System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting employee's work data; means for analyzing the collected work data; means for determining an optimum break timing and refresh method for the employee based on the analysis result; means for notifying the employee of the determined break timing and refresh method; means for collecting the refresh method executed by the employee and the result thereof; and means for updating a database based on the collected result and using it for the next analysis.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In today's corporate environment, employees are so focused on their work that it is difficult to take appropriate breaks, leading to the accumulation of mental and physical fatigue and stress. This situation not only leads to a decline in work performance, but is also likely to have a negative impact on employees' health. Furthermore, because the appropriate timing and method of taking a break vary from employee to employee, the effectiveness of general break guidance is limited. There is an urgent need to solve these problems and provide an environment where employees can always perform at their best. [Means for solving the problem]

[0005] This invention provides a system for collecting and analyzing employee work data. Specifically, it uses a means for collecting work data such as employee work hours, work content, work location, and stress level, and provides a means for analyzing the collected data. Based on the analysis results, it uses a means for determining optimal break timing and refreshment methods, and notifies employees of the results. Based on this notification, employees take appropriate breaks and perform refreshment methods. Furthermore, it provides a means for collecting and analyzing the refreshment methods performed by employees and their results, and updating the database to improve the accuracy of the next analysis. This makes it possible to efficiently reduce employee fatigue and stress and provide an environment in which employees can always maintain peak performance.

[0006] "Employee work data" refers to information generated in the course of an employee's work, and specifically includes information such as working hours, work content, work location, and stress level.

[0007] "Means of collection" refers to devices and systems for collecting and recording employee work data in real time, specifically PCs, smartphones, wearable devices, etc.

[0008] "Means of analysis" refers to the algorithms and AI models that process collected business data and extract useful information and patterns from that data.

[0009] "Break timing" refers to an appropriate time for an employee to take a temporary break from work.

[0010] "Refreshment methods" refer to specific actions and activities that employees take to reduce mental and physical fatigue in a short period of time and increase work efficiency.

[0011] "Means of notification" refers to devices or systems that inform employees of the decided break timing and methods of relaxation, specifically notification functions on PCs or smartphones.

[0012] "Database" refers to a system for systematically storing collected and analyzed data and using it for subsequent processing and analysis.

[0013] "AI model" refers to a mathematical model that uses machine learning and deep learning to analyze business data and predict employee stress levels and optimal break timing. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] This invention relates to a system that uses AI to analyze employee work data and proposes appropriate break timing and refreshment methods. Specific embodiments of this system are described below.

[0036] System configuration

[0037] The system consists of a terminal that collects employee work data, a server that performs analysis, and a terminal that sends notifications and re-collects data.

[0038] 1. Data Collection

[0039] The server connects to each employee's device (PC, smartphone, wearable device) and collects work data, including work hours, work content, work location, stress level, etc.

[0040] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[0041] Users (employees) input information such as self-evaluations in the form of a questionnaire at the start and end of work, and periodically (for example, every 1-2 hours).

[0042] 2. Data Analysis

[0043] The server receives the collected data and preprocesses it (e.g., normalizes and treats outliers).

[0044] The server inputs the preprocessed data into an AI model to predict stress and fatigue levels, thereby calculating the degree of stress and fatigue for each employee.

[0045] 3. Notifications and refresh suggestions

[0046] Based on the analysis results, the server determines the optimal break timing and method of refreshment for each employee.

[0047] The device notifies employees of break times based on the information received from the server, for example by displaying a message saying, "You will take a 15-minute break in 10 minutes."

[0048] The user receives the notification and can click on an option such as "OK" or "Later" as needed. If the notification is ignored, the server sets the timing for re-notification and prompts the user to take a break again after a certain period of time.

[0049] 4. Refresh execution and data collection

[0050] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[0051] The device collects performance data after a refresh, such as improved keystroke speed and fewer error messages.

[0052] The terminal transmits the execution data to the server.

[0053] 5. Feedback and model improvement

[0054] The server evaluates how effective the refreshment methods were in reducing employees' fatigue and stress.

[0055] The server readjusts the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next proposal.

[0056] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[0057] Specific examples

[0058] Example: Programmer Tanaka

[0059] The device recorded Tanaka's slowing keystroke speed and an increase in error messages, as well as his heart rate and self-assessment data, which revealed that he was experiencing a decline in concentration.

[0060] The server analyzed Tanaka's data and determined that his stress level was high.

[0061] The server notifies Tanaka, "In 10 minutes, you will take a 15-minute break and have a cup of coffee to refresh yourself."

[0062] The user (Mr. Tanaka) checks the notification and takes a break as suggested.

[0063] The device collects Tanaka's performance data again after the break and sends it to the server.

[0064] The server uses this data to update the AI ​​model and prepare for the next more precise analysis.

[0065] In this way, the system collects employees' work data, analyzes it using AI, and notifies them of the best times to take breaks and how to refresh themselves, thereby reducing employee stress and optimizing their performance.

[0066] The processing flow will be explained below.

[0067] The specific processing steps of the program and their operations will be explained below.

[0068] Step 1:

[0069] The device collects and records employees' work data in real time, such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[0070] Step 2:

[0071] The terminal displays a periodic questionnaire to the user (employee) and asks them to enter their self-evaluation (for example, their work progress or current physical condition).

[0072] Step 3:

[0073] The server receives the business data sent from the terminal and stores it in a database.

[0074] Step 4:

[0075] The server checks the received data and performs preprocessing (e.g., data normalization, missing value imputation, and outlier removal).

[0076] Step 5:

[0077] The server then feeds the pre-processed data into an AI model that predicts each employee's stress and fatigue levels using predictive algorithms based on past data and similar patterns.

[0078] Step 6:

[0079] The server determines the optimal break timing for each employee based on the prediction results of the AI ​​model.

[0080] Step 7:

[0081] The server references a database of each employee's hobbies and preferences and determines appropriate ways to refresh themselves, such as drinking coffee or stretching to take a break from work.

[0082] Step 8:

[0083] The server transmits the determined break timing and refreshment method to the terminal.

[0084] Step 9:

[0085] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[0086] Step 10:

[0087] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[0088] Step 11:

[0089] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[0090] Step 12:

[0091] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[0092] Step 13:

[0093] The terminal will again collect user performance data (e.g., improved keystroke speed, reduced error messages) after the refresh.

[0094] Step 14:

[0095] The terminal transmits the refreshed data to the server.

[0096] Step 15:

[0097] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[0098] Step 16:

[0099] The server stores the evaluation results in a database and uses them to retrain the AI ​​model, improving the accuracy of the next analysis.

[0100] By performing the above processing steps, it is possible to analyze employees' work data and suggest appropriate break times and methods for refreshing, thereby reducing employee stress and optimizing work performance.

[0101] Example 1

[0102] 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."

[0103] In today's work environment, many employees accumulate fatigue and stress due to long working hours, which can result in reduced productivity and adverse effects on their mental and physical health. To address this issue, a system that can suggest effective ways to refresh at the appropriate time is needed. However, existing methods have the challenge of making it difficult to provide appropriate timing and methods of refreshment that suit each employee's individual situation.

[0104] 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.

[0105] In this invention, the server includes means for collecting employee work data, pre-processing means for normalizing the collected work data and processing outliers, means for having an AI model for analyzing the pre-processed work data, means for determining the optimal break timing and refreshment method for the employee based on the analysis results, means for notifying the employee of the determined break timing and refreshment method, means for collecting data on the refreshment method performed by the employee and subsequent performance data, and means for readjusting the AI ​​model based on the collected results and updating the algorithm to be useful for the next analysis. This allows for the provision of appropriate break timing and refreshment methods according to the situation of each employee, thereby reducing employee stress and improving productivity.

[0106] "Work data" is data generated when an employee performs their work, including work hours, work content, work location, stress level, keyboard input speed, mouse movements, heart rate, and electrodermal activity.

[0107] "Preprocessing means" refers to the means of normalizing collected business data, removing outliers, and preparing the data in a form that is easy to analyze.

[0108] The "AI model" is a model that uses machine learning algorithms to predict employees' stress levels and fatigue levels from preprocessed data.

[0109] The "analysis method" is a method of inputting preprocessed business data into an AI model to calculate employee stress and fatigue levels.

[0110] "Break timing" refers to employees taking breaks at optimal times during work, and indicates time to reduce stress and fatigue.

[0111] "Refreshment methods" are methods suggested to reduce employees' stress and fatigue, such as drinking coffee or stretching their eyes.

[0112] The "notification means" is a means for informing employees of the break timing and refreshment method determined based on the analysis results.

[0113] "Performance data" is data collected after an employee performs a refresher technique, including recovery of keystroke speed and reduction in error messages.

[0114] "Readjustment measures" are measures for readjusting the AI ​​model based on collected performance data to improve the accuracy of the next analysis results.

[0115] The "algorithm updating means" is a means of updating the analysis algorithm to the latest version using a retuned AI model.

[0116] This invention is a system that collects employee work data, analyzes it using AI, and suggests appropriate break times and ways to refresh. The system continuously monitors employees' work status, aiming to reduce stress and fatigue and improve productivity.

[0117] System configuration

[0118] The system consists of a terminal that collects employee work data, a server that performs analysis, and a terminal that sends notifications and re-collects data.

[0119] Hardware and software used

[0120] Device: An input device such as a PC, smartphone, or wearable device.

[0121] Server: A central computer with data processing and analysis capabilities.

[0122] Software: Data collection software, pre-processing software, AI models, notification software, database management software.

[0123] Data collection

[0124] The server connects to each employee's device and collects work data, including work hours, work content, work location, and stress level.

[0125] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[0126] Users (employees) enter self-evaluations in the form of a questionnaire at the start and end of work, and periodically.

[0127] Data analysis

[0128] The server preprocesses the collected data, normalizes the data, and processes outliers.

[0129] The server inputs the preprocessed data into an AI model to predict the user's stress level and fatigue level.

[0130] Based on the analysis results, the server quantifies and evaluates the user's level of stress and fatigue.

[0131] Notifications and refresh suggestions

[0132] Based on the analysis results, the server determines the optimal timing for breaks and the method of refreshing.

[0133] The device notifies the user of the break based on the information received from the server. For example, it displays a message saying, "Take a 15-minute break in 10 minutes."

[0134] The user can check the notification and choose options such as "OK" or "Later." If the notification is ignored, the server will notify the user again.

[0135] Refresh execution and data collection

[0136] The user performs the suggested refreshment method, such as drinking a cup of coffee or doing eye stretches.

[0137] The device will collect performance data after the refresh, recording things like improved keystroke speed and fewer error messages.

[0138] The terminal retransmits the collected data to the server.

[0139] Feedback and model improvement

[0140] The server evaluates how effective the refreshment method was in reducing the user's fatigue and stress.

[0141] The server readjusts the AI ​​model based on the evaluation results to improve the accuracy of the next analysis results.

[0142] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[0143] Specific examples

[0144] Example: Programmer employee A

[0145] The device records Employee A's slowing keystroke speed and an increase in error messages, and also collects data on his heart rate and self-evaluation questionnaires that indicate his "declining ability to concentrate."

[0146] The server analyzes employee A's data and determines that his stress level is high.

[0147] The server notifies employee A, "Take a 15-minute break in 10 minutes and refresh yourself with some coffee."

[0148] The user (Employee A) checks the notification and takes a break as suggested.

[0149] The terminal collects the performance data of employee A again after the break and transmits it to the server.

[0150] The server updates the AI ​​model based on this data, preparing for the next more precise analysis.

[0151] Prompt Sentence Examples

[0152] Example: Describe a system that analyzes the stress level of programmer employee A based on his work data and suggests optimal break times and ways to refresh himself.

[0153] This system can reduce employee stress and optimize performance by informing them of the best time to take a break and how to refresh themselves based on their individual circumstances.

[0154] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0155] Step 1: Data collection

[0156] The user presses the start button on the terminal and fills out a self-evaluation questionnaire. The entered data is then saved directly on the terminal.

[0157] The device collects user input and real-time task data (such as keyboard typing speed, mouse movements, heart rate, and electrodermal activity), which is temporarily stored on the device.

[0158] The server receives work data periodically sent from the terminal, including work hours, work content, work location, stress level, etc.

[0159] (Input) User surveys, keyboard input speed, mouse movement, heart rate, etc.

[0160] (Output) The collected business data is sent to the server.

[0161] Specifically, when Tanaka starts work, he answers a questionnaire displayed on the device asking, "How are you feeling today?" and the device records his answers.

[0162] Step 2: Data Preprocessing

[0163] The server normalizes the business data received from the device and corrects or removes any abnormal values. For example, if the heart rate is extremely high, it determines that this is noise and corrects the value.

[0164] The server prepares the pre-processed data to be fed into the AI ​​model.

[0165] (Input) Collected business data.

[0166] (Output) Preprocessed data that has been normalized and outlier treated.

[0167] Specifically, the server analyzes Tanaka's heart rate data, and if an unnatural increase is detected, it corrects it to the correct value.

[0168] Step 3: Data analysis

[0169] The server then inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level, for example, assessing stress levels based on a decrease in keyboard typing speed or an increase in heart rate.

[0170] Based on the analysis results, the server quantifies and evaluates the user's level of stress and fatigue.

[0171] (Input) Preprocessed data.

[0172] (Output) Analysis results showing stress levels and fatigue levels.

[0173] Specifically, the server detects a decrease in Tanaka's keyboard input speed and evaluates this as a stress index.

[0174] Step 4: Notifications and refresh suggestions

[0175] The server then determines the optimal timing for rest and how to refresh based on the analysis results. For example, if the heart rate is high, it may recommend five minutes of deep breathing.

[0176] The device receives a notification from the server and displays a message to the user such as, "The next break is in 10 minutes. Have a cup of coffee and refresh yourself."

[0177] The user can review the notification and enter options such as "OK" or "later."

[0178] (Input) Analysis results.

[0179] (Output) Notification of break times and ways to refresh.

[0180] Specifically, the server notifies Tanaka that "we recommend that you take a coffee break in 10 minutes," and Tanaka responds with "Got it."

[0181] Step 5: Refresh execution and data collection

[0182] The user performs the suggested refreshment method (e.g., drinking coffee, stretching the eyes, etc.).

[0183] The terminal collects performance data after the refresh (e.g., recovery of keystroke speed, reduction in error messages).

[0184] The terminal retransmits the collected data to the server.

[0185] (Input) Performance data before the refresh is executed.

[0186] (Output) Performance data after the refresh is performed.

[0187] Specifically, Tanaka drinks coffee to refresh himself, and the data on his subsequent recovery of keyboard typing speed is collected on the device and sent to the server.

[0188] Step 6: Feedback and model improvement

[0189] The server evaluates how effective the refreshment methods were in reducing employees' fatigue and stress.

[0190] The server readjusts the AI ​​model based on the evaluation results to improve the accuracy of the next analysis results.

[0191] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[0192] (Input) Performance data after the refresh is executed.

[0193] (Output) Retuned AI models and updated algorithms.

[0194] Specifically, the server evaluates Tanaka's refreshed data and updates the model, improving the accuracy of suggestions from the next time onwards.

[0195] (Application example 1)

[0196] 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."

[0197] Conventional systems for suggesting breaks and refreshment methods determine the optimal timing by collecting and analyzing employees' work data, but the types of data available are limited, meaning that the suggestions cannot fully reflect the detailed stress and fatigue levels of individual employees.In addition, the limited number of devices available for notification makes it difficult to respond flexibly to the work environment.

[0198] 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.

[0199] In this invention, the server includes means for collecting employee work data, means for analyzing the collected work data, means for determining optimal break timing and refreshment methods for employees based on the analysis results, means for notifying employees of the determined break timing and refreshment methods, means for collecting the refreshment methods performed by employees and their results, means for updating the database based on the collected results and using them for the next analysis, means for collecting eye movements, heart rate, and body movements as work data in real time, and means for notifying workers of break timing and refreshment methods via their smart devices. This allows the system to reflect detailed stress and fatigue levels of employees in real time, making it possible to propose more accurate break timing and refreshment methods.

[0200] "Employee work data" refers to information generated by employees during work, such as time, content, location, heart rate, eye movements, body movements, and stress levels.

[0201] "Means for collection" refers to devices and systems for collecting employee work data in real time, including wearable devices such as smart glasses.

[0202] The "means of analysis" is a system that preprocesses collected business data and uses an AI model to analyze stress levels and fatigue levels.

[0203] The "means for determining" is a system that selects the optimal break timing and refreshment method for employees based on the analysis results.

[0204] The "notification means" refers to a device or application that notifies employees in real time of the determined break timing and refreshment method. Specifically, this is done via a smart device.

[0205] The "means of updating the database based on collected results" refers to a system that records the refresh methods used by employees and the results of those methods, updates the database based on those results, and uses them for the next analysis.

[0206] "Eye movement" is data on employees' eye movements and is an indicator used to evaluate their concentration level and stress level at work.

[0207] "Heart rate" is the number of times an employee's heart beats within a certain period of time, and is an indicator primarily used to assess stress and fatigue levels.

[0208] "Physical movement" refers to information about an employee's physical movement and is an indicator for evaluating work efficiency and fatigue level.

[0209] A "smart device" is a device that has Internet connectivity and is capable of data communication and application execution, and specifically includes smart glasses and smartphones.

[0210] The "AI model" is a model that uses machine learning algorithms to analyze business data and predict employees' stress levels and fatigue levels.

[0211] This invention is a system that uses AI to analyze employees' work data and propose optimal break timing and refreshment methods. The system consists of a means for collecting employees' work data, a means for analyzing the data, a means for determining and notifying break timing and refreshment methods, and a means for collecting and evaluating the refreshment results.

[0212] Hardware Configuration

[0213] The system consists of smart glasses for collecting data, wearable devices, and terminals such as PCs and smartphones. The data is collected on a server and analyzed. Specific devices used include Google Glass and Microsoft HoloLens.

[0214] Software Configuration

[0215] AI models are used for data analysis, and machine learning libraries such as TensorFlow are introduced, while data analysis libraries such as SciPy and NumPy are used for data preprocessing.

[0216] Process Overview

[0217] 1. Data Collection

[0218] The server collects real-time data on employees' eye movements, heart rate, and body movements via smart glasses or wearable devices.

[0219] The data is transmitted from the terminal to a server and recorded in real time.

[0220] 2. Data Analysis

[0221] The server preprocesses the collected data, eliminating outliers and normalizing it.

[0222] The preprocessed data is then fed into an AI model to analyze stress levels and fatigue levels.

[0223] 3. Notifications and refresh suggestions

[0224] Based on the analysis results, the server determines the optimal timing for breaks and the method of refreshing.

[0225] The decision will be communicated to employees via smart glasses or smartphones.

[0226] For example, a notification will appear saying, "Take a 15-minute break in 10 minutes and take some deep breaths."

[0227] 4. Collecting and Evaluating Refresh Results

[0228] After the employee executes the refresh method, the terminal recollects the resulting data.

[0229] The collected data is sent to a server and used by an AI model to evaluate its effectiveness.

[0230] 5. Feedback and model improvement

[0231] The effect of the refresh is evaluated and the AI ​​model is readjusted based on the results.

[0232] The database will be updated with new evaluation results and used for the next analysis.

[0233] Specific examples

[0234] For example, if a factory worker wears smart glasses and their eye movements deviate from a certain pattern, it will be judged as a "high stress level." The same applies if their heart rate rises sharply. In this case, the system will notify the worker to "take a break and take a short walk or stretch." After the worker takes a break as notified, the resulting data will be collected again and fed back into the AI ​​model.

[0235] Prompt Sentence Examples

[0236] Model inputs

[0237] Worker heart rate data:

[0238] [68, 70, 72, 75, 77, 80, 85, 90, 95, 100]

[0239] Worker gaze movement data (eye movement):

[0240] [0.5, 0.6, 1.2, 1.1, 0.8, 0.7, 1.3, 1.4, 1.5, 0.9]

[0241] Expected output

[0242] Your stress levels are high, so I suggest you take a 10-minute break and stretch.

[0243] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0244] Step 1:

[0245] The server collects data on employees' eye movements, heart rates, and body movements in real time via smart glasses or wearable devices. The input is raw data sent from each device, and the output is a business dataset that compiles this data. Specifically, all eye movement information obtained by eye tracking sensors and heart rate information obtained by heart rate sensors are aggregated on the server.

[0246] Step 2:

[0247] The terminal preprocesses the collected business data. The input is the raw data collected in step 1. Specific data processing includes removing outliers, normalizing the data, and removing noise. The output is a normalized dataset that has undergone these preprocessing steps. The server uses SciPy and NumPy to clean and format the data.

[0248] Step 3:

[0249] The server inputs the preprocessed data into an AI model to analyze stress levels and fatigue levels. The input is the normalized dataset generated in step 2. Specific data calculations include predicting stress levels using machine learning algorithms. The output is a numerical value indicating each employee's stress level and fatigue level. The server runs the AI ​​model using TensorFlow.

[0250] Step 4:

[0251] The server determines the optimal break timing and refreshment method based on the analysis results. The input is the analysis result from step 3. In concrete terms, the server determines the break timing and selects an appropriate refreshment method (e.g., stretching or deep breathing). The output is the determined break timing and refreshment method.

[0252] Step 5:

[0253] The terminal notifies the employee of the determined break timing and refreshment method. The input is the result of the decision made in step 4. The specific operation is to display a notification message on the smart glasses or smartphone. The output is a confirmation message of the notification. For example, the message "Take a 15-minute break in 10 minutes and take deep breaths" is displayed.

[0254] Step 6:

[0255] The user (employee) performs the suggested refreshment method based on the notification. The input is the notification message from step 5. Specifically, the employee performs a refreshment method such as stretching or deep breathing. The output is the result data of the refreshment method that was performed.

[0256] Step 7:

[0257] The device again collects data after the refresh method is executed. The input is the result data of the refresh method executed by the user. Specifically, heart rate, eye movement, etc. are collected again. The output is a performance dataset after the refresh.

[0258] Step 8:

[0259] The server evaluates the effectiveness of the refresh method based on the collected post-refresh data. The input is the post-refresh data from step 7. The specific data calculation involves comparing the data before and after the refresh and evaluating areas for improvement. The output is numerical data of the evaluation results.

[0260] Step 9:

[0261] The server readjusts the AI ​​model based on the evaluation results and updates the database. The input is the evaluation result from step 8. Specifically, it adjusts the parameters of the AI ​​model and adds new data points to the database. The output is an improved AI model and an updated database.

[0262] 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.

[0263] This invention relates to a system that uses AI to analyze employee work data and emotional data and suggests appropriate break timing and methods for refreshing. Specific embodiments of this system are described below.

[0264] System configuration

[0265] The system consists of a terminal that collects employees' work data and emotional data, a server that performs the analysis, and a terminal that sends notifications and re-collects the data.

[0266] 1. Data Collection

[0267] The server connects to each employee's device (PC, smartphone, wearable device) and collects work data and emotional data, including work hours, work content, work location, and stress level.

