system

Abstract

Provide a system. [Solution] a means for collecting business-related communication and document data; means for extracting keywords and topic candidates from the communication data and document data using natural language processing; means for dynamically selecting relevant participants based on the candidate topics; means for generating a meeting agenda based on the candidate topics and selected participants; and means for announcing the meeting agenda.

Description

[Technical Field]

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

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

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

[0004] In modern business activities, setting up and managing meetings requires a great deal of time and effort. Setting appropriate meeting agendas and selecting participants with relevant expertise is often difficult and inefficient. Furthermore, it is difficult to hold meaningful discussions and share information across departments, leading to a decline in work efficiency. Given this background, there is a growing need for a system that automates meeting setting and management, making it more efficient and effective. [Means for solving the problem]

[0005] The present invention provides a system including means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and selected participants, and means for notifying participants of the meeting agenda. This system reduces the time and effort required to set a meeting agenda and select participants, enabling smooth information sharing and business efficiency.

[0006] "Communication data" refers to data such as text, audio, and video used in business interactions, and specifically refers to information obtained from chat platforms, emails, etc.

[0007] "Document data" refers to data in file format that contains business-related documents and materials stored electronically, and refers to information obtained from cloud storage or file servers.

[0008] "Collection means" refers to a method or device for automatically capturing and storing communication and document data.

[0009] "Natural language processing" is a technology that allows computers to understand the language that humans use on a daily basis, and includes text analysis and speech recognition.

[0010] "Keywords" refer to important words and phrases extracted from communication data and document data and used to generate topic candidates.

[0011] "Agenda candidates" are proposed meeting topics derived from collected keywords and context.

[0012] "Dynamic selection means" refers to algorithms or methods for automatically selecting the most suitable meeting participants based on potential agenda items.

[0013] A "meeting agenda" is a planning document that includes the purpose, agenda, and participant list of a meeting, and is used to facilitate the progress of the meeting.

[0014] "Means for notifying" refers to a method or device for informing selected participants of the details of the meeting, including sending emails or creating calendar invitations. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention relates to a system that improves business efficiency by automating the setting of meeting agendas and participant selection. This system involves a series of operations: collecting business-related communication data and document data, extracting candidate agendas using natural language processing, dynamically selecting participants suitable for the agendas, and finally generating and notifying participants of the meeting agenda.

[0037] The overall system configuration is as follows:

[0038] 1. Data Collection

[0039] The server automatically collects data from communication platforms (e.g., chat systems, email) and document management systems (e.g., cloud storage, file servers) used in business.

[0040] The server connects to these platforms via API, periodically retrieves data, and stores it in a database.

[0041] 2. Natural Language Processing

[0042] The server analyzes the collected communication and document data using a natural language processing (NLP) engine to extract important keywords and potential topics.

[0043] NLP engines extract semantic information from data using techniques such as text tokenization, part-of-speech tagging, and named entity recognition (NER).

[0044] 3. Automatic extraction of agenda items

[0045] The server generates topic candidates based on the extracted keywords and phrases.

[0046] For example, from keywords such as "project delay" and "progress," topics such as "checking project progress" and "analyzing the cause of the delay" are generated.

[0047] 4. Dynamic Participant Management

[0048] The server automatically selects appropriate conference participants based on the aforementioned agenda candidates.

[0049] The server selects the most suitable participants based on information such as the participants' job titles, areas of expertise, and past meeting participation history. For example, members suitable for the agenda, such as "project managers" or "engineering team leaders," are selected.

[0050] 5. Agenda generation and notification

[0051] The server aggregates the topic candidates and selected participants and automatically generates the meeting agenda.

[0052] The server notifies participants of the generated agenda using methods such as email or calendar invitations.

[0053] Specific examples

[0054] Here is a concrete example of how the system works:

[0055] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[0056] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[0057] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[0058] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0059] 5. From the analysis results, potential topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" are generated.

[0060] 6. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A." This information is dynamically selected based on employee information stored in the database.

[0061] 7. The server generates the meeting agenda based on the topic and the selected participants, as follows:

[0062] Agenda:

[0063] 1. Analysis of the cause of delay in Project A

[0064] 2. Cost optimization measures

[0065] 3. Next Major Milestone

[0066] Participants:

[0067] Taro Tanaka (Project Manager)

[0068] Hanako Suzuki (Risk Management Officer)

[0069] Ichiro Yamada (Project member)

[0070] 8. The server notifies participants of the generated agenda via email and calendar invites.

[0071] In this way, the system reduces the workload involved in setting up and managing meetings, and supports the rapid and effective implementation of meetings.

[0072] The processing flow will be explained below.

[0073] Step 1:

[0074] Data collection

[0075] The server collects communication data from corporate chat platforms (e.g., Slack, MICROSOFT® TEAMS®) and email systems.

[0076] The server obtains this communication data via API, parses it in JSON format, and stores it in a database.

[0077] The server also retrieves document data from cloud storage (e.g., Google (registered trademark) Drive, SharePoint) and stores it in the database in the same way.

[0078] Step 2:

[0079] Analysis using natural language processing

[0080] The server inputs the collected communication data and document data into a natural language processing engine.

[0081] The server performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and significant phrases.

[0082] Step 3:

[0083] Automatic extraction of agenda items

[0084] The server applies an algorithm to generate topic candidates based on the keywords obtained from the NLP analysis results.

[0085] The server creates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization," and stores these in a database.

[0086] Step 4:

[0087] Dynamic Participant Management

[0088] The server selects the most suitable participants for the generated agenda candidates.

[0089] The server retrieves employee information such as past meeting history, areas of expertise, and job titles from a database, and dynamically selects participants deemed most suitable for each agenda item.

[0090] For example, the project manager, risk management officer, and project members are selected for the "cause analysis of project delays."

[0091] Step 5:

[0092] Generate a meeting agenda

[0093] The server automatically generates a meeting agenda based on the selected topic and participant list.

[0094] The server embeds the topic and participant information in the agenda template to create a formal meeting agenda.

[0095] Step 6:

[0096] Sending notifications

[0097] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[0098] The server interfaces with a mail server to send emails containing detailed agendas and meeting information, and uses a calendar system to generate meeting invites.

[0099] Example 1

[0100] 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."

[0101] In today's business environment, many companies and organizations generate large amounts of communication and document data. It is difficult to extract important agenda items from this data, select appropriate meeting participants, and efficiently schedule meetings, resulting in a decline in business efficiency. To solve this problem, a system that can effectively utilize communication and document data and automate meeting scheduling is needed.

[0102] 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.

[0103] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and the selected participants, and means for notifying the meeting agenda. This allows users to automate a series of processes from data collection to meeting setup and notification of the meeting agenda.

[0104] "Business-related communications data" refers to data including text messages, images, videos, and audio data sent and received via messaging platforms, emails, and other means used in business.

[0105] "Document data" refers to business-related data such as text documents, spreadsheets, presentation files, image files, and PDFs that are stored on remote storage or file servers.

[0106] "Natural language processing" is a technology for analyzing text data and extracting, generating, understanding, and translating information, and includes processes such as text tokenization, part-of-speech tagging, and named entity recognition.

[0107] "Keywords" are important words and phrases extracted from communication data and document data, and are information used to generate topic candidates.

[0108] "Agenda candidates" are themes or items to be discussed in the meeting, generated from the extracted keywords and phrases.

[0109] "Participants" are individuals selected to attend meetings related to the agenda, and are dynamically selected based on job title, area of ​​expertise, etc.

[0110] A "meeting agenda" is a list of specific topics or items to be discussed at a meeting, and is used to notify participants and guide the meeting.

[0111] "Notification medium" means the media and protocols for appropriately informing meeting participants of the generated meeting agenda, including email, calendar invites, messaging platform notifications, etc.

[0112] The present invention relates to a system that improves business efficiency by automating the setting of meeting agendas and participant selection. This system includes a series of operations: collecting business-related communication data and document data, extracting candidate agendas using natural language processing, dynamically selecting participants suitable for the agendas, and finally generating and notifying the meeting agenda.

[0113] The overall system configuration is as follows:

[0114] 1. Data Collection

[0115] Devices generate communication data such as chat systems and emails used for business, as well as document data stored in cloud storage and file servers every day. This data is sent from each employee's PC or mobile device. The server connects to communication platforms and document management systems via APIs to periodically collect this data.

[0116] Specific use cases include using Slack's API to retrieve chat messages and Google Drive's API to gather document data.

[0117] The server stores the retrieved data in a database.

[0118] 2. Natural Language Processing

[0119] The server inputs the communication data and document data stored in the database into a natural language processing (NLP) engine.

[0120] Examples of NLP engines used include SpaCy and NLTK.

[0121] Specific processing includes text tokenization, part-of-speech tagging, and named entity recognition (NER).

[0122] For example, if there is a chat message that says "Project A is delayed," the NLP engine breaks it down into words such as "project," "A," "delayed," and "being," and extracts important entities such as "Project A" and "delayed."

[0123] 3. Automatic extraction of agenda items

[0124] The server generates topic candidates based on the extracted keywords and phrases.

[0125] For example, from keywords such as "project delay" and "progress," topics such as "checking project progress" and "analyzing the cause of the delay" are generated.

[0126] 4. Dynamic Participant Management

[0127] The server dynamically selects appropriate conference participants based on the aforementioned agenda candidates.

[0128] The server selects the most suitable participants from a database based on their job title, area of ​​expertise, and past meeting attendance history, such as a project manager or engineering team leader.

[0129] 5. Agenda generation and notification

[0130] The server automatically generates a meeting agenda by aggregating the topic candidates and selected participants.

[0131] The server notifies participants of the generated agenda via email, calendar invites, or other means.

[0132] Specific examples are shown below:

[0133] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[0134] 2. The server periodically collects this communication data and document data obtained from cloud storage.

[0135] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are saved.

[0136] 4. The server inputs the saved data into an NLP engine (SpaCy) and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0137] 5. From the analysis results, potential topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" are generated.

[0138] 6. The server selects the project manager, risk management officer, and project members for the agenda item "Analyzing the causes of delays in Project A." This information is dynamically selected based on employee information stored in the database.

[0139] 7. The server generates the meeting agenda based on the topic and the selected participants, as follows:

[0140] Agenda:

[0141] 1. Analysis of the cause of delay in Project A

[0142] 2. Cost optimization measures

[0143] 3. Next Major Milestone

[0144] Participants:

[0145] Taro Tanaka (Project Manager)

[0146] Hanako Suzuki (Risk Management Officer)

[0147] Ichiro Yamada (Project member)

[0148] 8. The server notifies participants of the generated agenda via email and calendar invites.

[0149] Prompt statement:

[0150] "If there is communication data indicating that Project A is delayed and document data such as a monthly report, please explain a system that dynamically selects appropriate agenda candidates and participants based on them, generates a meeting agenda, and notifies them."

[0151] In this way, the system reduces the workload involved in setting up and managing a conference, and supports the rapid and effective implementation of conferences.

[0152] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0153] Step 1:

[0154] Terminals generate business-related communication data such as chat systems and emails, as well as document data from cloud storage and file servers, on a daily basis. These data are imported. The inputs are communication data (e.g., chat messages, emails) and document data (e.g., text files, PDFs), which are then saved. The output is the saved communication data and document data.

[0155] Specific behavior:

[0156] An employee sends a chat message on the device.

[0157] The device will upload the new document to cloud storage.

[0158] Step 2:

[0159] The server connects to communication platforms (e.g., Slack) and document management systems (e.g., Google Drive) via APIs to periodically collect data. The input is data obtained from the communication platform and document management system, and the output is data stored in a database. This allows communication data and document data to be periodically updated.

[0160] Specific behavior:

[0161] The server retrieves the latest chat messages from the Slack API.

[0162] The server retrieves the latest document file from the Google Drive API.

[0163] The server stores the collected data in a database.

[0164] Step 3:

[0165] The server inputs the communication data and document data stored in the database into a natural language processing engine. The input is the communication data and document data stored in the database, and the output is the analysis results of natural language processing. An NLP engine (e.g., SpaCy) is used to perform data tokenization, part-of-speech tagging, named entity recognition (NER), etc.

[0166] Specific behavior:

[0167] The server passes a chat message saying "Project A is delayed" to the NLP engine.

[0168] The server passes the document "This Month's Expense Report" to the NLP engine.

[0169] The NLP engine extracts keywords such as "project delay" and "cost optimization."

[0170] Step 4:

[0171] The server generates topic candidates based on the analysis results from the NLP engine. The input is keywords and phrases processed through natural language processing, and the output is topic candidates. Therefore, new topic candidates are listed based on important entities and keywords.

[0172] Specific behavior:

[0173] The server generates an agenda item "Checking project progress" from the keyword "project delay."

[0174] The server generates a "cost optimization measures" agenda from the keyword "cost optimization."

[0175] Step 5:

[0176] The server dynamically selects appropriate meeting participants based on the agenda candidates. The input is the agenda candidates and an employee database, and the output is a list of selected participants. The optimal participants are selected based on the participants' job titles, areas of expertise, and past meeting history.

[0177] Specific behavior:

[0178] The server searches the database for the project manager, risk manager, and project members.

[0179] The server creates a list of participants suitable for the "Project Progress Review" agenda item.

[0180] Step 6:

[0181] The server generates a meeting agenda based on the candidate agenda items and the selected participant information. The input is the candidate agenda items and the participant list, and the output is the generated meeting agenda. This completes the meeting agenda.

[0182] Specific behavior:

[0183] The server generates a meeting agenda like this:

[0184] Agenda:

[0185] 1. Analysis of the cause of delay in Project A

[0186] 2. Cost optimization measures

[0187] 3. Next Major Milestone

[0188] Participants:

[0189] Taro Tanaka (Project Manager)

[0190] Hanako Suzuki (Risk Management Officer)

[0191] Ichiro Yamada (Project member)

[0192] Step 7:

[0193] The server notifies participants of the generated meeting agenda using notification means (e.g., email, calendar invitation). The input is the generated meeting agenda, and the output is the notified meeting agenda. This allows participants to receive information about the meeting in a timely manner.

[0194] Specific behavior:

[0195] The server sends the meeting agenda to the participants by email.

[0196] The server sends calendar invites and adds the meeting to the attendees' schedules.

[0197] The above are the specific processing steps of this system. By describing the operations performed at each step and their inputs and outputs in detail, the overall flow and processing content of the system will become clear.

[0198] (Application example 1)

[0199] 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."

[0200] Conventional meeting scheduling systems have the problem of being inefficient, taking a lot of time and effort to set up meeting agendas and select participants. In particular, in-factory operations require the ability to dynamically extract necessary agendas and participants in real time and smoothly schedule meetings. Furthermore, there are many situations where a quick response according to the situation is required, and current systems are insufficient.

[0201] 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.

[0202] In this invention, the server includes: means for collecting communication data and document data related to business operations; means for extracting keywords and agenda candidates from the communication data and document data using natural language processing; means for dynamically selecting relevant participants based on the agenda candidates; means for generating a meeting agenda based on the agenda candidates and the selected participants; means for notifying the meeting agenda; means for collecting operation data of a communication platform and equipment operated in a factory; means for generating agenda candidates using the operation data of the communication platform and equipment; means for selecting appropriate workers in the factory as participants based on the agenda candidates; and means for notifying the meeting agenda by voice or display via a robot. This automates the setting of a meeting agenda and the selection of appropriate participants, enabling fast and efficient meeting management in a factory.

[0203] "Business-related communication data" refers to data obtained from various communication means used in business, such as chat platforms and email.

[0204] "Document data" refers to electronic file data such as business-related reports and materials.

[0205] "Natural language processing" is a technology for analyzing text data and extracting useful information such as keywords and topic candidates.

[0206] A "keyword" is a specific word or phrase extracted from text data.

[0207] "Agenda candidates" are meeting topics generated based on the extracted keywords.

[0208] "Participants" are people or parties who need to attend a meeting.

[0209] A "meeting agenda" is a document that lists the proceedings and agenda items for a meeting.

[0210] "Notification" is the act of informing participants of the generated meeting agenda.

[0211] A "communication platform" is a system for exchanging information such as chat systems and e-mail.

[0212] "Equipment operation data" refers to data that indicates the operating status and performance of equipment used in a factory.

[0213] A "robot" is an automated mechanical device with a voice assistant function and a display function.

[0214] "Voice notification" is a method in which a robot communicates information using voice.

[0215] "Display" refers to the method by which a robot displays and communicates information on a display.

[0216] MODE FOR CARRYING OUT THE INVENTION

[0217] Overall system configuration

[0218] This invention requires a system for supporting the scheduling of meetings within a factory. This system collects and automatically analyzes business-related communication and document data to select topics and participants, and generates and notifies meeting agendas.

[0219] Hardware Configuration

[0220] 1. Server: The main computational unit responsible for data collection, natural language processing, topic extraction, participant selection, agenda generation and notifications.

[0221] 2. Terminal: A PC or mobile device where communication data and document data are stored.

[0222] 3. Robot: A device that announces the meeting agenda by voice or display.

[0223] Software Configuration

[0224] 1. Natural language processing engine (e.g., nltk, sklearn): Software for extracting keywords and topic candidates from collected data.

[0225] 2. Email sending library (e.g. smtplib): Software for notifying participants of the generated agenda.

[0226] 3. Generative AI model: This is the AI ​​model used to analyze the collected data and generate the agenda.

[0227] Processing flow and operation explanation

[0228] Data collection

[0229] The server periodically uses APIs to collect data from communication platforms (chat systems and email) used within the factory and document data stored in the cloud. It also collects operational data from the equipment used in the factory. All of this data is stored in a database.

[0230] Natural Language Processing

[0231] The server analyzes the collected communication and document data using a natural language processing engine, tokenizing the text data and tagging it by part of speech to extract important keywords and potential topics.

[0232] Agenda extraction

[0233] The server generates candidate topics based on the extracted keywords and phrases. For example, keywords such as "maintenance delay" and "progress report" can be used to generate specific topics such as "checking the robot maintenance status" and "checking the product line progress."

[0234] Participant selection

[0235] The server dynamically selects appropriate meeting participants based on the generated agenda candidates, and automatically selects factory workers based on their past meeting participation history, job titles, and areas of expertise.

[0236] Agenda generation and notifications

[0237] The server automatically generates a meeting agenda based on the generated topic and selected participants. The generated agenda is notified to participants via email or calendar invitation. It is also notified to operators via voice assistants and displays on the factory robots.

[0238] Specific examples

[0239] For example, at the end of the day, the server collects chat messages such as "Robot A's maintenance is behind schedule" and data such as "Today's product line progress report." Based on this data, agenda candidates "Checking the maintenance status of Robot A" and "Checking the product line progress" are generated. Based on these agendas, maintenance personnel and production line leaders are selected as participants. The meeting agenda is generated as follows:

[0240] Agenda:

[0241] 1. Check the maintenance status of Robot A

[0242] 2. Check the progress of the product line

[0243] 3. Next Major Milestone

[0244] Participants:

[0245] Maintenance personnel

[0246] Production Line Leader

[0247] factory manager

[0248] The generated agenda is notified to participants via email, and the robot notifies the operator via voice and display.

[0249] Prompt Sentence Examples

[0250] "Automatically generate a meeting agenda. Use the following communication data: 'Robot A's maintenance is overdue', 'We need to check the progress of the production line'"

[0251] This automates the setting of meeting agendas and the selection of appropriate participants, enabling meetings within the factory to be run quickly and efficiently.

[0252] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0253] Step 1: Data collection

[0254] The server collects communication data from communication platforms and emails used within the factory. It also collects document data from cloud storage, including chat messages and reports. Specifically, it accesses these platforms via API and periodically retrieves the required data. The input is communication data and document data, and the output is the collected data.

[0255] Step 2: Natural Language Processing

[0256] The server inputs the collected communication data and document data into a natural language processing engine (e.g., nltk, sklearn) to analyze the text data. Here, the text is tokenized and tagged with parts of speech to extract important keywords and phrases. The input is the collected data, and the output is the extracted keywords and phrases. Specifically, TF-IDF vectorization and KMeans clustering are performed.

[0257] Step 3: Automatically extracting agenda items

[0258] The server generates meeting agenda candidates based on keywords and phrases extracted through natural language processing. For example, it generates the agenda "Checking the robot's maintenance status" from the keyword "maintenance delay." The input is the extracted keywords and phrases, and the output is the agenda candidate.

