System

Abstract

A system is provided.SOLUTION: A system comprising: means for obtaining a user's calendar information; means for collecting past meeting participation history; means for obtaining a user's job title and department information; means for applying a machine learning model to predict a user's tendency to participate in a meeting based on the collected data; means for displaying the predicted tendency to participate in a meeting score in a calendar application; means for making a notification to recommend a time slot with a low score when setting a meeting; and means for collecting user feedback and retraining the machine learning model.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] While the spread of remote work has led to an increase in online conferences, coordinating participants' free time requires mutual confirmation, which consumes time and effort for both the organizer and participants and reduces productivity. Another problem is that meetings with no intention of attending or regular meetings with few attendees appear on the calendar, causing confusion when scheduling meetings and delays in coordination. [Means for solving the problem]

[0005] To solve these problems, the present invention provides the following means. First, the present invention includes a means for acquiring a user's calendar information, a means for collecting past meeting attendance history, and a means for acquiring the user's job title and department information. Next, the present invention uses a means for applying a machine learning model to predict a user's tendency to attend meetings based on this data. Furthermore, the present invention includes a means for displaying the predicted meeting attendance propensity score in a calendar application and a means for notifying users when scheduling a meeting to recommend a time slot with a low score. Finally, the present invention incorporates a means for collecting user feedback and retraining the machine learning model to improve accuracy. The present invention also includes a means for visually displaying the predicted meeting attendance propensity score using color coding and icons, and a means for recommending time slots based on each user's attendance propensity.

[0006] This reduces the need to check each other's free time when scheduling a meeting, improving productivity. It also makes coordination more efficient by allowing optimal meeting scheduling based on the attendance trends of managers and the type of meeting.

[0007] "Calendar information" refers to schedule information that allows a user to manage appointments and meetings, and includes the title of a meeting, start time, end time, attendee list, frequency of the meeting, and the like.

[0008] "Meeting participation history" refers to a record of meetings that a user has attended in the past, and includes attendance history, invitation acceptance rate, web conference link click rate, etc.

[0009] "Position and department information" refers to information about the position and department to which the user belongs, and indicates the position and role of the user within the organization.

[0010] A "machine learning model" is an algorithm that learns from collected data and generates specific patterns and predictions.

[0011] "Meeting attendance propensity score" refers to numerical data calculated using a machine learning model that indicates the likelihood that a user will attend a particular meeting.

[0012] "Calendar application" refers to software with a particular interface that allows a user to manage their appointments and meetings.

[0013] "Notification" refers to messages or alerts sent to users to inform them about meeting setup or participation.

[0014] "Feedback" refers to ratings, opinions, and behavioral data collected from users, which is used to improve the system and retrain machine learning models.

[0015] "Display means" refers to methods and techniques for visually presenting information to the user, including screen display, color coding, icon display, etc. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

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

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] This invention relates to a system for achieving efficient time management for web conferences in a remote work environment. Specifically, it supports appropriate conference scheduling by predicting a user's conference participation tendency based on the user's calendar information, conference participation history, job title, and department information, and displaying the score in a calendar app.

[0038] The system mainly includes the following functions:

[0039] 1. Data Collection

[0040] The server obtains the user's calendar information, collects the user's past meeting participation history, and collects data such as attendance history, invitation acceptance rate, and web conference link click rate, as well as the user's job title and department information.

[0041] Examples:

[0042] User A has a "Weekly Meeting" scheduled for every Monday on his calendar, but he has only attended one of the past three meetings. This data is collected on the server. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[0043] 2. Data analysis and score calculation

[0044] The server applies a machine learning model to the collected data to predict each user's tendency to attend meetings. This prediction can be made using algorithms such as logistic regression or random forest. As a result, a participation propensity score is calculated for each user.

[0045] Examples:

[0046] User A's propensity score for attending "Weekly Meetings" is calculated to be 40%. User B's propensity score for attending important meetings is calculated to be 90%. This allows the server to quantify the user's probability of attending each meeting.

[0047] 3. Score display

[0048] The server displays this participation propensity score in the calendar application, using color coding and numerical values ​​to provide a visually easy-to-understand score.

[0049] Examples:

[0050] In the calendar app, entries for "Weekly Meetings" will show a "Participation Propensity Score of 40%," while entries for "Important Project Meetings" will show a 90%, making it easy to see at a glance which users are likely to attend.

[0051] 4. Meeting setup notification

[0052] The device (e.g., the user's PC or smartphone) will send a notification recommending avoiding time slots with low scores based on the predicted score when setting up a meeting. This notification will help with meeting setup and participant adjustments.

[0053] Examples:

[0054] When User C tries to schedule a new meeting, the calendar app recommends avoiding the "Weekly Meeting" time slot because it has a low score. This allows important meeting times to be adjusted to have a higher attendance rate.

[0055] 5. Collecting User Feedback

[0056] The server collects user feedback, including reasons for attending meetings and user impressions of the system, and uses this feedback to retrain the machine learning model to improve prediction accuracy.

[0057] Examples:

[0058] If a user provides feedback such as "this meeting was important to me, but I didn't attend because it had a low score," the model can be adjusted based on that feedback, allowing the system to provide a more accurate attendance propensity score.

[0059] These features enable efficient adjustment of meeting schedules in a remote work environment, which is expected to improve productivity. The system recommends time slots based on users' attendance trends and optimizes important meetings to increase attendance rates, thereby streamlining overall meeting management.

[0060] The processing flow will be explained below.

[0061] Step 1:

[0062] The server retrieves the user's calendar information, specifically, details of the user's appointments and meetings (title, start time, end time, attendee list, frequency, etc.) through the API.

[0063] Step 2:

[0064] The server collects past conference participation history. The collected data includes attendance history for web conferences, invitation acceptance rates, and web conference link click rates. This allows for an understanding of conference participation trends.

[0065] Step 3:

[0066] The server retrieves the user's job title and department information. The server retrieves the user's job title and department data from the HR system or database to clarify the user's role.

[0067] Step 4:

[0068] The server preprocesses the collected data by cleaning it, interpolating missing values, removing outliers, and making it suitable for machine learning models.

[0069] Step 5:

[0070] The server applies machine learning models to predict each user's tendency to attend meetings, using models such as logistic regression and random forests to calculate a participation propensity score based on data such as past attendance history and job title.

[0071] Step 6:

[0072] The server calculates the meeting attendance tendency score and displays it in the calendar app. Specifically, the score is displayed numerically and color-coded on each meeting entry, allowing users to understand it visually.

[0073] Step 7:

[0074] The device will notify you when a meeting is scheduled. When a user tries to schedule a new meeting, the device will send a notification recommending the best time based on the user's participation propensity score. This will help ensure that important members are able to attend.

[0075] Step 8:

[0076] The server collects feedback from users and stores the feedback provided by users to the system (e.g., opinions on the importance of the meeting and their willingness to attend) in a database.

[0077] Step 9:

[0078] The server retrains the machine learning model based on the collected feedback. By incorporating user feedback, the model's accuracy can be improved to provide a more accurate meeting attendance propensity score.

[0079] Through these steps, this system streamlines meeting scheduling in a remote work environment and improves user productivity.

[0080] Example 1

[0081] 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."

[0082] In remote work environments, productivity needs to be improved by streamlining user meeting schedules and optimizing meeting scheduling based on attendance trends. However, conventional systems struggle to adjust schedules while fully considering users' job titles and past meeting participation history. Furthermore, they do not recommend avoiding time slots with low attendance propensity scores when scheduling meetings, resulting in low attendance rates for important meetings. Additionally, they lacked a way to visually display meeting attendance propensity scores or a way to retrain machine learning models based on user feedback.

[0083] 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.

[0084] In this invention, the server includes means for acquiring user calendar information, means for collecting past conference participation history, means for acquiring user job title and department information, means for applying a machine learning model to predict the user's conference participation propensity based on the collected data, means for displaying the predicted conference participation propensity score in a calendar application, means for notifying the user to recommend a time slot with a low score when scheduling a conference, means for collecting user feedback and retraining the machine learning model, and means for reflecting the calculation results of the conference participation propensity score in actual conference scheduling and schedule management. This enables optimal schedule adjustment based on the user's conference participation propensity, which is expected to improve the attendance rate for important meetings and increase overall productivity.

[0085] "Calendar information" is digital schedule data including the user's plans, meeting dates and times, titles, participant information, and the like.

[0086] "Conference participation history" is data on conferences that a user has attended in the past, and includes information such as attendance history, invitation acceptance rate, and click rate of web conference links.

[0087] "Position and department information" is data relating to the position name and department name within the organization to which the user belongs.

[0088] A "machine learning model" is an algorithm or method used to analyze collected data and predict users' meeting participation trends.

[0089] A "meeting attendance propensity score" is a numerical value calculated using a machine learning model that indicates the probability that a user will attend a particular meeting.

[0090] A "calendar application" is software or a system that allows a user to manage their digital schedule and view appointments and meetings.

[0091] "Notification" is a means of sending alerts or messages to users to convey specific information.

[0092] "User feedback" refers to data on opinions and evaluations provided by users, such as reasons for attending or not attending a meeting and impressions of using the system.

[0093] "Retraining" is the process of improving the accuracy of a machine learning model based on new data collected.

[0094] "Visual display of scores" refers to a method of displaying the meeting attendance tendency scores in an easy-to-understand manner using color coding and icons.

[0095] This system is designed to streamline time management for web conferences in remote work environments. Specifically, the server collects users' calendar information, meeting participation history, job title, and department information, and then applies a machine learning model based on this information to predict the user's tendency to attend meetings. Furthermore, this participation tendency score is displayed in the calendar application, and when scheduling a meeting, a notification is sent recommending avoiding time slots with low scores, thereby helping to schedule appropriate meetings. Finally, user feedback is collected and the machine learning model is retrained to improve prediction accuracy. The specific implementation steps and the hardware and software used are described below.

[0096] Data collection

[0097] The server obtains the user's calendar information. Specifically, it uses the Google Calendar API or Microsoft Outlook Calendar API to collect each user's calendar entries along with the date, time, title, and participant information. The server also collects data such as past meeting participation history, attendance history, invitation acceptance rate, and web conference link click rate. It also obtains the user's job title and department information.

[0098] Examples:

[0099] User A has a "Weekly Meeting" scheduled for every Monday in his Google Calendar, but he has only attended one of the past three meetings. This data is collected on the server using the Google Calendar API. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[0100] Data analysis and score calculation

[0101] The server applies a machine learning model to the data collected. Specific algorithms include logistic regression and random forest, and Python libraries such as Scikit-learn and TensorFlow are used. This predicts each user's tendency to attend meetings and calculates a participation propensity score.

[0102] Examples:

[0103] User A's propensity score for attending "Weekly Meetings" is calculated to be 40%. User B's propensity score for attending important meetings is calculated to be 90%. This allows the server to quantify the user's probability of attending each meeting.

[0104] Score display

[0105] The server converts the calculated participation propensity score into a visually understandable format. Specifically, it generates code using HTML and JavaScript to display the score as a color code or a number. This score data is then sent to the calendar application for display.

[0106] Examples:

[0107] In a calendar app, the entry for User A's "Weekly Meeting" shows a "Participation Propensity Score of 40%." User B's "Important Project Meeting" shows a 90%, making it easy to see at a glance which users are likely to attend.

[0108] Meeting setup notifications

[0109] The device obtains the date and time and user list of the newly scheduled meeting, checks the existing attendance tendency score, and sends a notification recommending avoiding time slots with low scores. This allows meetings to be scheduled at appropriate time slots, improving the attendance rate of important meetings.

[0110] Examples:

[0111] When User C tries to schedule a new meeting, the calendar app recommends avoiding the "Weekly Meeting" time slot because it has a low score. This allows important meeting times to be adjusted to have a higher attendance rate.

[0112] Collecting user feedback

[0113] The server sends a request for feedback to users after the meeting or periodically. The server collects information from users about their reasons for attending or not attending meetings and their impressions of using the system using a web form or in-app feedback function. This feedback is used to retrain the machine learning model and improve the accuracy of predictions.

[0114] Examples:

[0115] If a user provides feedback such as "this meeting was important to me, but I didn't attend because it had a low score," the model can be adjusted based on that feedback, allowing the system to provide a more accurate attendance propensity score.

[0116] Prompt Sentence Examples

[0117] "Please look at User A's calendar and calculate and display this week's meeting attendance propensity score."

[0118] "When scheduling meetings, please remind me to avoid times with low attendance propensity scores."

[0119] This system makes it possible to efficiently adjust meeting schedules in a remote work environment, which is expected to improve productivity.

[0120] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0121] Step 1:

[0122] The server obtains the user's calendar information. To do this, it uses the Google Calendar API or the Microsoft Outlook Calendar API. The input is the user's access privileges to the calendar API, and the output is calendar entry data including the date, time, title, and attendee information of the user's appointments and meetings. This calendar entry data is then saved on the server.

[0123] Specific behavior:

[0124] The server sends an API request to retrieve the user's calendar information.

[0125] The server retrieves the calendar entries and stores them in a database.

[0126] Step 2:

[0127] The server collects past conference participation history. The input is calendar entries and participation history data, and the output is historical data including each user's attendance history, invitation acceptance rate, and web conference link click rate. This historical data is organized and stored on the server.

[0128] Specific behavior:

[0129] The server extracts past meeting data from the calendar entries.

[0130] The server calculates the user's attendance history, invitation acceptance rate, and web conference link click rate, and stores them in a database.

[0131] Step 3:

[0132] The server retrieves the user's job title and department information. The input is a database of job titles and departments within the company, and the output is the job title and department information for each user. This information is then stored in the integrated database.

[0133] Specific behavior:

[0134] The server accesses a personnel database within the company to obtain the user's job title and department information.

[0135] The server stores the acquired job title and department information in a database.

[0136] Step 4:

[0137] The server applies a machine learning model to the collected data. The input is calendar information, meeting attendance history, job title, and department information, and the output is the user's meeting attendance propensity score. Python's Scikit-learn and TensorFlow are used for data processing and learning.

[0138] Specific behavior:

[0139] The server extracts the necessary data from the database and performs preprocessing.

[0140] The server inputs data into the machine learning model, performs learning, and calculates the participation propensity score.

[0141] The server stores the calculated score in a database.

[0142] Step 5:

[0143] The server displays the predicted meeting attendance propensity score in a calendar application. The input is the attendance propensity score, and the output is a visually color-coded and / or numerically displayed score.

[0144] Specific behavior:

[0145] The server sends a request to the calendar application to display the score.

[0146] The calendar application displays the score in color.

[0147] Step 6:

[0148] The terminal sends a notification to the user recommending avoiding low-score time slots when setting up a meeting. The input is the date and time of the new meeting and the attendance tendency score, and the output is a notification to the user.

[0149] Specific behavior:

[0150] The terminal obtains the setting information for the new conference.

[0151] The device will check your existing participation propensity score and send you notifications recommending you avoid low-scoring times.

[0152] Step 7:

[0153] The server collects and analyzes user feedback, with the input being the user feedback data and the output being the analyzed feedback data, and then retrains the machine learning model based on this.

[0154] Specific behavior:

[0155] The server sends notifications to users to collect feedback.

[0156] The server collects feedback data from users and stores it in a database.

[0157] The server analyzes the feedback data and retrains the machine learning model.

[0158] (Application example 1)

[0159] 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."

[0160] In work environments such as remote work and factories, meeting time management is important, and there is a particular need to improve meeting attendance rates. However, it is difficult to predict the schedules and attendance tendencies of individual users and workers and then set optimal meeting times based on these. Furthermore, conventional systems have difficulty using user feedback to improve accuracy, leaving room for improvement. Therefore, there is a need for a system that can improve meeting attendance rates and optimize work efficiency.

[0161] 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.

[0162] In this invention, the server includes means for acquiring user schedule information, means for collecting past conference participation history, means for acquiring user job title and department information, means for applying a generative AI model to predict a user's conference participation tendency based on the collected data, means for displaying the predicted conference participation tendency score in a schedule management application, means for notifying users to avoid time periods with low scores when scheduling a conference, means for collecting user feedback and retraining the generative AI model, and means for supporting the efficiency of industrial machine maintenance and production planning meetings. This makes it possible to improve meeting attendance rates and optimize overall business efficiency.

[0163] "User schedule information" refers to data such as the date, time, location, and content of appointments and meetings set by the user.

[0164] "Conference participation history" refers to historical data of conferences that a user has attended in the past, and includes information such as whether or not the user attended, the frequency of attendance, and the attendance rate.

[0165] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to predict user behavior patterns and trends from data.

[0166] A "schedule management application" refers to a software application that helps users manage their schedules and meetings.

[0167] The "conference attendance propensity score" is a numerical representation of the likelihood that a user will attend a particular conference, and is calculated based on a prediction.

[0168] "Notification mechanism" refers to a digital messaging or alert system that notifies users of specific information or recommendations.

[0169] "Feedback" refers to opinions, impressions, and evaluations collected from users regarding their use of the system.

[0170] "Industrial machinery" refers to various machinery and equipment used in manufacturing and production industries, including those that require maintenance and management.

[0171] "Maintenance" refers to repairs and maintenance work carried out to keep industrial machinery in working order.

[0172] A "manufacturing planning meeting" refers to a meeting held to improve the efficiency of the production line and adjust schedules.

[0173] This invention is a system for streamlining time management in remote work environments, industrial machinery maintenance meetings, production planning meetings, etc. Specifically, it predicts a user's tendency to attend meetings based on the user's schedule information, meeting participation history, job title, and department information, and displays this in a schedule management application.

[0174] Program Description

[0175] The server collects users' schedule information, past meeting attendance history, job title, and department information. This data is used to apply a generative AI model to predict a meeting attendance propensity score. The score is displayed in the schedule management application, and notifications are sent when scheduling meetings to avoid low-scoring times. Furthermore, feedback from users can be collected to retrain the generative AI model and improve prediction accuracy. This can improve meeting attendance rates and optimize overall business efficiency.

[0176] Hardware and software used

[0177] 1. Hardware:

[0178] Factory Server

[0179] User's Digital Dashboard

[0180] 2. Software:

[0181] Python

[0182] Pandas (data processing library)

[0183] Scikit-learn (machine learning library)

[0184] The server retrieves users' schedule information, meeting attendance history, job title, and department information using a Python script that reads CSV files. It then applies a machine learning algorithm (here, a logistic regression model) to analyze the collected data. Using Python and Scikit-learn, the model is trained and a meeting attendance propensity score is calculated for each user.

[0185] The scores are then displayed on a digital dashboard using Pandas and related visualization libraries. The scores are visually represented using color coding and icons, providing a user-friendly format. The server also uses the scores to create meeting scheduling recommendations and sends them to the user's device. User feedback is also collected and used to retrain the generative AI model.

[0186] As a concrete example of data, the shift information and meeting attendance history of factory workers are shown below:

[0187] Shift Information:

[0188] employee_id, shift_hours, past_meeting_attendance

[0189] 1, 8, 0.75

[0190] 2, 6, 0.90

[0191] Meeting participation history:

[0192] employee_id, attendance_probability

[0193] 1, 0.80

[0194] 2, 0.95

[0195] Prompt Sentence Examples

[0196] For the generative AI model, enter the following prompt:

[0197] Please generate Python code that calculates the meeting attendance tendency score based on shift information and meeting attendance history, and sends notifications at specific times. The example data is as follows.

[0198] Shift Information:

[0199] employee_id, shift_hours, past_meeting_attendance

[0200] 1, 8, 0.75

[0201] 2, 6, 0.90

[0202] Meeting participation history:

[0203] employee_id, attendance_probability

[0204] 1, 0.80

[0205] 2, 0.95

[0206] The above configuration can optimize users' meeting schedules and significantly improve work efficiency in remote work and industrial environments.

[0207] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0208] Step 1: Data collection

[0209] The server reads user schedule information, past meeting attendance history, job title, and department information from CSV files. Specifically, it uses Python's Pandas library to read and integrate this data as a data frame. The input includes each user's shift information (employee_id, shift_hours, past_meeting_attendance) and meeting attendance history (employee_id, attendance_probability). The output is a dataset that integrates this data.

[0210] Step 2: Data analysis and score calculation

[0211] The server applies a generative AI model (logistic regression algorithm) to the collected data to predict each user's meeting attendance propensity score. The input includes shift information and meeting attendance history, and the output includes each user's meeting attendance propensity score. This score represents the probability of attendance, expressed as a number ranging from 0 to 1.

