System

Abstract

A system is provided.SOLUTION: A means for acquiring a plurality of pieces of data including an occurrence event and a statement content input by a user, a means for transmitting the acquired data to a server side, a means for receiving the transmitted data, and recording the data in a database, a means for executing a generated AI model based on the recorded data, and determining a possibility of power harassment by collating with a past case, and a means for generating a determination result and a countermeasure corresponding thereto; The system includes a means for preparing display data, a means for transmitting a generated result and a countermeasure to a user terminal, a means for generating a video simulation and transmitting a highly realistic video to the user terminal, and a means for performing immediate escalation to a department in charge when it is determined that the determination result is bad.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] Power harassment in the workplace is a serious problem that harms employees' mental health and reduces workplace productivity. The problem that this invention aims to solve is that the conventional subjective judgment of power harassment by humans is prone to intervention and lacks fairness. It has also been pointed out that the problem can become more serious if victims are unable to obtain immediate and appropriate countermeasures. Another issue is the lack of realistic training materials, and insufficient education to prevent power harassment. [Means for solving the problem]

[0005] The present invention solves these problems by providing a system that includes a means for acquiring multiple pieces of data, including incidents and statements entered by users, a means for transmitting the acquired data to a server, a means for receiving the transmitted data and recording it in a database, a means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment, a means for generating the determination results and corresponding countermeasures and preparing display data, a means for transmitting the generated results and countermeasures to the user's terminal, a means for generating a video simulation and transmitting highly realistic video to the user's terminal, and a means for immediately escalating to the relevant department if the determination results are deemed to be malicious.

[0006] This system not only allows users to quickly and fairly determine the possibility of power harassment, but also provides appropriate countermeasures immediately. Furthermore, training using highly realistic video simulations can enhance the effectiveness of power harassment prevention. This will improve the work environment and maintain the mental health of employees.

[0007] "User" refers to an individual who uses the system to report or seek advice on suspected power harassment.

[0008] "Terminal" refers to an electronic device, such as a computer or smartphone, that a user uses to input information and receive results.

[0009] "Server" refers to the central control system that receives, stores, processes and analyzes data sent by users.

[0010] "Data" refers to all information entered by the user, such as the event that occurred, what was said, the trigger, the location, the action, the intonation of the statement, relationships, work history, and past content.

[0011] A "generative AI model" refers to an artificial intelligence algorithm that learns from a large number of past cases, analyzes input data, and determines the possibility of power harassment.

[0012] "Power harassment assessment" refers to the process in which a generative AI model scores the likelihood of power harassment based on input data and provides the results.

[0013] "Countermeasures" refers to suggestions regarding specific actions and measures that users should take if a possible case of power harassment is determined.

[0014] "Video simulation" refers to a highly realistic scene reproduction video generated based on input data, and is used for training and educational purposes.

[0015] "Escalation" refers to the process of urgently reporting to the relevant department and requesting a response when extremely serious acts of power harassment are determined to have occurred.

[0016] "Database" refers to an information management system that accumulates data submitted by users and stores it in a usable state for subsequent processing.

[0017] "Input data" refers to all information such as events that occur and statements that users provide through their terminals. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] The present invention relates to a system that acquires detailed data related to power harassment input by users, and uses a generative AI model to determine power harassment and provide countermeasures. This system is composed of a user terminal, a server, a database, a generative AI model, and a video generation function.

[0040] System Program

[0041] User Input Processing

[0042] User

[0043] The user uses a terminal to input detailed information about the suspected power harassment behavior. Specifically, the user inputs the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "When reporting on the progress of the project"), the location (e.g., "Conference room"), the behavior (e.g., "Scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "Very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "3 years"), and past experiences (e.g., "I have received similar scolding three times in the past")).

[0044] Terminal

[0045] The terminal receives data input by the user, converts it into an appropriate format, and transmits it to the server.

[0046] Data transmission and AI judgment

[0047] server

[0048] The server receives the data sent from the terminal and records it in a database.

[0049] A generative AI model is run based on the recorded data, and the possibility of power harassment is judged by comparing it with past cases. The AI ​​model has learned from past court decisions and cases from other companies and the company itself, and evaluates the input data to generate a score for the possibility of power harassment.

[0050] Providing judgment results and countermeasures

[0051] server

[0052] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model and generates a judgment result. It also creates countermeasures that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the human resources department").

[0053] The judgment result and countermeasures are sent to the user's terminal.

[0054] Terminal

[0055] The terminal displays the judgment result and countermeasures received from the server to the user.

[0056] Generate video simulations

[0057] server

[0058] The server runs a model to generate a video simulation based on user input data. The generated simulation video is highly realistic, allowing users to visually confirm the scene.

[0059] The generated video is stored on the server and sent to the user's terminal.

[0060] Terminal

[0061] The terminal displays the video simulation sent from the server in a viewable form for the user.

[0062] Immediate Escalation

[0063] server

[0064] If the result is deemed to be extremely egregious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[0065] Specific examples

[0066] Example 1: Victim input

[0067] 1. User Input

[0068] The user input a scenario in which their boss suddenly raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[0069] Relationship: "Boss"

[0070] Work experience: 3 years

[0071] Location: "Conference Room"

[0072] 2. Data Transmission

[0073] The terminal formats the data and sends it to the server.

[0074] 3. AI Judgment and Result Display

[0075] The server receives the data and uses an AI model to score the likelihood of power harassment (e.g., 80%).

[0076] The server generates the judgment result and countermeasures and sends them to the terminal.

[0077] The terminal displays the judgment result and countermeasures to the user.

[0078] 4. Video Simulation Generation

[0079] The server generates a video simulation and transmits it to the terminal.

[0080] The terminal displays the simulation video in a viewable form for the user.

[0081] 5. Escalation

[0082] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[0083] This system contributes to preventing power harassment and improving the work environment by fairly and objectively assessing the possibility of power harassment, providing quick countermeasures, and using highly educational video simulations.

[0084] The processing flow will be explained below.

[0085] Step 1:

[0086] The user enters detailed information about suspected power harassment on the device, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[0087] Step 2:

[0088] The terminal takes the data entered by the user, converts it into the appropriate format, and sends it to the server, ensuring that the data is processed in a consistent format.

[0089] Step 3:

[0090] The server receives the data sent from the device, records it in a database, and stores it for further processing.

[0091] Step 4:

[0092] The server runs a generative AI model based on the recorded data. The generative AI model has learned from past court decisions and cases from other companies and the company, and evaluates the input data to generate a score for the possibility of power harassment.

[0093] Step 5:

[0094] Based on the results of the AI ​​model, the server generates a judgment result indicating the possibility of power harassment, and then creates countermeasures for the user to take (e.g., "consult with a higher-level manager" or "file a complaint with the human resources department") based on the judgment result.

[0095] Step 6:

[0096] The server sends the generated results and countermeasures data to the user's device, allowing the user to quickly check the results and countermeasures.

[0097] Step 7:

[0098] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to understand the possibility of power harassment and the actions they should take.

[0099] Step 8:

[0100] The server runs a model to generate a video simulation based on user input data. The generated simulation video is highly realistic and reproduces the actual scene.

[0101] Step 9:

[0102] The server stores the generated video simulation on the server and transmits a link or file to the user's terminal.

[0103] Step 10:

[0104] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user, thereby enabling the user to visually understand the situation.

[0105] Step 11:

[0106] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[0107] Through the specific processing flow described above, this system responds to power harassment issues fairly and efficiently, and provides prompt countermeasures. In addition, by using highly effective video simulations, it also contributes to preventing power harassment.

[0108] Example 1

[0109] 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."

[0110] Power harassment (pawahara) has become a serious social problem in modern workplaces, and there is a growing demand for systems that can fairly and quickly evaluate and provide appropriate countermeasures. However, current technology lacks the accuracy to make judgments based on detailed incident data or a method to provide visual evidence. This can lead to subjective judgments of power harassment and insufficient countermeasures. There is also a need for a system that can quickly escalate egregious cases of power harassment.

[0111] 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.

[0112] In this invention, the server includes: means for acquiring multiple data sets, including incidents and comments entered by users; means for transmitting the acquired data to the server; means for receiving the transmitted data and recording it in a database; means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment; means for generating a determination result and corresponding countermeasures and preparing display data; means for transmitting the generated results and countermeasures to the user's terminal; means for generating a video simulation based on detailed input data and transmitting highly realistic video to the user's terminal; and means for immediately escalating the generated result to the responsible department if the generated determination result is deemed to be extremely severe. This enables objective and rapid evaluation of cases of power harassment and provides appropriate countermeasures. Furthermore, providing visual evidence can deepen users' understanding and contribute to improving the work environment. Furthermore, severe cases can be quickly escalated to promote appropriate responses.

[0113] A "user" is a person who uses a terminal to input detailed information about a suspected case of power harassment.

[0114] "Occurrence of an incident" refers to a specific act or occurrence of power harassment.

[0115] "Content of remarks" refers to the specific words and actions that were spoken or taken during the power harassment situation.

[0116] "Data" refers to information entered by the user regarding the event, statement, trigger, location, action, intonation of statement, relationships, work history and period, and past content.

[0117] "Server" means a computer system that receives, processes, and stores data sent from a user terminal.

[0118] "Database" means a system for recording and storing data received by the server.

[0119] A "generative AI model" is an artificial intelligence algorithm that assesses the possibility of power harassment based on past cases and court precedents.

[0120] "Assessment" is the process by which the generative AI model scores the possibility of power harassment based on the data.

[0121] "Countermeasures" are specific instructions that suggest actions or measures that the user should take based on the judgment results.

[0122] "Video simulation" refers to highly realistic videos and animations that are generated based on data entered by the user.

[0123] "Escalation" is the process of contacting the department in charge of emergency response when a serious case of power harassment is determined.

[0124] The present invention relates to a system that acquires detailed data related to power harassment input by users, and uses a generative AI model to determine power harassment and provide countermeasures. This system is composed of a user terminal, a server, a database, a generative AI model, and a video generation function.

