system

Abstract

We provide the system. [Solution] To detect workplace harassment, means of collecting data, A method for analyzing collected data to evaluate the degree of harassment, A means of generating specific improvement measures based on evaluation results, A means of notifying the user of the generated improvement measures, A system that includes this.

Description

【Technical Field】 , , , 【0004】 , , , , 【0005】 , , , , , , 【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 performed by at least one processor, 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】 Japanese Patent Application Laid-Open No. 2022-180282 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0004】 <00​​​​​​​This invention provides a series of systems for effectively detecting and preventing workplace harassment. The system of this invention has means for collecting data from sources inside and outside the workplace and for evaluating the degree of harassment by analyzing this data. Furthermore, by providing means for generating specific improvement measures based on the evaluation results and notifying the user, it is possible to prevent the occurrence of harassment. In this way, this invention supports the promotion of a healthy and diverse workplace environment. 【0006】 "Means of data collection" refers to the technologies and processes for automatically obtaining relevant data from sources both inside and outside the workplace. 【0007】 "Methods for analyzing and evaluating the degree of harassment" refer to technologies and algorithms that use natural language processing and image analysis based on collected data to quantify the possibility and risk of harassment. 【0008】 "Means for generating specific improvement measures" refer to generative models and algorithms that present action guidelines for problem solving based on the evaluated risk score. 【0009】 "Means of notifying the user" refers to communication and interface technologies for displaying generated suggestions on the user's terminal or management system. [Brief explanation of the drawing] 【0010】 [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 a data processing device and a 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] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention] 【0011】 Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings. 【0012】 First, let's explain the terminology used in the following explanation. 【0013】 In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple 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. 【0014】 In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor. 【0015】 In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs, various parameters, and the like. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like. 【0016】 In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark). 【0017】 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." 【0018】 [First Embodiment] 【0019】 Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment. 【0020】 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. 【0021】 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). 【0022】 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. 【0023】 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. 【0024】 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. 【0025】 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. 【0026】 Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14. 【0027】 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. 【0028】 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. 【0029】 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. 【0030】 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". 【0031】 This invention is a system for effectively detecting and improving workplace harassment issues. This system aims to improve the workplace environment through a process of collecting and analyzing data and generating specific improvement measures. 【0032】 The server first collects data from internal workplace communication tools, external social networking services (SNS), email systems, and other sources. It utilizes technologies such as APIs and crawlers to gather necessary data from various information sources. This collected data includes text, images, and audio, making it possible to centrally collect diverse forms of information. 【0033】 Next, the server analyzes the collected data. For text data, natural language processing techniques are used to tokenize and perform sentiment analysis to detect keywords and emotional expressions that constitute harassment. For image and video data, computer vision techniques are used to analyze specific behaviors and situations to detect abnormal behavior. 【0034】 Based on the analysis results, the server quantifies the degree of harassment and calculates a risk score. This score allows for a quantitative assessment of the potential level of problems. 【0035】 The server then generates specific improvement measures based on the risk score. This improvement is generated using a generative AI model, creating actionable guidelines such as "recommend participation in training to improve communication skills" or "suggest the introduction of specific management methods." 【0036】 Users are notified of these suggestions via their devices. The devices provide an interface that allows users to easily review the suggestions and implement improvements as needed. 【0037】 As a concrete example, the system periodically collects data from a company's internal chat system where a server is located. If it detects that a particular employee is emotionally conflicted or engaging in inappropriate conversation, measures to improve communication are automatically suggested to that employee. The HR personnel, as users of the system, can then review the content and take appropriate action. In this way, the present invention is a system that supports harassment response in corporate human resource management and enables the creation of a healthier and more diverse workplace environment. 【0038】 The following describes the processing flow. 【0039】 Step 1: 【0040】 The server collects data from both inside and outside the workplace through APIs and crawlers. This includes data from communication tools and email systems used by employees, as well as data from publicly available social media platforms. 【0041】 Step 2: 【0042】 The server preprocesses the collected raw data. This preprocessing includes filtering out unnecessary information, formatting conversion, and text cleaning, preparing the data for analysis. 【0043】 Step 3: 【0044】 The server uses natural language processing techniques to analyze text data. Specifically, it performs tokenization, sentiment analysis, and extraction of important keywords to identify language patterns that may indicate harassment. 【0045】 Step 4: 【0046】 The server applies computer vision technology to image and video data to analyze specific behaviors and situations. This process includes detecting abnormal behavior and analyzing the behavioral patterns of specific individuals. 【0047】 Step 5: 【0048】 The server calculates a harassment risk score based on the analysis results. This score is used to quantitatively evaluate the degree of problems for each employee and in specific situations. 【0049】 Step 6: 【0050】 The server generates specific improvement measures based on the calculated risk score. This process utilizes a generative AI model to derive training suggestions and management method recommendations for behavioral improvement. 【0051】 Step 7: 【0052】 The server notifies the user's terminal of the generated improvement suggestions. The terminal displays the received notification on its interface to make it easier for the user to review the suggestions. 【0053】 Step 8: 【0054】 Users review the improvement measures notified on their devices and take action as needed. For example, they might issue training instructions to the relevant employees or promptly follow up with managers. 【0055】 Step 9: 【0056】 The server monitors user feedback and the results of implemented improvements, accumulating this data to help continuously improve the system. 【0057】 (Example 1) 【0058】 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." 【0059】 Workplace harassment is a serious problem that has profound impacts on individuals and organizations as a whole. However, traditional methods have made it difficult to detect potential signs of harassment early and effectively provide concrete corrective measures. Furthermore, there has been a lack of appropriate means for integrating and analyzing various data formats and generating dynamic and individualized corrective measures. 【0060】 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. 【0061】 In this invention, the server includes means for collecting information, means for analyzing the collected information to evaluate the degree of harassment, means for quantifying the risk based on the analysis results, and means for dynamically generating improvement measures using generative AI technology. This enables early detection of harassment and the provision of individualized and effective improvement measures. 【0062】 "Means of information gathering" refers to the function of acquiring data such as text, images, and audio from various sources both inside and outside the workplace. 【0063】 "Means of analyzing information to assess the degree of harassment" refers to a function that analyzes collected information, identifies signs of harassment, and assesses their severity. 【0064】 "Means for generating specific improvement measures" refers to a function for creating effective action guidelines and measures to improve harassment based on the analysis results. 【0065】 "Means of notifying users" refers to functions that provide users with generated improvement measures and analysis results, and effectively communicate necessary information. 【0066】 "Means of quantifying risk" refers to a function for quantitatively evaluating the potential risk of harassment and expressing it numerically. 【0067】 "Means for dynamically generating improvement measures using generative AI technology" refers to a function that utilizes artificial intelligence technology to automatically generate individualized and optimal improvement measures in response to analysis results. 【0068】 This invention is designed as a system for effectively detecting and addressing workplace harassment issues. The system primarily consists of servers and terminals and utilizes various hardware and software technologies. 【0069】 The server collects data from workplace communication tools, external social networking services (SNS), email systems, and other sources. By utilizing APIs (Application Programming Interfaces) and web crawler technologies, it's possible to centralize various forms of information during data collection. The collected data primarily includes text, images, and audio. 【0070】 The collected data is analyzed on a server. Natural language processing techniques are used for the analysis, and text data undergoes tokenization and sentiment analysis. This allows for the detection of keywords and emotional expressions related to harassment. In addition, computer vision techniques are used to analyze image and audio data to identify specific behaviors and situations, thereby detecting abnormal behavior. 【0071】 The analysis results are used to quantify the degree of harassment and calculate a risk score. Based on this score, the server generates appropriate improvement measures. A generative AI model is used to generate improvement measures, creating specific action guidelines such as "recommend participation in training to improve communication skills" or "suggest the introduction of a specific management method." An example of a prompt used here is the text, "Generate suggestions for improving communication in the workplace." 【0072】 The generated improvement measures are notified to the user via a terminal. The terminal provides an interface that allows the user to easily review the suggestions and implement the improvements as needed. In this way, the system actively supports harassment response in corporate human resources management, enabling employees to create a healthier and more diverse workplace environment. 【0073】 As a concrete example, a server periodically collects data from a company's chat system and detects emotional conflicts or inappropriate conversations between specific employees. As a result, communication improvement measures are automatically suggested to the employees in question, and HR personnel, as users, can review the content and take appropriate action. In this way, the present invention plays an important role in improving workplace health. 【0074】 The flow of the specific processing in Example 1 will be explained using Figure 11. 【0075】 Step 1: 【0076】 The server begins collecting information. Inputs include workplace communication tools, external social networking services (SNS), and email systems, from which data is retrieved using APIs and web crawler technologies. Outputs include text, image, and audio data, which are aggregated in the server's database. Specifically, the server accesses each information source at scheduled times, downloads the contents, and stores them. 【0077】 Step 2: 【0078】 The server analyzes the collected data. Input includes various data formats collected in Step 1. Output provides sentiment scores and abnormal behavior detection results as analysis results. For text data, tokenization and sentiment analysis are performed using natural language processing techniques to extract sentiment indicators and harassment-related keywords. For image and audio data, specific behaviors and alarming situations are identified through computer vision and audio analysis. In practice, the server processes data sequentially and generates analysis results using an analysis library. 【0079】 Step 3: 【0080】 The server uses the analysis results to calculate a harassment risk score. The emotion score and anomaly detection data obtained in step 2 are used as input. The output generates a risk score, quantifying potential problems. This enables the server to evaluate risk levels based on quantified indicators. 【0081】 Step 4: 【0082】 The server generates specific improvement measures based on the risk score. A generative AI model is used in this process. A risk score is given as input, and specific suggestions such as "We recommend participation in training to improve communication skills" are generated as output. The prompt text used as input to the generative AI model is "Generate suggestions for improving communication in the workplace." The server registers the generated suggestions in the database. 【0083】 Step 5: 【0084】 The terminal notifies the user of the generated improvement measures. The input is improvement measures provided from the server. The output is a notification message displayed on the interface for the user. Specifically, the terminal sends a notification through its alert function, allowing the user to immediately review the content and take appropriate action. 【0085】 (Application Example 1) 【0086】 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." 【0087】 In the workplace, harassment is a serious problem that negatively impacts the health and efficiency of an organization. Conventional methods make it difficult to detect and address harassment, often leading to delays in responding once the problem becomes apparent. This invention aims to solve these problems and promote improvement of the workplace environment. 【0088】 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. 【0089】 In this invention, the server includes means for collecting data to detect workplace harassment, means for analyzing the collected data to evaluate the degree of harassment, means for generating specific improvement measures based on the evaluation results, means for notifying the user of the generated improvement measures, and means installed on a smartphone that analyzes text data in real time, determines the harassment risk, and sends a notification. This enables early detection of harassment and the implementation of appropriate countermeasures. 【0090】 A "smartphone" is a portable electronic device equipped with communication capabilities that can run applications. 【0091】 "Installed" means to incorporate software into a computer or electronic device and make it usable. 【0092】 "Real-time" refers to the property of performing data processing or analysis at the moment it is requested, or almost instantly. 【0093】 "Text data" refers to digital information composed of characters, and is data stored in the form of sentences or strings. 【0094】 "Analysis" is the process of breaking down collected information and data into its details, understanding them, and extracting meaning and patterns. 【0095】 "Harassment risk" is an indicator that shows the degree to which harassment is likely to occur in the workplace. 【0096】 A "notification" is a communication or alert that conveys specific information to a recipient. 【0097】 The system implementing this invention is for detecting workplace harassment and proposing countermeasures. The server collects data from workplace communication tools and external information sources. The collected data includes text, images, and audio. This allows for the centralized handling of diverse information. 【0098】 The data analysis utilizes natural language processing techniques, specifically NLTK and spaCy, to analyze text data. During the analysis process, tokenization and sentiment analysis are performed to detect harassment-related keywords and emotional expressions. A smartphone is used as the hardware, and the application runs on it. Furthermore, OpenCV is used as computer vision technology to analyze image data and detect specific behaviors. Based on the analysis results, the harassment risk is quantified, and a risk score is calculated. This score indicates the potential degree of the problem. 【0099】 By utilizing a generative AI model, we propose specific improvement measures. This involves using OpenAI's GPT (Grade Point Test) as the generative AI model. For example, it might generate a specific action plan for the user, such as "We recommend participating in a workshop to improve communication skills." This suggestion is sent as a notification to the user's smartphone, allowing them to immediately review and implement the improvement measures. 