[0212] Step 3: View the score

[0213] The server visualizes the data to display the predicted meeting attendance propensity scores in the schedule management application. Specifically, the scores are visually displayed on a digital dashboard using color coding and icons. The input includes the meeting attendance propensity scores for each user, and the output includes the visually displayed scores.

[0214] Step 4: Notification of meeting settings

[0215] The server creates a recommendation notification to avoid low-scoring time slots when scheduling a meeting and sends it to the user's device. The input includes the predicted meeting attendance propensity score, and the output is the recommendation notification message. For example, it generates a message like "Employee ID: 1 should be scheduled for a meeting outside their current shift hours."

[0216] Step 5: Gather user feedback and retrain

[0217] The server collects user feedback after the meeting. The input includes user opinions and usage impressions, and the output is feedback data. This feedback is used to retrain the generative AI model to improve prediction accuracy. Specifically, the newly collected feedback data is added to the existing dataset, and the machine learning model is retrained.

[0218] Through the above processing steps, a system is realized that aims to improve meeting attendance rates and optimize business efficiency.

[0219] 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.

[0220] This invention relates to a system for streamlining the scheduling of web conferences in a remote work environment and improving productivity. In particular, by combining it with an emotion engine that recognizes user emotions, it more accurately predicts conference participation trends and supports optimal conference scheduling.

[0221] The system mainly includes the following functions:

[0222] 1. Data Collection

[0223] The server retrieves the user's calendar information via an API. It also collects the user's past meeting participation history and obtains data such as attendance history, invitation acceptance rate, and web conference link click rate. In addition, it retrieves the user's job title and department information from the HR system or database.

[0224] Examples:

[0225] User A has a "Weekly Meeting" scheduled for every Monday on his calendar, but he has only attended one of the past three meetings. This data is collected by the server. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[0226] 2. Emotional Data Collection

[0227] The device uses an emotion engine that recognizes the user's emotions in real time, acquiring emotion data and understanding the emotional state (e.g., stress, satisfaction, dissatisfaction) that influences participation tendencies.

[0228] Examples:

[0229] When User C attended a particular meeting, the emotion engine recognized high stress. This information is sent to the server and used to predict meeting attendance trends.

[0230] 3. Data analysis and score calculation

[0231] The server applies a machine learning model based on the collected calendar information, meeting participation history, job title and department information, and emotion data to predict each user's tendency to participate in meetings, thereby calculating a participation propensity score.

[0232] Examples:

[0233] User A's propensity score for attending the "Weekly Meeting" is calculated as 40% based on their emotional data and past history. This prediction takes into account their past participation frequency and emotional state.

[0234] 4. Score display

[0235] The server displays the calculated meeting attendance tendency score in the calendar app. The score is provided in a visually understandable format using color coding and numerical values. The server also displays the user's emotional state, allowing the user to share their state.

[0236] Examples:

[0237] In the calendar app, the entry for "Weekly Meeting" will show "Participation Propensity Score 40%" along with the stress level.

[0238] 5. Meeting setup notifications

[0239] The device provides notifications when setting up a meeting. When a user tries to set up a new meeting, the device sends a notification recommending the best time slot based on the participation propensity score and emotion data.

[0240] Examples:

[0241] If User C tries to schedule a new meeting, the calendar app will show that the "Weekly Meeting" time slot has a low score and a high stress level, and recommend that they avoid that time.

[0242] 6. Collecting User Feedback

[0243] The server collects feedback from users, including reasons for attending meetings, impressions of using the system, and opinions based on sentiment data.

[0244] Examples:

[0245] If a user provides feedback such as "This meeting was important, but I was too stressed and didn't attend," the model is retrained based on that feedback.

[0246] 7. Retraining the model

[0247] The server retrains the machine learning model based on the collected feedback, allowing the system to calculate a more accurate meeting attendance propensity score.

[0248] Through these functions, the system streamlines the adjustment of meeting schedules in a remote work environment, improving productivity. By taking into account the user's emotional state, it is possible to set up meetings that are easier to participate in, further improving the efficiency of overall meeting management.

[0249] The processing flow will be explained below.

[0250] Step 1:

[0251] The server obtains the user's calendar information. Specifically, it collects detailed information about the user's schedules and meetings (title, start time, end time, attendee list, frequency, etc.) through the API. This allows it to understand what meetings the user is participating in.

[0252] Step 2:

[0253] The server collects past conference participation history, including attendance history, invitation acceptance rates, and click rates for web conference links, allowing for analysis of participation trends for each conference.

[0254] Step 3:

[0255] The server retrieves the user's job title and department information. The server retrieves data about the user's job title and department from the HR system or database to clarify the user's role.

[0256] Step 4:

[0257] The device activates an emotion engine to recognize the user's emotions. The emotion engine acquires real-time emotion data from the user using technologies such as voice recognition and facial expression analysis.

[0258] Step 5:

[0259] The server receives the emotion data sent from the emotion engine and stores it in a database, where the emotional state (e.g., stress, satisfaction, dissatisfaction) is recorded.

[0260] Step 6:

[0261] The server preprocesses all data collected (calendar information, meeting attendance history, job title / department information, and emotion data) by cleaning the data, interpolating missing values, and removing outliers, and converting it into a format suitable for machine learning models.

[0262] Step 7:

[0263] The server applies machine learning models to predict each user's tendency to attend meetings, using models such as logistic regression and random forests, and calculates a participation propensity score that takes into account past participation history and emotional data.

[0264] Step 8:

[0265] The server calculates the meeting attendance tendency score and displays it in the calendar app. The score is presented in a visually easy-to-understand format using color coding and numerical values. The user's emotional state can also be displayed.

[0266] Step 9:

[0267] The device will notify users when a meeting is scheduled. When a user tries to schedule a new meeting, the device will send a notification recommending the best time based on attendance propensity scores and sentiment data. This will help ensure that important members are able to attend.

[0268] Step 10:

[0269] The server collects feedback from users, who provide reasons for attending meetings, their impressions of using the system, and opinions based on emotional data, and stores the feedback in a database.

[0270] Step 11:

[0271] The server retrains the machine learning model based on the collected feedback. By incorporating user feedback, the model's accuracy can be improved to provide a more accurate meeting attendance propensity score.

[0272] Through these steps, the system streamlines the scheduling of meetings in a remote work environment, improving user productivity. By taking emotional states into account, it becomes possible to set up meetings that are easier to participate in, further improving the efficiency of overall meeting management.

[0273] Example 2

[0274] 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."

[0275] In a remote work environment, conventional systems adjust the schedule of web conferences without taking into account the emotional state of the user, which has led to problems such as a decrease in conference participation rates and an increase in user stress. There is a need to solve these problems and realize optimal conference settings that reflect the emotional state of the user.

[0276] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0277] In this invention, the server includes means for acquiring user schedule information, means for collecting past meeting attendance history, means for acquiring user job title and department information, means for recognizing emotional states in real time and collecting data, means for predicting the user's tendency to attend meetings by applying a machine learning model based on the collected calendar information, meeting attendance history, job title and department information, and emotional data, means for displaying the predicted meeting attendance propensity score in a calendar application, means for notifying the user of time periods with low scores when scheduling a meeting, and means for collecting user feedback and retraining the machine learning model. This makes it possible to adjust an optimal meeting schedule taking into account the user's emotional state.

[0278] "User schedule information" refers to information about schedules and events that a user has registered in a calendar or schedule management application.

[0279] "Past meeting attendance history" refers to data such as attendance status of meetings a user has previously attended, invitation acceptance rate, and click rate of web conference links.

[0280] "User position and department information" is information about the user's job title and the department to which the user belongs, and is obtained from an HR system or database.

[0281] "Means for recognizing emotional states and collecting data in real time" refers to technology that uses built-in cameras and microphones to capture the user's facial expressions and tone of voice, and then analyzes them using an emotion engine.

[0282] "Emotion data" is data that represents the user's emotional state, such as stress, satisfaction, or dissatisfaction.

[0283] "Machine learning model" refers to an algorithm or statistical model used to predict users' meeting participation trends based on collected data.

[0284] The "meeting attendance propensity score" is a numerical value calculated based on a machine learning model that indicates the likelihood that a user will attend a particular meeting.

[0285] A "calendar application" is software that allows users to manage their schedules and has a function to display meeting attendance propensity scores.

[0286] "Means of notification" refers to technical means for notifying users of information or recommendations in real time, such as push notifications or alerts.

[0287] "Means for collecting feedback" refers to a function for collecting opinions and impressions from users and using them to improve the system.

[0288] "Retraining" is the process of updating a machine learning model based on collected feedback to improve its prediction accuracy.

[0289] This invention relates to a system that streamlines scheduling of web conferences in a remote work environment and improves productivity. In particular, by combining it with an emotion engine that recognizes user emotions, it more accurately predicts conference participation trends and supports optimal conference scheduling.

[0290] Data collection

[0291] First, the server obtains the user's schedule information. The server collects data from schedule management applications such as Google Calendar and Outlook Calendar via API. Next, the server accesses the HR system or internal database to obtain the user's job title and department information. At the same time, it collects the user's past meeting attendance history and stores it in the database.

[0292] For example, User A has a "Weekly Meeting" scheduled on his calendar, but he has only attended one of the past three meetings. This data is collected by the server. User B is a manager and tends to invite his subordinates to many meetings.

[0293] Emotional Data Collection

[0294] The device uses the built-in camera and microphone to recognize the user's emotional state in real time. It uses an emotion engine (e.g., Microsoft Azure's Emotion API) to capture the user's facial expressions and tone of voice and analyzes the emotional data, such as stress, satisfaction, and dissatisfaction. This information is sent to a server and stored in a database.

[0295] For example, when user C is attending a particular meeting, the emotion engine recognizes high stress and sends this information to the server.

[0296] Data analysis and score calculation

[0297] The server applies a machine learning model (e.g., scikit-learn or TensorFlow) based on the collected calendar information, meeting participation history, job title / department information, and emotion data to predict each user's tendency to participate in meetings. This calculates a participation propensity score, which is then stored in a database.

[0298] For example, the participation propensity score for user A in the "Weekly Meeting" is calculated as 40% based on emotional data and past history. This prediction takes into account the frequency of past participation and emotional state.

[0299] Score display

[0300] The server displays the calculated meeting attendance tendency score in a calendar application, which displays the score in a visually easy-to-understand format.

[0301] For example, an entry for "Weekly Meeting" will display "Participation Propensity Score 40%" along with the emotional state.

[0302] Meeting setup notifications

[0303] When a user attempts to schedule a new meeting, the device receives the score and emotion data from the server and generates a notification suggesting the optimal meeting time and structure, which is displayed to the user as a push notification or alert.

[0304] For example, if User C tries to schedule a new meeting, the calendar application may indicate that the "Weekly Meeting" time slot has a low score, indicating a high stress state, and recommend that they avoid that time.

[0305] Gathering user feedback and retraining the model

[0306] The server collects user opinions and feedback through the calendar application and a dedicated feedback form. This feedback is stored in a database and analyzed. Based on the collected feedback, the server retrains the machine learning model to improve prediction accuracy.

[0307] For example, if a user provides feedback such as "This meeting was important, but I was too stressed and didn't attend," the server uses this data to retrain the model.

[0308] Prompt Sentence Examples

[0309] "Calculate the attendance propensity score for the next weekly meeting based on the user's sentiment data and meeting attendance history."

[0310] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0311] Step 1:

[0312] Input: User's calendar information (Google Calendar, Outlook, etc.), HR system or database information (job title, department information), past meeting attendance history (attendance status, invitation acceptance rate, etc.)

[0313] Specific operation: The server collects the user's schedule information from Google Calendar or Outlook Calendar via API. The server then accesses the HR system or internal database to obtain the user's job title and department information. It also obtains the user's past meeting attendance history and stores it in the database.

[0314] Output: The retrieved calendar information, job title and department information, and meeting attendance history are saved in a database.

[0315] Step 2:

[0316] Input: User's facial expression and voice data

[0317] How it works: The device uses the built-in camera and microphone to capture the user's facial expressions and tone of voice in real time. An emotion engine (e.g., Microsoft Azure's Emotion API) is used to analyze the user's emotional state (stress, satisfaction, dissatisfaction, etc.) from the captured data. This information is then sent to a server.

[0318] Output: The analyzed emotion data is sent to the server and stored in a database.

[0319] Step 3:

[0320] Input: User's calendar information, job title and department information, past meeting attendance history, emotional data

[0321] Specific operation: The server integrates the collected data and uses a machine learning model (e.g., scikit-learn or TensorFlow) to predict users' meeting participation trends. Specifically, it analyzes each user's meeting participation patterns based on past data and calculates the participation propensity score for each meeting.

[0322] Output: The calculated meeting attendance propensity score is saved in the database.

[0323] Step 4:

[0324] Input: Meeting participation propensity score

[0325] Specific operation: The server sends the meeting attendance tendency score to the calendar application UI. The calendar application displays the received score using color coding and numerical values, providing the user with a visually easy-to-understand format.

[0326] Output: The meeting attendance propensity score is displayed in the user's calendar application.

[0327] Step 5:

[0328] Input: Meeting participation tendency score and emotion data from the server

[0329] Specific operation: When a user attempts to schedule a new meeting, the device receives the meeting participation propensity score and emotion data from the server. The device generates and displays a notification to the user suggesting the optimal meeting time and configuration.

[0330] Output: The user will be notified with a suggestion for the best meeting time.

[0331] Step 6:

[0332] Input: User feedback (opinions about meeting attendance, impressions on using the system, opinions based on emotional data)

[0333] What it does: The server collects user feedback through the calendar application and a dedicated feedback form, stores the collected feedback in a database, and formats it for analysis.

[0334] Output: The collected feedback is stored in a database.

[0335] Step 7:

[0336] Input: Collected feedback data

[0337] What happens: The server takes in new feedback data and combines it with existing training data to retrain the machine learning model and generate a more accurate meeting attendance propensity score.

[0338] Output: The updated machine learning model is used for the next prediction.

[0339] (Application example 2)

[0340] 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."

[0341] Improving the efficiency of store staff work methods and the working environment are important issues in today's brick-and-mortar stores. In particular, the impact of staff emotional states on work efficiency and customer service cannot be ignored. However, there is currently no way to understand in real time how staff are working, making it difficult to create optimal work schedules. Furthermore, there is no established method for appropriately avoiding work that is likely to overload staff. Furthermore, there is no efficient way to collect staff feedback and optimize work based on it. This increases staff stress, ultimately leading to a decrease in customer satisfaction.

[0342] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0343] In this invention, the server includes: means for acquiring a user's calendar information; means for collecting past meeting participation history; means for acquiring the user's job title and department information; means for acquiring emotional data and recognizing the user's state; means for applying a machine learning model to predict the user's tendency to attend meetings based on the collected data; means for displaying the predicted meeting participation propensity score in a calendar application; means for notifying the user to recommend a time slot with a low score when scheduling a meeting; means for collecting user feedback and retraining the machine learning model; and means for visually displaying the optimized work schedule and the impact of the emotional data. This makes it possible to grasp the current emotional state of store staff in real time and provide an optimal work schedule that takes their emotional state into account. Furthermore, by efficiently collecting staff feedback and continuously optimizing work based on it, an improved working environment and increased customer satisfaction can be achieved.

[0344] "User" refers to a person who uses the system.

[0345] "Calendar information" refers to information that records a user's schedule and plans.

[0346] "Conference participation history" refers to a record of conferences that a user has participated in in the past.

[0347] "Job title and department information" refers to information about the department and job to which the user belongs.

[0348] "Collected data" refers to calendar information, meeting participation history, job title and department information, emotional data, etc.

[0349] A "machine learning model" is an algorithm that learns patterns based on collected data and makes predictions and analyses.

[0350] A "calendar application" is software for managing a user's schedule.

[0351] "Emotional Data" refers to data that records and analyzes the emotional state of a user.

[0352] "User state" refers to the user's current emotional and psychological state.

[0353] "Work schedule" refers to the planned work to be performed by a user.

[0354] The "predicted conference attendance propensity score" is a score calculated by a machine learning model that indicates the likelihood of a user attending a conference.

[0355] "Notification" is a means of conveying information to the user through an interface.

[0356] "Feedback" refers to opinions and evaluations from users of the system.

[0357] An "optimized work schedule" refers to a work schedule that is adjusted taking into account the user's emotional state and work efficiency.

[0358] "Visual display means" refers to methods of visually presenting information to users using graphs, icons, color coding, etc.

[0359] This invention aims to apply a meeting schedule adjustment system in a remote work environment to optimizing work schedules in brick-and-mortar stores. The system utilizes user emotional data to grasp the real-time emotional state of staff and propose optimal work schedules.

[0360] The server first obtains the user's calendar information, including the user's past conference attendance history and current schedule. It also obtains the user's job title and department information to understand their work characteristics.

[0361] The device then collects the user's real-time emotional data, which is acquired through smart glasses or other wearable devices worn by the user, and uses this emotional data to measure the user's current stress level, satisfaction, dissatisfaction, etc.

[0362] The server applies machine learning models to the collected data to predict each user's work participation trends. Specifically, it integrates and analyzes past meeting history, job title and department information, and real-time emotion data. This analysis utilizes an emotion engine and schedule optimization algorithm to calculate the optimal schedule to maximize the user's work efficiency.

[0363] The optimized work schedule is displayed on the user's smart glasses or smartphone. The schedule is color-coded and displayed with icons for easy visual understanding. The impact of emotional data is also displayed visually, allowing staff to work while being aware of their own state.

[0364] Furthermore, the server collects user feedback and retrains the machine learning model based on it. The feedback includes opinions based on the results of applying the work schedule and emotional data. This allows the system to continuously improve and provide more accurate work schedules.

[0365] For example, if a staff member gives feedback that they felt high stress during their last work session, the system will use that information to adjust their next work schedule. Specifically, if the emotion engine detects that a staff member's current stress level is high, the system will prioritize relaxing tasks.

[0366] Below is an example of a prompt sentence.

[0367] Example prompt sentence:

[0368] “Sort the following tasks based on the optimal work schedule. Consider the emotional state of your staff and create a schedule that is least stressful.

[0369] Product display

[0370] Customer Service

[0371] Stock Check

[0372] Sales Report

[0373] User's current mild stress level is 8."

[0374] In this way, the present invention can be used in brick-and-mortar stores to improve the working environment for staff and increase work efficiency.

[0375] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0376] Step 1:

[0377] The server retrieves the user's calendar information.

[0378] Input: User ID

[0379] Output: User's calendar information (schedule and events)

[0380] What it does: The server retrieves the user's calendar information through an API request, providing data that includes the user's events and past meeting history.

[0381] Step 2:

[0382] The server retrieves the user's job title and department information.

[0383] Input: User ID

[0384] Output: User's job title and department information

[0385] What happens: The server sends an API request to an HR system or database to retrieve information about the user's job title and department.

[0386] Step 3:

[0387] The device collects the user's emotional data in real time.

[0388] Input: User ID, Emotion Engine

[0389] Output: User's emotional data (stress level, satisfaction, dissatisfaction, etc.)

[0390] Specific operation: The smart glasses or wearable device worn by the user uses an emotion engine to collect emotion data in real time and transmits the data to a server.

[0391] Step 4:

[0392] The server applies a machine learning model based on the collected data to predict the user's work participation tendencies.

[0393] Input: Calendar information, job title and department information, emotion data

[0394] Output: Work participation propensity score

[0395] Specific operation: The server integrates these data into a single dataset and applies machine learning algorithms (e.g., random forests, neural networks) to predict the user's work participation tendencies.

[0396] Step 5:

[0397] The server displays the predicted work attendance propensity score in a calendar application.

[0398] Input: Work participation propensity score

[0399] Output: Work participation propensity score displayed as a color or icon

[0400] Specific operation: The server incorporates the score into the calendar application and displays the results on the user's device in a visually easy-to-understand format (color coding and icons).

[0401] Step 6:

[0402] The device will notify you to recommend a time slot with a low score when setting up a meeting.

[0403] Input: Work participation propensity score

[0404] Output: Best time suggestion

[0405] Specific operation: The device sends a notification to the user, recommending that the user avoid times when the work attendance propensity score is low.

[0406] Step 7:

[0407] The server collects user feedback and retrains the machine learning model.

[0408] Input: User feedback

[0409] Output: An improved machine learning model

[0410] How it works: The server collects user feedback (e.g., "This time of day is stressful") and uses it to retrain the machine learning model to improve accuracy.