[0268] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[0269] The device uses an emotion engine to perform voice analysis, facial recognition, and text analysis to collect user (employee) emotional data in real time, including emotional states such as anger, sadness, joy, and fear.

[0270] Users (employees) input their emotional state by self-evaluating in the form of a questionnaire at the start and end of work, and periodically (for example, every 1-2 hours).

[0271] 2. Data Analysis

[0272] The server checks the received business data and emotion data and stores them in a database.

[0273] The server performs preprocessing (e.g., data normalization, missing value imputation, and outlier removal).

[0274] The server then inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level. By incorporating emotional data into the analysis results, more accurate predictions can be made.

[0275] 3. Notifications and refresh suggestions

[0276] Based on the analysis results, the server determines the optimal break timing for each employee.

[0277] The server refers to the employee's hobby and preference database and determines the appropriate way to refresh themselves, suggesting, for example, drinking coffee or stretching to take a break from work.

[0278] The server transmits the determined break timing and refreshment method to the terminal.

[0279] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[0280] 4. Refresh execution and data collection

[0281] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[0282] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[0283] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[0284] The terminal collects performance data after the refresh (e.g., recovery of key input speed, reduction in error messages) and sends it to the server.

[0285] 5. Feedback and model improvement

[0286] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[0287] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis.

[0288] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[0289] Specific examples

[0290] Example: Programmer Tanaka

[0291] The device records Tanaka's slowing keystroke speed and rising heart rate, and the emotion engine determines from Tanaka's voice and face that he is "highly stressed" or "feeling angry."

[0292] The server analyzed Tanaka's data and determined that he was in a "high stress state."

[0293] The server notifies Tanaka that he should take a 15-minute break in 10 minutes and also suggests that he should have a cup of coffee to refresh himself.

[0294] The user (Mr. Tanaka) checks the notification and takes a break as suggested.

[0295] The device recollects Tanaka's performance and emotional data after the break and sends it to the server.

[0296] The server updates the AI ​​model based on this data, preparing for the next more precise analysis.

[0297] In this way, the system collects and analyzes employees' work and emotional data, and notifies them of appropriate break times and ways to refresh themselves, thereby reducing employee stress and optimizing their work performance.

[0298] The processing flow will be explained below.

[0299] The specific processing steps of the system are described below.

[0300] Step 1:

[0301] The device collects and records employees' work data (keystroke speed, mouse movements, application usage history, heart rate, electrodermal activity, etc.) in real time.

[0302] Step 2:

[0303] The device uses an emotion engine to perform voice analysis, facial recognition, and text analysis to collect and record the user's (employee's) emotional data in real time. Emotional data includes emotional states such as anger, sadness, joy, and fear.

[0304] Step 3:

[0305] The device displays regular self-evaluations in the form of a questionnaire to the user (employee), prompting them to enter information about their work progress, current physical condition, and emotional state.

[0306] Step 4:

[0307] The terminal transmits the collected business data and emotional data to the server at regular intervals.

[0308] Step 5:

[0309] The server stores the received data in a database and performs preprocessing, which includes data normalization, missing value imputation, and outlier removal.

[0310] Step 6:

[0311] The server inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level. Emotional data is also incorporated into the analysis results to make more accurate predictions.

[0312] Step 7:

[0313] Based on the analysis results, the server determines the optimal break timing for each employee.

[0314] Step 8:

[0315] The server refers to the employee's hobby and preference database and determines the appropriate way to refresh themselves, suggesting, for example, drinking coffee or stretching to take a break from work.

[0316] Step 9:

[0317] The server transmits the determined break timing and refreshment method to the terminal.

[0318] Step 10:

[0319] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[0320] Step 11:

[0321] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[0322] Step 12:

[0323] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[0324] Step 13:

[0325] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[0326] Step 14:

[0327] After the device is refreshed, the user's performance data (e.g., recovery of keystroke speed, reduction in error messages) and emotional data are collected again.

[0328] Step 15:

[0329] The terminal transmits the refreshed data to the server.

[0330] Step 16:

[0331] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[0332] Step 17:

[0333] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis.

[0334] Step 18:

[0335] The server stores the evaluation results and analysis data in a database for subsequent analysis.

[0336] By executing the above processing steps, the system can analyze employees' work data and emotional data and suggest appropriate break times and ways to refresh themselves. This reduces employee stress and optimizes work performance. As a concrete example, the system makes real-time suggestions to programmer Tanaka about when to take a break and how to refresh himself. Even when Tanaka is busy with a project, he can refresh himself at the appropriate time and maintain work efficiency.

[0337] Example 2

[0338] 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."

[0339] In today's business environment, properly managing employee stress and fatigue and improving work efficiency are important issues. However, traditional methods have made it difficult to grasp the state of each employee in real time and suggest appropriate break times and ways to refresh. As a result, many companies are not managing their employees' stress levels properly, resulting in decreased work performance and increased health risks.

[0340] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting work data and emotion data, means for analyzing the collected work data and emotion data, and means for determining optimal break timing and refreshment methods based on the analysis results. This makes it possible to grasp employee stress and fatigue in real time and propose optimal break timing and refreshment methods to each employee.

[0341] "Business data" refers to information generated when employees perform their work, including work hours, work content, work location, keystroke speed, mouse movements, application program usage history, heart rate, etc.

[0342] "Emotional data" refers to information that indicates an employee's emotional state, including emotional states such as anger, sadness, joy, and fear obtained through voice analysis, facial recognition, and text analysis.

[0343] "Analysis results" refers to information obtained after analyzing collected data using AI models, etc., and predicts the stress levels and fatigue levels of each employee.

[0344] "Break timing" refers to the time of break recommended for employees based on the analysis results.

[0345] "Refreshment methods" refer to actions or methods that employees are encouraged to take during their breaks, such as drinking coffee or stretching.

[0346] "Notification means" refers to the means of informing employees of the determined break timing and method of refreshment, and notifications are given via PCs, smartphones, wearable devices, etc.

[0347] A "database" is a repository of information that stores collected data and analysis results and allows reuse as needed.

[0348] A "generative AI model" is a model that includes algorithms for analyzing data and making predictions using AI.

[0349] MODE FOR CARRYING OUT THE INVENTION

[0350] This invention is a system that uses AI to analyze employees' work data and emotional data and suggests appropriate break times and ways to refresh. The system consists of the following main components:

[0351] Data collection

[0352] The device connects

[0353] The server establishes a secure connection with each employee's device (PC, smartphone, wearable device, etc.), enabling real-time data collection. For example, this includes the PCs and smartwatches employees use to do their work.

[0354] Business data collection

[0355] The devices record real-time work data, such as keystroke speed, mouse movements, and application usage history, which is used to gain a detailed understanding of how employees are performing their work. For example, keystroke speed during programming tasks and the type of software they are using are recorded.

[0356] Collecting Emotional Data

[0357] The device uses an emotion engine that performs voice analysis, facial recognition, and text analysis to analyze the user's emotional state in real time and digitize it. For example, the device classifies the user's emotional state into categories such as "stress" or "joy" based on the voice during a call or facial expressions captured via a webcam.

[0358] Enter your self-assessment

[0359] Users enter their emotional state in the form of a self-assessment questionnaire at the start and end of work and at regular intervals. For example, every 1-2 hours, they answer a questionnaire in which they rate their "current level of fatigue" on a scale of 1-5.

[0360] Data analysis

[0361] Receiving and storing data

[0362] The server receives the collected work data and emotion data and stores it in a database. This allows for centralized management, including past data. For example, fluctuations in an employee's stress level over the past month can also be recorded.

[0363] Data Preprocessing

[0364] The server performs preprocessing such as data normalization, missing value imputation, and outlier removal. This process improves the quality of the data to obtain accurate analysis results. For example, incomplete data sets and extreme values ​​are removed.

[0365] Prediction by AI model

[0366] The server inputs the preprocessed data into a generative AI model to predict each employee's stress level and fatigue level. The generative AI model is an algorithm that learns from past data and predicts future conditions. For example, it predicts that "Programmer Tanaka" is currently in a high stress state.

[0367] Notifications and refresh suggestions

[0368] Deciding when to take a break

[0369] Based on the analysis results, the server determines the optimal break timing for each employee. For example, it may decide that "Tanaka-san" should take a 15-minute break in 10 minutes.

[0370] Suggestions for refreshing

[0371] The server refers to a database of employees' hobbies and preferences to determine appropriate ways to refresh themselves, such as "drinking coffee" or "doing eye stretching exercises."

[0372] Sending notifications

[0373] The server then sends the determined break timing and refreshment method to the terminal, which then displays a notification to the employee. For example, a notification such as "Let's have a cup of coffee and refresh yourself" will appear on Tanaka's PC.

[0374] Refresh execution and data collection

[0375] User Verification

[0376] The user checks the notification and selects an option such as "OK" or "later." For example, Tanaka selects "OK."

[0377] Snooze settings

[0378] If "Later" is selected, the server sets the timing of the re-notification and sends the break notification to the terminal again after a certain time. For example, the setting "Re-notify after 30 minutes" is set.

[0379] Execute the refresh method

[0380] The user performs the suggested refreshment method. For example, Tanaka drinks coffee.

[0381] Performance Data Collection

[0382] The device collects performance data after the refresh and sends it to the server. For example, Tanaka's keystroke speed after the refresh is recorded.

[0383] Feedback and model improvement

[0384] Evaluation of effectiveness

[0385] The server evaluates the effectiveness of the refreshment method and analyzes the degree of improvement in stress and fatigue. For example, it evaluates changes in Tanaka's stress level.

[0386] Retraining the AI ​​model

[0387] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis, ensuring that analysis is always based on the latest data, resulting in more accurate predictions.

[0388] Data storage

[0389] The server stores the evaluation results and analysis data in a database for subsequent analysis, which can be used for future data analysis and model improvement.

[0390] Prompt Sentence Examples

[0391] "Design a system that analyzes employees' work data and emotional data to suggest appropriate break times and ways to refresh. The following input data is provided: work hours, work content, work location, keystroke speed, mouse movement, heart rate, voice analysis data, and facial recognition data. Use this data to predict employees' stress levels and fatigue levels."

[0392] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0393] Step 1:

[0394] Data collection

[0395] The server establishes a connection: The server establishes a secure connection with each employee's device (PC, smartphone, wearable device), allowing data to be sent and received in real time.

[0396] Input: Connection request from employee's device.

[0397] Data processing / calculation: Establish a secure connection and perform authentication.

[0398] Output: Establishment of a secure communication channel with the device.

[0399] Specific operation: Tanaka's smartwatch connects to the server and starts sending data.

[0400] Devices collect business data: Devices collect business data in real time, such as keystroke speed, mouse movements, and application usage history.

[0401] Input: Employee operation data (keystrokes, mouse movements, etc.).

[0402] Data processing / calculation: Recording of operational data and converting it into the required format.

[0403] Output: Formatted business data.

[0404] Specific operation: Tanaka's keystroke speed and mouse movements are recorded while he works in Excel.

[0405] Device collects emotional data: The device collects the user's emotional data using voice analysis, facial recognition, and text analysis.

[0406] Input: Audio data, facial image data, text data.

[0407] Data processing / calculation: Emotional states are analyzed using an emotion engine and converted into data.

[0408] Output: Emotion data (e.g. anger, joy, sadness, etc.).

[0409] Specific operation: While Tanaka is on the phone, the emotion of "anger" is determined and recorded as data.

[0410] Users enter self-evaluations: Users answer self-evaluation questionnaires at the start and end of work, and at regular intervals.

[0411] Input: User's self-assessment survey responses.

[0412] Data processing / calculation: Formatting the survey responses for storage in the database.

[0413] Output: Formatted self-assessment data.

[0414] Specific actions: Tanaka inputs his "current level of fatigue" every hour on a 5-point scale.

[0415] Step 2:

[0416] Data analysis

[0417] The server receives and stores the data: The server stores the received business data and emotion data in a database.

[0418] Input: Business and sentiment data collected in real time.

[0419] Data processing / calculation: Convert into a format that can be saved in a database.

[0420] Output: Business and sentiment data stored in a database.

[0421] Specific action: Tanaka's data is registered in the database.

[0422] The server performs data preprocessing: The server normalizes the data, imputes missing values, removes outliers, etc.

[0423] Input: Stored business and sentiment data.

[0424] Data processing / calculation: data normalization, missing value imputation, outlier removal.

[0425] Output: Clean data after preprocessing.

[0426] Specific action: Tanaka's data is normalized and incomplete data is complemented.

[0427] The server makes predictions using an AI model: The preprocessed data is fed into a generative AI model to predict stress levels and fatigue levels.

[0428] Input: Preprocessed business data and sentiment data.

[0429] Data processing / calculation: Perform predictive analysis using generative AI models.

[0430] Output: Predicted stress and fatigue levels for each employee.

[0431] Specific action: Tanaka's data is input into the AI ​​model, and it is predicted that he is in a state of high stress.

[0432] Step 3:

[0433] Notifications and refresh suggestions

[0434] The server determines the timing of breaks: Based on the analysis results, the server determines the optimal timing of breaks for each employee.

[0435] Inputs: Forecast results and current business progress.

[0436] Data processing / calculation: Calculate the optimal break timing.

[0437] Output: Break timing decision result.

[0438] Specific action: Tanaka decides the timing, "I will take a 15-minute break in 10 minutes."

[0439] The server suggests a refreshing method: It refers to the hobby and preference database and suggests an appropriate refreshing method.

[0440] Input: Hobby and preference information for each employee and analysis results.

[0441] Data processing / calculation: Determine the appropriate refresh method.

[0442] Output: The proposed refresh method.

[0443] Specific action: A method for "drinking coffee" is suggested to Tanaka.

[0444] The server sends a notification: The determined break timing and refreshment method are sent to the terminal. The terminal displays the notification to the employee.

[0445] Input: Break timing and refreshment method decisions.

[0446] Data processing / calculation: generating and sending notification messages.

[0447] Output: Notifications that appear on the device.

[0448] Specific action: A message will appear on Tanaka's PC saying, "Take a 15-minute break in 10 minutes and have a cup of coffee."

[0449] Step 4:

[0450] Refresh execution and data collection

[0451] User acknowledges notification: Employee acknowledges notification and selects an option such as "Got it" or "Later."

[0452] Input: Notification message.

[0453] Data processing / calculation: Record the user's selection results.

[0454] Output: The user's selection.

[0455] Specific action: Tanaka selects "Got it."

[0456] Server will set snooze: If the user selects "Later", this sets the snooze timing. The server will snooze after a certain time.

[0457] Input: User's choice ("Later").

[0458] Data processing / calculation: Calculation and setting of re-notification timing.

[0459] Output: Snooze settings.

[0460] Specific operation: If Tanaka selects "Later", the server will schedule a reminder for 30 minutes.

[0461] User executes the refresh method: The user executes the proposed refresh method.

[0462] Input: Refresh suggestion for notification message.

[0463] Data processing / calculation: Record of refresh methods performed.

[0464] Output: State data after refresh.

[0465] Specific action: Tanaka drinks coffee.

[0466] The terminal collects performance data: The terminal collects performance data after the refresh and sends it to the server.

[0467] Input: User operation data and physiological data.

[0468] Data processing / calculation: recording and transmission of performance data.

[0469] Output: Performance data after refresh.

[0470] Specific action: After Tanaka's break, his input speed improves and the data is sent from his device to the server.

[0471] Step 5:

[0472] Feedback and model improvement

[0473] The server evaluates the effect: The effect of the refreshment method is evaluated and the degree of improvement in stress and fatigue is analyzed.

[0474] Input: Performance and emotion data after refresh.

[0475] Data processing / calculation: Calculation and evaluation of the degree of improvement.

[0476] Output: Evaluation results.

[0477] Specific action: Tanaka's change in stress level is evaluated.

[0478] The server retrains the AI ​​model: The AI ​​model is retrained based on the evaluation results to improve the accuracy of the next analysis.

[0479] Input: Assessment results and stored historical data.

[0480] Data processing / computation: Retraining AI models and updating algorithms.

[0481] Output: The updated AI model.

[0482] Specific action: The AI ​​model is updated based on new data.

[0483] The server stores the data: The evaluation results and analysis data are stored in a database for future analysis.

[0484] Input: Evaluation results and analysis data.

[0485] Data processing / calculation: Saving to database.

[0486] Output: Saved evaluation results and analysis data.

[0487] Specific actions: Updated evaluation results and analysis data are added to the database.

[0488] (Application example 2)

[0489] 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."

[0490] In today's work environment, properly collecting and analyzing employee work data and emotional data to suggest effective break timing and refreshment methods is important for reducing employee stress and improving productivity. Similarly, for robots operating in factories, collecting operational and sensor data and suggesting appropriate maintenance timing and methods is essential for maximizing robot operating efficiency and minimizing downtime. While systems exist to solve these issues individually, integrated systems are still lacking. Therefore, there is a need for the development of an integrated system that suggests optimal break and maintenance methods for both employees and robots.

[0491] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting work data from in-house devices, means for analyzing the collected work data and emotion data, means for determining optimal break timing and refreshment methods for employees based on the analysis results, means for notifying employees of the determined break timing and refreshment methods, means for collecting refreshment methods performed by employees and their results, means for updating the database based on the collected results and using them for the next analysis, means for collecting robot operation data and sensor data, means for analyzing the collected robot data and detecting signs of stress and fatigue, and means for proposing appropriate maintenance timing and methods based on the detected results. This makes it possible to reduce employee stress, improve productivity, maximize robot operating efficiency, and minimize downtime.

[0492] "Internal devices" are devices such as PCs, smartphones, and wearable devices that are used to collect employee business data and emotional data.

[0493] "Work data" refers to data including an employee's working hours, work content, work location, and stress level.

[0494] "Emotion data" is data that indicates emotional states such as anger, sadness, joy, and fear, obtained through voice analysis, facial recognition, and text analysis.

[0495] "Robot operation data" refers to data such as motor temperature, operating time, and error rate collected from sensors on robots operating in factories, etc.

[0496] "Sensor data" refers to real-time data collected by sensors mounted on the robot.

[0497] An "AI model" is an artificial intelligence algorithm used to analyze business data, emotional data, and robot operation data to predict stress and fatigue and suggest breaks and maintenance.

[0498] "Break timing" refers to the time that employees should take a break for work efficiency and health management.

[0499] "Refreshment methods" refer to actions and activities that employees should take to smoothly continue their work, such as drinking coffee or stretching.

[0500] "Maintenance timing" refers to the time when maintenance necessary to maintain the operating efficiency of a robot should be performed.

[0501] "Maintenance method" refers to the specific measures and actions that should be taken to keep the robot functioning optimally, such as oiling and replacing parts.

[0502] This invention includes a system that collects employee work data and emotional data, analyzes it in real time, and proposes optimal break timing and refreshment methods, and a system that analyzes the operation data of factory robots and proposes appropriate maintenance timing and methods. Specific embodiments of this invention are described below.

[0503] System configuration

[0504] This system consists of a means of collecting employee data, a server that analyzes the collected data, a means of issuing instructions based on the analysis results, and a means of collecting the results of refreshes and maintenance based on the instructions. Each means is explained in detail below.

[0505] Data collection from internal devices

[0506] The server connects to in-house devices such as PCs, smartphones, and wearable devices to collect employee work data. Data collected from these devices includes work hours, work content, work location, and stress level. Emotional data, such as anger, sadness, joy, and fear, is also collected in real time using voice analysis, facial recognition, and text analysis.

[0507] Data analysis

[0508] The server performs preprocessing on the collected work data and emotion data, such as normalizing it, filling in missing values, and removing outliers, and then inputs the data into an AI model to predict each employee's stress level and fatigue level, which can then individually suggest optimal break times and ways to refresh.

[0509] For factory robots, the server collects real-time data from each sensor and uses AI models to analyze signs of stress and fatigue based on data such as motor temperature, operating time, and error rate.

[0510] Notifications and refresh / maintenance suggestions

[0511] Based on the analysis results, the server notifies employees of the optimal timing for breaks and how to refresh themselves. For example, it may notify employees that they should take a 15-minute break in 10 minutes, along with a suggestion to refresh themselves with a cup of coffee. For factory robots, it suggests the appropriate timing and method for maintenance. For example, it may suggest that robots be stopped to cool down or that they need oil.

[0512] Collection of results and feedback

[0513] The refresh methods performed by employees and the results of factory robot maintenance are collected again and sent to the server, which updates the database based on these results and retrains the AI ​​model to improve the accuracy of the next analysis.

[0514] Specific examples

[0515] A concrete example is the program Tanaka. The device records Tanaka's slowing keystroke speed and rising heart rate, and judges her to be "highly stressed" based on her voice and facial expression. The server analyzes this data, recognizes her as being in a "high stress state," and notifies her to "take a 15-minute break in 10 minutes," and suggests that she "have a cup of coffee and refresh herself." Tanaka follows the suggestion and takes a break, and the device again collects performance data after the break and sends it to the server. The server then updates the AI ​​model based on this data.

[0516] For example, in the case of a factory robot, if the motor temperature rises too high while the robot is operating, a sensor will report this abnormality to the server, which will then send a message to the robot to stop it to cool down, and a maintenance technician will inject oil or take other cooling measures.

[0517] Prompt Sentence Examples

[0518] "Create an AI model that predicts the stress level of a robot and suggests appropriate maintenance."

[0519] "Write a Python program to analyze stress and fatigue levels using data from factory robot sensors."

[0520] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0521] Step 1: Data collection

[0522] The server collects employee work data and emotional data from the devices. Specifically, it obtains data such as work hours, work content, work location, stress level, heart rate, and electrodermal activity from employees' PCs, smartphones, and wearable devices. It also performs voice analysis, facial recognition, and text analysis to collect emotional data such as anger, sadness, joy, and fear in real time. The input is sensor data from each device, and the output is the collected work data and emotional data.

[0523] Step 2: Preprocessing

[0524] The server performs preprocessing on the collected data, such as normalizing, filling in missing values, and removing outliers. Specifically, it detects and removes outliers and fills in missing data. The input is the collected raw data, and the output is the preprocessed data.

[0525] Step 3: Data analysis

[0526] The server inputs the preprocessed data into an AI model to predict the stress level and fatigue level for each employee. Specifically, the data is input into the AI ​​model, and the predicted stress level and fatigue level are output. The input is the preprocessed data, and the output is the predicted stress level and fatigue level.

[0527] Step 4: Notification based on analysis results

[0528] Based on the analysis results, the server determines the optimal timing for breaks and ways to refresh, and notifies employees via their devices. Specifically, it sends notifications such as "Take a 15-minute break in 10 minutes" or "Refresh yourself with a cup of coffee." The input is the predicted stress level and fatigue level, and the output is notifications on break timing and ways to refresh.

[0529] Step 5: Employee execution and collection of results

[0530] The employee performs the refresh method, and the terminal re-collects the subsequent performance data. Specifically, it again records the recovery of keystroke speed, the reduction of error messages, etc. The input is the employee's performance data after the method is performed, and the output is the collected result data.

[0531] Step 6: Feedback and model improvement

[0532] The server evaluates the effectiveness of the refreshment method based on the collected result data and retrains the AI ​​model. Specifically, it analyzes the degree of improvement in stress and fatigue and updates the algorithm. The input is the result data, and the output is an improved AI model.