【0100】 For example, if a chat between employees is deemed to have a high risk of harassment, a notification will be sent to the employee's smartphone stating, "We have detected conflict and negative emotions in recent conversations. Please consider participating in a communication workshop to improve the situation." An example of a prompt for the generating AI model would be, "Detect conflict and negative emotions in employees' recent conversations and suggest ways to improve the situation." This system promotes a healthier workplace environment while simultaneously prompting employees to become more aware of the need for improvement. 【0101】 The flow of a specific process in Application Example 1 will be explained using Figure 12. 【0102】 Step 1: 【0103】 The server collects data from workplace communication tools and external information sources. Inputs include email, chat, and social media data, which are retrieved using APIs and crawlers. Outputs are aggregated datasets of text, images, and audio data. This allows for the effective centralization of data from diverse sources. 【0104】 Step 2: 【0105】 The server analyzes the collected data. The input is the dataset collected in the previous step. For text data, tokenization is performed using natural language processing techniques (e.g., NLTK, spaCy), and sentiment analysis is conducted. The output is the analysis result, which includes keywords and emotional expressions necessary for harassment determination. This extracts the characteristics of the communication content. 【0106】 Step 3: 【0107】 The server uses computer vision technology on image data. The input is the collected image data. Using OpenCV, it executes a behavior identification algorithm to detect abnormal behavior. The output is an analysis showing the possibility of harassment behavior. This allows for the assessment of harassment risk from visual information. 【0108】 Step 4: 【0109】 The server calculates a harassment risk score based on the analysis results. The input consists of analysis results obtained from text and image data. Natural language processing and computer vision analysis results are integrated to quantify the degree of risk. The output is a specific risk score, making it possible to quantitatively demonstrate the severity of the problem. 【0110】 Step 5: 【0111】 The server proposes specific improvement measures using a generative AI model. The inputs are risk scores and analysis results. Using a generative AI model (e.g., OpenAI GPT), it generates actionable guidelines for improving the workplace environment. The output is a proposal for specific improvement measures. This provides effective countermeasures against risks. 【0112】 Step 6: 【0113】 The device notifies the user of suggested improvements. The input is the suggested improvements received from the server. The notification is displayed to the user through an appropriate interface on their smartphone. The output is an information display confirming the action guidelines. The user can then immediately take action after reviewing the suggestions. 【0114】 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. 【0115】 This invention provides a system for detecting and preventing workplace harassment, and by incorporating an emotion engine that recognizes user emotions, it enables more precise analysis and the provision of improvement measures. 【0116】 The server first collects data from multiple sources, such as workplace communication tools, email, and social media. This stage involves securing data access rights through an authentication process and appropriately obtaining the necessary information. The data includes various formats, such as text, images, and audio. 【0117】 Next, the collected data is preprocessed on the server. This preprocessing includes text filtering, image resolution adjustment, and audio clearing. The server analyzes the text data using natural language processing techniques, particularly sentiment analysis, to extract conversational tones and key phrases. It also utilizes computer vision techniques to analyze image and video data, determining emotions from facial expressions and gestures. 【0118】 Furthermore, the server uses an emotion engine to directly recognize emotions from the user's voice and facial expression data. This, combined with a harassment risk score based on the analysis data, enables more accurate situational analysis. 【0119】 Based on the risk score calculated from the analyzed data, the server generates specific improvement measures. This generation utilizes an AI model, suggesting concrete actions such as "proposing individual counseling" or "recommending participation in a course on more appropriate communication techniques." 【0120】 The terminal notifies the user of the generated improvement suggestions. The suggestions are displayed on the user's terminal in an easy-to-understand interface, allowing for quick review and action. Based on these suggestions, the user can decide on and implement appropriate behavioral guidance and training in the workplace. 【0121】 As a concrete example, if an employee receives offensive remarks during work, the server recognizes this change in real time through its emotion engine and alerts the system to the possibility of harassment. Subsequently, a human resources representative (user) can check the alert on their terminal and quickly take appropriate action against the employee in question. In this way, the present invention is a system that enables a rapid and accurate response to harassment issues in corporate human resources management and supports the maintenance of a better work environment. 【0122】 The following describes the processing flow. 【0123】 Step 1: 【0124】 The server collects data from workplace communication platforms, email, and social networking services using APIs or crawling technologies. This collection includes a user authentication process to ensure data access is limited to authorized users. 【0125】 Step 2: 【0126】 The server preprocesses the collected data. This process involves noise reduction, text cleaning, and format conversion of image and audio data to prepare the data for analysis. 【0127】 Step 3: 【0128】 The server uses natural language processing technology to analyze text data and perform sentiment analysis. During this process, it detects emotional expressions and keywords related to potential harassment within the text. 【0129】 Step 4: 【0130】 The server uses computer vision technology to analyze image and video data, particularly recognizing emotions from the user's facial expressions and movements. This process identifies abnormal behavior and changes in emotions. 【0131】 Step 5: 【0132】 Using an emotion engine, the server further identifies the user's emotional state from the voice data and monitors emotional changes in real time. This data is integrated with other emotional data. 【0133】 Step 6: 【0134】 The server integrates these analysis results with sentiment data to score the harassment risk. This score assesses the potential risk of the problem and generates an alert if it exceeds a threshold. 【0135】 Step 7: 【0136】 The server uses a generative AI model to design specific improvement measures based on the harassment risk score. These proposed measures may include individual counseling or participation in educational programs. 【0137】 Step 8: 【0138】 The server sends the generated improvement suggestions to the terminal, and the terminal displays the suggestions on the user interface, allowing the user to quickly review the content. 【0139】 Step 9: 【0140】 Users review the improvement suggestions presented on their devices and, if necessary, implement training and follow-up actions for the relevant employees. They also provide feedback back to the server, contributing to the continuous improvement of the system. 【0141】 (Example 2) 【0142】 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 will be referred to as the "terminal." 【0143】 The challenge lies in maintaining healthy workplace relationships and protecting employees' mental health by early detection of harassment in the workplace and promptly providing appropriate corrective measures. Furthermore, conventional systems may be insufficient in detection, or corrective measures may be too general and lack specificity, thus requiring more precise analysis and individualized responses. 【0144】 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. 【0145】 In this invention, the server includes means for collecting data from communication devices, means for preprocessing and analyzing the collected data in various formats, and means for generating improvement measures using a generative AI model. This makes it possible to analyze workplace communication from multiple perspectives and quickly propose specific and effective improvement measures. 【0146】 "Communication equipment" refers to devices used to send and receive digital data both within and outside the workplace. 【0147】 "Means of data collection" refers to the methods and techniques for obtaining necessary data from various sources within the workplace. 【0148】 "Data in various formats" refers to information that includes different data formats such as text, images, and audio. 【0149】 "Preprocessing" refers to the process of preparing collected data to make it suitable for analysis. 【0150】 A "generative AI model" is a computer program that uses machine learning algorithms to create new information or suggestions. 【0151】 "Means for generating improvement measures" refers to the processes and techniques for constructing action plans and countermeasures based on analysis results. 【0152】 "Means of notifying a display device" refers to a method of displaying the generated information in a visual or audio format so that the user can confirm it. 【0153】 A description of the embodiment for carrying out the invention will be provided. 【0154】 This system is designed as a solution for detecting and preventing workplace harassment. The server collects data from communication devices used within the workplace, utilizing communication platform APIs and email server protocols. The specific hardware consists of cloud servers and local servers, while the software uses scripts and programs for data collection. 【0155】 The data includes various formats, and the server preprocesses them. Text data undergoes sentiment analysis using natural language processing techniques. The software used includes natural language processing models such as BERT and GPT. Image data is analyzed using OpenCV, and audio data is cleared using FFmpeg. 【0156】 The server utilizes a generative AI model to generate specific improvement measures based on the analysis results. This model is built using programming languages ​​such as Python and R, and employs libraries such as TENSORFLOW® and PyTorch. The generated improvement measures are notified to the user via the terminal, enabling rapid response. 【0157】 As a concrete example, if an employee receives an inappropriate comment, the server analyzes the text data in real time and calculates a risk score. An alert is displayed on the terminal, and the user takes appropriate action. An example of a prompt message would be: "Analyze workplace communication data, determine the risk of harassment, and propose specific improvement measures." 【0158】 The flow of the specific processing in Example 2 will be explained using Figure 13. 【0159】 Step 1: 【0160】 The server collects data from various communication devices within the workplace. Specifically, it retrieves emails, chat messages, and social media posts via APIs and protocols. At this stage, the input is log information from the communication platform, and the output is a collection of raw data such as text, images, and audio. 【0161】 Step 2: 【0162】 The server preprocesses the collected data. Text data is processed to remove unwanted information using regular expressions and filtering techniques. Image data is resized and prepared for face recognition using OpenCV, and audio data is denoised using FFmpeg. The input is the raw data obtained in step 1, and the output is a dataset prepared for analysis. 【0163】 Step 3: 【0164】 The server analyzes the pre-processed data. It extracts emotions from text data using natural language processing techniques and recognizes facial expressions from images using computer vision techniques. In speech analysis, an emotion engine extracts speech features. The input is the dataset from step 2, and the output is the analysis results regarding emotions, tone, and facial expressions. 【0165】 Step 4: 【0166】 The server generates improvement measures using a generative AI model. Taking the analysis results as input, the generative AI model generates relevant improvement action suggestions. For example, if the risk score is high, it might suggest specific training or counseling. The output is a concrete action plan. 【0167】 Step 5: 【0168】 The terminal notifies the user of improvement suggestions received from the server. Information is displayed in a user-friendly format through templates and dashboards. The input is the action plan generated in step 4, and the output is an easily understandable information display. The user reviews this notification and decides on specific actions to implement in the workplace. 【0169】 (Application Example 2) 【0170】 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". 【0171】 Preventing harassment in the workplace is crucial, but traditional methods struggle to detect emotional changes and risks in real time. This can lead to delays in appropriate responses and a deterioration of the work environment. Therefore, there is a need for a system that can efficiently and quickly detect signs of harassment and enable responsible personnel to respond immediately. 【0172】 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. 【0173】 In this invention, the server includes means for collecting data to detect workplace harassment, means for analyzing the collected data to evaluate the degree of harassment, and means for generating specific improvement measures based on the evaluation results. This makes it possible to implement workplace environment improvement measures in a timely manner by sending an alert to the person in charge when the risk score exceeds a certain value. 【0174】 "Means of data collection" refers to devices or systems that acquire necessary information from various sources within the workplace, ensuring access rights through authentication and collecting diverse data formats such as text, images, and audio. 【0175】 "Methods for analyzing and evaluating the degree of harassment" refers to a process of analyzing collected data using algorithms and AI to quantify the possibility of harassment based on the tone of conversation, facial expressions, and gestures, and then evaluating its degree. 【0176】 "Means for generating specific improvement measures" refers to a function in which, based on evaluation results, the AI ​​model proposes appropriate action plans and countermeasures, and presents concrete improvement plans to HR personnel and supervisors. 【0177】 "Means for notifying users of generated improvement measures" refers to an interface or system that informs users of proposed measures in an easy-to-understand manner, enabling users to take immediate action by providing notifications quickly and appropriately. 【0178】 A "risk score" is an index that quantifies the likelihood of harassment based on analyzed data, and it is used as a basis for determining the priority of responses. 【0179】 The "means of sending alerts to responsible parties" refer to a system that sends real-time warnings to relevant parties when the risk score is high, based on established criteria, thereby encouraging a swift response. 【0180】 The server is equipped with means to collect various types of data generated within the workplace. Specifically, it retrieves data from sources such as email, instant messages, and voice calls via APIs and dedicated protocols. During this process, an authentication process is used to ensure data access rights and maintain security. 【0181】 Next, the server runs a program to analyze the collected data. This program includes a natural language processing engine and emotion recognition algorithms, implemented using frameworks such as TensorFlow and Keras in Python. NLTK is used for analyzing text data, while OpenCV and Dlib are used for analyzing image and video data. This allows for a detailed analysis of conversational tone and facial expressions, and the calculation of a harassment risk score. 【0182】 The terminal receives the risk score and specific corrective actions generated by the server and notifies the responsible person. Firebase Cloud Messaging is used to send this notification, ensuring that alerts are delivered quickly and reliably. Users, i.e., HR personnel, can then review the details from this notification and quickly take the necessary actions according to the corrective actions. 【0183】 For example, if an employee receives an offensive remark during a weekly meeting, the server immediately performs a sentiment analysis of the conversation, calculates a risk score, and generates appropriate corrective measures. In this case, an alert is immediately sent to the HR department, and specific actions to resolve the problem are proposed. 【0184】 An example of an input prompt for a generative AI model is: "Analyze the conversation log to identify inappropriate expressions and determine if further follow-up is needed." 【0185】 The flow of a specific process in Application Example 2 will be explained using Figure 14. 【0186】 Step 1: 【0187】 The server collects data from workplace communication tools and email systems. Inputs include various data sources accessed via APIs and protocols. Output is raw data obtained in text, audio, and image data formats. At this stage, security is ensured through an authentication process, and permissions for data retrieval are verified. 【0188】 Step 2: 【0189】 The server preprocesses the collected data. The input is the raw data obtained in step 1. The output is data processed into a format suitable for analysis. Specifically, it removes noise from text data, adjusts the resolution of image data, and clarifies audio data. 【0190】 Step 3: 【0191】 The server analyzes data using natural language processing and computer vision technologies. Inputs include pre-processed text, audio, and image data. Outputs are harassment risk scores based on emotional tone and facial expression data obtained through sentiment analysis. At this stage, emotion models using TensorFlow and Keras, and facial expression analysis using OpenCV and Dlib are implemented. 【0192】 Step 4: 【0193】 The server generates specific improvement measures based on the analyzed data. The input is the risk score calculated in step 3. The output is a list of improvement measures and action suggestions generated by the AI ​​model. The generating AI model is used to output improvement suggestions based on specific prompts. 