[0411] Step 8:

[0412] The server visually displays the optimized work schedule and the impact of emotional data.

[0413] Input: Optimized work schedule, emotional data

[0414] Output: Visually displayed work schedule and the impact of emotional data

[0415] Specific operation: The server visually displays the optimized schedule and emotion data in a calendar application or smart glasses, allowing users to work while being aware of their own state.

[0416] 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.

[0417] 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.

[0418] 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.

[0419] [Second embodiment]

[0420] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0421] 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.

[0422] 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).

[0423] 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.

[0424] 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.

[0425] 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).

[0426] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0427] 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.

[0428] 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.

[0429] 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.

[0430] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0431] 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."

[0432] This invention relates to a system for achieving efficient time management for web conferences in a remote work environment. Specifically, it supports appropriate conference scheduling by predicting a user's conference participation tendency based on the user's calendar information, conference participation history, job title, and department information, and displaying the score in a calendar app.

[0433] The system mainly includes the following functions:

[0434] 1. Data Collection

[0435] The server obtains the user's calendar information, collects the user's past meeting participation history, and collects data such as attendance history, invitation acceptance rate, and web conference link click rate, as well as the user's job title and department information.

[0436] Examples:

[0437] User A has a "Weekly Meeting" scheduled for every Monday on his calendar, but he has only attended one of the past three meetings. This data is collected on the server. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[0438] 2. Data analysis and score calculation

[0439] The server applies a machine learning model to the collected data to predict each user's tendency to attend meetings. This prediction can be made using algorithms such as logistic regression or random forest. As a result, a participation propensity score is calculated for each user.

[0440] Examples:

[0441] User A's propensity score for attending "Weekly Meetings" is calculated to be 40%. User B's propensity score for attending important meetings is calculated to be 90%. This allows the server to quantify the user's probability of attending each meeting.

[0442] 3. Score display

[0443] The server displays this participation propensity score in the calendar application, using color coding and numerical values ​​to provide a visually easy-to-understand score.

[0444] Examples:

[0445] In the calendar app, entries for "Weekly Meetings" will show a "Participation Propensity Score of 40%," while entries for "Important Project Meetings" will show a 90%, making it easy to see at a glance which users are likely to attend.

[0446] 4. Meeting setup notification

[0447] The device (e.g., the user's PC or smartphone) will send a notification recommending avoiding time slots with low scores based on the predicted score when setting up a meeting. This notification will help with meeting setup and participant adjustments.

[0448] Examples:

[0449] When User C tries to schedule a new meeting, the calendar app recommends avoiding the "Weekly Meeting" time slot because it has a low score. This allows important meeting times to be adjusted to have a higher attendance rate.

[0450] 5. Collecting User Feedback

[0451] The server collects user feedback, including reasons for attending meetings and user impressions of the system, and uses this feedback to retrain the machine learning model to improve prediction accuracy.

[0452] Examples:

[0453] If a user provides feedback such as "this meeting was important to me, but I didn't attend because it had a low score," the model can be adjusted based on that feedback, allowing the system to provide a more accurate attendance propensity score.

[0454] These features enable efficient adjustment of meeting schedules in a remote work environment, which is expected to improve productivity. The system recommends time slots based on users' attendance trends and optimizes important meetings to increase attendance rates, thereby streamlining overall meeting management.

[0455] The processing flow will be explained below.

[0456] Step 1:

[0457] The server retrieves the user's calendar information, specifically, details of the user's appointments and meetings (title, start time, end time, attendee list, frequency, etc.) through the API.

[0458] Step 2:

[0459] The server collects past conference participation history. The collected data includes attendance history for web conferences, invitation acceptance rates, and web conference link click rates. This allows for an understanding of conference participation trends.

[0460] Step 3:

[0461] The server retrieves the user's job title and department information. The server retrieves the user's job title and department data from the HR system or database to clarify the user's role.

[0462] Step 4:

[0463] The server preprocesses the collected data by cleaning it, interpolating missing values, removing outliers, and making it suitable for machine learning models.

[0464] Step 5:

[0465] The server applies machine learning models to predict each user's tendency to attend meetings, using models such as logistic regression and random forests to calculate a participation propensity score based on data such as past attendance history and job title.

[0466] Step 6:

[0467] The server calculates the meeting attendance tendency score and displays it in the calendar app. Specifically, the score is displayed numerically and color-coded on each meeting entry, allowing users to understand it visually.

[0468] Step 7:

[0469] The device will notify you when a meeting is scheduled. When a user tries to schedule a new meeting, the device will send a notification recommending the best time based on the user's participation propensity score. This will help ensure that important members are able to attend.

[0470] Step 8:

[0471] The server collects feedback from users and stores the feedback provided by users to the system (e.g., opinions on the importance of the meeting and their willingness to attend) in a database.

[0472] Step 9:

[0473] The server retrains the machine learning model based on the collected feedback. By incorporating user feedback, the model's accuracy can be improved to provide a more accurate meeting attendance propensity score.

[0474] Through these steps, this system streamlines meeting scheduling in a remote work environment and improves user productivity.

[0475] Example 1

[0476] 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."

[0477] In remote work environments, productivity needs to be improved by streamlining user meeting schedules and optimizing meeting scheduling based on attendance trends. However, conventional systems struggle to adjust schedules while fully considering users' job titles and past meeting participation history. Furthermore, they do not recommend avoiding time slots with low attendance propensity scores when scheduling meetings, resulting in low attendance rates for important meetings. Additionally, they lacked a way to visually display meeting attendance propensity scores or a way to retrain machine learning models based on user feedback.

[0478] 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.

[0479] In this invention, the server includes means for acquiring user calendar information, means for collecting past conference participation history, means for acquiring user job title and department information, means for applying a machine learning model to predict the user's conference participation propensity based on the collected data, means for displaying the predicted conference participation propensity score in a calendar application, means for notifying the user to recommend a time slot with a low score when scheduling a conference, means for collecting user feedback and retraining the machine learning model, and means for reflecting the calculation results of the conference participation propensity score in actual conference scheduling and schedule management. This enables optimal schedule adjustment based on the user's conference participation propensity, which is expected to improve the attendance rate for important meetings and increase overall productivity.

[0480] "Calendar information" is digital schedule data including the user's plans, meeting dates and times, titles, participant information, and the like.

[0481] "Conference participation history" is data on conferences that a user has attended in the past, and includes information such as attendance history, invitation acceptance rate, and click rate of web conference links.

[0482] "Position and department information" is data relating to the position name and department name within the organization to which the user belongs.

[0483] A "machine learning model" is an algorithm or method used to analyze collected data and predict users' meeting participation trends.

[0484] A "meeting attendance propensity score" is a numerical value calculated using a machine learning model that indicates the probability that a user will attend a particular meeting.

[0485] A "calendar application" is software or a system that allows a user to manage their digital schedule and view appointments and meetings.

[0486] "Notification" is a means of sending alerts or messages to users to convey specific information.

[0487] "User feedback" refers to data on opinions and evaluations provided by users, such as reasons for attending or not attending a meeting and impressions of using the system.

[0488] "Retraining" is the process of improving the accuracy of a machine learning model based on new data collected.

[0489] "Visual display of scores" refers to a method of displaying the meeting attendance tendency scores in an easy-to-understand manner using color coding and icons.

[0490] This system is designed to streamline time management for web conferences in remote work environments. Specifically, the server collects users' calendar information, meeting participation history, job title, and department information, and then applies a machine learning model based on this information to predict the user's tendency to attend meetings. Furthermore, this participation tendency score is displayed in the calendar application, and when scheduling a meeting, a notification is sent recommending avoiding time slots with low scores, thereby helping to schedule appropriate meetings. Finally, user feedback is collected and the machine learning model is retrained to improve prediction accuracy. The specific implementation steps and the hardware and software used are described below.

[0491] Data collection

[0492] The server obtains the user's calendar information. Specifically, it uses the Google Calendar API or Microsoft Outlook Calendar API to collect each user's calendar entries along with the date, time, title, and participant information. The server also collects data such as past meeting participation history, attendance history, invitation acceptance rate, and web conference link click rate. It also obtains the user's job title and department information.

[0493] Examples:

[0494] User A has a "Weekly Meeting" scheduled for every Monday in his Google Calendar, but he has only attended one of the past three meetings. This data is collected on the server using the Google Calendar API. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[0495] Data analysis and score calculation

[0496] The server applies a machine learning model to the data collected. Specific algorithms include logistic regression and random forest, and Python libraries such as Scikit-learn and TensorFlow are used. This predicts each user's tendency to attend meetings and calculates a participation propensity score.

[0497] Examples:

[0498] User A's propensity score for attending "Weekly Meetings" is calculated to be 40%. User B's propensity score for attending important meetings is calculated to be 90%. This allows the server to quantify the user's probability of attending each meeting.

[0499] Score display

[0500] The server converts the calculated participation propensity score into a visually understandable format. Specifically, it generates code using HTML and JavaScript to display the score as a color code or a number. This score data is then sent to the calendar application for display.

[0501] Examples:

[0502] In a calendar app, the entry for User A's "Weekly Meeting" shows a "Participation Propensity Score of 40%." User B's "Important Project Meeting" shows a 90%, making it easy to see at a glance which users are likely to attend.

[0503] Meeting setup notifications

[0504] The device obtains the date and time and user list of the newly scheduled meeting, checks the existing attendance tendency score, and sends a notification recommending avoiding time slots with low scores. This allows meetings to be scheduled at appropriate time slots, improving the attendance rate of important meetings.

[0505] Examples:

[0506] When User C tries to schedule a new meeting, the calendar app recommends avoiding the "Weekly Meeting" time slot because it has a low score. This allows important meeting times to be adjusted to have a higher attendance rate.

[0507] Collecting user feedback

[0508] The server sends a request for feedback to users after the meeting or periodically. The server collects information from users about their reasons for attending or not attending meetings and their impressions of using the system using a web form or in-app feedback function. This feedback is used to retrain the machine learning model and improve the accuracy of predictions.

[0509] Examples:

[0510] If a user provides feedback such as "this meeting was important to me, but I didn't attend because it had a low score," the model can be adjusted based on that feedback, allowing the system to provide a more accurate attendance propensity score.

[0511] Prompt Sentence Examples

[0512] "Please look at User A's calendar and calculate and display this week's meeting attendance propensity score."

[0513] "When scheduling meetings, please remind me to avoid times with low attendance propensity scores."

[0514] This system makes it possible to efficiently adjust meeting schedules in a remote work environment, which is expected to improve productivity.

[0515] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0516] Step 1:

[0517] The server obtains the user's calendar information. To do this, it uses the Google Calendar API or the Microsoft Outlook Calendar API. The input is the user's access privileges to the calendar API, and the output is calendar entry data including the date, time, title, and attendee information of the user's appointments and meetings. This calendar entry data is then saved on the server.

[0518] Specific behavior:

[0519] The server sends an API request to retrieve the user's calendar information.

[0520] The server retrieves the calendar entries and stores them in a database.

[0521] Step 2:

[0522] The server collects past conference participation history. The input is calendar entries and participation history data, and the output is historical data including each user's attendance history, invitation acceptance rate, and web conference link click rate. This historical data is organized and stored on the server.

[0523] Specific behavior:

[0524] The server extracts past meeting data from the calendar entries.

[0525] The server calculates the user's attendance history, invitation acceptance rate, and web conference link click rate, and stores them in a database.

[0526] Step 3:

[0527] The server retrieves the user's job title and department information. The input is a database of job titles and departments within the company, and the output is the job title and department information for each user. This information is then stored in the integrated database.

[0528] Specific behavior:

[0529] The server accesses a personnel database within the company to obtain the user's job title and department information.

[0530] The server stores the acquired job title and department information in a database.

[0531] Step 4:

[0532] The server applies a machine learning model to the collected data. The input is calendar information, meeting attendance history, job title, and department information, and the output is the user's meeting attendance propensity score. Python's Scikit-learn and TensorFlow are used for data processing and learning.

[0533] Specific behavior:

[0534] The server extracts the necessary data from the database and performs preprocessing.

[0535] The server inputs data into the machine learning model, performs learning, and calculates the participation propensity score.

[0536] The server stores the calculated score in a database.

[0537] Step 5:

[0538] The server displays the predicted meeting attendance propensity score in a calendar application. The input is the attendance propensity score, and the output is a visually color-coded and / or numerically displayed score.

[0539] Specific behavior:

[0540] The server sends a request to the calendar application to display the score.

[0541] The calendar application displays the score in color.

[0542] Step 6:

[0543] The terminal sends a notification to the user recommending avoiding low-score time slots when setting up a meeting. The input is the date and time of the new meeting and the attendance tendency score, and the output is a notification to the user.

[0544] Specific behavior:

[0545] The terminal obtains the setting information for the new conference.

[0546] The device will check your existing participation propensity score and send you notifications recommending you avoid low-scoring times.

[0547] Step 7:

[0548] The server collects and analyzes user feedback, with the input being the user feedback data and the output being the analyzed feedback data, and then retrains the machine learning model based on this.

[0549] Specific behavior:

[0550] The server sends notifications to users to collect feedback.

[0551] The server collects feedback data from users and stores it in a database.

[0552] The server analyzes the feedback data and retrains the machine learning model.

[0553] (Application example 1)

[0554] 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."

[0555] In work environments such as remote work and factories, meeting time management is important, and there is a particular need to improve meeting attendance rates. However, it is difficult to predict the schedules and attendance tendencies of individual users and workers and then set optimal meeting times based on these. Furthermore, conventional systems have difficulty using user feedback to improve accuracy, leaving room for improvement. Therefore, there is a need for a system that can improve meeting attendance rates and optimize work efficiency.

[0556] 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.

[0557] In this invention, the server includes means for acquiring user schedule information, means for collecting past conference participation history, means for acquiring user job title and department information, means for applying a generative AI model to predict a user's conference participation tendency based on the collected data, means for displaying the predicted conference participation tendency score in a schedule management application, means for notifying users to avoid time periods with low scores when scheduling a conference, means for collecting user feedback and retraining the generative AI model, and means for supporting the efficiency of industrial machine maintenance and production planning meetings. This makes it possible to improve meeting attendance rates and optimize overall business efficiency.

[0558] "User schedule information" refers to data such as the date, time, location, and content of appointments and meetings set by the user.

[0559] "Conference participation history" refers to historical data of conferences that a user has attended in the past, and includes information such as whether or not the user attended, the frequency of attendance, and the attendance rate.

[0560] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to predict user behavior patterns and trends from data.

[0561] A "schedule management application" refers to a software application that helps users manage their schedules and meetings.

[0562] The "conference attendance propensity score" is a numerical representation of the likelihood that a user will attend a particular conference, and is calculated based on a prediction.

[0563] "Notification mechanism" refers to a digital messaging or alert system that notifies users of specific information or recommendations.

[0564] "Feedback" refers to opinions, impressions, and evaluations collected from users regarding their use of the system.

[0565] "Industrial machinery" refers to various machinery and equipment used in manufacturing and production industries, including those that require maintenance and management.

[0566] "Maintenance" refers to repairs and maintenance work carried out to keep industrial machinery in working order.

[0567] A "manufacturing planning meeting" refers to a meeting held to improve the efficiency of the production line and adjust schedules.

[0568] This invention is a system for streamlining time management in remote work environments, industrial machinery maintenance meetings, production planning meetings, etc. Specifically, it predicts a user's tendency to attend meetings based on the user's schedule information, meeting participation history, job title, and department information, and displays this in a schedule management application.

[0569] Program Description

[0570] The server collects users' schedule information, past meeting attendance history, job title, and department information. This data is used to apply a generative AI model to predict a meeting attendance propensity score. The score is displayed in the schedule management application, and notifications are sent when scheduling meetings to avoid low-scoring times. Furthermore, feedback from users can be collected to retrain the generative AI model and improve prediction accuracy. This can improve meeting attendance rates and optimize overall business efficiency.

[0571] Hardware and software used

[0572] 1. Hardware:

[0573] Factory Server

[0574] User's Digital Dashboard

[0575] 2. Software:

[0576] Python

[0577] Pandas (data processing library)

[0578] Scikit-learn (machine learning library)

[0579] The server retrieves users' schedule information, meeting attendance history, job title, and department information using a Python script that reads CSV files. It then applies a machine learning algorithm (here, a logistic regression model) to analyze the collected data. Using Python and Scikit-learn, the model is trained and a meeting attendance propensity score is calculated for each user.

[0580] The scores are then displayed on a digital dashboard using Pandas and related visualization libraries. The scores are visually represented using color coding and icons, providing a user-friendly format. The server also uses the scores to create meeting scheduling recommendations and sends them to the user's device. User feedback is also collected and used to retrain the generative AI model.

[0581] As a concrete example of data, the shift information and meeting attendance history of factory workers are shown below:

[0582] Shift Information:

[0583] employee_id, shift_hours, past_meeting_attendance

[0584] 1, 8, 0.75

[0585] 2, 6, 0.90

[0586] Meeting participation history:

[0587] employee_id, attendance_probability

[0588] 1, 0.80

[0589] 2, 0.95

[0590] Prompt Sentence Examples

[0591] For the generative AI model, enter the following prompt:

[0592] Please generate Python code that calculates the meeting attendance tendency score based on shift information and meeting attendance history, and sends notifications at specific times. The example data is as follows.

[0593] Shift Information:

[0594] employee_id, shift_hours, past_meeting_attendance

[0595] 1, 8, 0.75

[0596] 2, 6, 0.90

[0597] Meeting participation history:

[0598] employee_id, attendance_probability

[0599] 1, 0.80

[0600] 2, 0.95

[0601] The above configuration can optimize users' meeting schedules and significantly improve work efficiency in remote work and industrial environments.

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

[0603] Step 1: Data collection

[0604] The server reads user schedule information, past meeting attendance history, job title, and department information from CSV files. Specifically, it uses Python's Pandas library to read and integrate this data as a data frame. The input includes each user's shift information (employee_id, shift_hours, past_meeting_attendance) and meeting attendance history (employee_id, attendance_probability). The output is a dataset that integrates this data.

[0605] Step 2: Data analysis and score calculation

[0606] The server applies a generative AI model (logistic regression algorithm) to the collected data to predict each user's meeting attendance propensity score. The input includes shift information and meeting attendance history, and the output includes each user's meeting attendance propensity score. This score represents the probability of attendance, expressed as a number ranging from 0 to 1.

[0607] Step 3: View the score

[0608] The server visualizes the data to display the predicted meeting attendance propensity scores in the schedule management application. Specifically, the scores are visually displayed on a digital dashboard using color coding and icons. The input includes the meeting attendance propensity scores for each user, and the output includes the visually displayed scores.

[0609] Step 4: Notification of meeting settings

[0610] The server creates a recommendation notification to avoid low-scoring time slots when scheduling a meeting and sends it to the user's device. The input includes the predicted meeting attendance propensity score, and the output is the recommendation notification message. For example, it generates a message like "Employee ID: 1 should be scheduled for a meeting outside their current shift hours."

[0611] Step 5: Gather user feedback and retrain

[0612] The server collects user feedback after the meeting. The input includes user opinions and usage impressions, and the output is feedback data. This feedback is used to retrain the generative AI model to improve prediction accuracy. Specifically, the newly collected feedback data is added to the existing dataset, and the machine learning model is retrained.

[0613] Through the above processing steps, a system is realized that aims to improve meeting attendance rates and optimize business efficiency.

[0614] 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.

[0615] This invention relates to a system for streamlining the scheduling of web conferences in a remote work environment and improving productivity. In particular, by combining it with an emotion engine that recognizes user emotions, it more accurately predicts conference participation trends and supports optimal conference scheduling.

[0616] The system mainly includes the following functions:

[0617] 1. Data Collection

[0618] The server retrieves the user's calendar information via an API. It also collects the user's past meeting participation history and obtains data such as attendance history, invitation acceptance rate, and web conference link click rate. In addition, it retrieves the user's job title and department information from the HR system or database.

[0619] Examples:

[0620] User A has a "Weekly Meeting" scheduled for every Monday on his calendar, but he has only attended one of the past three meetings. This data is collected by the server. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[0621] 2. Emotional Data Collection

[0622] The device uses an emotion engine that recognizes the user's emotions in real time, acquiring emotion data and understanding the emotional state (e.g., stress, satisfaction, dissatisfaction) that influences participation tendencies.