[0259] Step 4: Dynamic Participant Management

[0260] The server automatically selects appropriate meeting participants based on the generated agenda candidates. It selects the best participants for each agenda based on their past meeting participation history and job titles. The input is the agenda candidates, and the output is the selected participants. Specifically, it queries and filters relevant participant information from a database.

[0261] Step 5: Generate the agenda

[0262] The server automatically generates a meeting agenda based on the candidate agenda items and selected participants. The input is the candidate agenda items and participant information, and the output is the generated meeting agenda. Specifically, the server creates the agenda document using a text template.

[0263] Step 6: Announce the agenda

[0264] The server notifies the generated meeting agenda. The agenda is notified to participants using email notifications and calendar invitations. It also notifies operators via voice or display on the factory robots. The input is the generated meeting agenda, and the output is the notified agenda. Specifically, emails are sent using an email sending library (e.g., smtplib), and voice instructions and display are given through the robot's API.

[0265] 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.

[0266] This invention relates to a system that automates the setting of meeting agendas and participant selection, as well as recognizes user emotions to improve the quality of meetings. This system collects business-related communication data and document data, extracts candidate topics using natural language processing, dynamically selects participants suitable for the topics, and combines an emotion engine to recognize user emotions and reflect them in the topics and agendas.

[0267] The overall system configuration is as follows:

[0268] 1. Data Collection

[0269] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[0270] The server retrieves this communication data via API, parses it in JSON format, and stores it in a database. It also retrieves document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in a database.

[0271] 2. Natural Language Processing

[0272] The server inputs the collected communication and document data into a natural language processing (NLP) engine.

[0273] The server performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and significant phrases.

[0274] 3. Emotion Engine

[0275] The server sends the text extracted from the communication data and document data to an emotion engine to analyze the user's emotion.

[0276] The server recognizes emotions such as positive, negative, and neutral, and stores the results in a database.

[0277] 4. Automatic extraction of agenda items

[0278] The server applies an algorithm to generate topic candidates based on the keywords obtained from the NLP analysis results and the analysis results of the emotion engine.

[0279] The server creates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and emotional information, and stores them in a database.

[0280] 5. Dynamic Participant Management

[0281] The server selects the most suitable participants for the generated agenda candidates.

[0282] The server retrieves employee information such as past meeting history, areas of expertise, and job titles from the database, and dynamically selects the most appropriate participants for each agenda item. For example, for "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[0283] 6. Generate a meeting agenda

[0284] The server automatically generates a meeting agenda based on the selected topic and participant list.

[0285] The server embeds the topic and participant information in the agenda template and modifies the generated agenda to reflect the user's emotional information.

[0286] Based on the results of the emotion engine, the server prioritizes topics that are generating high emotions and reflects them in the agenda.

[0287] 7. Sending Notifications

[0288] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[0289] The server interfaces with the mail server to send emails containing detailed agendas and meeting information, and uses the calendar system to generate meeting invites.

[0290] Specific examples

[0291] Here is a concrete example of how the system works:

[0292] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[0293] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[0294] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[0295] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0296] 5. At the same time, the sentiment engine determines that the user has a strong negative sentiment towards the message "Project A is delayed."

[0297] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[0298] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[0299] 8. The server generates the meeting agenda based on the selected topic and participant list, as follows:

[0300] Agenda:

[0301] 1. Analysis of the cause of delay in Project A

[0302] 2. Cost optimization measures

[0303] 3. Next Major Milestone

[0304] Participants:

[0305] Taro Tanaka (Project Manager)

[0306] Hanako Suzuki (Risk Management Officer)

[0307] Ichiro Yamada (Project member)

[0308] 9. The server notifies participants of the generated agenda via email and calendar invites.

[0309] 10. Based on the analysis results of the emotion engine, the server makes modifications such as placing topics for which the user has particularly negative emotions at the top of the agenda.

[0310] In this way, the system reduces the workload involved in setting up and managing meetings, and supports prompt and effective meeting conduct through emotionally sensitive agenda setting.

[0311] The processing flow will be explained below.

[0312] Step 1:

[0313] Data collection

[0314] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[0315] The server obtains this communication data via API, parses it in JSON format, and stores it in a database.

[0316] The server also collects document data from cloud storage (e.g., Google Drive, SharePoint) via API and stores it in a database.

[0317] Step 2:

[0318] Performing Natural Language Processing (NLP)

[0319] The server inputs the collected communication and document data into a natural language processing (NLP) engine.

[0320] The server performs text tokenization, morphological analysis, part-of-speech tagging, and named entity recognition (NER) to extract important keywords and phrases.

[0321] Step 3:

[0322] emotion recognition

[0323] The server sends the communication data and the text extracted from the document data to the emotion engine.

[0324] The server uses an emotion engine to analyze the sentiment of each piece of text and classify it into categories such as positive, negative, or neutral.

[0325] The server stores the emotion recognition results in a database.

[0326] Step 4:

[0327] Automatic extraction of agenda items

[0328] The server runs an algorithm that integrates the results of NLP analysis and emotion recognition to generate topic candidates.

[0329] The server generates candidate topics such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and emotional information, and stores them in a database.

[0330] Step 5:

[0331] Dynamic Participant Management

[0332] The server dynamically selects the most suitable participants for the generated agenda candidates.

[0333] The server retrieves information such as employee job titles, areas of expertise, and past meeting participation history from a database and selects the most appropriate participants for each agenda item.

[0334] For example, for the agenda item "Analysis of the causes of project delays," the project manager, risk management officer, and project members are selected.

[0335] Step 6:

[0336] Generate a meeting agenda

[0337] The server automatically generates a meeting agenda based on the selected topic and participant list.

[0338] Based on the results of the emotion engine, the server makes adjustments such as placing topics with high negative emotions at the beginning of the agenda.

[0339] Step 7:

[0340] Sending notifications

[0341] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[0342] The server interfaces with a mail server to send emails containing detailed agendas and meeting information, and uses a calendar system to generate meeting invites.

[0343] Specific examples

[0344] Here is a concrete example of how the system works:

[0345] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[0346] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[0347] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[0348] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0349] 5. At the same time, the emotion engine recognizes emotions from the user's communication data and document data and determines that the text "Project A is delayed" indicates a strong negative emotion.

[0350] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[0351] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[0352] 8. The server generates the meeting agenda based on the selected topic and participant list, as follows:

[0353] Agenda:

[0354] 1. Analysis of the cause of delay in Project A

[0355] 2. Cost optimization measures

[0356] 3. Next Major Milestone

[0357] Participants:

[0358] Taro Tanaka (Project Manager)

[0359] Hanako Suzuki (Risk Management Officer)

[0360] Ichiro Yamada (Project member)

[0361] 9. The server notifies participants of the generated agenda via email and calendar invites.

[0362] 10. Based on the analysis results of the emotion engine, the server makes adjustments to improve the quality of the meeting, such as placing topics for which users have particularly negative emotions at the top of the agenda.

[0363] In this way, the system reduces the workload involved in setting up and managing meetings, and supports prompt and effective meeting conduct through emotionally sensitive agenda setting.

[0364] Example 2

[0365] 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."

[0366] Conventionally, setting the agenda and selecting participants for a meeting have been done manually, which has the problem of requiring a great deal of effort and time. Furthermore, it has been difficult to consider the user's feelings when setting the agenda, which has also made it difficult to improve the quality of the meeting. The present invention aims to solve these problems and improve the quality of meetings through automated agenda setting and participant selection.

[0367] 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.

[0368] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and selected participants, means for notifying the meeting agenda, means for analyzing user emotions from the communication data and document data, and means for determining the priority of agenda items based on the emotion analysis results. This reduces the workload associated with setting and managing meetings, and enables more effective meetings by setting agendas that take emotions into consideration.

[0369] "Business-related communications data" refers to messages and their metadata exchanged through chat platforms and email systems used within a company.

[0370] "Document data" refers to digital information such as document files, reports, and presentation materials obtained from cloud storage or corporate databases.

[0371] "Natural language processing" refers to techniques and algorithms that enable computers to understand, interpret, and generate human language.

[0372] "Keywords" refer to important words and phrases extracted from communication data and document data through natural language processing.

[0373] "Agenda candidates" refer to items to be discussed at a meeting that are generated based on keywords extracted from communication data and document data.

[0374] "Participants" refers to employees or personnel who should be invited to the meeting.

[0375] "Meeting agenda" refers to a meeting schedule generated based on candidate topics and a selected participant list.

[0376] "Sentiment analysis" refers to the process of identifying emotions contained in user communication and document data and classifying those emotions as positive, negative, neutral, etc.

[0377] "Priority" refers to the criteria used to determine the order in which meeting topics are discussed based on the results of sentiment analysis.

[0378] This invention relates to a system that automates the setting of meeting agendas and participant selection, as well as recognizes user emotions to improve the quality of meetings. This system collects business-related communication data and document data, extracts candidate topics using natural language processing (NLP), dynamically selects participants suitable for the topics, and combines an emotion engine to recognize user emotions and reflect them in the topics and agendas.

[0379] Overall system configuration

[0380] Data collection

[0381] The server collects communication data from the company's internal chat platform (e.g., Slack, Microsoft Teams) and email system. It obtains the data via API, parses it in JSON format, and stores it in a database. It also obtains document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in a database.

[0382] Natural Language Processing (NLP)

[0383] The server inputs the collected communication and document data into an NLP engine (e.g., Google Cloud NLP, AWS® Comprehend), which performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and important phrases.

[0384] Emotion Engine

[0385] The server sends the text extracted from the communication data and document data to an emotion engine (e.g., IBM Watson® Tone Analyzer) to analyze the user's emotions. The classification results into emotion categories such as positive, negative, and neutral are stored in a database.

[0386] Automatic extraction of agenda items

[0387] The server applies an algorithm based on the results of NLP analysis and sentiment analysis to generate topic candidates. For example, it generates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and sentiment information.

[0388] Dynamic Participant Management

[0389] The server selects the most suitable participants for each candidate agenda item. Employee information such as past meeting history, areas of expertise, and job titles is retrieved from the database, and the most suitable participants are dynamically selected for each agenda item. For example, for the agenda item "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[0390] Generate a meeting agenda

[0391] The server automatically generates a meeting agenda based on the selected topic and participant list. It embeds the topic and participant information in an agenda template and modifies the generated agenda to reflect emotional information. Based on the results of the emotional analysis, it prioritizes topics for which users have particularly strong emotions and reflects them in the agenda.

[0392] Sending notifications

[0393] The server notifies the generated meeting agenda to the relevant participants via email and calendar invites. It works with the mail server to send emails containing the detailed agenda and meeting information, and creates meeting invites using the calendar system.

[0394] Specific examples

[0395] Here is a concrete example of how the system works:

[0396] 1. Work-related chat messages and emails accumulate daily on devices (employee computers and mobile devices).

[0397] 2. The server collects this communication data and document data from cloud storage on a daily basis.

[0398] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[0399] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0400] 5. At the same time, the emotion engine determines that the user is expressing strong negative emotion towards the message "Project A is delayed."

[0401] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[0402] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[0403] 8. The server generates a meeting agenda based on the selected topic and participant list:

[0404] Agenda:

[0405] 1. Analysis of the cause of delay in Project A

[0406] 2. Cost optimization measures

[0407] 3. Next Major Milestone

[0408] Participants:

[0409] Taro Tanaka (Project Manager)

[0410] Hanako Suzuki (Risk Management Officer)

[0411] Ichiro Yamada (Project member)

[0412] 9. The server notifies participants of the generated agenda via email and calendar invites.

[0413] 10. Based on the analysis results of the emotion engine, the server places topics for which the user has particularly negative emotions at the top of the agenda.

[0414] Prompt Sentence Examples

[0415] The following prompts are used to input the generative AI model:

[0416] Please create a program that extracts topics from business chat messages containing negative sentiment, such as "Project A is delayed," and generates an agenda by selecting the most appropriate meeting participants. Please also provide an example of the generated agenda.

[0417] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0418] Step 1: Data collection

[0419] The server collects communication data from the company's internal chat platforms (e.g., chat service A, chat service B) and email systems. Specifically, it uses APIs to obtain text-based message data and email data from these services, parses it into JSON format, and stores it in a database.

[0420] Input: Data obtained from APIs of chat platforms and email systems.

[0421] Data processing: Convert the acquired data into JSON format.

[0422] Output: JSON formatted communication and email data, stored in a database.

[0423] Step 2: Collect document data from cloud storage

[0424] The server retrieves document data (PDF, Word files, etc.) from cloud storage (e.g., storage service A, storage service B) and stores it in a database. It periodically scans documents using the storage API to retrieve new and updated documents.

[0425] Input: Data obtained from cloud storage API.

[0426] Data processing: Convert the acquired document data into JSON format.

[0427] Output: Document data in JSON format, stored in the database.

[0428] Step 3: Natural Language Processing (NLP)

[0429] The server inputs the communication and document data stored in the database into NLP engines (e.g., NLP engine A, NLP engine B). The NLP engines perform text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract important keywords and phrases.

[0430] Input: Correspondence and document data extracted from the database.

[0431] Data operations: text tokenization, part-of-speech tagging, NER processing.

[0432] Output: Keywords and important phrases, stored in a database.

[0433] Step 4: Emotion analysis using the emotion engine

[0434] The server sends the text analyzed by the NLP engine to an emotion engine (e.g., emotion analysis service A, emotion analysis service B) to analyze the user's emotions. It classifies the emotions into positive, negative, neutral, and other emotional characteristics, and stores the results in a database.

[0435] Input: Keywords and phrases obtained as a result of analysis by the NLP engine.

[0436] Data Computation: Determine the sentiment category of the text through sentiment analysis.

[0437] Output: Sentiment analysis results (positive, negative, neutral), stored in a database.

[0438] Step 5: Automatically extracting topic candidates

[0439] The server generates candidate topics by applying an algorithm based on the results of NLP analysis and sentiment analysis. This extracts topics with keywords such as "project delay" and "cost optimization," and generates specific themes such as "analysis of the causes of project delays" and "cost optimization measures."

[0440] Input: NLP analysis results, sentiment analysis results.

[0441] Data computation: Applying algorithms that combine keywords and sentiment information to generate agendas.

[0442] Output: Agenda candidates, stored in a database.

[0443] Step 6: Dynamic Participant Management

[0444] The server selects the most suitable participants for each candidate topic. Appropriate employees are dynamically selected based on information such as past meeting history, areas of expertise, and job titles. For example, for the topic "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[0445] Input: Agenda candidate, employee information retrieved from database.

[0446] Data calculation: Applying algorithms to select the most suitable participants based on meeting history, areas of expertise, and job title information.

[0447] Output: Participant list, stored in database.

[0448] Step 7: Generate the meeting agenda

[0449] The server automatically generates a meeting agenda based on the selected topics and participant list. It embeds the topic and participant information in a template and prioritizes topics about which users have particularly strong feelings, reflecting them in the agenda.

[0450] Input: candidate agenda items, participant list, sentiment analysis results.

[0451] Data operations: Embedding information into templates and prioritizing agenda items.

[0452] Output: The generated meeting agenda, stored in a database.

[0453] Step 8: Sending notifications

[0454] The server notifies the generated meeting agenda to interested participants via email and calendar invites. It works with a mail server to send emails containing the detailed agenda and meeting information, and uses a calendar system to create meeting invites.

[0455] Input: The generated meeting agenda.

[0456] Data Computing: Notification generation via email and calendar systems.

[0457] Output: Notification emails and calendar events sent.

[0458] (Application example 2)

[0459] 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."

[0460] Troubles and delays at production sites are issues that require prompt and appropriate countermeasures. However, the traditional method of setting up meetings and selecting participants places a heavy workload on employees, making efficient management difficult, especially in large factories. Furthermore, meetings often fail to take into account the feelings of employees when setting agendas and topics, resulting in ineffective discussions. This hinders the efficient operation of factories, creating a challenge.

[0461] 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.

[0462] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and the selected participants, means for analyzing user emotions, means for modifying the meeting agenda based on emotion information, and means for notifying the meeting agenda. This improves the efficiency of meeting setup and management for troubleshooting at production sites and enables effective discussions that take user emotions into consideration.

[0463] "Business-related communication data" refers to communication information used within a company, such as chat, email, and monitoring data.

[0464] "Document data" refers to document information such as reports, memos, and files stored in online storage or local storage.

[0465] "Natural language processing" is an information processing technology for analyzing and understanding human language.

[0466] "Keywords" are important words and phrases extracted from the data.

[0467] "Agenda candidate" refers to themes and issues extracted through natural language processing.

[0468] "Relevant participants" are people or positions who should attend the meeting based on the proposed agenda.

[0469] A "meeting agenda" is a document that includes the agenda and attendee list to direct the progress of a meeting.

[0470] "User emotion" is the emotional state of the user as assessed by the emotion analysis engine.

[0471] "Emotion information" is data obtained as a result of analyzing the user's emotions.

[0472] An "emotion engine" is an information processing system for analyzing emotions from communication data and voice data.

[0473] "Dynamic selection of relevant participants based on topic candidates" is a process of selecting the most suitable participants in real time, taking into account their past history and areas of expertise related to the topic.

[0474] A "conversation platform" is a communication tool such as a chat tool or video conferencing software used within a company.

[0475] "Online storage" is a data storage service that can be accessed via the Internet.

[0476] "Notifying a meeting agenda" refers to the act of informing relevant parties of the generated agenda via email or calendar invitation.

[0477] This invention is a conference management system for quickly and effectively resolving problems and delays at production sites. This system collects business-related communication data and document data, extracts candidate topics using natural language processing, dynamically selects participants, and uses an emotion engine to recognize user emotions and reflect them in the topics and agenda.

[0478] The server first collects real-time data from sensors and surveillance cameras. It also obtains production data and trouble reports from the factory management system (MES) via API and stores them in a database. Based on the collected data, the server uses a natural language processing engine (e.g., NLTK, Transformers) to extract important keywords and phrases and generate candidate topics.

[0479] The emotion engine (e.g., Emotion AI) analyzes emotions from collected data and user voice and chat. The emotion results are stored in a database as emotional states, such as positive, negative, or neutral. The server then modifies the meeting agenda based on this emotional information, prioritizing topics that indicate negative emotions in the agenda list.

[0480] The generated meeting agenda is then sent to participants via email or a calendar invitation system, using common business communication tools (e.g., Microsoft Outlook, Google Calendar).

[0481] For example, the server generates candidate topics such as "malfunction of machine A" or "delay on production line B" and selects appropriate participants such as machine maintenance personnel and operators. This information is then reflected in the negative emotions analyzed by the emotion engine, and notified as the generated meeting agenda.

[0482] As an example, the following prompt sentences can be input to a generative AI model:

[0483] "An AI robot collects sensor data and trouble reports from within a factory, inputs this data into a natural language processing and sentiment analysis engine, selects appropriate meeting topics and participants, and automatically generates and notifies meeting agendas. We will show you a specific example of how this works."

[0484] In this way, the present invention can realize an effective conference management system in the production site and improve the efficiency of troubleshooting.

[0485] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0486] Step 1:

[0487] The server collects real-time data from sensors and surveillance cameras. It acquires the data through the sensor API and the surveillance camera API and stores it in a database. The input is sensor data and video data, and the output is the collected data stored in the database.

[0488] Step 2:

[0489] The server obtains production data and trouble reports from the factory management system (MES) via API. This data is also stored in the database. The input is production data and trouble reports, and the output is business data stored in the database.

[0490] Step 3:

[0491] The server inputs collected data (sensor data, production data, trouble reports) into a natural language processing (NLP) engine, which performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and important phrases. The input is collected data, and the output is keywords and topic candidates.

[0492] Step 4:

[0493] The server inputs the collected communication data and the user's voice and chat data into an emotion engine (e.g., Emotion AI) to analyze the user's emotions. The emotion engine analyzes the emotional state, such as positive, negative, or neutral, and stores the results in a database. The input is communication data and voice data, and the output is the emotion analysis results.

[0494] Step 5:

[0495] The server generates topic candidates and dynamically selects relevant participants based on the processed NLP analysis results and sentiment analysis results. The inputs are the NLP analysis results and sentiment analysis results, and the outputs are topic candidates and a list of participants.

[0496] Step 6:

[0497] The server automatically generates a meeting agenda based on the generated topics and the selected participant list. It considers emotional information and prioritizes topics that show negative emotions. The inputs are candidate topics and the participant list, and the output is a revised meeting agenda.

