored in the database

[0921] Output: Denoised data

[0922] Step 4: Anomaly detection

[0923] The server then feeds the denoised data into an AI algorithm to detect anomalies, for example using a Long Short-Term Memory (LSTM) algorithm to detect anomalous patterns.

[0924] Input: Denoised data

[0925] Output: Detected anomalies

[0926] Step 5: Alert Generation

[0927] The server logs any detected anomalies and generates alerts if necessary, which are sent to users in real-time via email and / or WebSocket notifications.

[0928] Input: Detected anomaly

[0929] Output: Log entries, alert notifications

[0930] Step 6: Update the predictive model

[0931] The server uses anomalies and historical data to update the predictive model, specifically by retraining the machine learning algorithm using TensorFlow or PyTorch.

[0932] Input: Anomaly data, historical data

[0933] Output: Updated forecast model

[0934] Step 7: Consider the trend

[0935] The server modifies the forecast model to take into account recent trends and seasonal fluctuations, specifically including semi-annual and annual data in the forecast model.

[0936] Input: Updated forecast model

[0937] Output: A revised forecast model

[0938] Step 8: Risk assessment and maintenance plan generation

[0939] The server evaluates future risks based on the modified predictive model, automatically generates appropriate preventive maintenance plans, and determines maintenance timing using an automatic scheduling algorithm.

[0940] Input: Modified forecast model

[0941] Output: Risk assessment results, preventive maintenance plan

[0942] Step 9: Notification

[0943] The device will notify the user of the generated preventive maintenance plan and notify them when it is time to implement it, using a mobile app or desktop notification.

[0944] Input: Preventive Maintenance Plan

[0945] Output: User notification

[0946] Step 10: Perform preventative maintenance

[0947] Users receive notifications from their devices and can then perform preventative maintenance, such as rebooting the system, applying patches, or replacing hardware.

[0948] Input: Notification

[0949] Output: Preventive maintenance performed

[0950] Step 11: Progress Report

[0951] The terminal reports the progress of the performed maintenance to the server. For example, after the maintenance is completed, the terminal presses a confirmation button to send a completion status to the server.

[0952] Input: Preventive maintenance performed

[0953] Output: progress report

[0954] Step 12: Maintenance results analysis and feedback

[0955] The server analyzes the results of the maintenance performed by the user and feeds that data back into the predictive model. Specifically, it compares performance data before and after the maintenance to evaluate the effectiveness, and adds the analysis results to the dataset for retraining.

[0956] Input: Progress report

[0957] Output: Feedback data

[0958] Step 13: Model Improvement

[0959] The server improves the predictive model based on the feedback results, and ultimately the improved predictive model generates a more accurate next preventive maintenance plan.

[0960] Input: Feedback data

[0961] Output: An improved predictive model

[0962] (Application example 1)

[0963] 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."

[0964] Maintaining system reliability and stability is essential for modern electronic payment services. However, conventional monitoring systems are slow to detect anomalies and are unable to quickly generate appropriate maintenance plans. Furthermore, there are limited means of properly notifying users about the timing and methods of preventive maintenance, making it difficult to completely eliminate the risk of breakdowns and service outages.

[0965] 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.

[0966] In this invention, the server includes means for collecting data from an AI sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the results of the anomaly detection, means for automatically generating a maintenance plan for the predicted problems, means for notifying the user of the generated maintenance plan, and means for notifying the user of the anomaly and presenting a maintenance procedure via a smart device. This makes it possible to significantly improve the reliability and stability of payment services by detecting system anomalies early and immediately implementing appropriate preventive maintenance.

[0967] An "AI sensor" is a sensor equipped with artificial intelligence that monitors the status of the environment and system in real time and collects data.

[0968] "Data collection" is the process of compiling information from a system or environment using sensors or other data acquisition methods and storing it for analysis and processing.

[0969] "Anomaly detection" is the process of identifying unusual patterns and values ​​based on collected data and finding signs of system abnormalities or problems.

[0970] "Predicting future problems and failures" is a method of estimating possible future problems and system failures based on past data and current anomaly detection results, and taking measures in advance.

[0971] "Automatic generation of maintenance plans" is the process of automatically formulating specific maintenance procedures and schedules for predicted problems or failures.

[0972] "Notification" is an information transmission means for informing the user of abnormalities and the results of the maintenance plan generation.

[0973] A "smart device" is a mobile terminal or wearable device that has the ability to connect to the Internet and is a device that provides information to users and receives notifications.

[0974] "Feedback" is the process of recording the results and effectiveness of maintenance and preventative care to help improve future predictive models and optimize the system.

[0975] A "machine learning algorithm" is an algorithm that automatically learns from data, finds patterns and relationships, and makes predictions and classifications.

[0976] "Preventive maintenance" refers to maintenance activities carried out in advance to prevent system failures.

[0977] The system for implementing this invention consists of three main components: an AI sensor, a server, and a user terminal (smart device).

[0978] Data collection

[0979] The server collects data in real time from AI sensors, such as the number of transactions, response time, CPU usage, and memory consumption of the payment system, and has built-in functionality to obtain data via sensors and APIs.

[0980] Data analysis

[0981] The server performs anomaly detection based on the collected data. To do this, the first step is to remove noise from the data. After the clean data is generated, anomalies are detected using machine learning algorithms (e.g., RandomForestClassifier). When an anomaly is detected, the information is saved as a log and an alert is generated if necessary.

[0982] Generate future forecasts and maintenance plans

[0983] The server predicts potential future problems and failures based on the anomaly detection results. By analyzing past data and continually updating the prediction model, trends and seasonal fluctuations are taken into account. For predicted problems, the server automatically generates an appropriate maintenance plan.

[0984] Notifications and Feedback

[0985] The generated maintenance plan is sent to the user's smart device (such as a smartphone or tablet). The user can receive this notification and carry out preventive maintenance on the system. Once the maintenance is complete, the results are fed back to the server and used to improve the accuracy of the predictive model.

[0986] Specific examples

[0987] For example, if a server analyzes real-time data and detects a sudden increase in transaction response time, it will recognize this as an anomaly. Based on this information, the server can predict future load increases or system failures and send notifications to smart devices such as "System load is increasing. Please perform maintenance." Users can receive these notifications and perform maintenance to maintain system stability.

[0988] An example of a prompt is as follows:

[0989] "Please implement a function that analyzes transaction data from payment systems, detects anomalies, and automatically generates appropriate maintenance plans."

[0990] In this way, the present invention builds a system that improves the reliability and stability of electronic payment services by linking AI sensors, servers, and smart devices.

[0991] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0992] Step 1: Data collection

[0993] The server collects data in real time from the AI ​​sensors, including the number of transactions, response time, CPU usage, memory usage, etc. The data is retrieved through an API and sent to the server, providing the raw data needed to accurately monitor the system's operating status.

[0994] Step 2: Save data

[0995] The server stores the collected data in a database for subsequent analysis and to update the predictive model. During this process, data is backed up regularly to ensure data reliability and integrity.

[0996] Step 3: Data analysis

[0997] The server analyzes the data and removes noise. Outliers and missing values ​​are filtered out during the data cleaning process. After that, anomalies are detected using a machine learning algorithm (e.g., RandomForestClassifier) ​​based on the clean data. The clean data is given as input, and the output is the presence or absence of anomalies.

[0998] Step 4: Log anomalies and generate alerts

[0999] If an abnormality is detected, the server records the details in a log and generates an alert if necessary. The log stores detailed information about the abnormality and the time of occurrence. The alert is used for notification in the next step.

[1000] Step 5: Auto-generate maintenance plans

[1001] The server predicts future problems and failures based on the anomaly detection results. This involves statistical trend analysis that takes into account past data and current conditions. Detailed maintenance work to address predicted problems is automatically planned. The maintenance plan is generated as a result of calculations using the predictive model.

[1002] Step 6: Notification

[1003] The server notifies the user of the generated maintenance plan via their smart device (smartphone or tablet), where a push notification is sent via the application to prompt the user to take action. The notification includes specific maintenance procedures and the timing of their implementation.

[1004] Step 7: User performs preventative maintenance

[1005] Users receive notifications and can then perform preventative maintenance as instructed, such as rebooting the system or applying software patches, with the actions taken directly fed back to the next step.

[1006] Step 8: Feedback

[1007] Users provide feedback on the results of their maintenance work to the server, which then uses this feedback to retrain the predictive model, improving the accuracy of future anomaly detection and maintenance planning.

[1008] Through the above steps, the present invention provides a system that improves the reliability and stability of electronic payment services.

[1009] 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.

[1010] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, automatically generates appropriate maintenance plans, and provides more personalized notifications to users by combining it with an emotion engine that recognizes user emotions. This not only improves the reliability of the payment system, but also increases user satisfaction.

[1011] Program processing

[1012] Data collection

[1013] The server receives real-time data from the AI ​​sensors, such as the number of transactions, response time, CPU usage, memory usage, etc. This data is stored in a database and backed up regularly.

[1014] Data analysis

[1015] The server pre-processes the collected data to remove noise and improve the reliability of the data by using filtering algorithms to correct and remove outliers and missing data.

[1016] The server then runs the pre-processed data through AI algorithms to detect anomalies, for example by comparing the current data with normal data patterns from the past to determine whether they are abnormal. If an anomaly is detected, an alert is generated.

[1017] Update the forecast model

[1018] The server updates the forecasting model based on the data, retraining the machine learning model to take into account current trends and seasonal fluctuations, improving forecast accuracy.