[0623] Examples:

[0624] When User C attended a particular meeting, the emotion engine recognized high stress. This information is sent to the server and used to predict meeting attendance trends.

[0625] 3. Data analysis and score calculation

[0626] The server applies a machine learning model based on the collected calendar information, meeting participation history, job title and department information, and emotion data to predict each user's tendency to participate in meetings, thereby calculating a participation propensity score.

[0627] Examples:

[0628] User A's propensity score for attending the "Weekly Meeting" is calculated as 40% based on their emotional data and past history. This prediction takes into account their past participation frequency and emotional state.

[0629] 4. Score display

[0630] The server displays the calculated meeting attendance tendency score in the calendar app. The score is provided in a visually understandable format using color coding and numerical values. The server also displays the user's emotional state, allowing the user to share their state.

[0631] Examples:

[0632] In the calendar app, the entry for "Weekly Meeting" will show "Participation Propensity Score 40%" along with the stress level.

[0633] 5. Meeting setup notifications

[0634] The device provides notifications when setting up a meeting. When a user tries to set up a new meeting, the device sends a notification recommending the best time slot based on the participation propensity score and emotion data.

[0635] Examples:

[0636] If User C tries to schedule a new meeting, the calendar app will show that the "Weekly Meeting" time slot has a low score and a high stress level, and recommend that they avoid that time.

[0637] 6. Collecting User Feedback

[0638] The server collects feedback from users, including reasons for attending meetings, impressions of using the system, and opinions based on sentiment data.

[0639] Examples:

[0640] If a user provides feedback such as "This meeting was important, but I was too stressed and didn't attend," the model is retrained based on that feedback.

[0641] 7. Retraining the model

[0642] The server retrains the machine learning model based on the collected feedback, allowing the system to calculate a more accurate meeting attendance propensity score.

[0643] Through these functions, the system streamlines the adjustment of meeting schedules in a remote work environment, improving productivity. By taking into account the user's emotional state, it is possible to set up meetings that are easier to participate in, further improving the efficiency of overall meeting management.

[0644] The processing flow will be explained below.

[0645] Step 1:

[0646] The server obtains the user's calendar information. Specifically, it collects detailed information about the user's schedules and meetings (title, start time, end time, attendee list, frequency, etc.) through the API. This allows it to understand what meetings the user is participating in.

[0647] Step 2:

[0648] The server collects past conference participation history, including attendance history, invitation acceptance rates, and click rates for web conference links, allowing for analysis of participation trends for each conference.

[0649] Step 3:

[0650] The server retrieves the user's job title and department information. The server retrieves data about the user's job title and department from the HR system or database to clarify the user's role.

[0651] Step 4:

[0652] The device activates an emotion engine to recognize the user's emotions. The emotion engine acquires real-time emotion data from the user using technologies such as voice recognition and facial expression analysis.

[0653] Step 5:

[0654] The server receives the emotion data sent from the emotion engine and stores it in a database, where the emotional state (e.g., stress, satisfaction, dissatisfaction) is recorded.

[0655] Step 6:

[0656] The server preprocesses all data collected (calendar information, meeting attendance history, job title / department information, and emotion data) by cleaning the data, interpolating missing values, and removing outliers, and converting it into a format suitable for machine learning models.

[0657] Step 7:

[0658] The server applies machine learning models to predict each user's tendency to attend meetings, using models such as logistic regression and random forests, and calculates a participation propensity score that takes into account past participation history and emotional data.

[0659] Step 8:

[0660] The server calculates the meeting attendance tendency score and displays it in the calendar app. The score is presented in a visually easy-to-understand format using color coding and numerical values. The user's emotional state can also be displayed.

[0661] Step 9:

[0662] The device will notify users when a meeting is scheduled. When a user tries to schedule a new meeting, the device will send a notification recommending the best time based on attendance propensity scores and sentiment data. This will help ensure that important members are able to attend.

[0663] Step 10:

[0664] The server collects feedback from users, who provide reasons for attending meetings, their impressions of using the system, and opinions based on emotional data, and stores the feedback in a database.

[0665] Step 11:

[0666] The server retrains the machine learning model based on the collected feedback. By incorporating user feedback, the model's accuracy can be improved to provide a more accurate meeting attendance propensity score.

[0667] Through these steps, the system streamlines the scheduling of meetings in a remote work environment, improving user productivity. By taking emotional states into account, it becomes possible to set up meetings that are easier to participate in, further improving the efficiency of overall meeting management.

[0668] Example 2

[0669] 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."

[0670] In a remote work environment, conventional systems adjust the schedule of web conferences without taking into account the emotional state of the user, which has led to problems such as a decrease in conference participation rates and an increase in user stress. There is a need to solve these problems and realize optimal conference settings that reflect the emotional state of the user.

[0671] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0672] In this invention, the server includes means for acquiring user schedule information, means for collecting past meeting attendance history, means for acquiring user job title and department information, means for recognizing emotional states in real time and collecting data, means for predicting the user's tendency to attend meetings by applying a machine learning model based on the collected calendar information, meeting attendance history, job title and department information, and emotional data, means for displaying the predicted meeting attendance propensity score in a calendar application, means for notifying the user of time periods with low scores when scheduling a meeting, and means for collecting user feedback and retraining the machine learning model. This makes it possible to adjust an optimal meeting schedule taking into account the user's emotional state.

[0673] "User schedule information" refers to information about schedules and events that a user has registered in a calendar or schedule management application.

[0674] "Past meeting attendance history" refers to data such as attendance status of meetings a user has previously attended, invitation acceptance rate, and click rate of web conference links.

[0675] "User position and department information" is information about the user's job title and the department to which the user belongs, and is obtained from an HR system or database.

[0676] "Means for recognizing emotional states and collecting data in real time" refers to technology that uses built-in cameras and microphones to capture the user's facial expressions and tone of voice, and then analyzes them using an emotion engine.

[0677] "Emotion data" is data that represents the user's emotional state, such as stress, satisfaction, or dissatisfaction.

[0678] "Machine learning model" refers to an algorithm or statistical model used to predict users' meeting participation trends based on collected data.

[0679] The "meeting attendance propensity score" is a numerical value calculated based on a machine learning model that indicates the likelihood that a user will attend a particular meeting.

[0680] A "calendar application" is software that allows users to manage their schedules and has a function to display meeting attendance propensity scores.

[0681] "Means of notification" refers to technical means for notifying users of information or recommendations in real time, such as push notifications or alerts.

[0682] "Means for collecting feedback" refers to a function for collecting opinions and impressions from users and using them to improve the system.

[0683] "Retraining" is the process of updating a machine learning model based on collected feedback to improve its prediction accuracy.

[0684] This invention relates to a system that streamlines scheduling of web conferences in a remote work environment and improves productivity. In particular, by combining it with an emotion engine that recognizes user emotions, it more accurately predicts conference participation trends and supports optimal conference scheduling.

[0685] Data collection

[0686] First, the server obtains the user's schedule information. The server collects data from schedule management applications such as Google Calendar and Outlook Calendar via API. Next, the server accesses the HR system or internal database to obtain the user's job title and department information. At the same time, it collects the user's past meeting attendance history and stores it in the database.

[0687] For example, User A has a "Weekly Meeting" scheduled on his calendar, but he has only attended one of the past three meetings. This data is collected by the server. User B is a manager and tends to invite his subordinates to many meetings.

[0688] Emotional Data Collection

[0689] The device uses the built-in camera and microphone to recognize the user's emotional state in real time. It uses an emotion engine (e.g., Microsoft Azure's Emotion API) to capture the user's facial expressions and tone of voice and analyzes the emotional data, such as stress, satisfaction, and dissatisfaction. This information is sent to a server and stored in a database.

[0690] For example, when user C is attending a particular meeting, the emotion engine recognizes high stress and sends this information to the server.

[0691] Data analysis and score calculation

[0692] The server applies a machine learning model (e.g., scikit-learn or TensorFlow) based on the collected calendar information, meeting participation history, job title / department information, and emotion data to predict each user's tendency to participate in meetings. This calculates a participation propensity score, which is then stored in a database.

[0693] For example, the participation propensity score for user A in the "Weekly Meeting" is calculated as 40% based on emotional data and past history. This prediction takes into account the frequency of past participation and emotional state.

[0694] Score display

[0695] The server displays the calculated meeting attendance tendency score in a calendar application, which displays the score in a visually easy-to-understand format.

[0696] For example, an entry for "Weekly Meeting" will display "Participation Propensity Score 40%" along with the emotional state.

[0697] Meeting setup notifications

[0698] When a user attempts to schedule a new meeting, the device receives the score and emotion data from the server and generates a notification suggesting the optimal meeting time and structure, which is displayed to the user as a push notification or alert.

[0699] For example, if User C tries to schedule a new meeting, the calendar application may indicate that the "Weekly Meeting" time slot has a low score, indicating a high stress state, and recommend that they avoid that time.

[0700] Gathering user feedback and retraining the model

[0701] The server collects user opinions and feedback through the calendar application and a dedicated feedback form. This feedback is stored in a database and analyzed. Based on the collected feedback, the server retrains the machine learning model to improve prediction accuracy.

[0702] For example, if a user provides feedback such as "This meeting was important, but I was too stressed and didn't attend," the server uses this data to retrain the model.

[0703] Prompt Sentence Examples

[0704] "Calculate the attendance propensity score for the next weekly meeting based on the user's sentiment data and meeting attendance history."

[0705] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0706] Step 1:

[0707] Input: User's calendar information (Google Calendar, Outlook, etc.), HR system or database information (job title, department information), past meeting attendance history (attendance status, invitation acceptance rate, etc.)

[0708] Specific operation: The server collects the user's schedule information from Google Calendar or Outlook Calendar via API. The server then accesses the HR system or internal database to obtain the user's job title and department information. It also obtains the user's past meeting attendance history and stores it in the database.

[0709] Output: The retrieved calendar information, job title and department information, and meeting attendance history are saved in a database.

[0710] Step 2:

[0711] Input: User's facial expression and voice data

[0712] How it works: The device uses the built-in camera and microphone to capture the user's facial expressions and tone of voice in real time. An emotion engine (e.g., Microsoft Azure's Emotion API) is used to analyze the user's emotional state (stress, satisfaction, dissatisfaction, etc.) from the captured data. This information is then sent to a server.

[0713] Output: The analyzed emotion data is sent to the server and stored in a database.

[0714] Step 3:

[0715] Input: User's calendar information, job title and department information, past meeting attendance history, emotional data

[0716] Specific operation: The server integrates the collected data and uses a machine learning model (e.g., scikit-learn or TensorFlow) to predict users' meeting participation trends. Specifically, it analyzes each user's meeting participation patterns based on past data and calculates the participation propensity score for each meeting.

[0717] Output: The calculated meeting attendance propensity score is saved in the database.

[0718] Step 4:

[0719] Input: Meeting participation propensity score

[0720] Specific operation: The server sends the meeting attendance tendency score to the calendar application UI. The calendar application displays the received score using color coding and numerical values, providing the user with a visually easy-to-understand format.

[0721] Output: The meeting attendance propensity score is displayed in the user's calendar application.

[0722] Step 5:

[0723] Input: Meeting participation tendency score and emotion data from the server

[0724] Specific operation: When a user attempts to schedule a new meeting, the device receives the meeting participation propensity score and emotion data from the server. The device generates and displays a notification to the user suggesting the optimal meeting time and configuration.

[0725] Output: The user will be notified with a suggestion for the best meeting time.

[0726] Step 6:

[0727] Input: User feedback (opinions about meeting attendance, impressions on using the system, opinions based on emotional data)

[0728] What it does: The server collects user feedback through the calendar application and a dedicated feedback form, stores the collected feedback in a database, and formats it for analysis.

[0729] Output: The collected feedback is stored in a database.

[0730] Step 7:

[0731] Input: Collected feedback data

[0732] What happens: The server takes in new feedback data and combines it with existing training data to retrain the machine learning model and generate a more accurate meeting attendance propensity score.

[0733] Output: The updated machine learning model is used for the next prediction.

[0734] (Application example 2)

[0735] 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."

[0736] Improving the efficiency of store staff work methods and the working environment are important issues in today's brick-and-mortar stores. In particular, the impact of staff emotional states on work efficiency and customer service cannot be ignored. However, there is currently no way to understand in real time how staff are working, making it difficult to create optimal work schedules. Furthermore, there is no established method for appropriately avoiding work that is likely to overload staff. Furthermore, there is no efficient way to collect staff feedback and optimize work based on it. This increases staff stress, ultimately leading to a decrease in customer satisfaction.

[0737] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0738] In this invention, the server includes: means for acquiring a user's calendar information; means for collecting past meeting participation history; means for acquiring the user's job title and department information; means for acquiring emotional data and recognizing the user's state; means for applying a machine learning model to predict the user's tendency to attend meetings based on the collected data; means for displaying the predicted meeting participation propensity score in a calendar application; means for notifying the user to recommend a time slot with a low score when scheduling a meeting; means for collecting user feedback and retraining the machine learning model; and means for visually displaying the optimized work schedule and the impact of the emotional data. This makes it possible to grasp the current emotional state of store staff in real time and provide an optimal work schedule that takes their emotional state into account. Furthermore, by efficiently collecting staff feedback and continuously optimizing work based on it, an improved working environment and increased customer satisfaction can be achieved.

[0739] "User" refers to a person who uses the system.

[0740] "Calendar information" refers to information that records a user's schedule and plans.

[0741] "Conference participation history" refers to a record of conferences that a user has participated in in the past.

[0742] "Job title and department information" refers to information about the department and job to which the user belongs.

[0743] "Collected data" refers to calendar information, meeting participation history, job title and department information, emotional data, etc.

[0744] A "machine learning model" is an algorithm that learns patterns based on collected data and makes predictions and analyses.

[0745] A "calendar application" is software for managing a user's schedule.

[0746] "Emotional Data" refers to data that records and analyzes the emotional state of a user.

[0747] "User state" refers to the user's current emotional and psychological state.

[0748] "Work schedule" refers to the planned work to be performed by a user.

[0749] The "predicted conference attendance propensity score" is a score calculated by a machine learning model that indicates the likelihood of a user attending a conference.

[0750] "Notification" is a means of conveying information to the user through an interface.

[0751] "Feedback" refers to opinions and evaluations from users of the system.

[0752] An "optimized work schedule" refers to a work schedule that is adjusted taking into account the user's emotional state and work efficiency.

[0753] "Visual display means" refers to methods of visually presenting information to users using graphs, icons, color coding, etc.

[0754] This invention aims to apply a meeting schedule adjustment system in a remote work environment to optimizing work schedules in brick-and-mortar stores. The system utilizes user emotional data to grasp the real-time emotional state of staff and propose optimal work schedules.

[0755] The server first obtains the user's calendar information, including the user's past conference attendance history and current schedule. It also obtains the user's job title and department information to understand their work characteristics.

[0756] The device then collects the user's real-time emotional data, which is acquired through smart glasses or other wearable devices worn by the user, and uses this emotional data to measure the user's current stress level, satisfaction, dissatisfaction, etc.

[0757] The server applies machine learning models to the collected data to predict each user's work participation trends. Specifically, it integrates and analyzes past meeting history, job title and department information, and real-time emotion data. This analysis utilizes an emotion engine and schedule optimization algorithm to calculate the optimal schedule to maximize the user's work efficiency.

[0758] The optimized work schedule is displayed on the user's smart glasses or smartphone. The schedule is color-coded and displayed with icons for easy visual understanding. The impact of emotional data is also displayed visually, allowing staff to work while being aware of their own state.

[0759] Furthermore, the server collects user feedback and retrains the machine learning model based on it. The feedback includes opinions based on the results of applying the work schedule and emotional data. This allows the system to continuously improve and provide more accurate work schedules.

[0760] For example, if a staff member gives feedback that they felt high stress during their last work session, the system will use that information to adjust their next work schedule. Specifically, if the emotion engine detects that a staff member's current stress level is high, the system will prioritize relaxing tasks.

[0761] Below is an example of a prompt sentence.

[0762] Example prompt sentence:

[0763] “Sort the following tasks based on the optimal work schedule. Consider the emotional state of your staff and create a schedule that is least stressful.

[0764] Product display

[0765] Customer Service

[0766] Stock Check

[0767] Sales Report

[0768] User's current mild stress level is 8."

[0769] In this way, the present invention can be used in brick-and-mortar stores to improve the working environment for staff and increase work efficiency.

[0770] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0771] Step 1:

[0772] The server retrieves the user's calendar information.

[0773] Input: User ID

[0774] Output: User's calendar information (schedule and events)

[0775] What it does: The server retrieves the user's calendar information through an API request, providing data that includes the user's events and past meeting history.

[0776] Step 2:

[0777] The server retrieves the user's job title and department information.

[0778] Input: User ID

[0779] Output: User's job title and department information

[0780] What happens: The server sends an API request to an HR system or database to retrieve information about the user's job title and department.

[0781] Step 3:

[0782] The device collects the user's emotional data in real time.

[0783] Input: User ID, Emotion Engine

[0784] Output: User's emotional data (stress level, satisfaction, dissatisfaction, etc.)

[0785] Specific operation: The smart glasses or wearable device worn by the user uses an emotion engine to collect emotion data in real time and transmits the data to a server.

[0786] Step 4:

[0787] The server applies a machine learning model based on the collected data to predict the user's work participation tendencies.

[0788] Input: Calendar information, job title and department information, emotion data

[0789] Output: Work participation propensity score

[0790] Specific operation: The server integrates these data into a single dataset and applies machine learning algorithms (e.g., random forests, neural networks) to predict the user's work participation tendencies.

[0791] Step 5:

[0792] The server displays the predicted work attendance propensity score in a calendar application.

[0793] Input: Work participation propensity score

[0794] Output: Work participation propensity score displayed as a color or icon

[0795] Specific operation: The server incorporates the score into the calendar application and displays the results on the user's device in a visually easy-to-understand format (color coding and icons).

[0796] Step 6:

[0797] The device will notify you to recommend a time slot with a low score when setting up a meeting.

[0798] Input: Work participation propensity score

[0799] Output: Best time suggestion

[0800] Specific operation: The device sends a notification to the user, recommending that the user avoid times when the work attendance propensity score is low.

[0801] Step 7:

[0802] The server collects user feedback and retrains the machine learning model.

[0803] Input: User feedback

[0804] Output: An improved machine learning model

[0805] How it works: The server collects user feedback (e.g., "This time of day is stressful") and uses it to retrain the machine learning model to improve accuracy.

[0806] Step 8:

[0807] The server visually displays the optimized work schedule and the impact of emotional data.

[0808] Input: Optimized work schedule, emotional data

[0809] Output: Visually displayed work schedule and the impact of emotional data

[0810] Specific operation: The server visually displays the optimized schedule and emotion data in a calendar application or smart glasses, allowing users to work while being aware of their own state.

[0811] 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.

[0812] 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.

[0813] 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.

[0814] [Third embodiment]

[0815] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0816] 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.

[0817] 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).

[0818] 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.

[0819] 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.

[0820] 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).

[0821] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0822] 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.

[0823] 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.

[0824] 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.

[0825] 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.

[0826] 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."

[0827] This invention relates to a system for achieving efficient time management for web conferences in a remote work environment. Specifically, it supports appropriate conference scheduling by predicting a user's conference participation tendency based on the user's calendar information, conference participation history, job title, and department information, and displaying the score in a calendar app.

[0828] The system mainly includes the following functions:

[0829] 1. Data Collection

[0830] The server obtains the user's calendar information, collects the user's past meeting participation history, and collects data such as attendance history, invitation acceptance rate, and web conference link click rate, as well as the user's job title and department information.

[0831] Examples:

[0832] User A has a "Weekly Meeting" scheduled for every Monday on his calendar, but he has only attended one of the past three meetings. This data is collected on the server. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[0833] 2. Data analysis and score calculation

[0834] The server applies a machine learning model to the collected data to predict each user's tendency to attend meetings. This prediction can be made using algorithms such as logistic regression or random forest. As a result, a participation propensity score is calculated for each user.

[0835] Examples:

[0836] User A's propensity score for attending "Weekly Meetings" is calculated to be 40%. User B's propensity score for attending important meetings is calculated to be 90%. This allows the server to quantify the user's probability of attending each meeting.

[0837] 3. Score display

[0838] The server displays this participation propensity score in the calendar application, using color coding and numerical values ​​to provide a visually easy-to-understand score.