[0498] Step 7:

[0499] The server notifies participants of the generated meeting agenda via email and / or calendar invites. The input to the notification system is the modified meeting agenda, and the output is the meeting notice sent to participants.

[0500] 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.

[0501] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0502] 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.

[0503] [Second embodiment]

[0504] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0505] 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.

[0506] 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).

[0507] 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.

[0508] 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.

[0509] 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).

[0510] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0511] 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.

[0512] 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.

[0513] 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.

[0514] 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.

[0515] 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."

[0516] The present invention relates to a system that improves business efficiency by automating the setting of meeting agendas and participant selection. This system involves a series of operations: collecting business-related communication data and document data, extracting candidate agendas using natural language processing, dynamically selecting participants suitable for the agendas, and finally generating and notifying participants of the meeting agenda.

[0517] The overall system configuration is as follows:

[0518] 1. Data Collection

[0519] The server automatically collects data from communication platforms (e.g., chat systems, email) and document management systems (e.g., cloud storage, file servers) used in business.

[0520] The server connects to these platforms via API, periodically retrieves data, and stores it in a database.

[0521] 2. Natural Language Processing

[0522] The server analyzes the collected communication and document data using a natural language processing (NLP) engine to extract important keywords and potential topics.

[0523] NLP engines extract semantic information from data using techniques such as text tokenization, part-of-speech tagging, and named entity recognition (NER).

[0524] 3. Automatic extraction of agenda items

[0525] The server generates topic candidates based on the extracted keywords and phrases.

[0526] For example, from keywords such as "project delay" and "progress," topics such as "checking project progress" and "analyzing the cause of the delay" are generated.

[0527] 4. Dynamic Participant Management

[0528] The server automatically selects appropriate conference participants based on the aforementioned agenda candidates.

[0529] The server selects the most suitable participants based on information such as the participants' job titles, areas of expertise, and past meeting participation history. For example, members suitable for the agenda, such as "project managers" or "engineering team leaders," are selected.

[0530] 5. Agenda generation and notification

[0531] The server aggregates the topic candidates and selected participants and automatically generates the meeting agenda.

[0532] The server notifies participants of the generated agenda using methods such as email or calendar invitations.

[0533] Specific examples

[0534] Here is a concrete example of how the system works:

[0535] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[0536] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[0537] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[0538] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0539] 5. From the analysis results, potential topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" are generated.

[0540] 6. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A." This information is dynamically selected based on employee information stored in the database.

[0541] 7. The server generates the meeting agenda based on the topic and the selected participants, as follows:

[0542] Agenda:

[0543] 1. Analysis of the cause of delay in Project A

[0544] 2. Cost optimization measures

[0545] 3. Next Major Milestone

[0546] Participants:

[0547] Taro Tanaka (Project Manager)

[0548] Hanako Suzuki (Risk Management Officer)

[0549] Ichiro Yamada (Project member)

[0550] 8. The server notifies participants of the generated agenda via email and calendar invites.

[0551] In this way, the system reduces the workload involved in setting up and managing meetings, and supports the rapid and effective implementation of meetings.

[0552] The processing flow will be explained below.

[0553] Step 1:

[0554] Data collection

[0555] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[0556] The server obtains this communication data via API, parses it in JSON format, and stores it in a database.

[0557] The server also retrieves document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in the database as well.

[0558] Step 2:

[0559] Analysis using natural language processing

[0560] The server inputs the collected communication data and document data into a natural language processing engine.

[0561] The server performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and significant phrases.

[0562] Step 3:

[0563] Automatic extraction of agenda items

[0564] The server applies an algorithm to generate topic candidates based on the keywords obtained from the NLP analysis results.

[0565] The server creates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization," and stores these in a database.

[0566] Step 4:

[0567] Dynamic Participant Management

[0568] The server selects the most suitable participants for the generated agenda candidates.

[0569] The server retrieves employee information such as past meeting history, areas of expertise, and job titles from a database, and dynamically selects participants deemed most suitable for each agenda item.

[0570] For example, the project manager, risk management officer, and project members are selected for the "cause analysis of project delays."

[0571] Step 5:

[0572] Generate a meeting agenda

[0573] The server automatically generates a meeting agenda based on the selected topic and participant list.

[0574] The server embeds the topic and participant information in the agenda template to create a formal meeting agenda.

[0575] Step 6:

[0576] Sending notifications

[0577] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[0578] The server interfaces with a mail server to send emails containing detailed agendas and meeting information, and uses a calendar system to generate meeting invites.

[0579] Example 1

[0580] 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."

[0581] In today's business environment, many companies and organizations generate large amounts of communication and document data. It is difficult to extract important agenda items from this data, select appropriate meeting participants, and efficiently schedule meetings, resulting in a decline in business efficiency. To solve this problem, a system that can effectively utilize communication and document data and automate meeting scheduling is needed.

[0582] 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.

[0583] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and the selected participants, and means for notifying the meeting agenda. This allows users to automate a series of processes from data collection to meeting setup and notification of the meeting agenda.

[0584] "Business-related communications data" refers to data including text messages, images, videos, and audio data sent and received via messaging platforms, emails, and other means used in business.

[0585] "Document data" refers to business-related data such as text documents, spreadsheets, presentation files, image files, and PDFs that are stored on remote storage or file servers.

[0586] "Natural language processing" is a technology for analyzing text data and extracting, generating, understanding, and translating information, and includes processes such as text tokenization, part-of-speech tagging, and named entity recognition.

[0587] "Keywords" are important words and phrases extracted from communication data and document data, and are information used to generate topic candidates.

[0588] "Agenda candidates" are themes or items to be discussed in the meeting, generated from the extracted keywords and phrases.

[0589] "Participants" are individuals selected to attend meetings related to the agenda, and are dynamically selected based on job title, area of ​​expertise, etc.

[0590] A "meeting agenda" is a list of specific topics or items to be discussed at a meeting, and is used to notify participants and guide the meeting.

[0591] "Notification medium" means the media and protocols for appropriately informing meeting participants of the generated meeting agenda, including email, calendar invites, messaging platform notifications, etc.

[0592] The present invention relates to a system that improves business efficiency by automating the setting of meeting agendas and participant selection. This system includes a series of operations: collecting business-related communication data and document data, extracting candidate agendas using natural language processing, dynamically selecting participants suitable for the agendas, and finally generating and notifying the meeting agenda.

[0593] The overall system configuration is as follows:

[0594] 1. Data Collection

[0595] Devices generate communication data such as chat systems and emails used for business, as well as document data stored in cloud storage and file servers every day. This data is sent from each employee's PC or mobile device. The server connects to communication platforms and document management systems via APIs to periodically collect this data.

[0596] Specific use cases include using Slack's API to retrieve chat messages and Google Drive's API to gather document data.

[0597] The server stores the retrieved data in a database.

[0598] 2. Natural Language Processing

[0599] The server inputs the communication data and document data stored in the database into a natural language processing (NLP) engine.

[0600] Examples of NLP engines used include SpaCy and NLTK.

[0601] Specific processing includes text tokenization, part-of-speech tagging, and named entity recognition (NER).

[0602] For example, if there is a chat message that says "Project A is delayed," the NLP engine breaks it down into words such as "project," "A," "delayed," and "being," and extracts important entities such as "Project A" and "delayed."

[0603] 3. Automatic extraction of agenda items

[0604] The server generates topic candidates based on the extracted keywords and phrases.

[0605] For example, from keywords such as "project delay" and "progress," topics such as "checking project progress" and "analyzing the cause of the delay" are generated.

[0606] 4. Dynamic Participant Management

[0607] The server dynamically selects appropriate conference participants based on the aforementioned agenda candidates.

[0608] The server selects the most suitable participants from a database based on their job title, area of ​​expertise, and past meeting attendance history, such as a project manager or engineering team leader.

[0609] 5. Agenda generation and notification

[0610] The server automatically generates a meeting agenda by aggregating the topic candidates and selected participants.

[0611] The server notifies participants of the generated agenda via email, calendar invites, or other means.

[0612] Specific examples are shown below:

[0613] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[0614] 2. The server periodically collects this communication data and document data obtained from cloud storage.

[0615] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are saved.

[0616] 4. The server inputs the saved data into an NLP engine (SpaCy) and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0617] 5. From the analysis results, potential topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" are generated.

[0618] 6. The server selects the project manager, risk management officer, and project members for the agenda item "Analyzing the causes of delays in Project A." This information is dynamically selected based on employee information stored in the database.

[0619] 7. The server generates the meeting agenda based on the topic and the selected participants, as follows:

[0620] Agenda:

[0621] 1. Analysis of the cause of delay in Project A

[0622] 2. Cost optimization measures

[0623] 3. Next Major Milestone

[0624] Participants:

[0625] Taro Tanaka (Project Manager)

[0626] Hanako Suzuki (Risk Management Officer)

[0627] Ichiro Yamada (Project member)

[0628] 8. The server notifies participants of the generated agenda via email and calendar invites.

[0629] Prompt statement:

[0630] "If there is communication data indicating that Project A is delayed and document data such as a monthly report, please explain a system that dynamically selects appropriate agenda candidates and participants based on them, generates a meeting agenda, and notifies them."

[0631] In this way, the system reduces the workload involved in setting up and managing a conference, and supports the rapid and effective implementation of conferences.

[0632] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0633] Step 1:

[0634] Terminals generate business-related communication data such as chat systems and emails, as well as document data from cloud storage and file servers, on a daily basis. These data are imported. The inputs are communication data (e.g., chat messages, emails) and document data (e.g., text files, PDFs), which are then saved. The output is the saved communication data and document data.

[0635] Specific behavior:

[0636] An employee sends a chat message on the device.

[0637] The device will upload the new document to cloud storage.

[0638] Step 2:

[0639] The server connects to communication platforms (e.g., Slack) and document management systems (e.g., Google Drive) via APIs to periodically collect data. The input is data obtained from the communication platform and document management system, and the output is data stored in a database. This allows communication data and document data to be periodically updated.

[0640] Specific behavior:

[0641] The server retrieves the latest chat messages from the Slack API.

[0642] The server retrieves the latest document file from the Google Drive API.

[0643] The server stores the collected data in a database.

[0644] Step 3:

[0645] The server inputs the communication data and document data stored in the database into a natural language processing engine. The input is the communication data and document data stored in the database, and the output is the analysis results of natural language processing. An NLP engine (e.g., SpaCy) is used to perform data tokenization, part-of-speech tagging, named entity recognition (NER), etc.

[0646] Specific behavior:

[0647] The server passes a chat message saying "Project A is delayed" to the NLP engine.

[0648] The server passes the document "This Month's Expense Report" to the NLP engine.

[0649] The NLP engine extracts keywords such as "project delay" and "cost optimization."

[0650] Step 4:

[0651] The server generates topic candidates based on the analysis results from the NLP engine. The input is keywords and phrases processed through natural language processing, and the output is topic candidates. Therefore, new topic candidates are listed based on important entities and keywords.

[0652] Specific behavior:

[0653] The server generates an agenda item "Checking project progress" from the keyword "project delay."

[0654] The server generates a "cost optimization measures" agenda from the keyword "cost optimization."

[0655] Step 5:

[0656] The server dynamically selects appropriate meeting participants based on the agenda candidates. The input is the agenda candidates and an employee database, and the output is a list of selected participants. The optimal participants are selected based on the participants' job titles, areas of expertise, and past meeting history.

[0657] Specific behavior:

[0658] The server searches the database for the project manager, risk manager, and project members.

[0659] The server creates a list of participants suitable for the "Project Progress Review" agenda item.

[0660] Step 6:

[0661] The server generates a meeting agenda based on the candidate agenda items and the selected participant information. The input is the candidate agenda items and the participant list, and the output is the generated meeting agenda. This completes the meeting agenda.

[0662] Specific behavior:

[0663] The server generates a meeting agenda like this:

[0664] Agenda:

[0665] 1. Analysis of the cause of delay in Project A

[0666] 2. Cost optimization measures

[0667] 3. Next Major Milestone

[0668] Participants:

[0669] Taro Tanaka (Project Manager)

[0670] Hanako Suzuki (Risk Management Officer)

[0671] Ichiro Yamada (Project member)

[0672] Step 7:

[0673] The server notifies participants of the generated meeting agenda using notification means (e.g., email, calendar invitation). The input is the generated meeting agenda, and the output is the notified meeting agenda. This allows participants to receive information about the meeting in a timely manner.

[0674] Specific behavior:

[0675] The server sends the meeting agenda to the participants by email.

[0676] The server sends calendar invites and adds the meeting to the attendees' schedules.

[0677] The above are the specific processing steps of this system. By describing the operations performed at each step and their inputs and outputs in detail, the overall flow and processing content of the system will become clear.

[0678] (Application example 1)

[0679] 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."

[0680] Conventional meeting scheduling systems have the problem of being inefficient, taking a lot of time and effort to set up meeting agendas and select participants. In particular, in-factory operations require the ability to dynamically extract necessary agendas and participants in real time and smoothly schedule meetings. Furthermore, there are many situations where a quick response according to the situation is required, and current systems are insufficient.

[0681] 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.

[0682] In this invention, the server includes: means for collecting communication data and document data related to business operations; means for extracting keywords and agenda candidates from the communication data and document data using natural language processing; means for dynamically selecting relevant participants based on the agenda candidates; means for generating a meeting agenda based on the agenda candidates and the selected participants; means for notifying the meeting agenda; means for collecting operation data of a communication platform and equipment operated in a factory; means for generating agenda candidates using the operation data of the communication platform and equipment; means for selecting appropriate workers in the factory as participants based on the agenda candidates; and means for notifying the meeting agenda by voice or display via a robot. This automates the setting of a meeting agenda and the selection of appropriate participants, enabling fast and efficient meeting management in a factory.

[0683] "Business-related communication data" refers to data obtained from various communication means used in business, such as chat platforms and email.

[0684] "Document data" refers to electronic file data such as business-related reports and materials.

[0685] "Natural language processing" is a technology for analyzing text data and extracting useful information such as keywords and topic candidates.

[0686] A "keyword" is a specific word or phrase extracted from text data.

[0687] "Agenda candidates" are meeting topics generated based on the extracted keywords.

[0688] "Participants" are people or parties who need to attend a meeting.

[0689] A "meeting agenda" is a document that lists the proceedings and agenda items for a meeting.

[0690] "Notification" is the act of informing participants of the generated meeting agenda.

[0691] A "communication platform" is a system for exchanging information such as chat systems and e-mail.

[0692] "Equipment operation data" refers to data that indicates the operating status and performance of equipment used in a factory.

[0693] A "robot" is an automated mechanical device with a voice assistant function and a display function.

[0694] "Voice notification" is a method in which a robot communicates information using voice.

[0695] "Display" refers to the method by which a robot displays and communicates information on a display.

[0696] MODE FOR CARRYING OUT THE INVENTION

[0697] Overall system configuration

[0698] This invention requires a system for supporting the scheduling of meetings within a factory. This system collects and automatically analyzes business-related communication and document data to select topics and participants, and generates and notifies meeting agendas.

[0699] Hardware Configuration

[0700] 1. Server: The main computational unit responsible for data collection, natural language processing, topic extraction, participant selection, agenda generation and notifications.

[0701] 2. Terminal: A PC or mobile device where communication data and document data are stored.

[0702] 3. Robot: A device that announces the meeting agenda by voice or display.

[0703] Software Configuration

[0704] 1. Natural language processing engine (e.g., nltk, sklearn): Software for extracting keywords and topic candidates from collected data.

[0705] 2. Email sending library (e.g. smtplib): Software for notifying participants of the generated agenda.

[0706] 3. Generative AI model: This is the AI ​​model used to analyze the collected data and generate the agenda.

[0707] Processing flow and operation explanation

[0708] Data collection

[0709] The server periodically uses APIs to collect data from communication platforms (chat systems and email) used within the factory and document data stored in the cloud. It also collects operational data from the equipment used in the factory. All of this data is stored in a database.

[0710] Natural Language Processing

[0711] The server analyzes the collected communication and document data using a natural language processing engine, tokenizing the text data and tagging it by part of speech to extract important keywords and potential topics.

[0712] Agenda extraction

[0713] The server generates candidate topics based on the extracted keywords and phrases. For example, keywords such as "maintenance delay" and "progress report" can be used to generate specific topics such as "checking the robot maintenance status" and "checking the product line progress."

[0714] Participant selection

[0715] The server dynamically selects appropriate meeting participants based on the generated agenda candidates, and automatically selects factory workers based on their past meeting participation history, job titles, and areas of expertise.

[0716] Agenda generation and notifications

[0717] The server automatically generates a meeting agenda based on the generated topic and selected participants. The generated agenda is notified to participants via email or calendar invitation. It is also notified to operators via voice assistants and displays on the factory robots.

[0718] Specific examples

[0719] For example, at the end of the day, the server collects chat messages such as "Robot A's maintenance is behind schedule" and data such as "Today's product line progress report." Based on this data, agenda candidates "Checking the maintenance status of Robot A" and "Checking the product line progress" are generated. Based on these agendas, maintenance personnel and production line leaders are selected as participants. The meeting agenda is generated as follows:

[0720] Agenda:

[0721] 1. Check the maintenance status of Robot A

[0722] 2. Check the progress of the product line

[0723] 3. Next Major Milestone

[0724] Participants:

[0725] Maintenance personnel

[0726] Production Line Leader

[0727] factory manager

[0728] The generated agenda is notified to participants via email, and the robot notifies the operator via voice and display.

[0729] Prompt Sentence Examples

[0730] "Automatically generate a meeting agenda. Use the following communication data: 'Robot A's maintenance is overdue', 'We need to check the progress of the production line'"

[0731] This automates the setting of meeting agendas and the selection of appropriate participants, enabling meetings within the factory to be run quickly and efficiently.

[0732] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0733] Step 1: Data collection

[0734] The server collects communication data from communication platforms and emails used within the factory. It also collects document data from cloud storage, including chat messages and reports. Specifically, it accesses these platforms via API and periodically retrieves the required data. The input is communication data and document data, and the output is the collected data.

[0735] Step 2: Natural Language Processing

[0736] The server inputs the collected communication data and document data into a natural language processing engine (e.g., nltk, sklearn) to analyze the text data. Here, the text is tokenized and tagged with parts of speech to extract important keywords and phrases. The input is the collected data, and the output is the extracted keywords and phrases. Specifically, TF-IDF vectorization and KMeans clustering are performed.

[0737] Step 3: Automatically extracting agenda items

[0738] The server generates meeting agenda candidates based on keywords and phrases extracted through natural language processing. For example, it generates the agenda "Checking the robot's maintenance status" from the keyword "maintenance delay." The input is the extracted keywords and phrases, and the output is the agenda candidate.

[0739] Step 4: Dynamic Participant Management

[0740] The server automatically selects appropriate meeting participants based on the generated agenda candidates. It selects the best participants for each agenda based on their past meeting participation history and job titles. The input is the agenda candidates, and the output is the selected participants. Specifically, it queries and filters relevant participant information from a database.

[0741] Step 5: Generate the agenda

[0742] The server automatically generates a meeting agenda based on the candidate agenda items and selected participants. The input is the candidate agenda items and participant information, and the output is the generated meeting agenda. Specifically, the server creates the agenda document using a text template.

[0743] Step 6: Announce the agenda

[0744] The server notifies the generated meeting agenda. The agenda is notified to participants using email notifications and calendar invitations. It also notifies operators via voice or display on the factory robots. The input is the generated meeting agenda, and the output is the notified agenda. Specifically, emails are sent using an email sending library (e.g., smtplib), and voice instructions and display are given through the robot's API.

[0745] 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.

[0746] This invention relates to a system that automates the setting of meeting agendas and participant selection, as well as recognizes user emotions to improve the quality of meetings. This system collects business-related communication data and document data, extracts candidate topics using natural language processing, dynamically selects participants suitable for the topics, and combines an emotion engine to recognize user emotions and reflect them in the topics and agendas.

[0747] The overall system configuration is as follows:

[0748] 1. Data Collection

[0749] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[0750] The server retrieves this communication data via API, parses it in JSON format, and stores it in a database. It also retrieves document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in a database.

[0751] 2. Natural Language Processing

[0752] The server inputs the collected communication and document data into a natural language processing (NLP) engine.