[0125] User Input Processing

[0126] User

[0127] The user uses a terminal to input detailed information about suspected power harassment. For example, the incident ("My boss raised his voice and scolded me during a meeting"), the trigger ("When reporting on the progress of a project"), the location ("Conference room"), the nature of the behavior ("Scolding"), what was said ("Don't you understand that?"), the intonation of the remarks ("Very harsh tone"), the relationship ("Boss and subordinate"), work history or length of time spent with the same organization ("3 years"), and past experiences ("I have received similar scolding three times in the past").

[0128] Terminal

[0129] The device takes the data entered by the user, converts it into the appropriate format, and sends it to the server, either using a standard web browser or a mobile application.

[0130] Data transmission and AI judgment

[0131] server

[0132] The server receives the data sent from the device and records it in a database, which uses SQL-based or NoSQL-based technologies (e.g., MySQL (registered trademark), MongoDB).

[0133] The server runs a generative AI model based on the recorded data, which learns from past court decisions and similar cases, evaluates the input data, and assigns a score to indicate the likelihood of power harassment.

[0134] Providing judgment results and countermeasures

[0135] server

[0136] The server determines the possibility of power harassment based on the results of the AI ​​model and generates specific countermeasures, such as advice to "consult with a higher-level manager" or "file a complaint with the human resources department."

[0137] The generated results and countermeasures are sent to the user's terminal.

[0138] Terminal

[0139] The terminal displays the judgment result and countermeasures received from the server to the user using a general message box or notification function.

[0140] Generate video simulations

[0141] server

[0142] The server runs a model to generate a visual simulation based on user input data, using widely used 3D modeling tools (e.g., Unity, Unreal Engine).

[0143] The generated simulation images are highly realistic, allowing users to visually confirm the scene.

[0144] The generated video is stored on the server and sent to the user's terminal.

[0145] Terminal

[0146] The device displays the video simulation sent from the server in a viewable format for the user, using a video player or a dedicated viewer within the application.

[0147] Immediate Escalation

[0148] server

[0149] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the relevant department. A detailed report will be sent along with a notification urging immediate action. Notifications will be sent via email or a dedicated notification system (e.g., Slack, MICROSOFT® TEAMS®).

[0150] Specific examples

[0151] Here is an example prompt:

[0152] Please rate and score whether this case constitutes power harassment based on the following details:

[0153] Incident: The boss raised his voice and scolded the employee during a meeting.

[0154] Trigger: Project progress report

[0155] Location: Conference room

[0156] Action: Scolding

[0157] What he said: Don't you know that?

[0158] Speech intonation: Very strong tone

[0159] Relationship: Boss and subordinate

[0160] Work experience: 3 years

[0161] Past experiences: I have received similar reprimands three times in the past.

[0162] This system not only fairly and objectively assesses the possibility of power harassment and provides prompt countermeasures, but also contributes to preventing power harassment and improving the work environment by using highly educational video simulations.

[0163] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0164] Processing flow

[0165] Step 1: Getting User Input

[0166] Step 2: Send data

[0167] Step 3: Record the data

[0168] Step 4: Generative AI model decision

[0169] Step 5: Generate results and countermeasures

[0170] Step 6: Displaying the results and countermeasures

[0171] Step 7: Generate the video simulation

[0172] Step 8: Viewing the video simulation

[0173] Step 9: Immediate Escalation

[0174] Processing step details

[0175] Step 1: Getting User Input

[0176] User

[0177] The user uses the terminal to input detailed information about the behavior suspected of being power harassment.

[0178] Input: Occurrence, trigger, location, action, statement, intonation, relationship, work history, past content

[0179] Output: Detailed data collected from the user

[0180] What happens: A user enters data into an input field on a web form or mobile application and presses the submit button.

[0181] Step 2: Send data

[0182] Terminal

[0183] The terminal converts the data entered by the user into an appropriate format (such as JSON or XML) and sends it to the server.

[0184] Input: Detailed data entered by the user

[0185] Output: The data format sent to the server

[0186] What it does: Creates an HTTP POST request and sends data to the specified endpoint on the server.

[0187] Step 3: Record the data

[0188] server

[0189] The server receives the data sent from the terminal and records it in a database.

[0190] Input: Formatted data sent from the terminal

[0191] Output: Data recorded in a database

[0192] What it does: Parses the received data and inserts it appropriately into a SQL or NoSQL database.

[0193] Step 4: Generative AI model decision

[0194] server

[0195] The server runs a generative AI model based on the data recorded in the database and compares it with past cases to determine the possibility of power harassment.

[0196] Input: User data recorded in the database

[0197] Output: Power harassment likelihood score

[0198] Specific operation: Convert the recorded data into prompt sentences, input them into a generative AI model, and obtain scoring results.

[0199] Example prompt:

[0200] Please rate and score whether this case constitutes power harassment based on the following details:

[0201] Incident: The boss raised his voice and scolded the employee during a meeting.

[0202] Trigger: Project progress report

[0203] Location: Conference room

[0204] Action: Scolding

[0205] What he said: Don't you know that?

[0206] Speech intonation: Very strong tone

[0207] Relationship: Boss and subordinate

[0208] Work experience: 3 years

[0209] Past experiences: I have received similar reprimands three times in the past.

[0210] Step 5: Generate results and countermeasures

[0211] server

[0212] Based on the results of the AI ​​model's assessment, the server generates specific countermeasures that the user should take.

[0213] Input: Judgment score

[0214] Output: Verification result and countermeasures

[0215] Specific actions: Analyze the scoring results and generate countermeasures based on the results (e.g., "consult with upper management" or "file a complaint with the HR department").

[0216] Step 6: Displaying the results and countermeasures

[0217] Terminal

[0218] The terminal displays the judgment result and countermeasures received from the server to the user.

[0219] Input: Verification result and countermeasures sent from the server

[0220] Output: Verification results and countermeasures displayed to the user

[0221] Specific behavior: The judgment result and countermeasures will be displayed as a pop-up message or notification.

[0222] Step 7: Generate the video simulation

[0223] server

[0224] The server executes a model that generates a visual simulation based on user input data.

[0225] Input: User details

[0226] Output: Video simulation data

[0227] What it does: Uses 3D modeling tools (e.g. Unity, Unreal Engine) to generate highly realistic images based on input data.

[0228] Step 8: Viewing the video simulation

[0229] Terminal

[0230] The terminal displays the video simulation sent from the server to the user.

[0231] Input: Video simulation data

[0232] Output: Simulation video displayed to the user

[0233] Specific operation: Play the video using a video player or a dedicated viewer within the application.

[0234] Step 9: Immediate Escalation

[0235] server

[0236] If the result of the assessment is deemed to be extremely malicious, the server will immediately escalate the matter to the relevant department.

[0237] Input: Verdict (extremely malicious)

[0238] Output: Escalation notification to the responsible department

[0239] Actions: Send a detailed report and a notification to the relevant department to prompt immediate action via email or a dedicated notification system (e.g., Slack, Microsoft Teams).

[0240] (Application example 1)

[0241] 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."

[0242] Power harassment (pa-hara) has become a serious problem in modern workplaces. This problem damages the mental and physical health of employees, reduces work efficiency, and lowers workplace morale. However, the determination of power harassment is subjective and requires specific criteria and evidence, making it difficult to address. There is also the problem of a lack of information provided to victims to take appropriate action. There is a need for a system that can solve these issues and enable swift and objective responses.

[0243] 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.

[0244] In this invention, the server includes means for scoring cases using a generative AI model based on information entered by the user and proposing appropriate countermeasures, means for conducting a detailed analysis of the incident with reference to past court cases and related cases, and means for providing the user with the results of the assessment and countermeasures. This makes it possible to quickly and objectively assess the possibility of power harassment and provide appropriate countermeasures to the victim.

[0245] "The incidents and comments entered by the user" refers to detailed information about specific incidents and comments related to power harassment that the user encountered in the workplace.

[0246] "Multiple pieces of information" refer to multiple data points or attributes related to power harassment, including the events that occurred and the content of statements.

[0247] "Means of acquisition" refers to the methods and functions for collecting information entered by the user and converting it into an appropriate format.

[0248] "Means for sending to the server side" refers to the communication means or protocol for sending the acquired information to the server via the network.

[0249] "Means for receiving transmitted information and recording it on a recording medium" refers to a method or function for receiving transmitted information and storing it in a database or storage device.

[0250] "Means for running a generative AI model" refers to the method or function of using a trained artificial intelligence model to assess potential power harassment in light of past cases.

[0251] "Means for generating judgment results and corresponding countermeasures, and preparing information for display" refers to the method or function for creating appropriate countermeasures based on the judgment results obtained by the AI ​​model and converting them into the format required for display to the user.

[0252] "Means for sending the generated results and countermeasures to the user's terminal" refers to the communication means or protocol for appropriately delivering the generated results and countermeasures to the user's terminal.

[0253] "Means for generating video simulations and transmitting highly realistic video to user terminals" refers to methods and functions for creating video that realistically reproduces actual power harassment scenes based on information provided by users and transmitting the video to user terminals.

[0254] "Means for immediate escalation to the relevant department if the assessment result is deemed to be serious" refers to a method or function for immediately sending a notification to the appropriate organization or department in charge urging them to take emergency action if the case of power harassment is deemed to be serious.

[0255] "Means for scoring cases using a generative AI model based on information entered by the user and proposing appropriate countermeasures" refers to methods or functions for using an AI model to assess the risk of power harassment based on information provided by the user and proposing appropriate countermeasures.

[0256] "Means for conducting detailed analysis of events by referring to past court cases and related cases" refers to methods and functions for conducting detailed analysis of current cases based on related court cases and cases that occurred in the past.

[0257] This invention is a system in which a user inputs detailed information related to power harassment in the workplace, and a generative AI model determines the possibility of power harassment based on that information and proposes appropriate countermeasures. This system is configured to include a user terminal, a server, a database, a generative AI model, and an image generation function.

[0258] System Program

[0259] User Input Processing

[0260] User

[0261] Using a smartphone app, users input detailed information about suspected power harassment, including the incident, trigger, location, details of the behavior, remarks, intonation of remarks, relationships, work history, duration, and similar past cases.