[0533] Step 7: Robot Data Collection

[0534] The server collects operational and sensor data from the robots in the factory, specifically motor temperature, operating time, error rate, etc. The input is the robot sensor data, and the output is the collected robot data.

[0535] Step 8: Robot data analysis

[0536] The server inputs the collected robot data into an AI model to predict signs of stress and fatigue. Specifically, it analyzes stress levels and fatigue levels based on the data. The input is the collected robot data, and the output is the predicted stress levels and fatigue levels.

[0537] Step 9: Maintenance Recommendations

[0538] The server then suggests appropriate maintenance timing and methods based on the analysis results. Specifically, it makes suggestions such as "Stop the vehicle to cool it down" or "Oil injection is required." The input is the predicted stress level and fatigue level, and the output is a notification of the maintenance method.

[0539] Step 10: Collect maintenance results

[0540] The server collects the results of maintenance performed by the maintenance technician and uses them for future analysis. Specifically, it records the robot performance data after maintenance. The input is the performance data after maintenance, and the output is the collected result data.

[0541] 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.

[0542] 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.

[0543] 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.

[0544] [Second embodiment]

[0545] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0546] 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.

[0547] 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).

[0548] 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.

[0549] 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.

[0550] 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).

[0551] 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.

[0552] 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.

[0553] 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.

[0554] 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.

[0555] 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.

[0556] 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."

[0557] This invention relates to a system that uses AI to analyze employee work data and proposes appropriate break timing and refreshment methods. Specific embodiments of this system are described below.

[0558] System configuration

[0559] The system consists of a terminal that collects employee work data, a server that performs analysis, and a terminal that sends notifications and re-collects data.

[0560] 1. Data Collection

[0561] The server connects to each employee's device (PC, smartphone, wearable device) and collects work data, including work hours, work content, work location, stress level, etc.

[0562] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[0563] Users (employees) input information such as self-evaluations in the form of a questionnaire at the start and end of work, and periodically (for example, every 1-2 hours).

[0564] 2. Data Analysis

[0565] The server receives the collected data and preprocesses it (e.g., normalizes and treats outliers).

[0566] The server inputs the preprocessed data into an AI model to predict stress and fatigue levels, thereby calculating the degree of stress and fatigue for each employee.

[0567] 3. Notifications and refresh suggestions

[0568] Based on the analysis results, the server determines the optimal break timing and method of refreshment for each employee.

[0569] The device notifies employees of break times based on the information received from the server, for example by displaying a message saying, "You will take a 15-minute break in 10 minutes."

[0570] The user receives the notification and can click on an option such as "OK" or "Later" as needed. If the notification is ignored, the server sets the timing for re-notification and prompts the user to take a break again after a certain period of time.

[0571] 4. Refresh execution and data collection

[0572] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[0573] The device collects performance data after a refresh, such as improved keystroke speed and fewer error messages.

[0574] The terminal transmits the execution data to the server.

[0575] 5. Feedback and model improvement

[0576] The server evaluates how effective the refreshment methods were in reducing employees' fatigue and stress.

[0577] The server readjusts the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next proposal.

[0578] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[0579] Specific examples

[0580] Example: Programmer Tanaka

[0581] The device recorded Tanaka's slowing keystroke speed and an increase in error messages, as well as his heart rate and self-assessment data, which revealed that he was experiencing a decline in concentration.

[0582] The server analyzed Tanaka's data and determined that his stress level was high.

[0583] The server notifies Tanaka, "In 10 minutes, you will take a 15-minute break and have a cup of coffee to refresh yourself."

[0584] The user (Mr. Tanaka) checks the notification and takes a break as suggested.

[0585] The device collects Tanaka's performance data again after the break and sends it to the server.

[0586] The server uses this data to update the AI ​​model and prepare for the next more precise analysis.

[0587] In this way, the system collects employees' work data, analyzes it using AI, and notifies them of the best times to take breaks and how to refresh themselves, thereby reducing employee stress and optimizing their performance.

[0588] The processing flow will be explained below.

[0589] The specific processing steps of the program and their operations will be explained below.

[0590] Step 1:

[0591] The device collects and records employees' work data in real time, such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[0592] Step 2:

[0593] The terminal displays a periodic questionnaire to the user (employee) and asks them to enter their self-evaluation (for example, their work progress or current physical condition).

[0594] Step 3:

[0595] The server receives the business data sent from the terminal and stores it in a database.

[0596] Step 4:

[0597] The server checks the received data and performs preprocessing (e.g., data normalization, missing value imputation, and outlier removal).

[0598] Step 5:

[0599] The server then feeds the pre-processed data into an AI model that predicts each employee's stress and fatigue levels using predictive algorithms based on past data and similar patterns.

[0600] Step 6:

[0601] The server determines the optimal break timing for each employee based on the prediction results of the AI ​​model.

[0602] Step 7:

[0603] The server references a database of each employee's hobbies and preferences and determines appropriate ways to refresh themselves, such as drinking coffee or stretching to take a break from work.

[0604] Step 8:

[0605] The server transmits the determined break timing and refreshment method to the terminal.

[0606] Step 9:

[0607] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[0608] Step 10:

[0609] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[0610] Step 11:

[0611] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[0612] Step 12:

[0613] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[0614] Step 13:

[0615] The terminal will again collect user performance data (e.g., improved keystroke speed, reduced error messages) after the refresh.

[0616] Step 14:

[0617] The terminal transmits the refreshed data to the server.

[0618] Step 15:

[0619] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[0620] Step 16:

[0621] The server stores the evaluation results in a database and uses them to retrain the AI ​​model, improving the accuracy of the next analysis.

[0622] By performing the above processing steps, it is possible to analyze employees' work data and suggest appropriate break times and methods for refreshing, thereby reducing employee stress and optimizing work performance.

[0623] Example 1

[0624] 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."

[0625] In today's work environment, many employees accumulate fatigue and stress due to long working hours, which can result in reduced productivity and adverse effects on their mental and physical health. To address this issue, a system that can suggest effective ways to refresh at the appropriate time is needed. However, existing methods have the challenge of making it difficult to provide appropriate timing and methods of refreshment that suit each employee's individual situation.

[0626] 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.

[0627] In this invention, the server includes means for collecting employee work data, pre-processing means for normalizing the collected work data and processing outliers, means for having an AI model for analyzing the pre-processed work data, means for determining the optimal break timing and refreshment method for the employee based on the analysis results, means for notifying the employee of the determined break timing and refreshment method, means for collecting data on the refreshment method performed by the employee and subsequent performance data, and means for readjusting the AI ​​model based on the collected results and updating the algorithm to be useful for the next analysis. This allows for the provision of appropriate break timing and refreshment methods according to the situation of each employee, thereby reducing employee stress and improving productivity.

[0628] "Work data" is data generated when an employee performs their work, including work hours, work content, work location, stress level, keyboard input speed, mouse movements, heart rate, and electrodermal activity.

[0629] "Preprocessing means" refers to the means of normalizing collected business data, removing outliers, and preparing the data in a form that is easy to analyze.

[0630] The "AI model" is a model that uses machine learning algorithms to predict employees' stress levels and fatigue levels from preprocessed data.

[0631] The "analysis method" is a method of inputting preprocessed business data into an AI model to calculate employee stress and fatigue levels.

[0632] "Break timing" refers to employees taking breaks at optimal times during work, and indicates time to reduce stress and fatigue.

[0633] "Refreshment methods" are methods suggested to reduce employees' stress and fatigue, such as drinking coffee or stretching their eyes.

[0634] The "notification means" is a means for informing employees of the break timing and refreshment method determined based on the analysis results.

[0635] "Performance data" is data collected after an employee performs a refresher technique, including recovery of keystroke speed and reduction in error messages.

[0636] "Readjustment measures" are measures for readjusting the AI ​​model based on collected performance data to improve the accuracy of the next analysis results.

[0637] The "algorithm updating means" is a means of updating the analysis algorithm to the latest version using a retuned AI model.

[0638] This invention is a system that collects employee work data, analyzes it using AI, and suggests appropriate break times and ways to refresh. The system continuously monitors employees' work status, aiming to reduce stress and fatigue and improve productivity.

[0639] System configuration

[0640] The system consists of a terminal that collects employee work data, a server that performs analysis, and a terminal that sends notifications and re-collects data.

[0641] Hardware and software used

[0642] Device: An input device such as a PC, smartphone, or wearable device.

[0643] Server: A central computer with data processing and analysis capabilities.

[0644] Software: Data collection software, pre-processing software, AI models, notification software, database management software.

[0645] Data collection

[0646] The server connects to each employee's device and collects work data, including work hours, work content, work location, and stress level.

[0647] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[0648] Users (employees) enter self-evaluations in the form of a questionnaire at the start and end of work, and periodically.

[0649] Data analysis

[0650] The server preprocesses the collected data, normalizes the data, and processes outliers.

[0651] The server inputs the preprocessed data into an AI model to predict the user's stress level and fatigue level.

[0652] Based on the analysis results, the server quantifies and evaluates the user's level of stress and fatigue.

[0653] Notifications and refresh suggestions

[0654] Based on the analysis results, the server determines the optimal timing for breaks and the method of refreshing.

[0655] The device notifies the user of the break based on the information received from the server. For example, it displays a message saying, "Take a 15-minute break in 10 minutes."

[0656] The user can check the notification and choose options such as "OK" or "Later." If the notification is ignored, the server will notify the user again.

[0657] Refresh execution and data collection

[0658] The user performs the suggested refreshment method, such as drinking a cup of coffee or doing eye stretches.

[0659] The device will collect performance data after the refresh, recording things like improved keystroke speed and fewer error messages.

[0660] The terminal retransmits the collected data to the server.

[0661] Feedback and model improvement

[0662] The server evaluates how effective the refreshment method was in reducing the user's fatigue and stress.

[0663] The server readjusts the AI ​​model based on the evaluation results to improve the accuracy of the next analysis results.

[0664] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[0665] Specific examples

[0666] Example: Programmer employee A

[0667] The device records Employee A's slowing keystroke speed and an increase in error messages, and also collects data on his heart rate and self-evaluation questionnaires that indicate his "declining ability to concentrate."

[0668] The server analyzes employee A's data and determines that his stress level is high.

[0669] The server notifies employee A, "Take a 15-minute break in 10 minutes and refresh yourself with some coffee."

[0670] The user (Employee A) checks the notification and takes a break as suggested.

[0671] The terminal collects the performance data of employee A again after the break and transmits it to the server.

[0672] The server updates the AI ​​model based on this data, preparing for the next more precise analysis.

[0673] Prompt Sentence Examples

[0674] Example: Describe a system that analyzes the stress level of programmer employee A based on his work data and suggests optimal break times and ways to refresh himself.

[0675] This system can reduce employee stress and optimize performance by informing them of the best time to take a break and how to refresh themselves based on their individual circumstances.

[0676] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0677] Step 1: Data collection

[0678] The user presses the start button on the terminal and fills out a self-evaluation questionnaire. The entered data is then saved directly on the terminal.

[0679] The device collects user input and real-time task data (such as keyboard typing speed, mouse movements, heart rate, and electrodermal activity), which is temporarily stored on the device.

[0680] The server receives work data periodically sent from the terminal, including work hours, work content, work location, stress level, etc.

[0681] (Input) User surveys, keyboard input speed, mouse movement, heart rate, etc.

[0682] (Output) The collected business data is sent to the server.

[0683] Specifically, when Tanaka starts work, he answers a questionnaire displayed on the device asking, "How are you feeling today?" and the device records his answers.

[0684] Step 2: Data Preprocessing

[0685] The server normalizes the business data received from the device and corrects or removes any abnormal values. For example, if the heart rate is extremely high, it determines that this is noise and corrects the value.

[0686] The server prepares the pre-processed data to be fed into the AI ​​model.

[0687] (Input) Collected business data.

[0688] (Output) Preprocessed data that has been normalized and outlier treated.

[0689] Specifically, the server analyzes Tanaka's heart rate data, and if an unnatural increase is detected, it corrects it to the correct value.

[0690] Step 3: Data analysis

[0691] The server then inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level, for example, assessing stress levels based on a decrease in keyboard typing speed or an increase in heart rate.

[0692] Based on the analysis results, the server quantifies and evaluates the user's level of stress and fatigue.

[0693] (Input) Preprocessed data.

[0694] (Output) Analysis results showing stress levels and fatigue levels.

[0695] Specifically, the server detects a decrease in Tanaka's keyboard input speed and evaluates this as a stress index.

[0696] Step 4: Notifications and refresh suggestions

[0697] The server then determines the optimal timing for rest and how to refresh based on the analysis results. For example, if the heart rate is high, it may recommend five minutes of deep breathing.

[0698] The device receives a notification from the server and displays a message to the user such as, "The next break is in 10 minutes. Have a cup of coffee and refresh yourself."

[0699] The user can review the notification and enter options such as "OK" or "later."

[0700] (Input) Analysis results.

[0701] (Output) Notification of break times and ways to refresh.

[0702] Specifically, the server notifies Tanaka that "we recommend that you take a coffee break in 10 minutes," and Tanaka responds with "Got it."

[0703] Step 5: Refresh execution and data collection

[0704] The user performs the suggested refreshment method (e.g., drinking coffee, stretching the eyes, etc.).

[0705] The terminal collects performance data after the refresh (e.g., recovery of keystroke speed, reduction in error messages).

[0706] The terminal retransmits the collected data to the server.

[0707] (Input) Performance data before the refresh is executed.

[0708] (Output) Performance data after the refresh is performed.

[0709] Specifically, Tanaka drinks coffee to refresh himself, and the data on his subsequent recovery of keyboard typing speed is collected on the device and sent to the server.

[0710] Step 6: Feedback and model improvement

[0711] The server evaluates how effective the refreshment methods were in reducing employees' fatigue and stress.

[0712] The server readjusts the AI ​​model based on the evaluation results to improve the accuracy of the next analysis results.

[0713] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[0714] (Input) Performance data after the refresh is executed.

[0715] (Output) Retuned AI models and updated algorithms.

[0716] Specifically, the server evaluates Tanaka's refreshed data and updates the model, improving the accuracy of suggestions from the next time onwards.

[0717] (Application example 1)

[0718] 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."

[0719] Conventional systems for suggesting breaks and refreshment methods determine the optimal timing by collecting and analyzing employees' work data, but the types of data available are limited, meaning that the suggestions cannot fully reflect the detailed stress and fatigue levels of individual employees.In addition, the limited number of devices available for notification makes it difficult to respond flexibly to the work environment.

[0720] 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.

[0721] In this invention, the server includes means for collecting employee work data, means for analyzing the collected work data, means for determining optimal break timing and refreshment methods for employees based on the analysis results, means for notifying employees of the determined break timing and refreshment methods, means for collecting the refreshment methods performed by employees and their results, means for updating the database based on the collected results and using them for the next analysis, means for collecting eye movements, heart rate, and body movements as work data in real time, and means for notifying workers of break timing and refreshment methods via their smart devices. This allows the system to reflect detailed stress and fatigue levels of employees in real time, making it possible to propose more accurate break timing and refreshment methods.

[0722] "Employee work data" refers to information generated by employees during work, such as time, content, location, heart rate, eye movements, body movements, and stress levels.

[0723] "Means for collection" refers to devices and systems for collecting employee work data in real time, including wearable devices such as smart glasses.

[0724] The "means of analysis" is a system that preprocesses collected business data and uses an AI model to analyze stress levels and fatigue levels.

[0725] The "means for determining" is a system that selects the optimal break timing and refreshment method for employees based on the analysis results.

[0726] The "notification means" refers to a device or application that notifies employees in real time of the determined break timing and refreshment method. Specifically, this is done via a smart device.

[0727] The "means of updating the database based on collected results" refers to a system that records the refresh methods used by employees and the results of those methods, updates the database based on those results, and uses them for the next analysis.

[0728] "Eye movement" is data on employees' eye movements and is an indicator used to evaluate their concentration level and stress level at work.

[0729] "Heart rate" is the number of times an employee's heart beats within a certain period of time, and is an indicator primarily used to assess stress and fatigue levels.

[0730] "Physical movement" refers to information about an employee's physical movement and is an indicator for evaluating work efficiency and fatigue level.

[0731] A "smart device" is a device that has Internet connectivity and is capable of data communication and application execution, and specifically includes smart glasses and smartphones.

[0732] The "AI model" is a model that uses machine learning algorithms to analyze business data and predict employees' stress levels and fatigue levels.

[0733] This invention is a system that uses AI to analyze employees' work data and propose optimal break timing and refreshment methods. The system consists of a means for collecting employees' work data, a means for analyzing the data, a means for determining and notifying break timing and refreshment methods, and a means for collecting and evaluating the refreshment results.

[0734] Hardware Configuration

[0735] The system consists of smart glasses for collecting data, wearable devices, and terminals such as PCs and smartphones. The data is collected on a server and analyzed. Specific devices used include Google Glass and Microsoft HoloLens.

[0736] Software Configuration

[0737] AI models are used for data analysis, and machine learning libraries such as TensorFlow are introduced, while data analysis libraries such as SciPy and NumPy are used for data preprocessing.

[0738] Process Overview

[0739] 1. Data Collection

[0740] The server collects real-time data on employees' eye movements, heart rate, and body movements via smart glasses or wearable devices.

[0741] The data is transmitted from the terminal to a server and recorded in real time.

[0742] 2. Data Analysis

[0743] The server preprocesses the collected data, eliminating outliers and normalizing it.

[0744] The preprocessed data is then fed into an AI model to analyze stress levels and fatigue levels.

[0745] 3. Notifications and refresh suggestions

[0746] Based on the analysis results, the server determines the optimal timing for breaks and the method of refreshing.

[0747] The decision will be communicated to employees via smart glasses or smartphones.

[0748] For example, a notification will appear saying, "Take a 15-minute break in 10 minutes and take some deep breaths."

[0749] 4. Collecting and Evaluating Refresh Results

[0750] After the employee executes the refresh method, the terminal recollects the resulting data.

[0751] The collected data is sent to a server and used by an AI model to evaluate its effectiveness.

[0752] 5. Feedback and model improvement

[0753] The effect of the refresh is evaluated and the AI ​​model is readjusted based on the results.

[0754] The database will be updated with new evaluation results and used for the next analysis.

[0755] Specific examples

[0756] For example, if a factory worker wears smart glasses and their eye movements deviate from a certain pattern, it will be judged as a "high stress level." The same applies if their heart rate rises sharply. In this case, the system will notify the worker to "take a break and take a short walk or stretch." After the worker takes a break as notified, the resulting data will be collected again and fed back into the AI ​​model.

[0757] Prompt Sentence Examples

[0758] Model inputs

[0759] Worker heart rate data:

[0760] [68, 70, 72, 75, 77, 80, 85, 90, 95, 100]

[0761] Worker gaze movement data (eye movement):

[0762] [0.5, 0.6, 1.2, 1.1, 0.8, 0.7, 1.3, 1.4, 1.5, 0.9]

[0763] Expected output

[0764] Your stress levels are high, so I suggest you take a 10-minute break and stretch.

[0765] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0766] Step 1:

[0767] The server collects data on employees' eye movements, heart rates, and body movements in real time via smart glasses or wearable devices. The input is raw data sent from each device, and the output is a business dataset that compiles this data. Specifically, all eye movement information obtained by eye tracking sensors and heart rate information obtained by heart rate sensors are aggregated on the server.

[0768] Step 2:

[0769] The terminal preprocesses the collected business data. The input is the raw data collected in step 1. Specific data processing includes removing outliers, normalizing the data, and removing noise. The output is a normalized dataset that has undergone these preprocessing steps. The server uses SciPy and NumPy to clean and format the data.

[0770] Step 3:

[0771] The server inputs the preprocessed data into an AI model to analyze stress levels and fatigue levels. The input is the normalized dataset generated in step 2. Specific data calculations include predicting stress levels using machine learning algorithms. The output is a numerical value indicating each employee's stress level and fatigue level. The server runs the AI ​​model using TensorFlow.

[0772] Step 4:

[0773] The server determines the optimal break timing and refreshment method based on the analysis results. The input is the analysis result from step 3. In concrete terms, the server determines the break timing and selects an appropriate refreshment method (e.g., stretching or deep breathing). The output is the determined break timing and refreshment method.

[0774] Step 5:

[0775] The terminal notifies the employee of the determined break timing and refreshment method. The input is the result of the decision made in step 4. The specific operation is to display a notification message on the smart glasses or smartphone. The output is a confirmation message of the notification. For example, the message "Take a 15-minute break in 10 minutes and take deep breaths" is displayed.

[0776] Step 6:

[0777] The user (employee) performs the suggested refreshment method based on the notification. The input is the notification message from step 5. Specifically, the employee performs a refreshment method such as stretching or deep breathing. The output is the result data of the refreshment method that was performed.

[0778] Step 7:

[0779] The device again collects data after the refresh method is executed. The input is the result data of the refresh method executed by the user. Specifically, heart rate, eye movement, etc. are collected again. The output is a performance dataset after the refresh.

[0780] Step 8:

[0781] The server evaluates the effectiveness of the refresh method based on the collected post-refresh data. The input is the post-refresh data from step 7. The specific data calculation involves comparing the data before and after the refresh and evaluating areas for improvement. The output is numerical data of the evaluation results.

[0782] Step 9:

[0783] The server readjusts the AI ​​model based on the evaluation results and updates the database. The input is the evaluation result from step 8. Specifically, it adjusts the parameters of the AI ​​model and adds new data points to the database. The output is an improved AI model and an updated database.

[0784] 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.

[0785] This invention relates to a system that uses AI to analyze employee work data and emotional data and suggests appropriate break timing and methods for refreshing. Specific embodiments of this system are described below.

[0786] System configuration

[0787] The system consists of a terminal that collects employees' work data and emotional data, a server that performs the analysis, and a terminal that sends notifications and re-collects the data.

[0788] 1. Data Collection

[0789] The server connects to each employee's device (PC, smartphone, wearable device) and collects work data and emotional data, including work hours, work content, work location, and stress level.

[0790] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[0791] The device uses an emotion engine to perform voice analysis, facial recognition, and text analysis to collect user (employee) emotional data in real time, including emotional states such as anger, sadness, joy, and fear.

[0792] Users (employees) input their emotional state by self-evaluating in the form of a questionnaire at the start and end of work, and periodically (for example, every 1-2 hours).

[0793] 2. Data Analysis

[0794] The server checks the received business data and emotion data and stores them in a database.

[0795] The server performs preprocessing (e.g., data normalization, missing value imputation, and outlier removal).

[0796] The server then inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level. By incorporating emotional data into the analysis results, more accurate predictions can be made.

[0797] 3. Notifications and refresh suggestions

[0798] Based on the analysis results, the server determines the optimal break timing for each employee.

[0799] The server refers to the employee's hobby and preference database and determines the appropriate way to refresh themselves, suggesting, for example, drinking coffee or stretching to take a break from work.

