System

Abstract

A system is provided.SOLUTION: A system comprising: means for performing pattern learning of a comment using a generative AI model; means for analyzing a newly input comment; means for automatically diagnosing whether or not the newly input comment corresponds to a harassment action based on an analysis result; and means for transmitting a diagnostic result to a user device.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Workplace harassment poses a wide range of serious risks, including worker dissatisfaction, increased turnover, legal issues, and reduced productivity. However, appropriate responses are difficult due to difficulties in recognizing and reporting harassment, differences in corporate culture, and misunderstandings. Furthermore, traditional training programs and policy development are costly and time-consuming, and difficult to implement effectively for all employees. The present invention aims to solve these problems and provide a rapid and objective diagnosis and prevention of harassment. [Means for solving the problem]

[0005] The present invention relates to a system that includes a means for learning patterns from comment data using a generative AI model, a means for analyzing newly entered comments and situational data, a means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, and a means for transmitting the diagnosis results to a user's device. The system also includes a means for storing expert preventative measures and case information in a database and using them for analysis and automatic diagnosis, and a means for generating appropriate educational programs based on newly entered comments and situational data and providing them to users. This enables quick and accurate diagnosis of harassment in the workplace and the provision of effective preventative measures and educational programs.

[0006] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to extract features from large amounts of data and make predictions and classifications for new data.

[0007] "Comment Data" means written feedback, opinions, ratings, or reports entered by users, including information relating to a particular topic or situation.

[0008] "Pattern learning" refers to the process of extracting patterns and features from data and using them to analyze or predict new data.

[0009] "Newly entered comments and context data" refers to new feedback, opinions, ratings, or reports received by the system, including information not already registered in any existing database.

[0010] "Analysis" refers to the process of using algorithms and models to evaluate input data for truth, meaning, emotion, category, etc.

[0011] "Automatic diagnosis" refers to the process by which a system automatically determines whether input data falls under certain conditions (such as harassment) based on patterns and rules it has previously learned.

[0012] "Expert-based preventive measures and case information" refers to recommendations provided by harassment prevention experts and information based on specific past cases.

[0013] A "database" refers to an electronic structure for efficiently storing, retrieving, and managing large amounts of data.

[0014] "Educational Program" refers to educational materials and training modules that teach employees about recognizing and preventing harassment.

[0015] "User Device" refers to the electronic device (e.g., computer, smartphone, tablet) used by a User to enter comments and situational data and receive and display diagnostic results and educational programs. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] The system of the present invention is designed to provide pattern learning of comment data using a generative AI model, an automatic diagnostic function that combines preventive measures by experts and case information, and an educational program. Specific embodiments for implementing this system are described below.

[0038] System configuration

[0039] server

[0040] The server has the following functions:

[0041] 1. Data collection function

[0042] 2. Data preprocessing function

[0043] 3. Pattern learning function

[0044] 4. Automatic diagnosis function

[0045] 5. Result transmission function

[0046] 6. Educational program provision function

[0047] Terminal

[0048] The terminal has the following functions:

[0049] 1. Comment and status data input function

[0050] 2. Data transmission function

[0051] 3. Diagnostic result display function

[0052] 4. Educational program display and progress tracking function

[0053] Service flow

[0054] Data collection

[0055] When users enter comments into internal surveys, chat systems, etc., the server continuously collects them. Prevention measures and case information provided by experts are also collected and stored in a database.

[0056] Data Preprocessing

[0057] The server cleanses, denoises, and normalizes the collected comment data, and then stores the pre-processed data back in the database.

[0058] Pattern Learning

[0059] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[0060] Automatic diagnosis

[0061] Users input new comments and situational data from their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[0062] Sending and displaying results

[0063] The server generates a diagnostic result and sends it to the user's terminal, which displays the diagnostic result to the user.

[0064] Providing educational programs

[0065] Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through the terminal, and the progress is reported to the server.

[0066] Specific examples

[0067] Example 1: Comments about appearance

[0068] The user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which then analyzes the comment using a generative AI model. The analysis results in a diagnosis that "these are negative comments about my appearance, and are likely to constitute harassment." The results are displayed on the user's device, and appropriate educational programs are provided.

[0069] Example 2: Ignoring comments

[0070] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which then analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an intentional attempt to exclude certain individuals, and is likely to constitute harassment." The results are displayed on the user's device, and an appropriate educational program is provided.

[0071] This allows users and companies to quickly and accurately diagnose harassment and take appropriate measures. Furthermore, education programs can be used to continuously educate employees and improve the work environment.

[0072] The processing flow will be explained below.

[0073] Step 1: Data collection

[0074] The server continuously collects comments from various comment data sources (internal surveys, internal chats, review sites, etc.) and stores the collected comment data, along with preventive measures and case information from experts, in a database.

[0075] Step 2: Data Preprocessing

[0076] The server retrieves raw comment data from the database, performs noise removal (e.g., correcting emotional words and typos), performs text normalization (e.g., synonym unification, text segmentation), and stores the preprocessed data back in the database.

[0077] Step 3: Pattern learning

[0078] The server retrieves preprocessed comment data from the database. It uses this comment data to train a generative AI model (e.g., BERT or GPT-4). The model learns patterns of harassing behavior. It saves the trained model, making it available for prediction.

[0079] Step 4: Accepting input data

[0080] The user uses the terminal to enter comments and situation data into the input form, and then presses the "Send" button to send the data to the server.

[0081] Step 5: Analyze the data

[0082] The server receives comments and context data from the device and passes it through an analysis process. A generative AI model is used to review the input data, evaluate the meaning and sentiment of the analyzed data, and determine whether it matches a specific pattern.

[0083] Step 6: Automatic diagnosis

[0084] Based on the analysis results, the server refers to expert preventative measures and case studies to determine whether the input data constitutes harassment, and generates a diagnosis such as "This is harassment," "This is not harassment," or "It is difficult to determine."

[0085] Step 7: Submitting diagnostic results

[0086] The diagnostic results generated by the server are converted into a data format and sent to the user's terminal.

[0087] Step 8: View the diagnostic results

[0088] The terminal displays the diagnosis results received from the server to the user.

[0089] Step 9: Offering educational programs

[0090] The server selects an appropriate educational program based on the diagnosis results and sends a notification containing the educational program's URL and content to the user's device.

[0091] Step 10: Take the education program and track your progress

[0092] The user takes the educational program provided through the terminal, which tracks the progress of the educational program and periodically reports it to the server.

[0093] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. Furthermore, education programs can be used to continuously educate employees and improve the work environment.

[0094] Example 1

[0095] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0096] There is a need for a method to quickly and accurately identify harassment behaviors experienced by users in the workplace and provide appropriate countermeasures. There is also a growing need for educational and awareness programs to prevent harassment. To address these issues, the present invention provides an automated diagnostic system using a generative AI model, with the aim of improving the workplace environment.

[0097] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0098] In this invention, the server includes: means for a user to input comment data from a terminal and send it to the server; means for the server to collect the comment data and store it in a database; means for the server to cleanse and normalize the collected comment data; means for the server to train a generative AI model using the preprocessed comment data; means for the user to send newly input data from the terminal to the server and for the server to analyze the data using the generative AI model; means for the server to automatically diagnose whether the behavior constitutes harassment based on the analysis results; means for the server to generate and send the diagnostic results to the user's terminal; and means for the user to take an educational program via the terminal and report the user's progress to the server. This enables rapid and accurate diagnosis of harassment behavior and continuous education through appropriate educational programs.

[0099] A "user" is an individual who utilizes the system to input comments and situational data and receive diagnostic results and educational programs.

[0100] "Terminal" refers to a device used by a user to input comment data and situation data, send them to a server, and receive diagnostic results and educational programs.

[0101] "Server" refers to the system that collects, stores, and pre-processes comment data, trains and uses generative AI models to analyze the data, and generates and transmits diagnostic results to users' devices.

[0102] "Comment data" refers to text data of opinions and experiences about the work environment that users enter into internal surveys or chat systems.

[0103] "Data cleansing" is the process of removing noise and unnecessary information from collected comment data and putting it into a form suitable for analysis.

[0104] "Normalization" is the process of converting cleansed comment data into a consistent format to improve the accuracy of analysis.

[0105] A "generative AI model" is a model that is trained using machine learning algorithms to learn patterns of harassing behavior from comment data.

[0106] The "database" refers to a system for storing comment data, preventive measures by experts, case information, etc.

[0107] "Automatic diagnosis" refers to the process of using a generative AI model to analyze newly entered comments and situational data to determine whether they constitute harassing behavior.

[0108] "Diagnostic results" refer to an assessment of whether or not harassment has occurred, generated through an automated diagnostic process.

[0109] "Educational Program" refers to educational content provided to users with the aim of preventing and raising awareness of harassment.

[0110] "Progress" refers to the degree of completion of the educational program that the user has taken and the progress of their learning.

[0111] This invention provides a system that uses a generative AI model to learn patterns from comment data, an automatic diagnostic function that combines preventive measures and case information from experts, and an educational program. This system is implemented using the following hardware and software.

[0112] server

[0113] A server is a piece of hardware that has the following main functions:

[0114] 1. Data collection function

[0115] 2. Data preprocessing function

[0116] 3. Pattern learning function

[0117] 4. Automatic diagnosis function

[0118] 5. Result transmission function

[0119] 6. Educational program provision function

[0120] The server uses software such as Python, TensorFlow, and PyTorch to train generative AI models and analyze data. The server also houses a database system (e.g., MySQL or PostgreSQL) to store comment data, expert precautions, and case information.

[0121] Terminal

[0122] A terminal is a piece of hardware that has the following main functions:

[0123] 1. Comment and status data input function

[0124] 2. Data transmission function

[0125] 3. Diagnostic result display function

[0126] 4. Educational program display and progress tracking function

[0127] The terminal communicates data between the user and the server using a web browser or a dedicated application. The terminal sends data to the server using HTTP requests and receives diagnostic results and educational programs.

[0128] Service flow

[0129] Data collection

[0130] The user inputs comments and situation data from the device. For example, the user might input, "My boss has been frequently commenting on my appearance lately, and it makes me feel uncomfortable." This data is sent from the device to the server via an HTTP request. The server then stores the received comment data in a database.

[0131] Data Preprocessing

[0132] The server cleanses, removes noise, and normalizes the comment data stored in the database, and the data is then stored back into the database after this preprocessing.

[0133] Pattern Learning

[0134] The server uses the preprocessed comment data to train a generative AI model, a process that uses machine learning libraries such as TensorFlow and PyTorch.

[0135] Automatic diagnosis

[0136] The user inputs new comments and situational data from their device and sends it to the server. For example, data such as "In meetings, certain colleagues' opinions are being ignored" can be input. The server then uses a generative AI model to analyze the new data and automatically diagnose whether it constitutes harassment. Preventive measures and case information from experts are also referenced.

[0137] Sending and displaying diagnostic results

[0138] The server generates a diagnostic result and sends it to the user's terminal, which displays the diagnostic result to the user.

[0139] Providing educational programs

[0140] Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user's device. The user takes the educational program through their device, and their progress is reported to the server.

[0141] Specific examples

[0142] Example 1: Comments about appearance

[0143] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this data to the server, which then analyzes the comment using a generative AI model. The analysis results in a diagnosis that "these are negative comments about my appearance, and are likely to constitute harassment." The results are displayed on the user's device, and appropriate educational programs are provided.

[0144] Example 2: Ignoring comments

[0145] The user inputs the situation, such as, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which then analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an intentional attempt to exclude certain individuals, and is likely to constitute harassment." The results are displayed on the user's device, and an appropriate educational program is provided.

[0146] The system allows users and companies to quickly and accurately diagnose harassment and take appropriate measures. It also raises employee awareness through educational programs, promoting an improved work environment.

[0147] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0148] Step 1: Data entry

[0149] The user inputs comment data from the terminal. For example, "Recently, my boss has frequently commented on my appearance, and it makes me feel uncomfortable." The terminal sends the input comment data to the server via an HTTP request.

[0150] Input: Comment data entered by the user from the terminal

[0151] Output: Comment data sent to the server via an HTTP request

[0152] Step 2: Data collection

[0153] The server receives the comment data sent from the device and stores it in a database, for example, in a table called "comment_data."

[0154] Input: Comment data sent from the device

[0155] Output: Comment data to be saved in the database

[0156] Step 3: Data Preprocessing

[0157] The server retrieves raw comment data from the database and performs data cleansing, such as removing unnecessary whitespace and special characters from the text. Next, a normalization process converts all text to lowercase and standardizes synonyms. The preprocessed data is then re-stored in the "preprocessed_data" table.

[0158] Input: Comment data retrieved from the database

[0159] Output: Cleansed and normalized data

[0160] Step 4: Pattern learning

[0161] The server retrieves all data from the "preprocessed_data" table and prepares it as a training dataset. Then, it starts training the generative AI model using TensorFlow, specifically using the "fit()" method. After training, it saves the model to disk as "ai_model.h5".

[0162] Input: Preprocessed data

[0163] Output: A trained generative AI model

[0164] Step 5: Automatic diagnosis

[0165] The user enters a new comment on their device, such as "In meetings, certain colleagues' opinions are always ignored." The device sends the comment to the server via an HTTP request. The server then analyzes the received data by calling the saved generative AI model "ai_model.h5." The analysis results determine whether the comment constitutes harassment.

[0166] Input: New comment data sent from the device

[0167] Output: Diagnostic results from analysis

[0168] Step 6: Send and view diagnostic results

[0169] The server generates the analysis result, "This is an act of intentionally excluding a specific individual, and is likely to constitute harassment." Next, this diagnosis result is sent to the user's device via an HTTP response. The device then displays the received diagnosis result to the user.

[0170] Input: Analysis results

[0171] Output: Diagnostic results displayed on the user's terminal

[0172] Step 7: Offering educational programs

[0173] Based on the diagnosis results, the server selects an appropriate educational program regarding "intentional exclusion of specific individuals." It then sends the link and content of the selected educational program to the user's device via an HTTP response. The user then takes the educational program through their device and reports their progress to the server.

[0174] Input: Diagnostic results

[0175] Output: Educational program and progress information provided to the user's terminal

[0176] Through this series of processing steps, users and companies can quickly and accurately diagnose harassment behavior and take appropriate measures. Furthermore, education programs can be implemented to raise employee awareness and promote an improved work environment.

[0177] (Application example 1)

[0178] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0179] In today's workplace, harassment increases the mental and work burden on employees. Employees in brick-and-mortar stores, in particular, need prompt and accurate diagnosis and action against everyday harassment from customers, superiors, and coworkers. However, current manual reporting systems and training programs are slow to respond, resulting in long time-consuming resolutions for harassment cases.

[0180] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0181] In this invention, the server includes means for learning patterns in comment data using a generative AI model, means for analyzing newly entered comments and situation data, means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, means for transmitting the diagnosis results to the user's display device, means for transmitting comments and situation data entered by the user to the server, means for displaying an appropriate educational program based on the diagnosis results and tracking its progress, means for storing preventive measures and case information by experts in a storage device and using them for analysis and automatic diagnosis, and means for generating an appropriate educational program based on the newly entered comments and situation data, providing it to the user, and reporting its progress to the server. This enables rapid and accurate diagnosis of harassment behavior and provision of an appropriate educational program.

[0182] A "generative AI model" is an algorithm that uses artificial intelligence techniques to learn specific patterns and trends and analyze data autonomously.

[0183] "Comment data" refers to opinions or feedback in text form entered by users, and is typically in free-form format.

[0184] "Pattern learning" is the process of taking a given data set, identifying consistent features or trends in the data, and then allowing an algorithm to learn from those features.

[0185] "Contextual data" is data that contains detailed information about specific events or situations, describing a user's experience or environment.

[0186] "Analysis" is the process of examining collected data in detail and understanding its meaning and structure.

[0187] "Automatic diagnosis" is the process by which a system uses algorithms to identify problems or anomalies based on input data and generate a diagnostic result.

[0188] "Diagnosis results" are information indicating conclusions or judgments obtained through the process of analysis and automatic diagnosis.

[0189] A "user display device" is a device for visually presenting information to a user, examples of which include a smartphone or smart glasses.

[0190] A "storage device" is hardware or software for long-term storage of data or information.

[0191] "Educational Program" means a set of educational materials or training methods designed to teach specific knowledge or skills.

[0192] A system for implementing this invention uses a generative AI model to learn patterns in comment data and analyze newly entered comments and situational data to automatically diagnose harassment behavior and transmit the diagnosis results to the user's display device. Furthermore, it displays appropriate educational programs based on the diagnosis results and tracks their progress. It also stores preventive measures and case information developed by experts in a storage device and uses them for analysis and automatic diagnosis.

[0193] Hardware and Software Configuration

[0194] Server: The server has the computational capabilities to use generative AI models to learn patterns from comment data, and the ability to analyze newly entered comments and situational data. The database server stores preventive measures and case information from experts, which are used for analysis.

[0195] Terminal: A terminal is a display device such as a smartphone or smart glasses, which the user uses to input comments and situational data and check the diagnosis results. These terminals have the function of sending data from the user to the server and receiving diagnosis results and educational programs from the server.

[0196] Generative AI Model: A generative AI model is used to learn patterns of harassing behavior from comment data. This model is trained on the server and used to analyze new data.

[0197] Data processing and calculation

[0198] The server processes the data in the following steps:

[0199] 1. Data collection: Users input comments and situational data using their devices and send them to the server. Preventive measures and case information provided by experts are also collected and stored in a database.

[0200] 2. Data preprocessing: The server cleanses, denoises, and normalizes the collected comment data. After this preprocessing, the data is stored back in the database.

[0201] 3. Pattern Learning: The server uses the preprocessed comment data to train a generative AI model, which is used to learn patterns of harassing behavior.

[0202] 4. Automated Diagnosis: Users input new comments and situational data and send it to the server. The server uses a generative AI model to analyze this data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[0203] 5. Sending and displaying the results: The server generates the diagnostic results and sends them to the user's terminal, which then displays them to the user.

[0204] 6. Providing educational programs: Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through the terminal, and the progress is reported to the server.

[0205] Specific examples

[0206] Example 1: A user types into their smartphone, "My boss keeps making comments about my clothes and it's annoying me." The server receives this comment data and analyzes it using a generative AI model. The analysis results return a diagnosis that "comments about my clothes are inappropriate." An appropriate educational program is also provided along with the diagnosis.

[0207] Example 2: A user inputs, "My colleague always ignores what I say during meetings and tries to overwhelm me." The server receives this data and analyzes it using a generative AI model. The diagnosis is that "ignoring what I say constitutes harassment." The user confirms this result and is provided with relevant educational programs.

[0208] Prompt Sentence Examples

[0209] "An employee says: 'My boss keeps making comments about my attire and it's bothering me.' Make a diagnosis based on this comment and provide an appropriate training program."

[0210] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0211] Step 1: Data collection

[0212] The user uses the device to input comments and status data and sends it to the server. This input includes opinions and feedback in text format. The device receives the input data and sends it to the server using an HTTP request. The server receives this data and stores it in a database.

[0213] Step 2: Data Preprocessing

[0214] The server cleanses, removes noise, and normalizes the comment data stored in the database. For example, it removes unnecessary symbols and emojis such as "!!!" and "???." It also standardizes the same content expressed in different formats (e.g., "Hello" and "Hello"). After this preprocessing, the data is stored back in the database.

[0215] Step 3: Pattern learning

[0216] The server uses the preprocessed comment data to train the generative AI model. During pattern learning, the cleansed comments are used as input data and patterns of harassing behavior are identified as output. The model iteratively analyzes the data to improve its accuracy. This training process is run periodically.

[0217] Step 4: Automatic diagnosis

[0218] The user inputs new comments and situational data into the device and sends it to the server. The server then provides the received data to the generative AI model for analysis. This analysis determines whether the input data constitutes harassment, and a diagnosis result is generated as the output.

[0219] Step 5: Send and view results

[0220] The server generates the results of the automated diagnosis and sends them to the user's device. The device then visually displays the received results to the user. For example, it may display a result such as "This comment is likely to constitute harassment."

[0221] Step 6: Offering educational programs

[0222] Based on the diagnosis results, the server selects an appropriate educational program and sends it to the user's device. The user then takes the educational program provided through the device. The device periodically reports the user's progress to the server. The server uses this information to adjust the next educational program.