[0753] The server performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and significant phrases.

[0754] 3. Emotion Engine

[0755] The server sends the text extracted from the communication data and document data to an emotion engine to analyze the user's emotion.

[0756] The server recognizes emotions such as positive, negative, and neutral, and stores the results in a database.

[0757] 4. Automatic extraction of agenda items

[0758] The server applies an algorithm to generate topic candidates based on the keywords obtained from the NLP analysis results and the analysis results of the emotion engine.

[0759] The server creates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and emotional information, and stores them in a database.

[0760] 5. Dynamic Participant Management

[0761] The server selects the most suitable participants for the generated agenda candidates.

[0762] The server retrieves employee information such as past meeting history, areas of expertise, and job titles from the database, and dynamically selects the most appropriate participants for each agenda item. For example, for "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[0763] 6. Generate a meeting agenda

[0764] The server automatically generates a meeting agenda based on the selected topic and participant list.

[0765] The server embeds the topic and participant information in the agenda template and modifies the generated agenda to reflect the user's emotional information.

[0766] Based on the results of the emotion engine, the server prioritizes topics that are generating high emotions and reflects them in the agenda.

[0767] 7. Sending Notifications

[0768] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[0769] The server interfaces with the mail server to send emails containing detailed agendas and meeting information, and uses the calendar system to generate meeting invites.

[0770] Specific examples

[0771] Here is a concrete example of how the system works:

[0772] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[0773] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[0774] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[0775] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0776] 5. At the same time, the sentiment engine determines that the user has a strong negative sentiment towards the message "Project A is delayed."

[0777] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[0778] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[0779] 8. The server generates the meeting agenda based on the selected topic and participant list, as follows:

[0780] Agenda:

[0781] 1. Analysis of the cause of delay in Project A

[0782] 2. Cost optimization measures

[0783] 3. Next Major Milestone

[0784] Participants:

[0785] Taro Tanaka (Project Manager)

[0786] Hanako Suzuki (Risk Management Officer)

[0787] Ichiro Yamada (Project member)

[0788] 9. The server notifies participants of the generated agenda via email and calendar invites.

[0789] 10. Based on the analysis results of the emotion engine, the server makes modifications such as placing topics for which the user has particularly negative emotions at the top of the agenda.

[0790] In this way, the system reduces the workload involved in setting up and managing meetings, and supports prompt and effective meeting conduct through emotionally sensitive agenda setting.

[0791] The processing flow will be explained below.

[0792] Step 1:

[0793] Data collection

[0794] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[0795] The server obtains this communication data via API, parses it in JSON format, and stores it in a database.

[0796] The server also collects document data from cloud storage (e.g., Google Drive, SharePoint) via API and stores it in a database.

[0797] Step 2:

[0798] Performing Natural Language Processing (NLP)

[0799] The server inputs the collected communication and document data into a natural language processing (NLP) engine.

[0800] The server performs text tokenization, morphological analysis, part-of-speech tagging, and named entity recognition (NER) to extract important keywords and phrases.

[0801] Step 3:

[0802] emotion recognition

[0803] The server sends the communication data and the text extracted from the document data to the emotion engine.

[0804] The server uses an emotion engine to analyze the sentiment of each piece of text and classify it into categories such as positive, negative, or neutral.

[0805] The server stores the emotion recognition results in a database.

[0806] Step 4:

[0807] Automatic extraction of agenda items

[0808] The server runs an algorithm that integrates the results of NLP analysis and emotion recognition to generate topic candidates.

[0809] The server generates candidate topics such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and emotional information, and stores them in a database.

[0810] Step 5:

[0811] Dynamic Participant Management

[0812] The server dynamically selects the most suitable participants for the generated agenda candidates.

[0813] The server retrieves information such as employee job titles, areas of expertise, and past meeting participation history from a database and selects the most appropriate participants for each agenda item.

[0814] For example, for the agenda item "Analysis of the causes of project delays," the project manager, risk management officer, and project members are selected.

[0815] Step 6:

[0816] Generate a meeting agenda

[0817] The server automatically generates a meeting agenda based on the selected topic and participant list.

[0818] Based on the results of the emotion engine, the server makes adjustments such as placing topics with high negative emotions at the beginning of the agenda.

[0819] Step 7:

[0820] Sending notifications

[0821] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[0822] The server interfaces with a mail server to send emails containing detailed agendas and meeting information, and uses a calendar system to generate meeting invites.

[0823] Specific examples

[0824] Here is a concrete example of how the system works:

[0825] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[0826] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[0827] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[0828] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0829] 5. At the same time, the emotion engine recognizes emotions from the user's communication data and document data and determines that the text "Project A is delayed" indicates a strong negative emotion.

[0830] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[0831] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[0832] 8. The server generates the meeting agenda based on the selected topic and participant list, as follows:

[0833] Agenda:

[0834] 1. Analysis of the cause of delay in Project A

[0835] 2. Cost optimization measures

[0836] 3. Next Major Milestone

[0837] Participants:

[0838] Taro Tanaka (Project Manager)

[0839] Hanako Suzuki (Risk Management Officer)

[0840] Ichiro Yamada (Project member)

[0841] 9. The server notifies participants of the generated agenda via email and calendar invites.

[0842] 10. Based on the analysis results of the emotion engine, the server makes adjustments to improve the quality of the meeting, such as placing topics for which users have particularly negative emotions at the top of the agenda.

[0843] In this way, the system reduces the workload involved in setting up and managing meetings, and supports prompt and effective meeting conduct through emotionally sensitive agenda setting.

[0844] Example 2

[0845] 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."

[0846] Conventionally, setting the agenda and selecting participants for a meeting have been done manually, which has the problem of requiring a great deal of effort and time. Furthermore, it has been difficult to consider the user's feelings when setting the agenda, which has also made it difficult to improve the quality of the meeting. The present invention aims to solve these problems and improve the quality of meetings through automated agenda setting and participant selection.

[0847] 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.

[0848] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and selected participants, means for notifying the meeting agenda, means for analyzing user emotions from the communication data and document data, and means for determining the priority of agenda items based on the emotion analysis results. This reduces the workload associated with setting and managing meetings, and enables more effective meetings by setting agendas that take emotions into consideration.

[0849] "Business-related communications data" refers to messages and their metadata exchanged through chat platforms and email systems used within a company.

[0850] "Document data" refers to digital information such as document files, reports, and presentation materials obtained from cloud storage or corporate databases.

[0851] "Natural language processing" refers to techniques and algorithms that enable computers to understand, interpret, and generate human language.

[0852] "Keywords" refer to important words and phrases extracted from communication data and document data through natural language processing.

[0853] "Agenda candidates" refer to items to be discussed at a meeting that are generated based on keywords extracted from communication data and document data.

[0854] "Participants" refers to employees or personnel who should be invited to the meeting.

[0855] "Meeting agenda" refers to a meeting schedule generated based on candidate topics and a selected participant list.

[0856] "Sentiment analysis" refers to the process of identifying emotions contained in user communication and document data and classifying those emotions as positive, negative, neutral, etc.

[0857] "Priority" refers to the criteria used to determine the order in which meeting topics are discussed based on the results of sentiment analysis.

[0858] This invention relates to a system that automates the setting of meeting agendas and participant selection, as well as recognizes user emotions to improve the quality of meetings. This system collects business-related communication data and document data, extracts candidate topics using natural language processing (NLP), dynamically selects participants suitable for the topics, and combines an emotion engine to recognize user emotions and reflect them in the topics and agendas.

[0859] Overall system configuration

[0860] Data collection

[0861] The server collects communication data from the company's internal chat platform (e.g., Slack, Microsoft Teams) and email system. It obtains the data via API, parses it in JSON format, and stores it in a database. It also obtains document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in a database.

[0862] Natural Language Processing (NLP)

[0863] The server inputs the collected communication and document data into an NLP engine (e.g., Google Cloud NLP, AWS Comprehend), which performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and important phrases.

[0864] Emotion Engine

[0865] The server sends the text extracted from the communication data and document data to an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The results are classified into emotion categories such as positive, negative, and neutral, and stored in a database.

[0866] Automatic extraction of agenda items

[0867] The server applies an algorithm based on the results of NLP analysis and sentiment analysis to generate topic candidates. For example, it generates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and sentiment information.

[0868] Dynamic Participant Management

[0869] The server selects the most suitable participants for each candidate agenda item. Employee information such as past meeting history, areas of expertise, and job titles is retrieved from the database, and the most suitable participants are dynamically selected for each agenda item. For example, for the agenda item "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[0870] Generate a meeting agenda

[0871] The server automatically generates a meeting agenda based on the selected topic and participant list. It embeds the topic and participant information in an agenda template and modifies the generated agenda to reflect emotional information. Based on the results of the emotional analysis, it prioritizes topics for which users have particularly strong emotions and reflects them in the agenda.

[0872] Sending notifications

[0873] The server notifies the generated meeting agenda to the relevant participants via email and calendar invites. It works with the mail server to send emails containing the detailed agenda and meeting information, and creates meeting invites using the calendar system.

[0874] Specific examples

[0875] Here is a concrete example of how the system works:

[0876] 1. Work-related chat messages and emails accumulate daily on devices (employee computers and mobile devices).

[0877] 2. The server collects this communication data and document data from cloud storage on a daily basis.

[0878] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[0879] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[0880] 5. At the same time, the emotion engine determines that the user is expressing strong negative emotion towards the message "Project A is delayed."

[0881] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[0882] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[0883] 8. The server generates a meeting agenda based on the selected topic and participant list:

[0884] Agenda:

[0885] 1. Analysis of the cause of delay in Project A

[0886] 2. Cost optimization measures

[0887] 3. Next Major Milestone

[0888] Participants:

[0889] Taro Tanaka (Project Manager)

[0890] Hanako Suzuki (Risk Management Officer)

[0891] Ichiro Yamada (Project member)

[0892] 9. The server notifies participants of the generated agenda via email and calendar invites.

[0893] 10. Based on the analysis results of the emotion engine, the server places topics for which the user has particularly negative emotions at the top of the agenda.

[0894] Prompt Sentence Examples

[0895] The following prompts are used to input the generative AI model:

[0896] Please create a program that extracts topics from business chat messages containing negative sentiment, such as "Project A is delayed," and generates an agenda by selecting the most appropriate meeting participants. Please also provide an example of the generated agenda.

[0897] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0898] Step 1: Data collection

[0899] The server collects communication data from the company's internal chat platforms (e.g., chat service A, chat service B) and email systems. Specifically, it uses APIs to obtain text-based message data and email data from these services, parses it into JSON format, and stores it in a database.

[0900] Input: Data obtained from APIs of chat platforms and email systems.

[0901] Data processing: Convert the acquired data into JSON format.

[0902] Output: JSON formatted communication and email data, stored in a database.

[0903] Step 2: Collect document data from cloud storage

[0904] The server retrieves document data (PDF, Word files, etc.) from cloud storage (e.g., storage service A, storage service B) and stores it in a database. It periodically scans documents using the storage API to retrieve new and updated documents.

[0905] Input: Data obtained from cloud storage API.

[0906] Data processing: Convert the acquired document data into JSON format.

[0907] Output: Document data in JSON format, stored in the database.

[0908] Step 3: Natural Language Processing (NLP)

[0909] The server inputs the communication and document data stored in the database into NLP engines (e.g., NLP engine A, NLP engine B). The NLP engines perform text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract important keywords and phrases.

[0910] Input: Correspondence and document data extracted from the database.

[0911] Data operations: text tokenization, part-of-speech tagging, NER processing.

[0912] Output: Keywords and important phrases, stored in a database.

[0913] Step 4: Emotion analysis using the emotion engine

[0914] The server sends the text analyzed by the NLP engine to an emotion engine (e.g., emotion analysis service A, emotion analysis service B) to analyze the user's emotions. It classifies the emotions into positive, negative, neutral, and other emotional characteristics, and stores the results in a database.

[0915] Input: Keywords and phrases obtained as a result of analysis by the NLP engine.

[0916] Data Computation: Determine the sentiment category of the text through sentiment analysis.

[0917] Output: Sentiment analysis results (positive, negative, neutral), stored in a database.

[0918] Step 5: Automatically extracting topic candidates

[0919] The server generates candidate topics by applying an algorithm based on the results of NLP analysis and sentiment analysis. This extracts topics with keywords such as "project delay" and "cost optimization," and generates specific themes such as "analysis of the causes of project delays" and "cost optimization measures."

[0920] Input: NLP analysis results, sentiment analysis results.

[0921] Data computation: Applying algorithms that combine keywords and sentiment information to generate agendas.

[0922] Output: Agenda candidates, stored in a database.

[0923] Step 6: Dynamic Participant Management

[0924] The server selects the most suitable participants for each candidate topic. Appropriate employees are dynamically selected based on information such as past meeting history, areas of expertise, and job titles. For example, for the topic "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[0925] Input: Agenda candidate, employee information retrieved from database.

[0926] Data calculation: Applying algorithms to select the most suitable participants based on meeting history, areas of expertise, and job title information.

[0927] Output: Participant list, stored in database.

[0928] Step 7: Generate the meeting agenda

[0929] The server automatically generates a meeting agenda based on the selected topics and participant list. It embeds the topic and participant information in a template and prioritizes topics about which users have particularly strong feelings, reflecting them in the agenda.

[0930] Input: candidate agenda items, participant list, sentiment analysis results.

[0931] Data operations: Embedding information into templates and prioritizing agenda items.

[0932] Output: The generated meeting agenda, stored in a database.

[0933] Step 8: Sending notifications

[0934] The server notifies the generated meeting agenda to interested participants via email and calendar invites. It works with a mail server to send emails containing the detailed agenda and meeting information, and uses a calendar system to create meeting invites.

[0935] Input: The generated meeting agenda.

[0936] Data Computing: Notification generation via email and calendar systems.

[0937] Output: Notification emails and calendar events sent.

[0938] (Application example 2)

[0939] 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."

[0940] Troubles and delays at production sites are issues that require prompt and appropriate countermeasures. However, the traditional method of setting up meetings and selecting participants places a heavy workload on employees, making efficient management difficult, especially in large factories. Furthermore, meetings often fail to take into account the feelings of employees when setting agendas and topics, resulting in ineffective discussions. This hinders the efficient operation of factories, creating a challenge.

[0941] 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.

[0942] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and the selected participants, means for analyzing user emotions, means for modifying the meeting agenda based on emotion information, and means for notifying the meeting agenda. This improves the efficiency of meeting setup and management for troubleshooting at production sites and enables effective discussions that take user emotions into consideration.

[0943] "Business-related communication data" refers to communication information used within a company, such as chat, email, and monitoring data.

[0944] "Document data" refers to document information such as reports, memos, and files stored in online storage or local storage.

[0945] "Natural language processing" is an information processing technology for analyzing and understanding human language.

[0946] "Keywords" are important words and phrases extracted from the data.

[0947] "Agenda candidate" refers to themes and issues extracted through natural language processing.

[0948] "Relevant participants" are people or positions who should attend the meeting based on the proposed agenda.

[0949] A "meeting agenda" is a document that includes the agenda and attendee list to direct the progress of a meeting.

[0950] "User emotion" is the emotional state of the user as assessed by the emotion analysis engine.

[0951] "Emotion information" is data obtained as a result of analyzing the user's emotions.

[0952] An "emotion engine" is an information processing system for analyzing emotions from communication data and voice data.

[0953] "Dynamic selection of relevant participants based on topic candidates" is a process of selecting the most suitable participants in real time, taking into account their past history and areas of expertise related to the topic.

[0954] A "conversation platform" is a communication tool such as a chat tool or video conferencing software used within a company.

[0955] "Online storage" is a data storage service that can be accessed via the Internet.

[0956] "Notifying a meeting agenda" refers to the act of informing relevant parties of the generated agenda via email or calendar invitation.

[0957] This invention is a conference management system for quickly and effectively resolving problems and delays at production sites. This system collects business-related communication data and document data, extracts candidate topics using natural language processing, dynamically selects participants, and uses an emotion engine to recognize user emotions and reflect them in the topics and agenda.

[0958] The server first collects real-time data from sensors and surveillance cameras. It also obtains production data and trouble reports from the factory management system (MES) via API and stores them in a database. Based on the collected data, the server uses a natural language processing engine (e.g., NLTK, Transformers) to extract important keywords and phrases and generate candidate topics.

[0959] The emotion engine (e.g., Emotion AI) analyzes emotions from collected data and user voice and chat. The emotion results are stored in a database as emotional states, such as positive, negative, or neutral. The server then modifies the meeting agenda based on this emotional information, prioritizing topics that indicate negative emotions in the agenda list.

[0960] The generated meeting agenda is then sent to participants via email or a calendar invitation system, using common business communication tools (e.g., Microsoft Outlook, Google Calendar).

[0961] For example, the server generates candidate topics such as "malfunction of machine A" or "delay on production line B" and selects appropriate participants such as machine maintenance personnel and operators. This information is then reflected in the negative emotions analyzed by the emotion engine, and notified as the generated meeting agenda.

[0962] As an example, the following prompt sentences can be input to a generative AI model:

[0963] "An AI robot collects sensor data and trouble reports from within a factory, inputs this data into a natural language processing and sentiment analysis engine, selects appropriate meeting topics and participants, and automatically generates and notifies meeting agendas. We will show you a specific example of how this works."

[0964] In this way, the present invention can realize an effective conference management system in the production site and improve the efficiency of troubleshooting.

[0965] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0966] Step 1:

[0967] The server collects real-time data from sensors and surveillance cameras. It acquires the data through the sensor API and the surveillance camera API and stores it in a database. The input is sensor data and video data, and the output is the collected data stored in the database.

[0968] Step 2:

[0969] The server obtains production data and trouble reports from the factory management system (MES) via API. This data is also stored in the database. The input is production data and trouble reports, and the output is business data stored in the database.

[0970] Step 3:

[0971] The server inputs collected data (sensor data, production data, trouble reports) into a natural language processing (NLP) engine, which performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and important phrases. The input is collected data, and the output is keywords and topic candidates.

[0972] Step 4:

[0973] The server inputs the collected communication data and the user's voice and chat data into an emotion engine (e.g., Emotion AI) to analyze the user's emotions. The emotion engine analyzes the emotional state, such as positive, negative, or neutral, and stores the results in a database. The input is communication data and voice data, and the output is the emotion analysis results.

[0974] Step 5:

[0975] The server generates topic candidates and dynamically selects relevant participants based on the processed NLP analysis results and sentiment analysis results. The inputs are the NLP analysis results and sentiment analysis results, and the outputs are topic candidates and a list of participants.

[0976] Step 6:

[0977] The server automatically generates a meeting agenda based on the generated topics and the selected participant list. It considers emotional information and prioritizes topics that show negative emotions. The inputs are candidate topics and the participant list, and the output is a revised meeting agenda.

[0978] Step 7:

[0979] The server notifies participants of the generated meeting agenda via email and / or calendar invites. The input to the notification system is the modified meeting agenda, and the output is the meeting notice sent to participants.

[0980] 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.

[0981] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0982] In the above embodiment, an example 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.

[0983] [Third embodiment]

[0984] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0985] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0986] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0987] The 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.

[0988] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0989] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0990] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0991] 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.

[0992] 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.

[0993] 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.

[0994] 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.

[0995] 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."

[0996] The present invention relates to a system that improves business efficiency by automating the setting of meeting agendas and participant selection. This system involves a series of operations: collecting business-related communication data and document data, extracting candidate agendas using natural language processing, dynamically selecting participants suitable for the agendas, and finally generating and notifying participants of the meeting agenda.

[0997] The overall system configuration is as follows:

[0998] 1. Data Collection

[0999] The server automatically collects data from communication platforms (e.g., chat systems, email) and document management systems (e.g., cloud storage, file servers) used in business.

[1000] The server connects to these platforms via API, periodically retrieves data, and stores it in a database.

[1001] 2. Natural Language Processing

[1002] The server analyzes the collected communication and document data using a natural language processing (NLP) engine to extract important keywords and potential topics.

[1003] NLP engines extract semantic information from data using techniques such as text tokenization, part-of-speech tagging, and named entity recognition (NER).

[1004] 3. Automatic extraction of agenda items

[1005] The server generates topic candidates based on the extracted keywords and phrases.

[1006] For example, from keywords such as "project delay" and "progress," topics such as "checking project progress" and "analyzing the cause of the delay" are generated.