[0262] Smartphone app

[0263] The smartphone app acquires the information entered by the user, converts it into an appropriate format, and sends it to the server. The application is developed using Swift (iOS) and Kotlin (ANDROID (registered trademark)).

[0264] Data transmission and AI judgment

[0265] server

[0266] The server receives the information sent from the smartphone app and records it in a database using PostgreSQL.

[0267] Based on the recorded information, a generative AI model (using TENSORFLOW (registered trademark)) is run to compare it with past court cases and other cases to determine the possibility of power harassment. The generative AI model evaluates the input information and assigns a score to the possibility of power harassment.

[0268] Providing judgment results and countermeasures

[0269] server

[0270] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model, generates a judgment result, and also suggests countermeasures that the user should take.

[0271] The assessment results and countermeasures are sent to a smartphone app.

[0272] Smartphone app

[0273] The smartphone app displays the judgment results and countermeasures received from the server to the user.

[0274] Generate video simulations

[0275] server

[0276] The server generates a video simulation based on the user's input data, using Unity for video generation.

[0277] The generated simulation video is stored on a server and sent to a smartphone app.

[0278] Smartphone app

[0279] The smartphone app can visually display the video simulation sent from the server to the user.

[0280] Immediate Escalation

[0281] server

[0282] If the result is deemed malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting immediate action.

[0283] Specific examples

[0284] Example 1: Victim input

[0285] 1. User Input

[0286] The user input a scene in which their boss raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[0287] Relationship: Boss and subordinate

[0288] Work experience: 3 years

[0289] Location: Conference room

[0290] Previous similar incidents: Received similar reprimands three times

[0291] 2. Data Transmission

[0292] The smartphone app formats the data and sends it to the server.

[0293] 3. AI Judgment and Result Display

[0294] The server receives the data and uses an AI model to score the likelihood of power harassment. Example: 80%

[0295] The system generates a judgment result (high possibility of power harassment) and countermeasures (report to supervisor) and sends them to a smartphone.

[0296] 4. Video Simulation Generation

[0297] The server generates a video simulation and sends it to the smartphone.

[0298] The smartphone displays the simulation video in a viewable format for the user.

[0299] 5. Escalation

[0300] The server determines the case is serious and sends a notification to the HR department requesting immediate action.

[0301] Prompt Sentence Examples

[0302] Enter the following prompt into the generative AI model:

[0303] A user has submitted the following case of power harassment:

[0304] Incident: Your boss raises his voice and scolds you during a meeting.

[0305] What he said: Don't you know that?

[0306] Intonation: Very strong

[0307] Relationship: Boss and subordinate

[0308] Work experience: 3 years

[0309] Location: Conference room

[0310] Previous cases: Received similar reprimands three times

[0311] Based on this case, please assess and score the possibility of power harassment. Also, please suggest appropriate countermeasures.

[0312] In this way, the system can quickly provide an AI decision and appropriate response based on user input.

[0313] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0314] Step 1: User Input Processing

[0315] Users use a smartphone app to input information about suspected power harassment, including the incident, trigger, location, behavior, remarks, intonation of remarks, relationships, work history, duration, and similar past cases. The input information is then converted into an appropriate format within the smartphone app.

[0316] Input: Detailed information related to power harassment provided by the user

[0317] Output: Formatted information

[0318] Step 2: Submit your information

[0319] The device (smartphone app) sends the formatted information entered by the user to the server. HTTPS is used as the communication protocol.

[0320] Input: Formatted information

[0321] Output: Data sent to the server

[0322] Step 3: Record in the database

[0323] The server receives the information sent from the terminal and records it in a database. PostgreSQL is used for the database. The received data is properly organized by information item and saved in the database.

[0324] Input: Information sent from the device

[0325] Output: Information recorded in the database

[0326] Step 4: AI model decision

[0327] The server runs a generative AI model (using TensorFlow) based on the recorded information, which compares it with past court cases and other relevant cases to determine whether the person is engaging in power harassment. This process involves data analysis using neural networks.

[0328] Input: Information recorded in the database

[0329] Output: Possibility score of power harassment

[0330] Step 5: Generate results and countermeasures

[0331] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model and generates a judgment result. It also suggests countermeasures that the user should take (e.g., reporting to a supervisor, contacting human resources, etc.). The series of countermeasures are generated based on a template.

[0332] Input: Possibility score of power harassment, judgment result template

[0333] Output: Judgment results and countermeasures

[0334] Step 6: Submitting results and action plans

[0335] The server then sends the generated judgment results and countermeasures to the smartphone app, again using HTTPS as the communication protocol.

[0336] Input: Judgment result and countermeasures

[0337] Output: Data sent to smartphone app

[0338] Step 7: Viewing results and solutions

[0339] The device (smartphone app) displays the assessment results and countermeasures received from the server to the user. To display the results on an interface that is easy for the user to see, alerts and notifications are used in particular.

[0340] Input: Judgment result and countermeasures

[0341] Output: Displayed on the user's smartphone screen

[0342] Step 8: Generate the video simulation

[0343] The server generates a video simulation based on the user's input data. Unity is used for video generation, and the generated video is saved on the server.

[0344] Input: User-entered data

[0345] Output: Generated video simulation

[0346] Step 9: Send the video simulation

[0347] The server then sends the generated video simulation to a smartphone app, which also uses HTTPS as the communication protocol.

[0348] Input: Generated video simulation

[0349] Output: Data sent to smartphone app

[0350] Step 10: Viewing the video simulation

[0351] The terminal (smartphone app) displays the video simulation sent from the server in a viewable format for the user. An interactive UI is used to improve the user experience.

[0352] Input: Video simulation data

[0353] Output: Playback on the user's smartphone screen

[0354] Step 11: Immediate Escalation

[0355] If the server determines that the result is malicious, it immediately escalates it to the appropriate department, a process that automatically sends a detailed report along with a notification requesting immediate action.

[0356] Input: Malicious verdict

[0357] Output: Urgent notification to the relevant department

[0358] 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.

[0359] This invention relates to a system that uses a generative AI model to determine power harassment and provide countermeasures by combining detailed data related to power harassment entered by the user with an emotion engine that recognizes the emotions at the time. This system is composed of a user terminal, a server, a database, a generative AI model, an emotion engine, and a video generation function.

[0360] System Program

[0361] User Input Processing

[0362] User

[0363] Using a terminal, users input detailed information about suspected power harassment, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[0364] Terminal

[0365] The device receives data entered by the user, converts it into an appropriate format, and sends it to the server. At the same time, it also recognizes emotions from the user's facial expressions and voice, and sends this information to the server.

[0366] Data transmission and emotion recognition by emotion engine

[0367] server

[0368] The server receives the data sent from the device and records it in a database, including the user's emotional information recognized by the emotion engine.

[0369] The server runs a generative AI model based on the recorded data and emotional information. The generative AI model learns from past court cases and cases from other companies and the company, and evaluates the input data and emotional information to score the possibility of power harassment.

[0370] Providing judgment results and countermeasures

[0371] server

[0372] Based on the results of the AI ​​model, the server generates a judgment result indicating the possibility of power harassment. It also takes into account emotional information and creates countermeasures that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department").

[0373] The judgment result and countermeasures are sent to the user's terminal.

[0374] Terminal

[0375] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to comprehensively understand the possibility of power harassment and the actions they should take, including emotional information.

[0376] Generate video simulations

[0377] server

[0378] The server executes a model to generate a video simulation based on the user's input data and emotional information. The generated simulation video is highly realistic and reproduces the actual scene.

[0379] The generated video is stored on the server and sent to the user's terminal.

[0380] Terminal

[0381] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user, thereby enabling the user to visually understand the situation.

[0382] Immediate Escalation

[0383] server

[0384] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[0385] Specific examples

[0386] Example 1: Victim input

[0387] 1. User Input

[0388] The user input a scenario in which their boss suddenly raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[0389] Relationship: "Boss"

[0390] Work experience: 3 years

[0391] Location: "Conference Room"

[0392] Recognizes "fear" from the user's facial expression when inputting.

[0393] 2. Data Transmission

[0394] The device formats the data and sends it to the server, including the emotional information (fear).

[0395] 3. AI Judgment and Result Display

[0396] The server receives the data and uses an AI model to score the likelihood of power harassment (e.g., 80%).

[0397] The system generates a judgment result and countermeasures, taking into account emotional information, and sends them to the device.

[0398] The terminal displays the judgment result and countermeasures to the user.

[0399] 4. Video Simulation Generation

[0400] The server generates a video simulation and transmits it to the terminal.

[0401] The terminal displays the simulation video in a viewable form for the user.

[0402] 5. Escalation

[0403] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[0404] This system fairly and objectively assesses the possibility of power harassment and provides prompt countermeasures by taking into account the user's emotional information. It also contributes to preventing power harassment by using highly educational video simulations.

[0405] The processing flow will be explained below.

[0406] Step 1:

[0407] The user enters detailed information about suspected power harassment on the terminal, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[0408] Step 2:

[0409] When a user types, the device collects data such as the user's facial expression, voice, and typing speed in real time, and the emotion engine analyzes this data to recognize the user's emotional state (e.g., "fear," "anger," etc.).

[0410] Step 3:

[0411] The terminal acquires detailed information entered by the user and emotion data recognized by the emotion engine, and converts them into an appropriate format.

[0412] Step 4:

[0413] The terminal transmits the formatted detailed information and emotion data to the server.

[0414] Step 5:

[0415] The server receives the data sent from the device and records it in a database, where both detailed information and emotional data are stored.

[0416] Step 6:

[0417] The server runs a generative AI model based on the recorded data. The generative AI model learns from past court decisions and cases from other companies and the company, and evaluates the input data and emotional data to score the likelihood of power harassment.

[0418] Step 7:

[0419] The server generates a judgment result indicating the possibility of power harassment based on the results of the AI ​​model, and also takes into account the user's emotional information to create a response that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department").

[0420] Step 8:

[0421] The server transmits the generated judgment results, countermeasures, and emotion data to the user's terminal.

