system

Abstract

We provide the system. [Solution] Methods for collecting data from online communication tools used by users, The collected data is analyzed using natural language processing to extract the stress and emotions experienced by the user, A means for generating appropriate advice for the user based on the analysis results, A means of displaying the generated advice on the user's device and providing it to the user, A means of collecting user feedback and incorporating it into the generation of future advice, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] Stress caused by human relationships in the workplace or telework is a problem that seriously affects the performance and mental health of workers. However, it is difficult to accurately grasp the stress and emotions of individual employees and provide appropriate advice and support. Also, it can be difficult for employees themselves to become aware of what is causing their stress. Effective means for solving these problems and facilitating smooth communication are in demand.

Means for Solving the Problems

[0005] This invention is a system that collects data from online communication tools and analyzes the stress and emotions experienced by users using natural language processing. Specifically, the system includes means for collecting data from online communication tools used by users, means for analyzing the collected data using natural language processing to extract the stress and emotions experienced by users, means for generating appropriate advice for users based on the analysis results, means for displaying the generated advice on the user's terminal and providing it to the user, and means for collecting user feedback and reflecting it in the generation of advice for the next time. It may also include means for suggesting the use of industrial physician consultations or in-house counseling services to users, and means for integrating and collecting data from multiple online communication tools used by users. This will support the mental health of users and facilitate smoother workplace communication.

[0006] An "online communication tool" is software or an application used to communicate over the internet in formats such as text messages, voice, and video.

[0007] "Means of data collection" refers to methods or devices for obtaining information such as messages and emails from users' online communication tools.

[0008] "Natural language processing" is a technology that enables computers to understand and analyze human language, and it is a method for analyzing the meaning and emotions of text.

[0009] "Methods for extracting stress and emotions" refer to algorithms and programs used to identify the stress and emotions a user is experiencing from the collected data.

[0010] "Means of generating advice" refers to methods for creating templates and generating messages to provide advice and recommendations to users based on analysis results.

[0011] "Means of displaying advice on the user's device" refers to programs or interfaces for displaying advice messages generated on the device's screen.

[0012] "Means of collecting feedback" refer to methods and programs for gathering user reactions and opinions and using them to improve the system.

[0013] "Methods for proposing industrial physician consultations" refer to methods of generating messages and providing reservation links to users that suggest consultations with specialists.

[0014] "Means of integrating and collecting data from multiple online communication tools" refers to a system or program for integrating and centrally managing data collected from different communication tools. [Brief explanation of the drawing]

[0015] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of the data processing device and smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] Shows an emotion map to which a plurality of emotions are mapped. [Figure 10] Shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

[0016] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

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

[0018] In the following embodiments, a processor with a reference number (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of a plurality of arithmetic units. Further, the processor may be a single type of arithmetic unit or a combination of a plurality of types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0019] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0020] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0021] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0023] [First Embodiment]

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

[0025] As shown in Figure 1, the 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 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0028] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0032] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores the data generation model 58 and the 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 processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0035] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0036] This invention is a system designed to reduce communication stress in the workplace and during teleworking, and to facilitate smooth communication. The system collects data from online communication tools used by the user, analyzes it using natural language processing, and provides advice.

[0037] First, the server connects to the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data, such as messages and emails. For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0038] Next, the server analyzes the collected data using natural language processing (NLP) techniques. The server removes noise from the obtained data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, to identify the cause of stress from a message such as "I'm exhausted from too many meetings." In this case, the server determines that excessive meetings are the main cause of the user's stress.

[0039] Furthermore, the server generates specific advice for the user based on the analysis results. For example, it might suggest to your boss to reduce the number of meetings or to do five minutes of stretching after a meeting to refresh yourself.

[0040] Subsequently, a chat AI running on the device provides the user with generated advice. The user can receive and implement the advice through the chat window on the device. The chat AI also suggests, if necessary, the use of an occupational physician consultation or in-house counseling services. For example, it might suggest, "It seems you've been feeling stressed a lot lately. Why don't you try scheduling an appointment with an occupational physician?"

[0041] Finally, the process includes collecting user feedback. When users input their opinions and thoughts on the advice provided on their device, this feedback is sent to the server. The server analyzes the feedback and incorporates it into future advice generation. This continuous process allows the system to consistently provide optimal support tailored to the user's situation.

[0042] This allows users to receive specific support to alleviate their stress and worries, and to improve the quality of their relationships and communication in the workplace.

[0043] The following describes the processing flow.

[0044] Step 1:

[0045] The server obtains the necessary permissions to access the online communication tools used by the user (e.g., Slack, Gmail). This allows the server to collect information such as the user's messages and emails.

[0046] Step 2:

[0047] The server calls the APIs of online communication tools at a specified schedule (for example, every night at midnight) to collect user-sent and received messages and emails. This includes using the APIs of each tool to retrieve data.

[0048] Step 3:

[0049] The server performs filtering on the collected data to remove noise (e.g., system messages and spam emails). Regular expressions and keyword filtering techniques are used for noise reduction.

[0050] Step 4:

[0051] The server divides the noise-free data into sentences and converts them into a format suitable for natural language processing (NLP). This includes text tokenization and format conversion.

[0052] Step 5:

[0053] The server uses NLP models (e.g., sentiment analysis models) to classify the emotions (e.g., joy, anger, sadness) in each message or email. This helps identify the user's stress levels and emotional state.

[0054] Step 6:

[0055] Based on the sentiment analysis results, the server extracts keywords and phrases that are causing stress. For example, from the message "too many meetings," it identifies "excessive meetings" as a stressor.

[0056] Step 7:

[0057] Based on the analysis results, the server generates specific advice to provide to the user. This advice is template-based and includes content tailored to the individual user's situation.

[0058] Step 8:

[0059] The on-device chat AI presents generated advice to the user. The advice messages are displayed in a natural, conversational format through the chat window, making them easy for the user to understand.

[0060] Step 9:

[0061] If necessary, the terminal's chat AI will suggest to the user the use of occupational health physician consultations or in-house counseling services. For example, if the user's stress level is high, it will provide a dedicated booking link.

[0062] Step 10:

[0063] Users enter feedback on the advice provided in the chat window. This feedback includes opinions on the effectiveness of the advice and other comments.

[0064] Step 11:

[0065] The server analyzes user feedback and incorporates the results into future advice generation. This allows the system to continuously optimize itself and provide support tailored to the user's situation.

[0066] (Example 1)

[0067] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0068] In traditional online communication environments, there is a lack of concrete support to effectively reduce the stress users experience in the workplace or remotely and to facilitate smooth communication. Furthermore, there is a lack of means to identify the causes of stress and provide appropriate advice, making it difficult for users to find ways to cope with stress. Additionally, mechanisms for continuously improving support through feedback are insufficient.

[0069] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0070] In this invention, the server includes means for collecting data from online communication tools used by the user; means for removing irrelevant information from the collected data and dividing the text into sentences; means for performing sentiment analysis and keyword extraction from the divided text to analyze the user's sources of stress and emotions; means for creating prompt sentences using a generation AI model based on the analysis results and generating specific advice for the user; means for displaying the generated advice on the user's terminal and providing it to the user; and means for collecting user feedback and reflecting it in the generation of advice for the next time. This makes it possible to reduce the user's communication-related stress, provide appropriate advice, and continuously improve the support content based on the feedback.

[0071] An "online communication tool" is software or an application that allows users to communicate in real time using text, voice, or video over the internet.

[0072] "Means of data collection" refers to programs and functions that obtain necessary information and messages from users' online communication tools.

[0073] "Means of removing irrelevant information" refer to algorithms and filtering functions used to remove unwanted elements and noise from collected data.

[0074] "Methods for dividing text into sentences" refer to natural language processing techniques for dividing collected text data into smaller sentence units.

[0075] "Sentiment analysis" is an algorithm used to identify emotions such as positive, negative, and neutral from text data.

[0076] Keyword extraction is a natural language processing technique used to extract important words and phrases from text data.

[0077] A "source of stress" refers to an event or factor that causes stress for the user.

[0078] A "generative AI model" is a model that uses machine learning and artificial intelligence technologies to learn patterns from data and generate new information and suggestions.

[0079] A "prompt message" is an instruction given to an AI model to generate specific information or answers.

[0080] "Means for generating advice" refers to functions that create useful suggestions and instructions for users based on analysis results and AI models.

[0081] "Means of collecting feedback" refers to functions for receiving opinions and feedback from users, saving them as data, and analyzing them.

[0082] "Improving support services" means updating and adjusting the advice and support provided by the system based on collected feedback data to make them more effective.

[0083] This invention is a system designed to reduce the stress users experience from communication in the workplace or remotely, and to facilitate smooth communication. The system collects and analyzes data through multiple steps and provides appropriate advice.

[0084] First, the server connects to the online communication tool used by the user. Specifically, it connects using the Slack API and performs authentication. By obtaining the necessary permissions, the server collects messages and related information sent and received by the user. This data collection is performed automatically every night at midnight, and the collected data is stored in a database.

[0085] Next, the server analyzes the collected data using natural language processing (NLP) techniques. Specifically, it first removes noise, eliminating unnecessary information, emojis, URLs, etc. Then, it uses a natural language processing library (e.g., NLTK) to divide the cleaned-up text into sentences. Each of these resulting sentences is then subjected to sentiment analysis and keyword extraction.

[0086] The server uses a sentiment analysis algorithm to evaluate the sentiment of each message. For example, in a message like "I'm tired of too many meetings," "tired" is evaluated as a negative emotion. Furthermore, a keyword extraction algorithm is used to identify important words and phrases (e.g., "meetings" and "tired").

[0087] Based on the analysis results, the server uses a generative AI model to create prompt statements and generate specific advice for the user. These prompt statements are used as input to the generative AI model. As a specific example, the following prompt statement is input to the generative AI model:

[0088] What advice would you provide to help users reduce stress at work? For example, generate advice for a situation where a user says, "I'm exhausted because there are too many meetings."

[0089] This prompt allows the AI ​​model to generate appropriate advice regarding specific stressors.

[0090] Next, the generated advice is sent from the server to the user's device. It is displayed in the chat window on the device, where the user can review and implement the advice. For example, the advice might say, "Try suggesting to your boss that you reduce the number of meetings." The chat AI also suggests medical consultations or the use of internal support services as needed. For example, it might suggest, "You seem to be feeling stressed lately. Why don't you try scheduling a medical consultation?"

[0091] Finally, the process includes collecting user feedback. Through a feedback window on the device, users enter their opinions and impressions of the advice provided. This feedback is sent to the server and incorporated into future advice generation. This allows for continuous improvement of the support provided, ensuring that users receive the best possible support.

[0092] In this way, this system not only reduces the user's communication-related stress and provides appropriate advice, but also enables continuous improvement of support based on feedback.

[0093] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0094] Step 1: Initial setup and data collection

[0095] Subject: Server

[0096] The server connects to the online communication tool used by the user. For example, it uses a Slack API key for authentication. The server obtains the necessary permissions (such as reading messages and retrieving user information). Every day at midnight, it calls the API to collect messages sent and received by the user during that day. The API key and authentication information are used as input, and the message data is stored in a database as output.

[0097] Step 2: Data preprocessing

[0098] Subject: Server

[0099] The server removes unwanted noise from the collected data. Specifically, it removes emojis, URLs, unnecessary spaces, etc. It uses the collected message data as input and outputs cleaned-up text data.

[0100] Step 3: Split the text into sentence units.

[0101] Subject: Server

[0102] The server splits the cleaned text into sentences. It uses a natural language processing library (e.g., NLTK) to split each message into sentences. The input is the cleaned text data, and the output is a list of the split sentences.

[0103] Step 4: Sentiment analysis and keyword extraction

[0104] Subject: Server

[0105] The server performs sentiment analysis on segmented sentences, evaluating the tone and sentiment of the message. It also uses a keyword extraction algorithm (e.g., TF-IDF) to identify important keywords. The input is a list of segmented sentences, and the output is the sentiment evaluation result and a keyword list. For example, from the message "I'm tired because there are too many meetings," "tired" is evaluated as a negative emotion, and "meetings" and "tired" are extracted as important keywords.

[0106] Step 5: Generating Advice

[0107] Subject: Server

[0108] The server uses a generative AI model to create prompt sentences based on the results of sentiment analysis and keyword extraction, generating specific advice for the user. The input consists of sentiment evaluation results and a keyword list, and the output is the generated advice. A concrete example of a prompt sentence used is: "What advice would you provide to reduce the stress a user feels at work? For example, generate advice for the situation, 'I'm exhausted because there are too many meetings.'"

[0109] Step 6: Delivering Advice

[0110] Subject: terminal

[0111] The terminal displays advice received from the server in a chat window. Users can review and implement the advice through this window. The input is advice generated from the server, and the output is the advice displayed in the chat window. For example, advice such as "Suggest reducing the number of meetings to your boss" might be displayed.

[0112] Step 7: Present additional proposals

[0113] Subject: Chat AI

[0114] The chat AI suggests medical consultations and internal support services based on the user's situation. For example, it might suggest, "It seems you've been feeling stressed lately. Why not schedule a medical consultation?" The AI ​​uses the user's emotional analysis results and past advice history as input, and provides additional suggestions as output.

[0115] Step 8: Gathering and analyzing feedback

[0116] Subject: User and Server

[0117] Users input their opinions and feedback on the advice provided through a feedback window on their device. The server receives this feedback and analyzes it. The input is the user's feedback data, and the output is the analysis results of the feedback. These analysis results are reflected in the next advice generation. This allows for continuous improvement of the support provided.

[0118] (Application Example 1)

[0119] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0120] In modern workplaces and teleworking environments, employees are prone to experiencing stress related to their work and communication. In particular, employees in physical stores often experience significant stress when dealing with customers and processing business messages. However, it is difficult for employees themselves to accurately identify their stress levels and deal with them appropriately. This invention aims to provide concrete support methods to help employees alleviate their stress and ensure smooth work operations.

[0121] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0122] In this invention, the server includes means for collecting data from online communication tools used by the user, means for analyzing the collected data using natural language processing to extract the user's stress and emotions, means for generating appropriate advice for the user based on the analysis results, means for displaying the generated advice on the user's terminal and providing it to the user, means for collecting user feedback and reflecting it in the generation of future advice, means for collecting work message data from employees in physical stores and providing advice based on it, means for analyzing the collected work message data using natural language processing to identify the causes of employee stress, and means for analyzing the collected feedback and reflecting it in the generation of future advice. As a result, employees can reduce stress in their daily work and achieve smoother communication.

[0123] "Online communication tools" are software and services that allow users to exchange messages, make calls, and conduct video conferences over the internet.

[0124] "Natural language processing" is a general term for technologies that enable computers to understand, analyze, and generate human language (natural language).

[0125] "Sentiment analysis" is a technique that analyzes text data to determine whether its content is positive, negative, or neutral.

[0126] "Advice" refers to suggestions or suggestions that recommend appropriate measures or actions in response to a specific problem or situation.

[0127] A "device" is a computing device that a user directly operates (for example, a smartphone, tablet, or PC).

[0128] "Feedback" refers to the act or content of a user's response to advice or services provided, such as an opinion or comment.

[0129] A "physical store" is a commercial facility or business location located in a physical space where customer service and sales activities take place.

[0130] "Business message data" refers to the content of messages generated by users during their work, and includes data such as order processing, customer service records, and internal communications.

[0131] "Causes of stress" refer to the psychological and physical burdens and pressures that users experience while performing their work.

[0132] "Advice generation" is the process of creating advice based on analysis results to suggest appropriate actions and measures for the user.

[0133] This invention is a system designed to reduce communication stress in the workplace and during teleworking, and to enable smoother communication.

[0134] This system includes the following means:

[0135] 1. A means of collecting data from online communication tools used by users.

[0136] 2. A method for analyzing collected data using natural language processing to extract the stress and emotions experienced by users.

[0137] 3. A means of generating appropriate advice for the user based on the analysis results.

[0138] 4. A means of displaying the generated advice on the user's device and providing it to the user.

[0139] 5. A means of collecting user feedback and incorporating it into future advice generation.

[0140] 6. A means by which employees at physical stores collect business message data and provide advice based on that data.

[0141] 7. A method for identifying the causes of employee stress by analyzing collected business message data using natural language processing.

[0142] 8. A means of analyzing the collected feedback and incorporating it into the generation of future advice.

[0143] The server first connects to the user's online communication tool, obtains the necessary permissions, and collects data. For example, the server calls an API every night at midnight to retrieve messages sent and received by the user during that day.

[0144] Next, the server analyzes the collected data using natural language processing (NLP) techniques. It removes noise from the collected business message data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, to identify the cause of stress from a message such as "I'm exhausted from too many meetings."

[0145] Therefore, we use the Vader SentimentIntensityAnalyzer from the Natural Language Toolkit (NLTK). The server identifies negative emotions from the data obtained and generates advice based on them. For example, it might generate advice such as "Remember to take a break" or "Let's lower the priority of this task and do it another day."

[0146] The generated advice is provided through a chat window on the user's device. The user can receive and implement this advice. If necessary, the chat AI can also suggest to the user the use of an industrial physician consultation or in-house counseling services.

[0147] When a user submits feedback, it is sent to the server. The server analyzes this feedback and incorporates it into the generation of future advice. This continuous process allows the system to consistently provide optimal support tailored to the user's situation.

[0148] As a concrete example, consider a scenario where employee A at a physical store is replenishing sandwiches that have run out of fillings after lunchtime while also assisting customers. In this situation, work message data is collected and sentiment analysis is performed to identify that employee A is experiencing stress. In response to this situation, advice such as "Remember to take a break during the busy post-lunchtime period" is generated and provided to employee A's terminal.

[0149] Examples of prompts for a generative AI model:

[0150] "Employees at physical stores sometimes experience stress in their daily work. For example, they might feel stressed when they have to adjust sandwiches that have run out of fillings after lunchtime while also dealing with customers. What kind of advice would you offer them?"

[0151] This allows employees in physical stores to use smart assistants for stress management, thereby improving the quality and efficiency of their work.

[0152] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0153] Step 1:

[0154] The server connects to the online communication tool used by the user and obtains the necessary permissions.

[0155] Input: User's online communication tool authentication information.

[0156] Specific operation: The server uses the API to obtain an authentication token and gain permission to access the data.

[0157] Output: User's access permissions to message data.

[0158] Step 2:

[0159] The server collects messages sent and received by users.

[0160] Input: User's online communication tool credentials and API key.

[0161] Specific operation: The server retrieves messages sent and received within a specified time frame (e.g., 1 day) via the API.

[0162] Output: Acquired message data.

[0163] Step 3:

[0164] The server analyzes the collected message data using natural language processing (NLP) techniques.

[0165] Input: Collected message data.

[0166] Specific operation: The server uses the NLTK library to split messages into sentences and perform sentiment analysis.

[0167] Output: Sentiment score (positive, negative, neutral) for each message.

[0168] Step 4:

[0169] The server generates appropriate advice for the user based on the analysis results.

[0170] Input: Sentiment score.

[0171] Specific operation: The server uses a generative AI model to generate appropriate advice corresponding to the sentiment score.

[0172] Output: Generated advice.

[0173] Step 5:

[0174] The generated advice is displayed on the user's device and provided to the user.

[0175] Input: Generated advice.

[0176] Specific actions: Display advice in the chat window on the device. Send notifications as needed.

[0177] Output: Advice displayed to the user.

[0178] Step 6:

[0179] Users follow the advice and input the results and feedback.

[0180] Input: User feedback.

[0181] Specific operation: The user provides feedback through an input form on the device.