【0194】 Step 5: 【0195】 The terminal notifies the responsible person of the generated improvement plan. The input is the improvement plan and alert information sent from the server. The output is the improvement plan and warning message displayed on the responsible person's device. Firebase Cloud Messaging is used to send notifications and deliver information to the responsible person in real time. 【0196】 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. 【0197】 Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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. 【0198】 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. 【0199】 [Second Embodiment] 【0200】 Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment. 【0201】 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. 【0202】 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). 【0203】 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. 【0204】 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. 【0205】 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). 【0206】 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. 【0207】 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. 【0208】 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. 【0209】 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. 【0210】 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. 【0211】 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". 【0212】 This invention is a system for effectively detecting and improving workplace harassment issues. This system aims to improve the workplace environment through a process of collecting and analyzing data and generating specific improvement measures. 【0213】 The server first collects data from internal workplace communication tools, external social networking services (SNS), email systems, and other sources. It utilizes technologies such as APIs and crawlers to gather necessary data from various information sources. This collected data includes text, images, and audio, making it possible to centrally collect diverse forms of information. 【0214】 Next, the server analyzes the collected data. For text data, natural language processing techniques are used to tokenize and perform sentiment analysis to detect keywords and emotional expressions that constitute harassment. For image and video data, computer vision techniques are used to analyze specific behaviors and situations to detect abnormal behavior. 【0215】 Based on the analysis results, the server quantifies the degree of harassment and calculates a risk score. This score allows for a quantitative assessment of the potential level of problems. 【0216】 The server then generates specific improvement measures based on the risk score. This improvement is generated using a generative AI model, creating actionable guidelines such as "recommend participation in training to improve communication skills" or "suggest the introduction of specific management methods." 【0217】 Users are notified of these suggestions via their devices. The devices provide an interface that allows users to easily review the suggestions and implement improvements as needed. 【0218】 As a concrete example, the system periodically collects data from a company's internal chat system where a server is located. If it detects that a particular employee is emotionally conflicted or engaging in inappropriate conversation, measures to improve communication are automatically suggested to that employee. The HR personnel, as users of the system, can then review the content and take appropriate action. In this way, the present invention is a system that supports harassment response in corporate human resource management and enables the creation of a healthier and more diverse workplace environment. 【0219】 The following describes the processing flow. 【0220】 Step 1: 【0221】 The server collects data from both inside and outside the workplace through APIs and crawlers. This includes data from communication tools and email systems used by employees, as well as data from publicly available social media platforms. 【0222】 Step 2: 【0223】 The server preprocesses the collected raw data. This preprocessing includes filtering out unnecessary information, formatting conversion, and text cleaning, preparing the data for analysis. 【0224】 Step 3: 【0225】 The server uses natural language processing techniques to analyze text data. Specifically, it performs tokenization, sentiment analysis, and extraction of important keywords to identify language patterns that may indicate harassment. 【0226】 Step 4: 【0227】 The server applies computer vision technology to image and video data to analyze specific behaviors and situations. This process includes detecting abnormal behavior and analyzing the behavioral patterns of specific individuals. 【0228】 Step 5: 【0229】 The server calculates a harassment risk score based on the analysis results. This score is used to quantitatively evaluate the degree of problems for each employee and in specific situations. 【0230】 Step 6: 【0231】 The server generates specific improvement measures based on the calculated risk score. This process utilizes a generative AI model to derive training suggestions and management method recommendations for behavioral improvement. 【0232】 Step 7: 【0233】 The server notifies the user's terminal of the generated improvement suggestions. The terminal displays the received notification on its interface to make it easier for the user to review the suggestions. 【0234】 Step 8: 【0235】 Users review the improvement measures notified on their devices and take action as needed. For example, they might issue training instructions to the relevant employees or promptly follow up with managers. 【0236】 Step 9: 【0237】 The server monitors user feedback and the results of implemented improvements, accumulating this data to help continuously improve the system. 【0238】 (Example 1) 【0239】 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." 【0240】 Workplace harassment is a serious problem that has profound impacts on individuals and organizations as a whole. However, traditional methods have made it difficult to detect potential signs of harassment early and effectively provide concrete corrective measures. Furthermore, there has been a lack of appropriate means for integrating and analyzing various data formats and generating dynamic and individualized corrective measures. 【0241】 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. 【0242】 In this invention, the server includes means for collecting information, means for analyzing the collected information to evaluate the degree of harassment, means for quantifying the risk based on the analysis results, and means for dynamically generating improvement measures using generative AI technology. This enables early detection of harassment and the provision of individualized and effective improvement measures. 【0243】 "Means of information gathering" refers to the function of acquiring data such as text, images, and audio from various sources both inside and outside the workplace. 【0244】 "Means of analyzing information to assess the degree of harassment" refers to a function that analyzes collected information, identifies signs of harassment, and assesses their severity. 【0245】 "Means for generating specific improvement measures" refers to a function for creating effective action guidelines and measures to improve harassment based on the analysis results. 【0246】 "Means of notifying users" refers to functions that provide users with generated improvement measures and analysis results, and effectively communicate necessary information. 【0247】 "Means of quantifying risk" refers to a function for quantitatively evaluating the potential risk of harassment and expressing it numerically. 【0248】 "Means for dynamically generating improvement measures using generative AI technology" refers to a function that utilizes artificial intelligence technology to automatically generate individualized and optimal improvement measures in response to analysis results. 【0249】 This invention is designed as a system for effectively detecting and addressing workplace harassment issues. The system primarily consists of servers and terminals and utilizes various hardware and software technologies. 【0250】 The server collects data from workplace communication tools, external social networking services (SNS), email systems, and other sources. By utilizing APIs (Application Programming Interfaces) and web crawler technologies, it's possible to centralize various forms of information during data collection. The collected data primarily includes text, images, and audio. 【0251】 The collected data is analyzed on a server. Natural language processing techniques are used for the analysis, and text data undergoes tokenization and sentiment analysis. This allows for the detection of keywords and emotional expressions related to harassment. In addition, computer vision techniques are used to analyze image and audio data to identify specific behaviors and situations, thereby detecting abnormal behavior. 【0252】 The analysis results are used to quantify the degree of harassment and calculate a risk score. Based on this score, the server generates appropriate improvement measures. A generative AI model is used to generate improvement measures, creating specific action guidelines such as "recommend participation in training to improve communication skills" or "suggest the introduction of a specific management method." An example of a prompt used here is the text, "Generate suggestions for improving communication in the workplace." 【0253】 The generated improvement measures are notified to the user via a terminal. The terminal provides an interface that allows the user to easily review the suggestions and implement the improvements as needed. In this way, the system actively supports harassment response in corporate human resources management, enabling employees to create a healthier and more diverse workplace environment. 【0254】 As a concrete example, a server periodically collects data from a company's chat system and detects emotional conflicts or inappropriate conversations between specific employees. As a result, communication improvement measures are automatically suggested to the employees in question, and HR personnel, as users, can review the content and take appropriate action. In this way, the present invention plays an important role in improving workplace health. 【0255】 The flow of the specific processing in Example 1 will be explained using Figure 11. 【0256】 Step 1: 【0257】 The server begins collecting information. Inputs include workplace communication tools, external social networking services (SNS), and email systems, from which data is retrieved using APIs and web crawler technologies. Outputs include text, image, and audio data, which are aggregated in the server's database. Specifically, the server accesses each information source at scheduled times, downloads the contents, and stores them. 【0258】 Step 2: 【0259】 The server analyzes the collected data. Input includes various data formats collected in Step 1. Output provides sentiment scores and abnormal behavior detection results as analysis results. For text data, tokenization and sentiment analysis are performed using natural language processing techniques to extract sentiment indicators and harassment-related keywords. For image and audio data, specific behaviors and alarming situations are identified through computer vision and audio analysis. In practice, the server processes data sequentially and generates analysis results using an analysis library. 【0260】 Step 3: 【0261】 The server uses the analysis results to calculate a harassment risk score. The emotion score and anomaly detection data obtained in step 2 are used as input. The output generates a risk score, quantifying potential problems. This enables the server to evaluate risk levels based on quantified indicators. 【0262】 Step 4: 【0263】 The server generates specific improvement measures based on the risk score. A generative AI model is used in this process. A risk score is given as input, and specific suggestions such as "We recommend participation in training to improve communication skills" are generated as output. The prompt text used as input to the generative AI model is "Generate suggestions for improving communication in the workplace." The server registers the generated suggestions in the database. 【0264】 Step 5: 【0265】 The terminal notifies the user of the generated improvement measures. The input is improvement measures provided from the server. The output is a notification message displayed on the interface for the user. Specifically, the terminal sends a notification through its alert function, allowing the user to immediately review the content and take appropriate action. 【0266】 (Application Example 1) 【0267】 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." 【0268】 In the workplace, harassment is a serious problem that negatively impacts the health and efficiency of an organization. Conventional methods make it difficult to detect and address harassment, often leading to delays in responding once the problem becomes apparent. This invention aims to solve these problems and promote improvement of the workplace environment. 【0269】 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. 【0270】 In this invention, the server includes means for collecting data to detect workplace harassment, means for analyzing the collected data to evaluate the degree of harassment, means for generating specific improvement measures based on the evaluation results, means for notifying the user of the generated improvement measures, and means installed on a smartphone that analyzes text data in real time, determines the harassment risk, and sends a notification. This enables early detection of harassment and the implementation of appropriate countermeasures. 【0271】 A "smartphone" is a portable electronic device equipped with communication capabilities that can run applications. 【0272】 "Installed" means to incorporate software into a computer or electronic device and make it usable. 【0273】 "Real-time" refers to the property of performing data processing or analysis at the moment it is requested, or almost instantly. 【0274】 "Text data" refers to digital information composed of characters, and is data stored in the form of sentences or strings. 【0275】 "Analysis" is the process of breaking down collected information and data into its details, understanding them, and extracting meaning and patterns. 【0276】 "Harassment risk" is an indicator that shows the degree to which harassment is likely to occur in the workplace. 【0277】 A "notification" is a communication or alert that conveys specific information to a recipient. 【0278】 The system implementing this invention is for detecting workplace harassment and proposing countermeasures. The server collects data from workplace communication tools and external information sources. The collected data includes text, images, and audio. This allows for the centralized handling of diverse information. 【0279】 The data analysis utilizes natural language processing techniques, specifically NLTK and spaCy, to analyze text data. During the analysis process, tokenization and sentiment analysis are performed to detect harassment-related keywords and emotional expressions. A smartphone is used as the hardware, and the application runs on it. Furthermore, OpenCV is used as computer vision technology to analyze image data and detect specific behaviors. Based on the analysis results, the harassment risk is quantified, and a risk score is calculated. This score indicates the potential degree of the problem. 【0280】 By utilizing a generative AI model, we propose specific improvement measures. This involves using OpenAI's GPT as the generative AI model. For example, it might generate a specific action plan for the user, such as "We recommend participating in a workshop to improve communication skills." This suggestion is sent as a notification to the user's smartphone, allowing them to immediately review and implement the improvement measures. 【0281】 For example, if a chat between employees is deemed to have a high risk of harassment, a notification will be sent to the employee's smartphone stating, "We have detected conflict and negative emotions in recent conversations. Please consider participating in a communication workshop to improve the situation." An example of a prompt for the generating AI model would be, "Detect conflict and negative emotions in employees' recent conversations and suggest ways to improve the situation." This system promotes a healthier workplace environment while simultaneously prompting employees to become more aware of the need for improvement. 【0282】 The flow of the specific process in Application Example 1 will be described using FIG. 12. 【0283】 Step 1: 【0284】 The server collects data from communication tools within the workplace and external information sources. The inputs are data from emails, chats, and SNS, which are obtained using APIs or crawlers. The output is a dataset of summarized text, images, and audio data. This makes it possible to effectively integrate data from various information sources. 【0285】 Step 2: 【0286】 The server analyzes the collected data. The input is the dataset collected in the previous step. For text data, tokenization is performed using natural language processing techniques (e.g., NLTK, spaCy), and sentiment analysis is carried out. The output is an analysis result including keywords and emotional expressions necessary for harassment determination. This extracts the characteristics of the communication content. 【0287】 Step 3: 【0288】 The server uses computer vision technology for image data. The input is the collected image data. An action recognition algorithm is executed using OpenCV to detect abnormal behaviors. The output is an analysis result indicating the possibility of harassment behavior. This enables the evaluation of harassment risks from visual information. 【0289】 Step 4: 【0290】 The server calculates a harassment risk score based on the analysis results. The input is the analysis results obtained from text and image data. The analysis results of natural language processing and computer vision are integrated to quantify the degree of risk. The output is a specific risk score. This makes it possible to quantitatively indicate the severity of the problem. 【0291】 Step 5: 【0292】 The server proposes specific improvement measures using a generative AI model. The inputs are risk scores and analysis results. Using a generative AI model (e.g., OpenAI GPT), it generates actionable guidelines for improving the workplace environment. The output is a proposal for specific improvement measures. This provides effective countermeasures against risks. 