[0800] The server transmits the determined break timing and refreshment method to the terminal.

[0801] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[0802] 4. Refresh execution and data collection

[0803] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[0804] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[0805] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[0806] The terminal collects performance data after the refresh (e.g., recovery of key input speed, reduction in error messages) and sends it to the server.

[0807] 5. Feedback and model improvement

[0808] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[0809] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis.

[0810] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[0811] Specific examples

[0812] Example: Programmer Tanaka

[0813] The device records Tanaka's slowing keystroke speed and rising heart rate, and the emotion engine determines from Tanaka's voice and face that he is "highly stressed" or "feeling angry."

[0814] The server analyzed Tanaka's data and determined that he was in a "high stress state."

[0815] The server notifies Tanaka that he should take a 15-minute break in 10 minutes and also suggests that he should have a cup of coffee to refresh himself.

[0816] The user (Mr. Tanaka) checks the notification and takes a break as suggested.

[0817] The device recollects Tanaka's performance and emotional data after the break and sends it to the server.

[0818] The server updates the AI ​​model based on this data, preparing for the next more precise analysis.

[0819] In this way, the system collects and analyzes employees' work and emotional data, and notifies them of appropriate break times and ways to refresh themselves, thereby reducing employee stress and optimizing their work performance.

[0820] The processing flow will be explained below.

[0821] The specific processing steps of the system are described below.

[0822] Step 1:

[0823] The device collects and records employees' work data (keystroke speed, mouse movements, application usage history, heart rate, electrodermal activity, etc.) in real time.

[0824] Step 2:

[0825] The device uses an emotion engine to perform voice analysis, facial recognition, and text analysis to collect and record the user's (employee's) emotional data in real time. Emotional data includes emotional states such as anger, sadness, joy, and fear.

[0826] Step 3:

[0827] The device displays regular self-evaluations in the form of a questionnaire to the user (employee), prompting them to enter information about their work progress, current physical condition, and emotional state.

[0828] Step 4:

[0829] The terminal transmits the collected business data and emotional data to the server at regular intervals.

[0830] Step 5:

[0831] The server stores the received data in a database and performs preprocessing, which includes data normalization, missing value imputation, and outlier removal.

[0832] Step 6:

[0833] The server inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level. Emotional data is also incorporated into the analysis results to make more accurate predictions.

[0834] Step 7:

[0835] Based on the analysis results, the server determines the optimal break timing for each employee.

[0836] Step 8:

[0837] The server refers to the employee's hobby and preference database and determines the appropriate way to refresh themselves, suggesting, for example, drinking coffee or stretching to take a break from work.

[0838] Step 9:

[0839] The server transmits the determined break timing and refreshment method to the terminal.

[0840] Step 10:

[0841] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[0842] Step 11:

[0843] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[0844] Step 12:

[0845] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[0846] Step 13:

[0847] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[0848] Step 14:

[0849] After the device is refreshed, the user's performance data (e.g., recovery of keystroke speed, reduction in error messages) and emotional data are collected again.

[0850] Step 15:

[0851] The terminal transmits the refreshed data to the server.

[0852] Step 16:

[0853] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[0854] Step 17:

[0855] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis.

[0856] Step 18:

[0857] The server stores the evaluation results and analysis data in a database for subsequent analysis.

[0858] By executing the above processing steps, the system can analyze employees' work data and emotional data and suggest appropriate break times and ways to refresh themselves. This reduces employee stress and optimizes work performance. As a concrete example, the system makes real-time suggestions to programmer Tanaka about when to take a break and how to refresh himself. Even when Tanaka is busy with a project, he can refresh himself at the appropriate time and maintain work efficiency.

[0859] Example 2

[0860] 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."

[0861] In today's business environment, properly managing employee stress and fatigue and improving work efficiency are important issues. However, traditional methods have made it difficult to grasp the state of each employee in real time and suggest appropriate break times and ways to refresh. As a result, many companies are not managing their employees' stress levels properly, resulting in decreased work performance and increased health risks.

[0862] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting work data and emotion data, means for analyzing the collected work data and emotion data, and means for determining optimal break timing and refreshment methods based on the analysis results. This makes it possible to grasp employee stress and fatigue in real time and propose optimal break timing and refreshment methods to each employee.

[0863] "Business data" refers to information generated when employees perform their work, including work hours, work content, work location, keystroke speed, mouse movements, application program usage history, heart rate, etc.

[0864] "Emotional data" refers to information that indicates an employee's emotional state, including emotional states such as anger, sadness, joy, and fear obtained through voice analysis, facial recognition, and text analysis.

[0865] "Analysis results" refers to information obtained after analyzing collected data using AI models, etc., and predicts the stress levels and fatigue levels of each employee.

[0866] "Break timing" refers to the time of break recommended for employees based on the analysis results.

[0867] "Refreshment methods" refer to actions or methods that employees are encouraged to take during their breaks, such as drinking coffee or stretching.

[0868] "Notification means" refers to the means of informing employees of the determined break timing and method of refreshment, and notifications are given via PCs, smartphones, wearable devices, etc.

[0869] A "database" is a repository of information that stores collected data and analysis results and allows reuse as needed.

[0870] A "generative AI model" is a model that includes algorithms for analyzing data and making predictions using AI.

[0871] MODE FOR CARRYING OUT THE INVENTION

[0872] This invention is a system that uses AI to analyze employees' work data and emotional data and suggests appropriate break times and ways to refresh. The system consists of the following main components:

[0873] Data collection

[0874] The device connects

[0875] The server establishes a secure connection with each employee's device (PC, smartphone, wearable device, etc.), enabling real-time data collection. For example, this includes the PCs and smartwatches employees use to do their work.

[0876] Business data collection

[0877] The devices record real-time work data, such as keystroke speed, mouse movements, and application usage history, which is used to gain a detailed understanding of how employees are performing their work. For example, keystroke speed during programming tasks and the type of software they are using are recorded.

[0878] Collecting Emotional Data

[0879] The device uses an emotion engine that performs voice analysis, facial recognition, and text analysis to analyze the user's emotional state in real time and digitize it. For example, the device classifies the user's emotional state into categories such as "stress" or "joy" based on the voice during a call or facial expressions captured via a webcam.

[0880] Enter your self-assessment

[0881] Users enter their emotional state in the form of a self-assessment questionnaire at the start and end of work and at regular intervals. For example, every 1-2 hours, they answer a questionnaire in which they rate their "current level of fatigue" on a scale of 1-5.

[0882] Data analysis

[0883] Receiving and storing data

[0884] The server receives the collected work data and emotion data and stores it in a database. This allows for centralized management, including past data. For example, fluctuations in an employee's stress level over the past month can also be recorded.

[0885] Data Preprocessing

[0886] The server performs preprocessing such as data normalization, missing value imputation, and outlier removal. This process improves the quality of the data to obtain accurate analysis results. For example, incomplete data sets and extreme values ​​are removed.

[0887] Prediction by AI model

[0888] The server inputs the preprocessed data into a generative AI model to predict each employee's stress level and fatigue level. The generative AI model is an algorithm that learns from past data and predicts future conditions. For example, it predicts that "Programmer Tanaka" is currently in a high stress state.

[0889] Notifications and refresh suggestions

[0890] Deciding when to take a break

[0891] Based on the analysis results, the server determines the optimal break timing for each employee. For example, it may decide that "Tanaka-san" should take a 15-minute break in 10 minutes.

[0892] Suggestions for refreshing

[0893] The server refers to a database of employees' hobbies and preferences to determine appropriate ways to refresh themselves, such as "drinking coffee" or "doing eye stretching exercises."

[0894] Sending notifications

[0895] The server then sends the determined break timing and refreshment method to the terminal, which then displays a notification to the employee. For example, a notification such as "Let's have a cup of coffee and refresh yourself" will appear on Tanaka's PC.

[0896] Refresh execution and data collection

[0897] User Verification

[0898] The user checks the notification and selects an option such as "OK" or "later." For example, Tanaka selects "OK."

[0899] Snooze settings

[0900] If "Later" is selected, the server sets the timing of the re-notification and sends the break notification to the terminal again after a certain time. For example, the setting "Re-notify after 30 minutes" is set.

[0901] Execute the refresh method

[0902] The user performs the suggested refreshment method. For example, Tanaka drinks coffee.

[0903] Performance Data Collection

[0904] The device collects performance data after the refresh and sends it to the server. For example, Tanaka's keystroke speed after the refresh is recorded.

[0905] Feedback and model improvement

[0906] Evaluation of effectiveness

[0907] The server evaluates the effectiveness of the refreshment method and analyzes the degree of improvement in stress and fatigue. For example, it evaluates changes in Tanaka's stress level.

[0908] Retraining the AI ​​model

[0909] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis, ensuring that analysis is always based on the latest data, resulting in more accurate predictions.

[0910] Data storage

[0911] The server stores the evaluation results and analysis data in a database for subsequent analysis, which can be used for future data analysis and model improvement.

[0912] Prompt Sentence Examples

[0913] "Design a system that analyzes employees' work data and emotional data to suggest appropriate break times and ways to refresh. The following input data is provided: work hours, work content, work location, keystroke speed, mouse movement, heart rate, voice analysis data, and facial recognition data. Use this data to predict employees' stress levels and fatigue levels."

[0914] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0915] Step 1:

[0916] Data collection

[0917] The server establishes a connection: The server establishes a secure connection with each employee's device (PC, smartphone, wearable device), allowing data to be sent and received in real time.

[0918] Input: Connection request from employee's device.

[0919] Data processing / calculation: Establish a secure connection and perform authentication.

[0920] Output: Establishment of a secure communication channel with the device.

[0921] Specific operation: Tanaka's smartwatch connects to the server and starts sending data.

[0922] Devices collect business data: Devices collect business data in real time, such as keystroke speed, mouse movements, and application usage history.

[0923] Input: Employee operation data (keystrokes, mouse movements, etc.).

[0924] Data processing / calculation: Recording of operational data and converting it into the required format.

[0925] Output: Formatted business data.

[0926] Specific operation: Tanaka's keystroke speed and mouse movements are recorded while he works in Excel.

[0927] Device collects emotional data: The device collects the user's emotional data using voice analysis, facial recognition, and text analysis.

[0928] Input: Audio data, facial image data, text data.

[0929] Data processing / calculation: Emotional states are analyzed using an emotion engine and converted into data.

[0930] Output: Emotion data (e.g. anger, joy, sadness, etc.).

[0931] Specific operation: While Tanaka is on the phone, the emotion of "anger" is determined and recorded as data.

[0932] Users enter self-evaluations: Users answer self-evaluation questionnaires at the start and end of work, and at regular intervals.

[0933] Input: User's self-assessment survey responses.

[0934] Data processing / calculation: Formatting the survey responses for storage in the database.

[0935] Output: Formatted self-assessment data.

[0936] Specific actions: Tanaka inputs his "current level of fatigue" every hour on a 5-point scale.

[0937] Step 2:

[0938] Data analysis

[0939] The server receives and stores the data: The server stores the received business data and emotion data in a database.

[0940] Input: Business and sentiment data collected in real time.

[0941] Data processing / calculation: Convert into a format that can be saved in a database.

[0942] Output: Business and sentiment data stored in a database.

[0943] Specific action: Tanaka's data is registered in the database.

[0944] The server performs data preprocessing: The server normalizes the data, imputes missing values, removes outliers, etc.

[0945] Input: Stored business and sentiment data.

[0946] Data processing / calculation: data normalization, missing value imputation, outlier removal.

[0947] Output: Clean data after preprocessing.

[0948] Specific action: Tanaka's data is normalized and incomplete data is complemented.

[0949] The server makes predictions using an AI model: The preprocessed data is fed into a generative AI model to predict stress levels and fatigue levels.

[0950] Input: Preprocessed business data and sentiment data.

[0951] Data processing / calculation: Perform predictive analysis using generative AI models.

[0952] Output: Predicted stress and fatigue levels for each employee.

[0953] Specific action: Tanaka's data is input into the AI ​​model, and it is predicted that he is in a state of high stress.

[0954] Step 3:

[0955] Notifications and refresh suggestions

[0956] The server determines the timing of breaks: Based on the analysis results, the server determines the optimal timing of breaks for each employee.

[0957] Inputs: Forecast results and current business progress.

[0958] Data processing / calculation: Calculate the optimal break timing.

[0959] Output: Break timing decision result.

[0960] Specific action: Tanaka decides the timing, "I will take a 15-minute break in 10 minutes."

[0961] The server suggests a refreshing method: It refers to the hobby and preference database and suggests an appropriate refreshing method.

[0962] Input: Hobby and preference information for each employee and analysis results.

[0963] Data processing / calculation: Determine the appropriate refresh method.

[0964] Output: The proposed refresh method.

[0965] Specific action: A method for "drinking coffee" is suggested to Tanaka.

[0966] The server sends a notification: The determined break timing and refreshment method are sent to the terminal. The terminal displays the notification to the employee.

[0967] Input: Break timing and refreshment method decisions.

[0968] Data processing / calculation: generating and sending notification messages.

[0969] Output: Notifications that appear on the device.

[0970] Specific action: A message will appear on Tanaka's PC saying, "Take a 15-minute break in 10 minutes and have a cup of coffee."

[0971] Step 4:

[0972] Refresh execution and data collection

[0973] User acknowledges notification: Employee acknowledges notification and selects an option such as "Got it" or "Later."

[0974] Input: Notification message.

[0975] Data processing / calculation: Record the user's selection results.

[0976] Output: The user's selection.

[0977] Specific action: Tanaka selects "Got it."

[0978] Server will set snooze: If the user selects "Later", this sets the snooze timing. The server will snooze after a certain time.

[0979] Input: User's choice ("Later").

[0980] Data processing / calculation: Calculation and setting of re-notification timing.

[0981] Output: Snooze settings.

[0982] Specific operation: If Tanaka selects "Later", the server will schedule a reminder for 30 minutes.

[0983] User executes the refresh method: The user executes the proposed refresh method.

[0984] Input: Refresh suggestion for notification message.

[0985] Data processing / calculation: Record of refresh methods performed.

[0986] Output: State data after refresh.

[0987] Specific action: Tanaka drinks coffee.

[0988] The terminal collects performance data: The terminal collects performance data after the refresh and sends it to the server.

[0989] Input: User operation data and physiological data.

[0990] Data processing / calculation: recording and transmission of performance data.

[0991] Output: Performance data after refresh.

[0992] Specific action: After Tanaka's break, his input speed improves and the data is sent from his device to the server.

[0993] Step 5:

[0994] Feedback and model improvement

[0995] The server evaluates the effect: The effect of the refreshment method is evaluated and the degree of improvement in stress and fatigue is analyzed.

[0996] Input: Performance and emotion data after refresh.

[0997] Data processing / calculation: Calculation and evaluation of the degree of improvement.

[0998] Output: Evaluation results.

[0999] Specific action: Tanaka's change in stress level is evaluated.

[1000] The server retrains the AI ​​model: The AI ​​model is retrained based on the evaluation results to improve the accuracy of the next analysis.

[1001] Input: Assessment results and stored historical data.

[1002] Data processing / computation: Retraining AI models and updating algorithms.

[1003] Output: The updated AI model.

[1004] Specific action: The AI ​​model is updated based on new data.

[1005] The server stores the data: The evaluation results and analysis data are stored in a database for future analysis.

[1006] Input: Evaluation results and analysis data.

[1007] Data processing / calculation: Saving to database.

[1008] Output: Saved evaluation results and analysis data.

[1009] Specific actions: Updated evaluation results and analysis data are added to the database.

[1010] (Application example 2)

[1011] 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."

[1012] In today's work environment, properly collecting and analyzing employee work data and emotional data to suggest effective break timing and refreshment methods is important for reducing employee stress and improving productivity. Similarly, for robots operating in factories, collecting operational and sensor data and suggesting appropriate maintenance timing and methods is essential for maximizing robot operating efficiency and minimizing downtime. While systems exist to solve these issues individually, integrated systems are still lacking. Therefore, there is a need for the development of an integrated system that suggests optimal break and maintenance methods for both employees and robots.

[1013] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting work data from in-house devices, means for analyzing the collected work data and emotion data, means for determining optimal break timing and refreshment methods for employees based on the analysis results, means for notifying employees of the determined break timing and refreshment methods, means for collecting refreshment methods performed by employees and their results, means for updating the database based on the collected results and using them for the next analysis, means for collecting robot operation data and sensor data, means for analyzing the collected robot data and detecting signs of stress and fatigue, and means for proposing appropriate maintenance timing and methods based on the detected results. This makes it possible to reduce employee stress, improve productivity, maximize robot operating efficiency, and minimize downtime.

[1014] "Internal devices" are devices such as PCs, smartphones, and wearable devices that are used to collect employee business data and emotional data.

[1015] "Work data" refers to data including an employee's working hours, work content, work location, and stress level.

[1016] "Emotion data" is data that indicates emotional states such as anger, sadness, joy, and fear, obtained through voice analysis, facial recognition, and text analysis.

[1017] "Robot operation data" refers to data such as motor temperature, operating time, and error rate collected from sensors on robots operating in factories, etc.

[1018] "Sensor data" refers to real-time data collected by sensors mounted on the robot.

[1019] An "AI model" is an artificial intelligence algorithm used to analyze business data, emotional data, and robot operation data to predict stress and fatigue and suggest breaks and maintenance.

[1020] "Break timing" refers to the time that employees should take a break for work efficiency and health management.

[1021] "Refreshment methods" refer to actions and activities that employees should take to smoothly continue their work, such as drinking coffee or stretching.

[1022] "Maintenance timing" refers to the time when maintenance necessary to maintain the operating efficiency of a robot should be performed.

[1023] "Maintenance method" refers to the specific measures and actions that should be taken to keep the robot functioning optimally, such as oiling and replacing parts.

[1024] This invention includes a system that collects employee work data and emotional data, analyzes it in real time, and proposes optimal break timing and refreshment methods, and a system that analyzes the operation data of factory robots and proposes appropriate maintenance timing and methods. Specific embodiments of this invention are described below.

[1025] System configuration

[1026] This system consists of a means of collecting employee data, a server that analyzes the collected data, a means of issuing instructions based on the analysis results, and a means of collecting the results of refreshes and maintenance based on the instructions. Each means is explained in detail below.

[1027] Data collection from internal devices

[1028] The server connects to in-house devices such as PCs, smartphones, and wearable devices to collect employee work data. Data collected from these devices includes work hours, work content, work location, and stress level. Emotional data, such as anger, sadness, joy, and fear, is also collected in real time using voice analysis, facial recognition, and text analysis.

[1029] Data analysis

[1030] The server performs preprocessing on the collected work data and emotion data, such as normalizing it, filling in missing values, and removing outliers, and then inputs the data into an AI model to predict each employee's stress level and fatigue level, which can then individually suggest optimal break times and ways to refresh.

[1031] For factory robots, the server collects real-time data from each sensor and uses AI models to analyze signs of stress and fatigue based on data such as motor temperature, operating time, and error rate.

[1032] Notifications and refresh / maintenance suggestions

[1033] Based on the analysis results, the server notifies employees of the optimal timing for breaks and how to refresh themselves. For example, it may notify employees that they should take a 15-minute break in 10 minutes, along with a suggestion to refresh themselves with a cup of coffee. For factory robots, it suggests the appropriate timing and method for maintenance. For example, it may suggest that robots be stopped to cool down or that they need oil.

[1034] Collection of results and feedback

[1035] The refresh methods performed by employees and the results of factory robot maintenance are collected again and sent to the server, which updates the database based on these results and retrains the AI ​​model to improve the accuracy of the next analysis.

[1036] Specific examples

[1037] A concrete example is the program Tanaka. The device records Tanaka's slowing keystroke speed and rising heart rate, and judges her to be "highly stressed" based on her voice and facial expression. The server analyzes this data, recognizes her as being in a "high stress state," and notifies her to "take a 15-minute break in 10 minutes," and suggests that she "have a cup of coffee and refresh herself." Tanaka follows the suggestion and takes a break, and the device again collects performance data after the break and sends it to the server. The server then updates the AI ​​model based on this data.

[1038] For example, in the case of a factory robot, if the motor temperature rises too high while the robot is operating, a sensor will report this abnormality to the server, which will then send a message to the robot to stop it to cool down, and a maintenance technician will inject oil or take other cooling measures.

[1039] Prompt Sentence Examples

[1040] "Create an AI model that predicts the stress level of a robot and suggests appropriate maintenance."

[1041] "Write a Python program to analyze stress and fatigue levels using data from factory robot sensors."

[1042] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1043] Step 1: Data collection

[1044] The server collects employee work data and emotional data from the devices. Specifically, it obtains data such as work hours, work content, work location, stress level, heart rate, and electrodermal activity from employees' PCs, smartphones, and wearable devices. It also performs voice analysis, facial recognition, and text analysis to collect emotional data such as anger, sadness, joy, and fear in real time. The input is sensor data from each device, and the output is the collected work data and emotional data.

[1045] Step 2: Preprocessing

[1046] The server performs preprocessing on the collected data, such as normalizing, filling in missing values, and removing outliers. Specifically, it detects and removes outliers and fills in missing data. The input is the collected raw data, and the output is the preprocessed data.

[1047] Step 3: Data analysis

[1048] The server inputs the preprocessed data into an AI model to predict the stress level and fatigue level for each employee. Specifically, the data is input into the AI ​​model, and the predicted stress level and fatigue level are output. The input is the preprocessed data, and the output is the predicted stress level and fatigue level.

[1049] Step 4: Notification based on analysis results

[1050] Based on the analysis results, the server determines the optimal timing for breaks and ways to refresh, and notifies employees via their devices. Specifically, it sends notifications such as "Take a 15-minute break in 10 minutes" or "Refresh yourself with a cup of coffee." The input is the predicted stress level and fatigue level, and the output is notifications on break timing and ways to refresh.

[1051] Step 5: Employee execution and collection of results

[1052] The employee performs the refresh method, and the terminal re-collects the subsequent performance data. Specifically, it again records the recovery of keystroke speed, the reduction of error messages, etc. The input is the employee's performance data after the method is performed, and the output is the collected result data.

[1053] Step 6: Feedback and model improvement

[1054] The server evaluates the effectiveness of the refreshment method based on the collected result data and retrains the AI ​​model. Specifically, it analyzes the degree of improvement in stress and fatigue and updates the algorithm. The input is the result data, and the output is an improved AI model.

[1055] Step 7: Robot Data Collection

[1056] The server collects operational and sensor data from the robots in the factory, specifically motor temperature, operating time, error rate, etc. The input is the robot sensor data, and the output is the collected robot data.

[1057] Step 8: Robot data analysis

[1058] The server inputs the collected robot data into an AI model to predict signs of stress and fatigue. Specifically, it analyzes stress levels and fatigue levels based on the data. The input is the collected robot data, and the output is the predicted stress levels and fatigue levels.

[1059] Step 9: Maintenance Recommendations

[1060] The server then suggests appropriate maintenance timing and methods based on the analysis results. Specifically, it makes suggestions such as "Stop the vehicle to cool it down" or "Oil injection is required." The input is the predicted stress level and fatigue level, and the output is a notification of the maintenance method.