[1007] 4. Dynamic Participant Management

[1008] The server automatically selects appropriate conference participants based on the aforementioned agenda candidates.

[1009] The server selects the most suitable participants based on information such as the participants' job titles, areas of expertise, and past meeting participation history. For example, members suitable for the agenda, such as "project managers" or "engineering team leaders," are selected.

[1010] 5. Agenda generation and notification

[1011] The server aggregates the topic candidates and selected participants and automatically generates the meeting agenda.

[1012] The server notifies participants of the generated agenda using methods such as email or calendar invitations.

[1013] Specific examples

[1014] Here is a concrete example of how the system works:

[1015] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[1016] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[1017] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[1018] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1019] 5. From the analysis results, potential topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" are generated.

[1020] 6. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A." This information is dynamically selected based on employee information stored in the database.

[1021] 7. The server generates the meeting agenda based on the topic and the selected participants, as follows:

[1022] Agenda:

[1023] 1. Analysis of the cause of delay in Project A

[1024] 2. Cost optimization measures

[1025] 3. Next Major Milestone

[1026] Participants:

[1027] Taro Tanaka (Project Manager)

[1028] Hanako Suzuki (Risk Management Officer)

[1029] Ichiro Yamada (Project member)

[1030] 8. The server notifies participants of the generated agenda via email and calendar invites.

[1031] In this way, the system reduces the workload involved in setting up and managing meetings, and supports the rapid and effective implementation of meetings.

[1032] The processing flow will be explained below.

[1033] Step 1:

[1034] Data collection

[1035] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[1036] The server obtains this communication data via API, parses it in JSON format, and stores it in a database.

[1037] The server also retrieves document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in the database as well.

[1038] Step 2:

[1039] Analysis using natural language processing

[1040] The server inputs the collected communication data and document data into a natural language processing engine.

[1041] The server performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and significant phrases.

[1042] Step 3:

[1043] Automatic extraction of agenda items

[1044] The server applies an algorithm to generate topic candidates based on the keywords obtained from the NLP analysis results.

[1045] The server creates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization," and stores these in a database.

[1046] Step 4:

[1047] Dynamic Participant Management

[1048] The server selects the most suitable participants for the generated agenda candidates.

[1049] The server retrieves employee information such as past meeting history, areas of expertise, and job titles from a database, and dynamically selects participants deemed most suitable for each agenda item.

[1050] For example, the project manager, risk management officer, and project members are selected for the "cause analysis of project delays."

[1051] Step 5:

[1052] Generate a meeting agenda

[1053] The server automatically generates a meeting agenda based on the selected topic and participant list.

[1054] The server embeds the topic and participant information in the agenda template to create a formal meeting agenda.

[1055] Step 6:

[1056] Sending notifications

[1057] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[1058] The server interfaces with a mail server to send emails containing detailed agendas and meeting information, and uses a calendar system to generate meeting invites.

[1059] Example 1

[1060] 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."

[1061] In today's business environment, many companies and organizations generate large amounts of communication and document data. It is difficult to extract important agenda items from this data, select appropriate meeting participants, and efficiently schedule meetings, resulting in a decline in business efficiency. To solve this problem, a system that can effectively utilize communication and document data and automate meeting scheduling is needed.

[1062] 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.

[1063] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and the selected participants, and means for notifying the meeting agenda. This allows users to automate a series of processes from data collection to meeting setup and notification of the meeting agenda.

[1064] "Business-related communications data" refers to data including text messages, images, videos, and audio data sent and received via messaging platforms, emails, and other means used in business.

[1065] "Document data" refers to business-related data such as text documents, spreadsheets, presentation files, image files, and PDFs that are stored on remote storage or file servers.

[1066] "Natural language processing" is a technology for analyzing text data and extracting, generating, understanding, and translating information, and includes processes such as text tokenization, part-of-speech tagging, and named entity recognition.

[1067] "Keywords" are important words and phrases extracted from communication data and document data, and are information used to generate topic candidates.

[1068] "Agenda candidates" are themes or items to be discussed in the meeting, generated from the extracted keywords and phrases.

[1069] "Participants" are individuals selected to attend meetings related to the agenda, and are dynamically selected based on job title, area of ​​expertise, etc.

[1070] A "meeting agenda" is a list of specific topics or items to be discussed at a meeting, and is used to notify participants and guide the meeting.

[1071] "Notification medium" means the media and protocols for appropriately informing meeting participants of the generated meeting agenda, including email, calendar invites, messaging platform notifications, etc.

[1072] The present invention relates to a system that improves business efficiency by automating the setting of meeting agendas and participant selection. This system includes a series of operations: collecting business-related communication data and document data, extracting candidate agendas using natural language processing, dynamically selecting participants suitable for the agendas, and finally generating and notifying the meeting agenda.

[1073] The overall system configuration is as follows:

[1074] 1. Data Collection

[1075] Devices generate communication data such as chat systems and emails used for business, as well as document data stored in cloud storage and file servers every day. This data is sent from each employee's PC or mobile device. The server connects to communication platforms and document management systems via APIs to periodically collect this data.

[1076] Specific use cases include using Slack's API to retrieve chat messages and Google Drive's API to gather document data.

[1077] The server stores the retrieved data in a database.

[1078] 2. Natural Language Processing

[1079] The server inputs the communication data and document data stored in the database into a natural language processing (NLP) engine.

[1080] Examples of NLP engines used include SpaCy and NLTK.

[1081] Specific processing includes text tokenization, part-of-speech tagging, and named entity recognition (NER).

[1082] For example, if there is a chat message that says "Project A is delayed," the NLP engine breaks it down into words such as "project," "A," "delayed," and "being," and extracts important entities such as "Project A" and "delayed."

[1083] 3. Automatic extraction of agenda items

[1084] The server generates topic candidates based on the extracted keywords and phrases.

[1085] For example, from keywords such as "project delay" and "progress," topics such as "checking project progress" and "analyzing the cause of the delay" are generated.

[1086] 4. Dynamic Participant Management

[1087] The server dynamically selects appropriate conference participants based on the aforementioned agenda candidates.

[1088] The server selects the most suitable participants from a database based on their job title, area of ​​expertise, and past meeting attendance history, such as a project manager or engineering team leader.

[1089] 5. Agenda generation and notification

[1090] The server automatically generates a meeting agenda by aggregating the topic candidates and selected participants.

[1091] The server notifies participants of the generated agenda via email, calendar invites, or other means.

[1092] Specific examples are shown below:

[1093] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[1094] 2. The server periodically collects this communication data and document data obtained from cloud storage.

[1095] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are saved.

[1096] 4. The server inputs the saved data into an NLP engine (SpaCy) and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1097] 5. From the analysis results, potential topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" are generated.

[1098] 6. The server selects the project manager, risk management officer, and project members for the agenda item "Analyzing the causes of delays in Project A." This information is dynamically selected based on employee information stored in the database.

[1099] 7. The server generates the meeting agenda based on the topic and the selected participants, as follows:

[1100] Agenda:

[1101] 1. Analysis of the cause of delay in Project A

[1102] 2. Cost optimization measures

[1103] 3. Next Major Milestone

[1104] Participants:

[1105] Taro Tanaka (Project Manager)

[1106] Hanako Suzuki (Risk Management Officer)

[1107] Ichiro Yamada (Project member)

[1108] 8. The server notifies participants of the generated agenda via email and calendar invites.

[1109] Prompt statement:

[1110] "If there is communication data indicating that Project A is delayed and document data such as a monthly report, please explain a system that dynamically selects appropriate agenda candidates and participants based on them, generates a meeting agenda, and notifies them."

[1111] In this way, the system reduces the workload involved in setting up and managing a conference, and supports the rapid and effective implementation of conferences.

[1112] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1113] Step 1:

[1114] Terminals generate business-related communication data such as chat systems and emails, as well as document data from cloud storage and file servers, on a daily basis. These data are imported. The inputs are communication data (e.g., chat messages, emails) and document data (e.g., text files, PDFs), which are then saved. The output is the saved communication data and document data.

[1115] Specific behavior:

[1116] An employee sends a chat message on the device.

[1117] The device will upload the new document to cloud storage.

[1118] Step 2:

[1119] The server connects to communication platforms (e.g., Slack) and document management systems (e.g., Google Drive) via APIs to periodically collect data. The input is data obtained from the communication platform and document management system, and the output is data stored in a database. This allows communication data and document data to be periodically updated.

[1120] Specific behavior:

[1121] The server retrieves the latest chat messages from the Slack API.

[1122] The server retrieves the latest document file from the Google Drive API.

[1123] The server stores the collected data in a database.

[1124] Step 3:

[1125] The server inputs the communication data and document data stored in the database into a natural language processing engine. The input is the communication data and document data stored in the database, and the output is the analysis results of natural language processing. An NLP engine (e.g., SpaCy) is used to perform data tokenization, part-of-speech tagging, named entity recognition (NER), etc.

[1126] Specific behavior:

[1127] The server passes a chat message saying "Project A is delayed" to the NLP engine.

[1128] The server passes the document "This Month's Expense Report" to the NLP engine.

[1129] The NLP engine extracts keywords such as "project delay" and "cost optimization."

[1130] Step 4:

[1131] The server generates topic candidates based on the analysis results from the NLP engine. The input is keywords and phrases processed through natural language processing, and the output is topic candidates. Therefore, new topic candidates are listed based on important entities and keywords.

[1132] Specific behavior:

[1133] The server generates an agenda item "Checking project progress" from the keyword "project delay."

[1134] The server generates a "cost optimization measures" agenda from the keyword "cost optimization."

[1135] Step 5:

[1136] The server dynamically selects appropriate meeting participants based on the agenda candidates. The input is the agenda candidates and an employee database, and the output is a list of selected participants. The optimal participants are selected based on the participants' job titles, areas of expertise, and past meeting history.

[1137] Specific behavior:

[1138] The server searches the database for the project manager, risk manager, and project members.

[1139] The server creates a list of participants suitable for the "Project Progress Review" agenda item.

[1140] Step 6:

[1141] The server generates a meeting agenda based on the candidate agenda items and the selected participant information. The input is the candidate agenda items and the participant list, and the output is the generated meeting agenda. This completes the meeting agenda.

[1142] Specific behavior:

[1143] The server generates a meeting agenda like this:

[1144] Agenda:

[1145] 1. Analysis of the cause of delay in Project A

[1146] 2. Cost optimization measures

[1147] 3. Next Major Milestone

[1148] Participants:

[1149] Taro Tanaka (Project Manager)

[1150] Hanako Suzuki (Risk Management Officer)

[1151] Ichiro Yamada (Project member)

[1152] Step 7:

[1153] The server notifies participants of the generated meeting agenda using notification means (e.g., email, calendar invitation). The input is the generated meeting agenda, and the output is the notified meeting agenda. This allows participants to receive information about the meeting in a timely manner.

[1154] Specific behavior:

[1155] The server sends the meeting agenda to the participants by email.

[1156] The server sends calendar invites and adds the meeting to the attendees' schedules.

[1157] The above are the specific processing steps of this system. By describing the operations performed at each step and their inputs and outputs in detail, the overall flow and processing content of the system will become clear.

[1158] (Application example 1)

[1159] 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."

[1160] Conventional meeting scheduling systems have the problem of being inefficient, taking a lot of time and effort to set up meeting agendas and select participants. In particular, in-factory operations require the ability to dynamically extract necessary agendas and participants in real time and smoothly schedule meetings. Furthermore, there are many situations where a quick response according to the situation is required, and current systems are insufficient.

[1161] 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.

[1162] In this invention, the server includes: means for collecting communication data and document data related to business operations; means for extracting keywords and agenda candidates from the communication data and document data using natural language processing; means for dynamically selecting relevant participants based on the agenda candidates; means for generating a meeting agenda based on the agenda candidates and the selected participants; means for notifying the meeting agenda; means for collecting operation data of a communication platform and equipment operated in a factory; means for generating agenda candidates using the operation data of the communication platform and equipment; means for selecting appropriate workers in the factory as participants based on the agenda candidates; and means for notifying the meeting agenda by voice or display via a robot. This automates the setting of a meeting agenda and the selection of appropriate participants, enabling fast and efficient meeting management in a factory.

[1163] "Business-related communication data" refers to data obtained from various communication means used in business, such as chat platforms and email.

[1164] "Document data" refers to electronic file data such as business-related reports and materials.

[1165] "Natural language processing" is a technology for analyzing text data and extracting useful information such as keywords and topic candidates.

[1166] A "keyword" is a specific word or phrase extracted from text data.

[1167] "Agenda candidates" are meeting topics generated based on the extracted keywords.

[1168] "Participants" are people or parties who need to attend a meeting.

[1169] A "meeting agenda" is a document that lists the proceedings and agenda items for a meeting.

[1170] "Notification" is the act of informing participants of the generated meeting agenda.

[1171] A "communication platform" is a system for exchanging information such as chat systems and e-mail.

[1172] "Equipment operation data" refers to data that indicates the operating status and performance of equipment used in a factory.

[1173] A "robot" is an automated mechanical device with a voice assistant function and a display function.

[1174] "Voice notification" is a method in which a robot communicates information using voice.

[1175] "Display" refers to the method by which a robot displays and communicates information on a display.

[1176] MODE FOR CARRYING OUT THE INVENTION

[1177] Overall system configuration

[1178] This invention requires a system for supporting the scheduling of meetings within a factory. This system collects and automatically analyzes business-related communication and document data to select topics and participants, and generates and notifies meeting agendas.

[1179] Hardware Configuration

[1180] 1. Server: The main computational unit responsible for data collection, natural language processing, topic extraction, participant selection, agenda generation and notifications.

[1181] 2. Terminal: A PC or mobile device where communication data and document data are stored.

[1182] 3. Robot: A device that announces the meeting agenda by voice or display.

[1183] Software Configuration

[1184] 1. Natural language processing engine (e.g., nltk, sklearn): Software for extracting keywords and topic candidates from collected data.

[1185] 2. Email sending library (e.g. smtplib): Software for notifying participants of the generated agenda.

[1186] 3. Generative AI model: This is the AI ​​model used to analyze the collected data and generate the agenda.

[1187] Processing flow and operation explanation

[1188] Data collection

[1189] The server periodically uses APIs to collect data from communication platforms (chat systems and email) used within the factory and document data stored in the cloud. It also collects operational data from the equipment used in the factory. All of this data is stored in a database.

[1190] Natural Language Processing

[1191] The server analyzes the collected communication and document data using a natural language processing engine, tokenizing the text data and tagging it by part of speech to extract important keywords and potential topics.

[1192] Agenda extraction

[1193] The server generates candidate topics based on the extracted keywords and phrases. For example, keywords such as "maintenance delay" and "progress report" can be used to generate specific topics such as "checking the robot maintenance status" and "checking the product line progress."

[1194] Participant selection

[1195] The server dynamically selects appropriate meeting participants based on the generated agenda candidates, and automatically selects factory workers based on their past meeting participation history, job titles, and areas of expertise.

[1196] Agenda generation and notifications

[1197] The server automatically generates a meeting agenda based on the generated topic and selected participants. The generated agenda is notified to participants via email or calendar invitation. It is also notified to operators via voice assistants and displays on the factory robots.

[1198] Specific examples

[1199] For example, at the end of the day, the server collects chat messages such as "Robot A's maintenance is behind schedule" and data such as "Today's product line progress report." Based on this data, agenda candidates "Checking the maintenance status of Robot A" and "Checking the product line progress" are generated. Based on these agendas, maintenance personnel and production line leaders are selected as participants. The meeting agenda is generated as follows:

[1200] Agenda:

[1201] 1. Check the maintenance status of Robot A

[1202] 2. Check the progress of the product line

[1203] 3. Next Major Milestone

[1204] Participants:

[1205] Maintenance personnel

[1206] Production Line Leader

[1207] factory manager

[1208] The generated agenda is notified to participants via email, and the robot notifies the operator via voice and display.

[1209] Prompt Sentence Examples

[1210] "Automatically generate a meeting agenda. Use the following communication data: 'Robot A's maintenance is overdue', 'We need to check the progress of the production line'"

[1211] This automates the setting of meeting agendas and the selection of appropriate participants, enabling meetings within the factory to be run quickly and efficiently.

[1212] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1213] Step 1: Data collection

[1214] The server collects communication data from communication platforms and emails used within the factory. It also collects document data from cloud storage, including chat messages and reports. Specifically, it accesses these platforms via API and periodically retrieves the required data. The input is communication data and document data, and the output is the collected data.

[1215] Step 2: Natural Language Processing

[1216] The server inputs the collected communication data and document data into a natural language processing engine (e.g., nltk, sklearn) to analyze the text data. Here, the text is tokenized and tagged with parts of speech to extract important keywords and phrases. The input is the collected data, and the output is the extracted keywords and phrases. Specifically, TF-IDF vectorization and KMeans clustering are performed.

[1217] Step 3: Automatically extracting agenda items

[1218] The server generates meeting agenda candidates based on keywords and phrases extracted through natural language processing. For example, it generates the agenda "Checking the robot's maintenance status" from the keyword "maintenance delay." The input is the extracted keywords and phrases, and the output is the agenda candidate.

[1219] Step 4: Dynamic Participant Management

[1220] The server automatically selects appropriate meeting participants based on the generated agenda candidates. It selects the best participants for each agenda based on their past meeting participation history and job titles. The input is the agenda candidates, and the output is the selected participants. Specifically, it queries and filters relevant participant information from a database.

[1221] Step 5: Generate the agenda

[1222] The server automatically generates a meeting agenda based on the candidate agenda items and selected participants. The input is the candidate agenda items and participant information, and the output is the generated meeting agenda. Specifically, the server creates the agenda document using a text template.

[1223] Step 6: Announce the agenda

[1224] The server notifies the generated meeting agenda. The agenda is notified to participants using email notifications and calendar invitations. It also notifies operators via voice or display on the factory robots. The input is the generated meeting agenda, and the output is the notified agenda. Specifically, emails are sent using an email sending library (e.g., smtplib), and voice instructions and display are given through the robot's API.

[1225] 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.

[1226] This invention relates to a system that automates the setting of meeting agendas and participant selection, as well as recognizes user emotions to improve the quality of meetings. This system collects business-related communication data and document data, extracts candidate topics using natural language processing, dynamically selects participants suitable for the topics, and combines an emotion engine to recognize user emotions and reflect them in the topics and agendas.

[1227] The overall system configuration is as follows:

[1228] 1. Data Collection

[1229] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[1230] The server retrieves this communication data via API, parses it in JSON format, and stores it in a database. It also retrieves document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in a database.

[1231] 2. Natural Language Processing

[1232] The server inputs the collected communication and document data into a natural language processing (NLP) engine.

[1233] The server performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and significant phrases.

[1234] 3. Emotion Engine

[1235] The server sends the text extracted from the communication data and document data to an emotion engine to analyze the user's emotion.

[1236] The server recognizes emotions such as positive, negative, and neutral, and stores the results in a database.

[1237] 4. Automatic extraction of agenda items

[1238] The server applies an algorithm to generate topic candidates based on the keywords obtained from the NLP analysis results and the analysis results of the emotion engine.

[1239] The server creates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and emotional information, and stores them in a database.

[1240] 5. Dynamic Participant Management

[1241] The server selects the most suitable participants for the generated agenda candidates.

[1242] The server retrieves employee information such as past meeting history, areas of expertise, and job titles from the database, and dynamically selects the most appropriate participants for each agenda item. For example, for "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[1243] 6. Generate a meeting agenda

[1244] The server automatically generates a meeting agenda based on the selected topic and participant list.

[1245] The server embeds the topic and participant information in the agenda template and modifies the generated agenda to reflect the user's emotional information.

[1246] Based on the results of the emotion engine, the server prioritizes topics that are generating high emotions and reflects them in the agenda.

[1247] 7. Sending Notifications

[1248] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[1249] The server interfaces with the mail server to send emails containing detailed agendas and meeting information, and uses the calendar system to generate meeting invites.

[1250] Specific examples

[1251] Here is a concrete example of how the system works:

[1252] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[1253] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[1254] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[1255] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1256] 5. At the same time, the sentiment engine determines that the user has a strong negative sentiment towards the message "Project A is delayed."