[1019] Preventive maintenance plan development

[1020] The server evaluates future risks based on predictive models, automatically generates a maintenance plan that proposes appropriate preventive maintenance work, and defines specific actions.

[1021] The terminal notifies the user of the generated preventive maintenance plan and notifies them of the timing of its implementation. This notification includes the specific content, timing, and method of the maintenance.

[1022] emotion recognition

[1023] Before sending a notification, the device uses an emotion engine to recognize the user's emotions, for example, by reading emotions from facial expressions, voice tone, and text.

[1024] The device personalizes notification messages based on the recognized emotion: if the user has a negative emotion, it provides a more polite and supportive message, and if the user has a positive emotion, it adds a message of thanks or encouragement.

[1025] Maintenance

[1026] Based on notifications from the terminal, the user performs preventive maintenance work, such as rebooting the system, applying patches, or replacing hardware. Once the work is completed, the results are reported to the terminal.

[1027] Feedback and learning from results

[1028] The terminal reports the progress and results of the maintenance work performed by the user to the server, including the work performed, the date and time of the work, and the results.

[1029] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[1030] Specific examples

[1031] Case 1: Real-time anomaly detection and emotion recognition

[1032] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[1033] The server compares this spike with historical data and determines that it is abnormal, and an alert is generated.

[1034] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. For example, if the user is feeling anxious, the notification message will include a message such as, "The system load is increasing rapidly, but please rest assured that we are taking measures to address this."

[1035] Users are notified and can take corrective action to maintain stable system operation.

[1036] Case 2: Predictive maintenance planning and emotion recognition

[1037] The server uses past data to discover trends in the failure of specific components.

[1038] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[1039] The device recognizes the user's emotions before sending a notification, and if the user is expressing positive emotions, it will include a thank-you message such as, "Thank you for your continued cooperation. Please replace parts in advance to ensure safe operation."

[1040] The user is notified and replaces the part to prevent a breakdown.

[1041] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[1042] As described above, the system of the present invention utilizes AI sensors and data analysis, and further combines it with an emotion engine to achieve preventative maintenance and improved reliability of the payment system, while also improving user satisfaction.

[1043] The processing flow will be explained below.

[1044] Step 1:

[1045] The server obtains data from the AI ​​sensors in real time. Specifically, it polls the sensors via API to collect data such as the number of transactions, response time, CPU usage, and memory usage. This data is recorded in a database and is backed up to always reflect the latest status.

[1046] Step 2:

[1047] The server pre-processes the collected data, applying filtering algorithms to correct or remove outliers and missing data, a process that improves the accuracy of the data used for analysis.

[1048] Step 3:

[1049] The server performs anomaly detection based on pre-processed data. It uses AI algorithms to compare the current data with past normal data patterns and determine whether it is anomalous. If an anomaly is detected, it is logged and an alert is generated if necessary.

[1050] Step 4:

[1051] The server predicts future problems and outages based on the anomaly detection results. It uses machine learning models to build predictive models from past data and apply them to real-time data. This prediction is used to assess the risk of system problems occurring.

[1052] Step 5:

[1053] The server automatically generates maintenance plans for predicted problems, creating a maintenance schedule with specific preventative maintenance actions such as part replacement needs and system reboots.

[1054] Step 6:

[1055] Before notifying the user of the maintenance plan received from the server, the device uses an emotion engine to recognize the user's emotions. Emotion recognition is performed through the camera, microphone, text analysis, etc.

[1056] Step 7:

[1057] The device customizes the maintenance notification message based on the user's recognized emotions. For example, if the user is expressing negative emotions, the notification message will include a thoughtful message such as "Don't worry, our technical team is working on it." On the other hand, if the user is expressing positive emotions, the notification message will include words of gratitude and encouragement.

[1058] Step 8:

[1059] Users can perform preventative maintenance tasks based on customized notifications from their devices, such as rebooting the system, applying patches, or replacing hardware, and the results are reported back to the device once the task is complete.

[1060] Step 9:

[1061] The terminal reports the progress and results of the maintenance work performed by the user to the server, including details of the work performed, the date and time of the work, and the results.

[1062] Step 10:

[1063] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[1064] Step 11:

[1065] The server uses the feedback results to further optimize predictive models and maintenance plans, thereby preventing future problems and continuously improving system reliability.

[1066] Example 2

[1067] 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."

[1068] In conventional systems, it is common to detect anomalies and perform preventive maintenance based on the results. However, more efficient and personalized notifications are required to further improve system reliability and user satisfaction. Furthermore, conventional systems have difficulty in providing notifications that take user emotions into account, which can lead to user dissatisfaction. Therefore, there is a need for a system that can detect anomalies in real time, automatically generate appropriate maintenance plans, and personalize notifications based on user emotions.

[1069] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting data from an AI sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the anomaly detection results, means for automatically generating a maintenance plan for the predicted problems, means for notifying the user of the generated maintenance plan, and means for recognizing the user's emotions and personalizing the notification message. This not only enables the system reliability to be improved by detecting anomalies in real time and automatically generating a preventive maintenance plan, but also improves user satisfaction by personalized notifications based on emotion recognition.

[1070] An "AI sensor" is a sensor that uses artificial intelligence technology to collect and analyze data in real time.

[1071] "Real-time" refers to data and information being processed and acquired the moment it is generated.

[1072] "Data collection means" means means for collecting data through sensors or other measuring devices.

[1073] An "anomaly detection means" is a means for detecting abnormal conditions or patterns based on collected data.

[1074] The "prediction means" is a means for predicting future problems or failures based on the anomaly detection results.

[1075] The "automatic maintenance plan generation means" is a means for automatically generating a maintenance plan for a predicted problem.

[1076] The "notification means" is a means for notifying the user of the generated maintenance plan.

[1077] "Emotion recognition means" refers to a means for recognizing and analyzing the user's emotions.

[1078] "Personalization" means customizing content and formats to suit individual users.

[1079] "Past data analysis means" is a means for analyzing data collected in the past and identifying patterns.

[1080] The "prediction model update means" is a means for updating the prediction model to the latest state based on the analyzed data.

[1081] The "maintenance progress storage means" is a means for storing the progress and results of maintenance work.

[1082] The "maintenance result feedback means" is a means for collecting the results of the performed maintenance and reflecting them in the prediction model.

[1083] The "alert generation means" is a means for generating a warning message when an abnormality is detected.

[1084] A "system" is the entire device or network that combines and operates the above means.

[1085] This invention is a system that detects abnormalities in real time and automatically generates appropriate maintenance plans by recognizing the user's emotions and personalizing notification messages.

[1086] The server uses AI sensors to collect data in real time, including the number of transactions, response time, CPU usage, and memory usage, which is then stored in a database and backed up regularly.

[1087] The collected data is pre-processed using a filtering algorithm. The server corrects and removes outliers and missing data to remove noise, improving the reliability of the data. The pre-processed data is compared with past normal data patterns. The server performs anomaly detection using a machine learning model (e.g., TensorFlow model) and generates an alert if an anomaly is detected.

[1088] Based on the anomaly detection results, the server predicts future problems and outages. The predictive model is updated with the latest data and retrained to take into account trends and seasonal fluctuations, improving the accuracy of the predictions.

[1089] The server automatically generates a maintenance plan that proposes preventive maintenance work for predicted problems. The generated maintenance plan includes the content and timing of specific maintenance work (e.g., system restart, part replacement, etc.). The terminal notifies the user of this maintenance plan and informs them of the timing of its implementation.

[1090] Before sending a notification, the device uses an emotion engine to recognize the user's emotions. It uses an emotion recognition engine (e.g., OpenCV + deep learning model) that analyzes emotions from facial expressions, voice tone, and text. Based on the recognized emotion, the notification message is personalized. For example, if the user has negative emotions, a polite message of support will be added to the notification text. If the emotion is positive, a message of thanks or encouragement will be added.

[1091] Based on notifications from the terminal, the user performs preventive maintenance tasks, such as rebooting the system, applying patches, and replacing hardware. Once the task is complete, the user reports the results to the terminal.

[1092] The terminal reports the progress and results of the maintenance work performed by the user to the server. The report includes the work performed, the date and time, and the results. The server analyzes the reported maintenance results and feeds the data back into the prediction model. The fed-back data is used to improve the accuracy of the next prediction.

[1093] As a specific example, a server may analyze data obtained in real time from an AI sensor to detect a sudden increase in the number of transactions. If an abnormality is detected by comparing it with past data, an alert is generated, and the device uses an emotion engine to recognize emotions from the user's facial expressions and voice. Users who express anxiety may receive a thoughtful notification such as, "The system load is increasing rapidly, but please rest assured that we are taking measures." Users can then take measures upon receiving the notification and maintain stable system operation.

[1094] An example of a prompt sentence to be input to the generative AI model is, "Analyze the data obtained from the AI ​​sensor and detect anomalies. If an anomaly is detected, automatically generate an appropriate maintenance plan and recognize the user's emotions to personalize the notification message."

[1095] By doing all of this, the system can improve system reliability and user satisfaction by utilizing AI sensors and data analysis to detect anomalies in real time, automatically generating appropriate maintenance plans, and combining an emotion engine to provide personalized notifications to users.

[1096] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1097] Step 1: Data collection

[1098] The server receives data from the AI ​​sensors in real time, including the number of transactions, response time, CPU usage, memory usage, etc. The input is the data from the AI ​​sensors, and the output is the data stored in the database.

[1099] Specific behavior:

[1100] The server receives data packets sent from the AI ​​sensor.

[1101] The received data is written to a database and the data is backed up to storage at regular intervals.