[1061] Step 10: Collect maintenance results

[1062] The server collects the results of maintenance performed by the maintenance technician and uses them for future analysis. Specifically, it records the robot performance data after maintenance. The input is the performance data after maintenance, and the output is the collected result data.

[1063] 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.

[1064] 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.

[1065] 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.

[1066] [Third embodiment]

[1067] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1068] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1069] 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).

[1070] 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.

[1071] 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.

[1072] 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).

[1073] 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.

[1074] 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.

[1075] 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.

[1076] 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.

[1077] 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.

[1078] 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."

[1079] This invention relates to a system that uses AI to analyze employee work data and proposes appropriate break timing and refreshment methods. Specific embodiments of this system are described below.

[1080] System configuration

[1081] The system consists of a terminal that collects employee work data, a server that performs analysis, and a terminal that sends notifications and re-collects data.

[1082] 1. Data Collection

[1083] The server connects to each employee's device (PC, smartphone, wearable device) and collects work data, including work hours, work content, work location, stress level, etc.

[1084] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[1085] Users (employees) input information such as self-evaluations in the form of a questionnaire at the start and end of work, and periodically (for example, every 1-2 hours).

[1086] 2. Data Analysis

[1087] The server receives the collected data and preprocesses it (e.g., normalizes and treats outliers).

[1088] The server inputs the preprocessed data into an AI model to predict stress and fatigue levels, thereby calculating the degree of stress and fatigue for each employee.

[1089] 3. Notifications and refresh suggestions

[1090] Based on the analysis results, the server determines the optimal break timing and method of refreshment for each employee.

[1091] The device notifies employees of break times based on the information received from the server, for example by displaying a message saying, "You will take a 15-minute break in 10 minutes."

[1092] The user receives the notification and can click on an option such as "OK" or "Later" as needed. If the notification is ignored, the server sets the timing for re-notification and prompts the user to take a break again after a certain period of time.

[1093] 4. Refresh execution and data collection

[1094] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[1095] The device collects performance data after a refresh, such as improved keystroke speed and fewer error messages.

[1096] The terminal transmits the execution data to the server.

[1097] 5. Feedback and model improvement

[1098] The server evaluates how effective the refreshment methods were in reducing employees' fatigue and stress.

[1099] The server readjusts the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next proposal.

[1100] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[1101] Specific examples

[1102] Example: Programmer Tanaka

[1103] The device recorded Tanaka's slowing keystroke speed and an increase in error messages, as well as his heart rate and self-assessment data, which revealed that he was experiencing a decline in concentration.

[1104] The server analyzed Tanaka's data and determined that his stress level was high.

[1105] The server notifies Tanaka, "In 10 minutes, you will take a 15-minute break and have a cup of coffee to refresh yourself."

[1106] The user (Mr. Tanaka) checks the notification and takes a break as suggested.

[1107] The device collects Tanaka's performance data again after the break and sends it to the server.

[1108] The server uses this data to update the AI ​​model and prepare for the next more precise analysis.

[1109] In this way, the system collects employees' work data, analyzes it using AI, and notifies them of the best times to take breaks and how to refresh themselves, thereby reducing employee stress and optimizing their performance.

[1110] The processing flow will be explained below.

[1111] The specific processing steps of the program and their operations will be explained below.

[1112] Step 1:

[1113] The device collects and records employees' work data in real time, such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[1114] Step 2:

[1115] The terminal displays a periodic questionnaire to the user (employee) and asks them to enter their self-evaluation (for example, their work progress or current physical condition).

[1116] Step 3:

[1117] The server receives the business data sent from the terminal and stores it in a database.

[1118] Step 4:

[1119] The server checks the received data and performs preprocessing (e.g., data normalization, missing value imputation, and outlier removal).

[1120] Step 5:

[1121] The server then feeds the pre-processed data into an AI model that predicts each employee's stress and fatigue levels using predictive algorithms based on past data and similar patterns.

[1122] Step 6:

[1123] The server determines the optimal break timing for each employee based on the prediction results of the AI ​​model.

[1124] Step 7:

[1125] The server references a database of each employee's hobbies and preferences and determines appropriate ways to refresh themselves, such as drinking coffee or stretching to take a break from work.

[1126] Step 8:

[1127] The server transmits the determined break timing and refreshment method to the terminal.

[1128] Step 9:

[1129] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[1130] Step 10:

[1131] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[1132] Step 11:

[1133] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[1134] Step 12:

[1135] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[1136] Step 13:

[1137] The terminal will again collect user performance data (e.g., improved keystroke speed, reduced error messages) after the refresh.

[1138] Step 14:

[1139] The terminal transmits the refreshed data to the server.

[1140] Step 15:

[1141] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[1142] Step 16:

[1143] The server stores the evaluation results in a database and uses them to retrain the AI ​​model, improving the accuracy of the next analysis.

[1144] By performing the above processing steps, it is possible to analyze employees' work data and suggest appropriate break times and methods for refreshing, thereby reducing employee stress and optimizing work performance.

[1145] Example 1

[1146] 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."

[1147] In today's work environment, many employees accumulate fatigue and stress due to long working hours, which can result in reduced productivity and adverse effects on their mental and physical health. To address this issue, a system that can suggest effective ways to refresh at the appropriate time is needed. However, existing methods have the challenge of making it difficult to provide appropriate timing and methods of refreshment that suit each employee's individual situation.

[1148] 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.

[1149] In this invention, the server includes means for collecting employee work data, pre-processing means for normalizing the collected work data and processing outliers, means for having an AI model for analyzing the pre-processed work data, means for determining the optimal break timing and refreshment method for the employee based on the analysis results, means for notifying the employee of the determined break timing and refreshment method, means for collecting data on the refreshment method performed by the employee and subsequent performance data, and means for readjusting the AI ​​model based on the collected results and updating the algorithm to be useful for the next analysis. This allows for the provision of appropriate break timing and refreshment methods according to the situation of each employee, thereby reducing employee stress and improving productivity.

[1150] "Work data" is data generated when an employee performs their work, including work hours, work content, work location, stress level, keyboard input speed, mouse movements, heart rate, and electrodermal activity.

[1151] "Preprocessing means" refers to the means of normalizing collected business data, removing outliers, and preparing the data in a form that is easy to analyze.

[1152] The "AI model" is a model that uses machine learning algorithms to predict employees' stress levels and fatigue levels from preprocessed data.

[1153] The "analysis method" is a method of inputting preprocessed business data into an AI model to calculate employee stress and fatigue levels.

[1154] "Break timing" refers to employees taking breaks at optimal times during work, and indicates time to reduce stress and fatigue.

[1155] "Refreshment methods" are methods suggested to reduce employees' stress and fatigue, such as drinking coffee or stretching their eyes.

[1156] The "notification means" is a means for informing employees of the break timing and refreshment method determined based on the analysis results.

[1157] "Performance data" is data collected after an employee performs a refresher technique, including recovery of keystroke speed and reduction in error messages.

[1158] "Readjustment measures" are measures for readjusting the AI ​​model based on collected performance data to improve the accuracy of the next analysis results.

[1159] The "algorithm updating means" is a means of updating the analysis algorithm to the latest version using a retuned AI model.

[1160] This invention is a system that collects employee work data, analyzes it using AI, and suggests appropriate break times and ways to refresh. The system continuously monitors employees' work status, aiming to reduce stress and fatigue and improve productivity.

[1161] System configuration

[1162] The system consists of a terminal that collects employee work data, a server that performs analysis, and a terminal that sends notifications and re-collects data.

[1163] Hardware and software used

[1164] Device: An input device such as a PC, smartphone, or wearable device.

[1165] Server: A central computer with data processing and analysis capabilities.

[1166] Software: Data collection software, pre-processing software, AI models, notification software, database management software.

[1167] Data collection

[1168] The server connects to each employee's device and collects work data, including work hours, work content, work location, and stress level.

[1169] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[1170] Users (employees) enter self-evaluations in the form of a questionnaire at the start and end of work, and periodically.

[1171] Data analysis

[1172] The server preprocesses the collected data, normalizes the data, and processes outliers.

[1173] The server inputs the preprocessed data into an AI model to predict the user's stress level and fatigue level.

[1174] Based on the analysis results, the server quantifies and evaluates the user's level of stress and fatigue.

[1175] Notifications and refresh suggestions

[1176] Based on the analysis results, the server determines the optimal timing for breaks and the method of refreshing.

[1177] The device notifies the user of the break based on the information received from the server. For example, it displays a message saying, "Take a 15-minute break in 10 minutes."

[1178] The user can check the notification and choose options such as "OK" or "Later." If the notification is ignored, the server will notify the user again.

[1179] Refresh execution and data collection

[1180] The user performs the suggested refreshment method, such as drinking a cup of coffee or doing eye stretches.

[1181] The device will collect performance data after the refresh, recording things like improved keystroke speed and fewer error messages.

[1182] The terminal retransmits the collected data to the server.

[1183] Feedback and model improvement

[1184] The server evaluates how effective the refreshment method was in reducing the user's fatigue and stress.

[1185] The server readjusts the AI ​​model based on the evaluation results to improve the accuracy of the next analysis results.

[1186] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[1187] Specific examples

[1188] Example: Programmer employee A

[1189] The device records Employee A's slowing keystroke speed and an increase in error messages, and also collects data on his heart rate and self-evaluation questionnaires that indicate his "declining ability to concentrate."

[1190] The server analyzes employee A's data and determines that his stress level is high.

[1191] The server notifies employee A, "Take a 15-minute break in 10 minutes and refresh yourself with some coffee."

[1192] The user (Employee A) checks the notification and takes a break as suggested.

[1193] The terminal collects the performance data of employee A again after the break and transmits it to the server.

[1194] The server updates the AI ​​model based on this data, preparing for the next more precise analysis.

[1195] Prompt Sentence Examples

[1196] Example: Describe a system that analyzes the stress level of programmer employee A based on his work data and suggests optimal break times and ways to refresh himself.

[1197] This system can reduce employee stress and optimize performance by informing them of the best time to take a break and how to refresh themselves based on their individual circumstances.

[1198] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1199] Step 1: Data collection

[1200] The user presses the start button on the terminal and fills out a self-evaluation questionnaire. The entered data is then saved directly on the terminal.

[1201] The device collects user input and real-time task data (such as keyboard typing speed, mouse movements, heart rate, and electrodermal activity), which is temporarily stored on the device.

[1202] The server receives work data periodically sent from the terminal, including work hours, work content, work location, stress level, etc.

[1203] (Input) User surveys, keyboard input speed, mouse movement, heart rate, etc.

[1204] (Output) The collected business data is sent to the server.

[1205] Specifically, when Tanaka starts work, he answers a questionnaire displayed on the device asking, "How are you feeling today?" and the device records his answers.

[1206] Step 2: Data Preprocessing

[1207] The server normalizes the business data received from the device and corrects or removes any abnormal values. For example, if the heart rate is extremely high, it determines that this is noise and corrects the value.

[1208] The server prepares the pre-processed data to be fed into the AI ​​model.

[1209] (Input) Collected business data.

[1210] (Output) Preprocessed data that has been normalized and outlier treated.

[1211] Specifically, the server analyzes Tanaka's heart rate data, and if an unnatural increase is detected, it corrects it to the correct value.

[1212] Step 3: Data analysis

[1213] The server then inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level, for example, assessing stress levels based on a decrease in keyboard typing speed or an increase in heart rate.

[1214] Based on the analysis results, the server quantifies and evaluates the user's level of stress and fatigue.

[1215] (Input) Preprocessed data.

[1216] (Output) Analysis results showing stress levels and fatigue levels.

[1217] Specifically, the server detects a decrease in Tanaka's keyboard input speed and evaluates this as a stress index.

[1218] Step 4: Notifications and refresh suggestions

[1219] The server then determines the optimal timing for rest and how to refresh based on the analysis results. For example, if the heart rate is high, it may recommend five minutes of deep breathing.

[1220] The device receives a notification from the server and displays a message to the user such as, "The next break is in 10 minutes. Have a cup of coffee and refresh yourself."

[1221] The user can review the notification and enter options such as "OK" or "later."

[1222] (Input) Analysis results.

[1223] (Output) Notification of break times and ways to refresh.

[1224] Specifically, the server notifies Tanaka that "we recommend that you take a coffee break in 10 minutes," and Tanaka responds with "Got it."

[1225] Step 5: Refresh execution and data collection

[1226] The user performs the suggested refreshment method (e.g., drinking coffee, stretching the eyes, etc.).

[1227] The terminal collects performance data after the refresh (e.g., recovery of keystroke speed, reduction in error messages).

[1228] The terminal retransmits the collected data to the server.

[1229] (Input) Performance data before the refresh is executed.

[1230] (Output) Performance data after the refresh is performed.

[1231] Specifically, Tanaka drinks coffee to refresh himself, and the data on his subsequent recovery of keyboard typing speed is collected on the device and sent to the server.

[1232] Step 6: Feedback and model improvement

[1233] The server evaluates how effective the refreshment methods were in reducing employees' fatigue and stress.

[1234] The server readjusts the AI ​​model based on the evaluation results to improve the accuracy of the next analysis results.

[1235] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[1236] (Input) Performance data after the refresh is executed.

[1237] (Output) Retuned AI models and updated algorithms.

[1238] Specifically, the server evaluates Tanaka's refreshed data and updates the model, improving the accuracy of suggestions from the next time onwards.

[1239] (Application example 1)

[1240] 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."

[1241] Conventional systems for suggesting breaks and refreshment methods determine the optimal timing by collecting and analyzing employees' work data, but the types of data available are limited, meaning that the suggestions cannot fully reflect the detailed stress and fatigue levels of individual employees.In addition, the limited number of devices available for notification makes it difficult to respond flexibly to the work environment.

[1242] 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.

[1243] In this invention, the server includes means for collecting employee work data, means for analyzing the collected work data, means for determining optimal break timing and refreshment methods for employees based on the analysis results, means for notifying employees of the determined break timing and refreshment methods, means for collecting the refreshment methods performed by employees and their results, means for updating the database based on the collected results and using them for the next analysis, means for collecting eye movements, heart rate, and body movements as work data in real time, and means for notifying workers of break timing and refreshment methods via their smart devices. This allows the system to reflect detailed stress and fatigue levels of employees in real time, making it possible to propose more accurate break timing and refreshment methods.

[1244] "Employee work data" refers to information generated by employees during work, such as time, content, location, heart rate, eye movements, body movements, and stress levels.

[1245] "Means for collection" refers to devices and systems for collecting employee work data in real time, including wearable devices such as smart glasses.

[1246] The "means of analysis" is a system that preprocesses collected business data and uses an AI model to analyze stress levels and fatigue levels.

[1247] The "means for determining" is a system that selects the optimal break timing and refreshment method for employees based on the analysis results.

[1248] The "notification means" refers to a device or application that notifies employees in real time of the determined break timing and refreshment method. Specifically, this is done via a smart device.

[1249] The "means of updating the database based on collected results" refers to a system that records the refresh methods used by employees and the results of those methods, updates the database based on those results, and uses them for the next analysis.

[1250] "Eye movement" is data on employees' eye movements and is an indicator used to evaluate their concentration level and stress level at work.

[1251] "Heart rate" is the number of times an employee's heart beats within a certain period of time, and is an indicator primarily used to assess stress and fatigue levels.

[1252] "Physical movement" refers to information about an employee's physical movement and is an indicator for evaluating work efficiency and fatigue level.

[1253] A "smart device" is a device that has Internet connectivity and is capable of data communication and application execution, and specifically includes smart glasses and smartphones.

[1254] The "AI model" is a model that uses machine learning algorithms to analyze business data and predict employees' stress levels and fatigue levels.

[1255] This invention is a system that uses AI to analyze employees' work data and propose optimal break timing and refreshment methods. The system consists of a means for collecting employees' work data, a means for analyzing the data, a means for determining and notifying break timing and refreshment methods, and a means for collecting and evaluating the refreshment results.

[1256] Hardware Configuration

[1257] The system consists of smart glasses for collecting data, wearable devices, and terminals such as PCs and smartphones. The data is collected on a server and analyzed. Specific devices used include Google Glass and Microsoft HoloLens.

[1258] Software Configuration

[1259] AI models are used for data analysis, and machine learning libraries such as TensorFlow are introduced, while data analysis libraries such as SciPy and NumPy are used for data preprocessing.

[1260] Process Overview

[1261] 1. Data Collection

[1262] The server collects real-time data on employees' eye movements, heart rate, and body movements via smart glasses or wearable devices.

[1263] The data is transmitted from the terminal to a server and recorded in real time.

[1264] 2. Data Analysis

[1265] The server preprocesses the collected data, eliminating outliers and normalizing it.

[1266] The preprocessed data is then fed into an AI model to analyze stress levels and fatigue levels.

[1267] 3. Notifications and refresh suggestions

[1268] Based on the analysis results, the server determines the optimal timing for breaks and the method of refreshing.

[1269] The decision will be communicated to employees via smart glasses or smartphones.

[1270] For example, a notification will appear saying, "Take a 15-minute break in 10 minutes and take some deep breaths."

[1271] 4. Collecting and Evaluating Refresh Results

[1272] After the employee executes the refresh method, the terminal recollects the resulting data.

[1273] The collected data is sent to a server and used by an AI model to evaluate its effectiveness.

[1274] 5. Feedback and model improvement

[1275] The effect of the refresh is evaluated and the AI ​​model is readjusted based on the results.

[1276] The database will be updated with new evaluation results and used for the next analysis.

[1277] Specific examples

[1278] For example, if a factory worker wears smart glasses and their eye movements deviate from a certain pattern, it will be judged as a "high stress level." The same applies if their heart rate rises sharply. In this case, the system will notify the worker to "take a break and take a short walk or stretch." After the worker takes a break as notified, the resulting data will be collected again and fed back into the AI ​​model.

[1279] Prompt Sentence Examples

[1280] Model inputs

[1281] Worker heart rate data:

[1282] [68, 70, 72, 75, 77, 80, 85, 90, 95, 100]

[1283] Worker gaze movement data (eye movement):

[1284] [0.5, 0.6, 1.2, 1.1, 0.8, 0.7, 1.3, 1.4, 1.5, 0.9]

[1285] Expected output

[1286] Your stress levels are high, so I suggest you take a 10-minute break and stretch.

[1287] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1288] Step 1:

[1289] The server collects data on employees' eye movements, heart rates, and body movements in real time via smart glasses or wearable devices. The input is raw data sent from each device, and the output is a business dataset that compiles this data. Specifically, all eye movement information obtained by eye tracking sensors and heart rate information obtained by heart rate sensors are aggregated on the server.

[1290] Step 2:

[1291] The terminal preprocesses the collected business data. The input is the raw data collected in step 1. Specific data processing includes removing outliers, normalizing the data, and removing noise. The output is a normalized dataset that has undergone these preprocessing steps. The server uses SciPy and NumPy to clean and format the data.

[1292] Step 3:

[1293] The server inputs the preprocessed data into an AI model to analyze stress levels and fatigue levels. The input is the normalized dataset generated in step 2. Specific data calculations include predicting stress levels using machine learning algorithms. The output is a numerical value indicating each employee's stress level and fatigue level. The server runs the AI ​​model using TensorFlow.

[1294] Step 4:

[1295] The server determines the optimal break timing and refreshment method based on the analysis results. The input is the analysis result from step 3. In concrete terms, the server determines the break timing and selects an appropriate refreshment method (e.g., stretching or deep breathing). The output is the determined break timing and refreshment method.

[1296] Step 5:

[1297] The terminal notifies the employee of the determined break timing and refreshment method. The input is the result of the decision made in step 4. The specific operation is to display a notification message on the smart glasses or smartphone. The output is a confirmation message of the notification. For example, the message "Take a 15-minute break in 10 minutes and take deep breaths" is displayed.

[1298] Step 6:

[1299] The user (employee) performs the suggested refreshment method based on the notification. The input is the notification message from step 5. Specifically, the employee performs a refreshment method such as stretching or deep breathing. The output is the result data of the refreshment method that was performed.

[1300] Step 7:

[1301] The device again collects data after the refresh method is executed. The input is the result data of the refresh method executed by the user. Specifically, heart rate, eye movement, etc. are collected again. The output is a performance dataset after the refresh.

[1302] Step 8:

[1303] The server evaluates the effectiveness of the refresh method based on the collected post-refresh data. The input is the post-refresh data from step 7. The specific data calculation involves comparing the data before and after the refresh and evaluating areas for improvement. The output is numerical data of the evaluation results.

[1304] Step 9:

[1305] The server readjusts the AI ​​model based on the evaluation results and updates the database. The input is the evaluation result from step 8. Specifically, it adjusts the parameters of the AI ​​model and adds new data points to the database. The output is an improved AI model and an updated database.

[1306] 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.

[1307] This invention relates to a system that uses AI to analyze employee work data and emotional data and suggests appropriate break timing and methods for refreshing. Specific embodiments of this system are described below.

[1308] System configuration

[1309] The system consists of a terminal that collects employees' work data and emotional data, a server that performs the analysis, and a terminal that sends notifications and re-collects the data.

[1310] 1. Data Collection

[1311] The server connects to each employee's device (PC, smartphone, wearable device) and collects work data and emotional data, including work hours, work content, work location, and stress level.

[1312] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[1313] The device uses an emotion engine to perform voice analysis, facial recognition, and text analysis to collect user (employee) emotional data in real time, including emotional states such as anger, sadness, joy, and fear.

[1314] Users (employees) input their emotional state by self-evaluating in the form of a questionnaire at the start and end of work, and periodically (for example, every 1-2 hours).

[1315] 2. Data Analysis

[1316] The server checks the received business data and emotion data and stores them in a database.

[1317] The server performs preprocessing (e.g., data normalization, missing value imputation, and outlier removal).

[1318] The server then inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level. By incorporating emotional data into the analysis results, more accurate predictions can be made.

[1319] 3. Notifications and refresh suggestions

[1320] Based on the analysis results, the server determines the optimal break timing for each employee.

[1321] The server refers to the employee's hobby and preference database and determines the appropriate way to refresh themselves, suggesting, for example, drinking coffee or stretching to take a break from work.

[1322] The server transmits the determined break timing and refreshment method to the terminal.

[1323] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[1324] 4. Refresh execution and data collection

[1325] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[1326] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[1327] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[1328] The terminal collects performance data after the refresh (e.g., recovery of key input speed, reduction in error messages) and sends it to the server.

[1329] 5. Feedback and model improvement

[1330] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[1331] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis.

[1332] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[1333] Specific examples

[1334] Example: Programmer Tanaka

[1335] The device records Tanaka's slowing keystroke speed and rising heart rate, and the emotion engine determines from Tanaka's voice and face that he is "highly stressed" or "feeling angry."

[1336] The server analyzed Tanaka's data and determined that he was in a "high stress state."

[1337] The server notifies Tanaka that he should take a 15-minute break in 10 minutes and also suggests that he should have a cup of coffee to refresh himself.

[1338] The user (Mr. Tanaka) checks the notification and takes a break as suggested.