[0223] Prompt Sentence Examples

[0224] "An employee says: 'My boss keeps making comments about my attire and it's bothering me.' Make a diagnosis based on this comment and provide an appropriate training program."

[0225] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0226] The system of the present invention is equipped with a generative AI model that learns patterns from comment data, an automatic diagnostic function that combines preventive measures and case information from experts, the provision of educational programs, and an emotion engine that recognizes user emotions. Specific embodiments for implementing this system are described below.

[0227] System configuration

[0228] server

[0229] The server has the following functions:

[0230] 1. Data collection function

[0231] 2. Data preprocessing function

[0232] 3. Pattern learning function

[0233] 4. Automatic diagnosis function

[0234] 5. Analysis function using emotion engine

[0235] 6. Diagnostic and emotion analysis results transmission function

[0236] 7. Educational program provision function

[0237] 8. Database management functions

[0238] Terminal

[0239] The terminal has the following functions:

[0240] 1. Comment and status data input function

[0241] 2. Data transmission function

[0242] 3. Diagnostic and emotion analysis results display function

[0243] 4. Educational program display and progress tracking function

[0244] Service flow

[0245] Data collection

[0246] When users enter comments into internal surveys, chat systems, etc., the server continuously collects them. Prevention measures and case information provided by experts are also collected and stored in a database.

[0247] Data Preprocessing

[0248] The server cleanses, denoises, and normalizes the collected comment data, and then stores the pre-processed data back in the database.

[0249] Pattern Learning

[0250] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[0251] Automatic diagnosis

[0252] Users input new comments and situational data from their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[0253] Emotion analysis

[0254] The server uses an emotion engine to analyze the emotional state of comments and situational data entered by users, which are then classified as positive, negative, neutral, etc.

[0255] Sending and displaying results

[0256] The server generates diagnosis results and emotion analysis results and sends them to the user's terminal, which then displays the diagnosis results and emotion analysis results to the user.

[0257] Providing educational programs

[0258] Based on the diagnosis and emotion analysis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through their device, and their progress is reported to the server.

[0259] Specific examples

[0260] Example 1: Comments about appearance

[0261] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which analyzes the comment using a generative AI model. The analysis results in a diagnosis that "This is a negative comment about my appearance, and is likely to constitute harassment." The server then uses an emotion engine to analyze the comment as containing negative emotions. The results are displayed on the user's device, and further appropriate educational programs are provided.

[0262] Example 2: Ignoring comments

[0263] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an act of intentionally excluding a specific individual, and is likely to constitute harassment." The server then uses an emotion engine to analyze that the situation contains negative emotions. The results are displayed on the user's device, and an appropriate educational program is provided.

[0264] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. The emotion analysis function using the emotion engine also enables more accurate diagnosis and response. Furthermore, education programs provide continuous employee education, leading to an improved work environment.

[0265] The processing flow will be explained below.

[0266] Step 1: Collecting comment data

[0267] The server continuously collects comments from various comment data sources (internal surveys, internal chats, review sites, etc.), including data extraction using APIs and CSV file imports.

[0268] Step 2: Gather expert information

[0269] The server collects harassment prevention measures and case information provided by experts and stores them in a database, including loading static files and using input forms for experts.

[0270] Step 3: Preprocessing the data

[0271] The server retrieves raw comment data from the database, corrects emotional words and typos to remove noise, and standardizes synonyms and segments text to normalize it. The preprocessed data is then saved back to the database.

[0272] Step 4: Pattern learning

[0273] The server uses the preprocessed comment data as training data for a generative AI model, trains a generative AI model (e.g., BERT or GPT-4) to learn patterns of harassing behavior, and saves the trained model so it can be used for predictions.

[0274] Step 5: Enter comments and status data

[0275] The user uses the terminal to enter comments and situation data into the input form, and then presses the "Send" button to send the data to the server.

[0276] Step 6: Receive data and prepare for analysis

[0277] The device sends input data to the server, which receives it and prepares to pass the received comments and status data to the analysis process.

[0278] Step 7: Sentiment Analysis

[0279] The server uses an emotion engine to analyze the emotional state of comments and situational data received from users, which can be classified as positive, negative, neutral, etc.

[0280] Step 8: Analyze the data

[0281] The server uses the generated AI model to analyze the incoming data, taking into account the results of sentiment analysis, to evaluate the meaning and sentiment of the analyzed data and determine whether it matches a specific pattern.

[0282] Step 9: Automatic diagnosis

[0283] Based on the analysis results, the server automatically diagnoses whether the input data constitutes harassment, while also referencing expert preventative measures and case studies, and generates a diagnosis such as "This is harassment," "This is not harassment," or "It is difficult to determine."

[0284] Step 10: Sending diagnosis and sentiment analysis results

[0285] The server converts the diagnosis results and emotion analysis results into a data format and sends them to the user's terminal.

[0286] Step 11: Viewing the diagnosis and sentiment analysis results

[0287] The device displays the diagnosis results and emotion analysis results received from the server to the user, allowing the user to check the evaluation of their own comments and situations.

[0288] Step 12: Offering educational programs

[0289] The server selects an appropriate educational program based on the diagnosis and emotion analysis results, provides it to the user, and sends a notification to the user's device containing the educational program's URL and content.

[0290] Step 13: Take the education program and track your progress

[0291] The user takes the educational program provided through the terminal, which tracks the progress of the educational program and periodically reports it to the server.

[0292] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. The emotion analysis function using the emotion engine also enables more accurate diagnosis and response. Furthermore, education programs provide continuous employee education, leading to an improved work environment.

[0293] Example 2

[0294] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0295] It is important to quickly and accurately recognize harassment in the work environment and take measures to address it, but conventional systems tend to delay such recognition and countermeasures, making it difficult to alleviate employee stress and problems. Furthermore, conventional systems lack the functionality to analyze the emotional state of comments or to provide appropriate educational programs, preventing effective improvements to the work environment.

[0296] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0297] In this invention, the server includes means for learning patterns in comment data using a generative AI model, means for analyzing newly entered comments and situation data, means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, means having an emotion engine for analyzing the emotional state of the user's comments and situation data, and means for transmitting the diagnosis and emotion analysis results to the user's device. This enables rapid analysis of user input data and provides comprehensive diagnosis results including the user's emotional state, enabling continuous improvement of the work environment through appropriate educational programs.

[0298] A "generative AI model" is an algorithm that uses artificial intelligence technology to learn patterns and trends from data and make predictions and analyses on new data.

[0299] "Comment data" refers to text information, opinions, explanations of situations, etc. entered by users.

[0300] "Pattern learning" is the process of extracting specific trends or recurring features from data, understanding them, and incorporating them into a model.

[0301] "Automatic diagnosis" refers to the ability of a system to analyze data and make a determination about a specific condition or problem without human intervention.

[0302] An "emotion engine" is an algorithm that analyzes the emotions contained in text data and classifies them as positive, negative, neutral, etc.

[0303] "Diagnostic results" refer to the judgment information, such as whether or not harassment has occurred, that is shown as a result of the analysis by the generative AI model.

[0304] "Analysis results" refers to the overall results of the analysis performed by the generative AI model and emotion engine on the input data.

[0305] "Educational Program" means a course of study or training offered to enable a user to acquire specific knowledge or skills.

[0306] The system of the present invention collects user comment data and situational data and automatically analyzes and diagnoses them. This system is equipped with a pattern learning function using a generative AI model, an automatic diagnosis function, an analysis function using an emotion engine, and an educational program provision function. This enables the rapid and accurate recognition of harassment in the work environment and the implementation of countermeasures.

[0307] Server configuration and functions

[0308] Data collection features:

[0309] The server continuously collects data on comments entered by users using the company's internal surveys and chat systems, and stores the collected data in a database.

[0310] Data preprocessing functions:

[0311] The server cleanses the collected comment data, removes noise, and normalizes it. The preprocessed data is then stored in the database again.

[0312] Pattern learning function:

[0313] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[0314] Automatic diagnostic function:

[0315] Users input new comments and situational data through their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether it constitutes harassment.

[0316] Emotion engine analysis function:

[0317] The server uses an emotion engine to analyze the emotional state of comments and situational data entered by users, which are classified as positive, negative, neutral, etc.

[0318] Diagnostic and sentiment analysis results transmission function:

[0319] The server transmits the generated diagnosis results and emotion analysis results to the user's terminal, which displays these results to the user.

[0320] Educational program offerings:

[0321] The server selects an appropriate educational program based on the diagnosis and emotion analysis results and provides it to the user. The user takes the educational program through their device, and their progress is reported to the server.

[0322] Device configuration and functions

[0323] Comment and status data entry function:

[0324] It provides an interface for users to input comment data and situational data through an internal survey or chat system.

[0325] Data transmission function:

[0326] The terminal transmits the input data to the server.

[0327] Diagnostic and emotion analysis results display function:

[0328] The terminal displays the diagnosis results and emotion analysis results sent from the server to the user.

[0329] Educational Program Display & Progress Tracking Features:

[0330] The terminal displays the educational program provided by the server and tracks the user's learning progress.

[0331] Specific operation example

[0332] Example 1: Comments about appearance

[0333] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which analyzes the comment using a generative AI model. The analysis results in a diagnosis that "This is a negative comment about my appearance, and is likely to constitute harassment." The server then uses an emotion engine to analyze the comment as containing negative emotions. The results are displayed on the user's device, and further appropriate educational programs are provided.

[0334] Example 2: Ignoring comments

[0335] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an act of intentionally excluding a specific individual, and is likely to constitute harassment." The server then uses an emotion engine to analyze that the situation contains negative emotions. The results are displayed on the user's device, and an appropriate educational program is provided.

[0336] Prompt Sentence Examples

[0337] "My boss frequently makes comments about my appearance, which I find very unpleasant. What kind of diagnosis would a generative AI model make in this case?"

[0338] "My colleague keeps getting ignored in meetings. How would the emotion engine analyze this situation?"

[0339] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0340] Step 1:

[0341] A user enters a comment into an internal survey or chat system.

[0342] As a specific operation, the user inputs text such as "Recently, my boss has been frequently commenting on my appearance, and it makes me feel uncomfortable."

[0343] Input: Comment data entered by the user

[0344] Output: Comment data sent to the device

[0345] Step 2:

[0346] The terminal transmits the input comment data to the server.

[0347] As a specific operation, the terminal generates and transmits a request for transmitting the comment data input by the user to the server.

[0348] Input: Comment data entered into the terminal

[0349] Output: Comment data sent to the server

[0350] Step 3:

[0351] The server stores the received comment data in a database.

[0352] Specifically, the server stores the received comment data in an appropriate database table.

[0353] Input: Comment data received by the server

[0354] Output: Comment data stored in the database

[0355] Step 4:

[0356] The server retrieves the comment data from the database and performs data cleansing.

[0357] Specifically, the server reads unprocessed comment data from the database and performs noise removal and format conversion.

[0358] Input: Comment data stored in the database

[0359] Output: Cleansed comment data

[0360] Step 5:

[0361] The server normalizes the cleansed comment data and stores it back in the database.

[0362] Specifically, the server performs normalization processing, such as converting the text to lowercase, and stores it in the database.

[0363] Input: Cleansed comment data

[0364] Output: Normalized comment data

[0365] Step 6:

[0366] The server uses the normalized comment data to train a generative AI model.

[0367] Specifically, the server inputs the normalized data into the AI ​​model and performs training.

[0368] Input: Normalized comment data

[0369] Output: A trained generative AI model

[0370] Step 7:

[0371] The user inputs new comments and status data from the terminal, which then transmits the data to the server.

[0372] Specifically, the user inputs new text such as, "In daily meetings, certain colleagues are always ignored and denied the opportunity to speak," and the device sends this to the server.

[0373] Input: New comments and status data entered by the user

[0374] Output: New comments and status data sent to the server

[0375] Step 8:

[0376] The server uses the generated AI model to analyze newly received data.

[0377] Specifically, the server inputs new comments and situational data into the generative AI model and obtains the analysis results.

[0378] Input: New comments and status data sent to the server

[0379] Output: Parsed data

[0380] Step 9:

[0381] Based on the analysis results, the server automatically diagnoses whether the behavior constitutes harassment.

[0382] Specifically, the system refers to the analysis results and case information from experts to determine whether or not the behavior constitutes harassment.

[0383] Input: Parsed data

[0384] Output: Automatic diagnosis results

[0385] Step 10:

[0386] The server uses an emotion engine to analyze the emotional state of the comments and situation data.

[0387] Specifically, the server detects emotional keywords in the comments and classifies their emotional state into "positive," "negative," or "neutral."

[0388] Input: New comments and status data

[0389] Output: Emotion analysis results

[0390] Step 11:

[0391] The server generates diagnosis results and emotion analysis results and transmits them to the terminal.

[0392] Specifically, the server combines the diagnosis results and emotion analysis results into a single packet and sends it to the terminal.

[0393] Input: Automatic diagnosis results and emotion analysis results

[0394] Output: Result data sent to the terminal

[0395] Step 12:

[0396] The terminal displays the received diagnosis results and emotion analysis results to the user.

[0397] Specifically, the terminal provides the user with the diagnosis results and emotion analysis results as a screen display.

[0398] Input: Result data sent to the terminal

[0399] Output: Diagnosis results and sentiment analysis results displayed to the user

[0400] Step 13:

[0401] The server selects an appropriate educational program based on the diagnosis and emotion analysis results.

[0402] Specifically, the server selects and lists educational programs from a database according to the diagnostic results.

[0403] Input: Diagnosis results and emotion analysis results

[0404] Output: Selected educational programs

[0405] Step 14:

[0406] The server transmits the selected educational program to the terminal.

[0407] Specifically, the server sends links to educational programs and educational materials to the terminal.

[0408] Input: Selected Educational Program

[0409] Output: Educational program sent to the terminal

[0410] Step 15:

[0411] A user takes an educational program through a terminal, and the progress is reported to a server.

[0412] Specifically, the user watches an educational program on the terminal, and the learning progress is automatically reported to the server.

[0413] Input: Educational program progress data

[0414] Output: Progress data reported to the server

[0415] (Application example 2)

[0416] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0417] Traditional brick-and-mortar stores lack systems for quickly and accurately evaluating customer satisfaction and proposing concrete improvement measures. In particular, there is no way to accurately analyze the emotional state of customer feedback and immediately propose countermeasures for negative feedback. As a result, improvements in the customer service skills of store staff and service quality are delayed, leading to a decline in customer satisfaction.

[0418] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0419] In this invention, the server includes a means for pattern learning of comment data using a generative AI model, a means for collecting, preprocessing, and analyzing newly entered customer feedback, and a means for analyzing the emotional state of comments and situation data using an emotion analysis engine. This makes it possible to quickly and accurately analyze the emotional state of customer feedback and provide specific improvement measures and training programs for store staff.

[0420] A "generative AI model" is an artificial intelligence model that learns patterns based on collected data and analyzes new data.

[0421] "Comment Data" refers to feedback and opinions entered by users in the form of text.

[0422] An "emotion analysis engine" is software that analyzes and classifies emotional states from comments and situational data.

[0423] "Automatic diagnosis" is the process by which a generative AI model analyzes newly input data and determines whether a particular behavior constitutes harassment.

[0424] "Educational Program" refers to training and learning materials provided to users based on the analysis results.

[0425] "Expert preventive measures and case information" means information provided by experts and past cases regarding the prevention and countermeasures against harassment.

[0426] "Customer feedback" refers to the evaluations, opinions, and impressions that customers provide regarding services and products.

[0427] "Customer service skills improvement training" is an educational program designed to improve the customer service attitude and response techniques of store staff.

[0428] "Negative feedback" refers to evaluations or opinions that express customer dissatisfaction or annoyance.

[0429] "Progress tracking" is the process of monitoring and managing progress through a delivered educational program.

[0430] The present invention relates to a system for analyzing customer feedback in a physical store and improving the customer service skills of store staff. The specific system configuration and operation will be described below.

[0431] System configuration

[0432] server

[0433] The server has the following functions:

[0434] 1. Data collection function: Collect customer feedback from smartphones and tablets.

[0435] 2. Data preprocessing function: Cleanse the collected feedback data and remove noise.

[0436] 3. Pattern learning function: Trains generative AI models using preprocessed data.

[0437] 4. Automatic diagnosis function: Feedback data is analyzed using a generative AI model to diagnose specific behaviors.

[0438] 5. Sentiment Analysis Function: Uses a sentiment analysis engine to analyze the emotional state of the feedback.

[0439] 6. Diagnostic result transmission function: Transmits diagnostic results and emotion analysis results to the user's device.

[0440] 7. Educational program provision function: Based on the diagnostic results, select and provide appropriate educational programs.

[0441] Terminal

[0442] The terminal has the following features:

[0443] 1. Comment input function: Customers can input feedback using their smartphones or tablets.

[0444] 2. Data transmission function: Sends collected data to the server.

[0445] 3. Result display function: Displays the diagnosis results and emotion analysis results.

[0446] 4. Educational program progress tracking function: Tracks the progress of the educational program and reports the results to the server.

[0447] Program processing flow

[0448] 1. The user enters feedback using a smartphone or tablet. For example, they enter feedback such as, "I feel like the staff have been cold towards me lately."

[0449] 2. The device sends the input feedback data to the server.

[0450] 3. The server receives the feedback data and performs data preprocessing, which involves data cleansing and noise removal.

[0451] 4. Using the pre-processed data, the generative AI model analyzes the feedback content and diagnoses whether a particular behavior constitutes harassment.

[0452] 5. The server uses an emotion analysis engine to analyze the emotional state of the feedback, e.g., it is analyzed as "negative."

[0453] 6. The diagnosis results and emotion analysis results are sent to the store manager's terminal and displayed.

[0454] 7. Based on the results of the diagnosis, the server will provide an appropriate educational program (e.g., "Customer Service Skills Improvement Training").

[0455] Hardware and software used

[0456] Hardware:

[0457] Devices used by customers and store managers: smartphones, tablets, PCs, etc.

[0458] Server: High-performance data processing server

[0459] software:

[0460] Sentiment analysis engine: for example, a model using the transformers library

[0461] Data preprocessing and analysis: Programming languages ​​such as Python

[0462] Specific examples

[0463] As a concrete example, let's consider the case where a customer uses a smartphone. A customer inputs feedback such as, "I feel like the staff have been cold towards me lately." The data is sent to the server, where it is preprocessed and analyzed by the generative AI model. The analysis results indicate that "improvement in customer service attitude is necessary," and the sentiment analysis result is "negative." These results are then notified to the store manager, who provides training to improve customer service skills.

[0464] Prompt Sentence Examples

[0465] Feedback: "I feel like the staff have been cold lately."

[0466] Prompt for generative AI model: "Based on this feedback, please rate the customer's attitude."

[0467] In this way, the present invention makes it possible to analyze customer feedback quickly and accurately in a physical store and immediately propose countermeasures, thereby improving the customer service skills of store staff and increasing customer satisfaction.

[0468] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0469] Step 1:

[0470] The user inputs and sends feedback using a smartphone or tablet. For example, the input data might be, "Recently, I feel like the staff have been cold towards me." The device then sends this feedback to the server. The input is text data, and the output is data sent to the server.

[0471] Step 2:

[0472] The server cleanses and pre-processes the received feedback data, which includes denoising and normalizing the data. The input data is the raw feedback, and the output data is the cleansed text data.

[0473] Step 3:

[0474] The server inputs the preprocessed feedback data into a generative AI model and performs pattern learning. The model analyzes the data and diagnoses whether a particular behavior constitutes harassment. The input data is the preprocessed feedback text, and the output data is the diagnosis result.

[0475] Step 4:

[0476] The server uses a sentiment analysis engine to analyze the emotional state of the feedback, e.g., it is analyzed as "negative." The input data is the feedback text, and the output data is the sentiment analysis result.

[0477] Step 5:

[0478] The server sends the diagnosis results and emotion analysis results to the user's device. The input data here are the diagnosis results and emotion analysis results, and the output data is sent to the device. The user's device displays these results.

[0479] Step 6:

[0480] Based on the diagnosis results, the server selects and provides a relevant educational program. For example, "Training to improve customer service skills" may be suggested. The input data is the diagnosis results, and the output data is a suggested educational program. The content of the educational program is displayed on the user's device.