[1102] Step 2: Data Preprocessing

[1103] The server pre-processes the retrieved data: the input is the raw data retrieved from the database, and the output is the denoised data.

[1104] Specific behavior:

[1105] The server runs filtering algorithms to detect and correct outliers and missing data in the data.

[1106] Remove noise from the data set, for example by using a low-pass filter.

[1107] Step 3: Anomaly detection

[1108] The server uses the pre-processed data to perform anomaly detection, where the input is the pre-processed data and the output is the anomaly detection results and alert messages.

[1109] Specific behavior:

[1110] The server loads a machine learning model (e.g., a TensorFlow model) that has been trained on past normal data.

[1111] Newly acquired data is input into the model and a prediction is made as to whether it is an anomaly.

[1112] If an abnormality is detected, an alert message is generated and sent to the system administrator.

[1113] Step 4: Update the predictive model

[1114] The server updates the predictive model based on the latest data. The input is past data and current data, and the output is the updated predictive model.

[1115] Specific behavior:

[1116] The server extracts the last six months' worth of data from the database.

[1117] Use the extracted data to retrain machine learning models.

[1118] Save the retrained model for the next anomaly detection.

[1119] Step 5: Auto-generate maintenance plans

[1120] The server generates a maintenance plan based on the predictive model: the input is the updated predictive model and the anomaly detection results, and the output is a maintenance plan that suggests specific preventive maintenance actions.

[1121] Specific behavior:

[1122] The server uses a predictive model to calculate when the next maintenance is required.

[1123] Generate a list of specific maintenance tasks (e.g., part replacement, system reboot, etc.).

[1124] Generate notifications to send maintenance plans to users.

[1125] Step 6: Emotion Recognition and Notification Personalization

[1126] The device uses an emotion engine to recognize the user's emotions before sending a notification. The input is the user's facial expression, voice, and text information, and the output is a personalized notification message.

[1127] Specific behavior:

[1128] The device uses a camera and microphone to capture the user's facial expressions and voice.

[1129] Use an emotion recognition engine (e.g. OpenCV + deep learning model) to recognize user emotions.

[1130] If the user expresses negative emotions, add a "polite support message" to the notification text. If the user expresses positive emotions, add a message of thanks or encouragement.

[1131] Step 7: Perform maintenance

[1132] The user performs preventive maintenance work based on notifications from the terminal. The input is the maintenance plan notification from the terminal, and the output is the results of the performed maintenance.

[1133] Specific behavior:

[1134] The user follows the instructions on the terminal screen to perform maintenance tasks such as restarting the system and replacing parts.

[1135] After the work is completed, the results of the work (e.g., part replacement date and time, whether there were any problems) are entered into the input form on the terminal.

[1136] This data is transmitted from the terminal to the server.

[1137] Step 8: Feedback and learning from results

[1138] The terminal sends the maintenance results reported by the user to the server, which analyzes the received data and feeds it back into the predictive model. The input is the maintenance results reported by the user, and the output is an improved predictive model.

[1139] Specific behavior:

[1140] The terminal calls an API that sends the user's input data to the server.

[1141] The server stores the received data in a database and uses it the next time the predictive model is trained.

[1142] Retrain the predictive model based on the results of the maintenance performed to improve the model's accuracy.

[1143] Through these steps, the system can detect anomalies in real time, develop preventive maintenance plans, and provide personalized notifications based on user sentiment.

[1144] (Application example 2)

[1145] 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."

[1146] In current autonomous vehicles, system anomaly detection and maintenance planning are important. However, automated notifications without considering passenger emotions can cause anxiety and dissatisfaction. This can result in lower user satisfaction and a loss of trust in the system. Furthermore, conventional systems lack the ability to recognize emotions in real time and generate personalized notifications, resulting in a lack of consideration for passengers.

[1147] 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.

[1148] In this invention, the server includes means for collecting data from the AI ​​sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the anomaly detection results, means for automatically generating a maintenance plan for the predicted problems, means for notifying the generated maintenance plan, means for recognizing passenger emotions, and means for personalizing notification messages based on the recognized emotions. This enables not only real-time anomaly detection and predictive maintenance, but also personalized notifications according to passenger emotions.

[1149] An "AI sensor" is a device that uses artificial intelligence technology to measure and collect information on the environment and equipment status in real time.

[1150] "Real-time" is a concept that refers to processing and analysis being carried out immediately at the moment data is generated.

[1151] "Data collection" is the process of obtaining information from devices such as AI sensors and incorporating it into the system.

[1152] "Anomaly detection" is the process of identifying unusual behaviors or conditions based on collected data.

[1153] "Predicting future problems and outages" is the process of assessing the likelihood of future risks and outages based on past data.

[1154] A "maintenance plan" is a schedule of maintenance work required to keep equipment and systems in good condition.

[1155] "Automatically generating" means that the system generates plans and data according to certain rules and algorithms without user intervention.

[1156] "Notifying" is the act of conveying information or warnings to a user.

[1157] "Recognizing passenger emotions" is the process of determining the user's emotional state through facial and voice analysis.

[1158] "Personalization" refers to tailoring the content of services and messages provided to each user based on their characteristics and preferences.

[1159] The system that realizes this invention collects data in real time from AI sensors, detects anomalies, and automatically generates predictive maintenance plans. It also has the ability to recognize passenger emotions and provide personalized notifications based on those emotions. The specific implementation method for this system is described below.

[1160] First, the server receives real-time data from the AI ​​sensors. This data includes information about each vehicle component and its operating status, such as temperature, vibration, transaction count, and response time. The data is stored in a database and regularly backed up to ensure reliability.

[1161] The server then pre-processes the collected data. Specifically, it uses filtering algorithms to remove noise and correct or remove outliers and missing data, improving the reliability of the data. The pre-processed data is then passed through an AI algorithm for anomaly detection. It compares the current data with past normal data patterns to determine whether it is abnormal, and generates an alert if an anomaly is detected.

[1162] The server then updates the predictive model based on the data. It retrains the machine learning model, taking into account the latest trends and seasonal fluctuations, improving prediction accuracy. This automatically generates a maintenance plan that suggests appropriate preventive maintenance work. The generated maintenance plan includes specific actions and their implementation timing, and is notified to the user via their device.

[1163] Before sending a notification, the device uses an emotion engine to recognize the user's emotions. Emotion recognition involves analyzing camera footage and audio captured by the microphone to read emotions from facial expressions and voice tone. The device then personalizes the notification message based on the recognized emotion. For example, if the user has negative emotions, it will provide a more polite, supportive message, and if the user has positive emotions, it will include a message of thanks or encouragement.

[1164] As a concrete example, consider the case where an autonomous vehicle detects an abnormality while driving. The server analyzes data obtained from sensors and detects the abnormality. If the emotion engine recognizes the passenger's anxiety, the terminal will notify the passenger, "The system has detected an abnormality, but please rest assured that we are taking measures."

[1165] Additionally, if a part needs to be replaced in advance based on a predictive maintenance plan, the server will notify the passenger when the next breakdown is predicted and suggest replacing the part in advance. If the emotion engine recognizes a passenger's positive emotion, it will notify them with a thank-you message saying, "Thank you for your pleasant ride. We will carry out maintenance in advance."

[1166] Examples of prompts include the following:

[1167] Write a Python program for an anomaly detection and maintenance scheduler for an autonomous vehicle, as well as a system that recognizes passenger emotions and provides personalized notifications. Use a machine learning model for anomaly detection and a pre-trained emotion recognition model for emotion recognition. Notify passengers with appropriate messages based on the results.

[1168] This invention combines AI sensors and an emotion recognition engine to enable real-time anomaly detection and predictive maintenance in autonomous vehicles, as well as personalized notifications for passengers, thereby improving passenger safety and satisfaction.

[1169] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1170] Step 1:

[1171] The server collects data in real time from the AI ​​sensors. The input includes data describing the status of each vehicle component and system (e.g., temperature, vibration, number of transactions, response time). The server continuously stores this data in a database, which is backed up periodically. The output is the stored real-time data.

[1172] Step 2:

[1173] The server pre-processes the collected data. This process uses a filtering algorithm to remove noise from the data. The input includes the raw data collected in step 1. Data processing involves correcting and removing outliers and missing data, converting it into reliable data. The output is pre-processed, clean data.

[1174] Step 3:

[1175] The server uses the pre-processed data to detect anomalies. The input includes the cleaned data from step 2. The server uses a machine learning model (anomaly detection model) to compare the current data with historical normal data patterns. Data calculations identify outliers and generate alerts if anomalies are detected. The output is the anomaly detection results and an alert if required.

[1176] Step 4:

[1177] The server predicts future problems and failures based on the anomaly detection results. The input includes the anomaly detection results obtained in step 3. The server uses a predictive model (machine learning algorithm) to evaluate future risks based on past data. Data calculation involves trend analysis of abnormal data. The output is the evaluation result of the predicted problems and failures.

[1178] Step 5:

[1179] The server automatically generates a maintenance plan for the predicted problems. The input includes the prediction data obtained in step 4. The server uses a generative AI model to automatically generate appropriate preventive maintenance work and its schedule. Data calculations determine the ideal maintenance time and the required work content. The output is an automatically generated maintenance plan.

[1180] Step 6:

[1181] The terminal recognizes passengers' emotions before notifying the user of the generated maintenance plan. Inputs include camera footage and audio data. The terminal uses an emotion recognition engine (a pre-trained emotion recognition model) to read emotions from facial expressions and voice tone. Emotion analysis is performed as data processing. The output is the recognized emotion data.