[0182] Output: Provided feedback data.

[0183] Step 7:

[0184] The server analyzes the collected feedback and incorporates it into the generation of advice for the next time.

[0185] Input: User feedback data.

[0186] Specific operation: The server analyzes the feedback data and identifies areas for improvement needed to generate advice for the next time.

[0187] Output: Improved advice generation algorithm.

[0188] Step 8:

[0189] The server collects business message data from employees at physical stores and provides advice based on that data.

[0190] Input: Business message data collected by employees at physical stores.

[0191] Specific operation: The server analyzes the business message data from physical stores using natural language processing technology to identify the causes of employee stress.

[0192] Output: Advice to provide to employees.

[0193] Step 9:

[0194] The collected feedback will be analyzed and used to improve the generation of advice for the next time.

[0195] Input: Feedback data collected from employees and users.

[0196] Specific operation: The server analyzes the feedback data to gain insights that will improve the quality of future advice.

[0197] Output: Improved version of the advice that will be generated next time.

[0198] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0199] This invention is a system designed to reduce communication-related stress and facilitate smooth communication for users in the workplace or while teleworking. The system collects data from online communication tools, analyzes it using natural language processing, recognizes the user's emotions using an emotion engine, and provides advice—a comprehensive process.

[0200] First, the server accesses the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data (such as messages and emails). For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0201] Next, the server analyzes the collected data using natural language processing (NLP) techniques. The server removes noise from the obtained data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, identifying the cause of stress from a message such as "I'm tired because there are too many meetings." Based on this, the sentiment engine analyzes the user's emotions in real time and identifies specific emotional states (e.g., stress, joy, anger).

[0202] Furthermore, the server generates advice for the user based on the analysis results of the emotion engine. For example, it might create specific advice such as "suggest to your boss that you reduce the number of meetings" or "do 5 minutes of stretching to refresh yourself after a meeting." The emotion engine can adjust the content and timing of the advice according to the user's emotional state.

[0203] Subsequently, a chat AI running on the device provides the user with generated advice. The user can receive and implement the advice in a natural conversational format through the chat window on the device. The chat AI also suggests, if necessary, the use of an occupational physician consultation or in-house counseling services. For example, it might suggest, "It seems you've been feeling stressed a lot lately. Why don't you try scheduling an appointment with an occupational physician?"

[0204] Finally, the process includes collecting user feedback. When users input their opinions and thoughts on the advice provided on their device, this feedback is sent to the server. The server analyzes the feedback and incorporates it into future advice generation. This allows the system to continuously provide optimal support tailored to the user's situation.

[0205] This allows users to receive specific support to alleviate their stress and worries, improving the quality of workplace relationships and communication. Furthermore, the use of an emotion engine enables real-time responses to the user's emotional state, resulting in more effective support.

[0206] The following describes the processing flow.

[0207] Step 1:

[0208] The server obtains the necessary permissions to access the online communication tools (e.g., Slack, Gmail) that the user uses. For example, by granting access permissions to a tool, the server can retrieve data through the APIs of those tools.

[0209] Step 2:

[0210] The server calls APIs of online communication tools at a specified schedule (for example, every night at midnight) to collect users' sent and received messages and emails. This includes actions such as using the Slack API to retrieve chat history and using the Gmail API to collect emails tagged with specific labels.

[0211] Step 3:

[0212] The server performs filtering on the collected data to remove noise (e.g., system messages and spam). Regular expressions and keyword filtering techniques are used for noise reduction. For example, messages with a specific format or pattern are excluded.

[0213] Step 4:

[0214] The server divides the noise-free data into sentences and converts them into a format suitable for natural language processing (NLP). This includes processes such as text tokenization and format conversion. For example, each sentence is organized as a separate entry in JSON format.

[0215] Step 5:

[0216] The server uses NLP models (e.g., sentiment analysis models) to classify the emotions (e.g., joy, anger, sadness) in each message or email. This allows the server to identify the user's stress levels and emotional state.

[0217] Step 6:

[0218] The server uses an emotion engine to analyze the user's emotions in real time. The emotion engine recognizes emotions in real time from the collected data and reflects the results. For example, if the user is feeling stressed, it quantifies that stress level.

[0219] Step 7:

[0220] Based on the analysis results, the server generates specific advice to provide to the user. This uses a template-based generation method, including advice tailored to each user's situation. For example, it might provide specific advice such as "Suggest to your boss that you reduce the number of meetings."

[0221] Step 8:

[0222] The on-device chat AI presents the user with generated advice. The generated advice messages are displayed in a natural conversational format through the chat window. For example, it might display a message like, "Are you feeling tired from all the meetings lately? Why not suggest to your boss that we reduce the number of meetings?"

[0223] Step 9:

[0224] If necessary, the terminal's chat AI will suggest to the user the use of an occupational physician consultation or in-house counseling services. If the user's stress level is high, it can also provide a dedicated booking link. For example, it may display a suggestion such as, "If you are feeling stressed, we recommend booking an appointment with an occupational physician."

[0225] Step 10:

[0226] Users enter feedback on the advice provided in the chat window. This feedback may include comments on the effectiveness of the advice and other opinions. For example, they might enter a comment such as, "I followed the advice and reduced the number of meetings, and I feel much better."

[0227] Step 11:

[0228] The server analyzes user feedback and incorporates the results into future advice generation. This allows the system to continuously optimize itself and provide support tailored to the user's situation. For example, it relearns keywords and emotions derived from feedback to improve accuracy.

[0229] (Example 2)

[0230] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0231] In today's business environment, users frequently utilize online communication tools, but managing the stress and negative emotions that arise from these interactions is a challenge. This often impairs users' work efficiency and mental health. Furthermore, current systems lack sufficient means to analyze users' emotional states in real time and provide appropriate advice. Moreover, collecting feedback based on the advice provided and incorporating it into future advice is difficult. To address these challenges, a system is needed that efficiently analyzes users' online communication data, automatically generates appropriate advice, and delivers it in real time.

[0232] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0233] In this invention, the server includes means for collecting data from online communication tools used by the user, means for removing noise from the collected data and dividing the text into sentences, means for analyzing the collected data using natural language processing to extract the user's stress and emotions, means for analyzing the user's emotional state in real time using an emotion engine, means for generating appropriate advice for the user based on the analysis results, means for inputting prompt sentences into a generation AI model to automatically generate advice, means for displaying the generated advice on the user's terminal and providing it to the user, and means for collecting user feedback and reflecting it in the generation of advice for the next time. This makes it possible to analyze the user's emotional state in real time, automatically generate appropriate advice, and provide it to the user. Furthermore, it is possible to continuously improve the advice based on feedback, so it is expected that the user's stress in online communication and mental health will be reduced.

[0234] An "online communication tool" is a software application that allows users to communicate with each other via the internet using text messages, voice, video, and other methods.

[0235] "Means of data collection" refers to software modules that automatically retrieve messages and other data from online communication tools used by users.

[0236] "Noise removal methods" refer to software processing that removes unnecessary or duplicate information from collected data to improve the accuracy of the analysis.

[0237] A "means for dividing text into sentences" refers to a software algorithm that divides collected messages or data into individual sentences and performs preprocessing for natural language processing.

[0238] Natural Language Processing (NLP) is a technology that enables computers to understand and generate natural language, performing tasks such as sentiment analysis and keyword extraction.

[0239] An "emotion engine" is a software module that analyzes a user's emotional state based on the results of natural language processing, and identifies stress levels and types of emotions.

[0240] The "means for generating advice" refer to a software module that automatically creates specific and appropriate advice for the user based on the analysis results of the emotion engine.

[0241] A "generative AI model" is an artificial intelligence model that generates natural language based on input prompt sentences, and is a technology used to provide appropriate advice to users.

[0242] A "prompt statement" is an instruction statement that is input into a generative AI model to generate advice, and it is a sentence that includes a specific output request to the model.

[0243] "User's device" refers to an electronic device such as a computer or smartphone used by the user, and is a device for displaying the generated advice.

[0244] A "means for collecting feedback" refers to a software module that collects opinions and impressions from users regarding advice, and analyzes and stores that data.

[0245] This invention is a system designed to reduce communication-related stress and facilitate smooth communication for users in the workplace or while teleworking. The system collects data from online communication tools, analyzes it using natural language processing, recognizes the user's emotions using an emotion engine, and provides advice—a comprehensive process.

[0246] First, the server accesses the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data (such as messages and emails). For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0247] Next, the server preprocesses the collected data. Specifically, it removes noise from the collected data and divides the text into sentences. This preprocessing transforms the data into a format suitable for analysis. For example, the server filters out duplicate messages and meaningless data and divides the text into individual sentences.

[0248] The server then analyzes the pre-processed data using natural language processing (NLP) techniques. NLP libraries such as SpaCy and NLTK are used. Sentiment analysis and keyword extraction are performed to identify the cause of stress from messages like, for example, "I'm exhausted from too many meetings." As a concrete example, prompts such as, "The user is feeling overworked. Please suggest ways to refresh," can be input into a generating AI model to create appropriate advice.

[0249] Furthermore, the server uses an emotion engine to analyze the user's emotional state in real time. The emotion engine identifies the user's stress level and type of emotion based on the results of natural language processing. For example, a message like "I'm exhausted from too many meetings" would be judged as "high stress."

[0250] Next, the server generates appropriate advice for the user based on the analysis results of the emotion engine. It automatically generates advice by inputting prompts into a generation AI model (e.g., GPT-3). Specific examples of advice include "suggest to your boss that you reduce the number of meetings" and "do 5 minutes of stretching to refresh yourself after a meeting."

[0251] The device then provides the user with the generated advice. Through a chat window on the device, the user can receive and implement the advice in a natural, conversational format. For example, it might suggest, "It seems you've been feeling stressed lately. Why not schedule a consultation with an occupational physician?"

[0252] Finally, the device collects user feedback and sends it to the server. A feedback form is displayed in the device's chat window, and when the user enters their opinions and thoughts on the advice provided, that feedback is sent to the server. The server analyzes the feedback and incorporates it into the generation of future advice. This allows the system to continuously provide optimal support tailored to the user's situation.

[0253] This entire process allows users to receive specific support to alleviate their stress and worries, thereby improving the quality of their relationships and communication in the workplace. Furthermore, the use of an emotion engine enables real-time responses to the user's emotional state, resulting in more effective support.

[0254] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0255] Step 1: Data Collection

[0256] The server calls the APIs of online communication tools used by users (e.g., Slack) to collect message data that is sent and received on a daily basis.

[0257] Input: User's API authentication information for their online communication tool.

[0258] Output: Message data sent and received by the user during that day.

[0259] Specific actions:

[0260] The server executes scheduled jobs every night at 12:00.

[0261] The server authenticates with Slack using API keys and authentication tokens.

[0262] The server calls the API to retrieve messages sent and received by the user.

[0263] The retrieved message data is saved to the database.

[0264] Step 2: Data Preprocessing

[0265] The server removes unnecessary information (noise) from the collected message data and divides the text into sentences.

[0266] Input: Collected message data.

[0267] Output: Preprocessed text data.

[0268] Specific actions:

[0269] The server filters out duplicate messages and meaningless data (e.g., system messages) from the data.

[0270] The server divides the text into sentences and converts them into a format that is easy to parse.

[0271] Step 3: Natural Language Processing (NLP) Analysis

[0272] The server uses the pre-processed data to perform sentiment analysis and keyword extraction.

[0273] Input: Preprocessed text data.

[0274] Output: Sentimental states and keywords extracted from text data.

[0275] Specific operations:

[0276] The server performs sentiment analysis using an NLP library (e.g., SpaCy, NLTK).

[0277] Example: Messages containing emotions such as "tired" and "happy" are classified as positive, negative, or neutral.

[0278] The server extracts important keywords using techniques such as TF-IDF and Word2Vec.

[0279] Step 4: Analysis of emotional state in the emotion engine

[0280] The server uses the emotion engine to analyze the user's emotional state in real time.

[0281] Input: NLP analysis result.

[0282] Output: Quantified emotional state (e.g., stress level).

[0283] Specific operations:

[0284] The emotion engine uses the result of sentiment analysis to quantify the user's emotional state.

[0285] Example: From the message "Too many meetings and I'm tired", the emotion engine determines that "stress is high".

[0286] Step 5: Advice generation

[0287] [[ID=​​​​​​​​​​ Specific actions:

[0291] The server inputs prompt text into the generated AI model (e.g., GPT-3) and automatically generates advice.

[0292] Example prompt: "The user is feeling overworked. Please suggest ways for them to refresh themselves."

[0293] The server then provides the generated advice to the user in a personalized manner.

[0294] Step 6: Providing advice

[0295] The device notifies the user via chat AI to provide advice.

[0296] Input: Generated advice.

[0297] Output: Advice provided to the user.

[0298] Specific actions:

[0299] Advice will be displayed in the chat window on the device.

[0300] Examples: "Try suggesting we reduce the number of meetings," or "Let's do some 5-minute stretches to refresh ourselves."

[0301] The chat AI provides advice through a conversational format with the user.

[0302] Step 7: Gathering Feedback

[0303] The device collects user feedback and sends it to the server.

[0304] Input: User feedback.

[0305] Output: Feedback data sent to the server.

[0306] Specific operations:

[0307] The terminal displays a feedback form within the chat window.

[0308] Example: "Was this advice helpful?" "Do you have any other opinions?"

[0309] The user enters feedback and presses the send button.

[0310] Step 8: Analysis and reflection of feedback

[0311] The server analyzes the collected feedback and reflects it in the next advice generation process.

[0312] Input: Collected feedback data.

[0313] Output: Improved advice generation process.

[0314] Specific operations:

[0315] The server analyzes the feedback data and saves the obtained information as learning data.

[0316] Based on the feedback, the server improves the prompt text and advice content input into the generation AI model.

[0317] Through this series of processes, users can receive appropriate support to reduce their stress and worries, and it becomes possible to improve workplace human relations and the quality of communication.

[0318] (Application Example 2)

[0319] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0320] In modern workplaces and brick-and-mortar stores, employees often experience stress from communicating with customers and colleagues. This leads to problems such as decreased employee performance and customer satisfaction. Furthermore, there is a growing need to focus on stress management and implement effective support systems.

[0321] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting data from a communication tool used by the user, means for analyzing the collected data using natural language processing to extract the tension and emotions the user is experiencing, means for generating appropriate advice for the user based on the analysis results, means for displaying the generated advice on the user's display device and providing it to the user, means for collecting the user's reactions and reflecting them in the generation of advice for the next time, and means for detecting the user's state of tension in real time during the conversation and generating and providing appropriate suggestions. As a result, the user can receive concrete support to reduce stress caused by communication and enable smooth conversations, making it possible to improve interpersonal relationships and customer satisfaction in the workplace and physical stores.

[0322] A "communication tool" refers to an online platform or application that users use to exchange messages and information with other people.

[0323] "Natural language processing" refers to the process by which computers understand and analyze natural language text. Specifically, it includes techniques such as grammatical analysis, sentiment analysis, and keyword extraction.

[0324] "Tension" refers to the stress and anxiety that users experience during communication. This often involves psychological burden.

[0325] "Emotions" refer to the user's psychological state, including emotional states such as joy, anger, sadness, and surprise.

[0326] "Advice" refers to specific suggestions or advice that help users act appropriately.

[0327] A "display device" is a device that a user uses to visually confirm information. This includes smart glasses and smartphones.

[0328] "Response" refers to the feedback or actions a user takes in response to the advice they receive.

[0329] "Real-time" refers to processing or responding immediately without delay.

[0330] A "suggestion" refers to a suggestion of actions or measures provided based on the user's situation and emotional state.

[0331] "Support" refers to assistance and services provided to users to resolve stress and communication problems.

[0332] Embodiments of this invention include systems utilizing communication tools, natural language processing, sentiment analysis engines, display devices, and feedback collection means. Their specific configuration and operation are described below.

[0333] The system includes means for collecting data from communication tools that users use on a daily basis. For example, the server uses the API of the communication tool to retrieve messages sent and received by users at regular intervals. Since the collected data is not suitable for analysis in its raw form, the server analyzes this data using natural language processing (NLP) techniques.

[0334] During the analysis phase, text data is tokenized, and processes such as sentiment analysis and keyword extraction are performed. This allows for the identification of the user's tension and emotions. For example, if a user sends a message such as "I'm exhausted from having to deal with too many customers," this statement is analyzed as an indication of stress.

[0335] Next, based on the analysis results, the server uses an emotion analysis engine to evaluate the user's emotional state. It then generates appropriate advice corresponding to the results. For example, if the server determines that the user is feeling stressed, it will generate specific advice such as "take a short break" or "talk to your supervisor."

[0336] The generated advice is displayed in real time through the user's display device (e.g., smart glasses or smartphone). This allows the user to receive advice quickly during the conversation. It also has a function to detect the user's real-time level of tension and provide appropriate suggestions immediately.

[0337] Next, the system includes a means for users to provide feedback on the advice they receive. Users input their opinions and thoughts on the advice, and this feedback is sent to the server. The server analyzes the feedback and incorporates it into the next advice generation. This allows the system to continuously provide optimal support tailored to the user's situation.

[0338] The specific hardware and software used include:

[0339] Smart glasses: A device worn by store employees to display information in real time.

[0340] SmartGlassesAPI: An API that collects data from smart glasses and communicates with them.

[0341] TextAnalyzer (NLP Module): A module for natural language processing.

[0342] EmotionRecognitionEngine: Emotion Analysis Engine

[0343] ChatAI: A chat AI for generating and providing advice

[0344] As a concrete example, suppose a store employee is serving a customer and the customer asks for a special discount. In this case, the system detects in real time that the employee is feeling stressed and offers advice such as, "Relax and try offering the best possible discount." This reduces employee stress and improves customer satisfaction.

[0345] Examples of prompt statements are as follows:

[0346] "We will provide conversation data between store staff and customers. If the staff member is experiencing stress, please generate advice on how to handle the situation: {conversation data}"

[0347] This will enable employees to communicate more smoothly and improve the quality of customer service in the workplace and at physical stores.

[0348] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0349] Step 1:

[0350] The server collects data through the APIs of the communication tools used by the user. Specifically, it periodically calls the APIs to retrieve messages and chat logs. This input data consists of messages sent and received by the user. In this case, it also includes obtaining real-time conversation data using, for example, the data collection function of smart glasses.

[0351] Step 2:

[0352] The server stores the collected data as text data. Next, it processes the collected text data using natural language processing (NLP) techniques. The input data is the collected text data, and the output data is the analyzed text structure and extracted keywords. Specifically, it performs text tokenization, grammatical analysis, keyword extraction, and sentiment analysis.

[0353] Step 3:

[0354] The server uses a sentiment analysis engine to evaluate the user's emotional state based on the analysis results obtained through natural language processing. The input data consists of the analyzed text structure and keywords, while the output data is the user's emotional state (e.g., stress, joy, anger). Here, an emotion detection algorithm is used to identify emotions within the text and quantify the emotional state.

[0355] Step 4:

[0356] Based on the results of the emotion analysis engine, the server generates appropriate advice. The input data is the user's emotional state, and the output data is specific advice and suggestions. Using a generative AI model, it generates advice that is appropriate to the user's current situation. For example, if the user is feeling stressed, it will output specific advice such as "take a short break."

[0357] Step 5:

[0358] The generated advice is transmitted in real time to the user's display device, such as smart glasses or a smartphone. The input data is the generated advice, and the output data is the advice message displayed to the user. The smart glasses' display function is used to provide the user with advice visually in real time.