【0293】 Step 6: 【0294】 The device notifies the user of suggested improvements. The input is the suggested improvements received from the server. The notification is displayed to the user through an appropriate interface on their smartphone. The output is an information display confirming the action guidelines. The user can then immediately take action after reviewing the suggestions. 【0295】 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. 【0296】 This invention provides a system for detecting and preventing workplace harassment, and by incorporating an emotion engine that recognizes user emotions, it enables more precise analysis and the provision of improvement measures. 【0297】 The server first collects data from multiple sources, such as workplace communication tools, email, and social media. This stage involves securing data access rights through an authentication process and appropriately obtaining the necessary information. The data includes various formats, such as text, images, and audio. 【0298】 Next, the collected data is preprocessed on the server. This preprocessing includes text filtering, image resolution adjustment, and audio clearing. The server analyzes the text data using natural language processing techniques, particularly sentiment analysis, to extract conversational tones and key phrases. It also utilizes computer vision techniques to analyze image and video data, determining emotions from facial expressions and gestures. 【0299】 Furthermore, the server uses an emotion engine to directly recognize emotions from the user's voice and facial expression data. This, combined with a harassment risk score based on the analysis data, enables more accurate situational analysis. 【0300】 Based on the risk score calculated from the analyzed data, the server generates specific improvement measures. This generation utilizes an AI model, suggesting concrete actions such as "proposing individual counseling" or "recommending participation in a course on more appropriate communication techniques." 【0301】 The terminal notifies the user of the generated improvement suggestions. The suggestions are displayed on the user's terminal in an easy-to-understand interface, allowing for quick review and action. Based on these suggestions, the user can decide on and implement appropriate behavioral guidance and training in the workplace. 【0302】 As a concrete example, if an employee receives offensive remarks during work, the server recognizes this change in real time through its emotion engine and alerts the system to the possibility of harassment. Subsequently, a human resources representative (user) can check the alert on their terminal and quickly take appropriate action against the employee in question. In this way, the present invention is a system that enables a rapid and accurate response to harassment issues in corporate human resources management and supports the maintenance of a better work environment. 【0303】 The following describes the processing flow. 【0304】 Step 1: 【0305】 The server collects data using API or crawling technology from in-office communication platforms, emails, and SNS. This collection includes a process of user authentication to ensure data access within the permitted scope. 【0306】 Step 2: 【0307】 The server preprocesses the collected data. In this process, noise removal, text cleaning, and format conversion of image and audio data are performed to prepare the data in an analyzable form. 【0308】 Step 3: 【0309】 The server analyzes text data using natural language processing technology to perform sentiment analysis. In this process, detection of sentiment expressions and keywords related to potential harassment in the text is performed. 【0310】 Step 4: 【0311】 The server uses computer vision technology to analyze image and video data, especially to recognize emotions from the user's expressions and movements. This process identifies abnormalities in behavior and changes in emotions. 【0312】 Step 5: 【0313】 Using the emotion engine, the server further identifies the user's emotional state from voice data and monitors changes in emotions in real time. This data is integrated with other emotion data. 【0314】 Step 6: 【0315】 The server integrates these analysis results and emotion data to score the harassment risk. This score evaluates the potential risk of the problem and generates an alert if it exceeds the threshold. 【0316】 Step 7: 【0317】 The server uses a generative AI model to design specific improvement measures based on the harassment risk score. These proposed measures may include individual counseling or participation in educational programs. 【0318】 Step 8: 【0319】 The server sends the generated improvement suggestions to the terminal, and the terminal displays the suggestions on the user interface, allowing the user to quickly review the content. 【0320】 Step 9: 【0321】 Users review the improvement suggestions presented on their devices and, if necessary, implement training and follow-up actions for the relevant employees. They also provide feedback back to the server, contributing to the continuous improvement of the system. 【0322】 (Example 2) 【0323】 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". 【0324】 The challenge lies in maintaining healthy workplace relationships and protecting employees' mental health by early detection of harassment in the workplace and promptly providing appropriate corrective measures. Furthermore, conventional systems may be insufficient in detection, or corrective measures may be too general and lack specificity, thus requiring more precise analysis and individualized responses. 【0325】 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. 【0326】 In this invention, the server includes means for collecting data from communication devices, means for preprocessing and analyzing the collected data in various formats, and means for generating improvement measures using a generative AI model. This makes it possible to analyze workplace communication from multiple perspectives and quickly propose specific and effective improvement measures. 【0327】 "Communication equipment" refers to devices used to send and receive digital data both within and outside the workplace. 【0328】 "Means of data collection" refers to the methods and techniques for obtaining necessary data from various sources within the workplace. 【0329】 "Data in various formats" refers to information that includes different data formats such as text, images, and audio. 【0330】 "Preprocessing" refers to the process of preparing collected data to make it suitable for analysis. 【0331】 A "generative AI model" is a computer program that uses machine learning algorithms to create new information or suggestions. 【0332】 "Means for generating improvement measures" refers to the processes and techniques for constructing action plans and countermeasures based on analysis results. 【0333】 "Means of notifying a display device" refers to a method of displaying the generated information in a visual or audio format so that the user can confirm it. 【0334】 A description of the embodiment for carrying out the invention will be provided. 【0335】 This system is designed as a solution for detecting and preventing workplace harassment. The server collects data from communication devices used within the workplace, utilizing communication platform APIs and email server protocols. The specific hardware consists of cloud servers and local servers, while the software uses scripts and programs for data collection. 【0336】 The data includes various formats, and the server preprocesses them. Text data undergoes sentiment analysis using natural language processing techniques. The software used includes natural language processing models such as BERT and GPT. Image data is analyzed using OpenCV, and audio data is cleared using FFmpeg. 【0337】 The server utilizes a generative AI model to generate specific improvement measures based on the analysis results. This model is built using programming languages ​​such as Python and R, and employs TensorFlow and PyTorch libraries. The generated improvement measures are notified to the user via the terminal, enabling rapid response. 【0338】 As a concrete example, if an employee receives an inappropriate comment, the server analyzes the text data in real time and calculates a risk score. An alert is displayed on the terminal, and the user takes appropriate action. An example of a prompt message would be: "Analyze workplace communication data, determine the risk of harassment, and propose specific improvement measures." 【0339】 The flow of the specific processing in Example 2 will be explained using Figure 13. 【0340】 Step 1: 【0341】 The server collects data from various communication devices within the workplace. Specifically, it retrieves emails, chat messages, and social media posts via APIs and protocols. At this stage, the input is log information from the communication platform, and the output is a collection of raw data such as text, images, and audio. 【0342】 Step 2: 【0343】 The server preprocesses the collected data. Text data is processed to remove unwanted information using regular expressions and filtering techniques. Image data is resized and prepared for face recognition using OpenCV, and audio data is denoised using FFmpeg. The input is the raw data obtained in step 1, and the output is a dataset prepared for analysis. 【0344】 Step 3: 【0345】 The server analyzes the pre-processed data. It extracts emotions from text data using natural language processing techniques and recognizes facial expressions from images using computer vision techniques. In speech analysis, an emotion engine extracts speech features. The input is the dataset from step 2, and the output is the analysis results regarding emotions, tone, and facial expressions. 【0346】 Step 4: 【0347】 The server generates improvement measures using a generative AI model. Taking the analysis results as input, the generative AI model generates relevant improvement action suggestions. For example, if the risk score is high, it might suggest specific training or counseling. The output is a concrete action plan. 【0348】 Step 5: 【0349】 The terminal notifies the user of improvement suggestions received from the server. Information is displayed in a user-friendly format through templates and dashboards. The input is the action plan generated in step 4, and the output is an easily understandable information display. The user reviews this notification and decides on specific actions to implement in the workplace. 【0350】 (Application Example 2) 【0351】 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." 【0352】 Preventing harassment in the workplace is crucial, but traditional methods struggle to detect emotional changes and risks in real time. This can lead to delays in appropriate responses and a deterioration of the work environment. Therefore, there is a need for a system that can efficiently and quickly detect signs of harassment and enable responsible personnel to respond immediately. 【0353】 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. 【0354】 In this invention, the server includes means for collecting data to detect workplace harassment, means for analyzing the collected data to evaluate the degree of harassment, and means for generating specific improvement measures based on the evaluation results. This makes it possible to implement workplace environment improvement measures in a timely manner by sending an alert to the person in charge when the risk score exceeds a certain value. 【0355】 "Means of data collection" refers to devices or systems that acquire necessary information from various sources within the workplace, ensuring access rights through authentication and collecting diverse data formats such as text, images, and audio. 【0356】 "Methods for analyzing and evaluating the degree of harassment" refers to a process of analyzing collected data using algorithms and AI to quantify the possibility of harassment based on the tone of conversation, facial expressions, and gestures, and then evaluating its degree. 【0357】 "Means for generating specific improvement measures" refers to a function in which, based on evaluation results, the AI ​​model proposes appropriate action plans and countermeasures, and presents concrete improvement plans to HR personnel and supervisors. 【0358】 "Means for notifying users of generated improvement measures" refers to an interface or system that informs users of proposed measures in an easy-to-understand manner, enabling users to take immediate action by providing notifications quickly and appropriately. 【0359】 A "risk score" is an index that quantifies the likelihood of harassment based on analyzed data, and it is used as a basis for determining the priority of responses. 【0360】 The "means of sending alerts to responsible parties" refer to a system that sends real-time warnings to relevant parties when the risk score is high, based on established criteria, thereby encouraging a swift response. 【0361】 The server is equipped with means to collect various types of data generated within the workplace. Specifically, it retrieves data from sources such as email, instant messages, and voice calls via APIs and dedicated protocols. During this process, an authentication process is used to ensure data access rights and maintain security. 【0362】 Next, the server runs a program to analyze the collected data. This program includes a natural language processing engine and emotion recognition algorithms, implemented using frameworks such as TensorFlow and Keras in Python. NLTK is used for analyzing text data, while OpenCV and Dlib are used for analyzing image and video data. This allows for a detailed analysis of conversational tone and facial expressions, and the calculation of a harassment risk score. 【0363】 The terminal receives the risk score and specific corrective actions generated by the server and notifies the responsible person. Firebase Cloud Messaging is used to send this notification, ensuring that alerts are delivered quickly and reliably. Users, i.e., HR personnel, can then review the details from this notification and quickly take the necessary actions according to the corrective actions. 【0364】 For example, if an employee receives an offensive remark during a weekly meeting, the server immediately performs a sentiment analysis of the conversation, calculates a risk score, and generates appropriate corrective measures. In this case, an alert is immediately sent to the HR department, and specific actions to resolve the problem are proposed. 【0365】 An example of an input prompt for a generative AI model is: "Analyze the conversation log to identify inappropriate expressions and determine if further follow-up is needed." 【0366】 The flow of a specific process in Application Example 2 will be explained using Figure 14. 【0367】 Step 1: 【0368】 The server collects data from workplace communication tools and email systems. Inputs include various data sources accessed via APIs and protocols. Output is raw data obtained in text, audio, and image data formats. At this stage, security is ensured through an authentication process, and permissions for data retrieval are verified. 【0369】 Step 2: 【0370】 The server preprocesses the collected data. The input is the raw data obtained in step 1. The output is data processed into a format suitable for analysis. Specifically, it removes noise from text data, adjusts the resolution of image data, and clarifies audio data. 【0371】 Step 3: 【0372】 The server analyzes data using natural language processing and computer vision technologies. Inputs include pre-processed text, audio, and image data. Outputs are harassment risk scores based on emotional tone and facial expression data obtained through sentiment analysis. At this stage, emotion models using TensorFlow and Keras, and facial expression analysis using OpenCV and Dlib are implemented. 【0373】 Step 4: 【0374】 The server generates specific improvement measures based on the analyzed data. The input is the risk score calculated in step 3. The output is a list of improvement measures and action suggestions generated by the AI ​​model. The generating AI model is used to output improvement suggestions based on specific prompts. 【0375】 Step 5: 【0376】 The terminal notifies the responsible person of the generated improvement plan. The input is the improvement plan and alert information sent from the server. The output is the improvement plan and warning message displayed on the responsible person's device. Firebase Cloud Messaging is used to send notifications and deliver information to the responsible person in real time. 【0377】 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. 【0378】 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. 【0379】 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. 【0380】 [Third Embodiment] 【0381】 Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment. 【0382】 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. 【0383】 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). 【0384】 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. 【0385】 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. 【0386】 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). 【0387】 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. 【0388】 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. 【0389】 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. 【0390】 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. 【0391】 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. 【0392】 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". 【0393】 This invention is a system for effectively detecting and improving workplace harassment issues. This system aims to improve the workplace environment through a process of collecting and analyzing data and generating specific improvement measures. 【0394】 The server first collects data from internal workplace communication tools, external social networking services (SNS), email systems, and other sources. It utilizes technologies such as APIs and crawlers to gather necessary data from various information sources. This collected data includes text, images, and audio, making it possible to centrally collect diverse forms of information. 【0395】 Next, the server analyzes the collected data. For text data, natural language processing techniques are used to tokenize and perform sentiment analysis to detect keywords and emotional expressions that constitute harassment. For image and video data, computer vision techniques are used to analyze specific behaviors and situations to detect abnormal behavior. 