[0481] Step 7:

[0482] A user takes an educational program and reports his / her progress to the server via his / her terminal. The input data is the progress information of the educational program, and the output data is the progress report to the server.

[0483] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0484] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0485] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0486] [Second embodiment]

[0487] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0488] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0489] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0490] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0491] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0492] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0493] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0494] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0495] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0496] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0497] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0498] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0499] The system of the present invention is designed to provide pattern learning of comment data using a generative AI model, an automatic diagnostic function that combines preventive measures by experts and case information, and an educational program. Specific embodiments for implementing this system are described below.

[0500] System configuration

[0501] server

[0502] The server has the following functions:

[0503] 1. Data collection function

[0504] 2. Data preprocessing function

[0505] 3. Pattern learning function

[0506] 4. Automatic diagnosis function

[0507] 5. Result transmission function

[0508] 6. Educational program provision function

[0509] Terminal

[0510] The terminal has the following functions:

[0511] 1. Comment and status data input function

[0512] 2. Data transmission function

[0513] 3. Diagnostic result display function

[0514] 4. Educational program display and progress tracking function

[0515] Service flow

[0516] Data collection

[0517] When users enter comments into internal surveys, chat systems, etc., the server continuously collects them. Prevention measures and case information provided by experts are also collected and stored in a database.

[0518] Data Preprocessing

[0519] The server cleanses, denoises, and normalizes the collected comment data, and then stores the pre-processed data back in the database.

[0520] Pattern Learning

[0521] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[0522] Automatic diagnosis

[0523] Users input new comments and situational data from their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[0524] Sending and displaying results

[0525] The server generates a diagnostic result and sends it to the user's terminal, which displays the diagnostic result to the user.

[0526] Providing educational programs

[0527] Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through the terminal, and the progress is reported to the server.

[0528] Specific examples

[0529] Example 1: Comments about appearance

[0530] The user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which then analyzes the comment using a generative AI model. The analysis results in a diagnosis that "these are negative comments about my appearance, and are likely to constitute harassment." The results are displayed on the user's device, and appropriate educational programs are provided.

[0531] Example 2: Ignoring comments

[0532] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which then analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an intentional attempt to exclude certain individuals, and is likely to constitute harassment." The results are displayed on the user's device, and an appropriate educational program is provided.

[0533] This allows users and companies to quickly and accurately diagnose harassment and take appropriate measures. Furthermore, education programs can be used to continuously educate employees and improve the work environment.

[0534] The processing flow will be explained below.

[0535] Step 1: Data collection

[0536] The server continuously collects comments from various comment data sources (internal surveys, internal chats, review sites, etc.) and stores the collected comment data, along with preventive measures and case information from experts, in a database.

[0537] Step 2: Data Preprocessing

[0538] The server retrieves raw comment data from the database, performs noise removal (e.g., correcting emotional words and typos), performs text normalization (e.g., synonym unification, text segmentation), and stores the preprocessed data back in the database.

[0539] Step 3: Pattern learning

[0540] The server retrieves preprocessed comment data from the database. It uses this comment data to train a generative AI model (e.g., BERT or GPT-4). The model learns patterns of harassing behavior. It saves the trained model, making it available for prediction.

[0541] Step 4: Accepting input data

[0542] The user uses the terminal to enter comments and situation data into the input form, and then presses the "Send" button to send the data to the server.

[0543] Step 5: Analyze the data

[0544] The server receives comments and context data from the device and passes it through an analysis process. A generative AI model is used to review the input data, evaluate the meaning and sentiment of the analyzed data, and determine whether it matches a specific pattern.

[0545] Step 6: Automatic diagnosis

[0546] Based on the analysis results, the server refers to expert preventative measures and case studies to determine whether the input data constitutes harassment, and generates a diagnosis such as "This is harassment," "This is not harassment," or "It is difficult to determine."

[0547] Step 7: Submitting diagnostic results

[0548] The diagnostic results generated by the server are converted into a data format and sent to the user's terminal.

[0549] Step 8: View the diagnostic results

[0550] The terminal displays the diagnosis results received from the server to the user.

[0551] Step 9: Offering educational programs

[0552] The server selects an appropriate educational program based on the diagnosis results and sends a notification containing the educational program's URL and content to the user's device.

[0553] Step 10: Take the education program and track your progress

[0554] The user takes the educational program provided through the terminal, which tracks the progress of the educational program and periodically reports it to the server.

[0555] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. Furthermore, education programs can be used to continuously educate employees and improve the work environment.

[0556] Example 1

[0557] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0558] There is a need for a method to quickly and accurately identify harassment behaviors experienced by users in the workplace and provide appropriate countermeasures. There is also a growing need for educational and awareness programs to prevent harassment. To address these issues, the present invention provides an automated diagnostic system using a generative AI model, with the aim of improving the workplace environment.

[0559] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0560] In this invention, the server includes: means for a user to input comment data from a terminal and send it to the server; means for the server to collect the comment data and store it in a database; means for the server to cleanse and normalize the collected comment data; means for the server to train a generative AI model using the preprocessed comment data; means for the user to send newly input data from the terminal to the server and for the server to analyze the data using the generative AI model; means for the server to automatically diagnose whether the behavior constitutes harassment based on the analysis results; means for the server to generate and send the diagnostic results to the user's terminal; and means for the user to take an educational program via the terminal and report the user's progress to the server. This enables rapid and accurate diagnosis of harassment behavior and continuous education through appropriate educational programs.

[0561] A "user" is an individual who utilizes the system to input comments and situational data and receive diagnostic results and educational programs.

[0562] "Terminal" refers to a device used by a user to input comment data and situation data, send them to a server, and receive diagnostic results and educational programs.

[0563] "Server" refers to the system that collects, stores, and pre-processes comment data, trains and uses generative AI models to analyze the data, and generates and transmits diagnostic results to users' devices.

[0564] "Comment data" refers to text data of opinions and experiences about the work environment that users enter into internal surveys or chat systems.

[0565] "Data cleansing" is the process of removing noise and unnecessary information from collected comment data and putting it into a form suitable for analysis.

[0566] "Normalization" is the process of converting cleansed comment data into a consistent format to improve the accuracy of analysis.

[0567] A "generative AI model" is a model that is trained using machine learning algorithms to learn patterns of harassing behavior from comment data.

[0568] The "database" refers to a system for storing comment data, preventive measures by experts, case information, etc.

[0569] "Automatic diagnosis" refers to the process of using a generative AI model to analyze newly entered comments and situational data to determine whether they constitute harassing behavior.

[0570] "Diagnostic results" refer to an assessment of whether or not harassment has occurred, generated through an automated diagnostic process.

[0571] "Educational Program" refers to educational content provided to users with the aim of preventing and raising awareness of harassment.

[0572] "Progress" refers to the degree of completion of the educational program that the user has taken and the progress of their learning.

[0573] This invention provides a system that uses a generative AI model to learn patterns from comment data, an automatic diagnostic function that combines preventive measures and case information from experts, and an educational program. This system is implemented using the following hardware and software.

[0574] server

[0575] A server is a piece of hardware that has the following main functions:

[0576] 1. Data collection function

[0577] 2. Data preprocessing function

[0578] 3. Pattern learning function

[0579] 4. Automatic diagnosis function

[0580] 5. Result transmission function

[0581] 6. Educational program provision function

[0582] The server uses software such as Python, TensorFlow, and PyTorch to train generative AI models and analyze data. The server also houses a database system (e.g., MySQL or PostgreSQL) to store comment data, expert precautions, and case information.

[0583] Terminal

[0584] A terminal is a piece of hardware that has the following main functions:

[0585] 1. Comment and status data input function

[0586] 2. Data transmission function

[0587] 3. Diagnostic result display function

[0588] 4. Educational program display and progress tracking function

[0589] The terminal communicates data between the user and the server using a web browser or a dedicated application. The terminal sends data to the server using HTTP requests and receives diagnostic results and educational programs.

[0590] Service flow

[0591] Data collection

[0592] The user inputs comments and situation data from the device. For example, the user might input, "My boss has been frequently commenting on my appearance lately, and it makes me feel uncomfortable." This data is sent from the device to the server via an HTTP request. The server then stores the received comment data in a database.

[0593] Data Preprocessing

[0594] The server cleanses, removes noise, and normalizes the comment data stored in the database, and the data is then stored back into the database after this preprocessing.

[0595] Pattern Learning

[0596] The server uses the preprocessed comment data to train a generative AI model, a process that uses machine learning libraries such as TensorFlow and PyTorch.

[0597] Automatic diagnosis

[0598] The user inputs new comments and situational data from their device and sends it to the server. For example, data such as "In meetings, certain colleagues' opinions are being ignored" can be input. The server then uses a generative AI model to analyze the new data and automatically diagnose whether it constitutes harassment. Preventive measures and case information from experts are also referenced.

[0599] Sending and displaying diagnostic results

[0600] The server generates a diagnostic result and sends it to the user's terminal, which displays the diagnostic result to the user.

[0601] Providing educational programs

[0602] Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user's device. The user takes the educational program through their device, and their progress is reported to the server.

[0603] Specific examples

[0604] Example 1: Comments about appearance

[0605] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this data to the server, which then analyzes the comment using a generative AI model. The analysis results in a diagnosis that "these are negative comments about my appearance, and are likely to constitute harassment." The results are displayed on the user's device, and appropriate educational programs are provided.

[0606] Example 2: Ignoring comments

[0607] The user inputs the situation, such as, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which then analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an intentional attempt to exclude certain individuals, and is likely to constitute harassment." The results are displayed on the user's device, and an appropriate educational program is provided.

[0608] The system allows users and companies to quickly and accurately diagnose harassment and take appropriate measures. It also raises employee awareness through educational programs, promoting an improved work environment.

[0609] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0610] Step 1: Data entry

[0611] The user inputs comment data from the terminal. For example, "Recently, my boss has frequently commented on my appearance, and it makes me feel uncomfortable." The terminal sends the input comment data to the server via an HTTP request.

[0612] Input: Comment data entered by the user from the terminal

[0613] Output: Comment data sent to the server via an HTTP request

[0614] Step 2: Data collection

[0615] The server receives the comment data sent from the device and stores it in a database, for example, in a table called "comment_data."

[0616] Input: Comment data sent from the device

[0617] Output: Comment data to be saved in the database

[0618] Step 3: Data Preprocessing

[0619] The server retrieves raw comment data from the database and performs data cleansing, such as removing unnecessary whitespace and special characters from the text. Next, a normalization process converts all text to lowercase and standardizes synonyms. The preprocessed data is then re-stored in the "preprocessed_data" table.

[0620] Input: Comment data retrieved from the database

[0621] Output: Cleansed and normalized data

[0622] Step 4: Pattern learning

[0623] The server retrieves all data from the "preprocessed_data" table and prepares it as a training dataset. Then, it starts training the generative AI model using TensorFlow, specifically using the "fit()" method. After training, it saves the model to disk as "ai_model.h5".

[0624] Input: Preprocessed data

[0625] Output: A trained generative AI model

[0626] Step 5: Automatic diagnosis

[0627] The user enters a new comment on their device, such as "In meetings, certain colleagues' opinions are always ignored." The device sends the comment to the server via an HTTP request. The server then analyzes the received data by calling the saved generative AI model "ai_model.h5." The analysis results determine whether the comment constitutes harassment.

[0628] Input: New comment data sent from the device

[0629] Output: Diagnostic results from analysis

[0630] Step 6: Send and view diagnostic results

[0631] The server generates the analysis result, "This is an act of intentionally excluding a specific individual, and is likely to constitute harassment." Next, this diagnosis result is sent to the user's device via an HTTP response. The device then displays the received diagnosis result to the user.

[0632] Input: Analysis results

[0633] Output: Diagnostic results displayed on the user's terminal

[0634] Step 7: Offering educational programs

[0635] Based on the diagnosis results, the server selects an appropriate educational program regarding "intentional exclusion of specific individuals." It then sends the link and content of the selected educational program to the user's device via an HTTP response. The user then takes the educational program through their device and reports their progress to the server.

[0636] Input: Diagnostic results

[0637] Output: Educational program and progress information provided to the user's terminal

[0638] Through this series of processing steps, users and companies can quickly and accurately diagnose harassment behavior and take appropriate measures. Furthermore, education programs can be implemented to raise employee awareness and promote an improved work environment.

[0639] (Application example 1)

[0640] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0641] In today's workplace, harassment increases the mental and work burden on employees. Employees in brick-and-mortar stores, in particular, need prompt and accurate diagnosis and action against everyday harassment from customers, superiors, and coworkers. However, current manual reporting systems and training programs are slow to respond, resulting in long time-consuming resolutions for harassment cases.

[0642] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0643] In this invention, the server includes means for learning patterns in comment data using a generative AI model, means for analyzing newly entered comments and situation data, means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, means for transmitting the diagnosis results to the user's display device, means for transmitting comments and situation data entered by the user to the server, means for displaying an appropriate educational program based on the diagnosis results and tracking its progress, means for storing preventive measures and case information by experts in a storage device and using them for analysis and automatic diagnosis, and means for generating an appropriate educational program based on the newly entered comments and situation data, providing it to the user, and reporting its progress to the server. This enables rapid and accurate diagnosis of harassment behavior and provision of an appropriate educational program.

[0644] A "generative AI model" is an algorithm that uses artificial intelligence techniques to learn specific patterns and trends and analyze data autonomously.

[0645] "Comment data" refers to opinions or feedback in text form entered by users, and is typically in free-form format.

[0646] "Pattern learning" is the process of taking a given data set, identifying consistent features or trends in the data, and then allowing an algorithm to learn from those features.

[0647] "Contextual data" is data that contains detailed information about specific events or situations, describing a user's experience or environment.

[0648] "Analysis" is the process of examining collected data in detail and understanding its meaning and structure.

[0649] "Automatic diagnosis" is the process by which a system uses algorithms to identify problems or anomalies based on input data and generate a diagnostic result.

[0650] "Diagnosis results" are information indicating conclusions or judgments obtained through the process of analysis and automatic diagnosis.

[0651] A "user display device" is a device for visually presenting information to a user, examples of which include a smartphone or smart glasses.

[0652] A "storage device" is hardware or software for long-term storage of data or information.

[0653] "Educational Program" means a set of educational materials or training methods designed to teach specific knowledge or skills.

[0654] A system for implementing this invention uses a generative AI model to learn patterns in comment data and analyze newly entered comments and situational data to automatically diagnose harassment behavior and transmit the diagnosis results to the user's display device. Furthermore, it displays appropriate educational programs based on the diagnosis results and tracks their progress. It also stores preventive measures and case information developed by experts in a storage device and uses them for analysis and automatic diagnosis.

[0655] Hardware and Software Configuration

[0656] Server: The server has the computational capabilities to use generative AI models to learn patterns from comment data, and the ability to analyze newly entered comments and situational data. The database server stores preventive measures and case information from experts, which are used for analysis.

[0657] Terminal: A terminal is a display device such as a smartphone or smart glasses, which the user uses to input comments and situational data and check the diagnosis results. These terminals have the function of sending data from the user to the server and receiving diagnosis results and educational programs from the server.

[0658] Generative AI Model: A generative AI model is used to learn patterns of harassing behavior from comment data. This model is trained on the server and used to analyze new data.

[0659] Data processing and calculation

[0660] The server processes the data in the following steps:

[0661] 1. Data collection: Users input comments and situational data using their devices and send them to the server. Preventive measures and case information provided by experts are also collected and stored in a database.

[0662] 2. Data preprocessing: The server cleanses, denoises, and normalizes the collected comment data. After this preprocessing, the data is stored back in the database.

[0663] 3. Pattern Learning: The server uses the preprocessed comment data to train a generative AI model, which is used to learn patterns of harassing behavior.

[0664] 4. Automated Diagnosis: Users input new comments and situational data and send it to the server. The server uses a generative AI model to analyze this data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[0665] 5. Sending and displaying the results: The server generates the diagnostic results and sends them to the user's terminal, which then displays them to the user.

[0666] 6. Providing educational programs: Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through the terminal, and the progress is reported to the server.

[0667] Specific examples

[0668] Example 1: A user types into their smartphone, "My boss keeps making comments about my clothes and it's annoying me." The server receives this comment data and analyzes it using a generative AI model. The analysis results return a diagnosis that "comments about my clothes are inappropriate." An appropriate educational program is also provided along with the diagnosis.

[0669] Example 2: A user inputs, "My colleague always ignores what I say during meetings and tries to overwhelm me." The server receives this data and analyzes it using a generative AI model. The diagnosis is that "ignoring what I say constitutes harassment." The user confirms this result and is provided with relevant educational programs.

[0670] Prompt Sentence Examples

[0671] "An employee says: 'My boss keeps making comments about my attire and it's bothering me.' Make a diagnosis based on this comment and provide an appropriate training program."

[0672] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0673] Step 1: Data collection

[0674] The user uses the device to input comments and status data and sends it to the server. This input includes opinions and feedback in text format. The device receives the input data and sends it to the server using an HTTP request. The server receives this data and stores it in a database.

[0675] Step 2: Data Preprocessing

[0676] The server cleanses, removes noise, and normalizes the comment data stored in the database. For example, it removes unnecessary symbols and emojis such as "!!!" and "???." It also standardizes the same content expressed in different formats (e.g., "Hello" and "Hello"). After this preprocessing, the data is stored back in the database.

[0677] Step 3: Pattern learning

[0678] The server uses the preprocessed comment data to train the generative AI model. During pattern learning, the cleansed comments are used as input data and patterns of harassing behavior are identified as output. The model iteratively analyzes the data to improve its accuracy. This training process is run periodically.

[0679] Step 4: Automatic diagnosis

[0680] The user inputs new comments and situational data into the device and sends it to the server. The server then provides the received data to the generative AI model for analysis. This analysis determines whether the input data constitutes harassment, and a diagnosis result is generated as the output.

[0681] Step 5: Send and view results

[0682] The server generates the results of the automated diagnosis and sends them to the user's device. The device then visually displays the received results to the user. For example, it may display a result such as "This comment is likely to constitute harassment."

[0683] Step 6: Offering educational programs

[0684] Based on the diagnosis results, the server selects an appropriate educational program and sends it to the user's device. The user then takes the educational program provided through the device. The device periodically reports the user's progress to the server. The server uses this information to adjust the next educational program.

[0685] Prompt Sentence Examples

[0686] "An employee says: 'My boss keeps making comments about my attire and it's bothering me.' Make a diagnosis based on this comment and provide an appropriate training program."

[0687] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0688] The system of the present invention is equipped with a generative AI model that learns patterns from comment data, an automatic diagnostic function that combines preventive measures and case information from experts, the provision of educational programs, and an emotion engine that recognizes user emotions. Specific embodiments for implementing this system are described below.

[0689] System configuration

[0690] server

[0691] The server has the following functions:

[0692] 1. Data collection function

[0693] 2. Data preprocessing function

[0694] 3. Pattern learning function

[0695] 4. Automatic diagnosis function

[0696] 5. Analysis function using emotion engine

[0697] 6. Diagnostic and emotion analysis results transmission function

[0698] 7. Educational program provision function

[0699] 8. Database management functions

[0700] Terminal

[0701] The terminal has the following functions:

[0702] 1. Comment and status data input function

[0703] 2. Data transmission function

[0704] 3. Diagnostic and emotion analysis results display function

[0705] 4. Educational program display and progress tracking function

[0706] Service flow

[0707] Data collection

[0708] When users enter comments into internal surveys, chat systems, etc., the server continuously collects them. Prevention measures and case information provided by experts are also collected and stored in a database.

[0709] Data Preprocessing

[0710] The server cleanses, denoises, and normalizes the collected comment data, and then stores the pre-processed data back in the database.

[0711] Pattern Learning

[0712] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[0713] Automatic diagnosis

[0714] Users input new comments and situational data from their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[0715] Emotion analysis

[0716] The server uses an emotion engine to analyze the emotional state of comments and situational data entered by users, which are then classified as positive, negative, neutral, etc.

[0717] Sending and displaying results

[0718] The server generates diagnosis results and emotion analysis results and sends them to the user's terminal, which then displays the diagnosis results and emotion analysis results to the user.

[0719] Providing educational programs

[0720] Based on the diagnosis and emotion analysis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through their device, and their progress is reported to the server.