[1339] The device recollects Tanaka's performance and emotional data after the break and sends it to the server.

[1340] The server updates the AI ​​model based on this data, preparing for the next more precise analysis.

[1341] In this way, the system collects and analyzes employees' work and emotional data, and notifies them of appropriate break times and ways to refresh themselves, thereby reducing employee stress and optimizing their work performance.

[1342] The processing flow will be explained below.

[1343] The specific processing steps of the system are described below.

[1344] Step 1:

[1345] The device collects and records employees' work data (keystroke speed, mouse movements, application usage history, heart rate, electrodermal activity, etc.) in real time.

[1346] Step 2:

[1347] The device uses an emotion engine to perform voice analysis, facial recognition, and text analysis to collect and record the user's (employee's) emotional data in real time. Emotional data includes emotional states such as anger, sadness, joy, and fear.

[1348] Step 3:

[1349] The device displays regular self-evaluations in the form of a questionnaire to the user (employee), prompting them to enter information about their work progress, current physical condition, and emotional state.

[1350] Step 4:

[1351] The terminal transmits the collected business data and emotional data to the server at regular intervals.

[1352] Step 5:

[1353] The server stores the received data in a database and performs preprocessing, which includes data normalization, missing value imputation, and outlier removal.

[1354] Step 6:

[1355] The server inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level. Emotional data is also incorporated into the analysis results to make more accurate predictions.

[1356] Step 7:

[1357] Based on the analysis results, the server determines the optimal break timing for each employee.

[1358] Step 8:

[1359] The server refers to the employee's hobby and preference database and determines the appropriate way to refresh themselves, suggesting, for example, drinking coffee or stretching to take a break from work.

[1360] Step 9:

[1361] The server transmits the determined break timing and refreshment method to the terminal.

[1362] Step 10:

[1363] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[1364] Step 11:

[1365] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[1366] Step 12:

[1367] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[1368] Step 13:

[1369] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[1370] Step 14:

[1371] After the device is refreshed, the user's performance data (e.g., recovery of keystroke speed, reduction in error messages) and emotional data are collected again.

[1372] Step 15:

[1373] The terminal transmits the refreshed data to the server.

[1374] Step 16:

[1375] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[1376] Step 17:

[1377] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis.

[1378] Step 18:

[1379] The server stores the evaluation results and analysis data in a database for subsequent analysis.

[1380] By executing the above processing steps, the system can analyze employees' work data and emotional data and suggest appropriate break times and ways to refresh themselves. This reduces employee stress and optimizes work performance. As a concrete example, the system makes real-time suggestions to programmer Tanaka about when to take a break and how to refresh himself. Even when Tanaka is busy with a project, he can refresh himself at the appropriate time and maintain work efficiency.

[1381] Example 2

[1382] 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."

[1383] In today's business environment, properly managing employee stress and fatigue and improving work efficiency are important issues. However, traditional methods have made it difficult to grasp the state of each employee in real time and suggest appropriate break times and ways to refresh. As a result, many companies are not managing their employees' stress levels properly, resulting in decreased work performance and increased health risks.

[1384] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting work data and emotion data, means for analyzing the collected work data and emotion data, and means for determining optimal break timing and refreshment methods based on the analysis results. This makes it possible to grasp employee stress and fatigue in real time and propose optimal break timing and refreshment methods to each employee.

[1385] "Business data" refers to information generated when employees perform their work, including work hours, work content, work location, keystroke speed, mouse movements, application program usage history, heart rate, etc.

[1386] "Emotional data" refers to information that indicates an employee's emotional state, including emotional states such as anger, sadness, joy, and fear obtained through voice analysis, facial recognition, and text analysis.

[1387] "Analysis results" refers to information obtained after analyzing collected data using AI models, etc., and predicts the stress levels and fatigue levels of each employee.

[1388] "Break timing" refers to the time of break recommended for employees based on the analysis results.

[1389] "Refreshment methods" refer to actions or methods that employees are encouraged to take during their breaks, such as drinking coffee or stretching.

[1390] "Notification means" refers to the means of informing employees of the determined break timing and method of refreshment, and notifications are given via PCs, smartphones, wearable devices, etc.

[1391] A "database" is a repository of information that stores collected data and analysis results and allows reuse as needed.

[1392] A "generative AI model" is a model that includes algorithms for analyzing data and making predictions using AI.

[1393] MODE FOR CARRYING OUT THE INVENTION

[1394] This invention is a system that uses AI to analyze employees' work data and emotional data and suggests appropriate break times and ways to refresh. The system consists of the following main components:

[1395] Data collection

[1396] The device connects

[1397] The server establishes a secure connection with each employee's device (PC, smartphone, wearable device, etc.), enabling real-time data collection. For example, this includes the PCs and smartwatches employees use to do their work.

[1398] Business data collection

[1399] The devices record real-time work data, such as keystroke speed, mouse movements, and application usage history, which is used to gain a detailed understanding of how employees are performing their work. For example, keystroke speed during programming tasks and the type of software they are using are recorded.

[1400] Collecting Emotional Data

[1401] The device uses an emotion engine that performs voice analysis, facial recognition, and text analysis to analyze the user's emotional state in real time and digitize it. For example, the device classifies the user's emotional state into categories such as "stress" or "joy" based on the voice during a call or facial expressions captured via a webcam.

[1402] Enter your self-assessment

[1403] Users enter their emotional state in the form of a self-assessment questionnaire at the start and end of work and at regular intervals. For example, every 1-2 hours, they answer a questionnaire in which they rate their "current level of fatigue" on a scale of 1-5.

[1404] Data analysis

[1405] Receiving and storing data

[1406] The server receives the collected work data and emotion data and stores it in a database. This allows for centralized management, including past data. For example, fluctuations in an employee's stress level over the past month can also be recorded.

[1407] Data Preprocessing

[1408] The server performs preprocessing such as data normalization, missing value imputation, and outlier removal. This process improves the quality of the data to obtain accurate analysis results. For example, incomplete data sets and extreme values ​​are removed.

[1409] Prediction by AI model

[1410] The server inputs the preprocessed data into a generative AI model to predict each employee's stress level and fatigue level. The generative AI model is an algorithm that learns from past data and predicts future conditions. For example, it predicts that "Programmer Tanaka" is currently in a high stress state.

[1411] Notifications and refresh suggestions

[1412] Deciding when to take a break

[1413] Based on the analysis results, the server determines the optimal break timing for each employee. For example, it may decide that "Tanaka-san" should take a 15-minute break in 10 minutes.

[1414] Suggestions for refreshing

[1415] The server refers to a database of employees' hobbies and preferences to determine appropriate ways to refresh themselves, such as "drinking coffee" or "doing eye stretching exercises."

[1416] Sending notifications

[1417] The server then sends the determined break timing and refreshment method to the terminal, which then displays a notification to the employee. For example, a notification such as "Let's have a cup of coffee and refresh yourself" will appear on Tanaka's PC.

[1418] Refresh execution and data collection

[1419] User Verification

[1420] The user checks the notification and selects an option such as "OK" or "later." For example, Tanaka selects "OK."

[1421] Snooze settings

[1422] If "Later" is selected, the server sets the timing of the re-notification and sends the break notification to the terminal again after a certain time. For example, the setting "Re-notify after 30 minutes" is set.

[1423] Execute the refresh method

[1424] The user performs the suggested refreshment method. For example, Tanaka drinks coffee.

[1425] Performance Data Collection

[1426] The device collects performance data after the refresh and sends it to the server. For example, Tanaka's keystroke speed after the refresh is recorded.

[1427] Feedback and model improvement

[1428] Evaluation of effectiveness

[1429] The server evaluates the effectiveness of the refreshment method and analyzes the degree of improvement in stress and fatigue. For example, it evaluates changes in Tanaka's stress level.

[1430] Retraining the AI ​​model

[1431] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis, ensuring that analysis is always based on the latest data, resulting in more accurate predictions.

[1432] Data storage

[1433] The server stores the evaluation results and analysis data in a database for subsequent analysis, which can be used for future data analysis and model improvement.

[1434] Prompt Sentence Examples

[1435] "Design a system that analyzes employees' work data and emotional data to suggest appropriate break times and ways to refresh. The following input data is provided: work hours, work content, work location, keystroke speed, mouse movement, heart rate, voice analysis data, and facial recognition data. Use this data to predict employees' stress levels and fatigue levels."

[1436] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1437] Step 1:

[1438] Data collection

[1439] The server establishes a connection: The server establishes a secure connection with each employee's device (PC, smartphone, wearable device), allowing data to be sent and received in real time.

[1440] Input: Connection request from employee's device.

[1441] Data processing / calculation: Establish a secure connection and perform authentication.

[1442] Output: Establishment of a secure communication channel with the device.

[1443] Specific operation: Tanaka's smartwatch connects to the server and starts sending data.

[1444] Devices collect business data: Devices collect business data in real time, such as keystroke speed, mouse movements, and application usage history.

[1445] Input: Employee operation data (keystrokes, mouse movements, etc.).

[1446] Data processing / calculation: Recording of operational data and converting it into the required format.

[1447] Output: Formatted business data.

[1448] Specific operation: Tanaka's keystroke speed and mouse movements are recorded while he works in Excel.

[1449] Device collects emotional data: The device collects the user's emotional data using voice analysis, facial recognition, and text analysis.

[1450] Input: Audio data, facial image data, text data.

[1451] Data processing / calculation: Emotional states are analyzed using an emotion engine and converted into data.

[1452] Output: Emotion data (e.g. anger, joy, sadness, etc.).

[1453] Specific operation: While Tanaka is on the phone, the emotion of "anger" is determined and recorded as data.

[1454] Users enter self-evaluations: Users answer self-evaluation questionnaires at the start and end of work, and at regular intervals.

[1455] Input: User's self-assessment survey responses.

[1456] Data processing / calculation: Formatting the survey responses for storage in the database.

[1457] Output: Formatted self-assessment data.

[1458] Specific actions: Tanaka inputs his "current level of fatigue" every hour on a 5-point scale.

[1459] Step 2:

[1460] Data analysis

[1461] The server receives and stores the data: The server stores the received business data and emotion data in a database.

[1462] Input: Business and sentiment data collected in real time.

[1463] Data processing / calculation: Convert into a format that can be saved in a database.

[1464] Output: Business and sentiment data stored in a database.

[1465] Specific action: Tanaka's data is registered in the database.

[1466] The server performs data preprocessing: The server normalizes the data, imputes missing values, removes outliers, etc.

[1467] Input: Stored business and sentiment data.

[1468] Data processing / calculation: data normalization, missing value imputation, outlier removal.

[1469] Output: Clean data after preprocessing.

[1470] Specific action: Tanaka's data is normalized and incomplete data is complemented.

[1471] The server makes predictions using an AI model: The preprocessed data is fed into a generative AI model to predict stress levels and fatigue levels.

[1472] Input: Preprocessed business data and sentiment data.

[1473] Data processing / calculation: Perform predictive analysis using generative AI models.

[1474] Output: Predicted stress and fatigue levels for each employee.

[1475] Specific action: Tanaka's data is input into the AI ​​model, and it is predicted that he is in a state of high stress.

[1476] Step 3:

[1477] Notifications and refresh suggestions

[1478] The server determines the timing of breaks: Based on the analysis results, the server determines the optimal timing of breaks for each employee.

[1479] Inputs: Forecast results and current business progress.

[1480] Data processing / calculation: Calculate the optimal break timing.

[1481] Output: Break timing decision result.

[1482] Specific action: Tanaka decides the timing, "I will take a 15-minute break in 10 minutes."

[1483] The server suggests a refreshing method: It refers to the hobby and preference database and suggests an appropriate refreshing method.

[1484] Input: Hobby and preference information for each employee and analysis results.

[1485] Data processing / calculation: Determine the appropriate refresh method.

[1486] Output: The proposed refresh method.

[1487] Specific action: A method for "drinking coffee" is suggested to Tanaka.

[1488] The server sends a notification: The determined break timing and refreshment method are sent to the terminal. The terminal displays the notification to the employee.

[1489] Input: Break timing and refreshment method decisions.

[1490] Data processing / calculation: generating and sending notification messages.

[1491] Output: Notifications that appear on the device.

[1492] Specific action: A message will appear on Tanaka's PC saying, "Take a 15-minute break in 10 minutes and have a cup of coffee."

[1493] Step 4:

[1494] Refresh execution and data collection

[1495] User acknowledges notification: Employee acknowledges notification and selects an option such as "Got it" or "Later."

[1496] Input: Notification message.

[1497] Data processing / calculation: Record the user's selection results.

[1498] Output: The user's selection.

[1499] Specific action: Tanaka selects "Got it."

[1500] Server will set snooze: If the user selects "Later", this sets the snooze timing. The server will snooze after a certain time.

[1501] Input: User's choice ("Later").

[1502] Data processing / calculation: Calculation and setting of re-notification timing.

[1503] Output: Snooze settings.

[1504] Specific operation: If Tanaka selects "Later", the server will schedule a reminder for 30 minutes.

[1505] User executes the refresh method: The user executes the proposed refresh method.

[1506] Input: Refresh suggestion for notification message.

[1507] Data processing / calculation: Record of refresh methods performed.

[1508] Output: State data after refresh.

[1509] Specific action: Tanaka drinks coffee.

[1510] The terminal collects performance data: The terminal collects performance data after the refresh and sends it to the server.

[1511] Input: User operation data and physiological data.

[1512] Data processing / calculation: recording and transmission of performance data.

[1513] Output: Performance data after refresh.

[1514] Specific action: After Tanaka's break, his input speed improves and the data is sent from his device to the server.

[1515] Step 5:

[1516] Feedback and model improvement

[1517] The server evaluates the effect: The effect of the refreshment method is evaluated and the degree of improvement in stress and fatigue is analyzed.

[1518] Input: Performance and emotion data after refresh.

[1519] Data processing / calculation: Calculation and evaluation of the degree of improvement.

[1520] Output: Evaluation results.

[1521] Specific action: Tanaka's change in stress level is evaluated.

[1522] The server retrains the AI ​​model: The AI ​​model is retrained based on the evaluation results to improve the accuracy of the next analysis.

[1523] Input: Assessment results and stored historical data.

[1524] Data processing / computation: Retraining AI models and updating algorithms.

[1525] Output: The updated AI model.

[1526] Specific action: The AI ​​model is updated based on new data.

[1527] The server stores the data: The evaluation results and analysis data are stored in a database for future analysis.

[1528] Input: Evaluation results and analysis data.

[1529] Data processing / calculation: Saving to database.

[1530] Output: Saved evaluation results and analysis data.

[1531] Specific actions: Updated evaluation results and analysis data are added to the database.

[1532] (Application example 2)

[1533] 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."

[1534] In today's work environment, properly collecting and analyzing employee work data and emotional data to suggest effective break timing and refreshment methods is important for reducing employee stress and improving productivity. Similarly, for robots operating in factories, collecting operational and sensor data and suggesting appropriate maintenance timing and methods is essential for maximizing robot operating efficiency and minimizing downtime. While systems exist to solve these issues individually, integrated systems are still lacking. Therefore, there is a need for the development of an integrated system that suggests optimal break and maintenance methods for both employees and robots.

[1535] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting work data from in-house devices, means for analyzing the collected work data and emotion data, means for determining optimal break timing and refreshment methods for employees based on the analysis results, means for notifying employees of the determined break timing and refreshment methods, means for collecting refreshment methods performed by employees and their results, means for updating the database based on the collected results and using them for the next analysis, means for collecting robot operation data and sensor data, means for analyzing the collected robot data and detecting signs of stress and fatigue, and means for proposing appropriate maintenance timing and methods based on the detected results. This makes it possible to reduce employee stress, improve productivity, maximize robot operating efficiency, and minimize downtime.

[1536] "Internal devices" are devices such as PCs, smartphones, and wearable devices that are used to collect employee business data and emotional data.

[1537] "Work data" refers to data including an employee's working hours, work content, work location, and stress level.

[1538] "Emotion data" is data that indicates emotional states such as anger, sadness, joy, and fear, obtained through voice analysis, facial recognition, and text analysis.

[1539] "Robot operation data" refers to data such as motor temperature, operating time, and error rate collected from sensors on robots operating in factories, etc.

[1540] "Sensor data" refers to real-time data collected by sensors mounted on the robot.

[1541] An "AI model" is an artificial intelligence algorithm used to analyze business data, emotional data, and robot operation data to predict stress and fatigue and suggest breaks and maintenance.

[1542] "Break timing" refers to the time that employees should take a break for work efficiency and health management.

[1543] "Refreshment methods" refer to actions and activities that employees should take to smoothly continue their work, such as drinking coffee or stretching.

[1544] "Maintenance timing" refers to the time when maintenance necessary to maintain the operating efficiency of a robot should be performed.

[1545] "Maintenance method" refers to the specific measures and actions that should be taken to keep the robot functioning optimally, such as oiling and replacing parts.

[1546] This invention includes a system that collects employee work data and emotional data, analyzes it in real time, and proposes optimal break timing and refreshment methods, and a system that analyzes the operation data of factory robots and proposes appropriate maintenance timing and methods. Specific embodiments of this invention are described below.

[1547] System configuration

[1548] This system consists of a means of collecting employee data, a server that analyzes the collected data, a means of issuing instructions based on the analysis results, and a means of collecting the results of refreshes and maintenance based on the instructions. Each means is explained in detail below.

[1549] Data collection from internal devices

[1550] The server connects to in-house devices such as PCs, smartphones, and wearable devices to collect employee work data. Data collected from these devices includes work hours, work content, work location, and stress level. Emotional data, such as anger, sadness, joy, and fear, is also collected in real time using voice analysis, facial recognition, and text analysis.

[1551] Data analysis

[1552] The server performs preprocessing on the collected work data and emotion data, such as normalizing it, filling in missing values, and removing outliers, and then inputs the data into an AI model to predict each employee's stress level and fatigue level, which can then individually suggest optimal break times and ways to refresh.

[1553] For factory robots, the server collects real-time data from each sensor and uses AI models to analyze signs of stress and fatigue based on data such as motor temperature, operating time, and error rate.

[1554] Notifications and refresh / maintenance suggestions

[1555] Based on the analysis results, the server notifies employees of the optimal timing for breaks and how to refresh themselves. For example, it may notify employees that they should take a 15-minute break in 10 minutes, along with a suggestion to refresh themselves with a cup of coffee. For factory robots, it suggests the appropriate timing and method for maintenance. For example, it may suggest that robots be stopped to cool down or that they need oil.

[1556] Collection of results and feedback

[1557] The refresh methods performed by employees and the results of factory robot maintenance are collected again and sent to the server, which updates the database based on these results and retrains the AI ​​model to improve the accuracy of the next analysis.

[1558] Specific examples

[1559] A concrete example is the program Tanaka. The device records Tanaka's slowing keystroke speed and rising heart rate, and judges her to be "highly stressed" based on her voice and facial expression. The server analyzes this data, recognizes her as being in a "high stress state," and notifies her to "take a 15-minute break in 10 minutes," and suggests that she "have a cup of coffee and refresh herself." Tanaka follows the suggestion and takes a break, and the device again collects performance data after the break and sends it to the server. The server then updates the AI ​​model based on this data.

[1560] For example, in the case of a factory robot, if the motor temperature rises too high while the robot is operating, a sensor will report this abnormality to the server, which will then send a message to the robot to stop it to cool down, and a maintenance technician will inject oil or take other cooling measures.

[1561] Prompt Sentence Examples

[1562] "Create an AI model that predicts the stress level of a robot and suggests appropriate maintenance."

[1563] "Write a Python program to analyze stress and fatigue levels using data from factory robot sensors."

[1564] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1565] Step 1: Data collection

[1566] The server collects employee work data and emotional data from the devices. Specifically, it obtains data such as work hours, work content, work location, stress level, heart rate, and electrodermal activity from employees' PCs, smartphones, and wearable devices. It also performs voice analysis, facial recognition, and text analysis to collect emotional data such as anger, sadness, joy, and fear in real time. The input is sensor data from each device, and the output is the collected work data and emotional data.

[1567] Step 2: Preprocessing

[1568] The server performs preprocessing on the collected data, such as normalizing, filling in missing values, and removing outliers. Specifically, it detects and removes outliers and fills in missing data. The input is the collected raw data, and the output is the preprocessed data.

[1569] Step 3: Data analysis

[1570] The server inputs the preprocessed data into an AI model to predict the stress level and fatigue level for each employee. Specifically, the data is input into the AI ​​model, and the predicted stress level and fatigue level are output. The input is the preprocessed data, and the output is the predicted stress level and fatigue level.

[1571] Step 4: Notification based on analysis results

[1572] Based on the analysis results, the server determines the optimal timing for breaks and ways to refresh, and notifies employees via their devices. Specifically, it sends notifications such as "Take a 15-minute break in 10 minutes" or "Refresh yourself with a cup of coffee." The input is the predicted stress level and fatigue level, and the output is notifications on break timing and ways to refresh.

[1573] Step 5: Employee execution and collection of results

[1574] The employee performs the refresh method, and the terminal re-collects the subsequent performance data. Specifically, it again records the recovery of keystroke speed, the reduction of error messages, etc. The input is the employee's performance data after the method is performed, and the output is the collected result data.

[1575] Step 6: Feedback and model improvement

[1576] The server evaluates the effectiveness of the refreshment method based on the collected result data and retrains the AI ​​model. Specifically, it analyzes the degree of improvement in stress and fatigue and updates the algorithm. The input is the result data, and the output is an improved AI model.

[1577] Step 7: Robot Data Collection

[1578] The server collects operational and sensor data from the robots in the factory, specifically motor temperature, operating time, error rate, etc. The input is the robot sensor data, and the output is the collected robot data.

[1579] Step 8: Robot data analysis

[1580] The server inputs the collected robot data into an AI model to predict signs of stress and fatigue. Specifically, it analyzes stress levels and fatigue levels based on the data. The input is the collected robot data, and the output is the predicted stress levels and fatigue levels.

[1581] Step 9: Maintenance Recommendations

[1582] The server then suggests appropriate maintenance timing and methods based on the analysis results. Specifically, it makes suggestions such as "Stop the vehicle to cool it down" or "Oil injection is required." The input is the predicted stress level and fatigue level, and the output is a notification of the maintenance method.

[1583] Step 10: Collect maintenance results

[1584] The server collects the results of maintenance performed by the maintenance technician and uses them for future analysis. Specifically, it records the robot performance data after maintenance. The input is the performance data after maintenance, and the output is the collected result data.

[1585] 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.

[1586] 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.

[1587] 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.

[1588] [Fourth embodiment]

[1589] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1590] 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.

[1591] 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).

[1592] 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.

[1593] 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.

[1594] 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).

[1595] 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.

[1596] 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.

[1597] 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.

[1598] 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.

[1599] 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.

[1600] 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.

[1601] 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."

[1602] This invention relates to a system that uses AI to analyze employee work data and proposes appropriate break timing and refreshment methods. Specific embodiments of this system are described below.