【0396】 Based on the analysis results, the server quantifies the degree of harassment and calculates a risk score. This score allows for a quantitative assessment of the potential level of problems. 【0397】 The server then generates specific improvement measures based on the risk score. This improvement is generated using a generative AI model, creating actionable guidelines such as "recommend participation in training to improve communication skills" or "suggest the introduction of specific management methods." 【0398】 Users are notified of these suggestions via their devices. The devices provide an interface that allows users to easily review the suggestions and implement improvements as needed. 【0399】 As a concrete example, the system periodically collects data from a company's internal chat system where a server is located. If it detects that a particular employee is emotionally conflicted or engaging in inappropriate conversation, measures to improve communication are automatically suggested to that employee. The HR personnel, as users of the system, can then review the content and take appropriate action. In this way, the present invention is a system that supports harassment response in corporate human resource management and enables the creation of a healthier and more diverse workplace environment. 【0400】 The following describes the processing flow. 【0401】 Step 1: 【0402】 The server collects data from both inside and outside the workplace through APIs and crawlers. This includes data from communication tools and email systems used by employees, as well as data from publicly available social media platforms. 【0403】 Step 2: 【0404】 The server preprocesses the collected raw data. This preprocessing includes filtering out unnecessary information, formatting conversion, and text cleaning, preparing the data for analysis. 【0405】 Step 3: 【0406】 The server uses natural language processing techniques to analyze text data. Specifically, it performs tokenization, sentiment analysis, and extraction of important keywords to identify language patterns that may indicate harassment. 【0407】 Step 4: 【0408】 The server applies computer vision technology to image and video data to analyze specific behaviors and situations. This process includes detecting abnormal behavior and analyzing the behavioral patterns of specific individuals. 【0409】 Step 5: 【0410】 The server calculates a harassment risk score based on the analysis results. This score is used to quantitatively evaluate the degree of problems for each employee and in specific situations. 【0411】 Step 6: 【0412】 The server generates specific improvement measures based on the calculated risk score. This process utilizes a generative AI model to derive training suggestions and management method recommendations for behavioral improvement. 【0413】 Step 7: 【0414】 The server notifies the user's terminal of the generated improvement suggestions. The terminal displays the received notification on its interface to make it easier for the user to review the suggestions. 【0415】 Step 8: 【0416】 Users review the improvement measures notified on their devices and take action as needed. For example, they might issue training instructions to the relevant employees or promptly follow up with managers. 【0417】 Step 9: 【0418】 The server monitors user feedback and the results of implemented improvements, accumulating this data to help continuously improve the system. 【0419】 (Example 1) 【0420】 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." 【0421】 Workplace harassment is a serious problem that has profound impacts on individuals and organizations as a whole. However, traditional methods have made it difficult to detect potential signs of harassment early and effectively provide concrete corrective measures. Furthermore, there has been a lack of appropriate means for integrating and analyzing various data formats and generating dynamic and individualized corrective measures. 【0422】 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. 【0423】 In this invention, the server includes means for collecting information, means for analyzing the collected information to evaluate the degree of harassment, means for quantifying the risk based on the analysis results, and means for dynamically generating improvement measures using generative AI technology. This enables early detection of harassment and the provision of individualized and effective improvement measures. 【0424】 "Means of information gathering" refers to the function of acquiring data such as text, images, and audio from various sources both inside and outside the workplace. 【0425】 "Means of analyzing information to assess the degree of harassment" refers to a function that analyzes collected information, identifies signs of harassment, and assesses their severity. 【0426】 "Means for generating specific improvement measures" refers to a function for creating effective action guidelines and measures to improve harassment based on the analysis results. 【0427】 "Means of notifying users" refers to functions that provide users with generated improvement measures and analysis results, and effectively communicate necessary information. 【0428】 "Means of quantifying risk" refers to a function for quantitatively evaluating the potential risk of harassment and expressing it numerically. 【0429】 "Means for dynamically generating improvement measures using generative AI technology" refers to a function that utilizes artificial intelligence technology to automatically generate individualized and optimal improvement measures in response to analysis results. 【0430】 This invention is designed as a system for effectively detecting and addressing workplace harassment issues. The system primarily consists of servers and terminals and utilizes various hardware and software technologies. 【0431】 The server collects data from workplace communication tools, external social networking services (SNS), email systems, and other sources. By utilizing APIs (Application Programming Interfaces) and web crawler technologies, it's possible to centralize various forms of information during data collection. The collected data primarily includes text, images, and audio. 【0432】 The collected data is analyzed on a server. Natural language processing techniques are used for the analysis, and text data undergoes tokenization and sentiment analysis. This allows for the detection of keywords and emotional expressions related to harassment. In addition, computer vision techniques are used to analyze image and audio data to identify specific behaviors and situations, thereby detecting abnormal behavior. 【0433】 The analysis results are used to quantify the degree of harassment and calculate a risk score. Based on this score, the server generates appropriate improvement measures. A generative AI model is used to generate improvement measures, creating specific action guidelines such as "recommend participation in training to improve communication skills" or "suggest the introduction of a specific management method." An example of a prompt used here is the text, "Generate suggestions for improving communication in the workplace." 【0434】 The generated improvement measures are notified to the user via a terminal. The terminal provides an interface that allows the user to easily review the suggestions and implement the improvements as needed. In this way, the system actively supports harassment response in corporate human resources management, enabling employees to create a healthier and more diverse workplace environment. 【0435】 As a concrete example, a server periodically collects data from a company's chat system and detects emotional conflicts or inappropriate conversations between specific employees. As a result, communication improvement measures are automatically suggested to the employees in question, and HR personnel, as users, can review the content and take appropriate action. In this way, the present invention plays an important role in improving workplace health. 【0436】 The flow of the specific processing in Example 1 will be explained using Figure 11. 【0437】 Step 1: 【0438】 The server begins collecting information. Inputs include workplace communication tools, external social networking services (SNS), and email systems, from which data is retrieved using APIs and web crawler technologies. Outputs include text, image, and audio data, which are aggregated in the server's database. Specifically, the server accesses each information source at scheduled times, downloads the contents, and stores them. 【0439】 Step 2: 【0440】 The server analyzes the collected data. Input includes various data formats collected in Step 1. Output provides sentiment scores and abnormal behavior detection results as analysis results. For text data, tokenization and sentiment analysis are performed using natural language processing techniques to extract sentiment indicators and harassment-related keywords. For image and audio data, specific behaviors and alarming situations are identified through computer vision and audio analysis. In practice, the server processes data sequentially and generates analysis results using an analysis library. 【0441】 Step 3: 【0442】 The server uses the analysis results to calculate a harassment risk score. The emotion score and anomaly detection data obtained in step 2 are used as input. The output generates a risk score, quantifying potential problems. This enables the server to evaluate risk levels based on quantified indicators. 【0443】 Step 4: 【0444】 The server generates specific improvement measures based on the risk score. A generative AI model is used in this process. A risk score is given as input, and specific suggestions such as "We recommend participation in training to improve communication skills" are generated as output. The prompt text used as input to the generative AI model is "Generate suggestions for improving communication in the workplace." The server registers the generated suggestions in the database. 【0445】 Step 5: 【0446】 The terminal notifies the user of the generated improvement measures. The input is improvement measures provided from the server. The output is a notification message displayed on the interface for the user. Specifically, the terminal sends a notification through its alert function, allowing the user to immediately review the content and take appropriate action. 【0447】 (Application Example 1) 【0448】 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." 【0449】 In the workplace, harassment is a serious problem that negatively impacts the health and efficiency of an organization. Conventional methods make it difficult to detect and address harassment, often leading to delays in responding once the problem becomes apparent. This invention aims to solve these problems and promote improvement of the workplace environment. 【0450】 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. 【0451】 In this invention, the server includes means for collecting data to detect workplace harassment, means for analyzing the collected data to evaluate the degree of harassment, means for generating specific improvement measures based on the evaluation results, means for notifying the user of the generated improvement measures, and means installed on a smartphone that analyzes text data in real time, determines the harassment risk, and sends a notification. This enables early detection of harassment and the implementation of appropriate countermeasures. 【0452】 A "smartphone" is a portable electronic device equipped with communication capabilities that can run applications. 【0453】 "Installed" means to incorporate software into a computer or electronic device and make it usable. 【0454】 "Real-time" refers to the property of performing data processing or analysis at the moment it is requested, or almost instantly. 【0455】 "Text data" refers to digital information composed of characters, and is data stored in the form of sentences or strings. 【0456】 "Analysis" is the process of breaking down collected information and data into its details, understanding them, and extracting meaning and patterns. 【0457】 "Harassment risk" is an indicator that shows the degree to which harassment is likely to occur in the workplace. 【0458】 A "notification" is a communication or alert that conveys specific information to a recipient. 【0459】 The system implementing this invention is for detecting workplace harassment and proposing countermeasures. The server collects data from workplace communication tools and external information sources. The collected data includes text, images, and audio. This allows for the centralized handling of diverse information. 【0460】 The data analysis utilizes natural language processing techniques, specifically NLTK and spaCy, to analyze text data. During the analysis process, tokenization and sentiment analysis are performed to detect harassment-related keywords and emotional expressions. A smartphone is used as the hardware, and the application runs on it. Furthermore, OpenCV is used as computer vision technology to analyze image data and detect specific behaviors. Based on the analysis results, the harassment risk is quantified, and a risk score is calculated. This score indicates the potential degree of the problem. 【0461】 By utilizing a generative AI model, we propose specific improvement measures. This involves using OpenAI's GPT as the generative AI model. For example, it might generate a specific action plan for the user, such as "We recommend participating in a workshop to improve communication skills." This suggestion is sent as a notification to the user's smartphone, allowing them to immediately review and implement the improvement measures. 【0462】 For example, if a chat between employees is deemed to have a high risk of harassment, a notification will be sent to the employee's smartphone stating, "We have detected conflict and negative emotions in recent conversations. Please consider participating in a communication workshop to improve the situation." An example of a prompt for the generating AI model would be, "Detect conflict and negative emotions in employees' recent conversations and suggest ways to improve the situation." This system promotes a healthier workplace environment while simultaneously prompting employees to become more aware of the need for improvement. 【0463】 The flow of a specific process in Application Example 1 will be explained using Figure 12. 【0464】 Step 1: 【0465】 The server collects data from workplace communication tools and external information sources. Inputs include email, chat, and social media data, which are retrieved using APIs and crawlers. Outputs are aggregated datasets of text, images, and audio data. This allows for the effective centralization of data from diverse sources. 【0466】 Step 2: 【0467】 The server analyzes the collected data. The input is the dataset collected in the previous step. For text data, tokenization is performed using natural language processing techniques (e.g., NLTK, spaCy), and sentiment analysis is conducted. The output is the analysis result, which includes keywords and emotional expressions necessary for harassment determination. This extracts the characteristics of the communication content. 【0468】 Step 3: 【0469】 The server uses computer vision technology on image data. The input is the collected image data. Using OpenCV, it executes a behavior identification algorithm to detect abnormal behavior. The output is an analysis showing the possibility of harassment behavior. This allows for the assessment of harassment risk from visual information. 【0470】 Step 4: 【0471】 The server calculates a harassment risk score based on the analysis results. The input consists of analysis results obtained from text and image data. Natural language processing and computer vision analysis results are integrated to quantify the degree of risk. The output is a specific risk score, making it possible to quantitatively demonstrate the severity of the problem. 【0472】 Step 5: 【0473】 The server proposes specific improvement measures using a generative AI model. The inputs are risk scores and analysis results. Using a generative AI model (e.g., OpenAI GPT), it generates actionable guidelines for improving the workplace environment. The output is a proposal for specific improvement measures. This provides effective countermeasures against risks. 【0474】 Step 6: 【0475】 The device notifies the user of suggested improvements. The input is the suggested improvements received from the server. The notification is displayed to the user through an appropriate interface on their smartphone. The output is an information display confirming the action guidelines. The user can then immediately take action after reviewing the suggestions. 【0476】 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. 【0477】 This invention provides a system for detecting and preventing workplace harassment, and by incorporating an emotion engine that recognizes user emotions, it enables more precise analysis and the provision of improvement measures. 【0478】 The server first collects data from multiple sources, such as workplace communication tools, email, and social media. This stage involves securing data access rights through an authentication process and appropriately obtaining the necessary information. The data includes various formats, such as text, images, and audio. 【0479】 Next, the collected data is preprocessed on the server. This preprocessing includes text filtering, image resolution adjustment, and audio clearing. The server analyzes the text data using natural language processing techniques, particularly sentiment analysis, to extract conversational tones and key phrases. It also utilizes computer vision techniques to analyze image and video data, determining emotions from facial expressions and gestures. 【0480】 Furthermore, the server uses an emotion engine to directly recognize emotions from the user's voice and facial expression data. This, combined with a harassment risk score based on the analysis data, enables more accurate situational analysis. 【0481】 Based on the risk score calculated from the analyzed data, the server generates specific improvement measures. This generation utilizes an AI model, suggesting concrete actions such as "proposing individual counseling" or "recommending participation in a course on more appropriate communication techniques." 【0482】 The terminal notifies the user of the generated improvement suggestions. The suggestions are displayed on the user's terminal in an easy-to-understand interface, allowing for quick review and action. Based on these suggestions, the user can decide on and implement appropriate behavioral guidance and training in the workplace. 