[0721] Specific examples

[0722] Example 1: Comments about appearance

[0723] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which analyzes the comment using a generative AI model. The analysis results in a diagnosis that "This is a negative comment about my appearance, and is likely to constitute harassment." The server then uses an emotion engine to analyze the comment as containing negative emotions. The results are displayed on the user's device, and further appropriate educational programs are provided.

[0724] Example 2: Ignoring comments

[0725] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an act of intentionally excluding a specific individual, and is likely to constitute harassment." The server then uses an emotion engine to analyze that the situation contains negative emotions. The results are displayed on the user's device, and an appropriate educational program is provided.

[0726] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. The emotion analysis function using the emotion engine also enables more accurate diagnosis and response. Furthermore, education programs provide continuous employee education, leading to an improved work environment.

[0727] The processing flow will be explained below.

[0728] Step 1: Collecting comment data

[0729] The server continuously collects comments from various comment data sources (internal surveys, internal chats, review sites, etc.), including data extraction using APIs and CSV file imports.

[0730] Step 2: Gather expert information

[0731] The server collects harassment prevention measures and case information provided by experts and stores them in a database, including loading static files and using input forms for experts.

[0732] Step 3: Preprocessing the data

[0733] The server retrieves raw comment data from the database, corrects emotional words and typos to remove noise, and standardizes synonyms and segments text to normalize it. The preprocessed data is then saved back to the database.

[0734] Step 4: Pattern learning

[0735] The server uses the preprocessed comment data as training data for a generative AI model, trains a generative AI model (e.g., BERT or GPT-4) to learn patterns of harassing behavior, and saves the trained model so it can be used for predictions.

[0736] Step 5: Enter comments and status data

[0737] The user uses the terminal to enter comments and situation data into the input form, and then presses the "Send" button to send the data to the server.

[0738] Step 6: Receive data and prepare for analysis

[0739] The device sends input data to the server, which receives it and prepares to pass the received comments and status data to the analysis process.

[0740] Step 7: Sentiment Analysis

[0741] The server uses an emotion engine to analyze the emotional state of comments and situational data received from users, which can be classified as positive, negative, neutral, etc.

[0742] Step 8: Analyze the data

[0743] The server uses the generated AI model to analyze the incoming data, taking into account the results of sentiment analysis, to evaluate the meaning and sentiment of the analyzed data and determine whether it matches a specific pattern.

[0744] Step 9: Automatic diagnosis

[0745] Based on the analysis results, the server automatically diagnoses whether the input data constitutes harassment, while also referencing expert preventative measures and case studies, and generates a diagnosis such as "This is harassment," "This is not harassment," or "It is difficult to determine."

[0746] Step 10: Sending diagnosis and sentiment analysis results

[0747] The server converts the diagnosis results and emotion analysis results into a data format and sends them to the user's terminal.

[0748] Step 11: Viewing the diagnosis and sentiment analysis results

[0749] The device displays the diagnosis results and emotion analysis results received from the server to the user, allowing the user to check the evaluation of their own comments and situations.

[0750] Step 12: Offering educational programs

[0751] The server selects an appropriate educational program based on the diagnosis and emotion analysis results, provides it to the user, and sends a notification to the user's device containing the educational program's URL and content.

[0752] Step 13: Take the education program and track your progress

[0753] The user takes the educational program provided through the terminal, which tracks the progress of the educational program and periodically reports it to the server.

[0754] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. The emotion analysis function using the emotion engine also enables more accurate diagnosis and response. Furthermore, education programs provide continuous employee education, leading to an improved work environment.

[0755] Example 2

[0756] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0757] It is important to quickly and accurately recognize harassment in the work environment and take measures to address it, but conventional systems tend to delay such recognition and countermeasures, making it difficult to alleviate employee stress and problems. Furthermore, conventional systems lack the functionality to analyze the emotional state of comments or to provide appropriate educational programs, preventing effective improvements to the work environment.

[0758] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0759] In this invention, the server includes means for learning patterns in comment data using a generative AI model, means for analyzing newly entered comments and situation data, means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, means having an emotion engine for analyzing the emotional state of the user's comments and situation data, and means for transmitting the diagnosis and emotion analysis results to the user's device. This enables rapid analysis of user input data and provides comprehensive diagnosis results including the user's emotional state, enabling continuous improvement of the work environment through appropriate educational programs.

[0760] A "generative AI model" is an algorithm that uses artificial intelligence technology to learn patterns and trends from data and make predictions and analyses on new data.

[0761] "Comment data" refers to text information, opinions, explanations of situations, etc. entered by users.

[0762] "Pattern learning" is the process of extracting specific trends or recurring features from data, understanding them, and incorporating them into a model.

[0763] "Automatic diagnosis" refers to the ability of a system to analyze data and make a determination about a specific condition or problem without human intervention.

[0764] An "emotion engine" is an algorithm that analyzes the emotions contained in text data and classifies them as positive, negative, neutral, etc.

[0765] "Diagnostic results" refer to the judgment information, such as whether or not harassment has occurred, that is shown as a result of the analysis by the generative AI model.

[0766] "Analysis results" refers to the overall results of the analysis performed by the generative AI model and emotion engine on the input data.

[0767] "Educational Program" means a course of study or training offered to enable a user to acquire specific knowledge or skills.

[0768] The system of the present invention collects user comment data and situational data and automatically analyzes and diagnoses them. This system is equipped with a pattern learning function using a generative AI model, an automatic diagnosis function, an analysis function using an emotion engine, and an educational program provision function. This enables the rapid and accurate recognition of harassment in the work environment and the implementation of countermeasures.

[0769] Server configuration and functions

[0770] Data collection features:

[0771] The server continuously collects data on comments entered by users using the company's internal surveys and chat systems, and stores the collected data in a database.

[0772] Data preprocessing functions:

[0773] The server cleanses the collected comment data, removes noise, and normalizes it. The preprocessed data is then stored in the database again.

[0774] Pattern learning function:

[0775] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[0776] Automatic diagnostic function:

[0777] Users input new comments and situational data through their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether it constitutes harassment.

[0778] Emotion engine analysis function:

[0779] The server uses an emotion engine to analyze the emotional state of comments and situational data entered by users, which are classified as positive, negative, neutral, etc.

[0780] Diagnostic and sentiment analysis results transmission function:

[0781] The server transmits the generated diagnosis results and emotion analysis results to the user's terminal, which displays these results to the user.

[0782] Educational program offerings:

[0783] The server selects an appropriate educational program based on the diagnosis and emotion analysis results and provides it to the user. The user takes the educational program through their device, and their progress is reported to the server.

[0784] Device configuration and functions

[0785] Comment and status data entry function:

[0786] It provides an interface for users to input comment data and situational data through an internal survey or chat system.

[0787] Data transmission function:

[0788] The terminal transmits the input data to the server.

[0789] Diagnostic and emotion analysis results display function:

[0790] The terminal displays the diagnosis results and emotion analysis results sent from the server to the user.

[0791] Educational Program Display & Progress Tracking Features:

[0792] The terminal displays the educational program provided by the server and tracks the user's learning progress.

[0793] Specific operation example

[0794] Example 1: Comments about appearance

[0795] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which analyzes the comment using a generative AI model. The analysis results in a diagnosis that "This is a negative comment about my appearance, and is likely to constitute harassment." The server then uses an emotion engine to analyze the comment as containing negative emotions. The results are displayed on the user's device, and further appropriate educational programs are provided.

[0796] Example 2: Ignoring comments

[0797] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an act of intentionally excluding a specific individual, and is likely to constitute harassment." The server then uses an emotion engine to analyze that the situation contains negative emotions. The results are displayed on the user's device, and an appropriate educational program is provided.

[0798] Prompt Sentence Examples

[0799] "My boss frequently makes comments about my appearance, which I find very unpleasant. What kind of diagnosis would a generative AI model make in this case?"

[0800] "My colleague keeps getting ignored in meetings. How would the emotion engine analyze this situation?"

[0801] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0802] Step 1:

[0803] A user enters a comment into an internal survey or chat system.

[0804] As a specific operation, the user inputs text such as "Recently, my boss has been frequently commenting on my appearance, and it makes me feel uncomfortable."

[0805] Input: Comment data entered by the user

[0806] Output: Comment data sent to the device

[0807] Step 2:

[0808] The terminal transmits the input comment data to the server.

[0809] As a specific operation, the terminal generates and transmits a request for transmitting the comment data input by the user to the server.

[0810] Input: Comment data entered into the terminal

[0811] Output: Comment data sent to the server

[0812] Step 3:

[0813] The server stores the received comment data in a database.

[0814] Specifically, the server stores the received comment data in an appropriate database table.

[0815] Input: Comment data received by the server

[0816] Output: Comment data stored in the database

[0817] Step 4:

[0818] The server retrieves the comment data from the database and performs data cleansing.

[0819] Specifically, the server reads unprocessed comment data from the database and performs noise removal and format conversion.

[0820] Input: Comment data stored in the database

[0821] Output: Cleansed comment data

[0822] Step 5:

[0823] The server normalizes the cleansed comment data and stores it back in the database.

[0824] Specifically, the server performs normalization processing, such as converting the text to lowercase, and stores it in the database.

[0825] Input: Cleansed comment data

[0826] Output: Normalized comment data

[0827] Step 6:

[0828] The server uses the normalized comment data to train a generative AI model.

[0829] Specifically, the server inputs the normalized data into the AI ​​model and performs training.

[0830] Input: Normalized comment data

[0831] Output: A trained generative AI model

[0832] Step 7:

[0833] The user inputs new comments and status data from the terminal, which then transmits the data to the server.

[0834] Specifically, the user inputs new text such as, "In daily meetings, certain colleagues are always ignored and denied the opportunity to speak," and the device sends this to the server.

[0835] Input: New comments and status data entered by the user

[0836] Output: New comments and status data sent to the server

[0837] Step 8:

[0838] The server uses the generated AI model to analyze newly received data.

[0839] Specifically, the server inputs new comments and situational data into the generative AI model and obtains the analysis results.

[0840] Input: New comments and status data sent to the server

[0841] Output: Parsed data

[0842] Step 9:

[0843] Based on the analysis results, the server automatically diagnoses whether the behavior constitutes harassment.

[0844] Specifically, the system refers to the analysis results and case information from experts to determine whether or not the behavior constitutes harassment.

[0845] Input: Parsed data

[0846] Output: Automatic diagnosis results

[0847] Step 10:

[0848] The server uses an emotion engine to analyze the emotional state of the comments and situation data.

[0849] Specifically, the server detects emotional keywords in the comments and classifies their emotional state into "positive," "negative," or "neutral."

[0850] Input: New comments and status data

[0851] Output: Emotion analysis results

[0852] Step 11:

[0853] The server generates diagnosis results and emotion analysis results and transmits them to the terminal.

[0854] Specifically, the server combines the diagnosis results and emotion analysis results into a single packet and sends it to the terminal.

[0855] Input: Automatic diagnosis results and emotion analysis results

[0856] Output: Result data sent to the terminal

[0857] Step 12:

[0858] The terminal displays the received diagnosis results and emotion analysis results to the user.

[0859] Specifically, the terminal provides the user with the diagnosis results and emotion analysis results as a screen display.

[0860] Input: Result data sent to the terminal

[0861] Output: Diagnosis results and sentiment analysis results displayed to the user

[0862] Step 13:

[0863] The server selects an appropriate educational program based on the diagnosis and emotion analysis results.

[0864] Specifically, the server selects and lists educational programs from a database according to the diagnostic results.

[0865] Input: Diagnosis results and emotion analysis results

[0866] Output: Selected educational programs

[0867] Step 14:

[0868] The server transmits the selected educational program to the terminal.

[0869] Specifically, the server sends links to educational programs and educational materials to the terminal.

[0870] Input: Selected Educational Program

[0871] Output: Educational program sent to the terminal

[0872] Step 15:

[0873] A user takes an educational program through a terminal, and the progress is reported to a server.

[0874] Specifically, the user watches an educational program on the terminal, and the learning progress is automatically reported to the server.

[0875] Input: Educational program progress data

[0876] Output: Progress data reported to the server

[0877] (Application example 2)

[0878] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0879] Traditional brick-and-mortar stores lack systems for quickly and accurately evaluating customer satisfaction and proposing concrete improvement measures. In particular, there is no way to accurately analyze the emotional state of customer feedback and immediately propose countermeasures for negative feedback. As a result, improvements in the customer service skills of store staff and service quality are delayed, leading to a decline in customer satisfaction.

[0880] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0881] In this invention, the server includes a means for pattern learning of comment data using a generative AI model, a means for collecting, preprocessing, and analyzing newly entered customer feedback, and a means for analyzing the emotional state of comments and situation data using an emotion analysis engine. This makes it possible to quickly and accurately analyze the emotional state of customer feedback and provide specific improvement measures and training programs for store staff.

[0882] A "generative AI model" is an artificial intelligence model that learns patterns based on collected data and analyzes new data.

[0883] "Comment Data" refers to feedback and opinions entered by users in the form of text.

[0884] An "emotion analysis engine" is software that analyzes and classifies emotional states from comments and situational data.

[0885] "Automatic diagnosis" is the process by which a generative AI model analyzes newly input data and determines whether a particular behavior constitutes harassment.

[0886] "Educational Program" refers to training and learning materials provided to users based on the analysis results.

[0887] "Expert preventive measures and case information" means information provided by experts and past cases regarding the prevention and countermeasures against harassment.

[0888] "Customer feedback" refers to the evaluations, opinions, and impressions that customers provide regarding services and products.

[0889] "Customer service skills improvement training" is an educational program designed to improve the customer service attitude and response techniques of store staff.

[0890] "Negative feedback" refers to evaluations or opinions that express customer dissatisfaction or annoyance.

[0891] "Progress tracking" is the process of monitoring and managing progress through a delivered educational program.

[0892] The present invention relates to a system for analyzing customer feedback in a physical store and improving the customer service skills of store staff. The specific system configuration and operation will be described below.

[0893] System configuration

[0894] server

[0895] The server has the following functions:

[0896] 1. Data collection function: Collect customer feedback from smartphones and tablets.

[0897] 2. Data preprocessing function: Cleanse the collected feedback data and remove noise.

[0898] 3. Pattern learning function: Trains generative AI models using preprocessed data.

[0899] 4. Automatic diagnosis function: Feedback data is analyzed using a generative AI model to diagnose specific behaviors.

[0900] 5. Sentiment Analysis Function: Uses a sentiment analysis engine to analyze the emotional state of the feedback.

[0901] 6. Diagnostic result transmission function: Transmits diagnostic results and emotion analysis results to the user's device.

[0902] 7. Educational program provision function: Based on the diagnostic results, select and provide appropriate educational programs.

[0903] Terminal

[0904] The terminal has the following features:

[0905] 1. Comment input function: Customers can input feedback using their smartphones or tablets.

[0906] 2. Data transmission function: Sends collected data to the server.

[0907] 3. Result display function: Displays the diagnosis results and emotion analysis results.

[0908] 4. Educational program progress tracking function: Tracks the progress of the educational program and reports the results to the server.

[0909] Program processing flow

[0910] 1. The user enters feedback using a smartphone or tablet. For example, they enter feedback such as, "I feel like the staff have been cold towards me lately."

[0911] 2. The device sends the input feedback data to the server.

[0912] 3. The server receives the feedback data and performs data preprocessing, which involves data cleansing and noise removal.

[0913] 4. Using the pre-processed data, the generative AI model analyzes the feedback content and diagnoses whether a particular behavior constitutes harassment.

[0914] 5. The server uses an emotion analysis engine to analyze the emotional state of the feedback, e.g., it is analyzed as "negative."

[0915] 6. The diagnosis results and emotion analysis results are sent to the store manager's terminal and displayed.

[0916] 7. Based on the results of the diagnosis, the server will provide an appropriate educational program (e.g., "Customer Service Skills Improvement Training").

[0917] Hardware and software used

[0918] Hardware:

[0919] Devices used by customers and store managers: smartphones, tablets, PCs, etc.

[0920] Server: High-performance data processing server

[0921] software:

[0922] Sentiment analysis engine: for example, a model using the transformers library

[0923] Data preprocessing and analysis: Programming languages ​​such as Python

[0924] Specific examples

[0925] As a concrete example, let's consider the case where a customer uses a smartphone. A customer inputs feedback such as, "I feel like the staff have been cold towards me lately." The data is sent to the server, where it is preprocessed and analyzed by the generative AI model. The analysis results indicate that "improvement in customer service attitude is necessary," and the sentiment analysis result is "negative." These results are then notified to the store manager, who provides training to improve customer service skills.

[0926] Prompt Sentence Examples

[0927] Feedback: "I feel like the staff have been cold lately."

[0928] Prompt for generative AI model: "Based on this feedback, please rate the customer's attitude."

[0929] In this way, the present invention makes it possible to analyze customer feedback quickly and accurately in a physical store and immediately propose countermeasures, thereby improving the customer service skills of store staff and increasing customer satisfaction.

[0930] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0931] Step 1:

[0932] The user inputs and sends feedback using a smartphone or tablet. For example, the input data might be, "Recently, I feel like the staff have been cold towards me." The device then sends this feedback to the server. The input is text data, and the output is data sent to the server.

[0933] Step 2:

[0934] The server cleanses and pre-processes the received feedback data, which includes denoising and normalizing the data. The input data is the raw feedback, and the output data is the cleansed text data.

[0935] Step 3:

[0936] The server inputs the preprocessed feedback data into a generative AI model and performs pattern learning. The model analyzes the data and diagnoses whether a particular behavior constitutes harassment. The input data is the preprocessed feedback text, and the output data is the diagnosis result.

[0937] Step 4:

[0938] The server uses a sentiment analysis engine to analyze the emotional state of the feedback, e.g., it is analyzed as "negative." The input data is the feedback text, and the output data is the sentiment analysis result.

[0939] Step 5:

[0940] The server sends the diagnosis results and emotion analysis results to the user's device. The input data here are the diagnosis results and emotion analysis results, and the output data is sent to the device. The user's device displays these results.

[0941] Step 6:

[0942] Based on the diagnosis results, the server selects and provides a relevant educational program. For example, "Training to improve customer service skills" may be suggested. The input data is the diagnosis results, and the output data is a suggested educational program. The content of the educational program is displayed on the user's device.

[0943] Step 7:

[0944] A user takes an educational program and reports his / her progress to the server via his / her terminal. The input data is the progress information of the educational program, and the output data is the progress report to the server.

[0945] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0946] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0947] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0948] [Third embodiment]

[0949] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0950] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0951] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0952] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0953] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0954] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0955] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0956] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0957] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0958] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0959] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0960] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0961] The system of the present invention is designed to provide pattern learning of comment data using a generative AI model, an automatic diagnostic function that combines preventive measures by experts and case information, and an educational program. Specific embodiments for implementing this system are described below.

[0962] System configuration

[0963] server

[0964] The server has the following functions:

[0965] 1. Data collection function

[0966] 2. Data preprocessing function

[0967] 3. Pattern learning function

[0968] 4. Automatic diagnosis function

[0969] 5. Result transmission function

[0970] 6. Educational program provision function

[0971] Terminal

[0972] The terminal has the following functions:

[0973] 1. Comment and status data input function

[0974] 2. Data transmission function

[0975] 3. Diagnostic result display function

[0976] 4. Educational program display and progress tracking function

[0977] Service flow

[0978] Data collection

[0979] When users enter comments into internal surveys, chat systems, etc., the server continuously collects them. Prevention measures and case information provided by experts are also collected and stored in a database.

[0980] Data Preprocessing

[0981] The server cleanses, denoises, and normalizes the collected comment data, and then stores the pre-processed data back in the database.

[0982] Pattern Learning

[0983] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[0984] Automatic diagnosis

[0985] Users input new comments and situational data from their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[0986] Sending and displaying results

[0987] The server generates a diagnostic result and sends it to the user's terminal, which displays the diagnostic result to the user.

[0988] Providing educational programs

[0989] Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through the terminal, and the progress is reported to the server.

[0990] Specific examples

[0991] Example 1: Comments about appearance

[0992] The user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which then analyzes the comment using a generative AI model. The analysis results in a diagnosis that "these are negative comments about my appearance, and are likely to constitute harassment." The results are displayed on the user's device, and appropriate educational programs are provided.

[0993] Example 2: Ignoring comments

[0994] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which then analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an intentional attempt to exclude certain individuals, and is likely to constitute harassment." The results are displayed on the user's device, and an appropriate educational program is provided.