[0359] Step 6:

[0360] When a user enters feedback on advice, that feedback data is sent to the server. The input data is the user's feedback, and the output data is the feedback information stored on the server. The user enters the feedback through smart glasses.

[0361] Step 7:

[0362] The server analyzes the collected feedback data and incorporates it into the next advice generation. The input data is feedback information, and the output data is the improved advice generation algorithm. Specifically, a machine learning algorithm is used to analyze the feedback data and tune the AI ​​model for generation.

[0363] The above outlines the specific processing steps required to implement this system. This process allows users to receive support in real time, reducing communication stress.

[0364] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0365] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0366] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0367] [Second Embodiment]

[0368] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0369] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0370] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0371] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0372] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0373] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0374] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0375] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0376] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0377] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0378] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0379] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0380] This invention is a system designed to reduce communication stress in the workplace and during teleworking, and to facilitate smooth communication. The system collects data from online communication tools used by the user, analyzes it using natural language processing, and provides advice.

[0381] First, the server connects to the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data, such as messages and emails. For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0382] Next, the server analyzes the collected data using natural language processing (NLP) techniques. The server removes noise from the obtained data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, to identify the cause of stress from a message such as "I'm exhausted from too many meetings." In this case, the server determines that excessive meetings are the main cause of the user's stress.

[0383] Furthermore, the server generates specific advice for the user based on the analysis results. For example, it might suggest to your boss to reduce the number of meetings or to do five minutes of stretching after a meeting to refresh yourself.

[0384] Subsequently, a chat AI running on the device provides the user with generated advice. The user can receive and implement the advice through the chat window on the device. The chat AI also suggests, if necessary, the use of an occupational physician consultation or in-house counseling services. For example, it might suggest, "It seems you've been feeling stressed a lot lately. Why don't you try scheduling an appointment with an occupational physician?"

[0385] Finally, the process includes collecting user feedback. When users input their opinions and thoughts on the advice provided on their device, this feedback is sent to the server. The server analyzes the feedback and incorporates it into future advice generation. This continuous process allows the system to consistently provide optimal support tailored to the user's situation.

[0386] This allows users to receive specific support to alleviate their stress and worries, and to improve the quality of their relationships and communication in the workplace.

[0387] The following describes the processing flow.

[0388] Step 1:

[0389] The server obtains the necessary permissions to access the online communication tools used by the user (e.g., Slack, Gmail). This allows the server to collect information such as the user's messages and emails.

[0390] Step 2:

[0391] The server calls the APIs of online communication tools at a specified schedule (for example, every night at midnight) to collect user-sent and received messages and emails. This includes using the APIs of each tool to retrieve data.

[0392] Step 3:

[0393] The server performs filtering on the collected data to remove noise (e.g., system messages and spam emails). Regular expressions and keyword filtering techniques are used for noise reduction.

[0394] Step 4:

[0395] The server divides the noise-free data into sentences and converts them into a format suitable for natural language processing (NLP). This includes text tokenization and format conversion.

[0396] Step 5:

[0397] The server uses NLP models (e.g., sentiment analysis models) to classify the emotions (e.g., joy, anger, sadness) in each message or email. This helps identify the user's stress levels and emotional state.

[0398] Step 6:

[0399] Based on the sentiment analysis results, the server extracts keywords and phrases that are causing stress. For example, from the message "too many meetings," it identifies "excessive meetings" as a stressor.

[0400] Step 7:

[0401] Based on the analysis results, the server generates specific advice to provide to the user. This advice is template-based and includes content tailored to the individual user's situation.

[0402] Step 8:

[0403] The on-device chat AI presents generated advice to the user. The advice messages are displayed in a natural, conversational format through the chat window, making them easy for the user to understand.

[0404] Step 9:

[0405] If necessary, the terminal's chat AI will suggest to the user the use of occupational health physician consultations or in-house counseling services. For example, if the user's stress level is high, it will provide a dedicated booking link.

[0406] Step 10:

[0407] Users enter feedback on the advice provided in the chat window. This feedback includes opinions on the effectiveness of the advice and other comments.

[0408] Step 11:

[0409] The server analyzes user feedback and incorporates the results into future advice generation. This allows the system to continuously optimize itself and provide support tailored to the user's situation.

[0410] (Example 1)

[0411] Next, we will describe Example 1. 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".

[0412] In traditional online communication environments, there is a lack of concrete support to effectively reduce the stress users experience in the workplace or remotely and to facilitate smooth communication. Furthermore, there is a lack of means to identify the causes of stress and provide appropriate advice, making it difficult for users to find ways to cope with stress. Additionally, mechanisms for continuously improving support through feedback are insufficient.

[0413] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0414] In this invention, the server includes means for collecting data from online communication tools used by the user; means for removing irrelevant information from the collected data and dividing the text into sentences; means for performing sentiment analysis and keyword extraction from the divided text to analyze the user's sources of stress and emotions; means for creating prompt sentences using a generation AI model based on the analysis results and generating specific advice for the user; means for displaying the generated advice on the user's terminal and providing it to the user; and means for collecting user feedback and reflecting it in the generation of advice for the next time. This makes it possible to reduce the user's communication-related stress, provide appropriate advice, and continuously improve the support content based on the feedback.

[0415] An "online communication tool" is software or an application that allows users to communicate in real time using text, voice, or video over the internet.

[0416] "Means of data collection" refers to programs and functions that obtain necessary information and messages from users' online communication tools.

[0417] "Means of removing irrelevant information" refer to algorithms and filtering functions used to remove unwanted elements and noise from collected data.

[0418] "Methods for dividing text into sentences" refer to natural language processing techniques for dividing collected text data into smaller sentence units.

[0419] "Sentiment analysis" is an algorithm used to identify emotions such as positive, negative, and neutral from text data.

[0420] Keyword extraction is a natural language processing technique used to extract important words and phrases from text data.

[0421] A "source of stress" refers to an event or factor that causes stress for the user.

[0422] A "generative AI model" is a model that uses machine learning and artificial intelligence technologies to learn patterns from data and generate new information and suggestions.

[0423] A "prompt message" is an instruction given to an AI model to generate specific information or answers.

[0424] "Means for generating advice" refers to functions that create useful suggestions and instructions for users based on analysis results and AI models.

[0425] "Means of collecting feedback" refers to functions for receiving opinions and feedback from users, saving them as data, and analyzing them.

[0426] "Improving support services" means updating and adjusting the advice and support provided by the system based on collected feedback data to make them more effective.

[0427] This invention is a system designed to reduce the stress users experience from communication in the workplace or remotely, and to facilitate smooth communication. The system collects and analyzes data through multiple steps and provides appropriate advice.

[0428] First, the server connects to the online communication tool used by the user. Specifically, it connects using the Slack API and performs authentication. By obtaining the necessary permissions, the server collects messages and related information sent and received by the user. This data collection is performed automatically every night at midnight, and the collected data is stored in a database.

[0429] Next, the server analyzes the collected data using natural language processing (NLP) techniques. Specifically, it first removes noise, eliminating unnecessary information, emojis, URLs, etc. Then, it uses a natural language processing library (e.g., NLTK) to divide the cleaned-up text into sentences. Each of these resulting sentences is then subjected to sentiment analysis and keyword extraction.

[0430] The server uses a sentiment analysis algorithm to evaluate the sentiment of each message. For example, in a message like "I'm tired of too many meetings," "tired" is evaluated as a negative emotion. Furthermore, a keyword extraction algorithm is used to identify important words and phrases (e.g., "meetings" and "tired").

[0431] Based on the analysis results, the server uses a generative AI model to create prompt statements and generate specific advice for the user. These prompt statements are used as input to the generative AI model. As a specific example, the following prompt statement is input to the generative AI model:

[0432] What advice would you provide to help users reduce stress at work? For example, generate advice for a situation where a user says, "I'm exhausted because there are too many meetings."

[0433] This prompt allows the AI ​​model to generate appropriate advice regarding specific stressors.

[0434] Next, the generated advice is sent from the server to the user's device. It is displayed in the chat window on the device, where the user can review and implement the advice. For example, the advice might say, "Try suggesting to your boss that you reduce the number of meetings." The chat AI also suggests medical consultations or the use of internal support services as needed. For example, it might suggest, "You seem to be feeling stressed lately. Why don't you try scheduling a medical consultation?"

[0435] Finally, the process includes collecting user feedback. Through a feedback window on the device, users enter their opinions and impressions of the advice provided. This feedback is sent to the server and incorporated into future advice generation. This allows for continuous improvement of the support provided, ensuring that users receive the best possible support.

[0436] In this way, this system not only reduces the user's communication-related stress and provides appropriate advice, but also enables continuous improvement of support based on feedback.

[0437] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0438] Step 1: Initial setup and data collection

[0439] Subject: Server

[0440] The server connects to the online communication tool used by the user. For example, it uses a Slack API key for authentication. The server obtains the necessary permissions (such as reading messages and retrieving user information). Every day at midnight, it calls the API to collect messages sent and received by the user during that day. The API key and authentication information are used as input, and the message data is stored in a database as output.

[0441] Step 2: Data preprocessing

[0442] Subject: Server

[0443] The server removes unwanted noise from the collected data. Specifically, it removes emojis, URLs, unnecessary spaces, etc. It uses the collected message data as input and outputs cleaned-up text data.

[0444] Step 3: Split the text into sentence units.

[0445] Subject: Server

[0446] The server splits the cleaned text into sentences. It uses a natural language processing library (e.g., NLTK) to split each message into sentences. The input is the cleaned text data, and the output is a list of the split sentences.

[0447] Step 4: Sentiment analysis and keyword extraction

[0448] Subject: Server

[0449] The server performs sentiment analysis on segmented sentences, evaluating the tone and sentiment of the message. It also uses a keyword extraction algorithm (e.g., TF-IDF) to identify important keywords. The input is a list of segmented sentences, and the output is the sentiment evaluation result and a keyword list. For example, from the message "I'm tired because there are too many meetings," "tired" is evaluated as a negative emotion, and "meetings" and "tired" are extracted as important keywords.

[0450] Step 5: Generating Advice

[0451] Subject: Server

[0452] The server uses a generative AI model to create prompt sentences based on the results of sentiment analysis and keyword extraction, generating specific advice for the user. The input consists of sentiment evaluation results and a keyword list, and the output is the generated advice. A concrete example of a prompt sentence used is: "What advice would you provide to reduce the stress a user feels at work? For example, generate advice for the situation, 'I'm exhausted because there are too many meetings.'"

[0453] Step 6: Delivering Advice

[0454] Subject: terminal

[0455] The terminal displays advice received from the server in a chat window. Users can review and implement the advice through this window. The input is advice generated from the server, and the output is the advice displayed in the chat window. For example, advice such as "Suggest reducing the number of meetings to your boss" might be displayed.

[0456] Step 7: Present additional proposals

[0457] Subject: Chat AI

[0458] The chat AI suggests medical consultations and internal support services based on the user's situation. For example, it might suggest, "It seems you've been feeling stressed lately. Why not schedule a medical consultation?" The AI ​​uses the user's emotional analysis results and past advice history as input, and provides additional suggestions as output.

[0459] Step 8: Gathering and analyzing feedback

[0460] Subject: User and Server

[0461] Users input their opinions and feedback on the advice provided through a feedback window on their device. The server receives this feedback and analyzes it. The input is the user's feedback data, and the output is the analysis results of the feedback. These analysis results are reflected in the next advice generation. This allows for continuous improvement of the support provided.

[0462] (Application Example 1)

[0463] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0464] In modern workplaces and teleworking environments, employees are prone to experiencing stress related to their work and communication. In particular, employees in physical stores often experience significant stress when dealing with customers and processing business messages. However, it is difficult for employees themselves to accurately identify their stress levels and deal with them appropriately. This invention aims to provide concrete support methods to help employees alleviate their stress and ensure smooth work operations.

[0465] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0466] In this invention, the server includes means for collecting data from online communication tools used by the user, means for analyzing the collected data using natural language processing to extract the user's stress and emotions, means for generating appropriate advice for the user based on the analysis results, means for displaying the generated advice on the user's terminal and providing it to the user, means for collecting user feedback and reflecting it in the generation of future advice, means for collecting work message data from employees in physical stores and providing advice based on it, means for analyzing the collected work message data using natural language processing to identify the causes of employee stress, and means for analyzing the collected feedback and reflecting it in the generation of future advice. As a result, employees can reduce stress in their daily work and achieve smoother communication.

[0467] "Online communication tools" are software and services that allow users to exchange messages, make calls, and conduct video conferences over the internet.

[0468] "Natural language processing" is a general term for technologies that enable computers to understand, analyze, and generate human language (natural language).

[0469] "Sentiment analysis" is a technique that analyzes text data to determine whether its content is positive, negative, or neutral.

[0470] "Advice" refers to suggestions or suggestions that recommend appropriate measures or actions in response to a specific problem or situation.

[0471] A "device" is a computing device that a user directly operates (for example, a smartphone, tablet, or PC).

[0472] "Feedback" refers to the act or content of a user's response to advice or services provided, such as an opinion or comment.

[0473] A "physical store" is a commercial facility or business location located in a physical space where customer service and sales activities take place.

[0474] "Business message data" refers to the content of messages generated by users during their work, and includes data such as order processing, customer service records, and internal communications.

[0475] "Causes of stress" refer to the psychological and physical burdens and pressures that users experience while performing their work.

[0476] "Advice generation" is the process of creating advice based on analysis results to suggest appropriate actions and measures for the user.

[0477] This invention is a system designed to reduce communication stress in the workplace and during teleworking, and to enable smoother communication.

[0478] This system includes the following means:

[0479] 1. A means of collecting data from online communication tools used by users.

[0480] 2. A method for analyzing collected data using natural language processing to extract the stress and emotions experienced by users.

[0481] 3. A means of generating appropriate advice for the user based on the analysis results.

[0482] 4. A means of displaying the generated advice on the user's device and providing it to the user.

[0483] 5. A means of collecting user feedback and incorporating it into future advice generation.

[0484] 6. A means by which employees at physical stores collect business message data and provide advice based on that data.

[0485] 7. A method for identifying the causes of employee stress by analyzing collected business message data using natural language processing.

[0486] 8. A means of analyzing the collected feedback and incorporating it into the generation of future advice.

[0487] The server first connects to the user's online communication tool, obtains the necessary permissions, and collects data. For example, the server calls an API every night at midnight to retrieve messages sent and received by the user during that day.

[0488] Next, the server analyzes the collected data using natural language processing (NLP) techniques. It removes noise from the collected business message data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, to identify the cause of stress from a message such as "I'm exhausted from too many meetings."

[0489] Therefore, we use the Vader SentimentIntensityAnalyzer from the Natural Language Toolkit (NLTK). The server identifies negative emotions from the data obtained and generates advice based on them. For example, it might generate advice such as "Remember to take a break" or "Let's lower the priority of this task and do it another day."

[0490] The generated advice is provided through a chat window on the user's device. The user can receive and implement this advice. If necessary, the chat AI can also suggest to the user the use of an industrial physician consultation or in-house counseling services.

[0491] When a user submits feedback, it is sent to the server. The server analyzes this feedback and incorporates it into the generation of future advice. This continuous process allows the system to consistently provide optimal support tailored to the user's situation.

[0492] As a concrete example, consider a scenario where employee A at a physical store is replenishing sandwiches that have run out of fillings after lunchtime while also assisting customers. In this situation, work message data is collected and sentiment analysis is performed to identify that employee A is experiencing stress. In response to this situation, advice such as "Remember to take a break during the busy post-lunchtime period" is generated and provided to employee A's terminal.

[0493] Examples of prompts for a generative AI model:

[0494] "Employees at physical stores sometimes experience stress in their daily work. For example, they might feel stressed when they have to adjust sandwiches that have run out of fillings after lunchtime while also dealing with customers. What kind of advice would you offer them?"

[0495] This allows employees in physical stores to use smart assistants for stress management, thereby improving the quality and efficiency of their work.

[0496] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0497] Step 1:

[0498] The server connects to the online communication tool used by the user and obtains the necessary permissions.

[0499] Input: User's online communication tool authentication information.

[0500] Specific operation: The server uses the API to obtain an authentication token and gain permission to access the data.

[0501] Output: User's access permissions to message data.

[0502] Step 2:

[0503] The server collects messages sent and received by users.

[0504] Input: User's online communication tool credentials and API key.

[0505] Specific operation: The server retrieves messages sent and received within a specified time frame (e.g., 1 day) via the API.

[0506] Output: Acquired message data.

[0507] Step 3:

[0508] The server analyzes the collected message data using natural language processing (NLP) techniques.

[0509] Input: Collected message data.

[0510] Specific operation: The server uses the NLTK library to split messages into sentences and perform sentiment analysis.

[0511] Output: Sentiment score (positive, negative, neutral) for each message.

[0512] Step 4:

[0513] The server generates appropriate advice for the user based on the analysis results.

[0514] Input: Sentiment score.

[0515] Specific operation: The server uses a generative AI model to generate appropriate advice corresponding to the sentiment score.

[0516] Output: Generated advice.

[0517] Step 5:

[0518] The generated advice is displayed on the user's device and provided to the user.

[0519] Input: Generated advice.

[0520] Specific actions: Display advice in the chat window on the device. Send notifications as needed.

[0521] Output: Advice displayed to the user.

[0522] Step 6:

[0523] Users follow the advice and input the results and feedback.

[0524] Input: User feedback.

[0525] Specific operation: The user provides feedback through an input form on the device.

[0526] Output: Provided feedback data.

[0527] Step 7:

[0528] The server analyzes the collected feedback and incorporates it into the generation of advice for the next time.

[0529] Input: User feedback data.

[0530] Specific operation: The server analyzes the feedback data and identifies areas for improvement needed to generate advice for the next time.

[0531] Output: Improved advice generation algorithm.

[0532] Step 8:

[0533] The server collects business message data from employees at physical stores and provides advice based on that data.

[0534] Input: Business message data collected by employees at physical stores.

[0535] Specific operation: The server analyzes the business message data from physical stores using natural language processing technology to identify the causes of employee stress.

[0536] Output: Advice to provide to employees.

[0537] Step 9:

[0538] The collected feedback will be analyzed and used to improve the generation of advice for the next time.

[0539] Input: Feedback data collected from employees and users.

[0540] Specific operation: The server analyzes the feedback data to gain insights that will improve the quality of future advice.

[0541] Output: Improved version of the advice that will be generated next time.

[0542] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0543] This invention is a system designed to reduce communication-related stress and facilitate smooth communication for users in the workplace or while teleworking. The system collects data from online communication tools, analyzes it using natural language processing, recognizes the user's emotions using an emotion engine, and provides advice—a comprehensive process.

[0544] First, the server accesses the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data (such as messages and emails). For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0545] Next, the server analyzes the collected data using natural language processing (NLP) techniques. The server removes noise from the obtained data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, identifying the cause of stress from a message such as "I'm tired because there are too many meetings." Based on this, the sentiment engine analyzes the user's emotions in real time and identifies specific emotional states (e.g., stress, joy, anger).

[0546] Furthermore, the server generates advice for the user based on the analysis results of the emotion engine. For example, it might create specific advice such as "suggest to your boss that you reduce the number of meetings" or "do 5 minutes of stretching to refresh yourself after a meeting." The emotion engine can adjust the content and timing of the advice according to the user's emotional state.

[0547] Subsequently, a chat AI running on the device provides the user with generated advice. The user can receive and implement the advice in a natural conversational format through the chat window on the device. The chat AI also suggests, if necessary, the use of an occupational physician consultation or in-house counseling services. For example, it might suggest, "It seems you've been feeling stressed a lot lately. Why don't you try scheduling an appointment with an occupational physician?"