[0422] Step 9:

[0423] The terminal displays the judgment result, countermeasures, and emotion data received from the server to the user.

[0424] Step 10:

[0425] The server runs a model to generate a video simulation based on the user's input information and emotional data. The simulation video is highly realistic and reproduces specific scenes.

[0426] Step 11:

[0427] The server stores the generated video simulation and transmits it to the user's device.

[0428] Step 12:

[0429] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user.

[0430] Step 13:

[0431] If the result is deemed to be extremely egregious, the server will immediately escalate the issue to the appropriate department, which will include a detailed report and a notification requesting immediate action.

[0432] This processing flow allows the system to combine user input information and emotional data to accurately determine the possibility of power harassment and provide specific countermeasures. Furthermore, through video simulations, the system can contribute to education and raising awareness of power harassment prevention.

[0433] Example 2

[0434] 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."

[0435] Conventional power harassment response systems are primarily based solely on information entered by the user, and lack objectivity and consideration of emotional information. As a result, the determination of power harassment is subjective, and the countermeasures offered are limited. Furthermore, there is no simulation function for visually understanding the actual situation, which can make it difficult for users to select appropriate countermeasures. Furthermore, in extremely severe cases of power harassment, there is a lack of a rapid escalation system. The purpose of this system is to solve these problems.

[0436] 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.

[0437] In this invention, the server includes a means for acquiring emotional data from the user's facial expressions and voice, a means for executing an emotion engine based on the emotional data to evaluate the impact of power harassment, and a means for generating a video simulation and transmitting the highly realistic video to the user's terminal. This enables a fair and objective assessment of power harassment that takes into account the user's emotional information, provides appropriate countermeasures, and enables a visual understanding of the situation. Furthermore, in extremely severe cases, rapid escalation is possible.

[0438] "User" means an individual or end user who inputs information about power harassment into the system.

[0439] A "terminal" is a device used by a user to input information and receive analysis results and countermeasures.

[0440] A "server" is a computer system that receives data from a user and performs analysis and necessary processing.

[0441] "Database" means a data storage system for storing user-entered information and system-generated analysis results.

[0442] A "generative AI model" is an artificial intelligence algorithm that analyzes acts of power harassment and scores the possibility of power harassment based on past cases and court precedents.

[0443] The "emotion engine" is a system component that recognizes emotions from the user's facial expressions and voice, and generates emotion data based on that.

[0444] "Video simulation" is a visual simulation that recreates realistic scenes based on information and emotional data entered by the user.

[0445] "Escalation" is the process of notifying the relevant department of an emergency response to extremely serious cases of power harassment.

[0446] "Emotion data" is information about emotions recognized from the user's facial expressions and voice.

[0447] The "judgment result" is the evaluation score regarding the possibility of power harassment derived by the generative AI model and the countermeasures based on that score.

[0448] "Countermeasures" are action plans or recommendations that users should take if it is determined that there is a possibility of power harassment.

[0449] This invention relates to a system that combines detailed data related to power harassment entered by users with an emotion engine that recognizes the emotions at the time, and uses a generative AI model to determine power harassment and provide countermeasures. This system is configured to include a user terminal, a server, a database, a generative AI model, an emotion engine, and an image generation function.

[0450] User Input Processing

[0451] The user uses a device to input detailed information about the power harassment. Specifically, the user inputs details such as the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "reprimand"), the remarks (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history (e.g., "3 years"), and past experiences (e.g., "I've received similar reprimands three times in the past"). This input information is acquired along with the user's facial expressions and voice data, which are analyzed by the emotion engine. The device converts this data into an appropriate format and sends it to the server.

[0452] Data transmission and emotion recognition by emotion engine

[0453] The server receives the data sent from the device and records it in a database, including the user's emotional information recognized by the emotion engine. The server then runs a generative AI model based on the recorded data and emotional information. The generative AI model has learned from past court decisions and other cases, and evaluates the input data and emotional information to score the likelihood of power harassment.

[0454] Providing judgment results and countermeasures

[0455] The server generates a judgment result indicating the possibility of power harassment based on the results of the generative AI model. It also takes emotional information into consideration and creates countermeasures for the user to take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department"), and sends the judgment result and countermeasures to the user's device. The device displays the judgment result and countermeasures received from the server to the user, allowing the user to comprehensively understand the possibility of power harassment and the actions they should take.

[0456] Generate video simulations

[0457] The server generates a video simulation based on the user's input data and emotional information. This simulation video is highly realistic and designed to reproduce actual situations. The generated video is stored on the server and sent to the user's device. The device displays the video simulation sent from the server, allowing the user to visually confirm the situation.

[0458] Immediate Escalation

[0459] If the result is deemed to be extremely malicious, the server will immediately escalate the issue to the relevant department. The escalation notification will include a detailed report, and the relevant department will take immediate action.

[0460] Specific examples

[0461] As a specific example, the following case can be considered.

[0462] (Example 1: Victim input)

[0463] 1. User Input

[0464] The user enters, "My boss suddenly raised his voice during a meeting and scolded me, saying, 'Don't you understand that?'" The intonation of the remarks is "very strong," and the user also includes the detail that they felt pressured.

[0465] Select "Superior" as the relationship

[0466] Select "3 years" as work experience

[0467] Select "Meeting Room" as the location

[0468] Recognizes "fear" from the user's facial expression when inputting.

[0469] 2. Data Transmission

[0470] The device formats the data and sends it to the server, including the emotional information (fear).

[0471] 3. AI Judgment and Result Display

[0472] The server receives the data and uses a generative AI model to score the likelihood of power harassment (e.g., 80%).

[0473] The server also takes emotional information into consideration, generates a judgment result and countermeasures, and sends them to the terminal.

[0474] The terminal displays the judgment result and countermeasures to the user.

[0475] 4. Video Simulation Generation

[0476] The server generates a video simulation and transmits it to the terminal.

[0477] The terminal displays the simulation video in a viewable form for the user.

[0478] 5. Escalation

[0479] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[0480] Prompt Sentence Examples

[0481] Prompt example 1: Input of power harassment cases

[0482] "The boss raised his voice and scolded the employee during a meeting. It happened in a conference room during a project progress report. What was said: "Don't you understand that?", intonation: "Very strong tone", relationship: "Boss and subordinate", work history: "3 years", past experiences: "I've been scolded in a similar way three times in the past." The user felt fear while typing.

[0483] Based on this prompt, the system determines whether there is suspicion of power harassment and, if necessary, generates a video simulation or immediately escalates the situation.

[0484] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0485] Step 1: User Input Processing

[0486] The user inputs detailed information about the power harassment. This information includes the incident, trigger, location, behavior, remarks, intonation of remarks, relationships, work history, and past details. The device collects the data entered by the user and converts this information into digital format. The device also recognizes emotions from the user's facial expressions and voice data and collects emotional data.

[0487] Input: User-entered details, facial expressions, and voice data.

[0488] Output: Collected details and sentiment data.

[0489] Step 2: Send data

[0490] The device converts the collected details and emotion data into an appropriate format and transmits it to a server using a secure communication protocol to ensure data integrity.

[0491] Input: Details and sentiment data converted into digital form.

[0492] Output: The formatted data that is sent to the server.

[0493] Step 3: Database Recording

[0494] The server receives the data sent from the device and records it in a database, including the details and emotion data entered by the user.

[0495] Input: Formatted and submitted user details and sentiment data.

[0496] Output: Information recorded in the database.

[0497] Step 4: Run the generative AI model

[0498] The server runs a generative AI model based on the recorded data. The generative AI model has learned from past court decisions and similar cases, and analyzes the input data and emotional information to score the likelihood of power harassment.

[0499] Input: Details and emotion data recorded in the database.

[0500] Output: Scoring results indicating the possibility of power harassment.

[0501] Step 5: Generate judgment results

[0502] The server generates a judgment result indicating the possibility of power harassment based on the results of the generative AI model, and also considers emotional information to formulate countermeasures that the user should take.

[0503] Input: Scoring results and sentiment data.

[0504] Output: Verification result and countermeasures.

[0505] Step 6: Provide results and solutions

[0506] The server transmits the generated judgment result and countermeasure to the user's terminal.

[0507] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to understand the possibility of power harassment and the actions they should take.

[0508] Input: Verification result and countermeasures.

[0509] Output: Verification results and countermeasures displayed on the user's terminal.

[0510] Step 7: Generate video simulation

[0511] The server generates a video simulation using a generative AI model based on the user's input data and emotional information. This video is highly realistic and reproduces actual scenes. The server stores the generated video on its server and transmits it to the user's device.

[0512] Input: User details and emotion data.

[0513] Output: Video simulation.

[0514] Step 8: Image simulation display

[0515] The terminal receives the video simulation sent from the server and displays it in a form that the user can visually confirm, thereby enabling the user to visually understand the situation.

[0516] Input: Transmitted video simulation.

[0517] Output: The video simulation that is displayed to the user.

[0518] Step 9: Immediate Escalation

[0519] If the result is deemed to be extremely malicious, the server will immediately escalate the issue to the relevant department. The escalation notification will include a detailed report and require urgent action.

[0520] Input: Verdict and detailed report.

[0521] Output: Escalation notification sent to the responsible department.

[0522] (Application example 2)

[0523] 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."

[0524] Conventional power harassment detection systems require users to manually input detailed information, which makes it difficult to recognize emotions and make accurate judgments. Power harassment in factories is particularly difficult to detect, often negatively impacting work efficiency and the work environment. This has created a need for early problem detection and countermeasures.

[0525] 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.

[0526] In this invention, the server includes: means for acquiring multiple data sets including incidents and comments entered by the user; means for transmitting the acquired data to the server; means for receiving the transmitted data and recording it in a database; means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment; means for generating the determination results and corresponding countermeasures and preparing display data; means for transmitting the generated results and countermeasures to the user's terminal; means for generating a video simulation and transmitting highly realistic video to the user's terminal; means for immediately escalating the case to the relevant department if the determination results are deemed malicious; and means for a robot in a factory to monitor interactions between workers using surveillance cameras and voice recognition functions and detect possible acts of power harassment using an emotion engine and a generative AI model. This enables early detection and rapid response to power harassment in the work environment.