[1257] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[1258] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[1259] 8. The server generates the meeting agenda based on the selected topic and participant list, as follows:

[1260] Agenda:

[1261] 1. Analysis of the cause of delay in Project A

[1262] 2. Cost optimization measures

[1263] 3. Next Major Milestone

[1264] Participants:

[1265] Taro Tanaka (Project Manager)

[1266] Hanako Suzuki (Risk Management Officer)

[1267] Ichiro Yamada (Project member)

[1268] 9. The server notifies participants of the generated agenda via email and calendar invites.

[1269] 10. Based on the analysis results of the emotion engine, the server makes modifications such as placing topics for which the user has particularly negative emotions at the top of the agenda.

[1270] In this way, the system reduces the workload involved in setting up and managing meetings, and supports prompt and effective meeting conduct through emotionally sensitive agenda setting.

[1271] The processing flow will be explained below.

[1272] Step 1:

[1273] Data collection

[1274] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[1275] The server obtains this communication data via API, parses it in JSON format, and stores it in a database.

[1276] The server also collects document data from cloud storage (e.g., Google Drive, SharePoint) via API and stores it in a database.

[1277] Step 2:

[1278] Performing Natural Language Processing (NLP)

[1279] The server inputs the collected communication and document data into a natural language processing (NLP) engine.

[1280] The server performs text tokenization, morphological analysis, part-of-speech tagging, and named entity recognition (NER) to extract important keywords and phrases.

[1281] Step 3:

[1282] emotion recognition

[1283] The server sends the communication data and the text extracted from the document data to the emotion engine.

[1284] The server uses an emotion engine to analyze the sentiment of each piece of text and classify it into categories such as positive, negative, or neutral.

[1285] The server stores the emotion recognition results in a database.

[1286] Step 4:

[1287] Automatic extraction of agenda items

[1288] The server runs an algorithm that integrates the results of NLP analysis and emotion recognition to generate topic candidates.

[1289] The server generates candidate topics such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and emotional information, and stores them in a database.

[1290] Step 5:

[1291] Dynamic Participant Management

[1292] The server dynamically selects the most suitable participants for the generated agenda candidates.

[1293] The server retrieves information such as employee job titles, areas of expertise, and past meeting participation history from a database and selects the most appropriate participants for each agenda item.

[1294] For example, for the agenda item "Analysis of the causes of project delays," the project manager, risk management officer, and project members are selected.

[1295] Step 6:

[1296] Generate a meeting agenda

[1297] The server automatically generates a meeting agenda based on the selected topic and participant list.

[1298] Based on the results of the emotion engine, the server makes adjustments such as placing topics with high negative emotions at the beginning of the agenda.

[1299] Step 7:

[1300] Sending notifications

[1301] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[1302] The server interfaces with a mail server to send emails containing detailed agendas and meeting information, and uses a calendar system to generate meeting invites.

[1303] Specific examples

[1304] Here is a concrete example of how the system works:

[1305] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[1306] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[1307] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[1308] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1309] 5. At the same time, the emotion engine recognizes emotions from the user's communication data and document data and determines that the text "Project A is delayed" indicates a strong negative emotion.

[1310] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[1311] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[1312] 8. The server generates the meeting agenda based on the selected topic and participant list, as follows:

[1313] Agenda:

[1314] 1. Analysis of the cause of delay in Project A

[1315] 2. Cost optimization measures

[1316] 3. Next Major Milestone

[1317] Participants:

[1318] Taro Tanaka (Project Manager)

[1319] Hanako Suzuki (Risk Management Officer)

[1320] Ichiro Yamada (Project member)

[1321] 9. The server notifies participants of the generated agenda via email and calendar invites.

[1322] 10. Based on the analysis results of the emotion engine, the server makes adjustments to improve the quality of the meeting, such as placing topics for which users have particularly negative emotions at the top of the agenda.

[1323] In this way, the system reduces the workload involved in setting up and managing meetings, and supports prompt and effective meeting conduct through emotionally sensitive agenda setting.

[1324] Example 2

[1325] 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."

[1326] Conventionally, setting the agenda and selecting participants for a meeting have been done manually, which has the problem of requiring a great deal of effort and time. Furthermore, it has been difficult to consider the user's feelings when setting the agenda, which has also made it difficult to improve the quality of the meeting. The present invention aims to solve these problems and improve the quality of meetings through automated agenda setting and participant selection.

[1327] 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.

[1328] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and selected participants, means for notifying the meeting agenda, means for analyzing user emotions from the communication data and document data, and means for determining the priority of agenda items based on the emotion analysis results. This reduces the workload associated with setting and managing meetings, and enables more effective meetings by setting agendas that take emotions into consideration.

[1329] "Business-related communications data" refers to messages and their metadata exchanged through chat platforms and email systems used within a company.

[1330] "Document data" refers to digital information such as document files, reports, and presentation materials obtained from cloud storage or corporate databases.

[1331] "Natural language processing" refers to techniques and algorithms that enable computers to understand, interpret, and generate human language.

[1332] "Keywords" refer to important words and phrases extracted from communication data and document data through natural language processing.

[1333] "Agenda candidates" refer to items to be discussed at a meeting that are generated based on keywords extracted from communication data and document data.

[1334] "Participants" refers to employees or personnel who should be invited to the meeting.

[1335] "Meeting agenda" refers to a meeting schedule generated based on candidate topics and a selected participant list.

[1336] "Sentiment analysis" refers to the process of identifying emotions contained in user communication and document data and classifying those emotions as positive, negative, neutral, etc.

[1337] "Priority" refers to the criteria used to determine the order in which meeting topics are discussed based on the results of sentiment analysis.

[1338] This invention relates to a system that automates the setting of meeting agendas and participant selection, as well as recognizes user emotions to improve the quality of meetings. This system collects business-related communication data and document data, extracts candidate topics using natural language processing (NLP), dynamically selects participants suitable for the topics, and combines an emotion engine to recognize user emotions and reflect them in the topics and agendas.

[1339] Overall system configuration

[1340] Data collection

[1341] The server collects communication data from the company's internal chat platform (e.g., Slack, Microsoft Teams) and email system. It obtains the data via API, parses it in JSON format, and stores it in a database. It also obtains document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in a database.

[1342] Natural Language Processing (NLP)

[1343] The server inputs the collected communication and document data into an NLP engine (e.g., Google Cloud NLP, AWS Comprehend), which performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and important phrases.

[1344] Emotion Engine

[1345] The server sends the text extracted from the communication data and document data to an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The results are classified into emotion categories such as positive, negative, and neutral, and stored in a database.

[1346] Automatic extraction of agenda items

[1347] The server applies an algorithm based on the results of NLP analysis and sentiment analysis to generate topic candidates. For example, it generates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and sentiment information.

[1348] Dynamic Participant Management

[1349] The server selects the most suitable participants for each candidate agenda item. Employee information such as past meeting history, areas of expertise, and job titles is retrieved from the database, and the most suitable participants are dynamically selected for each agenda item. For example, for the agenda item "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[1350] Generate a meeting agenda

[1351] The server automatically generates a meeting agenda based on the selected topic and participant list. It embeds the topic and participant information in an agenda template and modifies the generated agenda to reflect emotional information. Based on the results of the emotional analysis, it prioritizes topics for which users have particularly strong emotions and reflects them in the agenda.

[1352] Sending notifications

[1353] The server notifies the generated meeting agenda to the relevant participants via email and calendar invites. It works with the mail server to send emails containing the detailed agenda and meeting information, and creates meeting invites using the calendar system.

[1354] Specific examples

[1355] Here is a concrete example of how the system works:

[1356] 1. Work-related chat messages and emails accumulate daily on devices (employee computers and mobile devices).

[1357] 2. The server collects this communication data and document data from cloud storage on a daily basis.

[1358] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[1359] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1360] 5. At the same time, the emotion engine determines that the user is expressing strong negative emotion towards the message "Project A is delayed."

[1361] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[1362] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[1363] 8. The server generates a meeting agenda based on the selected topic and participant list:

[1364] Agenda:

[1365] 1. Analysis of the cause of delay in Project A

[1366] 2. Cost optimization measures

[1367] 3. Next Major Milestone

[1368] Participants:

[1369] Taro Tanaka (Project Manager)

[1370] Hanako Suzuki (Risk Management Officer)

[1371] Ichiro Yamada (Project member)

[1372] 9. The server notifies participants of the generated agenda via email and calendar invites.

[1373] 10. Based on the analysis results of the emotion engine, the server places topics for which the user has particularly negative emotions at the top of the agenda.

[1374] Prompt Sentence Examples

[1375] The following prompts are used to input the generative AI model:

[1376] Please create a program that extracts topics from business chat messages containing negative sentiment, such as "Project A is delayed," and generates an agenda by selecting the most appropriate meeting participants. Please also provide an example of the generated agenda.

[1377] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1378] Step 1: Data collection

[1379] The server collects communication data from the company's internal chat platforms (e.g., chat service A, chat service B) and email systems. Specifically, it uses APIs to obtain text-based message data and email data from these services, parses it into JSON format, and stores it in a database.

[1380] Input: Data obtained from APIs of chat platforms and email systems.

[1381] Data processing: Convert the acquired data into JSON format.

[1382] Output: JSON formatted communication and email data, stored in a database.

[1383] Step 2: Collect document data from cloud storage

[1384] The server retrieves document data (PDF, Word files, etc.) from cloud storage (e.g., storage service A, storage service B) and stores it in a database. It periodically scans documents using the storage API to retrieve new and updated documents.

[1385] Input: Data obtained from cloud storage API.

[1386] Data processing: Convert the acquired document data into JSON format.

[1387] Output: Document data in JSON format, stored in the database.

[1388] Step 3: Natural Language Processing (NLP)

[1389] The server inputs the communication and document data stored in the database into NLP engines (e.g., NLP engine A, NLP engine B). The NLP engines perform text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract important keywords and phrases.

[1390] Input: Correspondence and document data extracted from the database.

[1391] Data operations: text tokenization, part-of-speech tagging, NER processing.

[1392] Output: Keywords and important phrases, stored in a database.

[1393] Step 4: Emotion analysis using the emotion engine

[1394] The server sends the text analyzed by the NLP engine to an emotion engine (e.g., emotion analysis service A, emotion analysis service B) to analyze the user's emotions. It classifies the emotions into positive, negative, neutral, and other emotional characteristics, and stores the results in a database.

[1395] Input: Keywords and phrases obtained as a result of analysis by the NLP engine.

[1396] Data Computation: Determine the sentiment category of the text through sentiment analysis.

[1397] Output: Sentiment analysis results (positive, negative, neutral), stored in a database.

[1398] Step 5: Automatically extracting topic candidates

[1399] The server generates candidate topics by applying an algorithm based on the results of NLP analysis and sentiment analysis. This extracts topics with keywords such as "project delay" and "cost optimization," and generates specific themes such as "analysis of the causes of project delays" and "cost optimization measures."

[1400] Input: NLP analysis results, sentiment analysis results.

[1401] Data computation: Applying algorithms that combine keywords and sentiment information to generate agendas.

[1402] Output: Agenda candidates, stored in a database.

[1403] Step 6: Dynamic Participant Management

[1404] The server selects the most suitable participants for each candidate topic. Appropriate employees are dynamically selected based on information such as past meeting history, areas of expertise, and job titles. For example, for the topic "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[1405] Input: Agenda candidate, employee information retrieved from database.

[1406] Data calculation: Applying algorithms to select the most suitable participants based on meeting history, areas of expertise, and job title information.

[1407] Output: Participant list, stored in database.

[1408] Step 7: Generate the meeting agenda

[1409] The server automatically generates a meeting agenda based on the selected topics and participant list. It embeds the topic and participant information in a template and prioritizes topics about which users have particularly strong feelings, reflecting them in the agenda.

[1410] Input: candidate agenda items, participant list, sentiment analysis results.

[1411] Data operations: Embedding information into templates and prioritizing agenda items.

[1412] Output: The generated meeting agenda, stored in a database.

[1413] Step 8: Sending notifications

[1414] The server notifies the generated meeting agenda to interested participants via email and calendar invites. It works with a mail server to send emails containing the detailed agenda and meeting information, and uses a calendar system to create meeting invites.

[1415] Input: The generated meeting agenda.

[1416] Data Computing: Notification generation via email and calendar systems.

[1417] Output: Notification emails and calendar events sent.

[1418] (Application example 2)

[1419] 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."

[1420] Troubles and delays at production sites are issues that require prompt and appropriate countermeasures. However, the traditional method of setting up meetings and selecting participants places a heavy workload on employees, making efficient management difficult, especially in large factories. Furthermore, meetings often fail to take into account the feelings of employees when setting agendas and topics, resulting in ineffective discussions. This hinders the efficient operation of factories, creating a challenge.

[1421] 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.

[1422] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and the selected participants, means for analyzing user emotions, means for modifying the meeting agenda based on emotion information, and means for notifying the meeting agenda. This improves the efficiency of meeting setup and management for troubleshooting at production sites and enables effective discussions that take user emotions into consideration.

[1423] "Business-related communication data" refers to communication information used within a company, such as chat, email, and monitoring data.

[1424] "Document data" refers to document information such as reports, memos, and files stored in online storage or local storage.

[1425] "Natural language processing" is an information processing technology for analyzing and understanding human language.

[1426] "Keywords" are important words and phrases extracted from the data.

[1427] "Agenda candidate" refers to themes and issues extracted through natural language processing.

[1428] "Relevant participants" are people or positions who should attend the meeting based on the proposed agenda.

[1429] A "meeting agenda" is a document that includes the agenda and attendee list to direct the progress of a meeting.

[1430] "User emotion" is the emotional state of the user as assessed by the emotion analysis engine.

[1431] "Emotion information" is data obtained as a result of analyzing the user's emotions.

[1432] An "emotion engine" is an information processing system for analyzing emotions from communication data and voice data.

[1433] "Dynamic selection of relevant participants based on topic candidates" is a process of selecting the most suitable participants in real time, taking into account their past history and areas of expertise related to the topic.

[1434] A "conversation platform" is a communication tool such as a chat tool or video conferencing software used within a company.

[1435] "Online storage" is a data storage service that can be accessed via the Internet.

[1436] "Notifying a meeting agenda" refers to the act of informing relevant parties of the generated agenda via email or calendar invitation.

[1437] This invention is a conference management system for quickly and effectively resolving problems and delays at production sites. This system collects business-related communication data and document data, extracts candidate topics using natural language processing, dynamically selects participants, and uses an emotion engine to recognize user emotions and reflect them in the topics and agenda.

[1438] The server first collects real-time data from sensors and surveillance cameras. It also obtains production data and trouble reports from the factory management system (MES) via API and stores them in a database. Based on the collected data, the server uses a natural language processing engine (e.g., NLTK, Transformers) to extract important keywords and phrases and generate candidate topics.

[1439] The emotion engine (e.g., Emotion AI) analyzes emotions from collected data and user voice and chat. The emotion results are stored in a database as emotional states, such as positive, negative, or neutral. The server then modifies the meeting agenda based on this emotional information, prioritizing topics that indicate negative emotions in the agenda list.

[1440] The generated meeting agenda is then sent to participants via email or a calendar invitation system, using common business communication tools (e.g., Microsoft Outlook, Google Calendar).

[1441] For example, the server generates candidate topics such as "malfunction of machine A" or "delay on production line B" and selects appropriate participants such as machine maintenance personnel and operators. This information is then reflected in the negative emotions analyzed by the emotion engine, and notified as the generated meeting agenda.

[1442] As an example, the following prompt sentences can be input to a generative AI model:

[1443] "An AI robot collects sensor data and trouble reports from within a factory, inputs this data into a natural language processing and sentiment analysis engine, selects appropriate meeting topics and participants, and automatically generates and notifies meeting agendas. We will show you a specific example of how this works."

[1444] In this way, the present invention can realize an effective conference management system in the production site and improve the efficiency of troubleshooting.

[1445] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1446] Step 1:

[1447] The server collects real-time data from sensors and surveillance cameras. It acquires the data through the sensor API and the surveillance camera API and stores it in a database. The input is sensor data and video data, and the output is the collected data stored in the database.

[1448] Step 2:

[1449] The server obtains production data and trouble reports from the factory management system (MES) via API. This data is also stored in the database. The input is production data and trouble reports, and the output is business data stored in the database.

[1450] Step 3:

[1451] The server inputs collected data (sensor data, production data, trouble reports) into a natural language processing (NLP) engine, which performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and important phrases. The input is collected data, and the output is keywords and topic candidates.

[1452] Step 4:

[1453] The server inputs the collected communication data and the user's voice and chat data into an emotion engine (e.g., Emotion AI) to analyze the user's emotions. The emotion engine analyzes the emotional state, such as positive, negative, or neutral, and stores the results in a database. The input is communication data and voice data, and the output is the emotion analysis results.

[1454] Step 5:

[1455] The server generates topic candidates and dynamically selects relevant participants based on the processed NLP analysis results and sentiment analysis results. The inputs are the NLP analysis results and sentiment analysis results, and the outputs are topic candidates and a list of participants.

[1456] Step 6:

[1457] The server automatically generates a meeting agenda based on the generated topics and the selected participant list. It considers emotional information and prioritizes topics that show negative emotions. The inputs are candidate topics and the participant list, and the output is a revised meeting agenda.

[1458] Step 7:

[1459] The server notifies participants of the generated meeting agenda via email and / or calendar invites. The input to the notification system is the modified meeting agenda, and the output is the meeting notice sent to participants.

[1460] 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.

[1461] 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.

[1462] 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.

[1463] [Fourth embodiment]

[1464] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1465] 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.

[1466] 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).

[1467] 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.

[1468] 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.

[1469] 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).

[1470] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1471] 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.

[1472] 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.

[1473] 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.

[1474] 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.

[1475] 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.

[1476] 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."

[1477] The present invention relates to a system that improves business efficiency by automating the setting of meeting agendas and participant selection. This system involves a series of operations: collecting business-related communication data and document data, extracting candidate agendas using natural language processing, dynamically selecting participants suitable for the agendas, and finally generating and notifying participants of the meeting agenda.

[1478] The overall system configuration is as follows:

[1479] 1. Data Collection

[1480] The server automatically collects data from communication platforms (e.g., chat systems, email) and document management systems (e.g., cloud storage, file servers) used in business.

[1481] The server connects to these platforms via API, periodically retrieves data, and stores it in a database.

[1482] 2. Natural Language Processing

[1483] The server analyzes the collected communication and document data using a natural language processing (NLP) engine to extract important keywords and potential topics.

[1484] NLP engines extract semantic information from data using techniques such as text tokenization, part-of-speech tagging, and named entity recognition (NER).

[1485] 3. Automatic extraction of agenda items

[1486] The server generates topic candidates based on the extracted keywords and phrases.

[1487] For example, from keywords such as "project delay" and "progress," topics such as "checking project progress" and "analyzing the cause of the delay" are generated.

[1488] 4. Dynamic Participant Management

[1489] The server automatically selects appropriate conference participants based on the aforementioned agenda candidates.

[1490] The server selects the most suitable participants based on information such as the participants' job titles, areas of expertise, and past meeting participation history. For example, members suitable for the agenda, such as "project managers" or "engineering team leaders," are selected.

[1491] 5. Agenda generation and notification

[1492] The server aggregates the topic candidates and selected participants and automatically generates the meeting agenda.

[1493] The server notifies participants of the generated agenda using methods such as email or calendar invitations.

[1494] Specific examples

[1495] Here is a concrete example of how the system works:

[1496] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[1497] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[1498] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[1499] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1500] 5. From the analysis results, potential topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" are generated.

[1501] 6. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A." This information is dynamically selected based on employee information stored in the database.

[1502] 7. The server generates the meeting agenda based on the topic and the selected participants, as follows:

[1503] Agenda:

[1504] 1. Analysis of the cause of delay in Project A

[1505] 2. Cost optimization measures

[1506] 3. Next Major Milestone

[1507] Participants:

[1508] Taro Tanaka (Project Manager)

[1509] Hanako Suzuki (Risk Management Officer)

[1510] Ichiro Yamada (Project member)

[1511] 8. The server notifies participants of the generated agenda via email and calendar invites.

[1512] In this way, the system reduces the workload involved in setting up and managing meetings, and supports the rapid and effective implementation of meetings.

[1513] The processing flow will be explained below.

[1514] Step 1:

[1515] Data collection

[1516] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[1517] The server obtains this communication data via API, parses it in JSON format, and stores it in a database.