[1603] System configuration

[1604] The system consists of a terminal that collects employee work data, a server that performs analysis, and a terminal that sends notifications and re-collects data.

[1605] 1. Data Collection

[1606] The server connects to each employee's device (PC, smartphone, wearable device) and collects work data, including work hours, work content, work location, stress level, etc.

[1607] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[1608] Users (employees) input information such as self-evaluations in the form of a questionnaire at the start and end of work, and periodically (for example, every 1-2 hours).

[1609] 2. Data Analysis

[1610] The server receives the collected data and preprocesses it (e.g., normalizes and treats outliers).

[1611] The server inputs the preprocessed data into an AI model to predict stress and fatigue levels, thereby calculating the degree of stress and fatigue for each employee.

[1612] 3. Notifications and refresh suggestions

[1613] Based on the analysis results, the server determines the optimal break timing and method of refreshment for each employee.

[1614] The device notifies employees of break times based on the information received from the server, for example by displaying a message saying, "You will take a 15-minute break in 10 minutes."

[1615] The user receives the notification and can click on an option such as "OK" or "Later" as needed. If the notification is ignored, the server sets the timing for re-notification and prompts the user to take a break again after a certain period of time.

[1616] 4. Refresh execution and data collection

[1617] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[1618] The device collects performance data after a refresh, such as improved keystroke speed and fewer error messages.

[1619] The terminal transmits the execution data to the server.

[1620] 5. Feedback and model improvement

[1621] The server evaluates how effective the refreshment methods were in reducing employees' fatigue and stress.

[1622] The server readjusts the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next proposal.

[1623] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[1624] Specific examples

[1625] Example: Programmer Tanaka

[1626] The device recorded Tanaka's slowing keystroke speed and an increase in error messages, as well as his heart rate and self-assessment data, which revealed that he was experiencing a decline in concentration.

[1627] The server analyzed Tanaka's data and determined that his stress level was high.

[1628] The server notifies Tanaka, "In 10 minutes, you will take a 15-minute break and have a cup of coffee to refresh yourself."

[1629] The user (Mr. Tanaka) checks the notification and takes a break as suggested.

[1630] The device collects Tanaka's performance data again after the break and sends it to the server.

[1631] The server uses this data to update the AI ​​model and prepare for the next more precise analysis.

[1632] In this way, the system collects employees' work data, analyzes it using AI, and notifies them of the best times to take breaks and how to refresh themselves, thereby reducing employee stress and optimizing their performance.

[1633] The processing flow will be explained below.

[1634] The specific processing steps of the program and their operations will be explained below.

[1635] Step 1:

[1636] The device collects and records employees' work data in real time, such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[1637] Step 2:

[1638] The terminal displays a periodic questionnaire to the user (employee) and asks them to enter their self-evaluation (for example, their work progress or current physical condition).

[1639] Step 3:

[1640] The server receives the business data sent from the terminal and stores it in a database.

[1641] Step 4:

[1642] The server checks the received data and performs preprocessing (e.g., data normalization, missing value imputation, and outlier removal).

[1643] Step 5:

[1644] The server then feeds the pre-processed data into an AI model that predicts each employee's stress and fatigue levels using predictive algorithms based on past data and similar patterns.

[1645] Step 6:

[1646] The server determines the optimal break timing for each employee based on the prediction results of the AI ​​model.

[1647] Step 7:

[1648] The server references a database of each employee's hobbies and preferences and determines appropriate ways to refresh themselves, such as drinking coffee or stretching to take a break from work.

[1649] Step 8:

[1650] The server transmits the determined break timing and refreshment method to the terminal.

[1651] Step 9:

[1652] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[1653] Step 10:

[1654] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[1655] Step 11:

[1656] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[1657] Step 12:

[1658] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[1659] Step 13:

[1660] The terminal will again collect user performance data (e.g., improved keystroke speed, reduced error messages) after the refresh.

[1661] Step 14:

[1662] The terminal transmits the refreshed data to the server.

[1663] Step 15:

[1664] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[1665] Step 16:

[1666] The server stores the evaluation results in a database and uses them to retrain the AI ​​model, improving the accuracy of the next analysis.

[1667] By performing the above processing steps, it is possible to analyze employees' work data and suggest appropriate break times and methods for refreshing, thereby reducing employee stress and optimizing work performance.

[1668] Example 1

[1669] 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."

[1670] In today's work environment, many employees accumulate fatigue and stress due to long working hours, which can result in reduced productivity and adverse effects on their mental and physical health. To address this issue, a system that can suggest effective ways to refresh at the appropriate time is needed. However, existing methods have the challenge of making it difficult to provide appropriate timing and methods of refreshment that suit each employee's individual situation.

[1671] 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.

[1672] In this invention, the server includes means for collecting employee work data, pre-processing means for normalizing the collected work data and processing outliers, means for having an AI model for analyzing the pre-processed work data, means for determining the optimal break timing and refreshment method for the employee based on the analysis results, means for notifying the employee of the determined break timing and refreshment method, means for collecting data on the refreshment method performed by the employee and subsequent performance data, and means for readjusting the AI ​​model based on the collected results and updating the algorithm to be useful for the next analysis. This allows for the provision of appropriate break timing and refreshment methods according to the situation of each employee, thereby reducing employee stress and improving productivity.

[1673] "Work data" is data generated when an employee performs their work, including work hours, work content, work location, stress level, keyboard input speed, mouse movements, heart rate, and electrodermal activity.

[1674] "Preprocessing means" refers to the means of normalizing collected business data, removing outliers, and preparing the data in a form that is easy to analyze.

[1675] The "AI model" is a model that uses machine learning algorithms to predict employees' stress levels and fatigue levels from preprocessed data.

[1676] The "analysis method" is a method of inputting preprocessed business data into an AI model to calculate employee stress and fatigue levels.

[1677] "Break timing" refers to employees taking breaks at optimal times during work, and indicates time to reduce stress and fatigue.

[1678] "Refreshment methods" are methods suggested to reduce employees' stress and fatigue, such as drinking coffee or stretching their eyes.

[1679] The "notification means" is a means for informing employees of the break timing and refreshment method determined based on the analysis results.

[1680] "Performance data" is data collected after an employee performs a refresher technique, including recovery of keystroke speed and reduction in error messages.

[1681] "Readjustment measures" are measures for readjusting the AI ​​model based on collected performance data to improve the accuracy of the next analysis results.

[1682] The "algorithm updating means" is a means of updating the analysis algorithm to the latest version using a retuned AI model.

[1683] This invention is a system that collects employee work data, analyzes it using AI, and suggests appropriate break times and ways to refresh. The system continuously monitors employees' work status, aiming to reduce stress and fatigue and improve productivity.

[1684] System configuration

[1685] The system consists of a terminal that collects employee work data, a server that performs analysis, and a terminal that sends notifications and re-collects data.

[1686] Hardware and software used

[1687] Device: An input device such as a PC, smartphone, or wearable device.

[1688] Server: A central computer with data processing and analysis capabilities.

[1689] Software: Data collection software, pre-processing software, AI models, notification software, database management software.

[1690] Data collection

[1691] The server connects to each employee's device and collects work data, including work hours, work content, work location, and stress level.

[1692] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[1693] Users (employees) enter self-evaluations in the form of a questionnaire at the start and end of work, and periodically.

[1694] Data analysis

[1695] The server preprocesses the collected data, normalizes the data, and processes outliers.

[1696] The server inputs the preprocessed data into an AI model to predict the user's stress level and fatigue level.

[1697] Based on the analysis results, the server quantifies and evaluates the user's level of stress and fatigue.

[1698] Notifications and refresh suggestions

[1699] Based on the analysis results, the server determines the optimal timing for breaks and the method of refreshing.

[1700] The device notifies the user of the break based on the information received from the server. For example, it displays a message saying, "Take a 15-minute break in 10 minutes."

[1701] The user can check the notification and choose options such as "OK" or "Later." If the notification is ignored, the server will notify the user again.

[1702] Refresh execution and data collection

[1703] The user performs the suggested refreshment method, such as drinking a cup of coffee or doing eye stretches.

[1704] The device will collect performance data after the refresh, recording things like improved keystroke speed and fewer error messages.

[1705] The terminal retransmits the collected data to the server.

[1706] Feedback and model improvement

[1707] The server evaluates how effective the refreshment method was in reducing the user's fatigue and stress.

[1708] The server readjusts the AI ​​model based on the evaluation results to improve the accuracy of the next analysis results.

[1709] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[1710] Specific examples

[1711] Example: Programmer employee A

[1712] The device records Employee A's slowing keystroke speed and an increase in error messages, and also collects data on his heart rate and self-evaluation questionnaires that indicate his "declining ability to concentrate."

[1713] The server analyzes employee A's data and determines that his stress level is high.

[1714] The server notifies employee A, "Take a 15-minute break in 10 minutes and refresh yourself with some coffee."

[1715] The user (Employee A) checks the notification and takes a break as suggested.

[1716] The terminal collects the performance data of employee A again after the break and transmits it to the server.

[1717] The server updates the AI ​​model based on this data, preparing for the next more precise analysis.

[1718] Prompt Sentence Examples

[1719] Example: Describe a system that analyzes the stress level of programmer employee A based on his work data and suggests optimal break times and ways to refresh himself.

[1720] This system can reduce employee stress and optimize performance by informing them of the best time to take a break and how to refresh themselves based on their individual circumstances.

[1721] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1722] Step 1: Data collection

[1723] The user presses the start button on the terminal and fills out a self-evaluation questionnaire. The entered data is then saved directly on the terminal.

[1724] The device collects user input and real-time task data (such as keyboard typing speed, mouse movements, heart rate, and electrodermal activity), which is temporarily stored on the device.

[1725] The server receives work data periodically sent from the terminal, including work hours, work content, work location, stress level, etc.

[1726] (Input) User surveys, keyboard input speed, mouse movement, heart rate, etc.

[1727] (Output) The collected business data is sent to the server.

[1728] Specifically, when Tanaka starts work, he answers a questionnaire displayed on the device asking, "How are you feeling today?" and the device records his answers.

[1729] Step 2: Data Preprocessing

[1730] The server normalizes the business data received from the device and corrects or removes any abnormal values. For example, if the heart rate is extremely high, it determines that this is noise and corrects the value.

[1731] The server prepares the pre-processed data to be fed into the AI ​​model.

[1732] (Input) Collected business data.

[1733] (Output) Preprocessed data that has been normalized and outlier treated.

[1734] Specifically, the server analyzes Tanaka's heart rate data, and if an unnatural increase is detected, it corrects it to the correct value.

[1735] Step 3: Data analysis

[1736] The server then inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level, for example, assessing stress levels based on a decrease in keyboard typing speed or an increase in heart rate.

[1737] Based on the analysis results, the server quantifies and evaluates the user's level of stress and fatigue.

[1738] (Input) Preprocessed data.

[1739] (Output) Analysis results showing stress levels and fatigue levels.

[1740] Specifically, the server detects a decrease in Tanaka's keyboard input speed and evaluates this as a stress index.

[1741] Step 4: Notifications and refresh suggestions

[1742] The server then determines the optimal timing for rest and how to refresh based on the analysis results. For example, if the heart rate is high, it may recommend five minutes of deep breathing.

[1743] The device receives a notification from the server and displays a message to the user such as, "The next break is in 10 minutes. Have a cup of coffee and refresh yourself."

[1744] The user can review the notification and enter options such as "OK" or "later."

[1745] (Input) Analysis results.

[1746] (Output) Notification of break times and ways to refresh.

[1747] Specifically, the server notifies Tanaka that "we recommend that you take a coffee break in 10 minutes," and Tanaka responds with "Got it."

[1748] Step 5: Refresh execution and data collection

[1749] The user performs the suggested refreshment method (e.g., drinking coffee, stretching the eyes, etc.).

[1750] The terminal collects performance data after the refresh (e.g., recovery of keystroke speed, reduction in error messages).

[1751] The terminal retransmits the collected data to the server.

[1752] (Input) Performance data before the refresh is executed.

[1753] (Output) Performance data after the refresh is performed.

[1754] Specifically, Tanaka drinks coffee to refresh himself, and the data on his subsequent recovery of keyboard typing speed is collected on the device and sent to the server.

[1755] Step 6: Feedback and model improvement

[1756] The server evaluates how effective the refreshment methods were in reducing employees' fatigue and stress.

[1757] The server readjusts the AI ​​model based on the evaluation results to improve the accuracy of the next analysis results.

[1758] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[1759] (Input) Performance data after the refresh is executed.

[1760] (Output) Retuned AI models and updated algorithms.

[1761] Specifically, the server evaluates Tanaka's refreshed data and updates the model, improving the accuracy of suggestions from the next time onwards.

[1762] (Application example 1)

[1763] 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."

[1764] Conventional systems for suggesting breaks and refreshment methods determine the optimal timing by collecting and analyzing employees' work data, but the types of data available are limited, meaning that the suggestions cannot fully reflect the detailed stress and fatigue levels of individual employees.In addition, the limited number of devices available for notification makes it difficult to respond flexibly to the work environment.

[1765] 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.

[1766] In this invention, the server includes means for collecting employee work data, means for analyzing the collected work data, means for determining optimal break timing and refreshment methods for employees based on the analysis results, means for notifying employees of the determined break timing and refreshment methods, means for collecting the refreshment methods performed by employees and their results, means for updating the database based on the collected results and using them for the next analysis, means for collecting eye movements, heart rate, and body movements as work data in real time, and means for notifying workers of break timing and refreshment methods via their smart devices. This allows the system to reflect detailed stress and fatigue levels of employees in real time, making it possible to propose more accurate break timing and refreshment methods.

[1767] "Employee work data" refers to information generated by employees during work, such as time, content, location, heart rate, eye movements, body movements, and stress levels.

[1768] "Means for collection" refers to devices and systems for collecting employee work data in real time, including wearable devices such as smart glasses.

[1769] The "means of analysis" is a system that preprocesses collected business data and uses an AI model to analyze stress levels and fatigue levels.

[1770] The "means for determining" is a system that selects the optimal break timing and refreshment method for employees based on the analysis results.

[1771] The "notification means" refers to a device or application that notifies employees in real time of the determined break timing and refreshment method. Specifically, this is done via a smart device.

[1772] The "means of updating the database based on collected results" refers to a system that records the refresh methods used by employees and the results of those methods, updates the database based on those results, and uses them for the next analysis.

[1773] "Eye movement" is data on employees' eye movements and is an indicator used to evaluate their concentration level and stress level at work.

[1774] "Heart rate" is the number of times an employee's heart beats within a certain period of time, and is an indicator primarily used to assess stress and fatigue levels.

[1775] "Physical movement" refers to information about an employee's physical movement and is an indicator for evaluating work efficiency and fatigue level.

[1776] A "smart device" is a device that has Internet connectivity and is capable of data communication and application execution, and specifically includes smart glasses and smartphones.

[1777] The "AI model" is a model that uses machine learning algorithms to analyze business data and predict employees' stress levels and fatigue levels.

[1778] This invention is a system that uses AI to analyze employees' work data and propose optimal break timing and refreshment methods. The system consists of a means for collecting employees' work data, a means for analyzing the data, a means for determining and notifying break timing and refreshment methods, and a means for collecting and evaluating the refreshment results.

[1779] Hardware Configuration

[1780] The system consists of smart glasses for collecting data, wearable devices, and terminals such as PCs and smartphones. The data is collected on a server and analyzed. Specific devices used include Google Glass and Microsoft HoloLens.

[1781] Software Configuration

[1782] AI models are used for data analysis, and machine learning libraries such as TensorFlow are introduced, while data analysis libraries such as SciPy and NumPy are used for data preprocessing.

[1783] Process Overview

[1784] 1. Data Collection

[1785] The server collects real-time data on employees' eye movements, heart rate, and body movements via smart glasses or wearable devices.

[1786] The data is transmitted from the terminal to a server and recorded in real time.

[1787] 2. Data Analysis

[1788] The server preprocesses the collected data, eliminating outliers and normalizing it.

[1789] The preprocessed data is then fed into an AI model to analyze stress levels and fatigue levels.

[1790] 3. Notifications and refresh suggestions

[1791] Based on the analysis results, the server determines the optimal timing for breaks and the method of refreshing.

[1792] The decision will be communicated to employees via smart glasses or smartphones.

[1793] For example, a notification will appear saying, "Take a 15-minute break in 10 minutes and take some deep breaths."

[1794] 4. Collecting and Evaluating Refresh Results

[1795] After the employee executes the refresh method, the terminal recollects the resulting data.

[1796] The collected data is sent to a server and used by an AI model to evaluate its effectiveness.

[1797] 5. Feedback and model improvement

[1798] The effect of the refresh is evaluated and the AI ​​model is readjusted based on the results.

[1799] The database will be updated with new evaluation results and used for the next analysis.

[1800] Specific examples

[1801] For example, if a factory worker wears smart glasses and their eye movements deviate from a certain pattern, it will be judged as a "high stress level." The same applies if their heart rate rises sharply. In this case, the system will notify the worker to "take a break and take a short walk or stretch." After the worker takes a break as notified, the resulting data will be collected again and fed back into the AI ​​model.

[1802] Prompt Sentence Examples

[1803] Model inputs

[1804] Worker heart rate data:

[1805] [68, 70, 72, 75, 77, 80, 85, 90, 95, 100]

[1806] Worker gaze movement data (eye movement):

[1807] [0.5, 0.6, 1.2, 1.1, 0.8, 0.7, 1.3, 1.4, 1.5, 0.9]

[1808] Expected output

[1809] Your stress levels are high, so I suggest you take a 10-minute break and stretch.

[1810] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1811] Step 1:

[1812] The server collects data on employees' eye movements, heart rates, and body movements in real time via smart glasses or wearable devices. The input is raw data sent from each device, and the output is a business dataset that compiles this data. Specifically, all eye movement information obtained by eye tracking sensors and heart rate information obtained by heart rate sensors are aggregated on the server.

[1813] Step 2:

[1814] The terminal preprocesses the collected business data. The input is the raw data collected in step 1. Specific data processing includes removing outliers, normalizing the data, and removing noise. The output is a normalized dataset that has undergone these preprocessing steps. The server uses SciPy and NumPy to clean and format the data.

[1815] Step 3:

[1816] The server inputs the preprocessed data into an AI model to analyze stress levels and fatigue levels. The input is the normalized dataset generated in step 2. Specific data calculations include predicting stress levels using machine learning algorithms. The output is a numerical value indicating each employee's stress level and fatigue level. The server runs the AI ​​model using TensorFlow.

[1817] Step 4:

[1818] The server determines the optimal break timing and refreshment method based on the analysis results. The input is the analysis result from step 3. In concrete terms, the server determines the break timing and selects an appropriate refreshment method (e.g., stretching or deep breathing). The output is the determined break timing and refreshment method.

[1819] Step 5:

[1820] The terminal notifies the employee of the determined break timing and refreshment method. The input is the result of the decision made in step 4. The specific operation is to display a notification message on the smart glasses or smartphone. The output is a confirmation message of the notification. For example, the message "Take a 15-minute break in 10 minutes and take deep breaths" is displayed.

[1821] Step 6:

[1822] The user (employee) performs the suggested refreshment method based on the notification. The input is the notification message from step 5. Specifically, the employee performs a refreshment method such as stretching or deep breathing. The output is the result data of the refreshment method that was performed.

[1823] Step 7:

[1824] The device again collects data after the refresh method is executed. The input is the result data of the refresh method executed by the user. Specifically, heart rate, eye movement, etc. are collected again. The output is a performance dataset after the refresh.

[1825] Step 8:

[1826] The server evaluates the effectiveness of the refresh method based on the collected post-refresh data. The input is the post-refresh data from step 7. The specific data calculation involves comparing the data before and after the refresh and evaluating areas for improvement. The output is numerical data of the evaluation results.

[1827] Step 9:

[1828] The server readjusts the AI ​​model based on the evaluation results and updates the database. The input is the evaluation result from step 8. Specifically, it adjusts the parameters of the AI ​​model and adds new data points to the database. The output is an improved AI model and an updated database.

[1829] 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.

[1830] This invention relates to a system that uses AI to analyze employee work data and emotional data and suggests appropriate break timing and methods for refreshing. Specific embodiments of this system are described below.

[1831] System configuration

[1832] The system consists of a terminal that collects employees' work data and emotional data, a server that performs the analysis, and a terminal that sends notifications and re-collects the data.

[1833] 1. Data Collection

[1834] The server connects to each employee's device (PC, smartphone, wearable device) and collects work data and emotional data, including work hours, work content, work location, and stress level.

[1835] The device records real-time data such as keystroke speed, mouse movements, application usage history, heart rate, and electrodermal activity.

[1836] The device uses an emotion engine to perform voice analysis, facial recognition, and text analysis to collect user (employee) emotional data in real time, including emotional states such as anger, sadness, joy, and fear.

[1837] Users (employees) input their emotional state by self-evaluating in the form of a questionnaire at the start and end of work, and periodically (for example, every 1-2 hours).

[1838] 2. Data Analysis

[1839] The server checks the received business data and emotion data and stores them in a database.

[1840] The server performs preprocessing (e.g., data normalization, missing value imputation, and outlier removal).

[1841] The server then inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level. By incorporating emotional data into the analysis results, more accurate predictions can be made.

[1842] 3. Notifications and refresh suggestions

[1843] Based on the analysis results, the server determines the optimal break timing for each employee.

[1844] The server refers to the employee's hobby and preference database and determines the appropriate way to refresh themselves, suggesting, for example, drinking coffee or stretching to take a break from work.

[1845] The server transmits the determined break timing and refreshment method to the terminal.

[1846] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[1847] 4. Refresh execution and data collection

[1848] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[1849] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[1850] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[1851] The terminal collects performance data after the refresh (e.g., recovery of key input speed, reduction in error messages) and sends it to the server.

[1852] 5. Feedback and model improvement

[1853] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[1854] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis.

[1855] The server stores the evaluation results and analysis data in a database for use in subsequent analyses.

[1856] Specific examples

[1857] Example: Programmer Tanaka

[1858] The device records Tanaka's slowing keystroke speed and rising heart rate, and the emotion engine determines from Tanaka's voice and face that he is "highly stressed" or "feeling angry."

[1859] The server analyzed Tanaka's data and determined that he was in a "high stress state."

[1860] The server notifies Tanaka that he should take a 15-minute break in 10 minutes and also suggests that he should have a cup of coffee to refresh himself.

[1861] The user (Mr. Tanaka) checks the notification and takes a break as suggested.

[1862] The device recollects Tanaka's performance and emotional data after the break and sends it to the server.

[1863] The server updates the AI ​​model based on this data, preparing for the next more precise analysis.

[1864] In this way, the system collects and analyzes employees' work and emotional data, and notifies them of appropriate break times and ways to refresh themselves, thereby reducing employee stress and optimizing their work performance.

[1865] The processing flow will be explained below.

[1866] The specific processing steps of the system are described below.

[1867] Step 1:

[1868] The device collects and records employees' work data (keystroke speed, mouse movements, application usage history, heart rate, electrodermal activity, etc.) in real time.