[0527] A "user terminal" is a device that a user operates to input information and check results.

[0528] A "server" is a computer system that receives, transmits, and processes data.

[0529] A "database" is a system that records acquired data and allows it to be searched and updated as needed.

[0530] A "generative AI model" is an artificial intelligence model that analyzes data and makes evaluations and predictions based on past cases and patterns.

[0531] The "emotion engine" is a system for recognizing emotions from user input data and audio / visual data.

[0532] A "surveillance camera" is a device that captures and records video in real time.

[0533] The "voice recognition function" is a function that converts voice into text and analyzes it.

[0534] "Power harassment" is the abuse of power in the workplace to cause mental or physical pain to others.

[0535] "Video simulation" is a technology that generates highly realistic images that reproduce real-world situations.

[0536] "Escalation" is the process of quickly notifying higher management or responsible departments when a problem occurs.

[0537] This invention is a system for detecting power harassment problems in factories and providing countermeasures, and is constructed by combining the following hardware and software.

[0538] Hardware configuration:

[0539] Surveillance camera (e.g., Logitech C920): Captures interactions between workers in real time and collects them as video data.

[0540] Microphone (e.g. RODE NT-USB): Collects audio while working and saves it as audio data.

[0541] Factory robots (e.g., Universal Robots UR series): Perform tasks within the factory and centrally process data from surveillance cameras and microphones.

[0542] Software configuration:

[0543] Speech recognition library (e.g., Google® Cloud Speech-to-Text API): Converts collected voice data into text and analyzes it.

[0544] Emotion recognition engine (e.g., Affectiva SDK): Recognizes the emotions of workers from collected video and audio data and stores them as data.

[0545] Generative AI model (e.g., OpenAI (registered trademark) GPT-4 (registered trademark)): Analyzes collected data based on past cases and precedents to assess the possibility of power harassment.

[0546] System operation description:

[0547] 1. Data Collection:

[0548] Surveillance cameras and microphones collect real-time images of work taking place inside the factory and store them as surveillance data.

[0549] The robot centrally manages this data and periodically sends it to a server.

[0550] 2. Emotion recognition:

[0551] The server sends the transmitted data to an emotion recognition engine to detect changes in the worker's emotions (e.g., "anger" or "fear").

[0552] The detected emotion data is input into a generative AI model.

[0553] 3. Power harassment judgment:

[0554] The generative AI model compares the input data with past cases and precedents to score the likelihood of power harassment.

[0555] Along with the judgment results, we will suggest countermeasures that the user should take.

[0556] 4. Providing solutions and escalation:

[0557] The results of the assessment and countermeasures are communicated to the worker via the robot's display and voice.

[0558] In serious cases, the server will automatically escalate the situation to the management department and prompt an emergency response.

[0559] Examples:

[0560] 1. Example: In a factory, a scene occurs in which worker A yells at worker B, saying, "Can't you even do something so simple?" A robot monitors this and automatically analyzes whether it is power harassment.

[0561] The surveillance camera captures the footage and recognizes Worker A's emotion as "anger."

[0562] The voice recognition library converts the voice data into text, and the emotion engine recognizes the emotion of Worker B, who felt "fear."

[0563] The generative AI model evaluates this and scores it as having an 80% chance of being power harassment.

[0564] Suggestion of countermeasures: The robot suggests "report to the HR department" and conveys this to Worker B. At the same time, the issue is escalated to the management department.

[0565] Example prompt sentence:

[0566] "A worker was scolded by his boss in a very harsh tone, saying, 'Can't you even do something as simple as this?' Please tell me if this situation is likely to be power harassment and what measures should be taken."

[0567] In this way, power harassment in the work environment can be detected early and dealt with promptly.

[0568] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0569] Step 1:

[0570] Data collection

[0571] The robot collects data entered by the user, including incidents and statements, as well as real-time video and audio data from surveillance cameras and microphones. The data entered by the user includes the incident, trigger, location, behavior, statements, intonation, the relationship between the perpetrator and victim, work history and duration, and past content. The collected data is converted into an appropriate format and sent to the server.

[0572] Input: User input data, surveillance camera video data, microphone audio data

[0573] Output: Sending collected data to a server

[0574] Step 2:

[0575] emotion recognition

[0576] The server receives the transmitted data and sends it to an emotion recognition engine to detect changes in the worker's emotions. The emotion recognition engine identifies emotions from the video and audio data and generates emotion data such as "anger" or "fear." This adds detailed information about the emotional changes.

[0577] Input: Collected video and audio data

[0578] Output: Emotion data generated by the emotion recognition engine

[0579] Step 3:

[0580] Power harassment judgment

[0581] The server inputs emotion data from the emotion recognition engine and detailed data entered by the user into a generative AI model. The generative AI model compares the data with past cases and legal precedents to determine the likelihood of power harassment. This score indicates the likelihood of power harassment and is used by users and the system to determine how seriously it should be taken.

[0582] Input: User details, emotion data

[0583] Output: Possibility score of power harassment

[0584] Step 4:

[0585] Presentation of judgment results and countermeasures

[0586] The server generates countermeasures based on the power harassment likelihood score obtained by the generative AI model. The results and countermeasures are formatted as display data and sent to the user's device. The countermeasures include specific actions such as "consulting with a higher-level manager" or "reporting to the human resources department."

[0587] Input: Possibility score of power harassment

[0588] Output: Sending countermeasures and judgment results to the user's device

[0589] Step 5:

[0590] Generate video simulations

[0591] The server runs a model that generates a video simulation based on the user's input data and emotional information. The model reproduces the real-world scene with high realism and transmits it to the user's device. The user can visually check the generated simulation video to gain a deeper understanding of the situation.

[0592] Input: User input data, emotional information

[0593] Output: Sending the video simulation to the user's device

[0594] Step 6:

[0595] escalation

[0596] If the server judges the result to be extremely malicious, it will immediately escalate the issue. When escalation occurs, a notification is sent to the responsible department requesting an emergency response, along with the judgment result and a detailed report. This allows the management department to take action quickly to resolve the problem.

[0597] Input: Verification results and detailed report

[0598] Output: Escalation notification to the responsible department

[0599] 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.

[0600] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0601] 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.

[0602] [Second embodiment]

[0603] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0604] 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.

[0605] 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).

[0606] 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.

[0607] 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.

[0608] 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).

[0609] 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.

[0610] 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.

[0611] 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.

[0612] 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.

[0613] In the smart glasses 214, the 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.

[0614] 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."

[0615] The present invention relates to a system that acquires detailed data related to power harassment input by users, and uses a generative AI model to determine power harassment and provide countermeasures. This system is composed of a user terminal, a server, a database, a generative AI model, and a video generation function.

[0616] System Program

[0617] User Input Processing

[0618] User

[0619] The user uses a terminal to input detailed information about the suspected power harassment behavior. Specifically, the user inputs the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "When reporting on the progress of the project"), the location (e.g., "Conference room"), the behavior (e.g., "Scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "Very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "3 years"), and past experiences (e.g., "I have received similar scolding three times in the past")).

[0620] Terminal

[0621] The terminal receives data input by the user, converts it into an appropriate format, and transmits it to the server.

[0622] Data transmission and AI judgment

[0623] server

[0624] The server receives the data sent from the terminal and records it in a database.

[0625] A generative AI model is run based on the recorded data, and the possibility of power harassment is judged by comparing it with past cases. The AI ​​model has learned from past court decisions and cases from other companies and the company itself, and evaluates the input data to generate a score for the possibility of power harassment.

[0626] Providing judgment results and countermeasures

[0627] server

[0628] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model and generates a judgment result. It also creates countermeasures that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the human resources department").

[0629] The judgment result and countermeasures are sent to the user's terminal.

[0630] Terminal

[0631] The terminal displays the judgment result and countermeasures received from the server to the user.

[0632] Generate video simulations

[0633] server

[0634] The server runs a model to generate a video simulation based on user input data. The generated simulation video is highly realistic, allowing users to visually confirm the scene.

[0635] The generated video is stored on the server and sent to the user's terminal.

[0636] Terminal

[0637] The terminal displays the video simulation sent from the server in a viewable form for the user.

[0638] Immediate Escalation

[0639] server

[0640] If the result is deemed to be extremely egregious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[0641] Specific examples

[0642] Example 1: Victim input

[0643] 1. User Input

[0644] The user input a scenario in which their boss suddenly raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[0645] Relationship: "Boss"

[0646] Work experience: 3 years

[0647] Location: "Conference Room"

[0648] 2. Data Transmission

[0649] The terminal formats the data and sends it to the server.

[0650] 3. AI Judgment and Result Display

[0651] The server receives the data and uses an AI model to score the likelihood of power harassment (e.g., 80%).

[0652] The server generates the judgment result and countermeasures and sends them to the terminal.

[0653] The terminal displays the judgment result and countermeasures to the user.

[0654] 4. Video Simulation Generation

[0655] The server generates a video simulation and transmits it to the terminal.

[0656] The terminal displays the simulation video in a viewable form for the user.

[0657] 5. Escalation

[0658] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[0659] This system contributes to preventing power harassment and improving the work environment by fairly and objectively assessing the possibility of power harassment, providing quick countermeasures, and using highly educational video simulations.

[0660] The processing flow will be explained below.

[0661] Step 1:

[0662] The user enters detailed information about suspected power harassment on the device, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[0663] Step 2:

[0664] The terminal takes the data entered by the user, converts it into the appropriate format, and sends it to the server, ensuring that the data is processed in a consistent format.

[0665] Step 3:

[0666] The server receives the data sent from the device, records it in a database, and stores it for further processing.

[0667] Step 4:

[0668] The server runs a generative AI model based on the recorded data. The generative AI model has learned from past court decisions and cases from other companies and the company, and evaluates the input data to generate a score for the possibility of power harassment.

[0669] Step 5:

[0670] Based on the results of the AI ​​model, the server generates a judgment result indicating the possibility of power harassment, and then creates countermeasures for the user to take (e.g., "consult with a higher-level manager" or "file a complaint with the human resources department") based on the judgment result.