【0483】 As a concrete example, if an employee receives offensive remarks during work, the server recognizes this change in real time through its emotion engine and alerts the system to the possibility of harassment. Subsequently, a human resources representative (user) can check the alert on their terminal and quickly take appropriate action against the employee in question. In this way, the present invention is a system that enables a rapid and accurate response to harassment issues in corporate human resources management and supports the maintenance of a better work environment. 【0484】 The following describes the processing flow. 【0485】 Step 1: 【0486】 The server collects data from workplace communication platforms, email, and social networking services using APIs or crawling technologies. This collection includes a user authentication process to ensure data access is limited to authorized users. 【0487】 Step 2: 【0488】 The server preprocesses the collected data. This process involves noise reduction, text cleaning, and format conversion of image and audio data to prepare the data for analysis. 【0489】 Step 3: 【0490】 The server uses natural language processing technology to analyze text data and perform sentiment analysis. During this process, it detects emotional expressions and keywords related to potential harassment within the text. 【0491】 Step 4: 【0492】 The server uses computer vision technology to analyze image and video data, particularly recognizing emotions from the user's facial expressions and movements. This process identifies abnormal behavior and changes in emotions. 【0493】 Step 5: 【0494】 Using an emotion engine, the server further identifies the user's emotional state from the voice data and monitors emotional changes in real time. This data is integrated with other emotional data. 【0495】 Step 6: 【0496】 The server integrates these analysis results with sentiment data to score the harassment risk. This score assesses the potential risk of the problem and generates an alert if it exceeds a threshold. 【0497】 Step 7: 【0498】 The server uses a generative AI model to design specific improvement measures based on the harassment risk score. These proposed measures may include individual counseling or participation in educational programs. 【0499】 Step 8: 【0500】 The server sends the generated improvement suggestions to the terminal, and the terminal displays the suggestions on the user interface, allowing the user to quickly review the content. 【0501】 Step 9: 【0502】 Users review the improvement suggestions presented on their devices and, if necessary, implement training and follow-up actions for the relevant employees. They also provide feedback back to the server, contributing to the continuous improvement of the system. 【0503】 (Example 2) 【0504】 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." 【0505】 The challenge lies in maintaining healthy workplace relationships and protecting employees' mental health by early detection of harassment in the workplace and promptly providing appropriate corrective measures. Furthermore, conventional systems may be insufficient in detection, or corrective measures may be too general and lack specificity, thus requiring more precise analysis and individualized responses. 【0506】 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. 【0507】 In this invention, the server includes means for collecting data from communication devices, means for preprocessing and analyzing the collected data in various formats, and means for generating improvement measures using a generative AI model. This makes it possible to analyze workplace communication from multiple perspectives and quickly propose specific and effective improvement measures. 【0508】 "Communication equipment" refers to devices used to send and receive digital data both within and outside the workplace. 【0509】 "Means of data collection" refers to the methods and techniques for obtaining necessary data from various sources within the workplace. 【0510】 "Data in various formats" refers to information that includes different data formats such as text, images, and audio. 【0511】 "Preprocessing" refers to the process of preparing collected data to make it suitable for analysis. 【0512】 A "generative AI model" is a computer program that uses machine learning algorithms to create new information or suggestions. 【0513】 "Means for generating improvement measures" refers to the processes and techniques for constructing action plans and countermeasures based on analysis results. 【0514】 "Means of notifying a display device" refers to a method of displaying the generated information in a visual or audio format so that the user can confirm it. 【0515】 A description of the embodiment for carrying out the invention will be provided. 【0516】 This system is designed as a solution for detecting and preventing workplace harassment. The server collects data from communication devices used within the workplace, utilizing communication platform APIs and email server protocols. The specific hardware consists of cloud servers and local servers, while the software uses scripts and programs for data collection. 【0517】 The data includes various formats, and the server preprocesses them. Text data undergoes sentiment analysis using natural language processing techniques. The software used includes natural language processing models such as BERT and GPT. Image data is analyzed using OpenCV, and audio data is cleared using FFmpeg. 【0518】 The server utilizes a generative AI model to generate specific improvement measures based on the analysis results. This model is built using programming languages ​​such as Python and R, and employs TensorFlow and PyTorch libraries. The generated improvement measures are notified to the user via the terminal, enabling rapid response. 【0519】 As a concrete example, if an employee receives an inappropriate comment, the server analyzes the text data in real time and calculates a risk score. An alert is displayed on the terminal, and the user takes appropriate action. An example of a prompt message would be: "Analyze workplace communication data, determine the risk of harassment, and propose specific improvement measures." 【0520】 The flow of the specific processing in Example 2 will be explained using Figure 13. 【0521】 Step 1: 【0522】 The server collects data from various communication devices within the workplace. Specifically, it retrieves emails, chat messages, and social media posts via APIs and protocols. At this stage, the input is log information from the communication platform, and the output is a collection of raw data such as text, images, and audio. 【0523】 Step 2: 【0524】 The server preprocesses the collected data. Text data is processed to remove unwanted information using regular expressions and filtering techniques. Image data is resized and prepared for face recognition using OpenCV, and audio data is denoised using FFmpeg. The input is the raw data obtained in step 1, and the output is a dataset prepared for analysis. 【0525】 Step 3: 【0526】 The server analyzes the pre-processed data. It extracts emotions from text data using natural language processing techniques and recognizes facial expressions from images using computer vision techniques. In speech analysis, an emotion engine extracts speech features. The input is the dataset from step 2, and the output is the analysis results regarding emotions, tone, and facial expressions. 【0527】 Step 4: 【0528】 The server generates improvement measures using a generative AI model. Taking the analysis results as input, the generative AI model generates relevant improvement action suggestions. For example, if the risk score is high, it might suggest specific training or counseling. The output is a concrete action plan. 【0529】 Step 5: 【0530】 The terminal notifies the user of improvement suggestions received from the server. Information is displayed in a user-friendly format through templates and dashboards. The input is the action plan generated in step 4, and the output is an easily understandable information display. The user reviews this notification and decides on specific actions to implement in the workplace. 【0531】 (Application Example 2) 【0532】 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." 【0533】 Preventing harassment in the workplace is crucial, but traditional methods struggle to detect emotional changes and risks in real time. This can lead to delays in appropriate responses and a deterioration of the work environment. Therefore, there is a need for a system that can efficiently and quickly detect signs of harassment and enable responsible personnel to respond immediately. 【0534】 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. 【0535】 In this invention, the server includes means for collecting data to detect workplace harassment, means for analyzing the collected data to evaluate the degree of harassment, and means for generating specific improvement measures based on the evaluation results. This makes it possible to implement workplace environment improvement measures in a timely manner by sending an alert to the person in charge when the risk score exceeds a certain value. 【0536】 "Means of data collection" refers to devices or systems that acquire necessary information from various sources within the workplace, ensuring access rights through authentication and collecting diverse data formats such as text, images, and audio. 【0537】 "Methods for analyzing and evaluating the degree of harassment" refers to a process of analyzing collected data using algorithms and AI to quantify the possibility of harassment based on the tone of conversation, facial expressions, and gestures, and then evaluating its degree. 【0538】 "Means for generating specific improvement measures" refers to a function in which, based on evaluation results, the AI ​​model proposes appropriate action plans and countermeasures, and presents concrete improvement plans to HR personnel and supervisors. 【0539】 "Means for notifying users of generated improvement measures" refers to an interface or system that informs users of proposed measures in an easy-to-understand manner, enabling users to take immediate action by providing notifications quickly and appropriately. 【0540】 A "risk score" is an index that quantifies the likelihood of harassment based on analyzed data, and it is used as a basis for determining the priority of responses. 【0541】 The "means of sending alerts to responsible parties" refer to a system that sends real-time warnings to relevant parties when the risk score is high, based on established criteria, thereby encouraging a swift response. 【0542】 The server is equipped with means to collect various types of data generated within the workplace. Specifically, it retrieves data from sources such as email, instant messages, and voice calls via APIs and dedicated protocols. During this process, an authentication process is used to ensure data access rights and maintain security. 【0543】 Next, the server runs a program to analyze the collected data. This program includes a natural language processing engine and emotion recognition algorithms, implemented using frameworks such as TensorFlow and Keras in Python. NLTK is used for analyzing text data, while OpenCV and Dlib are used for analyzing image and video data. This allows for a detailed analysis of conversational tone and facial expressions, and the calculation of a harassment risk score. 【0544】 The terminal receives the risk score and specific corrective actions generated by the server and notifies the responsible person. Firebase Cloud Messaging is used to send this notification, ensuring that alerts are delivered quickly and reliably. Users, i.e., HR personnel, can then review the details from this notification and quickly take the necessary actions according to the corrective actions. 【0545】 For example, if an employee receives an offensive remark during a weekly meeting, the server immediately performs a sentiment analysis of the conversation, calculates a risk score, and generates appropriate corrective measures. In this case, an alert is immediately sent to the HR department, and specific actions to resolve the problem are proposed. 【0546】 An example of an input prompt for a generative AI model is: "Analyze the conversation log to identify inappropriate expressions and determine if further follow-up is needed." 【0547】 The flow of a specific process in Application Example 2 will be explained using Figure 14. 【0548】 Step 1: 【0549】 The server collects data from workplace communication tools and email systems. Inputs include various data sources accessed via APIs and protocols. Output is raw data obtained in text, audio, and image data formats. At this stage, security is ensured through an authentication process, and permissions for data retrieval are verified. 【0550】 Step 2: 【0551】 The server preprocesses the collected data. The input is the raw data obtained in step 1. The output is data processed into a format suitable for analysis. Specifically, it removes noise from text data, adjusts the resolution of image data, and clarifies audio data. 【0552】 Step 3: 【0553】 The server analyzes data using natural language processing and computer vision technologies. Inputs include pre-processed text, audio, and image data. Outputs are harassment risk scores based on emotional tone and facial expression data obtained through sentiment analysis. At this stage, emotion models using TensorFlow and Keras, and facial expression analysis using OpenCV and Dlib are implemented. 【0554】 Step 4: 【0555】 The server generates specific improvement measures based on the analyzed data. The input is the risk score calculated in step 3. The output is a list of improvement measures and action suggestions generated by the AI ​​model. The generating AI model is used to output improvement suggestions based on specific prompts. 【0556】 Step 5: 【0557】 The terminal notifies the responsible person of the generated improvement plan. The input is the improvement plan and alert information sent from the server. The output is the improvement plan and warning message displayed on the responsible person's device. Firebase Cloud Messaging is used to send notifications and deliver information to the responsible person in real time. 【0558】 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. 【0559】 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. 【0560】 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. 【0561】 [Fourth Embodiment] 【0562】 Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment. 【0563】 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. 【0564】 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). 【0565】 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. 【0566】 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. 【0567】 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). 【0568】 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. 【0569】 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. 【0570】 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. 【0571】 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. 【0572】 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. 【0573】 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. 【0574】 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". 【0575】 This invention is a system for effectively detecting and improving workplace harassment issues. This system aims to improve the workplace environment through a process of collecting and analyzing data and generating specific improvement measures. 【0576】 The server first collects data from internal workplace communication tools, external social networking services (SNS), email systems, and other sources. It utilizes technologies such as APIs and crawlers to gather necessary data from various information sources. This collected data includes text, images, and audio, making it possible to centrally collect diverse forms of information. 【0577】 Next, the server analyzes the collected data. For text data, natural language processing techniques are used to tokenize and perform sentiment analysis to detect keywords and emotional expressions that constitute harassment. For image and video data, computer vision techniques are used to analyze specific behaviors and situations to detect abnormal behavior. 【0578】 Based on the analysis results, the server quantifies the degree of harassment and calculates a risk score. This score allows for a quantitative assessment of the potential level of problems. 【0579】 The server then generates specific improvement measures based on the risk score. This improvement is generated using a generative AI model, creating actionable guidelines such as "recommend participation in training to improve communication skills" or "suggest the introduction of specific management methods." 【0580】 Users are notified of these suggestions via their devices. The devices provide an interface that allows users to easily review the suggestions and implement improvements as needed. 【0581】 As a concrete example, the system periodically collects data from a company's internal chat system where a server is located. If it detects that a particular employee is emotionally conflicted or engaging in inappropriate conversation, measures to improve communication are automatically suggested to that employee. The HR personnel, as users of the system, can then review the content and take appropriate action. In this way, the present invention is a system that supports harassment response in corporate human resource management and enables the creation of a healthier and more diverse workplace environment. 【0582】 The following describes the processing flow. 【0583】 Step 1: 【0584】 The server collects data from both inside and outside the workplace through APIs and crawlers. This includes data from communication tools and email systems used by employees, as well as data from publicly available social media platforms. 【0585】 Step 2: 【0586】 The server preprocesses the collected raw data. This preprocessing includes filtering out unnecessary information, formatting conversion, and text cleaning, preparing the data for analysis. 【0587】 Step 3: 【0588】 The server uses natural language processing techniques to analyze text data. Specifically, it performs tokenization, sentiment analysis, and extraction of important keywords to identify language patterns that may indicate harassment. 【0589】 Step 4: 【0590】 The server applies computer vision technology to image and video data to analyze specific behaviors and situations. This process includes detecting abnormal behavior and analyzing the behavioral patterns of specific individuals. 