[0839] Examples:

[0840] In the calendar app, entries for "Weekly Meetings" will show a "Participation Propensity Score of 40%," while entries for "Important Project Meetings" will show a 90%, making it easy to see at a glance which users are likely to attend.

[0841] 4. Meeting setup notification

[0842] The device (e.g., the user's PC or smartphone) will send a notification recommending avoiding time slots with low scores based on the predicted score when setting up a meeting. This notification will help with meeting setup and participant adjustments.

[0843] Examples:

[0844] When User C tries to schedule a new meeting, the calendar app recommends avoiding the "Weekly Meeting" time slot because it has a low score. This allows important meeting times to be adjusted to have a higher attendance rate.

[0845] 5. Collecting User Feedback

[0846] The server collects user feedback, including reasons for attending meetings and user impressions of the system, and uses this feedback to retrain the machine learning model to improve prediction accuracy.

[0847] Examples:

[0848] If a user provides feedback such as "this meeting was important to me, but I didn't attend because it had a low score," the model can be adjusted based on that feedback, allowing the system to provide a more accurate attendance propensity score.

[0849] These features enable efficient adjustment of meeting schedules in a remote work environment, which is expected to improve productivity. The system recommends time slots based on users' attendance trends and optimizes important meetings to increase attendance rates, thereby streamlining overall meeting management.

[0850] The processing flow will be explained below.

[0851] Step 1:

[0852] The server retrieves the user's calendar information, specifically, details of the user's appointments and meetings (title, start time, end time, attendee list, frequency, etc.) through the API.

[0853] Step 2:

[0854] The server collects past conference participation history. The collected data includes attendance history for web conferences, invitation acceptance rates, and web conference link click rates. This allows for an understanding of conference participation trends.

[0855] Step 3:

[0856] The server retrieves the user's job title and department information. The server retrieves the user's job title and department data from the HR system or database to clarify the user's role.

[0857] Step 4:

[0858] The server preprocesses the collected data by cleaning it, interpolating missing values, removing outliers, and making it suitable for machine learning models.

[0859] Step 5:

[0860] The server applies machine learning models to predict each user's tendency to attend meetings, using models such as logistic regression and random forests to calculate a participation propensity score based on data such as past attendance history and job title.

[0861] Step 6:

[0862] The server calculates the meeting attendance tendency score and displays it in the calendar app. Specifically, the score is displayed numerically and color-coded on each meeting entry, allowing users to understand it visually.

[0863] Step 7:

[0864] The device will notify you when a meeting is scheduled. When a user tries to schedule a new meeting, the device will send a notification recommending the best time based on the user's participation propensity score. This will help ensure that important members are able to attend.

[0865] Step 8:

[0866] The server collects feedback from users and stores the feedback provided by users to the system (e.g., opinions on the importance of the meeting and their willingness to attend) in a database.

[0867] Step 9:

[0868] The server retrains the machine learning model based on the collected feedback. By incorporating user feedback, the model's accuracy can be improved to provide a more accurate meeting attendance propensity score.

[0869] Through these steps, this system streamlines meeting scheduling in a remote work environment and improves user productivity.

[0870] Example 1

[0871] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0872] In remote work environments, productivity needs to be improved by streamlining user meeting schedules and optimizing meeting scheduling based on attendance trends. However, conventional systems struggle to adjust schedules while fully considering users' job titles and past meeting participation history. Furthermore, they do not recommend avoiding time slots with low attendance propensity scores when scheduling meetings, resulting in low attendance rates for important meetings. Additionally, they lacked a way to visually display meeting attendance propensity scores or a way to retrain machine learning models based on user feedback.

[0873] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0874] In this invention, the server includes means for acquiring user calendar information, means for collecting past conference participation history, means for acquiring user job title and department information, means for applying a machine learning model to predict the user's conference participation propensity based on the collected data, means for displaying the predicted conference participation propensity score in a calendar application, means for notifying the user to recommend a time slot with a low score when scheduling a conference, means for collecting user feedback and retraining the machine learning model, and means for reflecting the calculation results of the conference participation propensity score in actual conference scheduling and schedule management. This enables optimal schedule adjustment based on the user's conference participation propensity, which is expected to improve the attendance rate for important meetings and increase overall productivity.

[0875] "Calendar information" is digital schedule data including the user's plans, meeting dates and times, titles, participant information, and the like.

[0876] "Conference participation history" is data on conferences that a user has attended in the past, and includes information such as attendance history, invitation acceptance rate, and click rate of web conference links.

[0877] "Position and department information" is data relating to the position name and department name within the organization to which the user belongs.

[0878] A "machine learning model" is an algorithm or method used to analyze collected data and predict users' meeting participation trends.

[0879] A "meeting attendance propensity score" is a numerical value calculated using a machine learning model that indicates the probability that a user will attend a particular meeting.

[0880] A "calendar application" is software or a system that allows a user to manage their digital schedule and view appointments and meetings.

[0881] "Notification" is a means of sending alerts or messages to users to convey specific information.

[0882] "User feedback" refers to data on opinions and evaluations provided by users, such as reasons for attending or not attending a meeting and impressions of using the system.

[0883] "Retraining" is the process of improving the accuracy of a machine learning model based on new data collected.

[0884] "Visual display of scores" refers to a method of displaying the meeting attendance tendency scores in an easy-to-understand manner using color coding and icons.

[0885] This system is designed to streamline time management for web conferences in remote work environments. Specifically, the server collects users' calendar information, meeting participation history, job title, and department information, and then applies a machine learning model based on this information to predict the user's tendency to attend meetings. Furthermore, this participation tendency score is displayed in the calendar application, and when scheduling a meeting, a notification is sent recommending avoiding time slots with low scores, thereby helping to schedule appropriate meetings. Finally, user feedback is collected and the machine learning model is retrained to improve prediction accuracy. The specific implementation steps and the hardware and software used are described below.

[0886] Data collection

[0887] The server obtains the user's calendar information. Specifically, it uses the Google Calendar API or Microsoft Outlook Calendar API to collect each user's calendar entries along with the date, time, title, and participant information. The server also collects data such as past meeting participation history, attendance history, invitation acceptance rate, and web conference link click rate. It also obtains the user's job title and department information.

[0888] Examples:

[0889] User A has a "Weekly Meeting" scheduled for every Monday in his Google Calendar, but he has only attended one of the past three meetings. This data is collected on the server using the Google Calendar API. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[0890] Data analysis and score calculation

[0891] The server applies a machine learning model to the data collected. Specific algorithms include logistic regression and random forest, and Python libraries such as Scikit-learn and TensorFlow are used. This predicts each user's tendency to attend meetings and calculates a participation propensity score.

[0892] Examples:

[0893] User A's propensity score for attending "Weekly Meetings" is calculated to be 40%. User B's propensity score for attending important meetings is calculated to be 90%. This allows the server to quantify the user's probability of attending each meeting.

[0894] Score display

[0895] The server converts the calculated participation propensity score into a visually understandable format. Specifically, it generates code using HTML and JavaScript to display the score as a color code or a number. This score data is then sent to the calendar application for display.

[0896] Examples:

[0897] In a calendar app, the entry for User A's "Weekly Meeting" shows a "Participation Propensity Score of 40%." User B's "Important Project Meeting" shows a 90%, making it easy to see at a glance which users are likely to attend.

[0898] Meeting setup notifications

[0899] The device obtains the date and time and user list of the newly scheduled meeting, checks the existing attendance tendency score, and sends a notification recommending avoiding time slots with low scores. This allows meetings to be scheduled at appropriate time slots, improving the attendance rate of important meetings.

[0900] Examples:

[0901] When User C tries to schedule a new meeting, the calendar app recommends avoiding the "Weekly Meeting" time slot because it has a low score. This allows important meeting times to be adjusted to have a higher attendance rate.

[0902] Collecting user feedback

[0903] The server sends a request for feedback to users after the meeting or periodically. The server collects information from users about their reasons for attending or not attending meetings and their impressions of using the system using a web form or in-app feedback function. This feedback is used to retrain the machine learning model and improve the accuracy of predictions.

[0904] Examples:

[0905] If a user provides feedback such as "this meeting was important to me, but I didn't attend because it had a low score," the model can be adjusted based on that feedback, allowing the system to provide a more accurate attendance propensity score.

[0906] Prompt Sentence Examples

[0907] "Please look at User A's calendar and calculate and display this week's meeting attendance propensity score."

[0908] "When scheduling meetings, please remind me to avoid times with low attendance propensity scores."

[0909] This system makes it possible to efficiently adjust meeting schedules in a remote work environment, which is expected to improve productivity.

[0910] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0911] Step 1:

[0912] The server obtains the user's calendar information. To do this, it uses the Google Calendar API or the Microsoft Outlook Calendar API. The input is the user's access privileges to the calendar API, and the output is calendar entry data including the date, time, title, and attendee information of the user's appointments and meetings. This calendar entry data is then saved on the server.

[0913] Specific behavior:

[0914] The server sends an API request to retrieve the user's calendar information.

[0915] The server retrieves the calendar entries and stores them in a database.

[0916] Step 2:

[0917] The server collects past conference participation history. The input is calendar entries and participation history data, and the output is historical data including each user's attendance history, invitation acceptance rate, and web conference link click rate. This historical data is organized and stored on the server.

[0918] Specific behavior:

[0919] The server extracts past meeting data from the calendar entries.

[0920] The server calculates the user's attendance history, invitation acceptance rate, and web conference link click rate, and stores them in a database.

[0921] Step 3:

[0922] The server retrieves the user's job title and department information. The input is a database of job titles and departments within the company, and the output is the job title and department information for each user. This information is then stored in the integrated database.

[0923] Specific behavior:

[0924] The server accesses a personnel database within the company to obtain the user's job title and department information.

[0925] The server stores the acquired job title and department information in a database.

[0926] Step 4:

[0927] The server applies a machine learning model to the collected data. The input is calendar information, meeting attendance history, job title, and department information, and the output is the user's meeting attendance propensity score. Python's Scikit-learn and TensorFlow are used for data processing and learning.

[0928] Specific behavior:

[0929] The server extracts the necessary data from the database and performs preprocessing.

[0930] The server inputs data into the machine learning model, performs learning, and calculates the participation propensity score.

[0931] The server stores the calculated score in a database.

[0932] Step 5:

[0933] The server displays the predicted meeting attendance propensity score in a calendar application. The input is the attendance propensity score, and the output is a visually color-coded and / or numerically displayed score.

[0934] Specific behavior:

[0935] The server sends a request to the calendar application to display the score.

[0936] The calendar application displays the score in color.

[0937] Step 6:

[0938] The terminal sends a notification to the user recommending avoiding low-score time slots when setting up a meeting. The input is the date and time of the new meeting and the attendance tendency score, and the output is a notification to the user.

[0939] Specific behavior:

[0940] The terminal obtains the setting information for the new conference.

[0941] The device will check your existing participation propensity score and send you notifications recommending you avoid low-scoring times.

[0942] Step 7:

[0943] The server collects and analyzes user feedback, with the input being the user feedback data and the output being the analyzed feedback data, and then retrains the machine learning model based on this.

[0944] Specific behavior:

[0945] The server sends notifications to users to collect feedback.

[0946] The server collects feedback data from users and stores it in a database.

[0947] The server analyzes the feedback data and retrains the machine learning model.

[0948] (Application example 1)

[0949] 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."

[0950] In work environments such as remote work and factories, meeting time management is important, and there is a particular need to improve meeting attendance rates. However, it is difficult to predict the schedules and attendance tendencies of individual users and workers and then set optimal meeting times based on these. Furthermore, conventional systems have difficulty using user feedback to improve accuracy, leaving room for improvement. Therefore, there is a need for a system that can improve meeting attendance rates and optimize work efficiency.

[0951] 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.

[0952] In this invention, the server includes means for acquiring user schedule information, means for collecting past conference participation history, means for acquiring user job title and department information, means for applying a generative AI model to predict a user's conference participation tendency based on the collected data, means for displaying the predicted conference participation tendency score in a schedule management application, means for notifying users to avoid time periods with low scores when scheduling a conference, means for collecting user feedback and retraining the generative AI model, and means for supporting the efficiency of industrial machine maintenance and production planning meetings. This makes it possible to improve meeting attendance rates and optimize overall business efficiency.

[0953] "User schedule information" refers to data such as the date, time, location, and content of appointments and meetings set by the user.

[0954] "Conference participation history" refers to historical data of conferences that a user has attended in the past, and includes information such as whether or not the user attended, the frequency of attendance, and the attendance rate.

[0955] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to predict user behavior patterns and trends from data.

[0956] A "schedule management application" refers to a software application that helps users manage their schedules and meetings.

[0957] The "conference attendance propensity score" is a numerical representation of the likelihood that a user will attend a particular conference, and is calculated based on a prediction.

[0958] "Notification mechanism" refers to a digital messaging or alert system that notifies users of specific information or recommendations.

[0959] "Feedback" refers to opinions, impressions, and evaluations collected from users regarding their use of the system.

[0960] "Industrial machinery" refers to various machinery and equipment used in manufacturing and production industries, including those that require maintenance and management.

[0961] "Maintenance" refers to repairs and maintenance work carried out to keep industrial machinery in working order.

[0962] A "manufacturing planning meeting" refers to a meeting held to improve the efficiency of the production line and adjust schedules.

[0963] This invention is a system for streamlining time management in remote work environments, industrial machinery maintenance meetings, production planning meetings, etc. Specifically, it predicts a user's tendency to attend meetings based on the user's schedule information, meeting participation history, job title, and department information, and displays this in a schedule management application.

[0964] Program Description

[0965] The server collects users' schedule information, past meeting attendance history, job title, and department information. This data is used to apply a generative AI model to predict a meeting attendance propensity score. The score is displayed in the schedule management application, and notifications are sent when scheduling meetings to avoid low-scoring times. Furthermore, feedback from users can be collected to retrain the generative AI model and improve prediction accuracy. This can improve meeting attendance rates and optimize overall business efficiency.

[0966] Hardware and software used

[0967] 1. Hardware:

[0968] Factory Server

[0969] User's Digital Dashboard

[0970] 2. Software:

[0971] Python

[0972] Pandas (data processing library)

[0973] Scikit-learn (machine learning library)

[0974] The server retrieves users' schedule information, meeting attendance history, job title, and department information using a Python script that reads CSV files. It then applies a machine learning algorithm (here, a logistic regression model) to analyze the collected data. Using Python and Scikit-learn, the model is trained and a meeting attendance propensity score is calculated for each user.

[0975] The scores are then displayed on a digital dashboard using Pandas and related visualization libraries. The scores are visually represented using color coding and icons, providing a user-friendly format. The server also uses the scores to create meeting scheduling recommendations and sends them to the user's device. User feedback is also collected and used to retrain the generative AI model.

[0976] As a concrete example of data, the shift information and meeting attendance history of factory workers are shown below:

[0977] Shift Information:

[0978] employee_id, shift_hours, past_meeting_attendance

[0979] 1, 8, 0.75

[0980] 2, 6, 0.90

[0981] Meeting participation history:

[0982] employee_id, attendance_probability

[0983] 1, 0.80

[0984] 2, 0.95

[0985] Prompt Sentence Examples

[0986] For the generative AI model, enter the following prompt:

[0987] Please generate Python code that calculates the meeting attendance tendency score based on shift information and meeting attendance history, and sends notifications at specific times. The example data is as follows.

[0988] Shift Information:

[0989] employee_id, shift_hours, past_meeting_attendance

[0990] 1, 8, 0.75

[0991] 2, 6, 0.90

[0992] Meeting participation history:

[0993] employee_id, attendance_probability

[0994] 1, 0.80

[0995] 2, 0.95

[0996] The above configuration can optimize users' meeting schedules and significantly improve work efficiency in remote work and industrial environments.

[0997] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0998] Step 1: Data collection

[0999] The server reads user schedule information, past meeting attendance history, job title, and department information from CSV files. Specifically, it uses Python's Pandas library to read and integrate this data as a data frame. The input includes each user's shift information (employee_id, shift_hours, past_meeting_attendance) and meeting attendance history (employee_id, attendance_probability). The output is a dataset that integrates this data.

[1000] Step 2: Data analysis and score calculation

[1001] The server applies a generative AI model (logistic regression algorithm) to the collected data to predict each user's meeting attendance propensity score. The input includes shift information and meeting attendance history, and the output includes each user's meeting attendance propensity score. This score represents the probability of attendance, expressed as a number ranging from 0 to 1.

[1002] Step 3: View the score

[1003] The server visualizes the data to display the predicted meeting attendance propensity scores in the schedule management application. Specifically, the scores are visually displayed on a digital dashboard using color coding and icons. The input includes the meeting attendance propensity scores for each user, and the output includes the visually displayed scores.

[1004] Step 4: Notification of meeting settings

[1005] The server creates a recommendation notification to avoid low-scoring time slots when scheduling a meeting and sends it to the user's device. The input includes the predicted meeting attendance propensity score, and the output is the recommendation notification message. For example, it generates a message like "Employee ID: 1 should be scheduled for a meeting outside their current shift hours."

[1006] Step 5: Gather user feedback and retrain

[1007] The server collects user feedback after the meeting. The input includes user opinions and usage impressions, and the output is feedback data. This feedback is used to retrain the generative AI model to improve prediction accuracy. Specifically, the newly collected feedback data is added to the existing dataset, and the machine learning model is retrained.

[1008] Through the above processing steps, a system is realized that aims to improve meeting attendance rates and optimize business efficiency.

[1009] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1010] This invention relates to a system for streamlining the scheduling of web conferences in a remote work environment and improving productivity. In particular, by combining it with an emotion engine that recognizes user emotions, it more accurately predicts conference participation trends and supports optimal conference scheduling.

[1011] The system mainly includes the following functions:

[1012] 1. Data Collection

[1013] The server retrieves the user's calendar information via an API. It also collects the user's past meeting participation history and obtains data such as attendance history, invitation acceptance rate, and web conference link click rate. In addition, it retrieves the user's job title and department information from the HR system or database.

[1014] Examples:

[1015] User A has a "Weekly Meeting" scheduled for every Monday on his calendar, but he has only attended one of the past three meetings. This data is collected by the server. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[1016] 2. Emotional Data Collection

[1017] The device uses an emotion engine that recognizes the user's emotions in real time, acquiring emotion data and understanding the emotional state (e.g., stress, satisfaction, dissatisfaction) that influences participation tendencies.

[1018] Examples:

[1019] When User C attended a particular meeting, the emotion engine recognized high stress. This information is sent to the server and used to predict meeting attendance trends.

[1020] 3. Data analysis and score calculation

[1021] The server applies a machine learning model based on the collected calendar information, meeting participation history, job title and department information, and emotion data to predict each user's tendency to participate in meetings, thereby calculating a participation propensity score.

[1022] Examples:

[1023] User A's propensity score for attending the "Weekly Meeting" is calculated as 40% based on their emotional data and past history. This prediction takes into account their past participation frequency and emotional state.

[1024] 4. Score display

[1025] The server displays the calculated meeting attendance tendency score in the calendar app. The score is provided in a visually understandable format using color coding and numerical values. The server also displays the user's emotional state, allowing the user to share their state.

[1026] Examples:

[1027] In the calendar app, the entry for "Weekly Meeting" will show "Participation Propensity Score 40%" along with the stress level.

[1028] 5. Meeting setup notifications

[1029] The device provides notifications when setting up a meeting. When a user tries to set up a new meeting, the device sends a notification recommending the best time slot based on the participation propensity score and emotion data.

[1030] Examples:

[1031] If User C tries to schedule a new meeting, the calendar app will show that the "Weekly Meeting" time slot has a low score and a high stress level, and recommend that they avoid that time.

[1032] 6. Collecting User Feedback

[1033] The server collects feedback from users, including reasons for attending meetings, impressions of using the system, and opinions based on sentiment data.

[1034] Examples:

[1035] If a user provides feedback such as "This meeting was important, but I was too stressed and didn't attend," the model is retrained based on that feedback.

[1036] 7. Retraining the model

[1037] The server retrains the machine learning model based on the collected feedback, allowing the system to calculate a more accurate meeting attendance propensity score.

[1038] Through these functions, the system streamlines the adjustment of meeting schedules in a remote work environment, improving productivity. By taking into account the user's emotional state, it is possible to set up meetings that are easier to participate in, further improving the efficiency of overall meeting management.

[1039] The processing flow will be explained below.