[1182] Step 7:

[1183] The device personalizes the notification message based on the recognized emotion. The input includes the emotion data obtained in step 6 and the maintenance plan generated in step 5. The device uses the emotion data to customize the maintenance notification message. Data processing involves adding corresponding wording according to the user's emotion. The output is a personalized notification message.

[1184] Step 8:

[1185] The user performs preventive maintenance based on the personalized notification sent from the terminal. The input includes the personalized notification message created in step 7. The user performs preventive maintenance tasks such as rebooting the system, applying patches, or replacing parts according to the notification. The output is the result of the performed maintenance task.

[1186] Step 9:

[1187] The terminal reports the results of the maintenance performed by the user to the server. The input includes the results of the maintenance work performed in step 8. The terminal sends the report content to the server, and the results are saved. Data processing involves aggregating and organizing the reported data. The output is the maintenance result data saved on the server.

[1188] Step 10:

[1189] The server analyzes the reported maintenance results and feeds them back into the next predictive model. The input includes the maintenance result data saved in step 9. The server analyzes the data to adjust and retrain the predictive model. Data calculations include improving the model to improve prediction accuracy. The output is a predictive model with improved accuracy.

[1190] 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.

[1191] 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.

[1192] 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.

[1193] [Fourth embodiment]

[1194] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1195] 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.

[1196] 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).

[1197] 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.

[1198] 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.

[1199] 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).

[1200] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1201] 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.

[1202] 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.

[1203] 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.

[1204] 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.

[1205] 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.

[1206] 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."

[1207] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, and automatically generates appropriate maintenance plans, thereby improving the reliability of payment systems.

[1208] Program processing

[1209] Data collection

[1210] The server receives real-time data from AI sensors, such as the number of transactions, response time, CPU usage, and memory usage, which are continuously collected.

[1211] The server stores the collected data in a database and backs it up as appropriate.

[1212] Data analysis

[1213] The server analyzes the data received in real time, removes noise, and then runs the analyzed data through an AI algorithm to detect anomalies.

[1214] If an anomaly is detected, the server logs it and generates an alert if necessary.

[1215] Update the forecast model

[1216] The server uses the data to update the forecasting model, taking into account current trends and seasonal fluctuations.

[1217] The server retrains the machine learning algorithm to improve prediction accuracy.

[1218] Preventive maintenance plan development

[1219] The server assesses future risks based on predictive models and automatically generates appropriate preventative maintenance plans based on the assessment results.

[1220] The terminal notifies the user of the generated preventive maintenance plan and informs the user when it should be carried out.

[1221] Maintenance

[1222] Users receive notifications from their devices and can perform preventative maintenance, such as rebooting the system, applying patches, or replacing hardware.

[1223] The terminal reports the progress of the performed maintenance to the server.

[1224] Feedback and learning from results

[1225] The server analyzes the results of the maintenance performed by the user and feeds the data back into the predictive model.

[1226] The server uses the feedback results to further improve the predictive model, which further optimizes the next preventive maintenance plan.

[1227] Specific examples

[1228] Case 1: Real-time anomaly detection

[1229] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[1230] The server compares this with past data and determines that this sudden increase is abnormal.

[1231] The terminal notifies the user of the abnormality and prompts them to implement system load balancing and traffic restrictions.

[1232] Users are notified and can take corrective action to maintain stable system operation.

[1233] Case 2: Predictive maintenance planning

[1234] The server uses past data to discover trends in the failure of specific components.

[1235] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[1236] The user is notified and replaces the part to prevent a breakdown.

[1237] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[1238] In this way, the system of the present invention utilizes AI sensors and data analysis to achieve preventive maintenance and improved reliability of the payment system.

[1239] The processing flow will be explained below.

[1240] Step 1:

[1241] The server receives real-time data from the AI ​​sensor. Specifically, it accesses the sensor through an API and collects data such as the number of transactions, response time, CPU usage, and memory usage. This data is periodically stored in a database.

[1242] Step 2:

[1243] The server pre-processes the collected data and removes noise, specifically by using filtering algorithms to correct and remove outliers and missing data, which improves the reliability of the data.

[1244] Step 3:

[1245] The server runs anomaly detection algorithms on the pre-processed data, for example comparing it with past normal data patterns to determine whether the current data is anomalous, and generates an alert if an anomaly is detected.

[1246] Step 4:

[1247] The server uses the anomaly detection results to predict future problems and outages. Specifically, it uses machine learning models to build predictive models from past data and apply them to real-time data to predict potential problems.

[1248] Step 5:

[1249] The server automatically generates maintenance plans for predicted problems, creating maintenance schedules that suggest specific actions, such as replacing specific parts or rebooting the system.

[1250] Step 6:

[1251] The terminal notifies the user of the maintenance plan received from the server. The notification includes the content, timing, and implementation method of the required maintenance. The user receives the notification and prepares for the maintenance.

[1252] Step 7:

[1253] Based on notifications from the terminal, the user performs preventive maintenance work, such as rebooting the system, applying patches, and replacing hardware. Once the work is completed, the results are reported to the terminal.

[1254] Step 8:

[1255] The terminal reports the progress and results of the maintenance work performed by the user to the server, including the work performed, the date and time of the work, and the results.

[1256] Step 9:

[1257] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[1258] Step 10:

[1259] The server uses the feedback results to further optimize predictive models and maintenance plans, thereby preventing future problems and continuously improving system reliability.

[1260] Example 1

[1261] 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."

[1262] Conventional systems lacked real-time data collection and anomaly detection, resulting in numerous limitations in the accuracy of predictive models and the automatic generation of maintenance plans. In particular, they lacked noise removal, consideration of trend fluctuations, and feedback on the status of preventive maintenance, which ultimately hindered efforts to improve the reliability of payment systems. Furthermore, inappropriate notification timing and alert generation for preventive maintenance plans hindered efficient maintenance.

[1263] 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.

[1264] In this invention, the server includes means for collecting data from the AI ​​sensor in real time, means for removing noise and detecting anomalies, means for recording the anomaly detection results in a log and generating alerts, means for analyzing past data and identifying patterns, means for updating the prediction model to take into account the latest trends and seasonal fluctuations, means for saving the maintenance progress status and analyzing the results, and means for feeding back the analysis results to improve the prediction model, thereby enabling real-time anomaly detection, improved prediction accuracy, and automatic generation and notification of efficient preventive maintenance plans.

[1265] An "AI sensor" is a sensor that uses machine learning algorithms to collect data from the environment in real time and analyze that data.

[1266] "Data collection" is the process in which the server acquires data from the AI ​​sensor in real time and stores it in a database or similar.

[1267] "Denoising" is the process of applying a filtering algorithm to remove unwanted noise from collected data.

[1268] Anomaly detection is the process of analyzing collected and denoised data to find patterns or signs that are out of the ordinary.

[1269] "Alert generation" is the process of creating an alert to notify a user when an anomaly is detected.

[1270] A "predictive model" is a machine learning algorithm or statistical model that uses historical data to forecast future movements and trends.

[1271] "Pattern analysis" is the process of analyzing historical data to find recurring rules or characteristics.

[1272] "Trend consideration" is the process of incorporating recent trends and seasonal fluctuations when updating a forecasting model.

[1273] "Maintenance progress" is information that records and tracks the implementation status of maintenance work such as preventive maintenance and repairs.

[1274] "Feedback" is the process of analyzing maintenance results and using that information to improve predictive models.

[1275] A "preventive maintenance plan" is a plan that automatically generates schedules and procedures for maintenance to be carried out in advance to prevent abnormalities and breakdowns.

[1276] "Notification" is the process by which users are informed of generated preventative maintenance plans and alerts.

[1277] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, and automatically generates appropriate preventive maintenance plans.

[1278] The server first has a means to obtain data such as the number of transactions, response time, CPU usage, and memory usage from the AI ​​sensor in real time. Specifically, the server periodically sends API requests to collect the latest data from the sensor. The collected data is stored in a database such as MySQL or PostgreSQL and is backed up as appropriate.

[1279] Next, the server analyzes the collected data and removes noise. This is done using algorithms such as moving average filters and low-pass filters. The noise-removed data is then analyzed using an AI algorithm to detect anomalies. For example, an LSTM (Long Short-Term Memory) anomaly detection algorithm is used. If an anomaly is detected, the server records the information in a log and notifies the user by sending an email or generating an alert via WebSocket.

[1280] The server then updates the predictive model based on abnormal data. This is done using machine learning libraries such as TensorFlow and PyTorch. The predictive model is updated to take into account the latest trends and seasonal fluctuations. Based on this updated predictive model, the server evaluates future risks and automatically generates appropriate preventive maintenance plans. For example, it uses an automatic scheduling algorithm to determine the timing of maintenance. The generated preventive maintenance plan is then notified to the user via their terminal, informing them of the timing of implementation.

[1281] During the preventive maintenance phase, users receive notifications from their devices and perform preventive maintenance such as rebooting the system, applying patches, and replacing hardware. The devices then report the progress of the maintenance to the server, which then analyzes the results of the user's maintenance and feeds the data back into the predictive model. Based on this feedback, the predictive model is further improved, and the next preventive maintenance plan is generated more precisely.

[1282] Specific examples

[1283] Case 1: Real-time anomaly detection

[1284] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[1285] The server compares this with past data and determines that this sudden increase is abnormal.

[1286] The terminal notifies the user of the abnormality and prompts them to implement system load balancing and traffic restrictions.

[1287] Users are notified and can take corrective action to maintain stable system operation.

[1288] Case 2: Predictive maintenance planning

[1289] The server uses past data to discover trends in the failure of specific components.