[1518] The server also retrieves document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in the database as well.

[1519] Step 2:

[1520] Analysis using natural language processing

[1521] The server inputs the collected communication data and document data into a natural language processing engine.

[1522] The server performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and significant phrases.

[1523] Step 3:

[1524] Automatic extraction of agenda items

[1525] The server applies an algorithm to generate topic candidates based on the keywords obtained from the NLP analysis results.

[1526] The server creates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization," and stores these in a database.

[1527] Step 4:

[1528] Dynamic Participant Management

[1529] The server selects the most suitable participants for the generated agenda candidates.

[1530] The server retrieves employee information such as past meeting history, areas of expertise, and job titles from a database, and dynamically selects participants deemed most suitable for each agenda item.

[1531] For example, the project manager, risk management officer, and project members are selected for the "cause analysis of project delays."

[1532] Step 5:

[1533] Generate a meeting agenda

[1534] The server automatically generates a meeting agenda based on the selected topic and participant list.

[1535] The server embeds the topic and participant information in the agenda template to create a formal meeting agenda.

[1536] Step 6:

[1537] Sending notifications

[1538] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[1539] The server interfaces with a mail server to send emails containing detailed agendas and meeting information, and uses a calendar system to generate meeting invites.

[1540] Example 1

[1541] 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."

[1542] In today's business environment, many companies and organizations generate large amounts of communication and document data. It is difficult to extract important agenda items from this data, select appropriate meeting participants, and efficiently schedule meetings, resulting in a decline in business efficiency. To solve this problem, a system that can effectively utilize communication and document data and automate meeting scheduling is needed.

[1543] 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.

[1544] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and the selected participants, and means for notifying the meeting agenda. This allows users to automate a series of processes from data collection to meeting setup and notification of the meeting agenda.

[1545] "Business-related communications data" refers to data including text messages, images, videos, and audio data sent and received via messaging platforms, emails, and other means used in business.

[1546] "Document data" refers to business-related data such as text documents, spreadsheets, presentation files, image files, and PDFs that are stored on remote storage or file servers.

[1547] "Natural language processing" is a technology for analyzing text data and extracting, generating, understanding, and translating information, and includes processes such as text tokenization, part-of-speech tagging, and named entity recognition.

[1548] "Keywords" are important words and phrases extracted from communication data and document data, and are information used to generate topic candidates.

[1549] "Agenda candidates" are themes or items to be discussed in the meeting, generated from the extracted keywords and phrases.

[1550] "Participants" are individuals selected to attend meetings related to the agenda, and are dynamically selected based on job title, area of ​​expertise, etc.

[1551] A "meeting agenda" is a list of specific topics or items to be discussed at a meeting, and is used to notify participants and guide the meeting.

[1552] "Notification medium" means the media and protocols for appropriately informing meeting participants of the generated meeting agenda, including email, calendar invites, messaging platform notifications, etc.

[1553] The present invention relates to a system that improves business efficiency by automating the setting of meeting agendas and participant selection. This system includes a series of operations: collecting business-related communication data and document data, extracting candidate agendas using natural language processing, dynamically selecting participants suitable for the agendas, and finally generating and notifying the meeting agenda.

[1554] The overall system configuration is as follows:

[1555] 1. Data Collection

[1556] Devices generate communication data such as chat systems and emails used for business, as well as document data stored in cloud storage and file servers every day. This data is sent from each employee's PC or mobile device. The server connects to communication platforms and document management systems via APIs to periodically collect this data.

[1557] Specific use cases include using Slack's API to retrieve chat messages and Google Drive's API to gather document data.

[1558] The server stores the retrieved data in a database.

[1559] 2. Natural Language Processing

[1560] The server inputs the communication data and document data stored in the database into a natural language processing (NLP) engine.

[1561] Examples of NLP engines used include SpaCy and NLTK.

[1562] Specific processing includes text tokenization, part-of-speech tagging, and named entity recognition (NER).

[1563] For example, if there is a chat message that says "Project A is delayed," the NLP engine breaks it down into words such as "project," "A," "delayed," and "being," and extracts important entities such as "Project A" and "delayed."

[1564] 3. Automatic extraction of agenda items

[1565] The server generates topic candidates based on the extracted keywords and phrases.

[1566] For example, from keywords such as "project delay" and "progress," topics such as "checking project progress" and "analyzing the cause of the delay" are generated.

[1567] 4. Dynamic Participant Management

[1568] The server dynamically selects appropriate conference participants based on the aforementioned agenda candidates.

[1569] The server selects the most suitable participants from a database based on their job title, area of ​​expertise, and past meeting attendance history, such as a project manager or engineering team leader.

[1570] 5. Agenda generation and notification

[1571] The server automatically generates a meeting agenda by aggregating the topic candidates and selected participants.

[1572] The server notifies participants of the generated agenda via email, calendar invites, or other means.

[1573] Specific examples are shown below:

[1574] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[1575] 2. The server periodically collects this communication data and document data obtained from cloud storage.

[1576] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are saved.

[1577] 4. The server inputs the saved data into an NLP engine (SpaCy) and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1578] 5. From the analysis results, potential topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" are generated.

[1579] 6. The server selects the project manager, risk management officer, and project members for the agenda item "Analyzing the causes of delays in Project A." This information is dynamically selected based on employee information stored in the database.

[1580] 7. The server generates the meeting agenda based on the topic and the selected participants, as follows:

[1581] Agenda:

[1582] 1. Analysis of the cause of delay in Project A

[1583] 2. Cost optimization measures

[1584] 3. Next Major Milestone

[1585] Participants:

[1586] Taro Tanaka (Project Manager)

[1587] Hanako Suzuki (Risk Management Officer)

[1588] Ichiro Yamada (Project member)

[1589] 8. The server notifies participants of the generated agenda via email and calendar invites.

[1590] Prompt statement:

[1591] "If there is communication data indicating that Project A is delayed and document data such as a monthly report, please explain a system that dynamically selects appropriate agenda candidates and participants based on them, generates a meeting agenda, and notifies them."

[1592] In this way, the system reduces the workload involved in setting up and managing a conference, and supports the rapid and effective implementation of conferences.

[1593] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1594] Step 1:

[1595] Terminals generate business-related communication data such as chat systems and emails, as well as document data from cloud storage and file servers, on a daily basis. These data are imported. The inputs are communication data (e.g., chat messages, emails) and document data (e.g., text files, PDFs), which are then saved. The output is the saved communication data and document data.

[1596] Specific behavior:

[1597] An employee sends a chat message on the device.

[1598] The device will upload the new document to cloud storage.

[1599] Step 2:

[1600] The server connects to communication platforms (e.g., Slack) and document management systems (e.g., Google Drive) via APIs to periodically collect data. The input is data obtained from the communication platform and document management system, and the output is data stored in a database. This allows communication data and document data to be periodically updated.

[1601] Specific behavior:

[1602] The server retrieves the latest chat messages from the Slack API.

[1603] The server retrieves the latest document file from the Google Drive API.

[1604] The server stores the collected data in a database.

[1605] Step 3:

[1606] The server inputs the communication data and document data stored in the database into a natural language processing engine. The input is the communication data and document data stored in the database, and the output is the analysis results of natural language processing. An NLP engine (e.g., SpaCy) is used to perform data tokenization, part-of-speech tagging, named entity recognition (NER), etc.

[1607] Specific behavior:

[1608] The server passes a chat message saying "Project A is delayed" to the NLP engine.

[1609] The server passes the document "This Month's Expense Report" to the NLP engine.

[1610] The NLP engine extracts keywords such as "project delay" and "cost optimization."

[1611] Step 4:

[1612] The server generates topic candidates based on the analysis results from the NLP engine. The input is keywords and phrases processed through natural language processing, and the output is topic candidates. Therefore, new topic candidates are listed based on important entities and keywords.

[1613] Specific behavior:

[1614] The server generates an agenda item "Checking project progress" from the keyword "project delay."

[1615] The server generates a "cost optimization measures" agenda from the keyword "cost optimization."

[1616] Step 5:

[1617] The server dynamically selects appropriate meeting participants based on the agenda candidates. The input is the agenda candidates and an employee database, and the output is a list of selected participants. The optimal participants are selected based on the participants' job titles, areas of expertise, and past meeting history.

[1618] Specific behavior:

[1619] The server searches the database for the project manager, risk manager, and project members.

[1620] The server creates a list of participants suitable for the "Project Progress Review" agenda item.

[1621] Step 6:

[1622] The server generates a meeting agenda based on the candidate agenda items and the selected participant information. The input is the candidate agenda items and the participant list, and the output is the generated meeting agenda. This completes the meeting agenda.

[1623] Specific behavior:

[1624] The server generates a meeting agenda like this:

[1625] Agenda:

[1626] 1. Analysis of the cause of delay in Project A

[1627] 2. Cost optimization measures

[1628] 3. Next Major Milestone

[1629] Participants:

[1630] Taro Tanaka (Project Manager)

[1631] Hanako Suzuki (Risk Management Officer)

[1632] Ichiro Yamada (Project member)

[1633] Step 7:

[1634] The server notifies participants of the generated meeting agenda using notification means (e.g., email, calendar invitation). The input is the generated meeting agenda, and the output is the notified meeting agenda. This allows participants to receive information about the meeting in a timely manner.

[1635] Specific behavior:

[1636] The server sends the meeting agenda to the participants by email.

[1637] The server sends calendar invites and adds the meeting to the attendees' schedules.

[1638] The above are the specific processing steps of this system. By describing the operations performed at each step and their inputs and outputs in detail, the overall flow and processing content of the system will become clear.

[1639] (Application example 1)

[1640] 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."

[1641] Conventional meeting scheduling systems have the problem of being inefficient, taking a lot of time and effort to set up meeting agendas and select participants. In particular, in-factory operations require the ability to dynamically extract necessary agendas and participants in real time and smoothly schedule meetings. Furthermore, there are many situations where a quick response according to the situation is required, and current systems are insufficient.

[1642] 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.

[1643] In this invention, the server includes: means for collecting communication data and document data related to business operations; means for extracting keywords and agenda candidates from the communication data and document data using natural language processing; means for dynamically selecting relevant participants based on the agenda candidates; means for generating a meeting agenda based on the agenda candidates and the selected participants; means for notifying the meeting agenda; means for collecting operation data of a communication platform and equipment operated in a factory; means for generating agenda candidates using the operation data of the communication platform and equipment; means for selecting appropriate workers in the factory as participants based on the agenda candidates; and means for notifying the meeting agenda by voice or display via a robot. This automates the setting of a meeting agenda and the selection of appropriate participants, enabling fast and efficient meeting management in a factory.

[1644] "Business-related communication data" refers to data obtained from various communication means used in business, such as chat platforms and email.

[1645] "Document data" refers to electronic file data such as business-related reports and materials.

[1646] "Natural language processing" is a technology for analyzing text data and extracting useful information such as keywords and topic candidates.

[1647] A "keyword" is a specific word or phrase extracted from text data.

[1648] "Agenda candidates" are meeting topics generated based on the extracted keywords.

[1649] "Participants" are people or parties who need to attend a meeting.

[1650] A "meeting agenda" is a document that lists the proceedings and agenda items for a meeting.

[1651] "Notification" is the act of informing participants of the generated meeting agenda.

[1652] A "communication platform" is a system for exchanging information such as chat systems and e-mail.

[1653] "Equipment operation data" refers to data that indicates the operating status and performance of equipment used in a factory.

[1654] A "robot" is an automated mechanical device with a voice assistant function and a display function.

[1655] "Voice notification" is a method in which a robot communicates information using voice.

[1656] "Display" refers to the method by which a robot displays and communicates information on a display.

[1657] MODE FOR CARRYING OUT THE INVENTION

[1658] Overall system configuration

[1659] This invention requires a system for supporting the scheduling of meetings within a factory. This system collects and automatically analyzes business-related communication and document data to select topics and participants, and generates and notifies meeting agendas.

[1660] Hardware Configuration

[1661] 1. Server: The main computational unit responsible for data collection, natural language processing, topic extraction, participant selection, agenda generation and notifications.

[1662] 2. Terminal: A PC or mobile device where communication data and document data are stored.

[1663] 3. Robot: A device that announces the meeting agenda by voice or display.

[1664] Software Configuration

[1665] 1. Natural language processing engine (e.g., nltk, sklearn): Software for extracting keywords and topic candidates from collected data.

[1666] 2. Email sending library (e.g. smtplib): Software for notifying participants of the generated agenda.

[1667] 3. Generative AI model: This is the AI ​​model used to analyze the collected data and generate the agenda.

[1668] Processing flow and operation explanation

[1669] Data collection

[1670] The server periodically uses APIs to collect data from communication platforms (chat systems and email) used within the factory and document data stored in the cloud. It also collects operational data from the equipment used in the factory. All of this data is stored in a database.

[1671] Natural Language Processing

[1672] The server analyzes the collected communication and document data using a natural language processing engine, tokenizing the text data and tagging it by part of speech to extract important keywords and potential topics.

[1673] Agenda extraction

[1674] The server generates candidate topics based on the extracted keywords and phrases. For example, keywords such as "maintenance delay" and "progress report" can be used to generate specific topics such as "checking the robot maintenance status" and "checking the product line progress."

[1675] Participant selection

[1676] The server dynamically selects appropriate meeting participants based on the generated agenda candidates, and automatically selects factory workers based on their past meeting participation history, job titles, and areas of expertise.

[1677] Agenda generation and notifications

[1678] The server automatically generates a meeting agenda based on the generated topic and selected participants. The generated agenda is notified to participants via email or calendar invitation. It is also notified to operators via voice assistants and displays on the factory robots.

[1679] Specific examples

[1680] For example, at the end of the day, the server collects chat messages such as "Robot A's maintenance is behind schedule" and data such as "Today's product line progress report." Based on this data, agenda candidates "Checking the maintenance status of Robot A" and "Checking the product line progress" are generated. Based on these agendas, maintenance personnel and production line leaders are selected as participants. The meeting agenda is generated as follows:

[1681] Agenda:

[1682] 1. Check the maintenance status of Robot A

[1683] 2. Check the progress of the product line

[1684] 3. Next Major Milestone

[1685] Participants:

[1686] Maintenance personnel

[1687] Production Line Leader

[1688] factory manager

[1689] The generated agenda is notified to participants via email, and the robot notifies the operator via voice and display.

[1690] Prompt Sentence Examples

[1691] "Automatically generate a meeting agenda. Use the following communication data: 'Robot A's maintenance is overdue', 'We need to check the progress of the production line'"

[1692] This automates the setting of meeting agendas and the selection of appropriate participants, enabling meetings within the factory to be run quickly and efficiently.

[1693] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1694] Step 1: Data collection

[1695] The server collects communication data from communication platforms and emails used within the factory. It also collects document data from cloud storage, including chat messages and reports. Specifically, it accesses these platforms via API and periodically retrieves the required data. The input is communication data and document data, and the output is the collected data.

[1696] Step 2: Natural Language Processing

[1697] The server inputs the collected communication data and document data into a natural language processing engine (e.g., nltk, sklearn) to analyze the text data. Here, the text is tokenized and tagged with parts of speech to extract important keywords and phrases. The input is the collected data, and the output is the extracted keywords and phrases. Specifically, TF-IDF vectorization and KMeans clustering are performed.

[1698] Step 3: Automatically extracting agenda items

[1699] The server generates meeting agenda candidates based on keywords and phrases extracted through natural language processing. For example, it generates the agenda "Checking the robot's maintenance status" from the keyword "maintenance delay." The input is the extracted keywords and phrases, and the output is the agenda candidate.

[1700] Step 4: Dynamic Participant Management

[1701] The server automatically selects appropriate meeting participants based on the generated agenda candidates. It selects the best participants for each agenda based on their past meeting participation history and job titles. The input is the agenda candidates, and the output is the selected participants. Specifically, it queries and filters relevant participant information from a database.

[1702] Step 5: Generate the agenda

[1703] The server automatically generates a meeting agenda based on the candidate agenda items and selected participants. The input is the candidate agenda items and participant information, and the output is the generated meeting agenda. Specifically, the server creates the agenda document using a text template.

[1704] Step 6: Announce the agenda

[1705] The server notifies the generated meeting agenda. The agenda is notified to participants using email notifications and calendar invitations. It also notifies operators via voice or display on the factory robots. The input is the generated meeting agenda, and the output is the notified agenda. Specifically, emails are sent using an email sending library (e.g., smtplib), and voice instructions and display are given through the robot's API.

[1706] 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.

[1707] This invention relates to a system that automates the setting of meeting agendas and participant selection, as well as recognizes user emotions to improve the quality of meetings. This system collects business-related communication data and document data, extracts candidate topics using natural language processing, dynamically selects participants suitable for the topics, and combines an emotion engine to recognize user emotions and reflect them in the topics and agendas.

[1708] The overall system configuration is as follows:

[1709] 1. Data Collection

[1710] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[1711] The server retrieves this communication data via API, parses it in JSON format, and stores it in a database. It also retrieves document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in a database.

[1712] 2. Natural Language Processing

[1713] The server inputs the collected communication and document data into a natural language processing (NLP) engine.

[1714] The server performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and significant phrases.

[1715] 3. Emotion Engine

[1716] The server sends the text extracted from the communication data and document data to an emotion engine to analyze the user's emotion.

[1717] The server recognizes emotions such as positive, negative, and neutral, and stores the results in a database.

[1718] 4. Automatic extraction of agenda items

[1719] The server applies an algorithm to generate topic candidates based on the keywords obtained from the NLP analysis results and the analysis results of the emotion engine.

[1720] The server creates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and emotional information, and stores them in a database.

[1721] 5. Dynamic Participant Management

[1722] The server selects the most suitable participants for the generated agenda candidates.

[1723] The server retrieves employee information such as past meeting history, areas of expertise, and job titles from the database, and dynamically selects the most appropriate participants for each agenda item. For example, for "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[1724] 6. Generate a meeting agenda

[1725] The server automatically generates a meeting agenda based on the selected topic and participant list.

[1726] The server embeds the topic and participant information in the agenda template and modifies the generated agenda to reflect the user's emotional information.

[1727] Based on the results of the emotion engine, the server prioritizes topics that are generating high emotions and reflects them in the agenda.

[1728] 7. Sending Notifications

[1729] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[1730] The server interfaces with the mail server to send emails containing detailed agendas and meeting information, and uses the calendar system to generate meeting invites.

[1731] Specific examples

[1732] Here is a concrete example of how the system works:

[1733] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[1734] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[1735] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[1736] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1737] 5. At the same time, the sentiment engine determines that the user has a strong negative sentiment towards the message "Project A is delayed."

[1738] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[1739] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[1740] 8. The server generates the meeting agenda based on the selected topic and participant list, as follows:

[1741] Agenda:

[1742] 1. Analysis of the cause of delay in Project A

[1743] 2. Cost optimization measures

[1744] 3. Next Major Milestone

[1745] Participants:

[1746] Taro Tanaka (Project Manager)

[1747] Hanako Suzuki (Risk Management Officer)

[1748] Ichiro Yamada (Project member)

[1749] 9. The server notifies participants of the generated agenda via email and calendar invites.

[1750] 10. Based on the analysis results of the emotion engine, the server makes modifications such as placing topics for which the user has particularly negative emotions at the top of the agenda.

[1751] In this way, the system reduces the workload involved in setting up and managing meetings, and supports prompt and effective meeting conduct through emotionally sensitive agenda setting.

[1752] The processing flow will be explained below.

[1753] Step 1:

[1754] Data collection

[1755] The server collects communication data from the company's internal chat platforms (e.g., Slack, Microsoft Teams) and email systems.

[1756] The server obtains this communication data via API, parses it in JSON format, and stores it in a database.

[1757] The server also collects document data from cloud storage (e.g., Google Drive, SharePoint) via API and stores it in a database.

[1758] Step 2:

[1759] Performing Natural Language Processing (NLP)

[1760] The server inputs the collected communication and document data into a natural language processing (NLP) engine.

[1761] The server performs text tokenization, morphological analysis, part-of-speech tagging, and named entity recognition (NER) to extract important keywords and phrases.

[1762] Step 3:

[1763] emotion recognition

[1764] The server sends the communication data and the text extracted from the document data to the emotion engine.

[1765] The server uses an emotion engine to analyze the sentiment of each piece of text and classify it into categories such as positive, negative, or neutral.