【0591】 Step 5: 【0592】 The server calculates a harassment risk score based on the analysis results. This score is used to quantitatively evaluate the degree of problems for each employee and in specific situations. 【0593】 Step 6: 【0594】 The server generates specific improvement measures based on the calculated risk score. This process utilizes a generative AI model to derive training suggestions and management method recommendations for behavioral improvement. 【0595】 Step 7: 【0596】 The server notifies the user's terminal of the generated improvement suggestions. The terminal displays the received notification on its interface to make it easier for the user to review the suggestions. 【0597】 Step 8: 【0598】 Users review the improvement measures notified on their devices and take action as needed. For example, they might issue training instructions to the relevant employees or promptly follow up with managers. 【0599】 Step 9: 【0600】 The server monitors user feedback and the results of implemented improvements, accumulating this data to help continuously improve the system. 【0601】 (Example 1) 【0602】 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". 【0603】 Workplace harassment is a serious problem that has profound impacts on individuals and organizations as a whole. However, traditional methods have made it difficult to detect potential signs of harassment early and effectively provide concrete corrective measures. Furthermore, there has been a lack of appropriate means for integrating and analyzing various data formats and generating dynamic and individualized corrective measures. 【0604】 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. 【0605】 In this invention, the server includes means for collecting information, means for analyzing the collected information to evaluate the degree of harassment, means for quantifying the risk based on the analysis results, and means for dynamically generating improvement measures using generative AI technology. This enables early detection of harassment and the provision of individualized and effective improvement measures. 【0606】 "Means of information gathering" refers to the function of acquiring data such as text, images, and audio from various sources both inside and outside the workplace. 【0607】 "Means of analyzing information to assess the degree of harassment" refers to a function that analyzes collected information, identifies signs of harassment, and assesses their severity. 【0608】 "Means for generating specific improvement measures" refers to a function for creating effective action guidelines and measures to improve harassment based on the analysis results. 【0609】 "Means of notifying users" refers to functions that provide users with generated improvement measures and analysis results, and effectively communicate necessary information. 【0610】 "Means of quantifying risk" refers to a function for quantitatively evaluating the potential risk of harassment and expressing it numerically. 【0611】 "Means for dynamically generating improvement measures using generative AI technology" refers to a function that utilizes artificial intelligence technology to automatically generate individualized and optimal improvement measures in response to analysis results. 【0612】 This invention is designed as a system for effectively detecting and addressing workplace harassment issues. The system primarily consists of servers and terminals and utilizes various hardware and software technologies. 【0613】 The server collects data from workplace communication tools, external social networking services (SNS), email systems, and other sources. By utilizing APIs (Application Programming Interfaces) and web crawler technologies, it's possible to centralize various forms of information during data collection. The collected data primarily includes text, images, and audio. 【0614】 The collected data is analyzed on a server. Natural language processing techniques are used for the analysis, and text data undergoes tokenization and sentiment analysis. This allows for the detection of keywords and emotional expressions related to harassment. In addition, computer vision techniques are used to analyze image and audio data to identify specific behaviors and situations, thereby detecting abnormal behavior. 【0615】 The analysis results are used to quantify the degree of harassment and calculate a risk score. Based on this score, the server generates appropriate improvement measures. A generative AI model is used to generate improvement measures, creating specific action guidelines such as "recommend participation in training to improve communication skills" or "suggest the introduction of a specific management method." An example of a prompt used here is the text, "Generate suggestions for improving communication in the workplace." 【0616】 The generated improvement measures are notified to the user via a terminal. The terminal provides an interface that allows the user to easily review the suggestions and implement the improvements as needed. In this way, the system actively supports harassment response in corporate human resources management, enabling employees to create a healthier and more diverse workplace environment. 【0617】 As a concrete example, a server periodically collects data from a company's chat system and detects emotional conflicts or inappropriate conversations between specific employees. As a result, communication improvement measures are automatically suggested to the employees in question, and HR personnel, as users, can review the content and take appropriate action. In this way, the present invention plays an important role in improving workplace health. 【0618】 The flow of the specific processing in Example 1 will be explained using Figure 11. 【0619】 Step 1: 【0620】 The server begins collecting information. Inputs include workplace communication tools, external social networking services (SNS), and email systems, from which data is retrieved using APIs and web crawler technologies. Outputs include text, image, and audio data, which are aggregated in the server's database. Specifically, the server accesses each information source at scheduled times, downloads the contents, and stores them. 【0621】 Step 2: 【0622】 The server analyzes the collected data. Input includes various data formats collected in Step 1. Output provides sentiment scores and abnormal behavior detection results as analysis results. For text data, tokenization and sentiment analysis are performed using natural language processing techniques to extract sentiment indicators and harassment-related keywords. For image and audio data, specific behaviors and alarming situations are identified through computer vision and audio analysis. In practice, the server processes data sequentially and generates analysis results using an analysis library. 【0623】 Step 3: 【0624】 The server uses the analysis results to calculate a harassment risk score. The emotion score and anomaly detection data obtained in step 2 are used as input. The output generates a risk score, quantifying potential problems. This enables the server to evaluate risk levels based on quantified indicators. 【0625】 Step 4: 【0626】 The server generates specific improvement measures based on the risk score. A generative AI model is used in this process. A risk score is given as input, and specific suggestions such as "We recommend participation in training to improve communication skills" are generated as output. The prompt text used as input to the generative AI model is "Generate suggestions for improving communication in the workplace." The server registers the generated suggestions in the database. 【0627】 Step 5: 【0628】 The terminal notifies the user of the generated improvement measures. The input is improvement measures provided from the server. The output is a notification message displayed on the interface for the user. Specifically, the terminal sends a notification through its alert function, allowing the user to immediately review the content and take appropriate action. 【0629】 (Application Example 1) 【0630】 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". 【0631】 In the workplace, harassment is a serious problem that negatively impacts the health and efficiency of an organization. Conventional methods make it difficult to detect and address harassment, often leading to delays in responding once the problem becomes apparent. This invention aims to solve these problems and promote improvement of the workplace environment. 【0632】 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. 【0633】 In this invention, the server includes means for collecting data to detect workplace harassment, means for analyzing the collected data to evaluate the degree of harassment, means for generating specific improvement measures based on the evaluation results, means for notifying the user of the generated improvement measures, and means installed on a smartphone that analyzes text data in real time, determines the harassment risk, and sends a notification. This enables early detection of harassment and the implementation of appropriate countermeasures. 【0634】 A "smartphone" is a portable electronic device equipped with communication capabilities that can run applications. 【0635】 "Installed" means to incorporate software into a computer or electronic device and make it usable. 【0636】 "Real-time" refers to the property of performing data processing or analysis at the moment it is requested, or almost instantly. 【0637】 "Text data" refers to digital information composed of characters, and is data stored in the form of sentences or strings. 【0638】 "Analysis" is the process of breaking down collected information and data into its details, understanding them, and extracting meaning and patterns. 【0639】 "Harassment risk" is an indicator that shows the degree to which harassment is likely to occur in the workplace. 【0640】 A "notification" is a communication or alert that conveys specific information to a recipient. 【0641】 The system implementing this invention is for detecting workplace harassment and proposing countermeasures. The server collects data from workplace communication tools and external information sources. The collected data includes text, images, and audio. This allows for the centralized handling of diverse information. 【0642】 The data analysis utilizes natural language processing techniques, specifically NLTK and spaCy, to analyze text data. During the analysis process, tokenization and sentiment analysis are performed to detect harassment-related keywords and emotional expressions. A smartphone is used as the hardware, and the application runs on it. Furthermore, OpenCV is used as computer vision technology to analyze image data and detect specific behaviors. Based on the analysis results, the harassment risk is quantified, and a risk score is calculated. This score indicates the potential degree of the problem. 【0643】 By utilizing a generative AI model, we propose specific improvement measures. This involves using OpenAI's GPT as the generative AI model. For example, it might generate a specific action plan for the user, such as "We recommend participating in a workshop to improve communication skills." This suggestion is sent as a notification to the user's smartphone, allowing them to immediately review and implement the improvement measures. 【0644】 For example, if a chat between employees is deemed to have a high risk of harassment, a notification will be sent to the employee's smartphone stating, "We have detected conflict and negative emotions in recent conversations. Please consider participating in a communication workshop to improve the situation." An example of a prompt for the generating AI model would be, "Detect conflict and negative emotions in employees' recent conversations and suggest ways to improve the situation." This system promotes a healthier workplace environment while simultaneously prompting employees to become more aware of the need for improvement. 【0645】 The flow of a specific process in Application Example 1 will be explained using Figure 12. 【0646】 Step 1: 【0647】 The server collects data from workplace communication tools and external information sources. Inputs include email, chat, and social media data, which are retrieved using APIs and crawlers. Outputs are aggregated datasets of text, images, and audio data. This allows for the effective centralization of data from diverse sources. 【0648】 Step 2: 【0649】 The server analyzes the collected data. The input is the dataset collected in the previous step. For text data, tokenization is performed using natural language processing techniques (e.g., NLTK, spaCy), and sentiment analysis is conducted. The output is the analysis result, which includes keywords and emotional expressions necessary for harassment determination. This extracts the characteristics of the communication content. 【0650】 Step 3: 【0651】 The server uses computer vision technology on image data. The input is the collected image data. Using OpenCV, it executes a behavior identification algorithm to detect abnormal behavior. The output is an analysis showing the possibility of harassment behavior. This allows for the assessment of harassment risk from visual information. 【0652】 Step 4: 【0653】 The server calculates a harassment risk score based on the analysis results. The input consists of analysis results obtained from text and image data. Natural language processing and computer vision analysis results are integrated to quantify the degree of risk. The output is a specific risk score, making it possible to quantitatively demonstrate the severity of the problem. 【0654】 Step 5: 【0655】 The server proposes specific improvement measures using a generative AI model. The inputs are risk scores and analysis results. Using a generative AI model (e.g., OpenAI GPT), it generates actionable guidelines for improving the workplace environment. The output is a proposal for specific improvement measures. This provides effective countermeasures against risks. 【0656】 Step 6: 【0657】 The device notifies the user of suggested improvements. The input is the suggested improvements received from the server. The notification is displayed to the user through an appropriate interface on their smartphone. The output is an information display confirming the action guidelines. The user can then immediately take action after reviewing the suggestions. 【0658】 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. 【0659】 This invention provides a system for detecting and preventing workplace harassment, and by incorporating an emotion engine that recognizes user emotions, it enables more precise analysis and the provision of improvement measures. 【0660】 The server first collects data from multiple sources, such as workplace communication tools, email, and social media. This stage involves securing data access rights through an authentication process and appropriately obtaining the necessary information. The data includes various formats, such as text, images, and audio. 【0661】 Next, the collected data is preprocessed on the server. This preprocessing includes text filtering, image resolution adjustment, and audio clearing. The server analyzes the text data using natural language processing techniques, particularly sentiment analysis, to extract conversational tones and key phrases. It also utilizes computer vision techniques to analyze image and video data, determining emotions from facial expressions and gestures. 【0662】 Furthermore, the server uses an emotion engine to directly recognize emotions from the user's voice and facial expression data. This, combined with a harassment risk score based on the analysis data, enables more accurate situational analysis. 【0663】 Based on the risk score calculated from the analyzed data, the server generates specific improvement measures. This generation utilizes an AI model, suggesting concrete actions such as "proposing individual counseling" or "recommending participation in a course on more appropriate communication techniques." 【0664】 The terminal notifies the user of the generated improvement suggestions. The suggestions are displayed on the user's terminal in an easy-to-understand interface, allowing for quick review and action. Based on these suggestions, the user can decide on and implement appropriate behavioral guidance and training in the workplace. 【0665】 As a concrete example, if an employee receives offensive remarks during work, the server recognizes this change in real time through its emotion engine and alerts the system to the possibility of harassment. Subsequently, a human resources representative (user) can check the alert on their terminal and quickly take appropriate action against the employee in question. In this way, the present invention is a system that enables a rapid and accurate response to harassment issues in corporate human resources management and supports the maintenance of a better work environment. 【0666】 The following describes the processing flow. 【0667】 Step 1: 【0668】 The server collects data from workplace communication platforms, email, and social networking services using APIs or crawling technologies. This collection includes a user authentication process to ensure data access is limited to authorized users. 【0669】 Step 2: 【0670】 The server preprocesses the collected data. This process involves noise reduction, text cleaning, and format conversion of image and audio data to prepare the data for analysis. 【0671】 Step 3: 【0672】 The server uses natural language processing technology to analyze text data and perform sentiment analysis. During this process, it detects emotional expressions and keywords related to potential harassment within the text. 【0673】 Step 4: 【0674】 The server uses computer vision technology to analyze image and video data, particularly recognizing emotions from the user's facial expressions and movements. This process identifies abnormal behavior and changes in emotions. 【0675】 Step 5: 【0676】 Using an emotion engine, the server further identifies the user's emotional state from the voice data and monitors emotional changes in real time. This data is integrated with other emotional data. 【0677】 Step 6: 【0678】 The server integrates these analysis results with sentiment data to score the harassment risk. This score assesses the potential risk of the problem and generates an alert if it exceeds a threshold. 【0679】 Step 7: 【0680】 The server uses a generative AI model to design specific improvement measures based on the harassment risk score. These proposed measures may include individual counseling or participation in educational programs. 