[0548] Finally, the process includes collecting user feedback. When users input their opinions and thoughts on the advice provided on their device, this feedback is sent to the server. The server analyzes the feedback and incorporates it into future advice generation. This allows the system to continuously provide optimal support tailored to the user's situation.

[0549] This allows users to receive specific support to alleviate their stress and worries, improving the quality of workplace relationships and communication. Furthermore, the use of an emotion engine enables real-time responses to the user's emotional state, resulting in more effective support.

[0550] The following describes the processing flow.

[0551] Step 1:

[0552] The server obtains the necessary permissions to access the online communication tools (e.g., Slack, Gmail) that the user uses. For example, by granting access permissions to a tool, the server can retrieve data through the APIs of those tools.

[0553] Step 2:

[0554] The server calls APIs of online communication tools at a specified schedule (for example, every night at midnight) to collect users' sent and received messages and emails. This includes actions such as using the Slack API to retrieve chat history and using the Gmail API to collect emails tagged with specific labels.

[0555] Step 3:

[0556] The server performs filtering on the collected data to remove noise (e.g., system messages and spam). Regular expressions and keyword filtering techniques are used for noise reduction. For example, messages with a specific format or pattern are excluded.

[0557] Step 4:

[0558] The server divides the noise-free data into sentences and converts them into a format suitable for natural language processing (NLP). This includes processes such as text tokenization and format conversion. For example, each sentence is organized as a separate entry in JSON format.

[0559] Step 5:

[0560] The server uses NLP models (e.g., sentiment analysis models) to classify the emotions (e.g., joy, anger, sadness) in each message or email. This allows the server to identify the user's stress levels and emotional state.

[0561] Step 6:

[0562] The server uses an emotion engine to analyze the user's emotions in real time. The emotion engine recognizes emotions in real time from the collected data and reflects the results. For example, if the user is feeling stressed, it quantifies that stress level.

[0563] Step 7:

[0564] Based on the analysis results, the server generates specific advice to provide to the user. This uses a template-based generation method, including advice tailored to each user's situation. For example, it might provide specific advice such as "Suggest to your boss that you reduce the number of meetings."

[0565] Step 8:

[0566] The on-device chat AI presents the user with generated advice. The generated advice messages are displayed in a natural conversational format through the chat window. For example, it might display a message like, "Are you feeling tired from all the meetings lately? Why not suggest to your boss that we reduce the number of meetings?"

[0567] Step 9:

[0568] If necessary, the terminal's chat AI will suggest to the user the use of an occupational physician consultation or in-house counseling services. If the user's stress level is high, it can also provide a dedicated booking link. For example, it may display a suggestion such as, "If you are feeling stressed, we recommend booking an appointment with an occupational physician."

[0569] Step 10:

[0570] Users enter feedback on the advice provided in the chat window. This feedback may include comments on the effectiveness of the advice and other opinions. For example, they might enter a comment such as, "I followed the advice and reduced the number of meetings, and I feel much better."

[0571] Step 11:

[0572] The server analyzes user feedback and incorporates the results into future advice generation. This allows the system to continuously optimize itself and provide support tailored to the user's situation. For example, it relearns keywords and emotions derived from feedback to improve accuracy.

[0573] (Example 2)

[0574] Next, we will describe Example 2. 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".

[0575] In today's business environment, users frequently utilize online communication tools, but managing the stress and negative emotions that arise from these interactions is a challenge. This often impairs users' work efficiency and mental health. Furthermore, current systems lack sufficient means to analyze users' emotional states in real time and provide appropriate advice. Moreover, collecting feedback based on the advice provided and incorporating it into future advice is difficult. To address these challenges, a system is needed that efficiently analyzes users' online communication data, automatically generates appropriate advice, and delivers it in real time.

[0576] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0577] In this invention, the server includes means for collecting data from online communication tools used by the user, means for removing noise from the collected data and dividing the text into sentences, means for analyzing the collected data using natural language processing to extract the user's stress and emotions, means for analyzing the user's emotional state in real time using an emotion engine, means for generating appropriate advice for the user based on the analysis results, means for inputting prompt sentences into a generation AI model to automatically generate advice, means for displaying the generated advice on the user's terminal and providing it to the user, and means for collecting user feedback and reflecting it in the generation of advice for the next time. This makes it possible to analyze the user's emotional state in real time, automatically generate appropriate advice, and provide it to the user. Furthermore, it is possible to continuously improve the advice based on feedback, so it is expected that the user's stress in online communication and mental health will be reduced.

[0578] An "online communication tool" is a software application that allows users to communicate with each other via the internet using text messages, voice, video, and other methods.

[0579] "Means of data collection" refers to software modules that automatically retrieve messages and other data from online communication tools used by users.

[0580] "Noise removal methods" refer to software processing that removes unnecessary or duplicate information from collected data to improve the accuracy of the analysis.

[0581] A "means for dividing text into sentences" refers to a software algorithm that divides collected messages or data into individual sentences and performs preprocessing for natural language processing.

[0582] Natural Language Processing (NLP) is a technology that enables computers to understand and generate natural language, performing tasks such as sentiment analysis and keyword extraction.

[0583] An "emotion engine" is a software module that analyzes a user's emotional state based on the results of natural language processing, and identifies stress levels and types of emotions.

[0584] The "means for generating advice" refer to a software module that automatically creates specific and appropriate advice for the user based on the analysis results of the emotion engine.

[0585] A "generative AI model" is an artificial intelligence model that generates natural language based on input prompt sentences, and is a technology used to provide appropriate advice to users.

[0586] A "prompt statement" is an instruction statement that is input into a generative AI model to generate advice, and it is a sentence that includes a specific output request to the model.

[0587] "User's device" refers to an electronic device such as a computer or smartphone used by the user, and is a device for displaying the generated advice.

[0588] A "means for collecting feedback" refers to a software module that collects opinions and impressions from users regarding advice, and analyzes and stores that data.

[0589] This invention is a system designed to reduce communication-related stress and facilitate smooth communication for users in the workplace or while teleworking. The system collects data from online communication tools, analyzes it using natural language processing, recognizes the user's emotions using an emotion engine, and provides advice—a comprehensive process.

[0590] First, the server accesses the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data (such as messages and emails). For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0591] Next, the server preprocesses the collected data. Specifically, it removes noise from the collected data and divides the text into sentences. This preprocessing transforms the data into a format suitable for analysis. For example, the server filters out duplicate messages and meaningless data and divides the text into individual sentences.

[0592] The server then analyzes the pre-processed data using natural language processing (NLP) techniques. NLP libraries such as SpaCy and NLTK are used. Sentiment analysis and keyword extraction are performed to identify the cause of stress from messages like, for example, "I'm exhausted from too many meetings." As a concrete example, prompts such as, "The user is feeling overworked. Please suggest ways to refresh," can be input into a generating AI model to create appropriate advice.

[0593] Furthermore, the server uses an emotion engine to analyze the user's emotional state in real time. The emotion engine identifies the user's stress level and type of emotion based on the results of natural language processing. For example, a message like "I'm exhausted from too many meetings" would be judged as "high stress."

[0594] Next, the server generates appropriate advice for the user based on the analysis results of the emotion engine. It automatically generates advice by inputting prompts into a generation AI model (e.g., GPT-3). Specific examples of advice include "suggest to your boss that you reduce the number of meetings" and "do 5 minutes of stretching to refresh yourself after a meeting."

[0595] The device then provides the user with the generated advice. Through a chat window on the device, the user can receive and implement the advice in a natural, conversational format. For example, it might suggest, "It seems you've been feeling stressed lately. Why not schedule a consultation with an occupational physician?"

[0596] Finally, the device collects user feedback and sends it to the server. A feedback form is displayed in the device's chat window, and when the user enters their opinions and thoughts on the advice provided, that feedback is sent to the server. The server analyzes the feedback and incorporates it into the generation of future advice. This allows the system to continuously provide optimal support tailored to the user's situation.

[0597] This entire process allows users to receive specific support to alleviate their stress and worries, thereby improving the quality of their relationships and communication in the workplace. Furthermore, the use of an emotion engine enables real-time responses to the user's emotional state, resulting in more effective support.

[0598] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0599] Step 1: Data Collection

[0600] The server calls the APIs of online communication tools used by users (e.g., Slack) to collect message data that is sent and received on a daily basis.

[0601] Input: User's API authentication information for their online communication tool.

[0602] Output: Message data sent and received by the user during that day.

[0603] Specific actions:

[0604] The server executes scheduled jobs every night at 12:00.

[0605] The server authenticates with Slack using API keys and authentication tokens.

[0606] The server calls the API to retrieve messages sent and received by the user.

[0607] The retrieved message data is saved to the database.

[0608] Step 2: Data Preprocessing

[0609] The server removes unnecessary information (noise) from the collected message data and divides the text into sentences.

[0610] Input: Collected message data.

[0611] Output: Preprocessed text data.

[0612] Specific actions:

[0613] The server filters out duplicate messages and meaningless data (e.g., system messages) from the data.

[0614] The server divides the text into sentences and converts them into a format that is easy to parse.

[0615] Step 3: Natural Language Processing (NLP) Analysis

[0616] The server uses the pre-processed data to perform sentiment analysis and keyword extraction.

[0617] Input: Preprocessed text data.

[0618] Output: Sentimental states and keywords extracted from text data.

[0619] Specific actions:

[0620] The server performs sentiment analysis using an NLP library (e.g., SpaCy, NLTK).

[0621] For example, messages containing emotions such as "tired" or "happy" are classified as positive, negative, or neutral.

[0622] The server extracts important keywords using technologies such as TF-IDF and Word2Vec.

[0623] Step 4: Emotional state analysis using the emotion engine

[0624] The server uses an emotion engine to analyze the user's emotional state in real time.

[0625] Input: NLP analysis results.

[0626] Output: Quantified emotional state (e.g., stress level).

[0627] Specific actions:

[0628] The emotion engine uses the results of emotion analysis to quantify the user's emotional state.

[0629] For example, the emotional engine interprets the message "I'm exhausted from too many meetings" as indicating high stress levels.

[0630] Step 5: Generate advice

[0631] Based on the analysis results of the emotion engine, the server generates specific advice for the user.

[0632] Input: Quantified result of emotional state.

[0633] Output: Generated advice.

[0634] Specific actions:

[0635] The server inputs prompt text into the generated AI model (e.g., GPT-3) and automatically generates advice.

[0636] Example prompt: "The user is feeling overworked. Please suggest ways for them to refresh themselves."

[0637] The server then provides the generated advice to the user in a personalized manner.

[0638] Step 6: Providing advice

[0639] The device notifies the user via chat AI to provide advice.

[0640] Input: Generated advice.

[0641] Output: Advice provided to the user.

[0642] Specific actions:

[0643] Advice will be displayed in the chat window on the device.

[0644] Examples: "Try suggesting we reduce the number of meetings," or "Let's do some 5-minute stretches to refresh ourselves."

[0645] The chat AI provides advice through a conversational format with the user.

[0646] Step 7: Gathering Feedback

[0647] The device collects user feedback and sends it to the server.

[0648] Input: User feedback.

[0649] Output: Feedback data sent to the server.

[0650] Specific actions:

[0651] The device displays a feedback form within the chat window.

[0652] Examples: "Was this advice helpful?" "Do you have any other comments?"

[0653] The user enters their feedback and presses the submit button.

[0654] Step 8: Analyze and implement feedback

[0655] The server analyzes the collected feedback and incorporates it into the next advice generation process.

[0656] Input: Collected feedback data.

[0657] Output: Improved advice generation process.

[0658] Specific actions:

[0659] The server analyzes the feedback data and saves the obtained information as training data.

[0660] Based on the feedback, the server improves the prompts and advice it inputs into the generated AI model.

[0661] This entire process allows users to receive appropriate support to alleviate their stress and worries, and to improve the quality of their relationships and communication in the workplace.

[0662] (Application Example 2)

[0663] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0664] In modern workplaces and brick-and-mortar stores, employees often experience stress from communicating with customers and colleagues. This leads to problems such as decreased employee performance and customer satisfaction. Furthermore, there is a growing need to focus on stress management and implement effective support systems.

[0665] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting data from a communication tool used by the user, means for analyzing the collected data using natural language processing to extract the tension and emotions the user is experiencing, means for generating appropriate advice for the user based on the analysis results, means for displaying the generated advice on the user's display device and providing it to the user, means for collecting the user's reactions and reflecting them in the generation of advice for the next time, and means for detecting the user's state of tension in real time during the conversation and generating and providing appropriate suggestions. As a result, the user can receive concrete support to reduce stress caused by communication and enable smooth conversations, making it possible to improve interpersonal relationships and customer satisfaction in the workplace and physical stores.

[0666] A "communication tool" refers to an online platform or application that users use to exchange messages and information with other people.

[0667] "Natural language processing" refers to the process by which computers understand and analyze natural language text. Specifically, it includes techniques such as grammatical analysis, sentiment analysis, and keyword extraction.

[0668] "Tension" refers to the stress and anxiety that users experience during communication. This often involves psychological burden.

[0669] "Emotions" refer to the user's psychological state, including emotional states such as joy, anger, sadness, and surprise.

[0670] "Advice" refers to specific suggestions or advice that help users act appropriately.

[0671] A "display device" is a device that a user uses to visually confirm information. This includes smart glasses and smartphones.

[0672] "Response" refers to the feedback or actions a user takes in response to the advice they receive.

[0673] "Real-time" refers to processing or responding immediately without delay.

[0674] A "suggestion" refers to a suggestion of actions or measures provided based on the user's situation and emotional state.

[0675] "Support" refers to assistance and services provided to users to resolve stress and communication problems.

[0676] Embodiments of this invention include systems utilizing communication tools, natural language processing, sentiment analysis engines, display devices, and feedback collection means. Their specific configuration and operation are described below.

[0677] The system includes means for collecting data from communication tools that users use on a daily basis. For example, the server uses the API of the communication tool to retrieve messages sent and received by users at regular intervals. Since the collected data is not suitable for analysis in its raw form, the server analyzes this data using natural language processing (NLP) techniques.

[0678] During the analysis phase, text data is tokenized, and processes such as sentiment analysis and keyword extraction are performed. This allows for the identification of the user's tension and emotions. For example, if a user sends a message such as "I'm exhausted from having to deal with too many customers," this statement is analyzed as an indication of stress.

[0679] Next, based on the analysis results, the server uses an emotion analysis engine to evaluate the user's emotional state. It then generates appropriate advice corresponding to the results. For example, if the server determines that the user is feeling stressed, it will generate specific advice such as "take a short break" or "talk to your supervisor."

[0680] The generated advice is displayed in real time through the user's display device (e.g., smart glasses or smartphone). This allows the user to receive advice quickly during the conversation. It also has a function to detect the user's real-time level of tension and provide appropriate suggestions immediately.

[0681] Next, the system includes a means for users to provide feedback on the advice they receive. Users input their opinions and thoughts on the advice, and this feedback is sent to the server. The server analyzes the feedback and incorporates it into the next advice generation. This allows the system to continuously provide optimal support tailored to the user's situation.

[0682] The specific hardware and software used include:

[0683] Smart glasses: A device worn by store employees to display information in real time.

[0684] SmartGlassesAPI: An API that collects data from smart glasses and communicates with them.

[0685] TextAnalyzer (NLP Module): A module for natural language processing.

[0686] EmotionRecognitionEngine: Emotion Analysis Engine

[0687] ChatAI: A chat AI for generating and providing advice

[0688] As a concrete example, suppose a store employee is serving a customer and the customer asks for a special discount. In this case, the system detects in real time that the employee is feeling stressed and offers advice such as, "Relax and try offering the best possible discount." This reduces employee stress and improves customer satisfaction.

[0689] Examples of prompt statements are as follows:

[0690] "We will provide conversation data between store staff and customers. If the staff member is experiencing stress, please generate advice on how to handle the situation: {conversation data}"

[0691] This will enable employees to communicate more smoothly and improve the quality of customer service in the workplace and at physical stores.

[0692] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0693] Step 1:

[0694] The server collects data through the APIs of the communication tools used by the user. Specifically, it periodically calls the APIs to retrieve messages and chat logs. This input data consists of messages sent and received by the user. In this case, it also includes obtaining real-time conversation data using, for example, the data collection function of smart glasses.

[0695] Step 2:

[0696] The server stores the collected data as text data. Next, it processes the collected text data using natural language processing (NLP) techniques. The input data is the collected text data, and the output data is the analyzed text structure and extracted keywords. Specifically, it performs text tokenization, grammatical analysis, keyword extraction, and sentiment analysis.

[0697] Step 3:

[0698] The server uses a sentiment analysis engine to evaluate the user's emotional state based on the analysis results obtained through natural language processing. The input data consists of the analyzed text structure and keywords, while the output data is the user's emotional state (e.g., stress, joy, anger). Here, an emotion detection algorithm is used to identify emotions within the text and quantify the emotional state.

[0699] Step 4:

[0700] Based on the results of the emotion analysis engine, the server generates appropriate advice. The input data is the user's emotional state, and the output data is specific advice and suggestions. Using a generative AI model, it generates advice that is appropriate to the user's current situation. For example, if the user is feeling stressed, it will output specific advice such as "take a short break."

[0701] Step 5:

[0702] The generated advice is transmitted in real time to the user's display device, such as smart glasses or a smartphone. The input data is the generated advice, and the output data is the advice message displayed to the user. The smart glasses' display function is used to provide the user with advice visually in real time.

[0703] Step 6:

[0704] When a user enters feedback on advice, that feedback data is sent to the server. The input data is the user's feedback, and the output data is the feedback information stored on the server. The user enters the feedback through smart glasses.

[0705] Step 7:

[0706] The server analyzes the collected feedback data and incorporates it into the next advice generation. The input data is feedback information, and the output data is the improved advice generation algorithm. Specifically, a machine learning algorithm is used to analyze the feedback data and tune the AI ​​model for generation.

[0707] The above outlines the specific processing steps required to implement this system. This process allows users to receive support in real time, reducing communication stress.

[0708] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0709] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0710] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0711] [Third Embodiment]

[0712] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0713] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0714] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0715] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0716] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0717] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0718] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0719] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0720] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0721] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0722] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0723] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0724] This invention is a system designed to reduce communication stress in the workplace and during teleworking, and to facilitate smooth communication. The system collects data from online communication tools used by the user, analyzes it using natural language processing, and provides advice.

[0725] First, the server connects to the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data, such as messages and emails. For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0726] Next, the server analyzes the collected data using natural language processing (NLP) techniques. The server removes noise from the obtained data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, to identify the cause of stress from a message such as "I'm exhausted from too many meetings." In this case, the server determines that excessive meetings are the main cause of the user's stress.

[0727] Furthermore, the server generates specific advice for the user based on the analysis results. For example, it might suggest to your boss to reduce the number of meetings or to do five minutes of stretching after a meeting to refresh yourself.

[0728] Subsequently, a chat AI running on the device provides the user with generated advice. The user can receive and implement the advice through the chat window on the device. The chat AI also suggests, if necessary, the use of an occupational physician consultation or in-house counseling services. For example, it might suggest, "It seems you've been feeling stressed a lot lately. Why don't you try scheduling an appointment with an occupational physician?"

[0729] Finally, the process includes collecting user feedback. When users input their opinions and thoughts on the advice provided on their device, this feedback is sent to the server. The server analyzes the feedback and incorporates it into future advice generation. This continuous process allows the system to consistently provide optimal support tailored to the user's situation.