[1290] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[1291] The user is notified and replaces the part to prevent a breakdown.

[1292] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[1293] Prompt Sentence Examples

[1294] Prompt example 1: Real-time anomaly detection

[1295] Create a program that detects sudden spikes in the number of transactions obtained from the AI ​​sensor and compares the situation with past data to determine whether it is an abnormality. If an abnormality is detected, notify the user and include content that encourages load balancing or traffic restrictions.

[1296] Example prompt 2: Predictive maintenance plan

[1297] Please generate a program that analyzes the failure trends of specific parts based on past data and notifies the user when the next failure is predicted. Also, please include a program that suggests parts replacement in advance to the user and provides feedback on the results to improve the prediction model.

[1298] As described above, this system utilizes AI sensors and data analysis to achieve preventive maintenance and improve the reliability of the payment system.

[1299] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1300] Step 1: Data Acquisition

[1301] The server receives data from the AI ​​sensors in real time. Specifically, the server periodically sends API requests to collect data from the sensors, such as response time, CPU usage, memory usage, and number of transactions.

[1302] Input: Data from AI sensors

[1303] Output: Real-time data captured

[1304] Step 2: Save data

[1305] The server stores the collected data in a database and performs regular backups. Specifically, data is inserted using SQL queries using MySQL or PostgreSQL, and the data is saved periodically using a backup script.

[1306] Input: Real-time data captured

[1307] Output: Data stored in the database

[1308] Step 3: Noise reduction

[1309] The server removes noise from the collected data by applying a moving average filter or low-pass filter to remove unnecessary data.

[1310] Input: Data stored in the database

[1311] Output: Denoised data

[1312] Step 4: Anomaly detection

[1313] The server then feeds the denoised data into an AI algorithm to detect anomalies, for example using a Long Short-Term Memory (LSTM) algorithm to detect anomalous patterns.

[1314] Input: Denoised data

[1315] Output: Detected anomalies

[1316] Step 5: Alert Generation

[1317] The server logs any detected anomalies and generates alerts if necessary, which are sent to users in real-time via email and / or WebSocket notifications.

[1318] Input: Detected anomaly

[1319] Output: Log entries, alert notifications

[1320] Step 6: Update the predictive model

[1321] The server uses anomalies and historical data to update the predictive model, specifically by retraining the machine learning algorithm using TensorFlow or PyTorch.

[1322] Input: Anomaly data, historical data

[1323] Output: Updated forecast model

[1324] Step 7: Consider the trend

[1325] The server modifies the forecast model to take into account recent trends and seasonal fluctuations, specifically including semi-annual and annual data in the forecast model.

[1326] Input: Updated forecast model

[1327] Output: A revised forecast model

[1328] Step 8: Risk assessment and maintenance plan generation

[1329] The server evaluates future risks based on the modified predictive model, automatically generates appropriate preventive maintenance plans, and determines maintenance timing using an automatic scheduling algorithm.

[1330] Input: Modified forecast model

[1331] Output: Risk assessment results, preventive maintenance plan

[1332] Step 9: Notification

[1333] The device will notify the user of the generated preventive maintenance plan and notify them when it is time to implement it, using a mobile app or desktop notification.

[1334] Input: Preventive Maintenance Plan

[1335] Output: User notification

[1336] Step 10: Perform preventative maintenance

[1337] Users receive notifications from their devices and can then perform preventative maintenance, such as rebooting the system, applying patches, or replacing hardware.

[1338] Input: Notification

[1339] Output: Preventive maintenance performed

[1340] Step 11: Progress Report

[1341] The terminal reports the progress of the performed maintenance to the server. For example, after the maintenance is completed, the terminal presses a confirmation button to send a completion status to the server.

[1342] Input: Preventive maintenance performed

[1343] Output: progress report

[1344] Step 12: Maintenance results analysis and feedback

[1345] The server analyzes the results of the maintenance performed by the user and feeds that data back into the predictive model. Specifically, it compares performance data before and after the maintenance to evaluate the effectiveness, and adds the analysis results to the dataset for retraining.

[1346] Input: Progress report

[1347] Output: Feedback data

[1348] Step 13: Model Improvement

[1349] The server improves the predictive model based on the feedback results, and ultimately the improved predictive model generates a more accurate next preventive maintenance plan.

[1350] Input: Feedback data

[1351] Output: An improved predictive model

[1352] (Application example 1)

[1353] 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."

[1354] Maintaining system reliability and stability is essential for modern electronic payment services. However, conventional monitoring systems are slow to detect anomalies and are unable to quickly generate appropriate maintenance plans. Furthermore, there are limited means of properly notifying users about the timing and methods of preventive maintenance, making it difficult to completely eliminate the risk of breakdowns and service outages.

[1355] 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.

[1356] In this invention, the server includes means for collecting data from an AI sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the results of the anomaly detection, means for automatically generating a maintenance plan for the predicted problems, means for notifying the user of the generated maintenance plan, and means for notifying the user of the anomaly and presenting a maintenance procedure via a smart device. This makes it possible to significantly improve the reliability and stability of payment services by detecting system anomalies early and immediately implementing appropriate preventive maintenance.

[1357] An "AI sensor" is a sensor equipped with artificial intelligence that monitors the status of the environment and system in real time and collects data.

[1358] "Data collection" is the process of compiling information from a system or environment using sensors or other data acquisition methods and storing it for analysis and processing.

[1359] "Anomaly detection" is the process of identifying unusual patterns and values ​​based on collected data and finding signs of system abnormalities or problems.

[1360] "Predicting future problems and failures" is a method of estimating possible future problems and system failures based on past data and current anomaly detection results, and taking measures in advance.

[1361] "Automatic generation of maintenance plans" is the process of automatically formulating specific maintenance procedures and schedules for predicted problems or failures.

[1362] "Notification" is an information transmission means for informing the user of abnormalities and the results of the maintenance plan generation.

[1363] A "smart device" is a mobile terminal or wearable device that has the ability to connect to the Internet and is a device that provides information to users and receives notifications.

[1364] "Feedback" is the process of recording the results and effectiveness of maintenance and preventative care to help improve future predictive models and optimize the system.

[1365] A "machine learning algorithm" is an algorithm that automatically learns from data, finds patterns and relationships, and makes predictions and classifications.

[1366] "Preventive maintenance" refers to maintenance activities carried out in advance to prevent system failures.

[1367] The system for implementing this invention consists of three main components: an AI sensor, a server, and a user terminal (smart device).

[1368] Data collection

[1369] The server collects data in real time from AI sensors, such as the number of transactions, response time, CPU usage, and memory consumption of the payment system, and has built-in functionality to obtain data via sensors and APIs.

[1370] Data analysis

[1371] The server performs anomaly detection based on the collected data. To do this, the first step is to remove noise from the data. After the clean data is generated, anomalies are detected using machine learning algorithms (e.g., RandomForestClassifier). When an anomaly is detected, the information is saved as a log and an alert is generated if necessary.

[1372] Generate future forecasts and maintenance plans

[1373] The server predicts potential future problems and failures based on the anomaly detection results. By analyzing past data and continually updating the prediction model, trends and seasonal fluctuations are taken into account. For predicted problems, the server automatically generates an appropriate maintenance plan.

[1374] Notifications and Feedback

[1375] The generated maintenance plan is sent to the user's smart device (such as a smartphone or tablet). The user can receive this notification and carry out preventive maintenance on the system. Once the maintenance is complete, the results are fed back to the server and used to improve the accuracy of the predictive model.

[1376] Specific examples

[1377] For example, if a server analyzes real-time data and detects a sudden increase in transaction response time, it will recognize this as an anomaly. Based on this information, the server can predict future load increases or system failures and send notifications to smart devices such as "System load is increasing. Please perform maintenance." Users can receive these notifications and perform maintenance to maintain system stability.

[1378] An example of a prompt is as follows:

[1379] "Please implement a function that analyzes transaction data from payment systems, detects anomalies, and automatically generates appropriate maintenance plans."

[1380] In this way, the present invention builds a system that improves the reliability and stability of electronic payment services by linking AI sensors, servers, and smart devices.

[1381] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1382] Step 1: Data collection

[1383] The server collects data in real time from the AI ​​sensors, including the number of transactions, response time, CPU usage, memory usage, etc. The data is retrieved through an API and sent to the server, providing the raw data needed to accurately monitor the system's operating status.

[1384] Step 2: Save data

[1385] The server stores the collected data in a database for subsequent analysis and to update the predictive model. During this process, data is backed up regularly to ensure data reliability and integrity.

[1386] Step 3: Data analysis

[1387] The server analyzes the data and removes noise. Outliers and missing values ​​are filtered out during the data cleaning process. After that, anomalies are detected using a machine learning algorithm (e.g., RandomForestClassifier) ​​based on the clean data. The clean data is given as input, and the output is the presence or absence of anomalies.

[1388] Step 4: Log anomalies and generate alerts

[1389] If an abnormality is detected, the server records the details in a log and generates an alert if necessary. The log stores detailed information about the abnormality and the time of occurrence. The alert is used for notification in the next step.

[1390] Step 5: Auto-generate maintenance plans

[1391] The server predicts future problems and failures based on the anomaly detection results. This involves statistical trend analysis that takes into account past data and current conditions. Detailed maintenance work to address predicted problems is automatically planned. The maintenance plan is generated as a result of calculations using the predictive model.

[1392] Step 6: Notification

[1393] The server notifies the user of the generated maintenance plan via their smart device (smartphone or tablet), where a push notification is sent via the application to prompt the user to take action. The notification includes specific maintenance procedures and the timing of their implementation.