[0995] This allows users and companies to quickly and accurately diagnose harassment and take appropriate measures. Furthermore, education programs can be used to continuously educate employees and improve the work environment.

[0996] The processing flow will be explained below.

[0997] Step 1: Data collection

[0998] The server continuously collects comments from various comment data sources (internal surveys, internal chats, review sites, etc.) and stores the collected comment data, along with preventive measures and case information from experts, in a database.

[0999] Step 2: Data Preprocessing

[1000] The server retrieves raw comment data from the database, performs noise removal (e.g., correcting emotional words and typos), performs text normalization (e.g., synonym unification, text segmentation), and stores the preprocessed data back in the database.

[1001] Step 3: Pattern learning

[1002] The server retrieves preprocessed comment data from the database. It uses this comment data to train a generative AI model (e.g., BERT or GPT-4). The model learns patterns of harassing behavior. It saves the trained model, making it available for prediction.

[1003] Step 4: Accepting input data

[1004] The user uses the terminal to enter comments and situation data into the input form, and then presses the "Send" button to send the data to the server.

[1005] Step 5: Analyze the data

[1006] The server receives comments and context data from the device and passes it through an analysis process. A generative AI model is used to review the input data, evaluate the meaning and sentiment of the analyzed data, and determine whether it matches a specific pattern.

[1007] Step 6: Automatic diagnosis

[1008] Based on the analysis results, the server refers to expert preventative measures and case studies to determine whether the input data constitutes harassment, and generates a diagnosis such as "This is harassment," "This is not harassment," or "It is difficult to determine."

[1009] Step 7: Submitting diagnostic results

[1010] The diagnostic results generated by the server are converted into a data format and sent to the user's terminal.

[1011] Step 8: View the diagnostic results

[1012] The terminal displays the diagnosis results received from the server to the user.

[1013] Step 9: Offering educational programs

[1014] The server selects an appropriate educational program based on the diagnosis results and sends a notification containing the educational program's URL and content to the user's device.

[1015] Step 10: Take the education program and track your progress

[1016] The user takes the educational program provided through the terminal, which tracks the progress of the educational program and periodically reports it to the server.

[1017] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. Furthermore, education programs can be used to continuously educate employees and improve the work environment.

[1018] Example 1

[1019] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1020] There is a need for a method to quickly and accurately identify harassment behaviors experienced by users in the workplace and provide appropriate countermeasures. There is also a growing need for educational and awareness programs to prevent harassment. To address these issues, the present invention provides an automated diagnostic system using a generative AI model, with the aim of improving the workplace environment.

[1021] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1022] In this invention, the server includes: means for a user to input comment data from a terminal and send it to the server; means for the server to collect the comment data and store it in a database; means for the server to cleanse and normalize the collected comment data; means for the server to train a generative AI model using the preprocessed comment data; means for the user to send newly input data from the terminal to the server and for the server to analyze the data using the generative AI model; means for the server to automatically diagnose whether the behavior constitutes harassment based on the analysis results; means for the server to generate and send the diagnostic results to the user's terminal; and means for the user to take an educational program via the terminal and report the user's progress to the server. This enables rapid and accurate diagnosis of harassment behavior and continuous education through appropriate educational programs.

[1023] A "user" is an individual who utilizes the system to input comments and situational data and receive diagnostic results and educational programs.

[1024] "Terminal" refers to a device used by a user to input comment data and situation data, send them to a server, and receive diagnostic results and educational programs.

[1025] "Server" refers to the system that collects, stores, and pre-processes comment data, trains and uses generative AI models to analyze the data, and generates and transmits diagnostic results to users' devices.

[1026] "Comment data" refers to text data of opinions and experiences about the work environment that users enter into internal surveys or chat systems.

[1027] "Data cleansing" is the process of removing noise and unnecessary information from collected comment data and putting it into a form suitable for analysis.

[1028] "Normalization" is the process of converting cleansed comment data into a consistent format to improve the accuracy of analysis.

[1029] A "generative AI model" is a model that is trained using machine learning algorithms to learn patterns of harassing behavior from comment data.

[1030] The "database" refers to a system for storing comment data, preventive measures by experts, case information, etc.

[1031] "Automatic diagnosis" refers to the process of using a generative AI model to analyze newly entered comments and situational data to determine whether they constitute harassing behavior.

[1032] "Diagnostic results" refer to an assessment of whether or not harassment has occurred, generated through an automated diagnostic process.

[1033] "Educational Program" refers to educational content provided to users with the aim of preventing and raising awareness of harassment.

[1034] "Progress" refers to the degree of completion of the educational program that the user has taken and the progress of their learning.

[1035] This invention provides a system that uses a generative AI model to learn patterns from comment data, an automatic diagnostic function that combines preventive measures and case information from experts, and an educational program. This system is implemented using the following hardware and software.

[1036] server

[1037] A server is a piece of hardware that has the following main functions:

[1038] 1. Data collection function

[1039] 2. Data preprocessing function

[1040] 3. Pattern learning function

[1041] 4. Automatic diagnosis function

[1042] 5. Result transmission function

[1043] 6. Educational program provision function

[1044] The server uses software such as Python, TensorFlow, and PyTorch to train generative AI models and analyze data. The server also houses a database system (e.g., MySQL or PostgreSQL) to store comment data, expert precautions, and case information.

[1045] Terminal

[1046] A terminal is a piece of hardware that has the following main functions:

[1047] 1. Comment and status data input function

[1048] 2. Data transmission function

[1049] 3. Diagnostic result display function

[1050] 4. Educational program display and progress tracking function

[1051] The terminal communicates data between the user and the server using a web browser or a dedicated application. The terminal sends data to the server using HTTP requests and receives diagnostic results and educational programs.

[1052] Service flow

[1053] Data collection

[1054] The user inputs comments and situation data from the device. For example, the user might input, "My boss has been frequently commenting on my appearance lately, and it makes me feel uncomfortable." This data is sent from the device to the server via an HTTP request. The server then stores the received comment data in a database.

[1055] Data Preprocessing

[1056] The server cleanses, removes noise, and normalizes the comment data stored in the database, and the data is then stored back into the database after this preprocessing.

[1057] Pattern Learning

[1058] The server uses the preprocessed comment data to train a generative AI model, a process that uses machine learning libraries such as TensorFlow and PyTorch.

[1059] Automatic diagnosis

[1060] The user inputs new comments and situational data from their device and sends it to the server. For example, data such as "In meetings, certain colleagues' opinions are being ignored" can be input. The server then uses a generative AI model to analyze the new data and automatically diagnose whether it constitutes harassment. Preventive measures and case information from experts are also referenced.

[1061] Sending and displaying diagnostic results

[1062] The server generates a diagnostic result and sends it to the user's terminal, which displays the diagnostic result to the user.

[1063] Providing educational programs

[1064] Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user's device. The user takes the educational program through their device, and their progress is reported to the server.

[1065] Specific examples

[1066] Example 1: Comments about appearance

[1067] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this data to the server, which then analyzes the comment using a generative AI model. The analysis results in a diagnosis that "these are negative comments about my appearance, and are likely to constitute harassment." The results are displayed on the user's device, and appropriate educational programs are provided.

[1068] Example 2: Ignoring comments

[1069] The user inputs the situation, such as, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which then analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an intentional attempt to exclude certain individuals, and is likely to constitute harassment." The results are displayed on the user's device, and an appropriate educational program is provided.

[1070] The system allows users and companies to quickly and accurately diagnose harassment and take appropriate measures. It also raises employee awareness through educational programs, promoting an improved work environment.

[1071] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1072] Step 1: Data entry

[1073] The user inputs comment data from the terminal. For example, "Recently, my boss has frequently commented on my appearance, and it makes me feel uncomfortable." The terminal sends the input comment data to the server via an HTTP request.

[1074] Input: Comment data entered by the user from the terminal

[1075] Output: Comment data sent to the server via an HTTP request

[1076] Step 2: Data collection

[1077] The server receives the comment data sent from the device and stores it in a database, for example, in a table called "comment_data."

[1078] Input: Comment data sent from the device

[1079] Output: Comment data to be saved in the database

[1080] Step 3: Data Preprocessing

[1081] The server retrieves raw comment data from the database and performs data cleansing, such as removing unnecessary whitespace and special characters from the text. Next, a normalization process converts all text to lowercase and standardizes synonyms. The preprocessed data is then re-stored in the "preprocessed_data" table.

[1082] Input: Comment data retrieved from the database

[1083] Output: Cleansed and normalized data

[1084] Step 4: Pattern learning

[1085] The server retrieves all data from the "preprocessed_data" table and prepares it as a training dataset. Then, it starts training the generative AI model using TensorFlow, specifically using the "fit()" method. After training, it saves the model to disk as "ai_model.h5".

[1086] Input: Preprocessed data

[1087] Output: A trained generative AI model

[1088] Step 5: Automatic diagnosis

[1089] The user enters a new comment on their device, such as "In meetings, certain colleagues' opinions are always ignored." The device sends the comment to the server via an HTTP request. The server then analyzes the received data by calling the saved generative AI model "ai_model.h5." The analysis results determine whether the comment constitutes harassment.

[1090] Input: New comment data sent from the device

[1091] Output: Diagnostic results from analysis

[1092] Step 6: Send and view diagnostic results

[1093] The server generates the analysis result, "This is an act of intentionally excluding a specific individual, and is likely to constitute harassment." Next, this diagnosis result is sent to the user's device via an HTTP response. The device then displays the received diagnosis result to the user.

[1094] Input: Analysis results

[1095] Output: Diagnostic results displayed on the user's terminal

[1096] Step 7: Offering educational programs

[1097] Based on the diagnosis results, the server selects an appropriate educational program regarding "intentional exclusion of specific individuals." It then sends the link and content of the selected educational program to the user's device via an HTTP response. The user then takes the educational program through their device and reports their progress to the server.

[1098] Input: Diagnostic results

[1099] Output: Educational program and progress information provided to the user's terminal

[1100] Through this series of processing steps, users and companies can quickly and accurately diagnose harassment behavior and take appropriate measures. Furthermore, education programs can be implemented to raise employee awareness and promote an improved work environment.

[1101] (Application example 1)

[1102] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1103] In today's workplace, harassment increases the mental and work burden on employees. Employees in brick-and-mortar stores, in particular, need prompt and accurate diagnosis and action against everyday harassment from customers, superiors, and coworkers. However, current manual reporting systems and training programs are slow to respond, resulting in long time-consuming resolutions for harassment cases.

[1104] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1105] In this invention, the server includes means for learning patterns in comment data using a generative AI model, means for analyzing newly entered comments and situation data, means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, means for transmitting the diagnosis results to the user's display device, means for transmitting comments and situation data entered by the user to the server, means for displaying an appropriate educational program based on the diagnosis results and tracking its progress, means for storing preventive measures and case information by experts in a storage device and using them for analysis and automatic diagnosis, and means for generating an appropriate educational program based on the newly entered comments and situation data, providing it to the user, and reporting its progress to the server. This enables rapid and accurate diagnosis of harassment behavior and provision of an appropriate educational program.

[1106] A "generative AI model" is an algorithm that uses artificial intelligence techniques to learn specific patterns and trends and analyze data autonomously.

[1107] "Comment data" refers to opinions or feedback in text form entered by users, and is typically in free-form format.

[1108] "Pattern learning" is the process of taking a given data set, identifying consistent features or trends in the data, and then allowing an algorithm to learn from those features.

[1109] "Contextual data" is data that contains detailed information about specific events or situations, describing a user's experience or environment.

[1110] "Analysis" is the process of examining collected data in detail and understanding its meaning and structure.

[1111] "Automatic diagnosis" is the process by which a system uses algorithms to identify problems or anomalies based on input data and generate a diagnostic result.

[1112] "Diagnosis results" are information indicating conclusions or judgments obtained through the process of analysis and automatic diagnosis.

[1113] A "user display device" is a device for visually presenting information to a user, examples of which include a smartphone or smart glasses.

[1114] A "storage device" is hardware or software for long-term storage of data or information.

[1115] "Educational Program" means a set of educational materials or training methods designed to teach specific knowledge or skills.

[1116] A system for implementing this invention uses a generative AI model to learn patterns in comment data and analyze newly entered comments and situational data to automatically diagnose harassment behavior and transmit the diagnosis results to the user's display device. Furthermore, it displays appropriate educational programs based on the diagnosis results and tracks their progress. It also stores preventive measures and case information developed by experts in a storage device and uses them for analysis and automatic diagnosis.

[1117] Hardware and Software Configuration

[1118] Server: The server has the computational capabilities to use generative AI models to learn patterns from comment data, and the ability to analyze newly entered comments and situational data. The database server stores preventive measures and case information from experts, which are used for analysis.

[1119] Terminal: A terminal is a display device such as a smartphone or smart glasses, which the user uses to input comments and situational data and check the diagnosis results. These terminals have the function of sending data from the user to the server and receiving diagnosis results and educational programs from the server.

[1120] Generative AI Model: A generative AI model is used to learn patterns of harassing behavior from comment data. This model is trained on the server and used to analyze new data.

[1121] Data processing and calculation

[1122] The server processes the data in the following steps:

[1123] 1. Data collection: Users input comments and situational data using their devices and send them to the server. Preventive measures and case information provided by experts are also collected and stored in a database.

[1124] 2. Data preprocessing: The server cleanses, denoises, and normalizes the collected comment data. After this preprocessing, the data is stored back in the database.

[1125] 3. Pattern Learning: The server uses the preprocessed comment data to train a generative AI model, which is used to learn patterns of harassing behavior.

[1126] 4. Automated Diagnosis: Users input new comments and situational data and send it to the server. The server uses a generative AI model to analyze this data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[1127] 5. Sending and displaying the results: The server generates the diagnostic results and sends them to the user's terminal, which then displays them to the user.

[1128] 6. Providing educational programs: Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through the terminal, and the progress is reported to the server.

[1129] Specific examples

[1130] Example 1: A user types into their smartphone, "My boss keeps making comments about my clothes and it's annoying me." The server receives this comment data and analyzes it using a generative AI model. The analysis results return a diagnosis that "comments about my clothes are inappropriate." An appropriate educational program is also provided along with the diagnosis.

[1131] Example 2: A user inputs, "My colleague always ignores what I say during meetings and tries to overwhelm me." The server receives this data and analyzes it using a generative AI model. The diagnosis is that "ignoring what I say constitutes harassment." The user confirms this result and is provided with relevant educational programs.

[1132] Prompt Sentence Examples

[1133] "An employee says: 'My boss keeps making comments about my attire and it's bothering me.' Make a diagnosis based on this comment and provide an appropriate training program."

[1134] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1135] Step 1: Data collection

[1136] The user uses the device to input comments and status data and sends it to the server. This input includes opinions and feedback in text format. The device receives the input data and sends it to the server using an HTTP request. The server receives this data and stores it in a database.

[1137] Step 2: Data Preprocessing

[1138] The server cleanses, removes noise, and normalizes the comment data stored in the database. For example, it removes unnecessary symbols and emojis such as "!!!" and "???." It also standardizes the same content expressed in different formats (e.g., "Hello" and "Hello"). After this preprocessing, the data is stored back in the database.

[1139] Step 3: Pattern learning

[1140] The server uses the preprocessed comment data to train the generative AI model. During pattern learning, the cleansed comments are used as input data and patterns of harassing behavior are identified as output. The model iteratively analyzes the data to improve its accuracy. This training process is run periodically.

[1141] Step 4: Automatic diagnosis

[1142] The user inputs new comments and situational data into the device and sends it to the server. The server then provides the received data to the generative AI model for analysis. This analysis determines whether the input data constitutes harassment, and a diagnosis result is generated as the output.

[1143] Step 5: Send and view results

[1144] The server generates the results of the automated diagnosis and sends them to the user's device. The device then visually displays the received results to the user. For example, it may display a result such as "This comment is likely to constitute harassment."

[1145] Step 6: Offering educational programs

[1146] Based on the diagnosis results, the server selects an appropriate educational program and sends it to the user's device. The user then takes the educational program provided through the device. The device periodically reports the user's progress to the server. The server uses this information to adjust the next educational program.

[1147] Prompt Sentence Examples

[1148] "An employee says: 'My boss keeps making comments about my attire and it's bothering me.' Make a diagnosis based on this comment and provide an appropriate training program."

[1149] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1150] The system of the present invention is equipped with a generative AI model that learns patterns from comment data, an automatic diagnostic function that combines preventive measures and case information from experts, the provision of educational programs, and an emotion engine that recognizes user emotions. Specific embodiments for implementing this system are described below.

[1151] System configuration

[1152] server

[1153] The server has the following functions:

[1154] 1. Data collection function

[1155] 2. Data preprocessing function

[1156] 3. Pattern learning function

[1157] 4. Automatic diagnosis function

[1158] 5. Analysis function using emotion engine

[1159] 6. Diagnostic and emotion analysis results transmission function

[1160] 7. Educational program provision function

[1161] 8. Database management functions

[1162] Terminal

[1163] The terminal has the following functions:

[1164] 1. Comment and status data input function

[1165] 2. Data transmission function

[1166] 3. Diagnostic and emotion analysis results display function

[1167] 4. Educational program display and progress tracking function

[1168] Service flow

[1169] Data collection

[1170] When users enter comments into internal surveys, chat systems, etc., the server continuously collects them. Prevention measures and case information provided by experts are also collected and stored in a database.

[1171] Data Preprocessing

[1172] The server cleanses, denoises, and normalizes the collected comment data, and then stores the pre-processed data back in the database.

[1173] Pattern Learning

[1174] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[1175] Automatic diagnosis

[1176] Users input new comments and situational data from their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[1177] Emotion analysis

[1178] The server uses an emotion engine to analyze the emotional state of comments and situational data entered by users, which are then classified as positive, negative, neutral, etc.

[1179] Sending and displaying results

[1180] The server generates diagnosis results and emotion analysis results and sends them to the user's terminal, which then displays the diagnosis results and emotion analysis results to the user.

[1181] Providing educational programs

[1182] Based on the diagnosis and emotion analysis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through their device, and their progress is reported to the server.

[1183] Specific examples

[1184] Example 1: Comments about appearance

[1185] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which analyzes the comment using a generative AI model. The analysis results in a diagnosis that "This is a negative comment about my appearance, and is likely to constitute harassment." The server then uses an emotion engine to analyze the comment as containing negative emotions. The results are displayed on the user's device, and further appropriate educational programs are provided.

[1186] Example 2: Ignoring comments

[1187] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an act of intentionally excluding a specific individual, and is likely to constitute harassment." The server then uses an emotion engine to analyze that the situation contains negative emotions. The results are displayed on the user's device, and an appropriate educational program is provided.

[1188] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. The emotion analysis function using the emotion engine also enables more accurate diagnosis and response. Furthermore, education programs provide continuous employee education, leading to an improved work environment.

[1189] The processing flow will be explained below.

[1190] Step 1: Collecting comment data

[1191] The server continuously collects comments from various comment data sources (internal surveys, internal chats, review sites, etc.), including data extraction using APIs and CSV file imports.

[1192] Step 2: Gather expert information

[1193] The server collects harassment prevention measures and case information provided by experts and stores them in a database, including loading static files and using input forms for experts.

[1194] Step 3: Preprocessing the data

[1195] The server retrieves raw comment data from the database, corrects emotional words and typos to remove noise, and standardizes synonyms and segments text to normalize it. The preprocessed data is then saved back to the database.

[1196] Step 4: Pattern learning

[1197] The server uses the preprocessed comment data as training data for a generative AI model, trains a generative AI model (e.g., BERT or GPT-4) to learn patterns of harassing behavior, and saves the trained model so it can be used for predictions.

[1198] Step 5: Enter comments and status data

[1199] The user uses the terminal to enter comments and situation data into the input form, and then presses the "Send" button to send the data to the server.

[1200] Step 6: Receive data and prepare for analysis

[1201] The device sends input data to the server, which receives it and prepares to pass the received comments and status data to the analysis process.

[1202] Step 7: Sentiment Analysis

[1203] The server uses an emotion engine to analyze the emotional state of comments and situational data received from users, which can be classified as positive, negative, neutral, etc.

[1204] Step 8: Analyze the data

[1205] The server uses the generated AI model to analyze the incoming data, taking into account the results of sentiment analysis, to evaluate the meaning and sentiment of the analyzed data and determine whether it matches a specific pattern.