[1766] The server stores the emotion recognition results in a database.

[1767] Step 4:

[1768] Automatic extraction of agenda items

[1769] The server runs an algorithm that integrates the results of NLP analysis and emotion recognition to generate topic candidates.

[1770] The server generates candidate topics such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and emotional information, and stores them in a database.

[1771] Step 5:

[1772] Dynamic Participant Management

[1773] The server dynamically selects the most suitable participants for the generated agenda candidates.

[1774] The server retrieves information such as employee job titles, areas of expertise, and past meeting participation history from a database and selects the most appropriate participants for each agenda item.

[1775] For example, for the agenda item "Analysis of the causes of project delays," the project manager, risk management officer, and project members are selected.

[1776] Step 6:

[1777] Generate a meeting agenda

[1778] The server automatically generates a meeting agenda based on the selected topic and participant list.

[1779] Based on the results of the emotion engine, the server makes adjustments such as placing topics with high negative emotions at the beginning of the agenda.

[1780] Step 7:

[1781] Sending notifications

[1782] The server notifies the generated meeting agenda to interested participants via email and calendar invites.

[1783] The server interfaces with a mail server to send emails containing detailed agendas and meeting information, and uses a calendar system to generate meeting invites.

[1784] Specific examples

[1785] Here is a concrete example of how the system works:

[1786] 1. Devices (e.g., employees' computers and mobile devices) store chat messages and emails related to daily work.

[1787] 2. The server collects this communication data and document data obtained from cloud storage on a daily basis.

[1788] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[1789] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1790] 5. At the same time, the emotion engine recognizes emotions from the user's communication data and document data and determines that the text "Project A is delayed" indicates a strong negative emotion.

[1791] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[1792] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[1793] 8. The server generates the meeting agenda based on the selected topic and participant list, as follows:

[1794] Agenda:

[1795] 1. Analysis of the cause of delay in Project A

[1796] 2. Cost optimization measures

[1797] 3. Next Major Milestone

[1798] Participants:

[1799] Taro Tanaka (Project Manager)

[1800] Hanako Suzuki (Risk Management Officer)

[1801] Ichiro Yamada (Project member)

[1802] 9. The server notifies participants of the generated agenda via email and calendar invites.

[1803] 10. Based on the analysis results of the emotion engine, the server makes adjustments to improve the quality of the meeting, such as placing topics for which users have particularly negative emotions at the top of the agenda.

[1804] In this way, the system reduces the workload involved in setting up and managing meetings, and supports prompt and effective meeting conduct through emotionally sensitive agenda setting.

[1805] Example 2

[1806] 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."

[1807] Conventionally, setting the agenda and selecting participants for a meeting have been done manually, which has the problem of requiring a great deal of effort and time. Furthermore, it has been difficult to consider the user's feelings when setting the agenda, which has also made it difficult to improve the quality of the meeting. The present invention aims to solve these problems and improve the quality of meetings through automated agenda setting and participant selection.

[1808] 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.

[1809] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and selected participants, means for notifying the meeting agenda, means for analyzing user emotions from the communication data and document data, and means for determining the priority of agenda items based on the emotion analysis results. This reduces the workload associated with setting and managing meetings, and enables more effective meetings by setting agendas that take emotions into consideration.

[1810] "Business-related communications data" refers to messages and their metadata exchanged through chat platforms and email systems used within a company.

[1811] "Document data" refers to digital information such as document files, reports, and presentation materials obtained from cloud storage or corporate databases.

[1812] "Natural language processing" refers to techniques and algorithms that enable computers to understand, interpret, and generate human language.

[1813] "Keywords" refer to important words and phrases extracted from communication data and document data through natural language processing.

[1814] "Agenda candidates" refer to items to be discussed at a meeting that are generated based on keywords extracted from communication data and document data.

[1815] "Participants" refers to employees or personnel who should be invited to the meeting.

[1816] "Meeting agenda" refers to a meeting schedule generated based on candidate topics and a selected participant list.

[1817] "Sentiment analysis" refers to the process of identifying emotions contained in user communication and document data and classifying those emotions as positive, negative, neutral, etc.

[1818] "Priority" refers to the criteria used to determine the order in which meeting topics are discussed based on the results of sentiment analysis.

[1819] This invention relates to a system that automates the setting of meeting agendas and participant selection, as well as recognizes user emotions to improve the quality of meetings. This system collects business-related communication data and document data, extracts candidate topics using natural language processing (NLP), dynamically selects participants suitable for the topics, and combines an emotion engine to recognize user emotions and reflect them in the topics and agendas.

[1820] Overall system configuration

[1821] Data collection

[1822] The server collects communication data from the company's internal chat platform (e.g., Slack, Microsoft Teams) and email system. It obtains the data via API, parses it in JSON format, and stores it in a database. It also obtains document data from cloud storage (e.g., Google Drive, SharePoint) and stores it in a database.

[1823] Natural Language Processing (NLP)

[1824] The server inputs the collected communication and document data into an NLP engine (e.g., Google Cloud NLP, AWS Comprehend), which performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and important phrases.

[1825] Emotion Engine

[1826] The server sends the text extracted from the communication data and document data to an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The results are classified into emotion categories such as positive, negative, and neutral, and stored in a database.

[1827] Automatic extraction of agenda items

[1828] The server applies an algorithm based on the results of NLP analysis and sentiment analysis to generate topic candidates. For example, it generates topic candidates such as "Analysis of the causes of project delays" and "Specific measures for cost optimization" based on keywords such as "project delays" and "cost optimization" and sentiment information.

[1829] Dynamic Participant Management

[1830] The server selects the most suitable participants for each candidate agenda item. Employee information such as past meeting history, areas of expertise, and job titles is retrieved from the database, and the most suitable participants are dynamically selected for each agenda item. For example, for the agenda item "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[1831] Generate a meeting agenda

[1832] The server automatically generates a meeting agenda based on the selected topic and participant list. It embeds the topic and participant information in an agenda template and modifies the generated agenda to reflect emotional information. Based on the results of the emotional analysis, it prioritizes topics for which users have particularly strong emotions and reflects them in the agenda.

[1833] Sending notifications

[1834] The server notifies the generated meeting agenda to the relevant participants via email and calendar invites. It works with the mail server to send emails containing the detailed agenda and meeting information, and creates meeting invites using the calendar system.

[1835] Specific examples

[1836] Here is a concrete example of how the system works:

[1837] 1. Work-related chat messages and emails accumulate daily on devices (employee computers and mobile devices).

[1838] 2. The server collects this communication data and document data from cloud storage on a daily basis.

[1839] 3. For example, chat messages such as "Project A is delayed" and document data such as "This month's expense report" are stored on the server.

[1840] 4. The server inputs the saved data into an NLP engine and obtains analysis results based on keywords such as "project delay" and "cost optimization."

[1841] 5. At the same time, the emotion engine determines that the user is expressing strong negative emotion towards the message "Project A is delayed."

[1842] 6. The server generates candidate topics such as "Analysis of the causes of delays in Project A" and "Cost optimization measures" from the analysis results and stores them in the database.

[1843] 7. The server selects the project manager, risk management officer, and project members for the topic "Analyzing the causes of delays in Project A."

[1844] 8. The server generates a meeting agenda based on the selected topic and participant list:

[1845] Agenda:

[1846] 1. Analysis of the cause of delay in Project A

[1847] 2. Cost optimization measures

[1848] 3. Next Major Milestone

[1849] Participants:

[1850] Taro Tanaka (Project Manager)

[1851] Hanako Suzuki (Risk Management Officer)

[1852] Ichiro Yamada (Project member)

[1853] 9. The server notifies participants of the generated agenda via email and calendar invites.

[1854] 10. Based on the analysis results of the emotion engine, the server places topics for which the user has particularly negative emotions at the top of the agenda.

[1855] Prompt Sentence Examples

[1856] The following prompts are used to input the generative AI model:

[1857] Please create a program that extracts topics from business chat messages containing negative sentiment, such as "Project A is delayed," and generates an agenda by selecting the most appropriate meeting participants. Please also provide an example of the generated agenda.

[1858] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1859] Step 1: Data collection

[1860] The server collects communication data from the company's internal chat platforms (e.g., chat service A, chat service B) and email systems. Specifically, it uses APIs to obtain text-based message data and email data from these services, parses it into JSON format, and stores it in a database.

[1861] Input: Data obtained from APIs of chat platforms and email systems.

[1862] Data processing: Convert the acquired data into JSON format.

[1863] Output: JSON formatted communication and email data, stored in a database.

[1864] Step 2: Collect document data from cloud storage

[1865] The server retrieves document data (PDF, Word files, etc.) from cloud storage (e.g., storage service A, storage service B) and stores it in a database. It periodically scans documents using the storage API to retrieve new and updated documents.

[1866] Input: Data obtained from cloud storage API.

[1867] Data processing: Convert the acquired document data into JSON format.

[1868] Output: Document data in JSON format, stored in the database.

[1869] Step 3: Natural Language Processing (NLP)

[1870] The server inputs the communication and document data stored in the database into NLP engines (e.g., NLP engine A, NLP engine B). The NLP engines perform text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract important keywords and phrases.

[1871] Input: Correspondence and document data extracted from the database.

[1872] Data operations: text tokenization, part-of-speech tagging, NER processing.

[1873] Output: Keywords and important phrases, stored in a database.

[1874] Step 4: Emotion analysis using the emotion engine

[1875] The server sends the text analyzed by the NLP engine to an emotion engine (e.g., emotion analysis service A, emotion analysis service B) to analyze the user's emotions. It classifies the emotions into positive, negative, neutral, and other emotional characteristics, and stores the results in a database.

[1876] Input: Keywords and phrases obtained as a result of analysis by the NLP engine.

[1877] Data Computation: Determine the sentiment category of the text through sentiment analysis.

[1878] Output: Sentiment analysis results (positive, negative, neutral), stored in a database.

[1879] Step 5: Automatically extracting topic candidates

[1880] The server generates candidate topics by applying an algorithm based on the results of NLP analysis and sentiment analysis. This extracts topics with keywords such as "project delay" and "cost optimization," and generates specific themes such as "analysis of the causes of project delays" and "cost optimization measures."

[1881] Input: NLP analysis results, sentiment analysis results.

[1882] Data computation: Applying algorithms that combine keywords and sentiment information to generate agendas.

[1883] Output: Agenda candidates, stored in a database.

[1884] Step 6: Dynamic Participant Management

[1885] The server selects the most suitable participants for each candidate topic. Appropriate employees are dynamically selected based on information such as past meeting history, areas of expertise, and job titles. For example, for the topic "Analyzing the causes of project delays," the project manager, risk management officer, and project members are selected.

[1886] Input: Agenda candidate, employee information retrieved from database.

[1887] Data calculation: Applying algorithms to select the most suitable participants based on meeting history, areas of expertise, and job title information.

[1888] Output: Participant list, stored in database.

[1889] Step 7: Generate the meeting agenda

[1890] The server automatically generates a meeting agenda based on the selected topics and participant list. It embeds the topic and participant information in a template and prioritizes topics about which users have particularly strong feelings, reflecting them in the agenda.

[1891] Input: candidate agenda items, participant list, sentiment analysis results.

[1892] Data operations: Embedding information into templates and prioritizing agenda items.

[1893] Output: The generated meeting agenda, stored in a database.

[1894] Step 8: Sending notifications

[1895] The server notifies the generated meeting agenda to interested participants via email and calendar invites. It works with a mail server to send emails containing the detailed agenda and meeting information, and uses a calendar system to create meeting invites.

[1896] Input: The generated meeting agenda.

[1897] Data Computing: Notification generation via email and calendar systems.

[1898] Output: Notification emails and calendar events sent.

[1899] (Application example 2)

[1900] 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."

[1901] Troubles and delays at production sites are issues that require prompt and appropriate countermeasures. However, the traditional method of setting up meetings and selecting participants places a heavy workload on employees, making efficient management difficult, especially in large factories. Furthermore, meetings often fail to take into account the feelings of employees when setting agendas and topics, resulting in ineffective discussions. This hinders the efficient operation of factories, creating a challenge.

[1902] 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.

[1903] In this invention, the server includes means for collecting business-related communication data and document data, means for extracting keywords and agenda candidates from the communication data and document data using natural language processing, means for dynamically selecting relevant participants based on the agenda candidates, means for generating a meeting agenda based on the agenda candidates and the selected participants, means for analyzing user emotions, means for modifying the meeting agenda based on emotion information, and means for notifying the meeting agenda. This improves the efficiency of meeting setup and management for troubleshooting at production sites and enables effective discussions that take user emotions into consideration.

[1904] "Business-related communication data" refers to communication information used within a company, such as chat, email, and monitoring data.

[1905] "Document data" refers to document information such as reports, memos, and files stored in online storage or local storage.

[1906] "Natural language processing" is an information processing technology for analyzing and understanding human language.

[1907] "Keywords" are important words and phrases extracted from the data.

[1908] "Agenda candidate" refers to themes and issues extracted through natural language processing.

[1909] "Relevant participants" are people or positions who should attend the meeting based on the proposed agenda.

[1910] A "meeting agenda" is a document that includes the agenda and attendee list to direct the progress of a meeting.

[1911] "User emotion" is the emotional state of the user as assessed by the emotion analysis engine.

[1912] "Emotion information" is data obtained as a result of analyzing the user's emotions.

[1913] An "emotion engine" is an information processing system for analyzing emotions from communication data and voice data.

[1914] "Dynamic selection of relevant participants based on topic candidates" is a process of selecting the most suitable participants in real time, taking into account their past history and areas of expertise related to the topic.

[1915] A "conversation platform" is a communication tool such as a chat tool or video conferencing software used within a company.

[1916] "Online storage" is a data storage service that can be accessed via the Internet.

[1917] "Notifying a meeting agenda" refers to the act of informing relevant parties of the generated agenda via email or calendar invitation.

[1918] This invention is a conference management system for quickly and effectively resolving problems and delays at production sites. This system collects business-related communication data and document data, extracts candidate topics using natural language processing, dynamically selects participants, and uses an emotion engine to recognize user emotions and reflect them in the topics and agenda.

[1919] The server first collects real-time data from sensors and surveillance cameras. It also obtains production data and trouble reports from the factory management system (MES) via API and stores them in a database. Based on the collected data, the server uses a natural language processing engine (e.g., NLTK, Transformers) to extract important keywords and phrases and generate candidate topics.

[1920] The emotion engine (e.g., Emotion AI) analyzes emotions from collected data and user voice and chat. The emotion results are stored in a database as emotional states, such as positive, negative, or neutral. The server then modifies the meeting agenda based on this emotional information, prioritizing topics that indicate negative emotions in the agenda list.

[1921] The generated meeting agenda is then sent to participants via email or a calendar invitation system, using common business communication tools (e.g., Microsoft Outlook, Google Calendar).

[1922] For example, the server generates candidate topics such as "malfunction of machine A" or "delay on production line B" and selects appropriate participants such as machine maintenance personnel and operators. This information is then reflected in the negative emotions analyzed by the emotion engine, and notified as the generated meeting agenda.

[1923] As an example, the following prompt sentences can be input to a generative AI model:

[1924] "An AI robot collects sensor data and trouble reports from within a factory, inputs this data into a natural language processing and sentiment analysis engine, selects appropriate meeting topics and participants, and automatically generates and notifies meeting agendas. We will show you a specific example of how this works."

[1925] In this way, the present invention can realize an effective conference management system in the production site and improve the efficiency of troubleshooting.

[1926] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1927] Step 1:

[1928] The server collects real-time data from sensors and surveillance cameras. It acquires the data through the sensor API and the surveillance camera API and stores it in a database. The input is sensor data and video data, and the output is the collected data stored in the database.

[1929] Step 2:

[1930] The server obtains production data and trouble reports from the factory management system (MES) via API. This data is also stored in the database. The input is production data and trouble reports, and the output is business data stored in the database.

[1931] Step 3:

[1932] The server inputs collected data (sensor data, production data, trouble reports) into a natural language processing (NLP) engine, which performs text tokenization, part-of-speech tagging, and named entity recognition (NER) to extract keywords and important phrases. The input is collected data, and the output is keywords and topic candidates.

[1933] Step 4:

[1934] The server inputs the collected communication data and the user's voice and chat data into an emotion engine (e.g., Emotion AI) to analyze the user's emotions. The emotion engine analyzes the emotional state, such as positive, negative, or neutral, and stores the results in a database. The input is communication data and voice data, and the output is the emotion analysis results.

[1935] Step 5:

[1936] The server generates topic candidates and dynamically selects relevant participants based on the processed NLP analysis results and sentiment analysis results. The inputs are the NLP analysis results and sentiment analysis results, and the outputs are topic candidates and a list of participants.

[1937] Step 6:

[1938] The server automatically generates a meeting agenda based on the generated topics and the selected participant list. It considers emotional information and prioritizes topics that show negative emotions. The inputs are candidate topics and the participant list, and the output is a revised meeting agenda.

[1939] Step 7:

[1940] The server notifies participants of the generated meeting agenda via email and / or calendar invites. The input to the notification system is the modified meeting agenda, and the output is the meeting notice sent to participants.

[1941] 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.

[1942] 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.

[1943] 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.

[1944] 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.

[1945] FIG. 9 illustrates 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 behaviors 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.

[1946] 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.

[1947] 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).

[1948] 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.

[1949] 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."

[1950] 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.

[1951] 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).

[1952] 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.

[1953] 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.

[1954] 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.

[1955] 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.

[1956] 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.

[1957] 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.

[1958] 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.

[1959] 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.

[1960] 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.

[1961] 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.

[1962] The following is further disclosed regarding the above embodiment.

[1963] (Claim 1)

[1964] a means for collecting business-related communication and document data;

[1965] means for extracting keywords and topic candidates from the communication data and document data using natural language processing;

[1966] means for dynamically selecting relevant participants based on the candidate topics;

[1967] means for generating a meeting agenda based on the candidate topics and selected participants;

[1968] The system includes means for announcing the meeting agenda.

[1969] (Claim 2)

[1970] The system of claim 1 , wherein the communication data is collected from a chat platform.

[1971] (Claim 3)

[1972] The system of claim 1 , wherein the document data is collected from cloud storage.

[1973] "Example 1"

[1974] (Claim 1)

[1975] a means for collecting business-related communication and document data;

[1976] means for extracting keywords and topic candidates from the communication data and document data using natural language processing;

[1977] means for dynamically selecting relevant participants based on the candidate topics;

[1978] means for generating a conference agenda based on the candidate agenda items and selected participants;

[1979] The system includes a means for notifying the meeting agenda.

[1980] (Claim 2)

[1981] The system of claim 1 , wherein the communication data is collected from a message platform.

[1982] (Claim 3)

[1983] The system of claim 1 , wherein the document data is collected from a remote storage.

[1984] "Application Example 1"

[1985] (Claim 1)

[1986] a means for collecting business-related communication and document data;

[1987] means for extracting keywords and topic candidates from the communication data and document data using natural language processing;

[1988] means for dynamically selecting relevant participants based on the candidate topics;

[1989] means for generating a meeting agenda based on the candidate topics and selected participants;

[1990] means for notifying the meeting agenda;

[1991] A means for collecting operational data of communication platforms and equipment operated within the factory;

[1992] means for generating candidate topics using operational data of the communication platform and devices;

[1993] A means for selecting appropriate workers in the factory as participants based on the candidate topics;

[1994] means for notifying the meeting agenda through a robot by voice or display;

[1995] A system including:

[1996] (Claim 2)

[1997] The system of claim 1 , wherein the communication data is collected from a chat platform.

[1998] (Claim 3)

[1999] The system of claim 1 , wherein the document data is collected from cloud storage.

[2000] "Example 2: Combining Emotion Engines"

[2001] (Claim 1)

[2002] a means for collecting business-related communication and document data;

[2003] means for extracting keywords and topic candidates from the communication data and document data using natural language processing;

[2004] means for dynamically selecting relevant participants b...

Claims

1. a means for collecting business-related communication and document data; means for extracting keywords and topic candidates from the communication data and document data using natural language processing; means for dynamically selecting relevant participants based on the candidate topics; means for generating a meeting agenda based on the candidate topics and selected participants; and means for announcing the meeting agenda.

2. The system of claim 1 , wherein the communication data is collected from a chat platform.

3. The system of claim 1 , wherein the document data is collected from cloud storage.

Images