[1394] Step 7: User performs preventative maintenance

[1395] Users receive notifications and can then perform preventative maintenance as instructed, such as rebooting the system or applying software patches, with the actions taken directly fed back to the next step.

[1396] Step 8: Feedback

[1397] Users provide feedback on the results of their maintenance work to the server, which then uses this feedback to retrain the predictive model, improving the accuracy of future anomaly detection and maintenance planning.

[1398] Through the above steps, the present invention provides a system that improves the reliability and stability of electronic payment services.

[1399] 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.

[1400] This invention is a system that collects data in real time from AI sensors, detects and predicts abnormalities based on this data, automatically generates appropriate maintenance plans, and provides more personalized notifications to users by combining it with an emotion engine that recognizes user emotions. This not only improves the reliability of the payment system, but also increases user satisfaction.

[1401] Program processing

[1402] Data collection

[1403] The server receives real-time data from the AI ​​sensors, such as the number of transactions, response time, CPU usage, memory usage, etc. This data is stored in a database and backed up regularly.

[1404] Data analysis

[1405] The server pre-processes the collected data to remove noise and improve the reliability of the data by using filtering algorithms to correct and remove outliers and missing data.

[1406] The server then runs the pre-processed data through AI algorithms to detect anomalies, for example by comparing the current data with normal data patterns from the past to determine whether they are abnormal. If an anomaly is detected, an alert is generated.

[1407] Update the forecast model

[1408] The server updates the forecasting model based on the data, retraining the machine learning model to take into account current trends and seasonal fluctuations, improving forecast accuracy.

[1409] Preventive maintenance plan development

[1410] The server evaluates future risks based on predictive models, automatically generates a maintenance plan that proposes appropriate preventive maintenance work, and defines specific actions.

[1411] The terminal notifies the user of the generated preventive maintenance plan and notifies them of the timing of its implementation. This notification includes the specific content, timing, and method of the maintenance.

[1412] emotion recognition

[1413] Before sending a notification, the device uses an emotion engine to recognize the user's emotions, for example, by reading emotions from facial expressions, voice tone, and text.

[1414] The device personalizes notification messages based on the recognized emotion: if the user has a negative emotion, it provides a more polite and supportive message, and if the user has a positive emotion, it adds a message of thanks or encouragement.

[1415] Maintenance

[1416] Based on notifications from the terminal, the user performs preventive maintenance work, such as rebooting the system, applying patches, or replacing hardware. Once the work is completed, the results are reported to the terminal.

[1417] Feedback and learning from results

[1418] The terminal reports the progress and results of the maintenance work performed by the user to the server, including the work performed, the date and time of the work, and the results.

[1419] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[1420] Specific examples

[1421] Case 1: Real-time anomaly detection and emotion recognition

[1422] The server analyzes the data obtained from the AI ​​sensor and detects sudden increases in the number of transactions.

[1423] The server compares this spike with historical data and determines that it is abnormal, and an alert is generated.

[1424] The device uses an emotion engine to recognize emotions from the user's facial expressions and voice. For example, if the user is feeling anxious, the notification message will include a message such as, "The system load is increasing rapidly, but please rest assured that we are taking measures to address this."

[1425] Users are notified and can take corrective action to maintain stable system operation.

[1426] Case 2: Predictive maintenance planning and emotion recognition

[1427] The server uses past data to discover trends in the failure of specific components.

[1428] The server notifies the user when the next failure is predicted and suggests replacing parts in advance.

[1429] The device recognizes the user's emotions before sending a notification, and if the user is expressing positive emotions, it will include a thank-you message such as, "Thank you for your continued cooperation. Please replace parts in advance to ensure safe operation."

[1430] The user is notified and replaces the part to prevent a breakdown.

[1431] The terminal reports the results of the performed part replacement to the server, which then feeds back the data to improve the prediction model.

[1432] As described above, the system of the present invention utilizes AI sensors and data analysis, and further combines it with an emotion engine to achieve preventative maintenance and improved reliability of the payment system, while also improving user satisfaction.

[1433] The processing flow will be explained below.

[1434] Step 1:

[1435] The server obtains data from the AI ​​sensors in real time. Specifically, it polls the sensors via API to collect data such as the number of transactions, response time, CPU usage, and memory usage. This data is recorded in a database and is backed up to always reflect the latest status.

[1436] Step 2:

[1437] The server pre-processes the collected data, applying filtering algorithms to correct or remove outliers and missing data, a process that improves the accuracy of the data used for analysis.

[1438] Step 3:

[1439] The server performs anomaly detection based on pre-processed data. It uses AI algorithms to compare the current data with past normal data patterns and determine whether it is anomalous. If an anomaly is detected, it is logged and an alert is generated if necessary.

[1440] Step 4:

[1441] The server predicts future problems and outages based on the anomaly detection results. It uses machine learning models to build predictive models from past data and apply them to real-time data. This prediction is used to assess the risk of system problems occurring.

[1442] Step 5:

[1443] The server automatically generates maintenance plans for predicted problems, creating a maintenance schedule with specific preventative maintenance actions such as part replacement needs and system reboots.

[1444] Step 6:

[1445] Before notifying the user of the maintenance plan received from the server, the device uses an emotion engine to recognize the user's emotions. Emotion recognition is performed through the camera, microphone, text analysis, etc.

[1446] Step 7:

[1447] The device customizes the maintenance notification message based on the user's recognized emotions. For example, if the user is expressing negative emotions, the notification message will include a thoughtful message such as "Don't worry, our technical team is working on it." On the other hand, if the user is expressing positive emotions, the notification message will include words of gratitude and encouragement.

[1448] Step 8:

[1449] Users can perform preventative maintenance tasks based on customized notifications from their devices, such as rebooting the system, applying patches, or replacing hardware, and the results are reported back to the device once the task is complete.

[1450] Step 9:

[1451] The terminal reports the progress and results of the maintenance work performed by the user to the server, including details of the work performed, the date and time of the work, and the results.

[1452] Step 10:

[1453] The server analyzes the reported maintenance results and feeds the data back into the predictive model, which is used to improve the accuracy of the next prediction.

[1454] Step 11:

[1455] The server uses the feedback results to further optimize predictive models and maintenance plans, thereby preventing future problems and continuously improving system reliability.

[1456] Example 2

[1457] 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."

[1458] In conventional systems, it is common to detect anomalies and perform preventive maintenance based on the results. However, more efficient and personalized notifications are required to further improve system reliability and user satisfaction. Furthermore, conventional systems have difficulty in providing notifications that take user emotions into account, which can lead to user dissatisfaction. Therefore, there is a need for a system that can detect anomalies in real time, automatically generate appropriate maintenance plans, and personalize notifications based on user emotions.

[1459] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting data from an AI sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the anomaly detection results, means for automatically generating a maintenance plan for the predicted problems, means for notifying the user of the generated maintenance plan, and means for recognizing the user's emotions and personalizing the notification message. This not only enables the system reliability to be improved by detecting anomalies in real time and automatically generating a preventive maintenance plan, but also improves user satisfaction by personalized notifications based on emotion recognition.

[1460] An "AI sensor" is a sensor that uses artificial intelligence technology to collect and analyze data in real time.

[1461] "Real-time" refers to data and information being processed and acquired the moment it is generated.

[1462] "Data collection means" means means for collecting data through sensors or other measuring devices.

[1463] An "anomaly detection means" is a means for detecting abnormal conditions or patterns based on collected data.

[1464] The "prediction means" is a means for predicting future problems or failures based on the anomaly detection results.

[1465] The "automatic maintenance plan generation means" is a means for automatically generating a maintenance plan for a predicted problem.

[1466] The "notification means" is a means for notifying the user of the generated maintenance plan.

[1467] "Emotion recognition means" refers to a means for recognizing and analyzing the user's emotions.

[1468] "Personalization" means customizing content and formats to suit individual users.

[1469] "Past data analysis means" is a means for analyzing data collected in the past and identifying patterns.

[1470] The "prediction model update means" is a means for updating the prediction model to the latest state based on the analyzed data.

[1471] The "maintenance progress storage means" is a means for storing the progress and results of maintenance work.

[1472] The "maintenance result feedback means" is a means for collecting the results of the performed maintenance and reflecting them in the prediction model.

[1473] The "alert generation means" is a means for generating a warning message when an abnormality is detected.

[1474] A "system" is the entire device or network that combines and operates the above means.

[1475] This invention is a system that detects abnormalities in real time and automatically generates appropriate maintenance plans by recognizing the user's emotions and personalizing notification messages.

[1476] The server uses AI sensors to collect data in real time, including the number of transactions, response time, CPU usage, and memory usage, which is then stored in a database and backed up regularly.

[1477] The collected data is pre-processed using a filtering algorithm. The server corrects and removes outliers and missing data to remove noise, improving the reliability of the data. The pre-processed data is compared with past normal data patterns. The server performs anomaly detection using a machine learning model (e.g., TensorFlow model) and generates an alert if an anomaly is detected.

[1478] Based on the anomaly detection results, the server predicts future problems and outages. The predictive model is updated with the latest data and retrained to take into account trends and seasonal fluctuations, improving the accuracy of the predictions.

[1479] The server automatically generates a maintenance plan that proposes preventive maintenance work for predicted problems. The generated maintenance plan includes the content and timing of specific maintenance work (e.g., system restart, part replacement, etc.). The terminal notifies the user of this maintenance plan and informs them of the timing of its implementation.