[1206] Step 9: Automatic diagnosis

[1207] Based on the analysis results, the server automatically diagnoses whether the input data constitutes harassment, while also referencing expert preventative measures and case studies, and generates a diagnosis such as "This is harassment," "This is not harassment," or "It is difficult to determine."

[1208] Step 10: Sending diagnosis and sentiment analysis results

[1209] The server converts the diagnosis results and emotion analysis results into a data format and sends them to the user's terminal.

[1210] Step 11: Viewing the diagnosis and sentiment analysis results

[1211] The device displays the diagnosis results and emotion analysis results received from the server to the user, allowing the user to check the evaluation of their own comments and situations.

[1212] Step 12: Offering educational programs

[1213] The server selects an appropriate educational program based on the diagnosis and emotion analysis results, provides it to the user, and sends a notification to the user's device containing the educational program's URL and content.

[1214] Step 13: Take the education program and track your progress

[1215] The user takes the educational program provided through the terminal, which tracks the progress of the educational program and periodically reports it to the server.

[1216] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. The emotion analysis function using the emotion engine also enables more accurate diagnosis and response. Furthermore, education programs provide continuous employee education, leading to an improved work environment.

[1217] Example 2

[1218] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1219] It is important to quickly and accurately recognize harassment in the work environment and take measures to address it, but conventional systems tend to delay such recognition and countermeasures, making it difficult to alleviate employee stress and problems. Furthermore, conventional systems lack the functionality to analyze the emotional state of comments or to provide appropriate educational programs, preventing effective improvements to the work environment.

[1220] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1221] In this invention, the server includes means for learning patterns in comment data using a generative AI model, means for analyzing newly entered comments and situation data, means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, means having an emotion engine for analyzing the emotional state of the user's comments and situation data, and means for transmitting the diagnosis and emotion analysis results to the user's device. This enables rapid analysis of user input data and provides comprehensive diagnosis results including the user's emotional state, enabling continuous improvement of the work environment through appropriate educational programs.

[1222] A "generative AI model" is an algorithm that uses artificial intelligence technology to learn patterns and trends from data and make predictions and analyses on new data.

[1223] "Comment data" refers to text information, opinions, explanations of situations, etc. entered by users.

[1224] "Pattern learning" is the process of extracting specific trends or recurring features from data, understanding them, and incorporating them into a model.

[1225] "Automatic diagnosis" refers to the ability of a system to analyze data and make a determination about a specific condition or problem without human intervention.

[1226] An "emotion engine" is an algorithm that analyzes the emotions contained in text data and classifies them as positive, negative, neutral, etc.

[1227] "Diagnostic results" refer to the judgment information, such as whether or not harassment has occurred, that is shown as a result of the analysis by the generative AI model.

[1228] "Analysis results" refers to the overall results of the analysis performed by the generative AI model and emotion engine on the input data.

[1229] "Educational Program" means a course of study or training offered to enable a user to acquire specific knowledge or skills.

[1230] The system of the present invention collects user comment data and situational data and automatically analyzes and diagnoses them. This system is equipped with a pattern learning function using a generative AI model, an automatic diagnosis function, an analysis function using an emotion engine, and an educational program provision function. This enables the rapid and accurate recognition of harassment in the work environment and the implementation of countermeasures.

[1231] Server configuration and functions

[1232] Data collection features:

[1233] The server continuously collects data on comments entered by users using the company's internal surveys and chat systems, and stores the collected data in a database.

[1234] Data preprocessing functions:

[1235] The server cleanses the collected comment data, removes noise, and normalizes it. The preprocessed data is then stored in the database again.

[1236] Pattern learning function:

[1237] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[1238] Automatic diagnostic function:

[1239] Users input new comments and situational data through their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether it constitutes harassment.

[1240] Emotion engine analysis function:

[1241] The server uses an emotion engine to analyze the emotional state of comments and situational data entered by users, which are classified as positive, negative, neutral, etc.

[1242] Diagnostic and sentiment analysis results transmission function:

[1243] The server transmits the generated diagnosis results and emotion analysis results to the user's terminal, which displays these results to the user.

[1244] Educational program offerings:

[1245] The server selects an appropriate educational program based on the diagnosis and emotion analysis results and provides it to the user. The user takes the educational program through their device, and their progress is reported to the server.

[1246] Device configuration and functions

[1247] Comment and status data entry function:

[1248] It provides an interface for users to input comment data and situational data through an internal survey or chat system.

[1249] Data transmission function:

[1250] The terminal transmits the input data to the server.

[1251] Diagnostic and emotion analysis results display function:

[1252] The terminal displays the diagnosis results and emotion analysis results sent from the server to the user.

[1253] Educational Program Display & Progress Tracking Features:

[1254] The terminal displays the educational program provided by the server and tracks the user's learning progress.

[1255] Specific operation example

[1256] Example 1: Comments about appearance

[1257] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which analyzes the comment using a generative AI model. The analysis results in a diagnosis that "This is a negative comment about my appearance, and is likely to constitute harassment." The server then uses an emotion engine to analyze the comment as containing negative emotions. The results are displayed on the user's device, and further appropriate educational programs are provided.

[1258] Example 2: Ignoring comments

[1259] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an act of intentionally excluding a specific individual, and is likely to constitute harassment." The server then uses an emotion engine to analyze that the situation contains negative emotions. The results are displayed on the user's device, and an appropriate educational program is provided.

[1260] Prompt Sentence Examples

[1261] "My boss frequently makes comments about my appearance, which I find very unpleasant. What kind of diagnosis would a generative AI model make in this case?"

[1262] "My colleague keeps getting ignored in meetings. How would the emotion engine analyze this situation?"

[1263] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1264] Step 1:

[1265] A user enters a comment into an internal survey or chat system.

[1266] As a specific operation, the user inputs text such as "Recently, my boss has been frequently commenting on my appearance, and it makes me feel uncomfortable."

[1267] Input: Comment data entered by the user

[1268] Output: Comment data sent to the device

[1269] Step 2:

[1270] The terminal transmits the input comment data to the server.

[1271] As a specific operation, the terminal generates and transmits a request for transmitting the comment data input by the user to the server.

[1272] Input: Comment data entered into the terminal

[1273] Output: Comment data sent to the server

[1274] Step 3:

[1275] The server stores the received comment data in a database.

[1276] Specifically, the server stores the received comment data in an appropriate database table.

[1277] Input: Comment data received by the server

[1278] Output: Comment data stored in the database

[1279] Step 4:

[1280] The server retrieves the comment data from the database and performs data cleansing.

[1281] Specifically, the server reads unprocessed comment data from the database and performs noise removal and format conversion.

[1282] Input: Comment data stored in the database

[1283] Output: Cleansed comment data

[1284] Step 5:

[1285] The server normalizes the cleansed comment data and stores it back in the database.

[1286] Specifically, the server performs normalization processing, such as converting the text to lowercase, and stores it in the database.

[1287] Input: Cleansed comment data

[1288] Output: Normalized comment data

[1289] Step 6:

[1290] The server uses the normalized comment data to train a generative AI model.

[1291] Specifically, the server inputs the normalized data into the AI ​​model and performs training.

[1292] Input: Normalized comment data

[1293] Output: A trained generative AI model

[1294] Step 7:

[1295] The user inputs new comments and status data from the terminal, which then transmits the data to the server.

[1296] Specifically, the user inputs new text such as, "In daily meetings, certain colleagues are always ignored and denied the opportunity to speak," and the device sends this to the server.

[1297] Input: New comments and status data entered by the user

[1298] Output: New comments and status data sent to the server

[1299] Step 8:

[1300] The server uses the generated AI model to analyze newly received data.

[1301] Specifically, the server inputs new comments and situational data into the generative AI model and obtains the analysis results.

[1302] Input: New comments and status data sent to the server

[1303] Output: Parsed data

[1304] Step 9:

[1305] Based on the analysis results, the server automatically diagnoses whether the behavior constitutes harassment.

[1306] Specifically, the system refers to the analysis results and case information from experts to determine whether or not the behavior constitutes harassment.

[1307] Input: Parsed data

[1308] Output: Automatic diagnosis results

[1309] Step 10:

[1310] The server uses an emotion engine to analyze the emotional state of the comments and situation data.

[1311] Specifically, the server detects emotional keywords in the comments and classifies their emotional state into "positive," "negative," or "neutral."

[1312] Input: New comments and status data

[1313] Output: Emotion analysis results

[1314] Step 11:

[1315] The server generates diagnosis results and emotion analysis results and transmits them to the terminal.

[1316] Specifically, the server combines the diagnosis results and emotion analysis results into a single packet and sends it to the terminal.

[1317] Input: Automatic diagnosis results and emotion analysis results

[1318] Output: Result data sent to the terminal

[1319] Step 12:

[1320] The terminal displays the received diagnosis results and emotion analysis results to the user.

[1321] Specifically, the terminal provides the user with the diagnosis results and emotion analysis results as a screen display.

[1322] Input: Result data sent to the terminal

[1323] Output: Diagnosis results and sentiment analysis results displayed to the user

[1324] Step 13:

[1325] The server selects an appropriate educational program based on the diagnosis and emotion analysis results.

[1326] Specifically, the server selects and lists educational programs from a database according to the diagnostic results.

[1327] Input: Diagnosis results and emotion analysis results

[1328] Output: Selected educational programs

[1329] Step 14:

[1330] The server transmits the selected educational program to the terminal.

[1331] Specifically, the server sends links to educational programs and educational materials to the terminal.

[1332] Input: Selected Educational Program

[1333] Output: Educational program sent to the terminal

[1334] Step 15:

[1335] A user takes an educational program through a terminal, and the progress is reported to a server.

[1336] Specifically, the user watches an educational program on the terminal, and the learning progress is automatically reported to the server.

[1337] Input: Educational program progress data

[1338] Output: Progress data reported to the server

[1339] (Application example 2)

[1340] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1341] Traditional brick-and-mortar stores lack systems for quickly and accurately evaluating customer satisfaction and proposing concrete improvement measures. In particular, there is no way to accurately analyze the emotional state of customer feedback and immediately propose countermeasures for negative feedback. As a result, improvements in the customer service skills of store staff and service quality are delayed, leading to a decline in customer satisfaction.

[1342] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1343] In this invention, the server includes a means for pattern learning of comment data using a generative AI model, a means for collecting, preprocessing, and analyzing newly entered customer feedback, and a means for analyzing the emotional state of comments and situation data using an emotion analysis engine. This makes it possible to quickly and accurately analyze the emotional state of customer feedback and provide specific improvement measures and training programs for store staff.

[1344] A "generative AI model" is an artificial intelligence model that learns patterns based on collected data and analyzes new data.

[1345] "Comment Data" refers to feedback and opinions entered by users in the form of text.

[1346] An "emotion analysis engine" is software that analyzes and classifies emotional states from comments and situational data.

[1347] "Automatic diagnosis" is the process by which a generative AI model analyzes newly input data and determines whether a particular behavior constitutes harassment.

[1348] "Educational Program" refers to training and learning materials provided to users based on the analysis results.

[1349] "Expert preventive measures and case information" means information provided by experts and past cases regarding the prevention and countermeasures against harassment.

[1350] "Customer feedback" refers to the evaluations, opinions, and impressions that customers provide regarding services and products.

[1351] "Customer service skills improvement training" is an educational program designed to improve the customer service attitude and response techniques of store staff.

[1352] "Negative feedback" refers to evaluations or opinions that express customer dissatisfaction or annoyance.

[1353] "Progress tracking" is the process of monitoring and managing progress through a delivered educational program.

[1354] The present invention relates to a system for analyzing customer feedback in a physical store and improving the customer service skills of store staff. The specific system configuration and operation will be described below.

[1355] System configuration

[1356] server

[1357] The server has the following functions:

[1358] 1. Data collection function: Collect customer feedback from smartphones and tablets.

[1359] 2. Data preprocessing function: Cleanse the collected feedback data and remove noise.

[1360] 3. Pattern learning function: Trains generative AI models using preprocessed data.

[1361] 4. Automatic diagnosis function: Feedback data is analyzed using a generative AI model to diagnose specific behaviors.

[1362] 5. Sentiment Analysis Function: Uses a sentiment analysis engine to analyze the emotional state of the feedback.

[1363] 6. Diagnostic result transmission function: Transmits diagnostic results and emotion analysis results to the user's device.

[1364] 7. Educational program provision function: Based on the diagnostic results, select and provide appropriate educational programs.

[1365] Terminal

[1366] The terminal has the following features:

[1367] 1. Comment input function: Customers can input feedback using their smartphones or tablets.

[1368] 2. Data transmission function: Sends collected data to the server.

[1369] 3. Result display function: Displays the diagnosis results and emotion analysis results.

[1370] 4. Educational program progress tracking function: Tracks the progress of the educational program and reports the results to the server.

[1371] Program processing flow

[1372] 1. The user enters feedback using a smartphone or tablet. For example, they enter feedback such as, "I feel like the staff have been cold towards me lately."

[1373] 2. The device sends the input feedback data to the server.

[1374] 3. The server receives the feedback data and performs data preprocessing, which involves data cleansing and noise removal.

[1375] 4. Using the pre-processed data, the generative AI model analyzes the feedback content and diagnoses whether a particular behavior constitutes harassment.

[1376] 5. The server uses an emotion analysis engine to analyze the emotional state of the feedback, e.g., it is analyzed as "negative."

[1377] 6. The diagnosis results and emotion analysis results are sent to the store manager's terminal and displayed.

[1378] 7. Based on the results of the diagnosis, the server will provide an appropriate educational program (e.g., "Customer Service Skills Improvement Training").

[1379] Hardware and software used

[1380] Hardware:

[1381] Devices used by customers and store managers: smartphones, tablets, PCs, etc.

[1382] Server: High-performance data processing server

[1383] software:

[1384] Sentiment analysis engine: for example, a model using the transformers library

[1385] Data preprocessing and analysis: Programming languages ​​such as Python

[1386] Specific examples

[1387] As a concrete example, let's consider the case where a customer uses a smartphone. A customer inputs feedback such as, "I feel like the staff have been cold towards me lately." The data is sent to the server, where it is preprocessed and analyzed by the generative AI model. The analysis results indicate that "improvement in customer service attitude is necessary," and the sentiment analysis result is "negative." These results are then notified to the store manager, who provides training to improve customer service skills.

[1388] Prompt Sentence Examples

[1389] Feedback: "I feel like the staff have been cold lately."

[1390] Prompt for generative AI model: "Based on this feedback, please rate the customer's attitude."

[1391] In this way, the present invention makes it possible to analyze customer feedback quickly and accurately in a physical store and immediately propose countermeasures, thereby improving the customer service skills of store staff and increasing customer satisfaction.

[1392] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1393] Step 1:

[1394] The user inputs and sends feedback using a smartphone or tablet. For example, the input data might be, "Recently, I feel like the staff have been cold towards me." The device then sends this feedback to the server. The input is text data, and the output is data sent to the server.

[1395] Step 2:

[1396] The server cleanses and pre-processes the received feedback data, which includes denoising and normalizing the data. The input data is the raw feedback, and the output data is the cleansed text data.

[1397] Step 3:

[1398] The server inputs the preprocessed feedback data into a generative AI model and performs pattern learning. The model analyzes the data and diagnoses whether a particular behavior constitutes harassment. The input data is the preprocessed feedback text, and the output data is the diagnosis result.

[1399] Step 4:

[1400] The server uses a sentiment analysis engine to analyze the emotional state of the feedback, e.g., it is analyzed as "negative." The input data is the feedback text, and the output data is the sentiment analysis result.

[1401] Step 5:

[1402] The server sends the diagnosis results and emotion analysis results to the user's device. The input data here are the diagnosis results and emotion analysis results, and the output data is sent to the device. The user's device displays these results.

[1403] Step 6:

[1404] Based on the diagnosis results, the server selects and provides a relevant educational program. For example, "Training to improve customer service skills" may be suggested. The input data is the diagnosis results, and the output data is a suggested educational program. The content of the educational program is displayed on the user's device.

[1405] Step 7:

[1406] A user takes an educational program and reports his / her progress to the server via his / her terminal. The input data is the progress information of the educational program, and the output data is the progress report to the server.

[1407] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1408] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1409] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1410] [Fourth embodiment]

[1411] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1412] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1413] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1414] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1415] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1416] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1417] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1418] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1419] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1420] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1421] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1422] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1423] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1424] The system of the present invention is designed to provide pattern learning of comment data using a generative AI model, an automatic diagnostic function that combines preventive measures by experts and case information, and an educational program. Specific embodiments for implementing this system are described below.

[1425] System configuration

[1426] server

[1427] The server has the following functions:

[1428] 1. Data collection function

[1429] 2. Data preprocessing function

[1430] 3. Pattern learning function

[1431] 4. Automatic diagnosis function

[1432] 5. Result transmission function

[1433] 6. Educational program provision function

[1434] Terminal

[1435] The terminal has the following functions:

[1436] 1. Comment and status data input function

[1437] 2. Data transmission function

[1438] 3. Diagnostic result display function

[1439] 4. Educational program display and progress tracking function

[1440] Service flow

[1441] Data collection

[1442] When users enter comments into internal surveys, chat systems, etc., the server continuously collects them. Prevention measures and case information provided by experts are also collected and stored in a database.

[1443] Data Preprocessing

[1444] The server cleanses, denoises, and normalizes the collected comment data, and then stores the pre-processed data back in the database.

[1445] Pattern Learning

[1446] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[1447] Automatic diagnosis

[1448] Users input new comments and situational data from their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[1449] Sending and displaying results

[1450] The server generates a diagnostic result and sends it to the user's terminal, which displays the diagnostic result to the user.

[1451] Providing educational programs

[1452] Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through the terminal, and the progress is reported to the server.

[1453] Specific examples

[1454] Example 1: Comments about appearance

[1455] The user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which then analyzes the comment using a generative AI model. The analysis results in a diagnosis that "these are negative comments about my appearance, and are likely to constitute harassment." The results are displayed on the user's device, and appropriate educational programs are provided.

[1456] Example 2: Ignoring comments

[1457] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which then analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an intentional attempt to exclude certain individuals, and is likely to constitute harassment." The results are displayed on the user's device, and an appropriate educational program is provided.

[1458] This allows users and companies to quickly and accurately diagnose harassment and take appropriate measures. Furthermore, education programs can be used to continuously educate employees and improve the work environment.

[1459] The processing flow will be explained below.

[1460] Step 1: Data collection

[1461] The server continuously collects comments from various comment data sources (internal surveys, internal chats, review sites, etc.) and stores the collected comment data, along with preventive measures and case information from experts, in a database.

[1462] Step 2: Data Preprocessing

[1463] The server retrieves raw comment data from the database, performs noise removal (e.g., correcting emotional words and typos), performs text normalization (e.g., synonym unification, text segmentation), and stores the preprocessed data back in the database.

[1464] Step 3: Pattern learning

[1465] The server retrieves preprocessed comment data from the database. It uses this comment data to train a generative AI model (e.g., BERT or GPT-4). The model learns patterns of harassing behavior. It saves the trained model, making it available for prediction.

[1466] Step 4: Accepting input data

[1467] The user uses the terminal to enter comments and situation data into the input form, and then presses the "Send" button to send the data to the server.

[1468] Step 5: Analyze the data

[1469] The server receives comments and context data from the device and passes it through an analysis process. A generative AI model is used to review the input data, evaluate the meaning and sentiment of the analyzed data, and determine whether it matches a specific pattern.

[1470] Step 6: Automatic diagnosis

[1471] Based on the analysis results, the server refers to expert preventative measures and case studies to determine whether the input data constitutes harassment, and generates a diagnosis such as "This is harassment," "This is not harassment," or "It is difficult to determine."

[1472] Step 7: Submitting diagnostic results

[1473] The diagnostic results generated by the server are converted into a data format and sent to the user's terminal.

[1474] Step 8: View the diagnostic results

[1475] The terminal displays the diagnosis results received from the server to the user.

[1476] Step 9: Offering educational programs

[1477] The server selects an appropriate educational program based on the diagnosis results and sends a notification containing the educational program's URL and content to the user's device.

[1478] Step 10: Take the education program and track your progress

[1479] The user takes the educational program provided through the terminal, which tracks the progress of the educational program and periodically reports it to the server.