[1869] Step 2:

[1870] The device uses an emotion engine to perform voice analysis, facial recognition, and text analysis to collect and record the user's (employee's) emotional data in real time. Emotional data includes emotional states such as anger, sadness, joy, and fear.

[1871] Step 3:

[1872] The device displays regular self-evaluations in the form of a questionnaire to the user (employee), prompting them to enter information about their work progress, current physical condition, and emotional state.

[1873] Step 4:

[1874] The terminal transmits the collected business data and emotional data to the server at regular intervals.

[1875] Step 5:

[1876] The server stores the received data in a database and performs preprocessing, which includes data normalization, missing value imputation, and outlier removal.

[1877] Step 6:

[1878] The server inputs the preprocessed data into an AI model to predict each employee's stress level and fatigue level. Emotional data is also incorporated into the analysis results to make more accurate predictions.

[1879] Step 7:

[1880] Based on the analysis results, the server determines the optimal break timing for each employee.

[1881] Step 8:

[1882] The server refers to the employee's hobby and preference database and determines the appropriate way to refresh themselves, suggesting, for example, drinking coffee or stretching to take a break from work.

[1883] Step 9:

[1884] The server transmits the determined break timing and refreshment method to the terminal.

[1885] Step 10:

[1886] The device will display notifications to employees, encouraging them to take a 15-minute break in 10 minutes, for example.

[1887] Step 11:

[1888] The user (employee) reviews the notification and selects an option such as "OK" or "Later."

[1889] Step 12:

[1890] If "later" is selected, the server sets the timing of the re-notification and transmits the break notification to the terminal again after a certain time.

[1891] Step 13:

[1892] The user performs the suggested refreshment method (e.g., drinking a cup of coffee, doing eye stretches).

[1893] Step 14:

[1894] After the device is refreshed, the user's performance data (e.g., recovery of keystroke speed, reduction in error messages) and emotional data are collected again.

[1895] Step 15:

[1896] The terminal transmits the refreshed data to the server.

[1897] Step 16:

[1898] The server evaluates the effectiveness of the refreshing method and analyzes the degree of improvement in stress and fatigue.

[1899] Step 17:

[1900] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis.

[1901] Step 18:

[1902] The server stores the evaluation results and analysis data in a database for subsequent analysis.

[1903] By executing the above processing steps, the system can analyze employees' work data and emotional data and suggest appropriate break times and ways to refresh themselves. This reduces employee stress and optimizes work performance. As a concrete example, the system makes real-time suggestions to programmer Tanaka about when to take a break and how to refresh himself. Even when Tanaka is busy with a project, he can refresh himself at the appropriate time and maintain work efficiency.

[1904] Example 2

[1905] 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."

[1906] In today's business environment, properly managing employee stress and fatigue and improving work efficiency are important issues. However, traditional methods have made it difficult to grasp the state of each employee in real time and suggest appropriate break times and ways to refresh. As a result, many companies are not managing their employees' stress levels properly, resulting in decreased work performance and increased health risks.

[1907] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting work data and emotion data, means for analyzing the collected work data and emotion data, and means for determining optimal break timing and refreshment methods based on the analysis results. This makes it possible to grasp employee stress and fatigue in real time and propose optimal break timing and refreshment methods to each employee.

[1908] "Business data" refers to information generated when employees perform their work, including work hours, work content, work location, keystroke speed, mouse movements, application program usage history, heart rate, etc.

[1909] "Emotional data" refers to information that indicates an employee's emotional state, including emotional states such as anger, sadness, joy, and fear obtained through voice analysis, facial recognition, and text analysis.

[1910] "Analysis results" refers to information obtained after analyzing collected data using AI models, etc., and predicts the stress levels and fatigue levels of each employee.

[1911] "Break timing" refers to the time of break recommended for employees based on the analysis results.

[1912] "Refreshment methods" refer to actions or methods that employees are encouraged to take during their breaks, such as drinking coffee or stretching.

[1913] "Notification means" refers to the means of informing employees of the determined break timing and method of refreshment, and notifications are given via PCs, smartphones, wearable devices, etc.

[1914] A "database" is a repository of information that stores collected data and analysis results and allows reuse as needed.

[1915] A "generative AI model" is a model that includes algorithms for analyzing data and making predictions using AI.

[1916] MODE FOR CARRYING OUT THE INVENTION

[1917] This invention is a system that uses AI to analyze employees' work data and emotional data and suggests appropriate break times and ways to refresh. The system consists of the following main components:

[1918] Data collection

[1919] The device connects

[1920] The server establishes a secure connection with each employee's device (PC, smartphone, wearable device, etc.), enabling real-time data collection. For example, this includes the PCs and smartwatches employees use to do their work.

[1921] Business data collection

[1922] The devices record real-time work data, such as keystroke speed, mouse movements, and application usage history, which is used to gain a detailed understanding of how employees are performing their work. For example, keystroke speed during programming tasks and the type of software they are using are recorded.

[1923] Collecting Emotional Data

[1924] The device uses an emotion engine that performs voice analysis, facial recognition, and text analysis to analyze the user's emotional state in real time and digitize it. For example, the device classifies the user's emotional state into categories such as "stress" or "joy" based on the voice during a call or facial expressions captured via a webcam.

[1925] Enter your self-assessment

[1926] Users enter their emotional state in the form of a self-assessment questionnaire at the start and end of work and at regular intervals. For example, every 1-2 hours, they answer a questionnaire in which they rate their "current level of fatigue" on a scale of 1-5.

[1927] Data analysis

[1928] Receiving and storing data

[1929] The server receives the collected work data and emotion data and stores it in a database. This allows for centralized management, including past data. For example, fluctuations in an employee's stress level over the past month can also be recorded.

[1930] Data Preprocessing

[1931] The server performs preprocessing such as data normalization, missing value imputation, and outlier removal. This process improves the quality of the data to obtain accurate analysis results. For example, incomplete data sets and extreme values ​​are removed.

[1932] Prediction by AI model

[1933] The server inputs the preprocessed data into a generative AI model to predict each employee's stress level and fatigue level. The generative AI model is an algorithm that learns from past data and predicts future conditions. For example, it predicts that "Programmer Tanaka" is currently in a high stress state.

[1934] Notifications and refresh suggestions

[1935] Deciding when to take a break

[1936] Based on the analysis results, the server determines the optimal break timing for each employee. For example, it may decide that "Tanaka-san" should take a 15-minute break in 10 minutes.

[1937] Suggestions for refreshing

[1938] The server refers to a database of employees' hobbies and preferences to determine appropriate ways to refresh themselves, such as "drinking coffee" or "doing eye stretching exercises."

[1939] Sending notifications

[1940] The server then sends the determined break timing and refreshment method to the terminal, which then displays a notification to the employee. For example, a notification such as "Let's have a cup of coffee and refresh yourself" will appear on Tanaka's PC.

[1941] Refresh execution and data collection

[1942] User Verification

[1943] The user checks the notification and selects an option such as "OK" or "later." For example, Tanaka selects "OK."

[1944] Snooze settings

[1945] If "Later" is selected, the server sets the timing of the re-notification and sends the break notification to the terminal again after a certain time. For example, the setting "Re-notify after 30 minutes" is set.

[1946] Execute the refresh method

[1947] The user performs the suggested refreshment method. For example, Tanaka drinks coffee.

[1948] Performance Data Collection

[1949] The device collects performance data after the refresh and sends it to the server. For example, Tanaka's keystroke speed after the refresh is recorded.

[1950] Feedback and model improvement

[1951] Evaluation of effectiveness

[1952] The server evaluates the effectiveness of the refreshment method and analyzes the degree of improvement in stress and fatigue. For example, it evaluates changes in Tanaka's stress level.

[1953] Retraining the AI ​​model

[1954] The server retrains the AI ​​model based on the evaluation results and updates the algorithm to improve the accuracy of the next analysis, ensuring that analysis is always based on the latest data, resulting in more accurate predictions.

[1955] Data storage

[1956] The server stores the evaluation results and analysis data in a database for subsequent analysis, which can be used for future data analysis and model improvement.

[1957] Prompt Sentence Examples

[1958] "Design a system that analyzes employees' work data and emotional data to suggest appropriate break times and ways to refresh. The following input data is provided: work hours, work content, work location, keystroke speed, mouse movement, heart rate, voice analysis data, and facial recognition data. Use this data to predict employees' stress levels and fatigue levels."

[1959] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1960] Step 1:

[1961] Data collection

[1962] The server establishes a connection: The server establishes a secure connection with each employee's device (PC, smartphone, wearable device), allowing data to be sent and received in real time.

[1963] Input: Connection request from employee's device.

[1964] Data processing / calculation: Establish a secure connection and perform authentication.

[1965] Output: Establishment of a secure communication channel with the device.

[1966] Specific operation: Tanaka's smartwatch connects to the server and starts sending data.

[1967] Devices collect business data: Devices collect business data in real time, such as keystroke speed, mouse movements, and application usage history.

[1968] Input: Employee operation data (keystrokes, mouse movements, etc.).

[1969] Data processing / calculation: Recording of operational data and converting it into the required format.

[1970] Output: Formatted business data.

[1971] Specific operation: Tanaka's keystroke speed and mouse movements are recorded while he works in Excel.

[1972] Device collects emotional data: The device collects the user's emotional data using voice analysis, facial recognition, and text analysis.

[1973] Input: Audio data, facial image data, text data.

[1974] Data processing / calculation: Emotional states are analyzed using an emotion engine and converted into data.

[1975] Output: Emotion data (e.g. anger, joy, sadness, etc.).

[1976] Specific operation: While Tanaka is on the phone, the emotion of "anger" is determined and recorded as data.

[1977] Users enter self-evaluations: Users answer self-evaluation questionnaires at the start and end of work, and at regular intervals.

[1978] Input: User's self-assessment survey responses.

[1979] Data processing / calculation: Formatting the survey responses for storage in the database.

[1980] Output: Formatted self-assessment data.

[1981] Specific actions: Tanaka inputs his "current level of fatigue" every hour on a 5-point scale.

[1982] Step 2:

[1983] Data analysis

[1984] The server receives and stores the data: The server stores the received business data and emotion data in a database.

[1985] Input: Business and sentiment data collected in real time.

[1986] Data processing / calculation: Convert into a format that can be saved in a database.

[1987] Output: Business and sentiment data stored in a database.

[1988] Specific action: Tanaka's data is registered in the database.

[1989] The server performs data preprocessing: The server normalizes the data, imputes missing values, removes outliers, etc.

[1990] Input: Stored business and sentiment data.

[1991] Data processing / calculation: data normalization, missing value imputation, outlier removal.

[1992] Output: Clean data after preprocessing.

[1993] Specific action: Tanaka's data is normalized and incomplete data is complemented.

[1994] The server makes predictions using an AI model: The preprocessed data is fed into a generative AI model to predict stress levels and fatigue levels.

[1995] Input: Preprocessed business data and sentiment data.

[1996] Data processing / calculation: Perform predictive analysis using generative AI models.

[1997] Output: Predicted stress and fatigue levels for each employee.

[1998] Specific action: Tanaka's data is input into the AI ​​model, and it is predicted that he is in a state of high stress.

[1999] Step 3:

[2000] Notifications and refresh suggestions

[2001] The server determines the timing of breaks: Based on the analysis results, the server determines the optimal timing of breaks for each employee.

[2002] Inputs: Forecast results and current business progress.

[2003] Data processing / calculation: Calculate the optimal break timing.

[2004] Output: Break timing decision result.

[2005] Specific action: Tanaka decides the timing, "I will take a 15-minute break in 10 minutes."

[2006] The server suggests a refreshing method: It refers to the hobby and preference database and suggests an appropriate refreshing method.

[2007] Input: Hobby and preference information for each employee and analysis results.

[2008] Data processing / calculation: Determine the appropriate refresh method.

[2009] Output: The proposed refresh method.

[2010] Specific action: A method for "drinking coffee" is suggested to Tanaka.

[2011] The server sends a notification: The determined break timing and refreshment method are sent to the terminal. The terminal displays the notification to the employee.

[2012] Input: Break timing and refreshment method decisions.

[2013] Data processing / calculation: generating and sending notification messages.

[2014] Output: Notifications that appear on the device.

[2015] Specific action: A message will appear on Tanaka's PC saying, "Take a 15-minute break in 10 minutes and have a cup of coffee."

[2016] Step 4:

[2017] Refresh execution and data collection

[2018] User acknowledges notification: Employee acknowledges notification and selects an option such as "Got it" or "Later."

[2019] Input: Notification message.

[2020] Data processing / calculation: Record the user's selection results.

[2021] Output: The user's selection.

[2022] Specific action: Tanaka selects "Got it."

[2023] Server will set snooze: If the user selects "Later", this sets the snooze timing. The server will snooze after a certain time.

[2024] Input: User's choice ("Later").

[2025] Data processing / calculation: Calculation and setting of re-notification timing.

[2026] Output: Snooze settings.

[2027] Specific operation: If Tanaka selects "Later", the server will schedule a reminder for 30 minutes.

[2028] User executes the refresh method: The user executes the proposed refresh method.

[2029] Input: Refresh suggestion for notification message.

[2030] Data processing / calculation: Record of refresh methods performed.

[2031] Output: State data after refresh.

[2032] Specific action: Tanaka drinks coffee.

[2033] The terminal collects performance data: The terminal collects performance data after the refresh and sends it to the server.

[2034] Input: User operation data and physiological data.

[2035] Data processing / calculation: recording and transmission of performance data.

[2036] Output: Performance data after refresh.

[2037] Specific action: After Tanaka's break, his input speed improves and the data is sent from his device to the server.

[2038] Step 5:

[2039] Feedback and model improvement

[2040] The server evaluates the effect: The effect of the refreshment method is evaluated and the degree of improvement in stress and fatigue is analyzed.

[2041] Input: Performance and emotion data after refresh.

[2042] Data processing / calculation: Calculation and evaluation of the degree of improvement.

[2043] Output: Evaluation results.

[2044] Specific action: Tanaka's change in stress level is evaluated.

[2045] The server retrains the AI ​​model: The AI ​​model is retrained based on the evaluation results to improve the accuracy of the next analysis.

[2046] Input: Assessment results and stored historical data.

[2047] Data processing / computation: Retraining AI models and updating algorithms.

[2048] Output: The updated AI model.

[2049] Specific action: The AI ​​model is updated based on new data.

[2050] The server stores the data: The evaluation results and analysis data are stored in a database for future analysis.

[2051] Input: Evaluation results and analysis data.

[2052] Data processing / calculation: Saving to database.

[2053] Output: Saved evaluation results and analysis data.

[2054] Specific actions: Updated evaluation results and analysis data are added to the database.

[2055] (Application example 2)

[2056] 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."

[2057] In today's work environment, properly collecting and analyzing employee work data and emotional data to suggest effective break timing and refreshment methods is important for reducing employee stress and improving productivity. Similarly, for robots operating in factories, collecting operational and sensor data and suggesting appropriate maintenance timing and methods is essential for maximizing robot operating efficiency and minimizing downtime. While systems exist to solve these issues individually, integrated systems are still lacking. Therefore, there is a need for the development of an integrated system that suggests optimal break and maintenance methods for both employees and robots.

[2058] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting work data from in-house devices, means for analyzing the collected work data and emotion data, means for determining optimal break timing and refreshment methods for employees based on the analysis results, means for notifying employees of the determined break timing and refreshment methods, means for collecting refreshment methods performed by employees and their results, means for updating the database based on the collected results and using them for the next analysis, means for collecting robot operation data and sensor data, means for analyzing the collected robot data and detecting signs of stress and fatigue, and means for proposing appropriate maintenance timing and methods based on the detected results. This makes it possible to reduce employee stress, improve productivity, maximize robot operating efficiency, and minimize downtime.

[2059] "Internal devices" are devices such as PCs, smartphones, and wearable devices that are used to collect employee business data and emotional data.

[2060] "Work data" refers to data including an employee's working hours, work content, work location, and stress level.

[2061] "Emotion data" is data that indicates emotional states such as anger, sadness, joy, and fear, obtained through voice analysis, facial recognition, and text analysis.

[2062] "Robot operation data" refers to data such as motor temperature, operating time, and error rate collected from sensors on robots operating in factories, etc.

[2063] "Sensor data" refers to real-time data collected by sensors mounted on the robot.

[2064] An "AI model" is an artificial intelligence algorithm used to analyze business data, emotional data, and robot operation data to predict stress and fatigue and suggest breaks and maintenance.

[2065] "Break timing" refers to the time that employees should take a break for work efficiency and health management.

[2066] "Refreshment methods" refer to actions and activities that employees should take to smoothly continue their work, such as drinking coffee or stretching.

[2067] "Maintenance timing" refers to the time when maintenance necessary to maintain the operating efficiency of a robot should be performed.

[2068] "Maintenance method" refers to the specific measures and actions that should be taken to keep the robot functioning optimally, such as oiling and replacing parts.

[2069] This invention includes a system that collects employee work data and emotional data, analyzes it in real time, and proposes optimal break timing and refreshment methods, and a system that analyzes the operation data of factory robots and proposes appropriate maintenance timing and methods. Specific embodiments of this invention are described below.

[2070] System configuration

[2071] This system consists of a means of collecting employee data, a server that analyzes the collected data, a means of issuing instructions based on the analysis results, and a means of collecting the results of refreshes and maintenance based on the instructions. Each means is explained in detail below.

[2072] Data collection from internal devices

[2073] The server connects to in-house devices such as PCs, smartphones, and wearable devices to collect employee work data. Data collected from these devices includes work hours, work content, work location, and stress level. Emotional data, such as anger, sadness, joy, and fear, is also collected in real time using voice analysis, facial recognition, and text analysis.

[2074] Data analysis

[2075] The server performs preprocessing on the collected work data and emotion data, such as normalizing it, filling in missing values, and removing outliers, and then inputs the data into an AI model to predict each employee's stress level and fatigue level, which can then individually suggest optimal break times and ways to refresh.

[2076] For factory robots, the server collects real-time data from each sensor and uses AI models to analyze signs of stress and fatigue based on data such as motor temperature, operating time, and error rate.

[2077] Notifications and refresh / maintenance suggestions

[2078] Based on the analysis results, the server notifies employees of the optimal timing for breaks and how to refresh themselves. For example, it may notify employees that they should take a 15-minute break in 10 minutes, along with a suggestion to refresh themselves with a cup of coffee. For factory robots, it suggests the appropriate timing and method for maintenance. For example, it may suggest that robots be stopped to cool down or that they need oil.

[2079] Collection of results and feedback

[2080] The refresh methods performed by employees and the results of factory robot maintenance are collected again and sent to the server, which updates the database based on these results and retrains the AI ​​model to improve the accuracy of the next analysis.

[2081] Specific examples

[2082] A concrete example is the program Tanaka. The device records Tanaka's slowing keystroke speed and rising heart rate, and judges her to be "highly stressed" based on her voice and facial expression. The server analyzes this data, recognizes her as being in a "high stress state," and notifies her to "take a 15-minute break in 10 minutes," and suggests that she "have a cup of coffee and refresh herself." Tanaka follows the suggestion and takes a break, and the device again collects performance data after the break and sends it to the server. The server then updates the AI ​​model based on this data.

[2083] For example, in the case of a factory robot, if the motor temperature rises too high while the robot is operating, a sensor will report this abnormality to the server, which will then send a message to the robot to stop it to cool down, and a maintenance technician will inject oil or take other cooling measures.

[2084] Prompt Sentence Examples

[2085] "Create an AI model that predicts the stress level of a robot and suggests appropriate maintenance."

[2086] "Write a Python program to analyze stress and fatigue levels using data from factory robot sensors."

[2087] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2088] Step 1: Data collection

[2089] The server collects employee work data and emotional data from the devices. Specifically, it obtains data such as work hours, work content, work location, stress level, heart rate, and electrodermal activity from employees' PCs, smartphones, and wearable devices. It also performs voice analysis, facial recognition, and text analysis to collect emotional data such as anger, sadness, joy, and fear in real time. The input is sensor data from each device, and the output is the collected work data and emotional data.

[2090] Step 2: Preprocessing

[2091] The server performs preprocessing on the collected data, such as normalizing, filling in missing values, and removing outliers. Specifically, it detects and removes outliers and fills in missing data. The input is the collected raw data, and the output is the preprocessed data.

[2092] Step 3: Data analysis

[2093] The server inputs the preprocessed data into an AI model to predict the stress level and fatigue level for each employee. Specifically, the data is input into the AI ​​model, and the predicted stress level and fatigue level are output. The input is the preprocessed data, and the output is the predicted stress level and fatigue level.

[2094] Step 4: Notification based on analysis results

[2095] Based on the analysis results, the server determines the optimal timing for breaks and ways to refresh, and notifies employees via their devices. Specifically, it sends notifications such as "Take a 15-minute break in 10 minutes" or "Refresh yourself with a cup of coffee." The input is the predicted stress level and fatigue level, and the output is notifications on break timing and ways to refresh.

[2096] Step 5: Employee execution and collection of results

[2097] The employee performs the refresh method, and the terminal re-collects the subsequent performance data. Specifically, it again records the recovery of keystroke speed, the reduction of error messages, etc. The input is the employee's performance data after the method is performed, and the output is the collected result data.

[2098] Step 6: Feedback and model improvement

[2099] The server evaluates the effectiveness of the refreshment method based on the collected result data and retrains the AI ​​model. Specifically, it analyzes the degree of improvement in stress and fatigue and updates the algorithm. The input is the result data, and the output is an improved AI model.

[2100] Step 7: Robot Data Collection

[2101] The server collects operational and sensor data from the robots in the factory, specifically motor temperature, operating time, error rate, etc. The input is the robot sensor data, and the output is the collected robot data.

[2102] Step 8: Robot data analysis

[2103] The server inputs the collected robot data into an AI model to predict signs of stress and fatigue. Specifically, it analyzes stress levels and fatigue levels based on the data. The input is the collected robot data, and the output is the predicted stress levels and fatigue levels.

[2104] Step 9: Maintenance Recommendations

[2105] The server then suggests appropriate maintenance timing and methods based on the analysis results. Specifically, it makes suggestions such as "Stop the vehicle to cool it down" or "Oil injection is required." The input is the predicted stress level and fatigue level, and the output is a notification of the maintenance method.

[2106] Step 10: Collect maintenance results

[2107] The server collects the results of maintenance performed by the maintenance technician and uses them for future analysis. Specifically, it records the robot performance data after maintenance. The input is the performance data after maintenance, and the output is the collected result data.

[2108] 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.

[2109] 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.

[2110] 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...

Claims

1. A means of collecting employee work data; A means for analyzing the collected business data; A means for determining the optimal break timing and refreshment method for employees based on the analysis results; A means of informing employees of the determined break timing and refreshment method; A means of collecting the refreshment methods implemented by employees and their results; Based on the collected results, we will update the database and use it for the next analysis. A system including:

2. 2. The system of claim 1, wherein the employee work data includes work hours, work content, work location, and stress level.

3. The system according to claim 1, wherein an AI model is used to analyze employee work data.

Images