[1480] Before sending a notification, the device uses an emotion engine to recognize the user's emotions. It uses an emotion recognition engine (e.g., OpenCV + deep learning model) that analyzes emotions from facial expressions, voice tone, and text. Based on the recognized emotion, the notification message is personalized. For example, if the user has negative emotions, a polite message of support will be added to the notification text. If the emotion is positive, a message of thanks or encouragement will be added.

[1481] Based on notifications from the terminal, the user performs preventive maintenance tasks, such as rebooting the system, applying patches, and replacing hardware. Once the task is complete, the user reports the results to the terminal.

[1482] The terminal reports the progress and results of the maintenance work performed by the user to the server. The report includes the work performed, the date and time, and the results. The server analyzes the reported maintenance results and feeds the data back into the prediction model. The fed-back data is used to improve the accuracy of the next prediction.

[1483] As a specific example, a server may analyze data obtained in real time from an AI sensor to detect a sudden increase in the number of transactions. If an abnormality is detected by comparing it with past data, an alert is generated, and the device uses an emotion engine to recognize emotions from the user's facial expressions and voice. Users who express anxiety may receive a thoughtful notification such as, "The system load is increasing rapidly, but please rest assured that we are taking measures." Users can then take measures upon receiving the notification and maintain stable system operation.

[1484] An example of a prompt sentence to be input to the generative AI model is, "Analyze the data obtained from the AI ​​sensor and detect anomalies. If an anomaly is detected, automatically generate an appropriate maintenance plan and recognize the user's emotions to personalize the notification message."

[1485] By doing all of this, the system can improve system reliability and user satisfaction by utilizing AI sensors and data analysis to detect anomalies in real time, automatically generating appropriate maintenance plans, and combining an emotion engine to provide personalized notifications to users.

[1486] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1487] Step 1: Data collection

[1488] The server receives data from the AI ​​sensors in real time, including the number of transactions, response time, CPU usage, memory usage, etc. The input is the data from the AI ​​sensors, and the output is the data stored in the database.

[1489] Specific behavior:

[1490] The server receives data packets sent from the AI ​​sensor.

[1491] The received data is written to a database and the data is backed up to storage at regular intervals.

[1492] Step 2: Data Preprocessing

[1493] The server pre-processes the retrieved data: the input is the raw data retrieved from the database, and the output is the denoised data.

[1494] Specific behavior:

[1495] The server runs filtering algorithms to detect and correct outliers and missing data in the data.

[1496] Remove noise from the data set, for example by using a low-pass filter.

[1497] Step 3: Anomaly detection

[1498] The server uses the pre-processed data to perform anomaly detection, where the input is the pre-processed data and the output is the anomaly detection results and alert messages.

[1499] Specific behavior:

[1500] The server loads a machine learning model (e.g., a TensorFlow model) that has been trained on past normal data.

[1501] Newly acquired data is input into the model and a prediction is made as to whether it is an anomaly.

[1502] If an abnormality is detected, an alert message is generated and sent to the system administrator.

[1503] Step 4: Update the predictive model

[1504] The server updates the predictive model based on the latest data. The input is past data and current data, and the output is the updated predictive model.

[1505] Specific behavior:

[1506] The server extracts the last six months' worth of data from the database.

[1507] Use the extracted data to retrain machine learning models.

[1508] Save the retrained model for the next anomaly detection.

[1509] Step 5: Auto-generate maintenance plans

[1510] The server generates a maintenance plan based on the predictive model: the input is the updated predictive model and the anomaly detection results, and the output is a maintenance plan that suggests specific preventive maintenance actions.

[1511] Specific behavior:

[1512] The server uses a predictive model to calculate when the next maintenance is required.

[1513] Generate a list of specific maintenance tasks (e.g., part replacement, system reboot, etc.).

[1514] Generate notifications to send maintenance plans to users.

[1515] Step 6: Emotion Recognition and Notification Personalization

[1516] The device uses an emotion engine to recognize the user's emotions before sending a notification. The input is the user's facial expression, voice, and text information, and the output is a personalized notification message.

[1517] Specific behavior:

[1518] The device uses a camera and microphone to capture the user's facial expressions and voice.

[1519] Use an emotion recognition engine (e.g. OpenCV + deep learning model) to recognize user emotions.

[1520] If the user expresses negative emotions, add a "polite support message" to the notification text. If the user expresses positive emotions, add a message of thanks or encouragement.

[1521] Step 7: Perform maintenance

[1522] The user performs preventive maintenance work based on notifications from the terminal. The input is the maintenance plan notification from the terminal, and the output is the results of the performed maintenance.

[1523] Specific behavior:

[1524] The user follows the instructions on the terminal screen to perform maintenance tasks such as restarting the system and replacing parts.

[1525] After the work is completed, the results of the work (e.g., part replacement date and time, whether there were any problems) are entered into the input form on the terminal.

[1526] This data is transmitted from the terminal to the server.

[1527] Step 8: Feedback and learning from results

[1528] The terminal sends the maintenance results reported by the user to the server, which analyzes the received data and feeds it back into the predictive model. The input is the maintenance results reported by the user, and the output is an improved predictive model.

[1529] Specific behavior:

[1530] The terminal calls an API that sends the user's input data to the server.

[1531] The server stores the received data in a database and uses it the next time the predictive model is trained.

[1532] Retrain the predictive model based on the results of the maintenance performed to improve the model's accuracy.

[1533] Through these steps, the system can detect anomalies in real time, develop preventive maintenance plans, and provide personalized notifications based on user sentiment.

[1534] (Application example 2)

[1535] 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."

[1536] In current autonomous vehicles, system anomaly detection and maintenance planning are important. However, automated notifications without considering passenger emotions can cause anxiety and dissatisfaction. This can result in lower user satisfaction and a loss of trust in the system. Furthermore, conventional systems lack the ability to recognize emotions in real time and generate personalized notifications, resulting in a lack of consideration for passengers.

[1537] 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.

[1538] In this invention, the server includes means for collecting data from the AI ​​sensor in real time, means for detecting anomalies based on the collected data, means for predicting future problems and failures based on the anomaly detection results, means for automatically generating a maintenance plan for the predicted problems, means for notifying the generated maintenance plan, means for recognizing passenger emotions, and means for personalizing notification messages based on the recognized emotions. This enables not only real-time anomaly detection and predictive maintenance, but also personalized notifications according to passenger emotions.

[1539] An "AI sensor" is a device that uses artificial intelligence technology to measure and collect information on the environment and equipment status in real time.

[1540] "Real-time" is a concept that refers to processing and analysis being carried out immediately at the moment data is generated.

[1541] "Data collection" is the process of obtaining information from devices such as AI sensors and incorporating it into the system.

[1542] "Anomaly detection" is the process of identifying unusual behaviors or conditions based on collected data.

[1543] "Predicting future problems and outages" is the process of assessing the likelihood of future risks and outages based on past data.

[1544] A "maintenance plan" is a schedule of maintenance work required to keep equipment and systems in good condition.

[1545] "Automatically generating" means that the system generates plans and data according to certain rules and algorithms without user intervention.

[1546] "Notifying" is the act of conveying information or warnings to a user.

[1547] "Recognizing passenger emotions" is the process of determining the user's emotional state through facial and voice analysis.

[1548] "Personalization" refers to tailoring the content of services and messages provided to each user based on their characteristics and preferences.

[1549] The system that realizes this invention collects data in real time from AI sensors, detects anomalies, and automatically generates predictive maintenance plans. It also has the ability to recognize passenger emotions and provide personalized notifications based on those emotions. The specific implementation method for this system is described below.

[1550] First, the server receives real-time data from the AI ​​sensors. This data includes information about each vehicle component and its operating status, such as temperature, vibration, transaction count, and response time. The data is stored in a database and regularly backed up to ensure reliability.

[1551] The server then pre-processes the collected data. Specifically, it uses filtering algorithms to remove noise and correct or remove outliers and missing data, improving the reliability of the data. The pre-processed data is then passed through an AI algorithm for anomaly detection. It compares the current data with past normal data patterns to determine whether it is abnormal, and generates an alert if an anomaly is detected.

[1552] The server then updates the predictive model based on the data. It retrains the machine learning model, taking into account the latest trends and seasonal fluctuations, improving prediction accuracy. This automatically generates a maintenance plan that suggests appropriate preventive maintenance work. The generated maintenance plan includes specific actions and their implementation timing, and is notified to the user via their device.

[1553] Before sending a notification, the device uses an emotion engine to recognize the user's emotions. Emotion recognition involves analyzing camera footage and audio captured by the microphone to read emotions from facial expressions and voice tone. The device then personalizes the notification message based on the recognized emotion. For example, if the user has negative emotions, it will provide a more polite, supportive message, and if the user has positive emotions, it will include a message of thanks or encouragement.

[1554] As a concrete example, consider the case where an autonomous vehicle detects an abnormality while driving. The server analyzes data obtained from sensors and detects the abnormality. If the emotion engine recognizes the passenger's anxiety, the terminal will notify the passenger, "The system has detected an abnormality, but please rest assured that we are taking measures."

[1555] Additionally, if a part needs to be replaced in advance based on a predictive maintenance plan, the server will notify the passenger when the next breakdown is predicted and suggest replacing the part in advance. If the emotion engine recognizes a passenger's positive emotion, it will notify them with a thank-you message saying, "Thank you for your pleasant ride. We will carry out maintenance in advance."

[1556] Examples of prompts include the following:

[1557] Write a Python program for an anomaly detection and maintenance scheduler for an autonomous vehicle, as well as a system that recognizes passenger emotions and provides personalized notifications. Use a machine learning model for anomaly detection and a pre-trained emotion recognition model for emotion recognition. Notify passengers with appropriate messages based on the results.