[0730] This allows users to receive specific support to alleviate their stress and worries, and to improve the quality of their relationships and communication in the workplace.

[0731] The following describes the processing flow.

[0732] Step 1:

[0733] The server obtains the necessary permissions to access the online communication tools used by the user (e.g., Slack, Gmail). This allows the server to collect information such as the user's messages and emails.

[0734] Step 2:

[0735] The server calls the APIs of online communication tools at a specified schedule (for example, every night at midnight) to collect user-sent and received messages and emails. This includes using the APIs of each tool to retrieve data.

[0736] Step 3:

[0737] The server performs filtering on the collected data to remove noise (e.g., system messages and spam emails). Regular expressions and keyword filtering techniques are used for noise reduction.

[0738] Step 4:

[0739] The server divides the noise-free data into sentences and converts them into a format suitable for natural language processing (NLP). This includes text tokenization and format conversion.

[0740] Step 5:

[0741] The server uses NLP models (e.g., sentiment analysis models) to classify the emotions (e.g., joy, anger, sadness) in each message or email. This helps identify the user's stress levels and emotional state.

[0742] Step 6:

[0743] Based on the sentiment analysis results, the server extracts keywords and phrases that are causing stress. For example, from the message "too many meetings," it identifies "excessive meetings" as a stressor.

[0744] Step 7:

[0745] Based on the analysis results, the server generates specific advice to provide to the user. This advice is template-based and includes content tailored to the individual user's situation.

[0746] Step 8:

[0747] The on-device chat AI presents generated advice to the user. The advice messages are displayed in a natural, conversational format through the chat window, making them easy for the user to understand.

[0748] Step 9:

[0749] If necessary, the terminal's chat AI will suggest to the user the use of occupational health physician consultations or in-house counseling services. For example, if the user's stress level is high, it will provide a dedicated booking link.

[0750] Step 10:

[0751] Users enter feedback on the advice provided in the chat window. This feedback includes opinions on the effectiveness of the advice and other comments.

[0752] Step 11:

[0753] The server analyzes user feedback and incorporates the results into future advice generation. This allows the system to continuously optimize itself and provide support tailored to the user's situation.

[0754] (Example 1)

[0755] Next, we will describe Example 1. 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."

[0756] In traditional online communication environments, there is a lack of concrete support to effectively reduce the stress users experience in the workplace or remotely and to facilitate smooth communication. Furthermore, there is a lack of means to identify the causes of stress and provide appropriate advice, making it difficult for users to find ways to cope with stress. Additionally, mechanisms for continuously improving support through feedback are insufficient.

[0757] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0758] In this invention, the server includes means for collecting data from online communication tools used by the user; means for removing irrelevant information from the collected data and dividing the text into sentences; means for performing sentiment analysis and keyword extraction from the divided text to analyze the user's sources of stress and emotions; means for creating prompt sentences using a generation AI model based on the analysis results and generating specific advice for the user; means for displaying the generated advice on the user's terminal and providing it to the user; and means for collecting user feedback and reflecting it in the generation of advice for the next time. This makes it possible to reduce the user's communication-related stress, provide appropriate advice, and continuously improve the support content based on the feedback.

[0759] An "online communication tool" is software or an application that allows users to communicate in real time using text, voice, or video over the internet.

[0760] "Means of data collection" refers to programs and functions that obtain necessary information and messages from users' online communication tools.

[0761] "Means of removing irrelevant information" refer to algorithms and filtering functions used to remove unwanted elements and noise from collected data.

[0762] "Methods for dividing text into sentences" refer to natural language processing techniques for dividing collected text data into smaller sentence units.

[0763] "Sentiment analysis" is an algorithm used to identify emotions such as positive, negative, and neutral from text data.

[0764] Keyword extraction is a natural language processing technique used to extract important words and phrases from text data.

[0765] A "source of stress" refers to an event or factor that causes stress for the user.

[0766] A "generative AI model" is a model that uses machine learning and artificial intelligence technologies to learn patterns from data and generate new information and suggestions.

[0767] A "prompt message" is an instruction given to an AI model to generate specific information or answers.

[0768] "Means for generating advice" refers to functions that create useful suggestions and instructions for users based on analysis results and AI models.

[0769] "Means of collecting feedback" refers to functions for receiving opinions and feedback from users, saving them as data, and analyzing them.

[0770] "Improving support services" means updating and adjusting the advice and support provided by the system based on collected feedback data to make them more effective.

[0771] This invention is a system designed to reduce the stress users experience from communication in the workplace or remotely, and to facilitate smooth communication. The system collects and analyzes data through multiple steps and provides appropriate advice.

[0772] First, the server connects to the online communication tool used by the user. Specifically, it connects using the Slack API and performs authentication. By obtaining the necessary permissions, the server collects messages and related information sent and received by the user. This data collection is performed automatically every night at midnight, and the collected data is stored in a database.

[0773] Next, the server analyzes the collected data using natural language processing (NLP) techniques. Specifically, it first removes noise, eliminating unnecessary information, emojis, URLs, etc. Then, it uses a natural language processing library (e.g., NLTK) to divide the cleaned-up text into sentences. Each of these resulting sentences is then subjected to sentiment analysis and keyword extraction.

[0774] The server uses a sentiment analysis algorithm to evaluate the sentiment of each message. For example, in a message like "I'm tired of too many meetings," "tired" is evaluated as a negative emotion. Furthermore, a keyword extraction algorithm is used to identify important words and phrases (e.g., "meetings" and "tired").

[0775] Based on the analysis results, the server uses a generative AI model to create prompt statements and generate specific advice for the user. These prompt statements are used as input to the generative AI model. As a specific example, the following prompt statement is input to the generative AI model:

[0776] What advice would you provide to help users reduce stress at work? For example, generate advice for a situation where a user says, "I'm exhausted because there are too many meetings."

[0777] This prompt allows the AI ​​model to generate appropriate advice regarding specific stressors.

[0778] Next, the generated advice is sent from the server to the user's device. It is displayed in the chat window on the device, where the user can review and implement the advice. For example, the advice might say, "Try suggesting to your boss that you reduce the number of meetings." The chat AI also suggests medical consultations or the use of internal support services as needed. For example, it might suggest, "You seem to be feeling stressed lately. Why don't you try scheduling a medical consultation?"

[0779] Finally, the process includes collecting user feedback. Through a feedback window on the device, users enter their opinions and impressions of the advice provided. This feedback is sent to the server and incorporated into future advice generation. This allows for continuous improvement of the support provided, ensuring that users receive the best possible support.

[0780] In this way, this system not only reduces the user's communication-related stress and provides appropriate advice, but also enables continuous improvement of support based on feedback.

[0781] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0782] Step 1: Initial setup and data collection

[0783] Subject: Server

[0784] The server connects to the online communication tool used by the user. For example, it uses a Slack API key for authentication. The server obtains the necessary permissions (such as reading messages and retrieving user information). Every day at midnight, it calls the API to collect messages sent and received by the user during that day. The API key and authentication information are used as input, and the message data is stored in a database as output.

[0785] Step 2: Data preprocessing

[0786] Subject: Server

[0787] The server removes unwanted noise from the collected data. Specifically, it removes emojis, URLs, unnecessary spaces, etc. It uses the collected message data as input and outputs cleaned-up text data.

[0788] Step 3: Split the text into sentence units.

[0789] Subject: Server

[0790] The server splits the cleaned text into sentences. It uses a natural language processing library (e.g., NLTK) to split each message into sentences. The input is the cleaned text data, and the output is a list of the split sentences.

[0791] Step 4: Sentiment analysis and keyword extraction

[0792] Subject: Server

[0793] The server performs sentiment analysis on segmented sentences, evaluating the tone and sentiment of the message. It also uses a keyword extraction algorithm (e.g., TF-IDF) to identify important keywords. The input is a list of segmented sentences, and the output is the sentiment evaluation result and a keyword list. For example, from the message "I'm tired because there are too many meetings," "tired" is evaluated as a negative emotion, and "meetings" and "tired" are extracted as important keywords.

[0794] Step 5: Generating Advice

[0795] Subject: Server

[0796] The server uses a generative AI model to create prompt sentences based on the results of sentiment analysis and keyword extraction, generating specific advice for the user. The input consists of sentiment evaluation results and a keyword list, and the output is the generated advice. A concrete example of a prompt sentence used is: "What advice would you provide to reduce the stress a user feels at work? For example, generate advice for the situation, 'I'm exhausted because there are too many meetings.'"

[0797] Step 6: Delivering Advice

[0798] Subject: terminal

[0799] The terminal displays advice received from the server in a chat window. Users can review and implement the advice through this window. The input is advice generated from the server, and the output is the advice displayed in the chat window. For example, advice such as "Suggest reducing the number of meetings to your boss" might be displayed.

[0800] Step 7: Present additional proposals

[0801] Subject: Chat AI

[0802] The chat AI suggests medical consultations and internal support services based on the user's situation. For example, it might suggest, "It seems you've been feeling stressed lately. Why not schedule a medical consultation?" The AI ​​uses the user's emotional analysis results and past advice history as input, and provides additional suggestions as output.

[0803] Step 8: Gathering and analyzing feedback

[0804] Subject: User and Server

[0805] Users input their opinions and feedback on the advice provided through a feedback window on their device. The server receives this feedback and analyzes it. The input is the user's feedback data, and the output is the analysis results of the feedback. These analysis results are reflected in the next advice generation. This allows for continuous improvement of the support provided.

[0806] (Application Example 1)

[0807] Next, we will explain Application Example 1. In the following explanation, 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."

[0808] In modern workplaces and teleworking environments, employees are prone to experiencing stress related to their work and communication. In particular, employees in physical stores often experience significant stress when dealing with customers and processing business messages. However, it is difficult for employees themselves to accurately identify their stress levels and deal with them appropriately. This invention aims to provide concrete support methods to help employees alleviate their stress and ensure smooth work operations.

[0809] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0810] In this invention, the server includes means for collecting data from online communication tools used by the user, means for analyzing the collected data using natural language processing to extract the user's stress and emotions, means for generating appropriate advice for the user based on the analysis results, means for displaying the generated advice on the user's terminal and providing it to the user, means for collecting user feedback and reflecting it in the generation of future advice, means for collecting work message data from employees in physical stores and providing advice based on it, means for analyzing the collected work message data using natural language processing to identify the causes of employee stress, and means for analyzing the collected feedback and reflecting it in the generation of future advice. As a result, employees can reduce stress in their daily work and achieve smoother communication.

[0811] "Online communication tools" are software and services that allow users to exchange messages, make calls, and conduct video conferences over the internet.

[0812] "Natural language processing" is a general term for technologies that enable computers to understand, analyze, and generate human language (natural language).

[0813] "Sentiment analysis" is a technique that analyzes text data to determine whether its content is positive, negative, or neutral.

[0814] "Advice" refers to suggestions or suggestions that recommend appropriate measures or actions in response to a specific problem or situation.

[0815] A "device" is a computing device that a user directly operates (for example, a smartphone, tablet, or PC).

[0816] "Feedback" refers to the act or content of a user's response to advice or services provided, such as an opinion or comment.

[0817] A "physical store" is a commercial facility or business location located in a physical space where customer service and sales activities take place.

[0818] "Business message data" refers to the content of messages generated by users during their work, and includes data such as order processing, customer service records, and internal communications.

[0819] "Causes of stress" refer to the psychological and physical burdens and pressures that users experience while performing their work.

[0820] "Advice generation" is the process of creating advice based on analysis results to suggest appropriate actions and measures for the user.

[0821] This invention is a system designed to reduce communication stress in the workplace and during teleworking, and to enable smoother communication.

[0822] This system includes the following means:

[0823] 1. A means of collecting data from online communication tools used by users.

[0824] 2. A method for analyzing collected data using natural language processing to extract the stress and emotions experienced by users.

[0825] 3. A means of generating appropriate advice for the user based on the analysis results.

[0826] 4. A means of displaying the generated advice on the user's device and providing it to the user.

[0827] 5. A means of collecting user feedback and incorporating it into future advice generation.

[0828] 6. A means by which employees at physical stores collect business message data and provide advice based on that data.

[0829] 7. A method for identifying the causes of employee stress by analyzing collected business message data using natural language processing.

[0830] 8. A means of analyzing the collected feedback and incorporating it into the generation of future advice.

[0831] The server first connects to the user's online communication tool, obtains the necessary permissions, and collects data. For example, the server calls an API every night at midnight to retrieve messages sent and received by the user during that day.

[0832] Next, the server analyzes the collected data using natural language processing (NLP) techniques. It removes noise from the collected business message data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, to identify the cause of stress from a message such as "I'm exhausted from too many meetings."

[0833] Therefore, we use the Vader SentimentIntensityAnalyzer from the Natural Language Toolkit (NLTK). The server identifies negative emotions from the data obtained and generates advice based on them. For example, it might generate advice such as "Remember to take a break" or "Let's lower the priority of this task and do it another day."

[0834] The generated advice is provided through a chat window on the user's device. The user can receive and implement this advice. If necessary, the chat AI can also suggest to the user the use of an industrial physician consultation or in-house counseling services.

[0835] When a user submits feedback, it is sent to the server. The server analyzes this feedback and incorporates it into the generation of future advice. This continuous process allows the system to consistently provide optimal support tailored to the user's situation.

[0836] As a concrete example, consider a scenario where employee A at a physical store is replenishing sandwiches that have run out of fillings after lunchtime while also assisting customers. In this situation, work message data is collected and sentiment analysis is performed to identify that employee A is experiencing stress. In response to this situation, advice such as "Remember to take a break during the busy post-lunchtime period" is generated and provided to employee A's terminal.

[0837] Examples of prompts for a generative AI model:

[0838] "Employees at physical stores sometimes experience stress in their daily work. For example, they might feel stressed when they have to adjust sandwiches that have run out of fillings after lunchtime while also dealing with customers. What kind of advice would you offer them?"

[0839] This allows employees in physical stores to use smart assistants for stress management, thereby improving the quality and efficiency of their work.

[0840] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0841] Step 1:

[0842] The server connects to the online communication tool used by the user and obtains the necessary permissions.

[0843] Input: User's online communication tool authentication information.

[0844] Specific operation: The server uses the API to obtain an authentication token and gain permission to access the data.

[0845] Output: User's access permissions to message data.

[0846] Step 2:

[0847] The server collects messages sent and received by users.

[0848] Input: User's online communication tool credentials and API key.

[0849] Specific operation: The server retrieves messages sent and received within a specified time frame (e.g., 1 day) via the API.

[0850] Output: Acquired message data.

[0851] Step 3:

[0852] The server analyzes the collected message data using natural language processing (NLP) techniques.

[0853] Input: Collected message data.

[0854] Specific operation: The server uses the NLTK library to split messages into sentences and perform sentiment analysis.

[0855] Output: Sentiment score (positive, negative, neutral) for each message.

[0856] Step 4:

[0857] The server generates appropriate advice for the user based on the analysis results.

[0858] Input: Sentiment score.

[0859] Specific operation: The server uses a generative AI model to generate appropriate advice corresponding to the sentiment score.

[0860] Output: Generated advice.

[0861] Step 5:

[0862] The generated advice is displayed on the user's device and provided to the user.

[0863] Input: Generated advice.

[0864] Specific actions: Display advice in the chat window on the device. Send notifications as needed.

[0865] Output: Advice displayed to the user.

[0866] Step 6:

[0867] Users follow the advice and input the results and feedback.

[0868] Input: User feedback.

[0869] Specific operation: The user provides feedback through an input form on the device.

[0870] Output: Provided feedback data.

[0871] Step 7:

[0872] The server analyzes the collected feedback and incorporates it into the generation of advice for the next time.

[0873] Input: User feedback data.

[0874] Specific operation: The server analyzes the feedback data and identifies areas for improvement needed to generate advice for the next time.

[0875] Output: Improved advice generation algorithm.

[0876] Step 8:

[0877] The server collects business message data from employees at physical stores and provides advice based on that data.

[0878] Input: Business message data collected by employees at physical stores.

[0879] Specific operation: The server analyzes the business message data from physical stores using natural language processing technology to identify the causes of employee stress.

[0880] Output: Advice to provide to employees.

[0881] Step 9:

[0882] The collected feedback will be analyzed and used to improve the generation of advice for the next time.

[0883] Input: Feedback data collected from employees and users.

[0884] Specific operation: The server analyzes the feedback data to gain insights that will improve the quality of future advice.

[0885] Output: Improved version of the advice that will be generated next time.

[0886] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0887] This invention is a system designed to reduce communication-related stress and facilitate smooth communication for users in the workplace or while teleworking. The system collects data from online communication tools, analyzes it using natural language processing, recognizes the user's emotions using an emotion engine, and provides advice—a comprehensive process.

[0888] First, the server accesses the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data (such as messages and emails). For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0889] Next, the server analyzes the collected data using natural language processing (NLP) techniques. The server removes noise from the obtained data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, identifying the cause of stress from a message such as "I'm tired because there are too many meetings." Based on this, the sentiment engine analyzes the user's emotions in real time and identifies specific emotional states (e.g., stress, joy, anger).

[0890] Furthermore, the server generates advice for the user based on the analysis results of the emotion engine. For example, it might create specific advice such as "suggest to your boss that you reduce the number of meetings" or "do 5 minutes of stretching to refresh yourself after a meeting." The emotion engine can adjust the content and timing of the advice according to the user's emotional state.

[0891] Subsequently, a chat AI running on the device provides the user with generated advice. The user can receive and implement the advice in a natural conversational format through the chat window on the device. The chat AI also suggests, if necessary, the use of an occupational physician consultation or in-house counseling services. For example, it might suggest, "It seems you've been feeling stressed a lot lately. Why don't you try scheduling an appointment with an occupational physician?"

[0892] Finally, the process includes collecting user feedback. When users input their opinions and thoughts on the advice provided on their device, this feedback is sent to the server. The server analyzes the feedback and incorporates it into future advice generation. This allows the system to continuously provide optimal support tailored to the user's situation.

[0893] This allows users to receive specific support to alleviate their stress and worries, improving the quality of workplace relationships and communication. Furthermore, the use of an emotion engine enables real-time responses to the user's emotional state, resulting in more effective support.

[0894] The following describes the processing flow.

[0895] Step 1:

[0896] The server obtains the necessary permissions to access the online communication tools (e.g., Slack, Gmail) that the user uses. For example, by granting access permissions to a tool, the server can retrieve data through the APIs of those tools.

[0897] Step 2:

[0898] The server calls APIs of online communication tools at a specified schedule (for example, every night at midnight) to collect users' sent and received messages and emails. This includes actions such as using the Slack API to retrieve chat history and using the Gmail API to collect emails tagged with specific labels.

[0899] Step 3:

[0900] The server performs filtering on the collected data to remove noise (e.g., system messages and spam). Regular expressions and keyword filtering techniques are used for noise reduction. For example, messages with a specific format or pattern are excluded.

[0901] Step 4:

[0902] The server divides the noise-free data into sentences and converts them into a format suitable for natural language processing (NLP). This includes processes such as text tokenization and format conversion. For example, each sentence is organized as a separate entry in JSON format.

[0903] Step 5:

[0904] The server uses NLP models (e.g., sentiment analysis models) to classify the emotions (e.g., joy, anger, sadness) in each message or email. This allows the server to identify the user's stress levels and emotional state.

[0905] Step 6:

[0906] The server uses an emotion engine to analyze the user's emotions in real time. The emotion engine recognizes emotions in real time from the collected data and reflects the results. For example, if the user is feeling stressed, it quantifies that stress level.