[1480] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. Furthermore, education programs can be used to continuously educate employees and improve the work environment.

[1481] Example 1

[1482] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1483] There is a need for a method to quickly and accurately identify harassment behaviors experienced by users in the workplace and provide appropriate countermeasures. There is also a growing need for educational and awareness programs to prevent harassment. To address these issues, the present invention provides an automated diagnostic system using a generative AI model, with the aim of improving the workplace environment.

[1484] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1485] In this invention, the server includes: means for a user to input comment data from a terminal and send it to the server; means for the server to collect the comment data and store it in a database; means for the server to cleanse and normalize the collected comment data; means for the server to train a generative AI model using the preprocessed comment data; means for the user to send newly input data from the terminal to the server and for the server to analyze the data using the generative AI model; means for the server to automatically diagnose whether the behavior constitutes harassment based on the analysis results; means for the server to generate and send the diagnostic results to the user's terminal; and means for the user to take an educational program via the terminal and report the user's progress to the server. This enables rapid and accurate diagnosis of harassment behavior and continuous education through appropriate educational programs.

[1486] A "user" is an individual who utilizes the system to input comments and situational data and receive diagnostic results and educational programs.

[1487] "Terminal" refers to a device used by a user to input comment data and situation data, send them to a server, and receive diagnostic results and educational programs.

[1488] "Server" refers to the system that collects, stores, and pre-processes comment data, trains and uses generative AI models to analyze the data, and generates and transmits diagnostic results to users' devices.

[1489] "Comment data" refers to text data of opinions and experiences about the work environment that users enter into internal surveys or chat systems.

[1490] "Data cleansing" is the process of removing noise and unnecessary information from collected comment data and putting it into a form suitable for analysis.

[1491] "Normalization" is the process of converting cleansed comment data into a consistent format to improve the accuracy of analysis.

[1492] A "generative AI model" is a model that is trained using machine learning algorithms to learn patterns of harassing behavior from comment data.

[1493] The "database" refers to a system for storing comment data, preventive measures by experts, case information, etc.

[1494] "Automatic diagnosis" refers to the process of using a generative AI model to analyze newly entered comments and situational data to determine whether they constitute harassing behavior.

[1495] "Diagnostic results" refer to an assessment of whether or not harassment has occurred, generated through an automated diagnostic process.

[1496] "Educational Program" refers to educational content provided to users with the aim of preventing and raising awareness of harassment.

[1497] "Progress" refers to the degree of completion of the educational program that the user has taken and the progress of their learning.

[1498] This invention provides a system that uses a generative AI model to learn patterns from comment data, an automatic diagnostic function that combines preventive measures and case information from experts, and an educational program. This system is implemented using the following hardware and software.

[1499] server

[1500] A server is a piece of hardware that has the following main functions:

[1501] 1. Data collection function

[1502] 2. Data preprocessing function

[1503] 3. Pattern learning function

[1504] 4. Automatic diagnosis function

[1505] 5. Result transmission function

[1506] 6. Educational program provision function

[1507] The server uses software such as Python, TensorFlow, and PyTorch to train generative AI models and analyze data. The server also houses a database system (e.g., MySQL or PostgreSQL) to store comment data, expert precautions, and case information.

[1508] Terminal

[1509] A terminal is a piece of hardware that has the following main functions:

[1510] 1. Comment and status data input function

[1511] 2. Data transmission function

[1512] 3. Diagnostic result display function

[1513] 4. Educational program display and progress tracking function

[1514] The terminal communicates data between the user and the server using a web browser or a dedicated application. The terminal sends data to the server using HTTP requests and receives diagnostic results and educational programs.

[1515] Service flow

[1516] Data collection

[1517] The user inputs comments and situation data from the device. For example, the user might input, "My boss has been frequently commenting on my appearance lately, and it makes me feel uncomfortable." This data is sent from the device to the server via an HTTP request. The server then stores the received comment data in a database.

[1518] Data Preprocessing

[1519] The server cleanses, removes noise, and normalizes the comment data stored in the database, and the data is then stored back into the database after this preprocessing.

[1520] Pattern Learning

[1521] The server uses the preprocessed comment data to train a generative AI model, a process that uses machine learning libraries such as TensorFlow and PyTorch.

[1522] Automatic diagnosis

[1523] The user inputs new comments and situational data from their device and sends it to the server. For example, data such as "In meetings, certain colleagues' opinions are being ignored" can be input. The server then uses a generative AI model to analyze the new data and automatically diagnose whether it constitutes harassment. Preventive measures and case information from experts are also referenced.

[1524] Sending and displaying diagnostic results

[1525] The server generates a diagnostic result and sends it to the user's terminal, which displays the diagnostic result to the user.

[1526] Providing educational programs

[1527] Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user's device. The user takes the educational program through their device, and their progress is reported to the server.

[1528] Specific examples

[1529] Example 1: Comments about appearance

[1530] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this data to the server, which then analyzes the comment using a generative AI model. The analysis results in a diagnosis that "these are negative comments about my appearance, and are likely to constitute harassment." The results are displayed on the user's device, and appropriate educational programs are provided.

[1531] Example 2: Ignoring comments

[1532] The user inputs the situation, such as, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which then analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an intentional attempt to exclude certain individuals, and is likely to constitute harassment." The results are displayed on the user's device, and an appropriate educational program is provided.

[1533] The system allows users and companies to quickly and accurately diagnose harassment and take appropriate measures. It also raises employee awareness through educational programs, promoting an improved work environment.

[1534] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1535] Step 1: Data entry

[1536] The user inputs comment data from the terminal. For example, "Recently, my boss has frequently commented on my appearance, and it makes me feel uncomfortable." The terminal sends the input comment data to the server via an HTTP request.

[1537] Input: Comment data entered by the user from the terminal

[1538] Output: Comment data sent to the server via an HTTP request

[1539] Step 2: Data collection

[1540] The server receives the comment data sent from the device and stores it in a database, for example, in a table called "comment_data."

[1541] Input: Comment data sent from the device

[1542] Output: Comment data to be saved in the database

[1543] Step 3: Data Preprocessing

[1544] The server retrieves raw comment data from the database and performs data cleansing, such as removing unnecessary whitespace and special characters from the text. Next, a normalization process converts all text to lowercase and standardizes synonyms. The preprocessed data is then re-stored in the "preprocessed_data" table.

[1545] Input: Comment data retrieved from the database

[1546] Output: Cleansed and normalized data

[1547] Step 4: Pattern learning

[1548] The server retrieves all data from the "preprocessed_data" table and prepares it as a training dataset. Then, it starts training the generative AI model using TensorFlow, specifically using the "fit()" method. After training, it saves the model to disk as "ai_model.h5".

[1549] Input: Preprocessed data

[1550] Output: A trained generative AI model

[1551] Step 5: Automatic diagnosis

[1552] The user enters a new comment on their device, such as "In meetings, certain colleagues' opinions are ignored." The device sends the comment to the server via an HTTP request. The server then analyzes the received data by calling the saved generative AI model "ai_model.h5." The analysis results determine whether the comment constitutes harassment.

[1553] Input: New comment data sent from the device

[1554] Output: Diagnostic results from analysis

[1555] Step 6: Send and view diagnostic results

[1556] The server generates the analysis result, "This is an act of intentionally excluding a specific individual, and is likely to constitute harassment." Next, this diagnosis result is sent to the user's device via an HTTP response. The device then displays the received diagnosis result to the user.

[1557] Input: Diagnostic results from analysis

[1558] Output: Diagnostic results displayed on the user's terminal

[1559] Step 7: Offering educational programs

[1560] Based on the diagnosis results, the server selects an appropriate educational program regarding "intentional exclusion of specific individuals." It then sends the link and content of the selected educational program to the user's device via an HTTP response. The user then takes the educational program through their device and reports their progress to the server.

[1561] Input: Diagnostic results

[1562] Output: Educational program and progress information provided to the user's terminal

[1563] Through this series of processing steps, users and companies can quickly and accurately diagnose harassment behavior and take appropriate measures. Furthermore, education programs can be implemented to raise employee awareness and promote an improved work environment.

[1564] (Application example 1)

[1565] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1566] In today's workplace, harassment increases the mental and work burden on employees. Employees in brick-and-mortar stores, in particular, need prompt and accurate diagnosis and action against everyday harassment from customers, superiors, and coworkers. However, current manual reporting systems and training programs are slow to respond, resulting in long time-consuming resolutions for harassment cases.

[1567] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1568] In this invention, the server includes means for learning patterns in comment data using a generative AI model, means for analyzing newly entered comments and situation data, means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, means for transmitting the diagnosis results to the user's display device, means for transmitting comments and situation data entered by the user to the server, means for displaying an appropriate educational program based on the diagnosis results and tracking its progress, means for storing preventive measures and case information by experts in a storage device and using them for analysis and automatic diagnosis, and means for generating an appropriate educational program based on the newly entered comments and situation data, providing it to the user, and reporting its progress to the server. This enables rapid and accurate diagnosis of harassment behavior and provision of an appropriate educational program.

[1569] A "generative AI model" is an algorithm that uses artificial intelligence techniques to learn specific patterns and trends and analyze data autonomously.

[1570] "Comment data" refers to opinions or feedback in text form entered by users, and is typically in free-form format.

[1571] "Pattern learning" is the process of taking a given data set, identifying consistent features or trends in the data, and then allowing an algorithm to learn from those features.

[1572] "Contextual data" is data that contains detailed information about specific events or situations, describing a user's experience or environment.

[1573] "Analysis" is the process of examining collected data in detail and understanding its meaning and structure.

[1574] "Automatic diagnosis" is the process by which a system uses algorithms to identify problems or anomalies based on input data and generate a diagnostic result.

[1575] "Diagnosis results" are information indicating conclusions or judgments obtained through the process of analysis and automatic diagnosis.

[1576] A "user display device" is a device for visually presenting information to a user, examples of which include a smartphone or smart glasses.

[1577] A "storage device" is hardware or software for long-term storage of data or information.

[1578] "Educational Program" means a set of educational materials or training methods designed to teach specific knowledge or skills.

[1579] A system for implementing this invention uses a generative AI model to learn patterns in comment data and analyze newly entered comments and situational data to automatically diagnose harassment behavior and transmit the diagnosis results to the user's display device. Furthermore, it displays appropriate educational programs based on the diagnosis results and tracks their progress. It also stores preventive measures and case information developed by experts in a storage device and uses them for analysis and automatic diagnosis.

[1580] Hardware and Software Configuration

[1581] Server: The server has the computational capabilities to use generative AI models to learn patterns from comment data, and the ability to analyze newly entered comments and situational data. The database server stores preventive measures and case information from experts, which are used for analysis.

[1582] Terminal: A terminal is a display device such as a smartphone or smart glasses, which the user uses to input comments and situational data and check the diagnosis results. These terminals have the function of sending data from the user to the server and receiving diagnosis results and educational programs from the server.

[1583] Generative AI Model: A generative AI model is used to learn patterns of harassing behavior from comment data. This model is trained on the server and used to analyze new data.

[1584] Data processing and calculation

[1585] The server processes the data in the following steps:

[1586] 1. Data collection: Users input comments and situational data using their devices and send them to the server. Preventive measures and case information provided by experts are also collected and stored in a database.

[1587] 2. Data preprocessing: The server cleanses, denoises, and normalizes the collected comment data. After this preprocessing, the data is stored back in the database.

[1588] 3. Pattern Learning: The server uses the preprocessed comment data to train a generative AI model, which is used to learn patterns of harassing behavior.

[1589] 4. Automated Diagnosis: Users input new comments and situational data and send it to the server. The server uses a generative AI model to analyze this data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[1590] 5. Sending and displaying the results: The server generates the diagnostic results and sends them to the user's terminal, which then displays them to the user.

[1591] 6. Providing educational programs: Based on the diagnosis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through the terminal, and the progress is reported to the server.

[1592] Specific examples

[1593] Example 1: A user types into their smartphone, "My boss keeps making comments about my clothes and it's annoying me." The server receives this comment data and analyzes it using a generative AI model. The analysis results return a diagnosis that "comments about my clothes are inappropriate." An appropriate educational program is also provided along with the diagnosis.

[1594] Example 2: A user inputs, "My colleague always ignores what I say during meetings and tries to overwhelm me." The server receives this data and analyzes it using a generative AI model. The diagnosis is that "ignoring what I say constitutes harassment." The user confirms this result and is provided with relevant educational programs.

[1595] Prompt Sentence Examples

[1596] "An employee says: 'My boss keeps making comments about my attire and it's bothering me.' Make a diagnosis based on this comment and provide an appropriate training program."

[1597] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1598] Step 1: Data collection

[1599] The user uses the device to input comments and status data and sends it to the server. This input includes opinions and feedback in text format. The device receives the input data and sends it to the server using an HTTP request. The server receives this data and stores it in a database.

[1600] Step 2: Data Preprocessing

[1601] The server cleanses, removes noise, and normalizes the comment data stored in the database. For example, it removes unnecessary symbols and emojis such as "!!!" and "???." It also standardizes the same content expressed in different formats (e.g., "Hello" and "Hello"). After this preprocessing, the data is stored back in the database.

[1602] Step 3: Pattern learning

[1603] The server uses the preprocessed comment data to train the generative AI model. During pattern learning, the cleansed comments are used as input data and patterns of harassing behavior are identified as output. The model iteratively analyzes the data to improve its accuracy. This training process is run periodically.

[1604] Step 4: Automatic diagnosis

[1605] The user inputs new comments and situational data into the device and sends it to the server. The server then provides the received data to the generative AI model for analysis. This analysis determines whether the input data constitutes harassment, and a diagnosis result is generated as the output.

[1606] Step 5: Send and view results

[1607] The server generates the results of the automated diagnosis and sends them to the user's device. The device then visually displays the received results to the user. For example, it may display a result such as "This comment is likely to constitute harassment."

[1608] Step 6: Offering educational programs

[1609] Based on the diagnosis results, the server selects an appropriate educational program and sends it to the user's device. The user then takes the educational program provided through the device. The device periodically reports the user's progress to the server. The server uses this information to adjust the next educational program.

[1610] Prompt Sentence Examples

[1611] "An employee says: 'My boss keeps making comments about my attire and it's bothering me.' Make a diagnosis based on this comment and provide an appropriate training program."

[1612] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1613] The system of the present invention is equipped with a generative AI model that learns patterns from comment data, an automatic diagnostic function that combines preventive measures and case information from experts, the provision of educational programs, and an emotion engine that recognizes user emotions. Specific embodiments for implementing this system are described below.

[1614] System configuration

[1615] server

[1616] The server has the following functions:

[1617] 1. Data collection function

[1618] 2. Data preprocessing function

[1619] 3. Pattern learning function

[1620] 4. Automatic diagnosis function

[1621] 5. Analysis function using emotion engine

[1622] 6. Diagnostic and emotion analysis results transmission function

[1623] 7. Educational program provision function

[1624] 8. Database management functions

[1625] Terminal

[1626] The terminal has the following functions:

[1627] 1. Comment and status data input function

[1628] 2. Data transmission function

[1629] 3. Diagnostic and emotion analysis results display function

[1630] 4. Educational program display and progress tracking function

[1631] Service flow

[1632] Data collection

[1633] When users enter comments into internal surveys, chat systems, etc., the server continuously collects them. Prevention measures and case information provided by experts are also collected and stored in a database.

[1634] Data Preprocessing

[1635] The server cleanses, denoises, and normalizes the collected comment data, and then stores the pre-processed data back in the database.

[1636] Pattern Learning

[1637] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[1638] Automatic diagnosis

[1639] Users input new comments and situational data from their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether the behavior constitutes harassment. Experts' preventive measures and case studies are also referenced.

[1640] Emotion analysis

[1641] The server uses an emotion engine to analyze the emotional state of comments and situational data entered by users, which are then classified as positive, negative, neutral, etc.

[1642] Sending and displaying results

[1643] The server generates diagnosis results and emotion analysis results and sends them to the user's terminal, which then displays the diagnosis results and emotion analysis results to the user.

[1644] Providing educational programs

[1645] Based on the diagnosis and emotion analysis results, the server selects an appropriate educational program and provides it to the user. The user takes the educational program provided through their device, and their progress is reported to the server.

[1646] Specific examples

[1647] Example 1: Comments about appearance

[1648] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which analyzes the comment using a generative AI model. The analysis results in a diagnosis that "This is a negative comment about my appearance, and is likely to constitute harassment." The server then uses an emotion engine to analyze the comment as containing negative emotions. The results are displayed on the user's device, and further appropriate educational programs are provided.

[1649] Example 2: Ignoring comments

[1650] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an act of intentionally excluding a specific individual, and is likely to constitute harassment." The server then uses an emotion engine to analyze that the situation contains negative emotions. The results are displayed on the user's device, and an appropriate educational program is provided.

[1651] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. The emotion analysis function using the emotion engine also enables more accurate diagnosis and response. Furthermore, education programs provide continuous employee education, leading to an improved work environment.

[1652] The processing flow will be explained below.

[1653] Step 1: Collecting comment data

[1654] The server continuously collects comments from various comment data sources (internal surveys, internal chats, review sites, etc.), including data extraction using APIs and CSV file imports.

[1655] Step 2: Gather expert information

[1656] The server collects harassment prevention measures and case information provided by experts and stores them in a database, including loading static files and using input forms for experts.

[1657] Step 3: Preprocessing the data

[1658] The server retrieves raw comment data from the database, corrects emotional words and typos to remove noise, and standardizes synonyms and segments text to normalize it. The preprocessed data is then saved back to the database.

[1659] Step 4: Pattern learning

[1660] The server uses the preprocessed comment data as training data for a generative AI model, trains a generative AI model (e.g., BERT or GPT-4) to learn patterns of harassing behavior, and saves the trained model so it can be used for predictions.

[1661] Step 5: Enter comments and status data

[1662] The user uses the terminal to enter comments and situation data into the input form, and then presses the "Send" button to send the data to the server.

[1663] Step 6: Receive data and prepare for analysis

[1664] The device sends input data to the server, which receives it and prepares to pass the received comments and status data to the analysis process.

[1665] Step 7: Sentiment Analysis

[1666] The server uses an emotion engine to analyze the emotional state of comments and situational data received from users, which can be classified as positive, negative, neutral, etc.

[1667] Step 8: Analyze the data

[1668] The server uses the generated AI model to analyze the incoming data, taking into account the results of sentiment analysis, to evaluate the meaning and sentiment of the analyzed data and determine whether it matches a specific pattern.

[1669] Step 9: Automatic diagnosis

[1670] Based on the analysis results, the server automatically diagnoses whether the input data constitutes harassment, while also referencing expert preventative measures and case studies, and generates a diagnosis such as "This is harassment," "This is not harassment," or "It is difficult to determine."

[1671] Step 10: Sending diagnosis and sentiment analysis results

[1672] The server converts the diagnosis results and emotion analysis results into a data format and sends them to the user's terminal.

[1673] Step 11: Viewing the diagnosis and sentiment analysis results

[1674] The device displays the diagnosis results and emotion analysis results received from the server to the user, allowing the user to check the evaluation of their own comments and situations.

[1675] Step 12: Offering educational programs

[1676] The server selects an appropriate educational program based on the diagnosis and emotion analysis results, provides it to the user, and sends a notification to the user's device containing the educational program's URL and content.

[1677] Step 13: Take the education program and track your progress

[1678] The user takes the educational program provided through the terminal, which tracks the progress of the educational program and periodically reports it to the server.

[1679] This allows users and companies to quickly and accurately diagnose harassment and take necessary measures. The emotion analysis function using the emotion engine also enables more accurate diagnosis and response. Furthermore, education programs provide continuous employee education, leading to an improved work environment.

[1680] Example 2

[1681] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1682] It is important to quickly and accurately recognize harassment in the work environment and take measures to address it, but conventional systems tend to delay such recognition and countermeasures, making it difficult to alleviate employee stress and problems. Furthermore, conventional systems lack the functionality to analyze the emotional state of comments or to provide appropriate educational programs, preventing effective improvements to the work environment.

[1683] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1684] In this invention, the server includes means for learning patterns in comment data using a generative AI model, means for analyzing newly entered comments and situation data, means for automatically diagnosing whether a comment constitutes harassment based on the analysis results, means having an emotion engine for analyzing the emotional state of the user's comments and situation data, and means for transmitting the diagnosis and emotion analysis results to the user's device. This enables rapid analysis of user input data and provides comprehensive diagnosis results including the user's emotional state, enabling continuous improvement of the work environment through appropriate educational programs.