[0671] Step 6:

[0672] The server sends the generated results and countermeasures data to the user's device, allowing the user to quickly check the results and countermeasures.

[0673] Step 7:

[0674] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to understand the possibility of power harassment and the actions they should take.

[0675] Step 8:

[0676] The server runs a model to generate a video simulation based on user input data. The generated simulation video is highly realistic and reproduces the actual scene.

[0677] Step 9:

[0678] The server stores the generated video simulation on the server and transmits a link or file to the user's terminal.

[0679] Step 10:

[0680] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user, thereby enabling the user to visually understand the situation.

[0681] Step 11:

[0682] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[0683] Through the specific processing flow described above, this system responds to power harassment issues fairly and efficiently, and provides prompt countermeasures. In addition, by using highly effective video simulations, it also contributes to preventing power harassment.

[0684] Example 1

[0685] 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."

[0686] Power harassment (pawahara) has become a serious social problem in modern workplaces, and there is a growing demand for systems that can fairly and quickly evaluate and provide appropriate countermeasures. However, current technology lacks the accuracy to make judgments based on detailed incident data or a method to provide visual evidence. This can lead to subjective judgments of power harassment and insufficient countermeasures. There is also a need for a system that can quickly escalate egregious cases of power harassment.

[0687] 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.

[0688] In this invention, the server includes: means for acquiring multiple data sets, including incidents and comments entered by users; means for transmitting the acquired data to the server; means for receiving the transmitted data and recording it in a database; means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment; means for generating a determination result and corresponding countermeasures and preparing display data; means for transmitting the generated results and countermeasures to the user's terminal; means for generating a video simulation based on detailed input data and transmitting highly realistic video to the user's terminal; and means for immediately escalating the generated result to the responsible department if the generated determination result is deemed to be extremely severe. This enables objective and rapid evaluation of cases of power harassment and provides appropriate countermeasures. Furthermore, providing visual evidence can deepen users' understanding and contribute to improving the work environment. Furthermore, severe cases can be quickly escalated to promote appropriate responses.

[0689] A "user" is a person who uses a terminal to input detailed information about a suspected case of power harassment.

[0690] "Occurrence of an incident" refers to a specific act or occurrence of power harassment.

[0691] "Content of remarks" refers to the specific words and actions that were spoken or taken during the power harassment situation.

[0692] "Data" refers to information entered by the user regarding the event, statement, trigger, location, action, intonation of statement, relationships, work history and period, and past content.

[0693] "Server" means a computer system that receives, processes, and stores data sent from a user terminal.

[0694] "Database" means a system for recording and storing data received by the server.

[0695] A "generative AI model" is an artificial intelligence algorithm that assesses the possibility of power harassment based on past cases and court precedents.

[0696] "Assessment" is the process by which the generative AI model scores the possibility of power harassment based on the data.

[0697] "Countermeasures" are specific instructions that suggest actions or measures that the user should take based on the judgment results.

[0698] "Video simulation" refers to highly realistic videos and animations that are generated based on data entered by the user.

[0699] "Escalation" is the process of contacting the department in charge of emergency response when a serious case of power harassment is determined.

[0700] The present invention relates to a system that acquires detailed data related to power harassment input by users, and uses a generative AI model to determine power harassment and provide countermeasures. This system is composed of a user terminal, a server, a database, a generative AI model, and a video generation function.

[0701] User Input Processing

[0702] User

[0703] The user uses a terminal to input detailed information about suspected power harassment. For example, the incident ("My boss raised his voice and scolded me during a meeting"), the trigger ("When reporting on the progress of a project"), the location ("Conference room"), the nature of the behavior ("Scolding"), what was said ("Don't you understand that?"), the intonation of the remarks ("Very harsh tone"), the relationship ("Boss and subordinate"), work history or length of time spent with the same organization ("3 years"), and past experiences ("I have received similar scolding three times in the past").

[0704] Terminal

[0705] The device takes the data entered by the user, converts it into the appropriate format, and sends it to the server, either using a standard web browser or a mobile application.

[0706] Data transmission and AI judgment

[0707] server

[0708] The server receives the data sent from the device and records it in a database, which uses SQL-based or NoSQL-based technologies (e.g., MySQL, MongoDB).

[0709] The server runs a generative AI model based on the recorded data, which learns from past court decisions and similar cases, evaluates the input data, and assigns a score to indicate the likelihood of power harassment.

[0710] Providing judgment results and countermeasures

[0711] server

[0712] The server determines the possibility of power harassment based on the results of the AI ​​model and generates specific countermeasures, such as advice to "consult with a higher-level manager" or "file a complaint with the human resources department."

[0713] The generated results and countermeasures are sent to the user's terminal.

[0714] Terminal

[0715] The terminal displays the judgment result and countermeasures received from the server to the user using a general message box or notification function.

[0716] Generate video simulations

[0717] server

[0718] The server runs a model to generate a visual simulation based on user input data, using widely used 3D modeling tools (e.g., Unity, Unreal Engine).

[0719] The generated simulation images are highly realistic, allowing users to visually confirm the scene.

[0720] The generated video is stored on the server and sent to the user's terminal.

[0721] Terminal

[0722] The device displays the video simulation sent from the server in a viewable format for the user, using a video player or a dedicated viewer within the application.

[0723] Immediate Escalation

[0724] server

[0725] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the relevant department. A detailed report will be sent along with a notification urging immediate action. Notifications will be sent via email or a dedicated notification system (e.g., Slack, Microsoft Teams).

[0726] Specific examples

[0727] Here is an example prompt:

[0728] Please rate and score whether this case constitutes power harassment based on the following details:

[0729] Incident: The boss raised his voice and scolded the employee during a meeting.

[0730] Trigger: Project progress report

[0731] Location: Conference room

[0732] Action: Scolding

[0733] What he said: Don't you know that?

[0734] Speech intonation: Very strong tone

[0735] Relationship: Boss and subordinate

[0736] Work experience: 3 years

[0737] Past experiences: I have received similar reprimands three times in the past.

[0738] This system not only fairly and objectively assesses the possibility of power harassment and provides prompt countermeasures, but also contributes to preventing power harassment and improving the work environment by using highly educational video simulations.

[0739] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0740] Processing flow

[0741] Step 1: Getting User Input

[0742] Step 2: Send data

[0743] Step 3: Record the data

[0744] Step 4: Generative AI model decision

[0745] Step 5: Generate results and countermeasures

[0746] Step 6: Displaying the results and countermeasures

[0747] Step 7: Generate the video simulation

[0748] Step 8: Viewing the video simulation

[0749] Step 9: Immediate Escalation

[0750] Processing step details

[0751] Step 1: Getting User Input

[0752] User

[0753] The user uses the terminal to input detailed information about the behavior suspected of being power harassment.

[0754] Input: Occurrence, trigger, location, action, statement, intonation, relationship, work history, past content

[0755] Output: Detailed data collected from the user

[0756] What happens: A user enters data into an input field on a web form or mobile application and presses the submit button.

[0757] Step 2: Send data

[0758] Terminal

[0759] The terminal converts the data entered by the user into an appropriate format (such as JSON or XML) and sends it to the server.

[0760] Input: Detailed data entered by the user

[0761] Output: The data format sent to the server

[0762] What it does: Creates an HTTP POST request and sends data to the specified endpoint on the server.

[0763] Step 3: Record the data

[0764] server

[0765] The server receives the data sent from the terminal and records it in a database.

[0766] Input: Formatted data sent from the terminal

[0767] Output: Data recorded in a database

[0768] What it does: Parses the received data and inserts it appropriately into a SQL or NoSQL database.

[0769] Step 4: Generative AI model decision

[0770] server

[0771] The server runs a generative AI model based on the data recorded in the database and compares it with past cases to determine the possibility of power harassment.

[0772] Input: User data recorded in the database

[0773] Output: Power harassment likelihood score

[0774] Specific operation: Convert the recorded data into prompt sentences, input them into a generative AI model, and obtain scoring results.

[0775] Example prompt:

[0776] Please rate and score whether this case constitutes power harassment based on the following details:

[0777] Incident: The boss raised his voice and scolded the employee during a meeting.

[0778] Trigger: Project progress report

[0779] Location: Conference room

[0780] Action: Scolding

[0781] What he said: Don't you know that?

[0782] Speech intonation: Very strong tone

[0783] Relationship: Boss and subordinate

[0784] Work experience: 3 years

[0785] Past experiences: I have received similar reprimands three times in the past.

[0786] Step 5: Generate results and countermeasures

[0787] server

[0788] Based on the results of the AI ​​model's assessment, the server generates specific countermeasures that the user should take.

[0789] Input: Judgment score

[0790] Output: Verification result and countermeasures

[0791] Specific actions: Analyze the scoring results and generate countermeasures based on the results (e.g., "consult with upper management" or "file a complaint with the HR department").

[0792] Step 6: Displaying the results and countermeasures

[0793] Terminal

[0794] The terminal displays the judgment result and countermeasures received from the server to the user.

[0795] Input: Verification result and countermeasures sent from the server

[0796] Output: Verification results and countermeasures displayed to the user

[0797] Specific behavior: The judgment result and countermeasures will be displayed as a pop-up message or notification.

[0798] Step 7: Generate the video simulation

[0799] server

[0800] The server executes a model that generates a visual simulation based on user input data.

[0801] Input: User details

[0802] Output: Video simulation data

[0803] What it does: Uses 3D modeling tools (e.g. Unity, Unreal Engine) to generate highly realistic images based on input data.

[0804] Step 8: Viewing the video simulation

[0805] Terminal

[0806] The terminal displays the video simulation sent from the server to the user.

[0807] Input: Video simulation data

[0808] Output: Simulation video displayed to the user

[0809] Specific operation: Play the video using a video player or a dedicated viewer within the application.

[0810] Step 9: Immediate Escalation

[0811] server

[0812] If the result of the assessment is deemed to be extremely malicious, the server will immediately escalate the matter to the relevant department.