[1040] Step 1:

[1041] The server obtains the user's calendar information. Specifically, it collects detailed information about the user's schedules and meetings (title, start time, end time, attendee list, frequency, etc.) through the API. This allows it to understand what meetings the user is participating in.

[1042] Step 2:

[1043] The server collects past conference participation history, including attendance history, invitation acceptance rates, and click rates for web conference links, allowing for analysis of participation trends for each conference.

[1044] Step 3:

[1045] The server retrieves the user's job title and department information. The server retrieves data about the user's job title and department from the HR system or database to clarify the user's role.

[1046] Step 4:

[1047] The device activates an emotion engine to recognize the user's emotions. The emotion engine acquires real-time emotion data from the user using technologies such as voice recognition and facial expression analysis.

[1048] Step 5:

[1049] The server receives the emotion data sent from the emotion engine and stores it in a database, where the emotional state (e.g., stress, satisfaction, dissatisfaction) is recorded.

[1050] Step 6:

[1051] The server preprocesses all data collected (calendar information, meeting attendance history, job title / department information, and emotion data) by cleaning the data, interpolating missing values, and removing outliers, and converting it into a format suitable for machine learning models.

[1052] Step 7:

[1053] The server applies machine learning models to predict each user's tendency to attend meetings, using models such as logistic regression and random forests, and calculates a participation propensity score that takes into account past participation history and emotional data.

[1054] Step 8:

[1055] The server calculates the meeting attendance tendency score and displays it in the calendar app. The score is presented in a visually easy-to-understand format using color coding and numerical values. The user's emotional state can also be displayed.

[1056] Step 9:

[1057] The device will notify users when a meeting is scheduled. When a user tries to schedule a new meeting, the device will send a notification recommending the best time based on attendance propensity scores and sentiment data. This will help ensure that important members are able to attend.

[1058] Step 10:

[1059] The server collects feedback from users, who provide reasons for attending meetings, their impressions of using the system, and opinions based on emotional data, and stores the feedback in a database.

[1060] Step 11:

[1061] The server retrains the machine learning model based on the collected feedback. By incorporating user feedback, the model's accuracy can be improved to provide a more accurate meeting attendance propensity score.

[1062] Through these steps, the system streamlines the scheduling of meetings in a remote work environment, improving user productivity. By taking emotional states into account, it becomes possible to set up meetings that are easier to participate in, further improving the efficiency of overall meeting management.

[1063] Example 2

[1064] 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."

[1065] In a remote work environment, conventional systems adjust the schedule of web conferences without taking into account the emotional state of the user, which has led to problems such as a decrease in conference participation rates and an increase in user stress. There is a need to solve these problems and realize optimal conference settings that reflect the emotional state of the user.

[1066] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1067] In this invention, the server includes means for acquiring user schedule information, means for collecting past meeting attendance history, means for acquiring user job title and department information, means for recognizing emotional states in real time and collecting data, means for predicting the user's tendency to attend meetings by applying a machine learning model based on the collected calendar information, meeting attendance history, job title and department information, and emotional data, means for displaying the predicted meeting attendance propensity score in a calendar application, means for notifying the user of time periods with low scores when scheduling a meeting, and means for collecting user feedback and retraining the machine learning model. This makes it possible to adjust an optimal meeting schedule taking into account the user's emotional state.

[1068] "User schedule information" refers to information about schedules and events that a user has registered in a calendar or schedule management application.

[1069] "Past meeting attendance history" refers to data such as attendance status of meetings a user has previously attended, invitation acceptance rate, and click rate of web conference links.

[1070] "User position and department information" is information about the user's job title and the department to which the user belongs, and is obtained from an HR system or database.

[1071] "Means for recognizing emotional states and collecting data in real time" refers to technology that uses built-in cameras and microphones to capture the user's facial expressions and tone of voice, and then analyzes them using an emotion engine.

[1072] "Emotion data" is data that represents the user's emotional state, such as stress, satisfaction, or dissatisfaction.

[1073] "Machine learning model" refers to an algorithm or statistical model used to predict users' meeting participation trends based on collected data.

[1074] The "meeting attendance propensity score" is a numerical value calculated based on a machine learning model that indicates the likelihood that a user will attend a particular meeting.

[1075] A "calendar application" is software that allows users to manage their schedules and has a function to display meeting attendance propensity scores.

[1076] "Means of notification" refers to technical means for notifying users of information or recommendations in real time, such as push notifications or alerts.

[1077] "Means for collecting feedback" refers to a function for collecting opinions and impressions from users and using them to improve the system.

[1078] "Retraining" is the process of updating a machine learning model based on collected feedback to improve its prediction accuracy.

[1079] This invention relates to a system that streamlines scheduling of web conferences in a remote work environment and improves productivity. In particular, by combining it with an emotion engine that recognizes user emotions, it more accurately predicts conference participation trends and supports optimal conference scheduling.

[1080] Data collection

[1081] First, the server obtains the user's schedule information. The server collects data from schedule management applications such as Google Calendar and Outlook Calendar via API. Next, the server accesses the HR system or internal database to obtain the user's job title and department information. At the same time, it collects the user's past meeting attendance history and stores it in the database.

[1082] For example, User A has a "Weekly Meeting" scheduled on his calendar, but he has only attended one of the past three meetings. This data is collected by the server. User B is a manager and tends to invite his subordinates to many meetings.

[1083] Emotional Data Collection

[1084] The device uses the built-in camera and microphone to recognize the user's emotional state in real time. It uses an emotion engine (e.g., Microsoft Azure's Emotion API) to capture the user's facial expressions and tone of voice and analyzes the emotional data, such as stress, satisfaction, and dissatisfaction. This information is sent to a server and stored in a database.

[1085] For example, when user C is attending a particular meeting, the emotion engine recognizes high stress and sends this information to the server.

[1086] Data analysis and score calculation

[1087] The server applies a machine learning model (e.g., scikit-learn or TensorFlow) based on the collected calendar information, meeting participation history, job title / department information, and emotion data to predict each user's tendency to participate in meetings. This calculates a participation propensity score, which is then stored in a database.

[1088] For example, the participation propensity score for user A in the "Weekly Meeting" is calculated as 40% based on emotional data and past history. This prediction takes into account the frequency of past participation and emotional state.

[1089] Score display

[1090] The server displays the calculated meeting attendance tendency score in a calendar application, which displays the score in a visually easy-to-understand format.

[1091] For example, an entry for "Weekly Meeting" will display "Participation Propensity Score 40%" along with the emotional state.

[1092] Meeting setup notifications

[1093] When a user attempts to schedule a new meeting, the device receives the score and emotion data from the server and generates a notification suggesting the optimal meeting time and structure, which is displayed to the user as a push notification or alert.

[1094] For example, if User C tries to schedule a new meeting, the calendar application may indicate that the "Weekly Meeting" time slot has a low score, indicating a high stress state, and recommend that they avoid that time.

[1095] Gathering user feedback and retraining the model

[1096] The server collects user opinions and feedback through the calendar application and a dedicated feedback form. This feedback is stored in a database and analyzed. Based on the collected feedback, the server retrains the machine learning model to improve prediction accuracy.

[1097] For example, if a user provides feedback such as "This meeting was important, but I was too stressed and didn't attend," the server uses this data to retrain the model.

[1098] Prompt Sentence Examples

[1099] "Calculate the attendance propensity score for the next weekly meeting based on the user's sentiment data and meeting attendance history."

[1100] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1101] Step 1:

[1102] Input: User's calendar information (Google Calendar, Outlook, etc.), HR system or database information (job title, department information), past meeting attendance history (attendance status, invitation acceptance rate, etc.)

[1103] Specific operation: The server collects the user's schedule information from Google Calendar or Outlook Calendar via API. The server then accesses the HR system or internal database to obtain the user's job title and department information. It also obtains the user's past meeting attendance history and stores it in the database.

[1104] Output: The retrieved calendar information, job title and department information, and meeting attendance history are saved in a database.

[1105] Step 2:

[1106] Input: User's facial expression and voice data

[1107] How it works: The device uses the built-in camera and microphone to capture the user's facial expressions and tone of voice in real time. An emotion engine (e.g., Microsoft Azure's Emotion API) is used to analyze the user's emotional state (stress, satisfaction, dissatisfaction, etc.) from the captured data. This information is then sent to a server.

[1108] Output: The analyzed emotion data is sent to the server and stored in a database.

[1109] Step 3:

[1110] Input: User's calendar information, job title and department information, past meeting attendance history, emotional data

[1111] Specific operation: The server integrates the collected data and uses a machine learning model (e.g., scikit-learn or TensorFlow) to predict users' meeting participation trends. Specifically, it analyzes each user's meeting participation patterns based on past data and calculates the participation propensity score for each meeting.

[1112] Output: The calculated meeting attendance propensity score is saved in the database.

[1113] Step 4:

[1114] Input: Meeting participation propensity score

[1115] Specific operation: The server sends the meeting attendance tendency score to the calendar application UI. The calendar application displays the received score using color coding and numerical values, providing the user with a visually easy-to-understand format.

[1116] Output: The meeting attendance propensity score is displayed in the user's calendar application.

[1117] Step 5:

[1118] Input: Meeting participation tendency score and emotion data from the server

[1119] Specific operation: When a user attempts to schedule a new meeting, the device receives the meeting participation propensity score and emotion data from the server. The device generates and displays a notification to the user suggesting the optimal meeting time and configuration.

[1120] Output: The user will be notified with a suggestion for the best meeting time.

[1121] Step 6:

[1122] Input: User feedback (opinions about meeting attendance, impressions on using the system, opinions based on emotional data)

[1123] What it does: The server collects user feedback through the calendar application and a dedicated feedback form, stores the collected feedback in a database, and formats it for analysis.

[1124] Output: The collected feedback is stored in a database.

[1125] Step 7:

[1126] Input: Collected feedback data

[1127] What happens: The server takes in new feedback data and combines it with existing training data to retrain the machine learning model and generate a more accurate meeting attendance propensity score.

[1128] Output: The updated machine learning model is used for the next prediction.

[1129] (Application example 2)

[1130] 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."

[1131] Improving the efficiency of store staff work methods and the working environment are important issues in today's brick-and-mortar stores. In particular, the impact of staff emotional states on work efficiency and customer service cannot be ignored. However, there is currently no way to understand in real time how staff are working, making it difficult to create optimal work schedules. Furthermore, there is no established method for appropriately avoiding work that is likely to overload staff. Furthermore, there is no efficient way to collect staff feedback and optimize work based on it. This increases staff stress, ultimately leading to a decrease in customer satisfaction.

[1132] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1133] In this invention, the server includes: means for acquiring a user's calendar information; means for collecting past meeting participation history; means for acquiring the user's job title and department information; means for acquiring emotional data and recognizing the user's state; means for applying a machine learning model to predict the user's tendency to attend meetings based on the collected data; means for displaying the predicted meeting participation propensity score in a calendar application; means for notifying the user to recommend a time slot with a low score when scheduling a meeting; means for collecting user feedback and retraining the machine learning model; and means for visually displaying the optimized work schedule and the impact of the emotional data. This makes it possible to grasp the current emotional state of store staff in real time and provide an optimal work schedule that takes their emotional state into account. Furthermore, by efficiently collecting staff feedback and continuously optimizing work based on it, an improved working environment and increased customer satisfaction can be achieved.

[1134] "User" refers to a person who uses the system.

[1135] "Calendar information" refers to information that records a user's schedule and plans.

[1136] "Conference participation history" refers to a record of conferences that a user has participated in in the past.

[1137] "Job title and department information" refers to information about the department and job to which the user belongs.

[1138] "Collected data" refers to calendar information, meeting participation history, job title and department information, emotional data, etc.

[1139] A "machine learning model" is an algorithm that learns patterns based on collected data and makes predictions and analyses.

[1140] A "calendar application" is software for managing a user's schedule.

[1141] "Emotional Data" refers to data that records and analyzes the emotional state of a user.

[1142] "User state" refers to the user's current emotional and psychological state.

[1143] "Work schedule" refers to the planned work to be performed by a user.

[1144] The "predicted conference attendance propensity score" is a score calculated by a machine learning model that indicates the likelihood of a user attending a conference.

[1145] "Notification" is a means of conveying information to the user through an interface.

[1146] "Feedback" refers to opinions and evaluations from users of the system.

[1147] An "optimized work schedule" refers to a work schedule that is adjusted taking into account the user's emotional state and work efficiency.

[1148] "Visual display means" refers to methods of visually presenting information to users using graphs, icons, color coding, etc.

[1149] This invention aims to apply a meeting schedule adjustment system in a remote work environment to optimizing work schedules in brick-and-mortar stores. The system utilizes user emotional data to grasp the real-time emotional state of staff and propose optimal work schedules.

[1150] The server first obtains the user's calendar information, including the user's past conference attendance history and current schedule. It also obtains the user's job title and department information to understand their work characteristics.

[1151] The device then collects the user's real-time emotional data, which is acquired through smart glasses or other wearable devices worn by the user, and uses this emotional data to measure the user's current stress level, satisfaction, dissatisfaction, etc.

[1152] The server applies machine learning models to the collected data to predict each user's work participation trends. Specifically, it integrates and analyzes past meeting history, job title and department information, and real-time emotion data. This analysis utilizes an emotion engine and schedule optimization algorithm to calculate the optimal schedule to maximize the user's work efficiency.

[1153] The optimized work schedule is displayed on the user's smart glasses or smartphone. The schedule is color-coded and displayed with icons for easy visual understanding. The impact of emotional data is also displayed visually, allowing staff to work while being aware of their own state.

[1154] Furthermore, the server collects user feedback and retrains the machine learning model based on it. The feedback includes opinions based on the results of applying the work schedule and emotional data. This allows the system to continuously improve and provide more accurate work schedules.

[1155] For example, if a staff member gives feedback that they felt high stress during their last work session, the system will use that information to adjust their next work schedule. Specifically, if the emotion engine detects that a staff member's current stress level is high, the system will prioritize relaxing tasks.

[1156] Below is an example of a prompt sentence.

[1157] Example prompt sentence:

[1158] “Sort the following tasks based on the optimal work schedule. Consider the emotional state of your staff and create a schedule that is least stressful.

[1159] Product display

[1160] Customer Service

[1161] Stock Check

[1162] Sales Report

[1163] User's current mild stress level is 8."

[1164] In this way, the present invention can be used in brick-and-mortar stores to improve the working environment for staff and increase work efficiency.

[1165] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1166] Step 1:

[1167] The server retrieves the user's calendar information.

[1168] Input: User ID

[1169] Output: User's calendar information (schedule and events)

[1170] What it does: The server retrieves the user's calendar information through an API request, providing data that includes the user's events and past meeting history.

[1171] Step 2:

[1172] The server retrieves the user's job title and department information.

[1173] Input: User ID

[1174] Output: User's job title and department information

[1175] What happens: The server sends an API request to an HR system or database to retrieve information about the user's job title and department.

[1176] Step 3:

[1177] The device collects the user's emotional data in real time.

[1178] Input: User ID, Emotion Engine

[1179] Output: User's emotional data (stress level, satisfaction, dissatisfaction, etc.)

[1180] Specific operation: The smart glasses or wearable device worn by the user uses an emotion engine to collect emotion data in real time and transmits the data to a server.

[1181] Step 4:

[1182] The server applies a machine learning model based on the collected data to predict the user's work participation tendencies.

[1183] Input: Calendar information, job title and department information, emotion data

[1184] Output: Work participation propensity score

[1185] Specific operation: The server integrates these data into a single dataset and applies machine learning algorithms (e.g., random forests, neural networks) to predict the user's work participation tendencies.

[1186] Step 5:

[1187] The server displays the predicted work attendance propensity score in a calendar application.

[1188] Input: Work participation propensity score

[1189] Output: Work participation propensity score displayed as a color or icon

[1190] Specific operation: The server incorporates the score into the calendar application and displays the results on the user's device in a visually easy-to-understand format (color coding and icons).

[1191] Step 6:

[1192] The device will notify you to recommend a time slot with a low score when setting up a meeting.

[1193] Input: Work participation propensity score

[1194] Output: Best time suggestion

[1195] Specific operation: The device sends a notification to the user, recommending that the user avoid times when the work attendance propensity score is low.

[1196] Step 7:

[1197] The server collects user feedback and retrains the machine learning model.

[1198] Input: User feedback

[1199] Output: An improved machine learning model

[1200] How it works: The server collects user feedback (e.g., "This time of day is stressful") and uses it to retrain the machine learning model to improve accuracy.

[1201] Step 8:

[1202] The server visually displays the optimized work schedule and the impact of emotional data.

[1203] Input: Optimized work schedule, emotional data

[1204] Output: Visually displayed work schedule and the impact of emotional data

[1205] Specific operation: The server visually displays the optimized schedule and emotion data in a calendar application or smart glasses, allowing users to work while being aware of their own state.

[1206] 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.

[1207] 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.

[1208] 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.

[1209] [Fourth embodiment]

[1210] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1211] 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.

[1212] 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).

[1213] 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.

[1214] 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.

[1215] 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).

[1216] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1217] 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.

[1218] 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.

[1219] 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.

[1220] 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.

[1221] 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.

[1222] 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."

[1223] This invention relates to a system for achieving efficient time management for web conferences in a remote work environment. Specifically, it supports appropriate conference scheduling by predicting a user's conference participation tendency based on the user's calendar information, conference participation history, job title, and department information, and displaying the score in a calendar app.

[1224] The system mainly includes the following functions:

[1225] 1. Data Collection

[1226] The server obtains the user's calendar information, collects the user's past meeting participation history, and collects data such as attendance history, invitation acceptance rate, and web conference link click rate, as well as the user's job title and department information.

[1227] Examples:

[1228] User A has a "Weekly Meeting" scheduled for every Monday on his calendar, but he has only attended one of the past three meetings. This data is collected on the server. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[1229] 2. Data analysis and score calculation

[1230] The server applies a machine learning model to the collected data to predict each user's tendency to attend meetings. This prediction can be made using algorithms such as logistic regression or random forest. As a result, a participation propensity score is calculated for each user.

[1231] Examples:

[1232] User A's propensity score for attending "Weekly Meetings" is calculated to be 40%. User B's propensity score for attending important meetings is calculated to be 90%. This allows the server to quantify the user's probability of attending each meeting.

[1233] 3. Score display

[1234] The server displays this participation propensity score in the calendar application, using color coding and numerical values ​​to provide a visually easy-to-understand score.

[1235] Examples:

[1236] In the calendar app, entries for "Weekly Meetings" will show a "Participation Propensity Score of 40%," while entries for "Important Project Meetings" will show a 90%, making it easy to see at a glance which users are likely to attend.

[1237] 4. Meeting setup notification

[1238] The device (e.g., the user's PC or smartphone) will send a notification recommending avoiding time slots with low scores based on the predicted score when setting up a meeting. This notification will help with meeting setup and participant adjustments.

[1239] Examples:

[1240] When User C tries to schedule a new meeting, the calendar app recommends avoiding the "Weekly Meeting" time slot because it has a low score. This allows important meeting times to be adjusted to have a higher attendance rate.

[1241] 5. Collecting User Feedback

[1242] The server collects user feedback, including reasons for attending meetings and user impressions of the system, and uses this feedback to retrain the machine learning model to improve prediction accuracy.

[1243] Examples:

[1244] If a user provides feedback such as "this meeting was important to me, but I didn't attend because it had a low score," the model can be adjusted based on that feedback, allowing the system to provide a more accurate attendance propensity score.

[1245] These features enable efficient adjustment of meeting schedules in a remote work environment, which is expected to improve productivity. The system recommends time slots based on users' attendance trends and optimizes important meetings to increase attendance rates, thereby streamlining overall meeting management.

[1246] The processing flow will be explained below.

[1247] Step 1:

[1248] The server retrieves the user's calendar information, specifically, details of the user's appointments and meetings (title, start time, end time, attendee list, frequency, etc.) through the API.

[1249] Step 2:

[1250] The server collects past conference participation history. The collected data includes attendance history for web conferences, invitation acceptance rates, and web conference link click rates. This allows for an understanding of conference participation trends.

[1251] Step 3:

[1252] The server retrieves the user's job title and department information. The server retrieves the user's job title and department data from the HR system or database to clarify the user's role.

[1253] Step 4:

[1254] The server preprocesses the collected data by cleaning it, interpolating missing values, removing outliers, and making it suitable for machine learning models.

[1255] Step 5:

[1256] The server applies machine learning models to predict each user's tendency to attend meetings, using models such as logistic regression and random forests to calculate a participation propensity score based on data such as past attendance history and job title.