[0907] Step 7:

[0908] Based on the analysis results, the server generates specific advice to provide to the user. This uses a template-based generation method, including advice tailored to each user's situation. For example, it might provide specific advice such as "Suggest to your boss that you reduce the number of meetings."

[0909] Step 8:

[0910] The on-device chat AI presents the user with generated advice. The generated advice messages are displayed in a natural conversational format through the chat window. For example, it might display a message like, "Are you feeling tired from all the meetings lately? Why not suggest to your boss that we reduce the number of meetings?"

[0911] Step 9:

[0912] If necessary, the terminal's chat AI will suggest to the user the use of an occupational physician consultation or in-house counseling services. If the user's stress level is high, it can also provide a dedicated booking link. For example, it may display a suggestion such as, "If you are feeling stressed, we recommend booking an appointment with an occupational physician."

[0913] Step 10:

[0914] Users enter feedback on the advice provided in the chat window. This feedback may include comments on the effectiveness of the advice and other opinions. For example, they might enter a comment such as, "I followed the advice and reduced the number of meetings, and I feel much better."

[0915] Step 11:

[0916] The server analyzes user feedback and incorporates the results into future advice generation. This allows the system to continuously optimize itself and provide support tailored to the user's situation. For example, it relearns keywords and emotions derived from feedback to improve accuracy.

[0917] (Example 2)

[0918] Next, we will describe Example 2. 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."

[0919] In today's business environment, users frequently utilize online communication tools, but managing the stress and negative emotions that arise from these interactions is a challenge. This often impairs users' work efficiency and mental health. Furthermore, current systems lack sufficient means to analyze users' emotional states in real time and provide appropriate advice. Moreover, collecting feedback based on the advice provided and incorporating it into future advice is difficult. To address these challenges, a system is needed that efficiently analyzes users' online communication data, automatically generates appropriate advice, and delivers it in real time.

[0920] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0921] In this invention, the server includes means for collecting data from online communication tools used by the user, means for removing noise from the collected data and dividing the text into sentences, means for analyzing the collected data using natural language processing to extract the user's stress and emotions, means for analyzing the user's emotional state in real time using an emotion engine, means for generating appropriate advice for the user based on the analysis results, means for inputting prompt sentences into a generation AI model to automatically generate advice, means for displaying the generated advice on the user's terminal and providing it to the user, and means for collecting user feedback and reflecting it in the generation of advice for the next time. This makes it possible to analyze the user's emotional state in real time, automatically generate appropriate advice, and provide it to the user. Furthermore, it is possible to continuously improve the advice based on feedback, so it is expected that the user's stress in online communication and mental health will be reduced.

[0922] An "online communication tool" is a software application that allows users to communicate with each other via the internet using text messages, voice, video, and other methods.

[0923] "Means of data collection" refers to software modules that automatically retrieve messages and other data from online communication tools used by users.

[0924] "Noise removal methods" refer to software processing that removes unnecessary or duplicate information from collected data to improve the accuracy of the analysis.

[0925] A "means for dividing text into sentences" refers to a software algorithm that divides collected messages or data into individual sentences and performs preprocessing for natural language processing.

[0926] Natural Language Processing (NLP) is a technology that enables computers to understand and generate natural language, performing tasks such as sentiment analysis and keyword extraction.

[0927] An "emotion engine" is a software module that analyzes a user's emotional state based on the results of natural language processing, and identifies stress levels and types of emotions.

[0928] The "means for generating advice" refer to a software module that automatically creates specific and appropriate advice for the user based on the analysis results of the emotion engine.

[0929] A "generative AI model" is an artificial intelligence model that generates natural language based on input prompt sentences, and is a technology used to provide appropriate advice to users.

[0930] A "prompt statement" is an instruction statement that is input into a generative AI model to generate advice, and it is a sentence that includes a specific output request to the model.

[0931] "User's device" refers to an electronic device such as a computer or smartphone used by the user, and is a device for displaying the generated advice.

[0932] A "means for collecting feedback" refers to a software module that collects opinions and impressions from users regarding advice, and analyzes and stores that data.

[0933] This invention is a system designed to reduce communication-related stress and facilitate smooth communication for users in the workplace or while teleworking. The system collects data from online communication tools, analyzes it using natural language processing, recognizes the user's emotions using an emotion engine, and provides advice—a comprehensive process.

[0934] First, the server accesses the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data (such as messages and emails). For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[0935] Next, the server preprocesses the collected data. Specifically, it removes noise from the collected data and divides the text into sentences. This preprocessing transforms the data into a format suitable for analysis. For example, the server filters out duplicate messages and meaningless data and divides the text into individual sentences.

[0936] The server then analyzes the pre-processed data using natural language processing (NLP) techniques. NLP libraries such as SpaCy and NLTK are used. Sentiment analysis and keyword extraction are performed to identify the cause of stress from messages like, for example, "I'm exhausted from too many meetings." As a concrete example, prompts such as, "The user is feeling overworked. Please suggest ways to refresh," can be input into a generating AI model to create appropriate advice.

[0937] Furthermore, the server uses an emotion engine to analyze the user's emotional state in real time. The emotion engine identifies the user's stress level and type of emotion based on the results of natural language processing. For example, a message like "I'm exhausted from too many meetings" would be judged as "high stress."

[0938] Next, the server generates appropriate advice for the user based on the analysis results of the emotion engine. It automatically generates advice by inputting prompts into a generation AI model (e.g., GPT-3). Specific examples of advice include "suggest to your boss that you reduce the number of meetings" and "do 5 minutes of stretching to refresh yourself after a meeting."

[0939] The device then provides the user with the generated advice. Through a chat window on the device, the user can receive and implement the advice in a natural, conversational format. For example, it might suggest, "It seems you've been feeling stressed lately. Why not schedule a consultation with an occupational physician?"

[0940] Finally, the device collects user feedback and sends it to the server. A feedback form is displayed in the device's chat window, and when the user enters their opinions and thoughts on the advice provided, that feedback is sent to the server. The server analyzes the feedback and incorporates it into the generation of future advice. This allows the system to continuously provide optimal support tailored to the user's situation.

[0941] This entire process allows users to receive specific support to alleviate their stress and worries, thereby improving the quality of their relationships and communication in the workplace. Furthermore, the use of an emotion engine enables real-time responses to the user's emotional state, resulting in more effective support.

[0942] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0943] Step 1: Data Collection

[0944] The server calls the APIs of online communication tools used by users (e.g., Slack) to collect message data that is sent and received on a daily basis.

[0945] Input: User's API authentication information for their online communication tool.

[0946] Output: Message data sent and received by the user during that day.

[0947] Specific actions:

[0948] The server executes scheduled jobs every night at 12:00.

[0949] The server authenticates with Slack using API keys and authentication tokens.

[0950] The server calls the API to retrieve messages sent and received by the user.

[0951] The retrieved message data is saved to the database.

[0952] Step 2: Data Preprocessing

[0953] The server removes unnecessary information (noise) from the collected message data and divides the text into sentences.

[0954] Input: Collected message data.

[0955] Output: Preprocessed text data.

[0956] Specific actions:

[0957] The server filters out duplicate messages and meaningless data (e.g., system messages) from the data.

[0958] The server divides the text into sentences and converts them into a format that is easy to parse.

[0959] Step 3: Natural Language Processing (NLP) Analysis

[0960] The server uses the pre-processed data to perform sentiment analysis and keyword extraction.

[0961] Input: Preprocessed text data.

[0962] Output: Sentimental states and keywords extracted from text data.

[0963] Specific actions:

[0964] The server performs sentiment analysis using an NLP library (e.g., SpaCy, NLTK).

[0965] For example, messages containing emotions such as "tired" or "happy" are classified as positive, negative, or neutral.

[0966] The server extracts important keywords using technologies such as TF-IDF and Word2Vec.

[0967] Step 4: Emotional state analysis using the emotion engine

[0968] The server uses an emotion engine to analyze the user's emotional state in real time.

[0969] Input: NLP analysis results.

[0970] Output: Quantified emotional state (e.g., stress level).

[0971] Specific actions:

[0972] The emotion engine uses the results of emotion analysis to quantify the user's emotional state.

[0973] For example, the emotional engine interprets the message "I'm exhausted from too many meetings" as indicating high stress levels.

[0974] Step 5: Generate advice

[0975] Based on the analysis results of the emotion engine, the server generates specific advice for the user.

[0976] Input: Quantified result of emotional state.

[0977] Output: Generated advice.

[0978] Specific actions:

[0979] The server inputs prompt text into the generated AI model (e.g., GPT-3) and automatically generates advice.

[0980] Example prompt: "The user is feeling overworked. Please suggest ways for them to refresh themselves."

[0981] The server then provides the generated advice to the user in a personalized manner.

[0982] Step 6: Providing advice

[0983] The device notifies the user via chat AI to provide advice.

[0984] Input: Generated advice.

[0985] Output: Advice provided to the user.

[0986] Specific actions:

[0987] Advice will be displayed in the chat window on the device.

[0988] Examples: "Try suggesting we reduce the number of meetings," or "Let's do some 5-minute stretches to refresh ourselves."

[0989] The chat AI provides advice through a conversational format with the user.

[0990] Step 7: Gathering Feedback

[0991] The device collects user feedback and sends it to the server.

[0992] Input: User feedback.

[0993] Output: Feedback data sent to the server.

[0994] Specific actions:

[0995] The device displays a feedback form within the chat window.

[0996] Examples: "Was this advice helpful?" "Do you have any other comments?"

[0997] The user enters their feedback and presses the submit button.

[0998] Step 8: Analyze and implement feedback

[0999] The server analyzes the collected feedback and incorporates it into the next advice generation process.

[1000] Input: Collected feedback data.

[1001] Output: Improved advice generation process.

[1002] Specific actions:

[1003] The server analyzes the feedback data and saves the obtained information as training data.

[1004] Based on the feedback, the server improves the prompts and advice it inputs into the generated AI model.

[1005] This entire process allows users to receive appropriate support to alleviate their stress and worries, and to improve the quality of their relationships and communication in the workplace.

[1006] (Application Example 2)

[1007] Next, we will explain application example 2. In the following explanation, 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."

[1008] In modern workplaces and brick-and-mortar stores, employees often experience stress from communicating with customers and colleagues. This leads to problems such as decreased employee performance and customer satisfaction. Furthermore, there is a growing need to focus on stress management and implement effective support systems.

[1009] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting data from a communication tool used by the user, means for analyzing the collected data using natural language processing to extract the tension and emotions the user is experiencing, means for generating appropriate advice for the user based on the analysis results, means for displaying the generated advice on the user's display device and providing it to the user, means for collecting the user's reactions and reflecting them in the generation of advice for the next time, and means for detecting the user's state of tension in real time during the conversation and generating and providing appropriate suggestions. As a result, the user can receive concrete support to reduce stress caused by communication and enable smooth conversations, making it possible to improve interpersonal relationships and customer satisfaction in the workplace and physical stores.

[1010] A "communication tool" refers to an online platform or application that users use to exchange messages and information with other people.

[1011] "Natural language processing" refers to the process by which computers understand and analyze natural language text. Specifically, it includes techniques such as grammatical analysis, sentiment analysis, and keyword extraction.

[1012] "Tension" refers to the stress and anxiety that users experience during communication. This often involves psychological burden.

[1013] "Emotions" refer to the user's psychological state, including emotional states such as joy, anger, sadness, and surprise.

[1014] "Advice" refers to specific suggestions or advice that help users act appropriately.

[1015] A "display device" is a device that a user uses to visually confirm information. This includes smart glasses and smartphones.

[1016] "Response" refers to the feedback or actions a user takes in response to the advice they receive.

[1017] "Real-time" refers to processing or responding immediately without delay.

[1018] A "suggestion" refers to a suggestion of actions or measures provided based on the user's situation and emotional state.

[1019] "Support" refers to assistance and services provided to users to resolve stress and communication problems.

[1020] Embodiments of this invention include systems utilizing communication tools, natural language processing, sentiment analysis engines, display devices, and feedback collection means. Their specific configuration and operation are described below.

[1021] The system includes means for collecting data from communication tools that users use on a daily basis. For example, the server uses the API of the communication tool to retrieve messages sent and received by users at regular intervals. Since the collected data is not suitable for analysis in its raw form, the server analyzes this data using natural language processing (NLP) techniques.

[1022] During the analysis phase, text data is tokenized, and processes such as sentiment analysis and keyword extraction are performed. This allows for the identification of the user's tension and emotions. For example, if a user sends a message such as "I'm exhausted from having to deal with too many customers," this statement is analyzed as an indication of stress.

[1023] Next, based on the analysis results, the server uses an emotion analysis engine to evaluate the user's emotional state. It then generates appropriate advice corresponding to the results. For example, if the server determines that the user is feeling stressed, it will generate specific advice such as "take a short break" or "talk to your supervisor."

[1024] The generated advice is displayed in real time through the user's display device (e.g., smart glasses or smartphone). This allows the user to receive advice quickly during the conversation. It also has a function to detect the user's real-time level of tension and provide appropriate suggestions immediately.

[1025] Next, the system includes a means for users to provide feedback on the advice they receive. Users input their opinions and thoughts on the advice, and this feedback is sent to the server. The server analyzes the feedback and incorporates it into the next advice generation. This allows the system to continuously provide optimal support tailored to the user's situation.

[1026] The specific hardware and software used include:

[1027] Smart glasses: A device worn by store employees to display information in real time.

[1028] SmartGlassesAPI: An API that collects data from smart glasses and communicates with them.

[1029] TextAnalyzer (NLP Module): A module for natural language processing.

[1030] EmotionRecognitionEngine: Emotion Analysis Engine

[1031] ChatAI: A chat AI for generating and providing advice

[1032] As a concrete example, suppose a store employee is serving a customer and the customer asks for a special discount. In this case, the system detects in real time that the employee is feeling stressed and offers advice such as, "Relax and try offering the best possible discount." This reduces employee stress and improves customer satisfaction.

[1033] Examples of prompt statements are as follows:

[1034] "We will provide conversation data between store staff and customers. If the staff member is experiencing stress, please generate advice on how to handle the situation: {conversation data}"

[1035] This will enable employees to communicate more smoothly and improve the quality of customer service in the workplace and at physical stores.

[1036] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1037] Step 1:

[1038] The server collects data through the APIs of the communication tools used by the user. Specifically, it periodically calls the APIs to retrieve messages and chat logs. This input data consists of messages sent and received by the user. In this case, it also includes obtaining real-time conversation data using, for example, the data collection function of smart glasses.

[1039] Step 2:

[1040] The server stores the collected data as text data. Next, it processes the collected text data using natural language processing (NLP) techniques. The input data is the collected text data, and the output data is the analyzed text structure and extracted keywords. Specifically, it performs text tokenization, grammatical analysis, keyword extraction, and sentiment analysis.

[1041] Step 3:

[1042] The server uses a sentiment analysis engine to evaluate the user's emotional state based on the analysis results obtained through natural language processing. The input data consists of the analyzed text structure and keywords, while the output data is the user's emotional state (e.g., stress, joy, anger). Here, an emotion detection algorithm is used to identify emotions within the text and quantify the emotional state.

[1043] Step 4:

[1044] Based on the results of the emotion analysis engine, the server generates appropriate advice. The input data is the user's emotional state, and the output data is specific advice and suggestions. Using a generative AI model, it generates advice that is appropriate to the user's current situation. For example, if the user is feeling stressed, it will output specific advice such as "take a short break."

[1045] Step 5:

[1046] The generated advice is transmitted in real time to the user's display device, such as smart glasses or a smartphone. The input data is the generated advice, and the output data is the advice message displayed to the user. The smart glasses' display function is used to provide the user with advice visually in real time.

[1047] Step 6:

[1048] When a user enters feedback on advice, that feedback data is sent to the server. The input data is the user's feedback, and the output data is the feedback information stored on the server. The user enters the feedback through smart glasses.

[1049] Step 7:

[1050] The server analyzes the collected feedback data and incorporates it into the next advice generation. The input data is feedback information, and the output data is the improved advice generation algorithm. Specifically, a machine learning algorithm is used to analyze the feedback data and tune the AI ​​model for generation.

[1051] The above outlines the specific processing steps required to implement this system. This process allows users to receive support in real time, reducing communication stress.

[1052] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1053] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1054] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1055] [Fourth Embodiment]

[1056] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1057] As shown in Figure 7, the 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.

[1058] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1059] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1060] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1061] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1062] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1063] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1064] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1065] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1066] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1067] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1068] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1069] This invention is a system designed to reduce communication stress in the workplace and during teleworking, and to facilitate smooth communication. The system collects data from online communication tools used by the user, analyzes it using natural language processing, and provides advice.

[1070] First, the server connects to the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data, such as messages and emails. For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[1071] Next, the server analyzes the collected data using natural language processing (NLP) techniques. The server removes noise from the obtained data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, to identify the cause of stress from a message such as "I'm exhausted from too many meetings." In this case, the server determines that excessive meetings are the main cause of the user's stress.

[1072] Furthermore, the server generates specific advice for the user based on the analysis results. For example, it might suggest to your boss to reduce the number of meetings or to do five minutes of stretching after a meeting to refresh yourself.

[1073] Subsequently, a chat AI running on the device provides the user with generated advice. The user can receive and implement the advice through the chat window on the device. The chat AI also suggests, if necessary, the use of an occupational physician consultation or in-house counseling services. For example, it might suggest, "It seems you've been feeling stressed a lot lately. Why don't you try scheduling an appointment with an occupational physician?"

[1074] Finally, the process includes collecting user feedback. When users input their opinions and thoughts on the advice provided on their device, this feedback is sent to the server. The server analyzes the feedback and incorporates it into future advice generation. This continuous process allows the system to consistently provide optimal support tailored to the user's situation.

[1075] This allows users to receive specific support to alleviate their stress and worries, and to improve the quality of their relationships and communication in the workplace.

[1076] The following describes the processing flow.

[1077] Step 1:

[1078] The server obtains the necessary permissions to access the online communication tools used by the user (e.g., Slack, Gmail). This allows the server to collect information such as the user's messages and emails.

[1079] Step 2:

[1080] The server calls the APIs of online communication tools at a specified schedule (for example, every night at midnight) to collect user-sent and received messages and emails. This includes using the APIs of each tool to retrieve data.

[1081] Step 3:

[1082] The server performs filtering on the collected data to remove noise (e.g., system messages and spam emails). Regular expressions and keyword filtering techniques are used for noise reduction.

[1083] Step 4:

[1084] The server divides the noise-free data into sentences and converts them into a format suitable for natural language processing (NLP). This includes text tokenization and format conversion.

[1085] Step 5:

[1086] The server uses NLP models (e.g., sentiment analysis models) to classify the emotions (e.g., joy, anger, sadness) in each message or email. This helps identify the user's stress levels and emotional state.

[1087] Step 6:

[1088] Based on the sentiment analysis results, the server extracts keywords and phrases that are causing stress. For example, from the message "too many meetings," it identifies "excessive meetings" as a stressor.

[1089] Step 7:

[1090] Based on the analysis results, the server generates specific advice to provide to the user. This advice is template-based and includes content tailored to the individual user's situation.

[1091] Step 8:

[1092] The on-device chat AI presents generated advice to the user. The advice messages are displayed in a natural, conversational format through the chat window, making them easy for the user to understand.

[1093] Step 9:

[1094] If necessary, the terminal's chat AI will suggest to the user the use of occupational health physician consultations or in-house counseling services. For example, if the user's stress level is high, it will provide a dedicated booking link.

[1095] Step 10:

[1096] Users enter feedback on the advice provided in the chat window. This feedback includes opinions on the effectiveness of the advice and other comments.