[0813] Input: Verdict (extremely malicious)

[0814] Output: Escalation notification to the responsible department

[0815] Actions: Send a detailed report and a notification to the relevant department to prompt immediate action via email or a dedicated notification system (e.g., Slack, Microsoft Teams).

[0816] (Application example 1)

[0817] 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."

[0818] Power harassment (pa-hara) has become a serious problem in modern workplaces. This problem damages the mental and physical health of employees, reduces work efficiency, and lowers workplace morale. However, the determination of power harassment is subjective and requires specific criteria and evidence, making it difficult to address. There is also the problem of a lack of information provided to victims to take appropriate action. There is a need for a system that can solve these issues and enable swift and objective responses.

[0819] 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.

[0820] In this invention, the server includes means for scoring cases using a generative AI model based on information entered by the user and proposing appropriate countermeasures, means for conducting a detailed analysis of the incident with reference to past court cases and related cases, and means for providing the user with the results of the assessment and countermeasures. This makes it possible to quickly and objectively assess the possibility of power harassment and provide appropriate countermeasures to the victim.

[0821] "The incidents and comments entered by the user" refers to detailed information about specific incidents and comments related to power harassment that the user encountered in the workplace.

[0822] "Multiple pieces of information" refer to multiple data points or attributes related to power harassment, including the events that occurred and the content of statements.

[0823] "Means of acquisition" refers to the methods and functions for collecting information entered by the user and converting it into an appropriate format.

[0824] "Means for sending to the server side" refers to the communication means or protocol for sending the acquired information to the server via the network.

[0825] "Means for receiving transmitted information and recording it on a recording medium" refers to a method or function for receiving transmitted information and storing it in a database or storage device.

[0826] "Means for running a generative AI model" refers to the method or function of using a trained artificial intelligence model to assess potential power harassment in light of past cases.

[0827] "Means for generating judgment results and corresponding countermeasures, and preparing information for display" refers to the method or function for creating appropriate countermeasures based on the judgment results obtained by the AI ​​model and converting them into the format required for display to the user.

[0828] "Means for sending the generated results and countermeasures to the user's terminal" refers to the communication means or protocol for appropriately delivering the generated results and countermeasures to the user's terminal.

[0829] "Means for generating video simulations and transmitting highly realistic video to user terminals" refers to methods and functions for creating video that realistically reproduces actual power harassment scenes based on information provided by users and transmitting the video to user terminals.

[0830] "Means for immediate escalation to the relevant department if the assessment result is deemed to be serious" refers to a method or function for immediately sending a notification to the appropriate organization or department in charge urging them to take emergency action if the case of power harassment is deemed to be serious.

[0831] "Means for scoring cases using a generative AI model based on information entered by the user and proposing appropriate countermeasures" refers to methods or functions for using an AI model to assess the risk of power harassment based on information provided by the user and proposing appropriate countermeasures.

[0832] "Means for conducting detailed analysis of events by referring to past court cases and related cases" refers to methods and functions for conducting detailed analysis of current cases based on related court cases and cases that occurred in the past.

[0833] This invention is a system in which a user inputs detailed information related to power harassment in the workplace, and a generative AI model determines the possibility of power harassment based on that information and proposes appropriate countermeasures. This system is configured to include a user terminal, a server, a database, a generative AI model, and an image generation function.

[0834] System Program

[0835] User Input Processing

[0836] User

[0837] Using a smartphone app, users input detailed information about suspected power harassment, including the incident, trigger, location, details of the behavior, remarks, intonation of remarks, relationships, work history, duration, and similar past cases.

[0838] Smartphone app

[0839] The smartphone app receives information entered by the user, converts it into an appropriate format, and sends it to the server. The application is developed using Swift (iOS) and Kotlin (Android).

[0840] Data transmission and AI judgment

[0841] server

[0842] The server receives the information sent from the smartphone app and records it in a database using PostgreSQL.

[0843] Based on the recorded information, a generative AI model (using TensorFlow) is run to compare it with past court cases and other cases to determine the possibility of power harassment. The generative AI model evaluates the input information and assigns a score to the possibility of power harassment.

[0844] Providing judgment results and countermeasures

[0845] server

[0846] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model, generates a judgment result, and also suggests countermeasures that the user should take.

[0847] The assessment results and countermeasures are sent to a smartphone app.

[0848] Smartphone app

[0849] The smartphone app displays the judgment results and countermeasures received from the server to the user.

[0850] Generate video simulations

[0851] server

[0852] The server generates a video simulation based on the user's input data, using Unity for video generation.

[0853] The generated simulation video is stored on a server and sent to a smartphone app.

[0854] Smartphone app

[0855] The smartphone app can visually display the video simulation sent from the server to the user.

[0856] Immediate Escalation

[0857] server

[0858] If the result is deemed malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting immediate action.

[0859] Specific examples

[0860] Example 1: Victim input

[0861] 1. User Input

[0862] The user input a scene in which their boss raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[0863] Relationship: Boss and subordinate

[0864] Work experience: 3 years

[0865] Location: Conference room

[0866] Previous similar incidents: Received similar reprimands three times

[0867] 2. Data Transmission

[0868] The smartphone app formats the data and sends it to the server.

[0869] 3. AI Judgment and Result Display

[0870] The server receives the data and uses an AI model to score the likelihood of power harassment. Example: 80%

[0871] The system generates a judgment result (high possibility of power harassment) and countermeasures (report to supervisor) and sends them to a smartphone.

[0872] 4. Video Simulation Generation

[0873] The server generates a video simulation and sends it to the smartphone.

[0874] The smartphone displays the simulation video in a viewable format for the user.

[0875] 5. Escalation

[0876] The server determines the case is serious and sends a notification to the HR department requesting immediate action.

[0877] Prompt Sentence Examples

[0878] Enter the following prompt into the generative AI model:

[0879] A user has submitted the following case of power harassment:

[0880] Incident: Your boss raises his voice and scolds you during a meeting.

[0881] What he said: Don't you know that?

[0882] Intonation: Very strong

[0883] Relationship: Boss and subordinate

[0884] Work experience: 3 years

[0885] Location: Conference room

[0886] Previous cases: Received similar reprimands three times

[0887] Based on this case, please assess and score the possibility of power harassment. Also, please suggest appropriate countermeasures.

[0888] In this way, the system can quickly provide an AI decision and appropriate response based on user input.

[0889] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0890] Step 1: User Input Processing

[0891] Users use a smartphone app to input information about suspected power harassment, including the incident, trigger, location, behavior, remarks, intonation of remarks, relationships, work history, duration, and similar past cases. The input information is then converted into an appropriate format within the smartphone app.

[0892] Input: Detailed information related to power harassment provided by the user

[0893] Output: Formatted information

[0894] Step 2: Submit your information

[0895] The device (smartphone app) sends the formatted information entered by the user to the server. HTTPS is used as the communication protocol.

[0896] Input: Formatted information

[0897] Output: Data sent to the server

[0898] Step 3: Record in the database

[0899] The server receives the information sent from the terminal and records it in a database. PostgreSQL is used for the database. The received data is properly organized by information item and saved in the database.

[0900] Input: Information sent from the device

[0901] Output: Information recorded in the database

[0902] Step 4: AI model decision

[0903] The server runs a generative AI model (using TensorFlow) based on the recorded information, which compares it with past court cases and other relevant cases to determine whether the person is engaging in power harassment. This process involves data analysis using neural networks.

[0904] Input: Information recorded in the database

[0905] Output: Possibility score of power harassment

[0906] Step 5: Generate results and countermeasures

[0907] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model and generates a judgment result. It also suggests countermeasures that the user should take (e.g., reporting to a supervisor, contacting human resources, etc.). The series of countermeasures are generated based on a template.

[0908] Input: Possibility score of power harassment, judgment result template

[0909] Output: Judgment results and countermeasures

[0910] Step 6: Submitting results and action plans

[0911] The server then sends the generated judgment results and countermeasures to the smartphone app, again using HTTPS as the communication protocol.

[0912] Input: Judgment result and countermeasures

[0913] Output: Data sent to smartphone app

[0914] Step 7: Viewing results and solutions

[0915] The device (smartphone app) displays the assessment results and countermeasures received from the server to the user. To display the results on an interface that is easy for the user to see, alerts and notifications are used in particular.

[0916] Input: Judgment result and countermeasures

[0917] Output: Displayed on the user's smartphone screen

[0918] Step 8: Generate the video simulation

[0919] The server generates a video simulation based on the user's input data. Unity is used for video generation, and the generated video is saved on the server.

[0920] Input: User-entered data

[0921] Output: Generated video simulation

[0922] Step 9: Send the video simulation

[0923] The server then sends the generated video simulation to a smartphone app, which also uses HTTPS as the communication protocol.

[0924] Input: Generated video simulation

[0925] Output: Data sent to smartphone app

[0926] Step 10: Viewing the video simulation

[0927] The terminal (smartphone app) displays the video simulation sent from the server in a viewable format for the user. An interactive UI is used to improve the user experience.

[0928] Input: Video simulation data

[0929] Output: Playback on the user's smartphone screen

[0930] Step 11: Immediate Escalation

[0931] If the server determines that the result is malicious, it immediately escalates it to the appropriate department, a process that automatically sends a detailed report along with a notification requesting immediate action.

[0932] Input: Malicious verdict

[0933] Output: Urgent notification to the relevant department

[0934] 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.

[0935] This invention relates to a system that uses a generative AI model to determine power harassment and provide countermeasures by combining detailed data related to power harassment entered by the user with an emotion engine that recognizes the emotions at the time. This system is composed of a user terminal, a server, a database, a generative AI model, an emotion engine, and a video generation function.

[0936] System Program

[0937] User Input Processing

[0938] User

[0939] Using a terminal, users input detailed information about suspected power harassment, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[0940] Terminal

[0941] The device receives data entered by the user, converts it into an appropriate format, and sends it to the server. At the same time, it also recognizes emotions from the user's facial expressions and voice, and sends this information to the server.