[1257] Step 6:

[1258] The server calculates the meeting attendance tendency score and displays it in the calendar app. Specifically, the score is displayed numerically and color-coded on each meeting entry, allowing users to understand it visually.

[1259] Step 7:

[1260] The device will notify you when a meeting is scheduled. When a user tries to schedule a new meeting, the device will send a notification recommending the best time based on the user's participation propensity score. This will help ensure that important members are able to attend.

[1261] Step 8:

[1262] The server collects feedback from users and stores the feedback provided by users to the system (e.g., opinions on the importance of the meeting and their willingness to attend) in a database.

[1263] Step 9:

[1264] The server retrains the machine learning model based on the collected feedback. By incorporating user feedback, the model's accuracy can be improved to provide a more accurate meeting attendance propensity score.

[1265] Through these steps, this system streamlines meeting scheduling in a remote work environment and improves user productivity.

[1266] Example 1

[1267] 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."

[1268] In remote work environments, productivity needs to be improved by streamlining user meeting schedules and optimizing meeting scheduling based on attendance trends. However, conventional systems struggle to adjust schedules while fully considering users' job titles and past meeting participation history. Furthermore, they do not recommend avoiding time slots with low attendance propensity scores when scheduling meetings, resulting in low attendance rates for important meetings. Additionally, they lacked a way to visually display meeting attendance propensity scores or a way to retrain machine learning models based on user feedback.

[1269] 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.

[1270] In this invention, the server includes means for acquiring user calendar information, means for collecting past conference participation history, means for acquiring user job title and department information, means for applying a machine learning model to predict the user's conference participation propensity based on the collected data, means for displaying the predicted conference participation propensity score in a calendar application, means for notifying the user to recommend a time slot with a low score when scheduling a conference, means for collecting user feedback and retraining the machine learning model, and means for reflecting the calculation results of the conference participation propensity score in actual conference scheduling and schedule management. This enables optimal schedule adjustment based on the user's conference participation propensity, which is expected to improve the attendance rate for important meetings and increase overall productivity.

[1271] "Calendar information" is digital schedule data including the user's plans, meeting dates and times, titles, participant information, and the like.

[1272] "Conference participation history" is data on conferences that a user has attended in the past, and includes information such as attendance history, invitation acceptance rate, and click rate of web conference links.

[1273] "Position and department information" is data relating to the position name and department name within the organization to which the user belongs.

[1274] A "machine learning model" is an algorithm or method used to analyze collected data and predict users' meeting participation trends.

[1275] A "meeting attendance propensity score" is a numerical value calculated using a machine learning model that indicates the probability that a user will attend a particular meeting.

[1276] A "calendar application" is software or a system that allows a user to manage their digital schedule and view appointments and meetings.

[1277] "Notification" is a means of sending alerts or messages to users to convey specific information.

[1278] "User feedback" refers to data on opinions and evaluations provided by users, such as reasons for attending or not attending a meeting and impressions of using the system.

[1279] "Retraining" is the process of improving the accuracy of a machine learning model based on new data collected.

[1280] "Visual display of scores" refers to a method of displaying the meeting attendance tendency scores in an easy-to-understand manner using color coding and icons.

[1281] This system is designed to streamline time management for web conferences in remote work environments. Specifically, the server collects users' calendar information, meeting participation history, job title, and department information, and then applies a machine learning model based on this information to predict the user's tendency to attend meetings. Furthermore, this participation tendency score is displayed in the calendar application, and when scheduling a meeting, a notification is sent recommending avoiding time slots with low scores, thereby helping to schedule appropriate meetings. Finally, user feedback is collected and the machine learning model is retrained to improve prediction accuracy. The specific implementation steps and the hardware and software used are described below.

[1282] Data collection

[1283] The server obtains the user's calendar information. Specifically, it uses the Google Calendar API or Microsoft Outlook Calendar API to collect each user's calendar entries along with the date, time, title, and participant information. The server also collects data such as past meeting participation history, attendance history, invitation acceptance rate, and web conference link click rate. It also obtains the user's job title and department information.

[1284] Examples:

[1285] User A has a "Weekly Meeting" scheduled for every Monday in his Google Calendar, but he has only attended one of the past three meetings. This data is collected on the server using the Google Calendar API. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[1286] Data analysis and score calculation

[1287] The server applies a machine learning model to the data collected. Specific algorithms include logistic regression and random forest, and Python libraries such as Scikit-learn and TensorFlow are used. This predicts each user's tendency to attend meetings and calculates a participation propensity score.

[1288] Examples:

[1289] User A's propensity score for attending "Weekly Meetings" is calculated to be 40%. User B's propensity score for attending important meetings is calculated to be 90%. This allows the server to quantify the user's probability of attending each meeting.

[1290] Score display

[1291] The server converts the calculated participation propensity score into a visually understandable format. Specifically, it generates code using HTML and JavaScript to display the score as a color code or a number. This score data is then sent to the calendar application for display.

[1292] Examples:

[1293] In a calendar app, the entry for User A's "Weekly Meeting" shows a "Participation Propensity Score of 40%." User B's "Important Project Meeting" shows a 90%, making it easy to see at a glance which users are likely to attend.

[1294] Meeting setup notifications

[1295] The device obtains the date and time and user list of the newly scheduled meeting, checks the existing attendance tendency score, and sends a notification recommending avoiding time slots with low scores. This allows meetings to be scheduled at appropriate time slots, improving the attendance rate of important meetings.

[1296] Examples:

[1297] When User C tries to schedule a new meeting, the calendar app recommends avoiding the "Weekly Meeting" time slot because it has a low score. This allows important meeting times to be adjusted to have a higher attendance rate.

[1298] Collecting user feedback

[1299] The server sends a request for feedback to users after the meeting or periodically. The server collects information from users about their reasons for attending or not attending meetings and their impressions of using the system using a web form or in-app feedback function. This feedback is used to retrain the machine learning model and improve the accuracy of predictions.

[1300] Examples:

[1301] If a user provides feedback such as "this meeting was important to me, but I didn't attend because it had a low score," the model can be adjusted based on that feedback, allowing the system to provide a more accurate attendance propensity score.

[1302] Prompt Sentence Examples

[1303] "Please look at User A's calendar and calculate and display this week's meeting attendance propensity score."

[1304] "When scheduling meetings, please remind me to avoid times with low attendance propensity scores."

[1305] This system makes it possible to efficiently adjust meeting schedules in a remote work environment, which is expected to improve productivity.

[1306] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1307] Step 1:

[1308] The server obtains the user's calendar information. To do this, it uses the Google Calendar API or the Microsoft Outlook Calendar API. The input is the user's access privileges to the calendar API, and the output is calendar entry data including the date, time, title, and attendee information of the user's appointments and meetings. This calendar entry data is then saved on the server.

[1309] Specific behavior:

[1310] The server sends an API request to retrieve the user's calendar information.

[1311] The server retrieves the calendar entries and stores them in a database.

[1312] Step 2:

[1313] The server collects past conference participation history. The input is calendar entries and participation history data, and the output is historical data including each user's attendance history, invitation acceptance rate, and web conference link click rate. This historical data is organized and stored on the server.

[1314] Specific behavior:

[1315] The server extracts past meeting data from the calendar entries.

[1316] The server calculates the user's attendance history, invitation acceptance rate, and web conference link click rate, and stores them in a database.

[1317] Step 3:

[1318] The server retrieves the user's job title and department information. The input is a database of job titles and departments within the company, and the output is the job title and department information for each user. This information is then stored in the integrated database.

[1319] Specific behavior:

[1320] The server accesses a personnel database within the company to obtain the user's job title and department information.

[1321] The server stores the acquired job title and department information in a database.

[1322] Step 4:

[1323] The server applies a machine learning model to the collected data. The input is calendar information, meeting attendance history, job title, and department information, and the output is the user's meeting attendance propensity score. Python's Scikit-learn and TensorFlow are used for data processing and learning.

[1324] Specific behavior:

[1325] The server extracts the necessary data from the database and performs preprocessing.

[1326] The server inputs data into the machine learning model, performs learning, and calculates the participation propensity score.

[1327] The server stores the calculated score in a database.

[1328] Step 5:

[1329] The server displays the predicted meeting attendance propensity score in a calendar application. The input is the attendance propensity score, and the output is a visually color-coded and / or numerically displayed score.

[1330] Specific behavior:

[1331] The server sends a request to the calendar application to display the score.

[1332] The calendar application displays the score in color.

[1333] Step 6:

[1334] The terminal sends a notification to the user recommending avoiding low-score time slots when setting up a meeting. The input is the date and time of the new meeting and the attendance tendency score, and the output is a notification to the user.

[1335] Specific behavior:

[1336] The terminal obtains the setting information for the new conference.

[1337] The device will check your existing participation propensity score and send you notifications recommending you avoid low-scoring times.

[1338] Step 7:

[1339] The server collects and analyzes user feedback, with the input being the user feedback data and the output being the analyzed feedback data, and then retrains the machine learning model based on this.

[1340] Specific behavior:

[1341] The server sends notifications to users to collect feedback.

[1342] The server collects feedback data from users and stores it in a database.

[1343] The server analyzes the feedback data and retrains the machine learning model.

[1344] (Application example 1)

[1345] 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."

[1346] In work environments such as remote work and factories, meeting time management is important, and there is a particular need to improve meeting attendance rates. However, it is difficult to predict the schedules and attendance tendencies of individual users and workers and then set optimal meeting times based on these. Furthermore, conventional systems have difficulty using user feedback to improve accuracy, leaving room for improvement. Therefore, there is a need for a system that can improve meeting attendance rates and optimize work efficiency.

[1347] 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.

[1348] In this invention, the server includes means for acquiring user schedule information, means for collecting past conference participation history, means for acquiring user job title and department information, means for applying a generative AI model to predict a user's conference participation tendency based on the collected data, means for displaying the predicted conference participation tendency score in a schedule management application, means for notifying users to avoid time periods with low scores when scheduling a conference, means for collecting user feedback and retraining the generative AI model, and means for supporting the efficiency of industrial machine maintenance and production planning meetings. This makes it possible to improve meeting attendance rates and optimize overall business efficiency.

[1349] "User schedule information" refers to data such as the date, time, location, and content of appointments and meetings set by the user.

[1350] "Conference participation history" refers to historical data of conferences that a user has attended in the past, and includes information such as whether or not the user attended, the frequency of attendance, and the attendance rate.

[1351] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to predict user behavior patterns and trends from data.

[1352] A "schedule management application" refers to a software application that helps users manage their schedules and meetings.

[1353] The "conference attendance propensity score" is a numerical representation of the likelihood that a user will attend a particular conference, and is calculated based on a prediction.

[1354] "Notification mechanism" refers to a digital messaging or alert system that notifies users of specific information or recommendations.

[1355] "Feedback" refers to opinions, impressions, and evaluations collected from users regarding their use of the system.

[1356] "Industrial machinery" refers to various machinery and equipment used in manufacturing and production industries, including those that require maintenance and management.

[1357] "Maintenance" refers to repairs and maintenance work carried out to keep industrial machinery in working order.

[1358] A "manufacturing planning meeting" refers to a meeting held to improve the efficiency of the production line and adjust schedules.

[1359] This invention is a system for streamlining time management in remote work environments, industrial machinery maintenance meetings, production planning meetings, etc. Specifically, it predicts a user's tendency to attend meetings based on the user's schedule information, meeting participation history, job title, and department information, and displays this in a schedule management application.

[1360] Program Description

[1361] The server collects users' schedule information, past meeting attendance history, job title, and department information. This data is used to apply a generative AI model to predict a meeting attendance propensity score. The score is displayed in the schedule management application, and notifications are sent when scheduling meetings to avoid low-scoring times. Furthermore, feedback from users can be collected to retrain the generative AI model and improve prediction accuracy. This can improve meeting attendance rates and optimize overall business efficiency.

[1362] Hardware and software used

[1363] 1. Hardware:

[1364] Factory Server

[1365] User's Digital Dashboard

[1366] 2. Software:

[1367] Python

[1368] Pandas (data processing library)

[1369] Scikit-learn (machine learning library)

[1370] The server retrieves users' schedule information, meeting attendance history, job title, and department information using a Python script that reads CSV files. It then applies a machine learning algorithm (here, a logistic regression model) to analyze the collected data. Using Python and Scikit-learn, the model is trained and a meeting attendance propensity score is calculated for each user.

[1371] The scores are then displayed on a digital dashboard using Pandas and related visualization libraries. The scores are visually represented using color coding and icons, providing a user-friendly format. The server also uses the scores to create meeting scheduling recommendations and sends them to the user's device. User feedback is also collected and used to retrain the generative AI model.

[1372] As a concrete example of data, the shift information and meeting attendance history of factory workers are shown below:

[1373] Shift Information:

[1374] employee_id, shift_hours, past_meeting_attendance

[1375] 1, 8, 0.75

[1376] 2, 6, 0.90

[1377] Meeting participation history:

[1378] employee_id, attendance_probability

[1379] 1, 0.80

[1380] 2, 0.95

[1381] Prompt Sentence Examples

[1382] For the generative AI model, enter the following prompt:

[1383] Please generate Python code that calculates the meeting attendance tendency score based on shift information and meeting attendance history, and sends notifications at specific times. The example data is as follows.

[1384] Shift Information:

[1385] employee_id, shift_hours, past_meeting_attendance

[1386] 1, 8, 0.75

[1387] 2, 6, 0.90

[1388] Meeting participation history:

[1389] employee_id, attendance_probability

[1390] 1, 0.80

[1391] 2, 0.95

[1392] The above configuration can optimize users' meeting schedules and significantly improve work efficiency in remote work and industrial environments.

[1393] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1394] Step 1: Data collection

[1395] The server reads user schedule information, past meeting attendance history, job title, and department information from CSV files. Specifically, it uses Python's Pandas library to read and integrate this data as a data frame. The input includes each user's shift information (employee_id, shift_hours, past_meeting_attendance) and meeting attendance history (employee_id, attendance_probability). The output is a dataset that integrates this data.

[1396] Step 2: Data analysis and score calculation

[1397] The server applies a generative AI model (logistic regression algorithm) to the collected data to predict each user's meeting attendance propensity score. The input includes shift information and meeting attendance history, and the output includes each user's meeting attendance propensity score. This score represents the probability of attendance, expressed as a number ranging from 0 to 1.

[1398] Step 3: View the score

[1399] The server visualizes the data to display the predicted meeting attendance propensity scores in the schedule management application. Specifically, the scores are visually displayed on a digital dashboard using color coding and icons. The input includes the meeting attendance propensity scores for each user, and the output includes the visually displayed scores.

[1400] Step 4: Notification of meeting settings

[1401] The server creates a recommendation notification to avoid low-scoring time slots when scheduling a meeting and sends it to the user's device. The input includes the predicted meeting attendance propensity score, and the output is the recommendation notification message. For example, it generates a message like "Employee ID: 1 should be scheduled for a meeting outside their current shift hours."

[1402] Step 5: Gather user feedback and retrain

[1403] The server collects user feedback after the meeting. The input includes user opinions and usage impressions, and the output is feedback data. This feedback is used to retrain the generative AI model to improve prediction accuracy. Specifically, the newly collected feedback data is added to the existing dataset, and the machine learning model is retrained.

[1404] Through the above processing steps, a system is realized that aims to improve meeting attendance rates and optimize business efficiency.

[1405] 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.

[1406] This invention relates to a system for streamlining the scheduling of web conferences in a remote work environment and improving productivity. In particular, by combining it with an emotion engine that recognizes user emotions, it more accurately predicts conference participation trends and supports optimal conference scheduling.

[1407] The system mainly includes the following functions:

[1408] 1. Data Collection

[1409] The server retrieves the user's calendar information via an API. It also collects the user's past meeting participation history and obtains data such as attendance history, invitation acceptance rate, and web conference link click rate. In addition, it retrieves the user's job title and department information from the HR system or database.

[1410] Examples:

[1411] User A has a "Weekly Meeting" scheduled for every Monday on his calendar, but he has only attended one of the past three meetings. This data is collected by the server. User B also attends many meetings, but tends to delegate certain important meetings to subordinates.

[1412] 2. Emotional Data Collection

[1413] The device uses an emotion engine that recognizes the user's emotions in real time, acquiring emotion data and understanding the emotional state (e.g., stress, satisfaction, dissatisfaction) that influences participation tendencies.

[1414] Examples:

[1415] When User C attended a particular meeting, the emotion engine recognized high stress. This information is sent to the server and used to predict meeting attendance trends.

[1416] 3. Data analysis and score calculation

[1417] The server applies a machine learning model based on the collected calendar information, meeting participation history, job title and department information, and emotion data to predict each user's tendency to participate in meetings, thereby calculating a participation propensity score.

[1418] Examples:

[1419] User A's propensity score for attending the "Weekly Meeting" is calculated as 40% based on their emotional data and past history. This prediction takes into account their past participation frequency and emotional state.

[1420] 4. Score display

[1421] The server displays the calculated meeting attendance tendency score in the calendar app. The score is provided in a visually understandable format using color coding and numerical values. The server also displays the user's emotional state, allowing the user to share their state.

[1422] Examples:

[1423] In the calendar app, the entry for "Weekly Meeting" will show "Participation Propensity Score 40%" along with the stress level.

[1424] 5. Meeting setup notifications

[1425] The device provides notifications when setting up a meeting. When a user tries to set up a new meeting, the device sends a notification recommending the best time slot based on the participation propensity score and emotion data.

[1426] Examples:

[1427] If User C tries to schedule a new meeting, the calendar app will show that the "Weekly Meeting" time slot has a low score and a high stress level, and recommend that they avoid that time.

[1428] 6. Collecting User Feedback

[1429] The server collects feedback from users, including reasons for attending meetings, impressions of using the system, and opinions based on sentiment data.

[1430] Examples:

[1431] If a user provides feedback such as "This meeting was important, but I was too stressed and didn't attend," the model is retrained based on that feedback.

[1432] 7. Retraining the model

[1433] The server retrains the machine learning model based on the collected feedback, allowing the system to calculate a more accurate meeting attendance propensity score.

[1434] Through these functions, the system streamlines the adjustment of meeting schedules in a remote work environment, improving productivity. By taking into account the user's emotional state, it is possible to set up meetings that are easier to participate in, further improving the efficiency of overall meeting management.

[1435] The processing flow will be explained below.

[1436] Step 1:

[1437] The server obtains the user's calendar information. Specifically, it collects detailed information about the user's schedules and meetings (title, start time, end time, attendee list, frequency, etc.) through the API. This allows it to understand what meetings the user is participating in.

[1438] Step 2:

[1439] The server collects past conference participation history, including attendance history, invitation acceptance rates, and click rates for web conference links, allowing for analysis of participation trends for each conference.

[1440] Step 3:

[1441] The server retrieves the user's job title and department information. The server retrieves data about the user's job title and department from the HR system or database to clarify the user's role.

[1442] Step 4:

[1443] The device activates an emotion engine to recognize the user's emotions. The emotion engine acquires real-time emotion data from the user using technologies such as voice recognition and facial expression analysis.

[1444] Step 5:

[1445] The server receives the emotion data sent from the emotion engine and stores it in a database, where the emotional state (e.g., stress, satisfaction, dissatisfaction) is recorded.

[1446] Step 6:

[1447] The server preprocesses all data collected (calendar information, meeting attendance history, job title / department information, and emotion data) by cleaning the data, interpolating missing values, and removing outliers, and converting it into a format suitable for machine learning models.

[1448] Step 7:

[1449] The server applies machine learning models to predict each user's tendency to attend meetings, using models such as logistic regression and random forests, and calculates a participation propensity score that takes into account past participation history and emotional data.

[1450] Step 8:

[1451] The server calculates the meeting attendance tendency score and displays it in the calendar app. The score is presented in a visually easy-to-understand format using color coding and numerical values. The user's emotional state can also be displayed.

[1452] Step 9:

[1453] The device will notify users when a meeting is scheduled. When a user tries to schedule a new meeting, the device will send a notification recommending the best time based on attendance propensity scores and sentiment data. This will help ensure that important members are able to attend.

[1454] Step 10:

[1455] The server collects feedback from users, who provide reasons for attending meetings, their impressions of using the system, and opinions based on emotional data, and stores the feedback in a database.