[1097] Step 11:

[1098] The server analyzes user feedback and incorporates the results into future advice generation. This allows the system to continuously optimize itself and provide support tailored to the user's situation.

[1099] (Example 1)

[1100] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1101] In traditional online communication environments, there is a lack of concrete support to effectively reduce the stress users experience in the workplace or remotely and to facilitate smooth communication. Furthermore, there is a lack of means to identify the causes of stress and provide appropriate advice, making it difficult for users to find ways to cope with stress. Additionally, mechanisms for continuously improving support through feedback are insufficient.

[1102] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1103] In this invention, the server includes means for collecting data from online communication tools used by the user; means for removing irrelevant information from the collected data and dividing the text into sentences; means for performing sentiment analysis and keyword extraction from the divided text to analyze the user's sources of stress and emotions; means for creating prompt sentences using a generation AI model based on the analysis results and generating specific advice for the user; means for displaying the generated advice on the user's terminal and providing it to the user; and means for collecting user feedback and reflecting it in the generation of advice for the next time. This makes it possible to reduce the user's communication-related stress, provide appropriate advice, and continuously improve the support content based on the feedback.

[1104] An "online communication tool" is software or an application that allows users to communicate in real time using text, voice, or video over the internet.

[1105] "Means of data collection" refers to programs and functions that obtain necessary information and messages from users' online communication tools.

[1106] "Means of removing irrelevant information" refer to algorithms and filtering functions used to remove unwanted elements and noise from collected data.

[1107] "Methods for dividing text into sentences" refer to natural language processing techniques for dividing collected text data into smaller sentence units.

[1108] "Sentiment analysis" is an algorithm used to identify emotions such as positive, negative, and neutral from text data.

[1109] Keyword extraction is a natural language processing technique used to extract important words and phrases from text data.

[1110] A "source of stress" refers to an event or factor that causes stress for the user.

[1111] A "generative AI model" is a model that uses machine learning and artificial intelligence technologies to learn patterns from data and generate new information and suggestions.

[1112] A "prompt message" is an instruction given to an AI model to generate specific information or answers.

[1113] "Means for generating advice" refers to functions that create useful suggestions and instructions for users based on analysis results and AI models.

[1114] "Means of collecting feedback" refers to functions for receiving opinions and feedback from users, saving them as data, and analyzing them.

[1115] "Improving support services" means updating and adjusting the advice and support provided by the system based on collected feedback data to make them more effective.

[1116] This invention is a system designed to reduce the stress users experience from communication in the workplace or remotely, and to facilitate smooth communication. The system collects and analyzes data through multiple steps and provides appropriate advice.

[1117] First, the server connects to the online communication tool used by the user. Specifically, it connects using the Slack API and performs authentication. By obtaining the necessary permissions, the server collects messages and related information sent and received by the user. This data collection is performed automatically every night at midnight, and the collected data is stored in a database.

[1118] Next, the server analyzes the collected data using natural language processing (NLP) techniques. Specifically, it first removes noise, eliminating unnecessary information, emojis, URLs, etc. Then, it uses a natural language processing library (e.g., NLTK) to divide the cleaned-up text into sentences. Each of these resulting sentences is then subjected to sentiment analysis and keyword extraction.

[1119] The server uses a sentiment analysis algorithm to evaluate the sentiment of each message. For example, in a message like "I'm tired of too many meetings," "tired" is evaluated as a negative emotion. Furthermore, a keyword extraction algorithm is used to identify important words and phrases (e.g., "meetings" and "tired").

[1120] Based on the analysis results, the server uses a generative AI model to create prompt statements and generate specific advice for the user. These prompt statements are used as input to the generative AI model. As a specific example, the following prompt statement is input to the generative AI model:

[1121] What advice would you provide to help users reduce stress at work? For example, generate advice for a situation where a user says, "I'm exhausted because there are too many meetings."

[1122] This prompt allows the AI ​​model to generate appropriate advice regarding specific stressors.

[1123] Next, the generated advice is sent from the server to the user's device. It is displayed in the chat window on the device, where the user can review and implement the advice. For example, the advice might say, "Try suggesting to your boss that you reduce the number of meetings." The chat AI also suggests medical consultations or the use of internal support services as needed. For example, it might suggest, "You seem to be feeling stressed lately. Why don't you try scheduling a medical consultation?"

[1124] Finally, the process includes collecting user feedback. Through a feedback window on the device, users enter their opinions and impressions of the advice provided. This feedback is sent to the server and incorporated into future advice generation. This allows for continuous improvement of the support provided, ensuring that users receive the best possible support.

[1125] In this way, this system not only reduces the user's communication-related stress and provides appropriate advice, but also enables continuous improvement of support based on feedback.

[1126] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1127] Step 1: Initial setup and data collection

[1128] Subject: Server

[1129] The server connects to the online communication tool used by the user. For example, it uses a Slack API key for authentication. The server obtains the necessary permissions (such as reading messages and retrieving user information). Every day at midnight, it calls the API to collect messages sent and received by the user during that day. The API key and authentication information are used as input, and the message data is stored in a database as output.

[1130] Step 2: Data preprocessing

[1131] Subject: Server

[1132] The server removes unwanted noise from the collected data. Specifically, it removes emojis, URLs, unnecessary spaces, etc. It uses the collected message data as input and outputs cleaned-up text data.

[1133] Step 3: Split the text into sentence units.

[1134] Subject: Server

[1135] The server splits the cleaned text into sentences. It uses a natural language processing library (e.g., NLTK) to split each message into sentences. The input is the cleaned text data, and the output is a list of the split sentences.

[1136] Step 4: Sentiment analysis and keyword extraction

[1137] Subject: Server

[1138] The server performs sentiment analysis on segmented sentences, evaluating the tone and sentiment of the message. It also uses a keyword extraction algorithm (e.g., TF-IDF) to identify important keywords. The input is a list of segmented sentences, and the output is the sentiment evaluation result and a keyword list. For example, from the message "I'm tired because there are too many meetings," "tired" is evaluated as a negative emotion, and "meetings" and "tired" are extracted as important keywords.

[1139] Step 5: Generating Advice

[1140] Subject: Server

[1141] The server uses a generative AI model to create prompt sentences based on the results of sentiment analysis and keyword extraction, generating specific advice for the user. The input consists of sentiment evaluation results and a keyword list, and the output is the generated advice. A concrete example of a prompt sentence used is: "What advice would you provide to reduce the stress a user feels at work? For example, generate advice for the situation, 'I'm exhausted because there are too many meetings.'"

[1142] Step 6: Delivering Advice

[1143] Subject: terminal

[1144] The terminal displays advice received from the server in a chat window. Users can review and implement the advice through this window. The input is advice generated from the server, and the output is the advice displayed in the chat window. For example, advice such as "Suggest reducing the number of meetings to your boss" might be displayed.

[1145] Step 7: Present additional proposals

[1146] Subject: Chat AI

[1147] The chat AI suggests medical consultations and internal support services based on the user's situation. For example, it might suggest, "It seems you've been feeling stressed lately. Why not schedule a medical consultation?" The AI ​​uses the user's emotional analysis results and past advice history as input, and provides additional suggestions as output.

[1148] Step 8: Gathering and analyzing feedback

[1149] Subject: User and Server

[1150] Users input their opinions and feedback on the advice provided through a feedback window on their device. The server receives this feedback and analyzes it. The input is the user's feedback data, and the output is the analysis results of the feedback. These analysis results are reflected in the next advice generation. This allows for continuous improvement of the support provided.

[1151] (Application Example 1)

[1152] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1153] In modern workplaces and teleworking environments, employees are prone to experiencing stress related to their work and communication. In particular, employees in physical stores often experience significant stress when dealing with customers and processing business messages. However, it is difficult for employees themselves to accurately identify their stress levels and deal with them appropriately. This invention aims to provide concrete support methods to help employees alleviate their stress and ensure smooth work operations.

[1154] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1155] In this invention, the server includes means for collecting data from online communication tools used by the user, means for analyzing the collected data using natural language processing to extract the user's stress and emotions, means for generating appropriate advice for the user based on the analysis results, means for displaying the generated advice on the user's terminal and providing it to the user, means for collecting user feedback and reflecting it in the generation of future advice, means for collecting work message data from employees in physical stores and providing advice based on it, means for analyzing the collected work message data using natural language processing to identify the causes of employee stress, and means for analyzing the collected feedback and reflecting it in the generation of future advice. As a result, employees can reduce stress in their daily work and achieve smoother communication.

[1156] "Online communication tools" are software and services that allow users to exchange messages, make calls, and conduct video conferences over the internet.

[1157] "Natural language processing" is a general term for technologies that enable computers to understand, analyze, and generate human language (natural language).

[1158] "Sentiment analysis" is a technique that analyzes text data to determine whether its content is positive, negative, or neutral.

[1159] "Advice" refers to suggestions or suggestions that recommend appropriate measures or actions in response to a specific problem or situation.

[1160] A "device" is a computing device that a user directly operates (for example, a smartphone, tablet, or PC).

[1161] "Feedback" refers to the act or content of a user's response to advice or services provided, such as an opinion or comment.

[1162] A "physical store" is a commercial facility or business location located in a physical space where customer service and sales activities take place.

[1163] "Business message data" refers to the content of messages generated by users during their work, and includes data such as order processing, customer service records, and internal communications.

[1164] "Causes of stress" refer to the psychological and physical burdens and pressures that users experience while performing their work.

[1165] "Advice generation" is the process of creating advice based on analysis results to suggest appropriate actions and measures for the user.

[1166] This invention is a system designed to reduce communication stress in the workplace and during teleworking, and to enable smoother communication.

[1167] This system includes the following means:

[1168] 1. A means of collecting data from online communication tools used by users.

[1169] 2. A method for analyzing collected data using natural language processing to extract the stress and emotions experienced by users.

[1170] 3. A means of generating appropriate advice for the user based on the analysis results.

[1171] 4. A means of displaying the generated advice on the user's device and providing it to the user.

[1172] 5. A means of collecting user feedback and incorporating it into future advice generation.

[1173] 6. A means by which employees at physical stores collect business message data and provide advice based on that data.

[1174] 7. A method for identifying the causes of employee stress by analyzing collected business message data using natural language processing.

[1175] 8. A means of analyzing the collected feedback and incorporating it into the generation of future advice.

[1176] The server first connects to the user's online communication tool, obtains the necessary permissions, and collects data. For example, the server calls an API every night at midnight to retrieve messages sent and received by the user during that day.

[1177] Next, the server analyzes the collected data using natural language processing (NLP) techniques. It removes noise from the collected business message data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, to identify the cause of stress from a message such as "I'm exhausted from too many meetings."

[1178] Therefore, we use the Vader SentimentIntensityAnalyzer from the Natural Language Toolkit (NLTK). The server identifies negative emotions from the data obtained and generates advice based on them. For example, it might generate advice such as "Remember to take a break" or "Let's lower the priority of this task and do it another day."

[1179] The generated advice is provided through a chat window on the user's device. The user can receive and implement this advice. If necessary, the chat AI can also suggest to the user the use of an industrial physician consultation or in-house counseling services.

[1180] When a user submits feedback, it is sent to the server. The server analyzes this feedback and incorporates it into the generation of future advice. This continuous process allows the system to consistently provide optimal support tailored to the user's situation.

[1181] As a concrete example, consider a scenario where employee A at a physical store is replenishing sandwiches that have run out of fillings after lunchtime while also assisting customers. In this situation, work message data is collected and sentiment analysis is performed to identify that employee A is experiencing stress. In response to this situation, advice such as "Remember to take a break during the busy post-lunchtime period" is generated and provided to employee A's terminal.

[1182] Examples of prompts for a generative AI model:

[1183] "Employees at physical stores sometimes experience stress in their daily work. For example, they might feel stressed when they have to adjust sandwiches that have run out of fillings after lunchtime while also dealing with customers. What kind of advice would you offer them?"

[1184] This allows employees in physical stores to use smart assistants for stress management, thereby improving the quality and efficiency of their work.

[1185] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1186] Step 1:

[1187] The server connects to the online communication tool used by the user and obtains the necessary permissions.

[1188] Input: User's online communication tool authentication information.

[1189] Specific operation: The server uses the API to obtain an authentication token and gain permission to access the data.

[1190] Output: User's access permissions to message data.

[1191] Step 2:

[1192] The server collects messages sent and received by users.

[1193] Input: User's online communication tool credentials and API key.

[1194] Specific operation: The server retrieves messages sent and received within a specified time frame (e.g., 1 day) via the API.

[1195] Output: Acquired message data.

[1196] Step 3:

[1197] The server analyzes the collected message data using natural language processing (NLP) techniques.

[1198] Input: Collected message data.

[1199] Specific operation: The server uses the NLTK library to split messages into sentences and perform sentiment analysis.

[1200] Output: Sentiment score (positive, negative, neutral) for each message.

[1201] Step 4:

[1202] The server generates appropriate advice for the user based on the analysis results.

[1203] Input: Sentiment score.

[1204] Specific operation: The server uses a generative AI model to generate appropriate advice corresponding to the sentiment score.

[1205] Output: Generated advice.

[1206] Step 5:

[1207] The generated advice is displayed on the user's device and provided to the user.

[1208] Input: Generated advice.

[1209] Specific actions: Display advice in the chat window on the device. Send notifications as needed.

[1210] Output: Advice displayed to the user.

[1211] Step 6:

[1212] Users follow the advice and input the results and feedback.

[1213] Input: User feedback.

[1214] Specific operation: The user provides feedback through an input form on the device.

[1215] Output: Provided feedback data.

[1216] Step 7:

[1217] The server analyzes the collected feedback and incorporates it into the generation of advice for the next time.

[1218] Input: User feedback data.

[1219] Specific operation: The server analyzes the feedback data and identifies areas for improvement needed to generate advice for the next time.

[1220] Output: Improved advice generation algorithm.

[1221] Step 8:

[1222] The server collects business message data from employees at physical stores and provides advice based on that data.

[1223] Input: Business message data collected by employees at physical stores.

[1224] Specific operation: The server analyzes the business message data from physical stores using natural language processing technology to identify the causes of employee stress.

[1225] Output: Advice to provide to employees.

[1226] Step 9:

[1227] The collected feedback will be analyzed and used to improve the generation of advice for the next time.

[1228] Input: Feedback data collected from employees and users.

[1229] Specific operation: The server analyzes the feedback data to gain insights that will improve the quality of future advice.

[1230] Output: Improved version of the advice that will be generated next time.

[1231] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1232] This invention is a system designed to reduce communication-related stress and facilitate smooth communication for users in the workplace or while teleworking. The system collects data from online communication tools, analyzes it using natural language processing, recognizes the user's emotions using an emotion engine, and provides advice—a comprehensive process.

[1233] First, the server accesses the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data (such as messages and emails). For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[1234] Next, the server analyzes the collected data using natural language processing (NLP) techniques. The server removes noise from the obtained data and divides the text into sentences. The analysis is performed through sentiment analysis and keyword extraction, for example, identifying the cause of stress from a message such as "I'm tired because there are too many meetings." Based on this, the sentiment engine analyzes the user's emotions in real time and identifies specific emotional states (e.g., stress, joy, anger).

[1235] Furthermore, the server generates advice for the user based on the analysis results of the emotion engine. For example, it might create specific advice such as "suggest to your boss that you reduce the number of meetings" or "do 5 minutes of stretching to refresh yourself after a meeting." The emotion engine can adjust the content and timing of the advice according to the user's emotional state.

[1236] Subsequently, a chat AI running on the device provides the user with generated advice. The user can receive and implement the advice in a natural conversational format through the chat window on the device. The chat AI also suggests, if necessary, the use of an occupational physician consultation or in-house counseling services. For example, it might suggest, "It seems you've been feeling stressed a lot lately. Why don't you try scheduling an appointment with an occupational physician?"

[1237] Finally, the process includes collecting user feedback. When users input their opinions and thoughts on the advice provided on their device, this feedback is sent to the server. The server analyzes the feedback and incorporates it into future advice generation. This allows the system to continuously provide optimal support tailored to the user's situation.

[1238] This allows users to receive specific support to alleviate their stress and worries, improving the quality of workplace relationships and communication. Furthermore, the use of an emotion engine enables real-time responses to the user's emotional state, resulting in more effective support.

[1239] The following describes the processing flow.

[1240] Step 1:

[1241] The server obtains the necessary permissions to access the online communication tools (e.g., Slack, Gmail) that the user uses. For example, by granting access permissions to a tool, the server can retrieve data through the APIs of those tools.

[1242] Step 2:

[1243] The server calls APIs of online communication tools at a specified schedule (for example, every night at midnight) to collect users' sent and received messages and emails. This includes actions such as using the Slack API to retrieve chat history and using the Gmail API to collect emails tagged with specific labels.

[1244] Step 3:

[1245] The server performs filtering on the collected data to remove noise (e.g., system messages and spam). Regular expressions and keyword filtering techniques are used for noise reduction. For example, messages with a specific format or pattern are excluded.

[1246] Step 4:

[1247] The server divides the noise-free data into sentences and converts them into a format suitable for natural language processing (NLP). This includes processes such as text tokenization and format conversion. For example, each sentence is organized as a separate entry in JSON format.

[1248] Step 5:

[1249] The server uses NLP models (e.g., sentiment analysis models) to classify the emotions (e.g., joy, anger, sadness) in each message or email. This allows the server to identify the user's stress levels and emotional state.

[1250] Step 6:

[1251] The server uses an emotion engine to analyze the user's emotions in real time. The emotion engine recognizes emotions in real time from the collected data and reflects the results. For example, if the user is feeling stressed, it quantifies that stress level.

[1252] Step 7:

[1253] Based on the analysis results, the server generates specific advice to provide to the user. This uses a template-based generation method, including advice tailored to each user's situation. For example, it might provide specific advice such as "Suggest to your boss that you reduce the number of meetings."

[1254] Step 8:

[1255] The on-device chat AI presents the user with generated advice. The generated advice messages are displayed in a natural conversational format through the chat window. For example, it might display a message like, "Are you feeling tired from all the meetings lately? Why not suggest to your boss that we reduce the number of meetings?"

[1256] Step 9:

[1257] If necessary, the terminal's chat AI will suggest to the user the use of an occupational physician consultation or in-house counseling services. If the user's stress level is high, it can also provide a dedicated booking link. For example, it may display a suggestion such as, "If you are feeling stressed, we recommend booking an appointment with an occupational physician."

[1258] Step 10:

[1259] Users enter feedback on the advice provided in the chat window. This feedback may include comments on the effectiveness of the advice and other opinions. For example, they might enter a comment such as, "I followed the advice and reduced the number of meetings, and I feel much better."

[1260] Step 11:

[1261] The server analyzes user feedback and incorporates the results into future advice generation. This allows the system to continuously optimize itself and provide support tailored to the user's situation. For example, it relearns keywords and emotions derived from feedback to improve accuracy.

[1262] (Example 2)

[1263] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1264] In today's business environment, users frequently utilize online communication tools, but managing the stress and negative emotions that arise from these interactions is a challenge. This often impairs users' work efficiency and mental health. Furthermore, current systems lack sufficient means to analyze users' emotional states in real time and provide appropriate advice. Moreover, collecting feedback based on the advice provided and incorporating it into future advice is difficult. To address these challenges, a system is needed that efficiently analyzes users' online communication data, automatically generates appropriate advice, and delivers it in real time.