[1558] This invention combines AI sensors and an emotion recognition engine to enable real-time anomaly detection and predictive maintenance in autonomous vehicles, as well as personalized notifications for passengers, thereby improving passenger safety and satisfaction.

[1559] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1560] Step 1:

[1561] The server collects data in real time from the AI ​​sensors. The input includes data describing the status of each vehicle component and system (e.g., temperature, vibration, number of transactions, response time). The server continuously stores this data in a database, which is backed up periodically. The output is the stored real-time data.

[1562] Step 2:

[1563] The server pre-processes the collected data. This process uses a filtering algorithm to remove noise from the data. The input includes the raw data collected in step 1. Data processing involves correcting and removing outliers and missing data, converting it into reliable data. The output is pre-processed, clean data.

[1564] Step 3:

[1565] The server uses the pre-processed data to detect anomalies. The input includes the cleaned data from step 2. The server uses a machine learning model (anomaly detection model) to compare the current data with historical normal data patterns. Data calculations identify outliers and generate alerts if anomalies are detected. The output is the anomaly detection results and an alert if required.

[1566] Step 4:

[1567] The server predicts future problems and failures based on the anomaly detection results. The input includes the anomaly detection results obtained in step 3. The server uses a predictive model (machine learning algorithm) to evaluate future risks based on past data. Data calculation involves trend analysis of abnormal data. The output is the evaluation result of the predicted problems and failures.

[1568] Step 5:

[1569] The server automatically generates a maintenance plan for the predicted problems. The input includes the prediction data obtained in step 4. The server uses a generative AI model to automatically generate appropriate preventive maintenance work and its schedule. Data calculations determine the ideal maintenance time and the required work content. The output is an automatically generated maintenance plan.

[1570] Step 6:

[1571] The terminal recognizes passengers' emotions before notifying the user of the generated maintenance plan. Inputs include camera footage and audio data. The terminal uses an emotion recognition engine (a pre-trained emotion recognition model) to read emotions from facial expressions and voice tone. Emotion analysis is performed as data processing. The output is the recognized emotion data.

[1572] Step 7:

[1573] The device personalizes the notification message based on the recognized emotion. The input includes the emotion data obtained in step 6 and the maintenance plan generated in step 5. The device uses the emotion data to customize the maintenance notification message. Data processing involves adding corresponding wording according to the user's emotion. The output is a personalized notification message.

[1574] Step 8:

[1575] The user performs preventive maintenance based on the personalized notification sent from the terminal. The input includes the personalized notification message created in step 7. The user performs preventive maintenance tasks such as rebooting the system, applying patches, or replacing parts according to the notification. The output is the result of the performed maintenance task.

[1576] Step 9:

[1577] The terminal reports the results of the maintenance performed by the user to the server. The input includes the results of the maintenance work performed in step 8. The terminal sends the report content to the server, and the results are saved. Data processing involves aggregating and organizing the reported data. The output is the maintenance result data saved on the server.

[1578] Step 10:

[1579] The server analyzes the reported maintenance results and feeds them back into the next predictive model. The input includes the maintenance result data saved in step 9. The server analyzes the data to adjust and retrain the predictive model. Data calculations include improving the model to improve prediction accuracy. The output is a predictive model with improved accuracy.

[1580] 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.

[1581] 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.

[1582] 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.

[1583] 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.

[1584] 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.

[1585] 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.

[1586] 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).

[1587] 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.

[1588] 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."

[1589] 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.

[1590] 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).

[1591] 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.

[1592] 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.

[1593] 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.

[1594] 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.

[1595] 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.

[1596] 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.

[1597] 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.

[1598] 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.

[1599] 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.

[1600] 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.

[1601] The following is further disclosed regarding the above embodiment.

[1602] (Claim 1)

[1603] A means of collecting data in real time from AI sensors, and

[1604] a means for detecting anomalies based on the collected data;

[1605] A means of predicting future problems or failures based on the anomaly detection results;

[1606] a means for automatically generating maintenance plans for predicted problems;

[1607] a means for communicating the generated maintenance plan;

[1608] A system including:

[1609] (Claim 2)

[1610] A means of analyzing historical data to identify patterns;

[1611] a means for updating the predictive model;

[1612] a means for saving the progress of the maintenance;

[1613] The system of claim 1 further comprising:

[1614] (Claim 3)

[1615] A means for prompting a user to perform preventive maintenance based on the generated maintenance plan;

[1616] a means of improving the predictive model by feeding back the results of preventive maintenance;

[1617] The system of claim 1 further comprising:

[1618] "Example 1"

[1619] (Claim 1)

[1620] A means of collecting data in real time from AI sensors, and

[1621] A means for removing noise and detecting anomalies based on the collected data;

[1622] a means for logging anomaly detection results and generating alerts;

[1623] A means of predicting future problems or failures based on the anomaly detection results;

[1624] means for automatically generating appropriate preventative maintenance plans for predicted problems;

[1625] a means for communicating the generated preventive maintenance plan;

[1626] A system including:

[1627] (Claim 2)

[1628] A means of analyzing historical data to identify patterns;

[1629] A means to update forecasting models to take into account current trends and seasonal fluctuations;

[1630] A means for storing the progress of maintenance and analyzing the results;

[1631] A means of improving the predictive model by feeding back the analysis results;

[1632] The system of claim 1 further comprising:

[1633] (Claim 3)

[1634] A means for prompting a user to perform preventive maintenance based on the generated preventive maintenance plan;

[1635] a means of improving the predictive model by feeding back the results of preventive maintenance;

[1636] The system of claim 1 further comprising:

[1637] "Application Example 1"

[1638] (Claim 1)

[1639] A means of collecting data in real time from AI sensors, and

[1640] a means for detecting anomalies based on the collected data;

[1641] A means of predicting future problems or failures based on the anomaly detection results;

[1642] a means for automatically generating maintenance plans for predicted problems;

[1643] a means for communicating the generated maintenance plan;

[1644] a means for notifying a user of an abnormality via a smart device and presenting a maintenance procedure;

[1645] A system including:

[1646] (Claim 2)

[1647] A means of analyzing historical data to identify patterns;

[1648] a means for updating the predictive model;

[1649] a means for saving the progress of the maintenance;

[1650] A means to retrain machine learning algorithms to improve their prediction accuracy; and

[1651] A means for improving the prediction model by feeding back the results of the execution of the generated maintenance plan;

[1652] The system of claim 1 further comprising:

[1653] (Claim 3)

[1654] A means for prompting a user to perform preventive maintenance based on the generated maintenance plan;

[1655] a means of improving the predictive model by feeding back the results of preventive maintenance;

[1656] A means to encourage users to take measures and provide response procedures via smart devices;

[1657] The system of claim 1 further comprising:

[1658] "Example 2: Combining Emotion Engines"

[1659] (Claim 1)

[1660] A means of collecting data in real time from AI sensors, and

[1661] a means for detecting anomalies based on the collected data;

[1662] A means of predicting future problems or failures based on the anomaly detection results;

[1663] a means for automatically generating maintenance plans for predicted problems;

[1664] a means for communicating the generated maintenance plan;

[1665] means for recognizing a user's emotions and personalizing notification messages;

[1666] A system including:

[1667] (Claim 2)

[1668] A means of analyzing historical data to identify patterns;

[1669] a means for updating the predictive model;

[1670] a means for saving the progress of the maintenance;

[1671] A means for improving the prediction model by feeding back the results of maintenance performed by the user;

[1672] The system of claim 1 further comprising:

[1673] (Claim 3)

[1674] A means for prompting a user to perform preventive maintenance based on the generated maintenance plan;

[1675] A means of detecting anomalies in real time and generating alerts;

[1676] The system of claim 1 further comprising:

[1677] "Application example 2 when combining emotion engines"

[1678] (Claim 1)

[1679] A means of collecting data in real time from AI sensors, and

[1680] a means for detecting anomalies based on the collected data;

[1681] A means of predicting future problems or failures based on the anomaly detection results;

[1682] a means for automatically generating maintenance plans for predicted problems;

[1683] a means for communicating the generated maintenance plan;

[1684] a means for recognizing passenger emotions;

[1685] means for personalizing notification messages based on the recognized emotions;

[1686] A system including:

[1687] (Claim 2)

[1688] A means of analyzing historical data to identify patterns;

[1689] a means for updating the predictive model;

[1690] a means for saving the progress of the maintenance;

[1691] Image analysis means for recognizing passenger emotions;

[1692] The system of claim 1 further comprising:

[1693] (Claim 3)

[1694] A means for prompting a user to perform preventive maintenance based on the generated maintenance plan;

[1695] a means of improving the predictive model by feeding back the results of preventive maintenance;

[1696] A means to detect anomalies from data acquired in real time and adjust notifications based on emotion recognition;

[1697] The system of claim 1 further comprising: [Explanation of symbols]

[1698] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting data in real time from AI sensors, and a means for detecting anomalies based on the collected data; A means of predicting future problems or failures based on the anomaly detection results; a means for automatically generating maintenance plans for predicted problems; a means for communicating the generated maintenance plan; A system including:

2. A means of analyzing historical data to identify patterns; a means for updating the predictive model; a means for saving the progress of the maintenance; The system of claim 1 further comprising:

3. A means for prompting a user to perform preventive maintenance based on the generated maintenance plan; a means of improving the predictive model by feeding back the results of preventive maintenance; The system of claim 1 further comprising:

Images