[1456] Step 11:

[1457] The server retrains the machine learning model based on the collected feedback. By incorporating user feedback, the model's accuracy can be improved to provide a more accurate meeting attendance propensity score.

[1458] Through these steps, the system streamlines the scheduling of meetings in a remote work environment, improving user productivity. By taking emotional states into account, it becomes possible to set up meetings that are easier to participate in, further improving the efficiency of overall meeting management.

[1459] Example 2

[1460] 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."

[1461] In a remote work environment, conventional systems adjust the schedule of web conferences without taking into account the emotional state of the user, which has led to problems such as a decrease in conference participation rates and an increase in user stress. There is a need to solve these problems and realize optimal conference settings that reflect the emotional state of the user.

[1462] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1463] In this invention, the server includes means for acquiring user schedule information, means for collecting past meeting attendance history, means for acquiring user job title and department information, means for recognizing emotional states in real time and collecting data, means for predicting the user's tendency to attend meetings by applying a machine learning model based on the collected calendar information, meeting attendance history, job title and department information, and emotional data, means for displaying the predicted meeting attendance propensity score in a calendar application, means for notifying the user of time periods with low scores when scheduling a meeting, and means for collecting user feedback and retraining the machine learning model. This makes it possible to adjust an optimal meeting schedule taking into account the user's emotional state.

[1464] "User schedule information" refers to information about schedules and events that a user has registered in a calendar or schedule management application.

[1465] "Past meeting attendance history" refers to data such as attendance status of meetings a user has previously attended, invitation acceptance rate, and click rate of web conference links.

[1466] "User position and department information" is information about the user's job title and the department to which the user belongs, and is obtained from an HR system or database.

[1467] "Means for recognizing emotional states and collecting data in real time" refers to technology that uses built-in cameras and microphones to capture the user's facial expressions and tone of voice, and then analyzes them using an emotion engine.

[1468] "Emotion data" is data that represents the user's emotional state, such as stress, satisfaction, or dissatisfaction.

[1469] "Machine learning model" refers to an algorithm or statistical model used to predict users' meeting participation trends based on collected data.

[1470] The "meeting attendance propensity score" is a numerical value calculated based on a machine learning model that indicates the likelihood that a user will attend a particular meeting.

[1471] A "calendar application" is software that allows users to manage their schedules and has a function to display meeting attendance propensity scores.

[1472] "Means of notification" refers to technical means for notifying users of information or recommendations in real time, such as push notifications or alerts.

[1473] "Means for collecting feedback" refers to a function for collecting opinions and impressions from users and using them to improve the system.

[1474] "Retraining" is the process of updating a machine learning model based on collected feedback to improve its prediction accuracy.

[1475] This invention relates to a system that streamlines scheduling of web conferences in a remote work environment and improves productivity. In particular, by combining it with an emotion engine that recognizes user emotions, it more accurately predicts conference participation trends and supports optimal conference scheduling.

[1476] Data collection

[1477] First, the server obtains the user's schedule information. The server collects data from schedule management applications such as Google Calendar and Outlook Calendar via API. Next, the server accesses the HR system or internal database to obtain the user's job title and department information. At the same time, it collects the user's past meeting attendance history and stores it in the database.

[1478] For example, User A has a "Weekly Meeting" scheduled on his calendar, but he has only attended one of the past three meetings. This data is collected by the server. User B is a manager and tends to invite his subordinates to many meetings.

[1479] Emotional Data Collection

[1480] The device uses the built-in camera and microphone to recognize the user's emotional state in real time. It uses an emotion engine (e.g., Microsoft Azure's Emotion API) to capture the user's facial expressions and tone of voice and analyzes the emotional data, such as stress, satisfaction, and dissatisfaction. This information is sent to a server and stored in a database.

[1481] For example, when user C is attending a particular meeting, the emotion engine recognizes high stress and sends this information to the server.

[1482] Data analysis and score calculation

[1483] The server applies a machine learning model (e.g., scikit-learn or TensorFlow) based on the collected calendar information, meeting participation history, job title / department information, and emotion data to predict each user's tendency to participate in meetings. This calculates a participation propensity score, which is then stored in a database.

[1484] For example, the participation propensity score for user A in the "Weekly Meeting" is calculated as 40% based on emotional data and past history. This prediction takes into account the frequency of past participation and emotional state.

[1485] Score display

[1486] The server displays the calculated meeting attendance tendency score in a calendar application, which displays the score in a visually easy-to-understand format.

[1487] For example, an entry for "Weekly Meeting" will display "Participation Propensity Score 40%" along with the emotional state.

[1488] Meeting setup notifications

[1489] When a user attempts to schedule a new meeting, the device receives the score and emotion data from the server and generates a notification suggesting the optimal meeting time and structure, which is displayed to the user as a push notification or alert.

[1490] For example, if User C tries to schedule a new meeting, the calendar application may indicate that the "Weekly Meeting" time slot has a low score, indicating a high stress state, and recommend that they avoid that time.

[1491] Gathering user feedback and retraining the model

[1492] The server collects user opinions and feedback through the calendar application and a dedicated feedback form. This feedback is stored in a database and analyzed. Based on the collected feedback, the server retrains the machine learning model to improve prediction accuracy.

[1493] For example, if a user provides feedback such as "This meeting was important, but I was too stressed and didn't attend," the server uses this data to retrain the model.

[1494] Prompt Sentence Examples

[1495] "Calculate the attendance propensity score for the next weekly meeting based on the user's sentiment data and meeting attendance history."

[1496] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1497] Step 1:

[1498] Input: User's calendar information (Google Calendar, Outlook, etc.), HR system or database information (job title, department information), past meeting attendance history (attendance status, invitation acceptance rate, etc.)

[1499] Specific operation: The server collects the user's schedule information from Google Calendar or Outlook Calendar via API. The server then accesses the HR system or internal database to obtain the user's job title and department information. It also obtains the user's past meeting attendance history and stores it in the database.

[1500] Output: The retrieved calendar information, job title and department information, and meeting attendance history are saved in a database.

[1501] Step 2:

[1502] Input: User's facial expression and voice data

[1503] How it works: The device uses the built-in camera and microphone to capture the user's facial expressions and tone of voice in real time. An emotion engine (e.g., Microsoft Azure's Emotion API) is used to analyze the user's emotional state (stress, satisfaction, dissatisfaction, etc.) from the captured data. This information is then sent to a server.

[1504] Output: The analyzed emotion data is sent to the server and stored in a database.

[1505] Step 3:

[1506] Input: User's calendar information, job title and department information, past meeting attendance history, emotional data

[1507] Specific operation: The server integrates the collected data and uses a machine learning model (e.g., scikit-learn or TensorFlow) to predict users' meeting participation trends. Specifically, it analyzes each user's meeting participation patterns based on past data and calculates the participation propensity score for each meeting.

[1508] Output: The calculated meeting attendance propensity score is saved in the database.

[1509] Step 4:

[1510] Input: Meeting participation propensity score

[1511] Specific operation: The server sends the meeting attendance tendency score to the calendar application UI. The calendar application displays the received score using color coding and numerical values, providing the user with a visually easy-to-understand format.

[1512] Output: The meeting attendance propensity score is displayed in the user's calendar application.

[1513] Step 5:

[1514] Input: Meeting participation tendency score and emotion data from the server

[1515] Specific operation: When a user attempts to schedule a new meeting, the device receives the meeting participation propensity score and emotion data from the server. The device generates and displays a notification to the user suggesting the optimal meeting time and configuration.

[1516] Output: The user will be notified with a suggestion for the best meeting time.

[1517] Step 6:

[1518] Input: User feedback (opinions about meeting attendance, impressions on using the system, opinions based on emotional data)

[1519] What it does: The server collects user feedback through the calendar application and a dedicated feedback form, stores the collected feedback in a database, and formats it for analysis.

[1520] Output: The collected feedback is stored in a database.

[1521] Step 7:

[1522] Input: Collected feedback data

[1523] What happens: The server takes in new feedback data and combines it with existing training data to retrain the machine learning model and generate a more accurate meeting attendance propensity score.

[1524] Output: The updated machine learning model is used for the next prediction.

[1525] (Application example 2)

[1526] 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."

[1527] Improving the efficiency of store staff work methods and the working environment are important issues in today's brick-and-mortar stores. In particular, the impact of staff emotional states on work efficiency and customer service cannot be ignored. However, there is currently no way to understand in real time how staff are working, making it difficult to create optimal work schedules. Furthermore, there is no established method for appropriately avoiding work that is likely to overload staff. Furthermore, there is no efficient way to collect staff feedback and optimize work based on it. This increases staff stress, ultimately leading to a decrease in customer satisfaction.

[1528] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1529] In this invention, the server includes: means for acquiring a user's calendar information; means for collecting past meeting participation history; means for acquiring the user's job title and department information; means for acquiring emotional data and recognizing the user's state; means for applying a machine learning model to predict the user's tendency to attend meetings based on the collected data; means for displaying the predicted meeting participation propensity score in a calendar application; means for notifying the user to recommend a time slot with a low score when scheduling a meeting; means for collecting user feedback and retraining the machine learning model; and means for visually displaying the optimized work schedule and the impact of the emotional data. This makes it possible to grasp the current emotional state of store staff in real time and provide an optimal work schedule that takes their emotional state into account. Furthermore, by efficiently collecting staff feedback and continuously optimizing work based on it, an improved working environment and increased customer satisfaction can be achieved.

[1530] "User" refers to a person who uses the system.

[1531] "Calendar information" refers to information that records a user's schedule and plans.

[1532] "Conference participation history" refers to a record of conferences that a user has participated in in the past.

[1533] "Job title and department information" refers to information about the department and job to which the user belongs.

[1534] "Collected data" refers to calendar information, meeting participation history, job title and department information, emotional data, etc.

[1535] A "machine learning model" is an algorithm that learns patterns based on collected data and makes predictions and analyses.

[1536] A "calendar application" is software for managing a user's schedule.

[1537] "Emotional Data" refers to data that records and analyzes the emotional state of a user.

[1538] "User state" refers to the user's current emotional and psychological state.

[1539] "Work schedule" refers to the planned work to be performed by a user.

[1540] The "predicted conference attendance propensity score" is a score calculated by a machine learning model that indicates the likelihood of a user attending a conference.

[1541] "Notification" is a means of conveying information to the user through an interface.

[1542] "Feedback" refers to opinions and evaluations from users of the system.

[1543] An "optimized work schedule" refers to a work schedule that is adjusted taking into account the user's emotional state and work efficiency.

[1544] "Visual display means" refers to methods of visually presenting information to users using graphs, icons, color coding, etc.

[1545] This invention aims to apply a meeting schedule adjustment system in a remote work environment to optimizing work schedules in brick-and-mortar stores. The system utilizes user emotional data to grasp the real-time emotional state of staff and propose optimal work schedules.

[1546] The server first obtains the user's calendar information, including the user's past conference attendance history and current schedule. It also obtains the user's job title and department information to understand their work characteristics.

[1547] The device then collects the user's real-time emotional data, which is acquired through smart glasses or other wearable devices worn by the user, and uses this emotional data to measure the user's current stress level, satisfaction, dissatisfaction, etc.

[1548] The server applies machine learning models to the collected data to predict each user's work participation trends. Specifically, it integrates and analyzes past meeting history, job title and department information, and real-time emotion data. This analysis utilizes an emotion engine and schedule optimization algorithm to calculate the optimal schedule to maximize the user's work efficiency.

[1549] The optimized work schedule is displayed on the user's smart glasses or smartphone. The schedule is color-coded and displayed with icons for easy visual understanding. The impact of emotional data is also displayed visually, allowing staff to work while being aware of their own state.

[1550] Furthermore, the server collects user feedback and retrains the machine learning model based on it. The feedback includes opinions based on the results of applying the work schedule and emotional data. This allows the system to continuously improve and provide more accurate work schedules.

[1551] For example, if a staff member gives feedback that they felt high stress during their last work session, the system will use that information to adjust their next work schedule. Specifically, if the emotion engine detects that a staff member's current stress level is high, the system will prioritize relaxing tasks.

[1552] Below is an example of a prompt sentence.

[1553] Example prompt sentence:

[1554] “Sort the following tasks based on the optimal work schedule. Consider the emotional state of your staff and create a schedule that is least stressful.

[1555] Product display

[1556] Customer Service

[1557] Stock Check

[1558] Sales Report

[1559] User's current mild stress level is 8."

[1560] In this way, the present invention can be used in brick-and-mortar stores to improve the working environment for staff and increase work efficiency.

[1561] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1562] Step 1:

[1563] The server retrieves the user's calendar information.

[1564] Input: User ID

[1565] Output: User's calendar information (schedule and events)

[1566] What it does: The server retrieves the user's calendar information through an API request, providing data that includes the user's events and past meeting history.

[1567] Step 2:

[1568] The server retrieves the user's job title and department information.

[1569] Input: User ID

[1570] Output: User's job title and department information

[1571] What happens: The server sends an API request to an HR system or database to retrieve information about the user's job title and department.

[1572] Step 3:

[1573] The device collects the user's emotional data in real time.

[1574] Input: User ID, Emotion Engine

[1575] Output: User's emotional data (stress level, satisfaction, dissatisfaction, etc.)

[1576] Specific operation: The smart glasses or wearable device worn by the user uses an emotion engine to collect emotion data in real time and transmits the data to a server.

[1577] Step 4:

[1578] The server applies a machine learning model based on the collected data to predict the user's work participation tendencies.

[1579] Input: Calendar information, job title and department information, emotion data

[1580] Output: Work participation propensity score

[1581] Specific operation: The server integrates these data into a single dataset and applies machine learning algorithms (e.g., random forests, neural networks) to predict the user's work participation tendencies.

[1582] Step 5:

[1583] The server displays the predicted work attendance propensity score in a calendar application.

[1584] Input: Work participation propensity score

[1585] Output: Work participation propensity score displayed as a color or icon

[1586] Specific operation: The server incorporates the score into the calendar application and displays the results on the user's device in a visually easy-to-understand format (color coding and icons).

[1587] Step 6:

[1588] The device will notify you to recommend a time slot with a low score when setting up a meeting.

[1589] Input: Work participation propensity score

[1590] Output: Best time suggestion

[1591] Specific operation: The device sends a notification to the user, recommending that the user avoid times when the work attendance propensity score is low.

[1592] Step 7:

[1593] The server collects user feedback and retrains the machine learning model.

[1594] Input: User feedback

[1595] Output: An improved machine learning model

[1596] How it works: The server collects user feedback (e.g., "This time of day is stressful") and uses it to retrain the machine learning model to improve accuracy.

[1597] Step 8:

[1598] The server visually displays the optimized work schedule and the impact of emotional data.

[1599] Input: Optimized work schedule, emotional data

[1600] Output: Visually displayed work schedule and the impact of emotional data

[1601] Specific operation: The server visually displays the optimized schedule and emotion data in a calendar application or smart glasses, allowing users to work while being aware of their own state.

[1602] 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.

[1603] 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.

[1604] 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 robot 414.

[1605] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1606] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1607] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1608] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1609] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1610] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1611] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1612] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1613] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1614] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1615] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1616] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1617] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1618] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1619] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1620] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1621] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1622] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1623] The following is further disclosed regarding the above embodiment.

[1624] (Claim 1)

[1625] means for obtaining a user's calendar information;

[1626] A means for collecting past conference participation history;

[1627] A means for obtaining user job title and department information;

[1628] a means for applying a machine learning model to predict users' meeting participation trends based on the collected data;

[1629] means for displaying the predicted meeting attendance propensity score in a calendar application;

[1630] A means of notifying users to recommend a time slot with a low score when setting up a meeting;

[1631] A means to collect user feedback and retrain the machine learning model;

[1632] A system including:

[1633] (Claim 2)

[1634] 10. The system of claim 1, further comprising means for visually displaying the predicted meeting attendance propensity score using color coding or icons.

[1635] (Claim 3)

[1636] The system according to claim 1, further comprising means for recommending a time slot based on attendance tendencies and attendance probabilities of each user.

[1637] "Example 1"

[1638] (Claim 1)

[1639] means for obtaining a user's calendar information;

[1640] A means for collecting past conference participation history;

[1641] A means for obtaining user job title and department information;

[1642] a means for applying a machine learning model to predict users' meeting participation trends based on the collected data;

[1643] means for displaying the predicted meeting attendance propensity score in a calendar application;

[1644] A means of notifying users to recommend a time slot with a low score when setting up a meeting;

[1645] A means to collect user feedback and retrain the machine learning model;

[1646] A means for reflecting the calculation result of the meeting participation tendency score in actual meeting setting and schedule management;

[1647] A system including:

[1648] (Claim 2)

[1649] 10. The system of claim 1, further comprising means for visually displaying the predicted meeting attendance propensity score using color coding or icons.

[1650] (Claim 3)

[1651] 2. The system according to claim 1, further comprising a means for issuing a recommendation notice to avoid low-score time slots when setting up a meeting by a terminal as a means for recommending time slots based on the attendance tendency and attendance probability of each user.

[1652] "Application Example 1"

[1653] (Claim 1)

[1654] A means for obtaining schedule information of a user;

[1655] A means for collecting past conference participation history;

[1656] A means for obtaining user job title and department information;

[1657] a means for applying a generative AI model to predict users' meeting participation trends based on the collected data; and

[1658] means for displaying the predicted meeting attendance propensity score in a schedule management application;

[1659] A means of notifying people to avoid low-scoring times when scheduling meetings;

[1660] A means to collect user feedback and retrain the generative AI model;

[1661] A means to support the efficiency of industrial machine maintenance and manufacturing planning meetings,

[1662] A system including:

[1663] (Claim 2)

[1664] 10. The system of claim 1, further comprising means for visually displaying the predicted meeting attendance propensity score using color coding or icons.

[1665] (Claim 3)

[1666] The system according to claim 1, further comprising means for recommending a time slot based on attendance tendencies and attendance probabilities of each user.

[1667] "Example 2: Combining Emotion Engines"

[1668] (Claim 1)

[1669] A means for acquiring schedule information of a user;

[1670] A means for collecting past meeting attendance history;

[1671] A means for obtaining user position and department information;

[1672] a means of recognizing and collecting data on emotional states in real time;

[1673] A method for predicting a user's tendency to attend meetings by applying a machine learning model based on the collected calendar information, meeting participation history, job title / department information, and emotion data;

[1674] means for displaying the predicted meeting attendance propensity score in a calendar application;

[1675] A means to notify low-scoring time periods when setting up a meeting,

[1676] A means to collect user feedback and retrain the machine learning model;

[1677] A system including:

[1678] (Claim 2)

[1679] 10. The system of claim 1, further comprising means for visually displaying the predicted meeting attendance propensity score using color coding or numerical values.

[1680] (Claim 3)

[1681] 10. The system of claim 1, further comprising means for recommending a time slot based on each user's attendance habits, attendance probability, and emotional state.

[1682] "Application example 2 when combining emotion engines"

[1683] (Claim 1)

[1684] means for obtaining a user's calendar information;

[1685] A means for collecting past conference participation history;

[1686] A means for obtaining user job title and department information;

[1687] a means for applying a machine learning model to predict users' meeting participation trends based on the collected data;

[1688] means for displaying the predicted meeting attendance propensity score in a calendar application;

[1689] A means of notifying users to recommend a time slot with a low score when setting up a meeting;

[1690] A means to collect user feedback and retrain the machine learning model;

[1691] A means for recognizing a user's state using the acquired emotion data and optimizing a work schedule based on the state;

[1692] A means to visually display the impact of optimized work schedules and sentiment data;

[1693] A system including:

[1694] (Claim 2)

[1695] 10. The system of claim 1, further comprising means for visually displaying the predicted meeting attendance propensity score using color coding or icons.

[1696] (Claim 3)

[1697] The system according to claim 1, further comprising means for recommending a time slot based on attendance tendencies and attendance probabilities of each user. [Explanation of symbols]

[1698] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for obtaining a user's calendar information; A means for collecting past conference participation history; A means for obtaining user job title and department information; a means for applying a machine learning model to predict users' meeting participation trends based on the collected data; means for displaying the predicted meeting attendance propensity score in a calendar application; A means of notifying users to recommend a time slot with a low score when setting up a meeting; A means to collect user feedback and retrain the machine learning model; A system including:

2. The system of claim 1 , further comprising means for visually displaying the predicted meeting attendance propensity score using color coding or icons.

3. The system according to claim 1, further comprising means for recommending a time slot based on the attendance tendency and attendance probability of each user.

Images