[1265] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1266] In this invention, the server includes means for collecting data from online communication tools used by the user, means for removing noise from the collected data and dividing the text into sentences, means for analyzing the collected data using natural language processing to extract the user's stress and emotions, means for analyzing the user's emotional state in real time using an emotion engine, means for generating appropriate advice for the user based on the analysis results, means for inputting prompt sentences into a generation AI model to automatically generate advice, means for displaying the generated advice on the user's terminal and providing it to the user, and means for collecting user feedback and reflecting it in the generation of advice for the next time. This makes it possible to analyze the user's emotional state in real time, automatically generate appropriate advice, and provide it to the user. Furthermore, it is possible to continuously improve the advice based on feedback, so it is expected that the user's stress in online communication and mental health will be reduced.

[1267] An "online communication tool" is a software application that allows users to communicate with each other via the internet using text messages, voice, video, and other methods.

[1268] "Means of data collection" refers to software modules that automatically retrieve messages and other data from online communication tools used by users.

[1269] "Noise removal methods" refer to software processing that removes unnecessary or duplicate information from collected data to improve the accuracy of the analysis.

[1270] A "means for dividing text into sentences" refers to a software algorithm that divides collected messages or data into individual sentences and performs preprocessing for natural language processing.

[1271] Natural Language Processing (NLP) is a technology that enables computers to understand and generate natural language, performing tasks such as sentiment analysis and keyword extraction.

[1272] An "emotion engine" is a software module that analyzes a user's emotional state based on the results of natural language processing, and identifies stress levels and types of emotions.

[1273] The "means for generating advice" refer to a software module that automatically creates specific and appropriate advice for the user based on the analysis results of the emotion engine.

[1274] A "generative AI model" is an artificial intelligence model that generates natural language based on input prompt sentences, and is a technology used to provide appropriate advice to users.

[1275] A "prompt statement" is an instruction statement that is input into a generative AI model to generate advice, and it is a sentence that includes a specific output request to the model.

[1276] "User's device" refers to an electronic device such as a computer or smartphone used by the user, and is a device for displaying the generated advice.

[1277] A "means for collecting feedback" refers to a software module that collects opinions and impressions from users regarding advice, and analyzes and stores that data.

[1278] This invention is a system designed to reduce communication-related stress and facilitate smooth communication for users in the workplace or while teleworking. The system collects data from online communication tools, analyzes it using natural language processing, recognizes the user's emotions using an emotion engine, and provides advice—a comprehensive process.

[1279] First, the server accesses the online communication tools that the user uses daily and obtains the necessary permissions. This allows the server to collect the user's communication data (such as messages and emails). For example, the server calls the Slack API every night at midnight to retrieve all messages sent and received by the user during that day.

[1280] Next, the server preprocesses the collected data. Specifically, it removes noise from the collected data and divides the text into sentences. This preprocessing transforms the data into a format suitable for analysis. For example, the server filters out duplicate messages and meaningless data and divides the text into individual sentences.

[1281] The server then analyzes the pre-processed data using natural language processing (NLP) techniques. NLP libraries such as SpaCy and NLTK are used. Sentiment analysis and keyword extraction are performed to identify the cause of stress from messages like, for example, "I'm exhausted from too many meetings." As a concrete example, prompts such as, "The user is feeling overworked. Please suggest ways to refresh," can be input into a generating AI model to create appropriate advice.

[1282] Furthermore, the server uses an emotion engine to analyze the user's emotional state in real time. The emotion engine identifies the user's stress level and type of emotion based on the results of natural language processing. For example, a message like "I'm exhausted from too many meetings" would be judged as "high stress."

[1283] Next, the server generates appropriate advice for the user based on the analysis results of the emotion engine. It automatically generates advice by inputting prompts into a generation AI model (e.g., GPT-3). Specific examples of advice include "suggest to your boss that you reduce the number of meetings" and "do 5 minutes of stretching to refresh yourself after a meeting."

[1284] The device then provides the user with the generated advice. Through a chat window on the device, the user can receive and implement the advice in a natural, conversational format. For example, it might suggest, "It seems you've been feeling stressed lately. Why not schedule a consultation with an occupational physician?"

[1285] Finally, the device collects user feedback and sends it to the server. A feedback form is displayed in the device's chat window, and when the user enters their opinions and thoughts on the advice provided, that feedback is sent to the server. The server analyzes the feedback and incorporates it into the generation of future advice. This allows the system to continuously provide optimal support tailored to the user's situation.

[1286] This entire process allows users to receive specific support to alleviate their stress and worries, thereby improving the quality of their relationships and communication in the workplace. Furthermore, the use of an emotion engine enables real-time responses to the user's emotional state, resulting in more effective support.

[1287] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1288] Step 1: Data Collection

[1289] The server calls the APIs of online communication tools used by users (e.g., Slack) to collect message data that is sent and received on a daily basis.

[1290] Input: User's API authentication information for their online communication tool.

[1291] Output: Message data sent and received by the user during that day.

[1292] Specific actions:

[1293] The server executes scheduled jobs every night at 12:00.

[1294] The server authenticates with Slack using API keys and authentication tokens.

[1295] The server calls the API to retrieve messages sent and received by the user.

[1296] The retrieved message data is saved to the database.

[1297] Step 2: Data Preprocessing

[1298] The server removes unnecessary information (noise) from the collected message data and divides the text into sentences.

[1299] Input: Collected message data.

[1300] Output: Preprocessed text data.

[1301] Specific actions:

[1302] The server filters out duplicate messages and meaningless data (e.g., system messages) from the data.

[1303] The server divides the text into sentences and converts them into a format that is easy to parse.

[1304] Step 3: Natural Language Processing (NLP) Analysis

[1305] The server uses the pre-processed data to perform sentiment analysis and keyword extraction.

[1306] Input: Preprocessed text data.

[1307] Output: Sentimental states and keywords extracted from text data.

[1308] Specific actions:

[1309] The server performs sentiment analysis using an NLP library (e.g., SpaCy, NLTK).

[1310] For example, messages containing emotions such as "tired" or "happy" are classified as positive, negative, or neutral.

[1311] The server extracts important keywords using technologies such as TF-IDF and Word2Vec.

[1312] Step 4: Emotional state analysis using the emotion engine

[1313] The server uses an emotion engine to analyze the user's emotional state in real time.

[1314] Input: NLP analysis results.

[1315] Output: Quantified emotional state (e.g., stress level).

[1316] Specific actions:

[1317] The emotion engine uses the results of emotion analysis to quantify the user's emotional state.

[1318] For example, the emotional engine interprets the message "I'm exhausted from too many meetings" as indicating high stress levels.

[1319] Step 5: Generate advice

[1320] Based on the analysis results of the emotion engine, the server generates specific advice for the user.

[1321] Input: Quantified result of emotional state.

[1322] Output: Generated advice.

[1323] Specific actions:

[1324] The server inputs prompt text into the generated AI model (e.g., GPT-3) and automatically generates advice.

[1325] Example prompt: "The user is feeling overworked. Please suggest ways for them to refresh themselves."

[1326] The server then provides the generated advice to the user in a personalized manner.

[1327] Step 6: Providing advice

[1328] The device notifies the user via chat AI to provide advice.

[1329] Input: Generated advice.

[1330] Output: Advice provided to the user.

[1331] Specific actions:

[1332] Advice will be displayed in the chat window on the device.

[1333] Examples: "Try suggesting we reduce the number of meetings," or "Let's do some 5-minute stretches to refresh ourselves."

[1334] The chat AI provides advice through a conversational format with the user.

[1335] Step 7: Gathering Feedback

[1336] The device collects user feedback and sends it to the server.

[1337] Input: User feedback.

[1338] Output: Feedback data sent to the server.

[1339] Specific actions:

[1340] The device displays a feedback form within the chat window.

[1341] Examples: "Was this advice helpful?" "Do you have any other comments?"

[1342] The user enters their feedback and presses the submit button.

[1343] Step 8: Analyze and implement feedback

[1344] The server analyzes the collected feedback and incorporates it into the next advice generation process.

[1345] Input: Collected feedback data.

[1346] Output: Improved advice generation process.

[1347] Specific actions:

[1348] The server analyzes the feedback data and saves the obtained information as training data.

[1349] Based on the feedback, the server improves the prompts and advice it inputs into the generated AI model.

[1350] This entire process allows users to receive appropriate support to alleviate their stress and worries, and to improve the quality of their relationships and communication in the workplace.

[1351] (Application Example 2)

[1352] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1353] In modern workplaces and brick-and-mortar stores, employees often experience stress from communicating with customers and colleagues. This leads to problems such as decreased employee performance and customer satisfaction. Furthermore, there is a growing need to focus on stress management and implement effective support systems.

[1354] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting data from a communication tool used by the user, means for analyzing the collected data using natural language processing to extract the tension and emotions the user is experiencing, means for generating appropriate advice for the user based on the analysis results, means for displaying the generated advice on the user's display device and providing it to the user, means for collecting the user's reactions and reflecting them in the generation of advice for the next time, and means for detecting the user's state of tension in real time during the conversation and generating and providing appropriate suggestions. As a result, the user can receive concrete support to reduce stress caused by communication and enable smooth conversations, making it possible to improve interpersonal relationships and customer satisfaction in the workplace and physical stores.

[1355] A "communication tool" refers to an online platform or application that users use to exchange messages and information with other people.

[1356] "Natural language processing" refers to the process by which computers understand and analyze natural language text. Specifically, it includes techniques such as grammatical analysis, sentiment analysis, and keyword extraction.

[1357] "Tension" refers to the stress and anxiety that users experience during communication. This often involves psychological burden.

[1358] "Emotions" refer to the user's psychological state, including emotional states such as joy, anger, sadness, and surprise.

[1359] "Advice" refers to specific suggestions or advice that help users act appropriately.

[1360] A "display device" is a device that a user uses to visually confirm information. This includes smart glasses and smartphones.

[1361] "Response" refers to the feedback or actions a user takes in response to the advice they receive.

[1362] "Real-time" refers to processing or responding immediately without delay.

[1363] A "suggestion" refers to a suggestion of actions or measures provided based on the user's situation and emotional state.

[1364] "Support" refers to assistance and services provided to users to resolve stress and communication problems.

[1365] Embodiments of this invention include systems utilizing communication tools, natural language processing, sentiment analysis engines, display devices, and feedback collection means. Their specific configuration and operation are described below.

[1366] The system includes means for collecting data from communication tools that users use on a daily basis. For example, the server uses the API of the communication tool to retrieve messages sent and received by users at regular intervals. Since the collected data is not suitable for analysis in its raw form, the server analyzes this data using natural language processing (NLP) techniques.

[1367] During the analysis phase, text data is tokenized, and processes such as sentiment analysis and keyword extraction are performed. This allows for the identification of the user's tension and emotions. For example, if a user sends a message such as "I'm exhausted from having to deal with too many customers," this statement is analyzed as an indication of stress.

[1368] Next, based on the analysis results, the server uses an emotion analysis engine to evaluate the user's emotional state. It then generates appropriate advice corresponding to the results. For example, if the server determines that the user is feeling stressed, it will generate specific advice such as "take a short break" or "talk to your supervisor."

[1369] The generated advice is displayed in real time through the user's display device (e.g., smart glasses or smartphone). This allows the user to receive advice quickly during the conversation. It also has a function to detect the user's real-time level of tension and provide appropriate suggestions immediately.

[1370] Next, the system includes a means for users to provide feedback on the advice they receive. Users input their opinions and thoughts on the advice, and this feedback is sent to the server. The server analyzes the feedback and incorporates it into the next advice generation. This allows the system to continuously provide optimal support tailored to the user's situation.

[1371] The specific hardware and software used include:

[1372] Smart glasses: A device worn by store employees to display information in real time.

[1373] SmartGlassesAPI: An API that collects data from smart glasses and communicates with them.

[1374] TextAnalyzer (NLP Module): A module for natural language processing.

[1375] EmotionRecognitionEngine: Emotion Analysis Engine

[1376] ChatAI: A chat AI for generating and providing advice

[1377] As a concrete example, suppose a store employee is serving a customer and the customer asks for a special discount. In this case, the system detects in real time that the employee is feeling stressed and offers advice such as, "Relax and try offering the best possible discount." This reduces employee stress and improves customer satisfaction.

[1378] Examples of prompt statements are as follows:

[1379] "We will provide conversation data between store staff and customers. If the staff member is experiencing stress, please generate advice on how to handle the situation: {conversation data}"

[1380] This will enable employees to communicate more smoothly and improve the quality of customer service in the workplace and at physical stores.

[1381] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1382] Step 1:

[1383] The server collects data through the APIs of the communication tools used by the user. Specifically, it periodically calls the APIs to retrieve messages and chat logs. This input data consists of messages sent and received by the user. In this case, it also includes obtaining real-time conversation data using, for example, the data collection function of smart glasses.

[1384] Step 2:

[1385] The server stores the collected data as text data. Next, it processes the collected text data using natural language processing (NLP) techniques. The input data is the collected text data, and the output data is the analyzed text structure and extracted keywords. Specifically, it performs text tokenization, grammatical analysis, keyword extraction, and sentiment analysis.

[1386] Step 3:

[1387] The server uses a sentiment analysis engine to evaluate the user's emotional state based on the analysis results obtained through natural language processing. The input data consists of the analyzed text structure and keywords, while the output data is the user's emotional state (e.g., stress, joy, anger). Here, an emotion detection algorithm is used to identify emotions within the text and quantify the emotional state.

[1388] Step 4:

[1389] Based on the results of the emotion analysis engine, the server generates appropriate advice. The input data is the user's emotional state, and the output data is specific advice and suggestions. Using a generative AI model, it generates advice that is appropriate to the user's current situation. For example, if the user is feeling stressed, it will output specific advice such as "take a short break."

[1390] Step 5:

[1391] The generated advice is transmitted in real time to the user's display device, such as smart glasses or a smartphone. The input data is the generated advice, and the output data is the advice message displayed to the user. The smart glasses' display function is used to provide the user with advice visually in real time.

[1392] Step 6:

[1393] When a user enters feedback on advice, that feedback data is sent to the server. The input data is the user's feedback, and the output data is the feedback information stored on the server. The user enters the feedback through smart glasses.

[1394] Step 7:

[1395] The server analyzes the collected feedback data and incorporates it into the next advice generation. The input data is feedback information, and the output data is the improved advice generation algorithm. Specifically, a machine learning algorithm is used to analyze the feedback data and tune the AI ​​model for generation.

[1396] The above outlines the specific processing steps required to implement this system. This process allows users to receive support in real time, reducing communication stress.

[1397] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1398] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1399] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1400] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1401] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1402] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1403] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1404] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1405] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1406] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1407] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1408] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1409] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1410] 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.

[1411] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1412] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1413] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1414] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1415] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1416] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1417] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1418] The following is further disclosed regarding the embodiments described above.

[1419] (Claim 1)

[1420] Methods for collecting data from online communication tools used by users,

[1421] The collected data is analyzed using natural language processing to extract the stress and emotions experienced by the user,

[1422] A means for generating appropriate advice for the user based on the analysis results,

[1423] A means of displaying the generated advice on the user's device and providing it to the user,

[1424] A means of collecting user feedback and incorporating it into the generation of future advice,

[1425] A system that includes this.

[1426] (Claim 2)

[1427] This further includes means of suggesting to users the use of industrial physician consultations and in-house counseling services.

[1428] The system according to claim 1.

[1429] (Claim 3)

[1430] This further includes means of integrating and collecting data from multiple online communication tools used by users.

[1431] The system according to claim 1.

[1432] "Example 1"

[1433] (Claim 1)

[1434] Methods for collecting data from online communication tools used by users,

[1435] A method for removing irrelevant information from collected data and dividing the text into sentences,

[1436] A method for analyzing user stressors and emotions by performing sentiment analysis and keyword extraction from segmented text,

[1437] A means of generating prompt sentences using a generation AI model based on analysis results and generating specific advice for the user,

[1438] A means of displaying the generated advice on the user's device and providing it to the user,

[1439] A means of collecting user feedback and incorporating it into the generation of future advice,

[1440] A system that includes this.

[1441] (Claim 2)

[1442] The system according to claim 1, further comprising means of suggesting to users medical consultations or the use of in-house support services.

[1443] (Claim 3)

[1444] The system according to claim 1, further comprising means for integrating and collecting data from multiple online communication tools used by the user.

[1445] "Application Example 1"

[1446] (Claim 1)

[1447] Methods for collecting data from online communication tools used by users,

[1448] The collected data is analyzed using natural language processing to extract the stress and emotions experienced by the user,

[1449] A means for generating appropriate advice for the user based on the analysis results,

[1450] A means of displaying the generated advice on the user's device and providing it to the user,

[1451] A means of collecting user feedback and incorporating it into the generation of future advice,

[1452] A means for employees in physical stores to collect business message data and provide advice based on that data,

[1453] A method for identifying the causes of employee stress by analyzing collected business message data using natural language processing,

[1454] A means to analyze the collected feedback and reflect it in the generation of future advice,

[1455] A system that includes this.

[1456] (Claim 2)

[1457] The system according to claim 1, further comprising means of suggesting to the user the use of an industrial physician consultation or in-house counseling services.

[1458] (Claim 3)

[1459] The system according to claim 1, further comprising means for integrating and collecting data from multiple online communication tools used by the user.

[1460] "Example 2 of combining an emotion engine"

[1461] (Claim 1)

[1462] Methods for collecting data from online communication tools used by users,

[1463] The collected data is analyzed using natural language processing to extract the stress and emotions experienced by the user,

[1464] A means for generating appropriate advice for the user based on the analysis results,

[1465] A means of displaying the generated advice on the user's device and providing it to the user,

[1466] A means of collecting user feedback and incorporating it into the generation of future advice,

[1467] A method for removing noise from collected data and dividing text into sentence units,

[1468] A method for analyzing a user's emotional state in real time using an emotion engine,

[1469] A method for automatically generating advice by inputting prompt sentences into a generative AI model,

[1470] A system that includes this.

[1471] (Claim 2)

[1472] This further includes means of suggesting to users the use of industrial physician consultations and in-house counseling services.

[1473] The system according to claim 1.

[1474] (Claim 3)

[1475] This further includes means of integrating and collecting data from multiple online communication tools used by users.

[1476] The system according to claim 1.

[1477] "Application example 2 when combining with an emotional engine"

[1478] (Claim 1)

[1479] Means of collecting data from communication tools used by users,

[1480] The collected data is analyzed using natural language processing to extract the tension and emotions the user is experiencing,

[1481] A means for generating appropriate advice for the user based on the analysis results,

[1482] A means of displaying the generated advice on the user's display device and providing it to the user,

[1483] A means of collecting user feedback and incorporating it into the next advice generation,

[1484] A means for detecting the user's state of tension in real time during a conversation and generating and providing appropriate suggestions,

[1485] A system that includes this.

[1486] (Claim 2)

[1487] This further includes means of suggesting to users the use of medical professional consultations or internal organizational counseling services.

[1488] The system according to claim 1.

[1489] (Claim 3)

[1490] This further includes means of integrating and collecting data from multiple communication tools used by the user.

[1491] The system according to claim 1. [Explanation of Symbols]

[1492] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Methods for collecting data from online communication tools used by users, The collected data is analyzed using natural language processing to extract the stress and emotions experienced by the user, A means for generating appropriate advice for the user based on the analysis results, A means of displaying the generated advice on the user's device and providing it to the user, A means of collecting user feedback and incorporating it into the generation of future advice, A system that includes this.

2. This further includes means of suggesting to users the use of industrial physician consultations and in-house counseling services. The system according to claim 1.

3. This further includes means of integrating and collecting data from multiple online communication tools used by users. The system according to claim 1.

Images