[1685] A "generative AI model" is an algorithm that uses artificial intelligence technology to learn patterns and trends from data and make predictions and analyses on new data.

[1686] "Comment data" refers to text information, opinions, explanations of situations, etc. entered by users.

[1687] "Pattern learning" is the process of extracting specific trends or recurring features from data, understanding them, and incorporating them into a model.

[1688] "Automatic diagnosis" refers to the ability of a system to analyze data and make a determination about a specific condition or problem without human intervention.

[1689] An "emotion engine" is an algorithm that analyzes the emotions contained in text data and classifies them as positive, negative, neutral, etc.

[1690] "Diagnostic results" refer to the judgment information, such as whether or not harassment has occurred, that is shown as a result of the analysis by the generative AI model.

[1691] "Analysis results" refers to the overall results of the analysis performed by the generative AI model and emotion engine on the input data.

[1692] "Educational Program" means a course of study or training offered to enable a user to acquire specific knowledge or skills.

[1693] The system of the present invention collects user comment data and situational data and automatically analyzes and diagnoses them. This system is equipped with a pattern learning function using a generative AI model, an automatic diagnosis function, an analysis function using an emotion engine, and an educational program provision function. This enables the rapid and accurate recognition of harassment in the work environment and the implementation of countermeasures.

[1694] Server configuration and functions

[1695] Data collection features:

[1696] The server continuously collects data on comments entered by users using the company's internal surveys and chat systems, and stores the collected data in a database.

[1697] Data preprocessing functions:

[1698] The server cleanses the collected comment data, removes noise, and normalizes it. The preprocessed data is then stored in the database again.

[1699] Pattern learning function:

[1700] The server uses the preprocessed comment data to train a generative AI model, which is then used to learn patterns of harassing behavior.

[1701] Automatic diagnostic function:

[1702] Users input new comments and situational data through their devices and send it to the server, which then uses a generative AI model to analyze the data and automatically diagnose whether it constitutes harassment.

[1703] Emotion engine analysis function:

[1704] The server uses an emotion engine to analyze the emotional state of comments and situational data entered by users, which are classified as positive, negative, neutral, etc.

[1705] Diagnostic and sentiment analysis results transmission function:

[1706] The server transmits the generated diagnosis results and emotion analysis results to the user's terminal, which displays these results to the user.

[1707] Educational program offerings:

[1708] The server selects an appropriate educational program based on the diagnosis and emotion analysis results and provides it to the user. The user takes the educational program through their device, and their progress is reported to the server.

[1709] Device configuration and functions

[1710] Comment and status data entry function:

[1711] It provides an interface for users to input comment data and situational data through an internal survey or chat system.

[1712] Data transmission function:

[1713] The terminal transmits the input data to the server.

[1714] Diagnostic and emotion analysis results display function:

[1715] The terminal displays the diagnosis results and emotion analysis results sent from the server to the user.

[1716] Educational Program Display & Progress Tracking Features:

[1717] The terminal displays the educational program provided by the server and tracks the user's learning progress.

[1718] Specific operation example

[1719] Example 1: Comments about appearance

[1720] A user types, "My boss has been frequently commenting on my appearance lately, and it's annoying." The device sends this to the server, which analyzes the comment using a generative AI model. The analysis results in a diagnosis that "This is a negative comment about my appearance, and is likely to constitute harassment." The server then uses an emotion engine to analyze the comment as containing negative emotions. The results are displayed on the user's device, and further appropriate educational programs are provided.

[1721] Example 2: Ignoring comments

[1722] The user inputs the situation, saying, "In daily meetings, certain colleagues are constantly being ignored and denied the opportunity to speak." The device sends this to the server, which analyzes the situation using a generative AI model. The analysis results in a diagnosis that "this is an act of intentionally excluding a specific individual, and is likely to constitute harassment." The server then uses an emotion engine to analyze that the situation contains negative emotions. The results are displayed on the user's device, and an appropriate educational program is provided.

[1723] Prompt Sentence Examples

[1724] "My boss frequently makes comments about my appearance, which I find very unpleasant. What kind of diagnosis would a generative AI model make in this case?"

[1725] "My colleague keeps getting ignored in meetings. How would the emotion engine analyze this situation?"

[1726] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1727] Step 1:

[1728] A user enters a comment into an internal survey or chat system.

[1729] As a specific operation, the user inputs text such as "Recently, my boss has been frequently commenting on my appearance, and it makes me feel uncomfortable."

[1730] Input: Comment data entered by the user

[1731] Output: Comment data sent to the device

[1732] Step 2:

[1733] The terminal transmits the input comment data to the server.

[1734] As a specific operation, the terminal generates and transmits a request for transmitting the comment data input by the user to the server.

[1735] Input: Comment data entered into the terminal

[1736] Output: Comment data sent to the server

[1737] Step 3:

[1738] The server stores the received comment data in a database.

[1739] Specifically, the server stores the received comment data in an appropriate database table.

[1740] Input: Comment data received by the server

[1741] Output: Comment data stored in the database

[1742] Step 4:

[1743] The server retrieves the comment data from the database and performs data cleansing.

[1744] Specifically, the server reads unprocessed comment data from the database and performs noise removal and format conversion.

[1745] Input: Comment data stored in the database

[1746] Output: Cleansed comment data

[1747] Step 5:

[1748] The server normalizes the cleansed comment data and stores it back in the database.

[1749] Specifically, the server performs normalization processing, such as converting the text to lowercase, and stores it in the database.

[1750] Input: Cleansed comment data

[1751] Output: Normalized comment data

[1752] Step 6:

[1753] The server uses the normalized comment data to train a generative AI model.

[1754] Specifically, the server inputs the normalized data into the AI ​​model and performs training.

[1755] Input: Normalized comment data

[1756] Output: A trained generative AI model

[1757] Step 7:

[1758] The user inputs new comments and status data from the terminal, which then transmits the data to the server.

[1759] Specifically, the user inputs new text such as, "In daily meetings, certain colleagues are always ignored and denied the opportunity to speak," and the device sends this to the server.

[1760] Input: New comments and status data entered by the user

[1761] Output: New comments and status data sent to the server

[1762] Step 8:

[1763] The server uses the generated AI model to analyze newly received data.

[1764] Specifically, the server inputs new comments and situational data into the generative AI model and obtains the analysis results.

[1765] Input: New comments and status data sent to the server

[1766] Output: Parsed data

[1767] Step 9:

[1768] Based on the analysis results, the server automatically diagnoses whether the behavior constitutes harassment.

[1769] Specifically, the system refers to the analysis results and case information from experts to determine whether or not the behavior constitutes harassment.

[1770] Input: Parsed data

[1771] Output: Automatic diagnosis results

[1772] Step 10:

[1773] The server uses an emotion engine to analyze the emotional state of the comments and situation data.

[1774] Specifically, the server detects emotional keywords in the comments and classifies their emotional state into "positive," "negative," or "neutral."

[1775] Input: New comments and status data

[1776] Output: Emotion analysis results

[1777] Step 11:

[1778] The server generates diagnosis results and emotion analysis results and transmits them to the terminal.

[1779] Specifically, the server combines the diagnosis results and emotion analysis results into a single packet and sends it to the terminal.

[1780] Input: Automatic diagnosis results and emotion analysis results

[1781] Output: Result data sent to the terminal

[1782] Step 12:

[1783] The terminal displays the received diagnosis results and emotion analysis results to the user.

[1784] Specifically, the terminal provides the user with the diagnosis results and emotion analysis results as a screen display.

[1785] Input: Result data sent to the terminal

[1786] Output: Diagnosis results and sentiment analysis results displayed to the user

[1787] Step 13:

[1788] The server selects an appropriate educational program based on the diagnosis and emotion analysis results.

[1789] Specifically, the server selects and lists educational programs from a database according to the diagnostic results.

[1790] Input: Diagnosis results and emotion analysis results

[1791] Output: Selected educational programs

[1792] Step 14:

[1793] The server transmits the selected educational program to the terminal.

[1794] Specifically, the server sends links to educational programs and educational materials to the terminal.

[1795] Input: Selected Educational Program

[1796] Output: Educational program sent to the terminal

[1797] Step 15:

[1798] A user takes an educational program through a terminal, and the progress is reported to a server.

[1799] Specifically, the user watches an educational program on the terminal, and the learning progress is automatically reported to the server.

[1800] Input: Educational program progress data

[1801] Output: Progress data reported to the server

[1802] (Application example 2)

[1803] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1804] Traditional brick-and-mortar stores lack systems for quickly and accurately evaluating customer satisfaction and proposing concrete improvement measures. In particular, there is no way to accurately analyze the emotional state of customer feedback and immediately propose countermeasures for negative feedback. As a result, improvements in the customer service skills of store staff and service quality are delayed, leading to a decline in customer satisfaction.

[1805] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1806] In this invention, the server includes a means for pattern learning of comment data using a generative AI model, a means for collecting, preprocessing, and analyzing newly entered customer feedback, and a means for analyzing the emotional state of comments and situation data using an emotion analysis engine. This makes it possible to quickly and accurately analyze the emotional state of customer feedback and provide specific improvement measures and training programs for store staff.

[1807] A "generative AI model" is an artificial intelligence model that learns patterns based on collected data and analyzes new data.

[1808] "Comment Data" refers to feedback and opinions entered by users in the form of text.

[1809] An "emotion analysis engine" is software that analyzes and classifies emotional states from comments and situational data.

[1810] "Automatic diagnosis" is the process by which a generative AI model analyzes newly input data and determines whether a particular behavior constitutes harassment.

[1811] "Educational Program" refers to training and learning materials provided to users based on the analysis results.

[1812] "Expert preventive measures and case information" means information provided by experts and past cases regarding the prevention and countermeasures against harassment.

[1813] "Customer feedback" refers to the evaluations, opinions, and impressions that customers provide regarding services and products.

[1814] "Customer service skills improvement training" is an educational program designed to improve the customer service attitude and response techniques of store staff.

[1815] "Negative feedback" refers to evaluations or opinions that express customer dissatisfaction or annoyance.

[1816] "Progress tracking" is the process of monitoring and managing progress through a delivered educational program.

[1817] The present invention relates to a system for analyzing customer feedback in a physical store and improving the customer service skills of store staff. The specific system configuration and operation will be described below.

[1818] System configuration

[1819] server

[1820] The server has the following functions:

[1821] 1. Data collection function: Collect customer feedback from smartphones and tablets.

[1822] 2. Data preprocessing function: Cleanse the collected feedback data and remove noise.

[1823] 3. Pattern learning function: Trains generative AI models using preprocessed data.

[1824] 4. Automatic diagnosis function: Feedback data is analyzed using a generative AI model to diagnose specific behaviors.

[1825] 5. Sentiment Analysis Function: Uses a sentiment analysis engine to analyze the emotional state of the feedback.

[1826] 6. Diagnostic result transmission function: Transmits diagnostic results and emotion analysis results to the user's device.

[1827] 7. Educational program provision function: Based on the diagnostic results, select and provide appropriate educational programs.

[1828] Terminal

[1829] The terminal has the following features:

[1830] 1. Comment input function: Customers can input feedback using their smartphones or tablets.

[1831] 2. Data transmission function: Sends collected data to the server.

[1832] 3. Result display function: Displays the diagnosis results and emotion analysis results.

[1833] 4. Educational program progress tracking function: Tracks the progress of the educational program and reports the results to the server.

[1834] Program processing flow

[1835] 1. The user enters feedback using a smartphone or tablet. For example, they enter feedback such as, "I feel like the staff have been cold towards me lately."

[1836] 2. The device sends the input feedback data to the server.

[1837] 3. The server receives the feedback data and performs data preprocessing, which involves data cleansing and noise removal.

[1838] 4. Using the pre-processed data, the generative AI model analyzes the feedback content and diagnoses whether a particular behavior constitutes harassment.

[1839] 5. The server uses an emotion analysis engine to analyze the emotional state of the feedback, e.g., it is analyzed as "negative."

[1840] 6. The diagnosis results and emotion analysis results are sent to the store manager's terminal and displayed.

[1841] 7. Based on the results of the diagnosis, the server will provide an appropriate educational program (e.g., "Customer Service Skills Improvement Training").

[1842] Hardware and software used

[1843] Hardware:

[1844] Devices used by customers and store managers: smartphones, tablets, PCs, etc.

[1845] Server: High-performance data processing server

[1846] software:

[1847] Sentiment analysis engine: for example, a model using the transformers library

[1848] Data preprocessing and analysis: Programming languages ​​such as Python

[1849] Specific examples

[1850] As a concrete example, let's consider the case where a customer uses a smartphone. A customer inputs feedback such as, "I feel like the staff have been cold towards me lately." The data is sent to the server, where it is preprocessed and analyzed by the generative AI model. The analysis results indicate that "improvement in customer service attitude is necessary," and the sentiment analysis result is "negative." These results are then notified to the store manager, who provides training to improve customer service skills.

[1851] Prompt Sentence Examples

[1852] Feedback: "I feel like the staff have been cold lately."

[1853] Prompt for generative AI model: "Based on this feedback, please rate the customer's attitude."

[1854] In this way, the present invention makes it possible to analyze customer feedback quickly and accurately in a physical store and immediately propose countermeasures, thereby improving the customer service skills of store staff and increasing customer satisfaction.

[1855] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1856] Step 1:

[1857] The user inputs and sends feedback using a smartphone or tablet. For example, the input data might be, "Recently, I feel like the staff have been cold towards me." The device then sends this feedback to the server. The input is text data, and the output is data sent to the server.

[1858] Step 2:

[1859] The server cleanses and pre-processes the received feedback data, which includes denoising and normalizing the data. The input data is the raw feedback, and the output data is the cleansed text data.

[1860] Step 3:

[1861] The server inputs the preprocessed feedback data into a generative AI model and performs pattern learning. The model analyzes the data and diagnoses whether a particular behavior constitutes harassment. The input data is the preprocessed feedback text, and the output data is the diagnosis result.

[1862] Step 4:

[1863] The server uses a sentiment analysis engine to analyze the emotional state of the feedback, e.g., it is analyzed as "negative." The input data is the feedback text, and the output data is the sentiment analysis result.

[1864] Step 5:

[1865] The server sends the diagnosis results and emotion analysis results to the user's device. The input data here are the diagnosis results and emotion analysis results, and the output data is sent to the device. The user's device displays these results.

[1866] Step 6:

[1867] Based on the diagnosis results, the server selects and provides a relevant educational program. For example, "Training to improve customer service skills" may be suggested. The input data is the diagnosis results, and the output data is a suggested educational program. The content of the educational program is displayed on the user's device.

[1868] Step 7:

[1869] A user takes an educational program and reports his / her progress to the server via his / her terminal. The input data is the progress information of the educational program, and the output data is the progress report to the server.

[1870] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1871] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1872] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1873] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1874] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1875] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1876] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1877] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1878] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1879] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1880] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1881] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1882] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1883] 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.

[1884] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1885] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1886] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1887] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1888] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1889] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1890] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1891] The following is further disclosed regarding the above embodiment.

[1892] (Claim 1)

[1893] A means of pattern learning comment data using a generative AI model; and

[1894] A means of analyzing newly entered comments and situation data;

[1895] A means of automatically diagnosing whether an act constitutes harassment based on the analysis results;

[1896] means for transmitting the diagnostic results to a user's terminal;

[1897] A system including:

[1898] (Claim 2)

[1899] 10. The system of claim 1, further comprising means for storing expert preventative measures and case information in a database for use in analysis and automatic diagnosis.

[1900] (Claim 3)

[1901] 2. The system according to claim 1, further comprising means for generating an appropriate educational program based on newly input comments and situation data, and providing the program to the user.

[1902] "Example 1"

[1903] (Claim 1)

[1904] A means for a user to input comment data from a terminal and transmit the data to a server;

[1905] A means by which the server collects and stores comment data in a database;

[1906] A means for data cleansing and normalizing the comment data collected by the server;

[1907] a means for the server to train a generative AI model using the preprocessed comment data;

[1908] A means for transmitting newly entered data from a user's device to a server, and for the server to analyze the data using the generated AI model;

[1909] A means for the server to automatically diagnose whether the behavior constitutes harassment based on the analysis results;

[1910] A means for the server to generate a diagnostic result and transmit it to the user's terminal;

[1911] A means for a user to take an educational program through a terminal and for the user's progress to be reported to a server;

[1912] A system including:

[1913] (Claim 2)

[1914] 10. The system of claim 1, further comprising means for storing expert preventative measures and case information in a database for use in analysis and automatic diagnosis.

[1915] (Claim 3)

[1916] 2. The system according to claim 1, further comprising means for generating an appropriate educational program based on newly input comments and situation data, and providing the program to the user.

[1917] "Application Example 1"

[1918] (Claim 1)

[1919] A means of pattern learning comment data using a generative AI model; and

[1920] A means of analyzing newly entered comments and situation data;

[1921] A means of automatically diagnosing whether an act constitutes harassment based on the analysis results;

[1922] means for transmitting the diagnostic results to a user's display device;

[1923] means for transmitting comments and status data input by a user to a server;

[1924] A means for displaying an appropriate educational program based on the diagnostic results and tracking its progress;

[1925] A system including:

[1926] (Claim 2)

[1927] 10. The system of claim 1, further comprising means for storing expert preventative measures and case information in a storage device for use in analysis and automated diagnosis.

[1928] (Claim 3)

[1929] 2. The system according to claim 1, further comprising means for generating an appropriate educational program based on newly input comments and situation data, providing the program to the user, and reporting the progress of the program to the server.

[1930] "Example 2: Combining Emotion Engines"

[1931] (Claim 1)

[1932] A means of pattern learning comment data using a generative AI model; and

[1933] A means of analyzing newly entered comments and situation data;

[1934] A means of automatically diagnosing whether an act constitutes harassment based on the analysis results;

[1935] means for analyzing the emotional state of a user's comments and situation data, the emotional engine;

[1936] means for transmitting the diagnosis result and the emotion analysis result to a user's terminal;

[1937] A system including:

[1938] (Claim 2)

[1939] 10. The system of claim 1, further comprising means for storing expert preventative measures and case information in a database for use in analysis and automatic diagnosis.

[1940] (Claim 3)

[1941] 2. The system according to claim 1, further comprising means for generating an appropriate educational program based on newly input comments and situation data, and providing the program to the user.

[1942] "Application example 2 when combining emotion engines"

[1943] (Claim 1)

[1944] A means of pattern learning comment data using a generative AI model; and

[1945] A means of analyzing newly entered comments and situation data;

[1946] A means of automatically diagnosing whether an act constitutes harassment based on the analysis results;

[1947] a means for analyzing the emotional state of the comments and situational data using a sentiment analysis engine;

[1948] means for transmitting the diagnosis result and the emotion analysis result to a user's terminal;

[1949] means for generating and providing an educational program to a user;

[1950] A system including:

[1951] (Claim 2)

[1952] 10. The system of claim 1, further comprising means for storing expert preventative measures and case information in a database for use in analysis and automatic diagnosis.

[1953] (Claim 3)

[1954] 10. The system of claim 1, further comprising means for collecting, pre-processing, and analyzing newly entered customer feedback using the generative AI model.

[1955] (Claim 4)

[1956] 4. The system according to claim 3, further comprising means for providing specific improvement measures and training programs for improving customer service skills to store staff based on the feedback analysis.

[1957] (Claim 5)

[1958] 2. The system according to claim 1, further comprising means for immediately suggesting countermeasures if the feedback is negative based on the diagnosis result and the emotion analysis result.

[1959] (Claim 6)

[1960] 5. The system of claim 4, further comprising means for tracking progress and sending periodic reports to store management. [Explanation of symbols]

[1961] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of pattern learning comment data using a generative AI model; and A means of analyzing newly entered comments and situation data; A means of automatically diagnosing whether an act constitutes harassment based on the analysis results; means for transmitting the diagnostic results to a user's terminal; A system including:

2. 10. The system of claim 1, further comprising means for storing expert preventative measures and case information in a database for use in analysis and automated diagnosis.

3. 2. The system according to claim 1, further comprising means for generating an appropriate educational program based on newly input comments and situation data, and providing the program to the user.

Images