【0681】 Step 8: 【0682】 The server sends the generated improvement suggestions to the terminal, and the terminal displays the suggestions on the user interface, allowing the user to quickly review the content. 【0683】 Step 9: 【0684】 Users review the improvement suggestions presented on their devices and, if necessary, implement training and follow-up actions for the relevant employees. They also provide feedback back to the server, contributing to the continuous improvement of the system. 【0685】 (Example 2) 【0686】 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". 【0687】 The challenge lies in maintaining healthy workplace relationships and protecting employees' mental health by early detection of harassment in the workplace and promptly providing appropriate corrective measures. Furthermore, conventional systems may be insufficient in detection, or corrective measures may be too general and lack specificity, thus requiring more precise analysis and individualized responses. 【0688】 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. 【0689】 In this invention, the server includes means for collecting data from communication devices, means for preprocessing and analyzing the collected data in various formats, and means for generating improvement measures using a generative AI model. This makes it possible to analyze workplace communication from multiple perspectives and quickly propose specific and effective improvement measures. 【0690】 "Communication equipment" refers to devices used to send and receive digital data both within and outside the workplace. 【0691】 "Means of data collection" refers to the methods and techniques for obtaining necessary data from various sources within the workplace. 【0692】 "Data in various formats" refers to information that includes different data formats such as text, images, and audio. 【0693】 "Preprocessing" refers to the process of preparing collected data to make it suitable for analysis. 【0694】 A "generative AI model" is a computer program that uses machine learning algorithms to create new information or suggestions. 【0695】 "Means for generating improvement measures" refers to the processes and techniques for constructing action plans and countermeasures based on analysis results. 【0696】 "Means of notifying a display device" refers to a method of displaying the generated information in a visual or audio format so that the user can confirm it. 【0697】 A description of the embodiment for carrying out the invention will be provided. 【0698】 This system is designed as a solution for detecting and preventing workplace harassment. The server collects data from communication devices used within the workplace, utilizing communication platform APIs and email server protocols. The specific hardware consists of cloud servers and local servers, while the software uses scripts and programs for data collection. 【0699】 The data includes various formats, and the server preprocesses them. Text data undergoes sentiment analysis using natural language processing techniques. The software used includes natural language processing models such as BERT and GPT. Image data is analyzed using OpenCV, and audio data is cleared using FFmpeg. 【0700】 The server utilizes a generative AI model to generate specific improvement measures based on the analysis results. This model is built using programming languages ​​such as Python and R, and employs TensorFlow and PyTorch libraries. The generated improvement measures are notified to the user via the terminal, enabling rapid response. 【0701】 As a concrete example, if an employee receives an inappropriate comment, the server analyzes the text data in real time and calculates a risk score. An alert is displayed on the terminal, and the user takes appropriate action. An example of a prompt message would be: "Analyze workplace communication data, determine the risk of harassment, and propose specific improvement measures." 【0702】 The flow of the specific processing in Example 2 will be explained using Figure 13. 【0703】 Step 1: 【0704】 The server collects data from various communication devices within the workplace. Specifically, it retrieves emails, chat messages, and social media posts via APIs and protocols. At this stage, the input is log information from the communication platform, and the output is a collection of raw data such as text, images, and audio. 【0705】 Step 2: 【0706】 The server preprocesses the collected data. Text data is processed to remove unwanted information using regular expressions and filtering techniques. Image data is resized and prepared for face recognition using OpenCV, and audio data is denoised using FFmpeg. The input is the raw data obtained in step 1, and the output is a dataset prepared for analysis. 【0707】 Step 3: 【0708】 The server analyzes the pre-processed data. It extracts emotions from text data using natural language processing techniques and recognizes facial expressions from images using computer vision techniques. In speech analysis, an emotion engine extracts speech features. The input is the dataset from step 2, and the output is the analysis results regarding emotions, tone, and facial expressions. 【0709】 Step 4: 【0710】 The server generates improvement measures using a generative AI model. Taking the analysis results as input, the generative AI model generates relevant improvement action suggestions. For example, if the risk score is high, it might suggest specific training or counseling. The output is a concrete action plan. 【0711】 Step 5: 【0712】 The terminal notifies the user of improvement suggestions received from the server. Information is displayed in a user-friendly format through templates and dashboards. The input is the action plan generated in step 4, and the output is an easily understandable information display. The user reviews this notification and decides on specific actions to implement in the workplace. 【0713】 (Application Example 2) 【0714】 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". 【0715】 Preventing harassment in the workplace is crucial, but traditional methods struggle to detect emotional changes and risks in real time. This can lead to delays in appropriate responses and a deterioration of the work environment. Therefore, there is a need for a system that can efficiently and quickly detect signs of harassment and enable responsible personnel to respond immediately. 【0716】 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. 【0717】 In this invention, the server includes means for collecting data to detect workplace harassment, means for analyzing the collected data to evaluate the degree of harassment, and means for generating specific improvement measures based on the evaluation results. This makes it possible to implement workplace environment improvement measures in a timely manner by sending an alert to the person in charge when the risk score exceeds a certain value. 【0718】 "Means of data collection" refers to devices or systems that acquire necessary information from various sources within the workplace, ensuring access rights through authentication and collecting diverse data formats such as text, images, and audio. 【0719】 "Methods for analyzing and evaluating the degree of harassment" refers to a process of analyzing collected data using algorithms and AI to quantify the possibility of harassment based on the tone of conversation, facial expressions, and gestures, and then evaluating its degree. 【0720】 "Means for generating specific improvement measures" refers to a function in which, based on evaluation results, the AI ​​model proposes appropriate action plans and countermeasures, and presents concrete improvement plans to HR personnel and supervisors. 【0721】 "Means for notifying users of generated improvement measures" refers to an interface or system that informs users of proposed measures in an easy-to-understand manner, enabling users to take immediate action by providing notifications quickly and appropriately. 【0722】 A "risk score" is an index that quantifies the likelihood of harassment based on analyzed data, and it is used as a basis for determining the priority of responses. 【0723】 The "means of sending alerts to responsible parties" refer to a system that sends real-time warnings to relevant parties when the risk score is high, based on established criteria, thereby encouraging a swift response. 【0724】 The server is equipped with means to collect various types of data generated within the workplace. Specifically, it retrieves data from sources such as email, instant messages, and voice calls via APIs and dedicated protocols. During this process, an authentication process is used to ensure data access rights and maintain security. 【0725】 Next, the server runs a program to analyze the collected data. This program includes a natural language processing engine and emotion recognition algorithms, implemented using frameworks such as TensorFlow and Keras in Python. NLTK is used for analyzing text data, while OpenCV and Dlib are used for analyzing image and video data. This allows for a detailed analysis of conversational tone and facial expressions, and the calculation of a harassment risk score. 【0726】 The terminal receives the risk score and specific corrective actions generated by the server and notifies the responsible person. Firebase Cloud Messaging is used to send this notification, ensuring that alerts are delivered quickly and reliably. Users, i.e., HR personnel, can then review the details from this notification and quickly take the necessary actions according to the corrective actions. 【0727】 For example, if an employee receives an offensive remark during a weekly meeting, the server immediately performs a sentiment analysis of the conversation, calculates a risk score, and generates appropriate corrective measures. In this case, an alert is immediately sent to the HR department, and specific actions to resolve the problem are proposed. 【0728】 An example of an input prompt for a generative AI model is: "Analyze the conversation log to identify inappropriate expressions and determine if further follow-up is needed." 【0729】 The flow of a specific process in Application Example 2 will be explained using Figure 14. 【0730】 Step 1: 【0731】 The server collects data from workplace communication tools and email systems. Inputs include various data sources accessed via APIs and protocols. Output is raw data obtained in text, audio, and image data formats. At this stage, security is ensured through an authentication process, and permissions for data retrieval are verified. 【0732】 Step 2: 【0733】 The server preprocesses the collected data. The input is the raw data obtained in step 1. The output is data processed into a format suitable for analysis. Specifically, it removes noise from text data, adjusts the resolution of image data, and clarifies audio data. 【0734】 Step 3: 【0735】 The server analyzes data using natural language processing and computer vision technologies. Inputs include pre-processed text, audio, and image data. Outputs are harassment risk scores based on emotional tone and facial expression data obtained through sentiment analysis. At this stage, emotion models using TensorFlow and Keras, and facial expression analysis using OpenCV and Dlib are implemented. 【0736】 Step 4: 【0737】 The server generates specific improvement measures based on the analyzed data. The input is the risk score calculated in step 3. The output is a list of improvement measures and action suggestions generated by the AI ​​model. The generating AI model is used to output improvement suggestions based on specific prompts. 【0738】 Step 5: 【0739】 The terminal notifies the responsible person of the generated improvement plan. The input is the improvement plan and alert information sent from the server. The output is the improvement plan and warning message displayed on the responsible person's device. Firebase Cloud Messaging is used to send notifications and deliver information to the responsible person in real time. 【0740】 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. 【0741】 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. 【0742】 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. 【0743】 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. 【0744】 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. 【0745】 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. 【0746】 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. 【0747】 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. 【0748】 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." 【0749】 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. 【0750】 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. 【0751】 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. 【0752】 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. 【0753】 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. 【0754】 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. 【0755】 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. 【0756】 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. 【0757】 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. 【0758】 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. 【0759】 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. 【0760】 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 as being incorporated by reference. 【0761】 The following is further disclosed regarding the embodiments described above. 【0762】 (Claim 1) 【0763】 To detect workplace harassment, means of collecting data, 【0764】 A method for analyzing collected data to evaluate the degree of harassment, 【0765】 A means of generating specific improvement measures based on evaluation results, 【0766】 A means of notifying the user of the generated improvement measures, 【0767】 A system that includes this. 【0768】 (Claim 2) 【0769】 The system according to claim 1, characterized in that the data analysis means analyzes text data using natural language processing technology and performs sentiment analysis. 【0770】 (Claim 3) 【0771】 The system according to claim 1, characterized in that the data collection means includes an authentication process for obtaining data from multiple sources. 【0772】 "Example 1" 【0773】 (Claim 1) 【0774】 To detect workplace harassment, means of collecting information, 【0775】 A means of analyzing the collected information to evaluate the degree of harassment, 【0776】 A means of generating specific improvement measures based on evaluation results, 【0777】 A means of notifying the user of the generated improvement measures, 【0778】 A means of quantifying risk based on the analysis results, 【0779】 A means of dynamically generating improvement measures using generative AI technology, 【0780】 A system that includes this. 【0781】 (Claim 2) 【0782】 The system according to claim 1, characterized in that the information analysis means analyzes language data using natural language processing technology and performs sentiment analysis. 【0783】 (Claim 3) 【0784】 The system according to claim 1, characterized in that the information gathering means includes an authentication procedure for obtaining information from multiple sources. 【0785】 "Application Example 1" 【0786】 (Claim 1) 【0787】 To detect workplace harassment, means of collecting data, 【0788】 A method for analyzing collected data to evaluate the degree of harassment, 【0789】 A means of generating specific improvement measures based on evaluation results, 【0790】 A means of notifying the user of the generated improvement measures, 【0791】 A means of installing on a smartphone, analyzing text data in real time, determining harassment risk, and sending notifications, 【0792】 A system that includes this. 【0793】 (Claim 2) 【0794】 The system according to claim 1, characterized in that the data analysis means analyzes text data using natural language processing technology and performs sentiment analysis. 【0795】 (Claim 3) 【0796】 The system according to claim 1, characterized in that the data collection means includes an authentication process for obtaining data from multiple sources. 【0797】 "Example 2 of combining an emotion engine" 【0798】 (Claim 1) 【0799】 A means of collecting data from communication devices to detect workplace harassment, 【0800】 Means for preprocessing and analyzing the collected data in various formats, 【0801】 A means of generating improvement measures using a generative AI model based on the analyzed data, 【0802】 A means for notifying a display device of the generated improvement measures, 【0803】 A system that includes this. 【0804】 (Claim 2) 【0805】 The system according to claim 1, characterized in that it performs sentiment analysis using natural language processing technology and recognizes emotions from text data. 【0806】 (Claim 3) 【0807】 The system according to claim 1, characterized by including an authentication process for obtaining multiple data from diverse sources. 【0808】 "Application example 2 when combining with an emotional engine" 【0809】 (Claim 1) 【0810】 To detect workplace harassment, means of collecting data, 【0811】 A method for analyzing collected data to evaluate the degree of harassment, 【0812】 A means of generating specific improvement measures based on evaluation results, 【0813】 A means of notifying the user of the generated improvement measures, 【0814】 The platform has a means of sending an alert to the person in charge when the risk score exceeds a certain value, 【0815】 A system that includes this. 【0816】 (Claim 2) 【0817】 The system according to claim 1, characterized in that the data analysis means analyzes text data using natural language processing technology, performs sentiment analysis, and analyzes facial expressions using facial recognition technology. 【0818】 (Claim 3) 【0819】 The system according to claim 1, characterized in that the data collection means includes an authentication process for obtaining data from multiple sources and performs sentiment analysis in real time. [Explanation of symbols] 【0820】 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

[Claim 1] To detect workplace harassment, means of collecting data, A method for analyzing collected data to evaluate the degree of harassment, A means of generating specific improvement measures based on evaluation results, A means of notifying the user of the generated improvement measures, A system that includes this. [Claim 2] The system according to claim 1, characterized in that the data analysis means analyzes text data using natural language processing technology and performs sentiment analysis. [Claim 3] The system according to claim 1, characterized in that the data collection means includes an authentication process for obtaining data from multiple sources.

Images