[0942] Data transmission and emotion recognition by emotion engine

[0943] server

[0944] The server receives the data sent from the device and records it in a database, including the user's emotional information recognized by the emotion engine.

[0945] The server runs a generative AI model based on the recorded data and emotional information. The generative AI model learns from past court cases and cases from other companies and the company, and evaluates the input data and emotional information to score the possibility of power harassment.

[0946] Providing judgment results and countermeasures

[0947] server

[0948] Based on the results of the AI ​​model, the server generates a judgment result indicating the possibility of power harassment. It also takes into account emotional information and creates countermeasures that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department").

[0949] The judgment result and countermeasures are sent to the user's terminal.

[0950] Terminal

[0951] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to comprehensively understand the possibility of power harassment and the actions they should take, including emotional information.

[0952] Generate video simulations

[0953] server

[0954] The server executes a model to generate a video simulation based on the user's input data and emotional information. The generated simulation video is highly realistic and reproduces the actual scene.

[0955] The generated video is stored on the server and sent to the user's terminal.

[0956] Terminal

[0957] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user, thereby enabling the user to visually understand the situation.

[0958] Immediate Escalation

[0959] server

[0960] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[0961] Specific examples

[0962] Example 1: Victim input

[0963] 1. User Input

[0964] The user input a scenario in which their boss suddenly raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[0965] Relationship: "Boss"

[0966] Work experience: 3 years

[0967] Location: "Conference Room"

[0968] Recognizes "fear" from the user's facial expression when inputting.

[0969] 2. Data Transmission

[0970] The device formats the data and sends it to the server, including the emotional information (fear).

[0971] 3. AI Judgment and Result Display

[0972] The server receives the data and uses an AI model to score the likelihood of power harassment (e.g., 80%).

[0973] The system generates a judgment result and countermeasures, taking into account emotional information, and sends them to the device.

[0974] The terminal displays the judgment result and countermeasures to the user.

[0975] 4. Video Simulation Generation

[0976] The server generates a video simulation and transmits it to the terminal.

[0977] The terminal displays the simulation video in a viewable form for the user.

[0978] 5. Escalation

[0979] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[0980] This system fairly and objectively assesses the possibility of power harassment and provides prompt countermeasures by taking into account the user's emotional information. It also contributes to preventing power harassment by using highly educational video simulations.

[0981] The processing flow will be explained below.

[0982] Step 1:

[0983] The user enters detailed information about suspected power harassment on the terminal, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[0984] Step 2:

[0985] When a user types, the device collects data such as the user's facial expression, voice, and typing speed in real time, and the emotion engine analyzes this data to recognize the user's emotional state (e.g., "fear," "anger," etc.).

[0986] Step 3:

[0987] The terminal acquires detailed information entered by the user and emotion data recognized by the emotion engine, and converts them into an appropriate format.

[0988] Step 4:

[0989] The terminal transmits the formatted detailed information and emotion data to the server.

[0990] Step 5:

[0991] The server receives the data sent from the device and records it in a database, where both detailed information and emotional data are stored.

[0992] Step 6:

[0993] The server runs a generative AI model based on the recorded data. The generative AI model learns from past court decisions and cases from other companies and the company, and evaluates the input data and emotional data to score the likelihood of power harassment.

[0994] Step 7:

[0995] The server generates a judgment result indicating the possibility of power harassment based on the results of the AI ​​model, and also takes into account the user's emotional information to create a response that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department").

[0996] Step 8:

[0997] The server transmits the generated judgment results, countermeasures, and emotion data to the user's terminal.

[0998] Step 9:

[0999] The terminal displays the judgment result, countermeasures, and emotion data received from the server to the user.

[1000] Step 10:

[1001] The server runs a model to generate a video simulation based on the user's input information and emotional data. The simulation video is highly realistic and reproduces specific scenes.

[1002] Step 11:

[1003] The server stores the generated video simulation and transmits it to the user's device.

[1004] Step 12:

[1005] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user.

[1006] Step 13:

[1007] If the result is deemed to be extremely egregious, the server will immediately escalate the issue to the appropriate department, which will include a detailed report and a notification requesting immediate action.

[1008] This processing flow allows the system to combine user input information and emotional data to accurately determine the possibility of power harassment and provide specific countermeasures. Furthermore, through video simulations, the system can contribute to education and raising awareness of power harassment prevention.

[1009] Example 2

[1010] 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."

[1011] Conventional power harassment response systems are primarily based solely on information entered by the user, and lack objectivity and consideration of emotional information. As a result, the determination of power harassment is subjective, and the countermeasures offered are limited. Furthermore, there is no simulation function for visually understanding the actual situation, which can make it difficult for users to select appropriate countermeasures. Furthermore, in extremely severe cases of power harassment, there is a lack of a rapid escalation system. The purpose of this system is to solve these problems.

[1012] 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.

[1013] In this invention, the server includes a means for acquiring emotional data from the user's facial expressions and voice, a means for executing an emotion engine based on the emotional data to evaluate the impact of power harassment, and a means for generating a video simulation and transmitting the highly realistic video to the user's terminal. This enables a fair and objective assessment of power harassment that takes into account the user's emotional information, provides appropriate countermeasures, and enables a visual understanding of the situation. Furthermore, in extremely severe cases, rapid escalation is possible.

[1014] "User" means an individual or end user who inputs information about power harassment into the system.

[1015] A "terminal" is a device used by a user to input information and receive analysis results and countermeasures.

[1016] A "server" is a computer system that receives data from a user and performs analysis and necessary processing.

[1017] "Database" means a data storage system for storing user-entered information and system-generated analysis results.

[1018] A "generative AI model" is an artificial intelligence algorithm that analyzes acts of power harassment and scores the possibility of power harassment based on past cases and court precedents.

[1019] The "emotion engine" is a system component that recognizes emotions from the user's facial expressions and voice, and generates emotion data based on that.

[1020] "Video simulation" is a visual simulation that recreates realistic scenes based on information and emotional data entered by the user.

[1021] "Escalation" is the process of notifying the relevant department of an emergency response to extremely serious cases of power harassment.

[1022] "Emotion data" is information about emotions recognized from the user's facial expressions and voice.

[1023] The "judgment result" is the evaluation score regarding the possibility of power harassment derived by the generative AI model and the countermeasures based on that score.

[1024] "Countermeasures" are action plans or recommendations that users should take if it is determined that there is a possibility of power harassment.

[1025] This invention relates to a system that combines detailed data related to power harassment entered by users with an emotion engine that recognizes the emotions at the time, and uses a generative AI model to determine power harassment and provide countermeasures. This system is configured to include a user terminal, a server, a database, a generative AI model, an emotion engine, and an image generation function.

[1026] User Input Processing

[1027] The user uses a device to input detailed information about the power harassment. Specifically, the user inputs details such as the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "reprimand"), the remarks (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history (e.g., "3 years"), and past experiences (e.g., "I've received similar reprimands three times in the past"). This input information is acquired along with the user's facial expressions and voice data, which are analyzed by the emotion engine. The device converts this data into an appropriate format and sends it to the server.

[1028] Data transmission and emotion recognition by emotion engine

[1029] The server receives the data sent from the device and records it in a database, including the user's emotional information recognized by the emotion engine. The server then runs a generative AI model based on the recorded data and emotional information. The generative AI model has learned from past court decisions and other cases, and evaluates the input data and emotional information to score the likelihood of power harassment.

[1030] Providing judgment results and countermeasures

[1031] The server generates a judgment result indicating the possibility of power harassment based on the results of the generative AI model. It also takes emotional information into consideration and creates countermeasures for the user to take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department"), and sends the judgment result and countermeasures to the user's device. The device displays the judgment result and countermeasures received from the server to the user, allowing the user to comprehensively understand the possibility of power harassment and the actions they should take.

[1032] Generate video simulations

[1033] The server generates a video simulation based on the user's input data and emotional information. This simulation video is highly realistic and designed to reproduce actual situations. The generated video is stored on the server and sent to the user's device. The device displays the video simulation sent from the server, allowing the user to visually confirm the situation.

[1034] Immediate Escalation

[1035] If the result is deemed to be extremely malicious, the server will immediately escalate the issue to the relevant department. The escalation notification will include a detailed report, and the relevant department will take immediate action.

[1036] Specific examples

[1037] As a specific example, the following case can be considered.

[1038] (Example 1: Victim input)

[1039] 1. User Input

[1040] The user enters, "My boss suddenly raised his voice during a meeting and scolded me, saying, 'Don't you understand that?'" The intonation of the remarks is "very strong," and the user also includes the detail that they felt pressured.

[1041] Select "Superior" as the relationship

[1042] Select "3 years" as work experience

[1043] Select "Meeting Room" as the location

[1044] Recognizes "fear" from the user's facial expression when inputting.

[1045] 2. Data Transmission

[1046] The device formats the data and sends it to the server, including the emotional information (fear).

[1047] 3. AI Judgment and Result Display

[1048] The server receives the data and uses a generative AI model to score the likelihood of power harassment (e.g., 80%).

[1049] The server also takes emotional information into consideration, generates a judgment result and countermeasures, and sends them to the terminal.

[1050] The terminal displays the judgment result and countermeasures to the user.

[1051] 4. Video Simulation Generation

[1052] The server generates a video simulation and transmits it to the terminal.

[1053] The terminal displays the simulation video in a viewable form for the user.

[1054] 5. Escalation

[1055] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[1056] Prompt Sentence Examples

[1057] Prompt example 1: Input of power harassment cases

[1058] "The boss raised his voice and scolded the employee during a meeting. It happened in a conference room during a project progress report. What was said: "Don't you understand that?", intonation: "Very strong tone", relationship: "Boss and subordinate", work history: "3 years", past experiences: "I've been scolded in a similar way three times in the past." The user felt fear while typing.

[1059] Based on this prompt, the system determines whether there is suspicion of power harassment and, if necessary, generates a video simulation or immediately escalates the situation.

[1060] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1061] Step 1: User Input Processing

[1062] The user inputs detailed information about the power harassment. This information includes the incident, trigger, location, behavior, remarks, intonation of remarks, relationships, work history, and past details. The device collects the data entered by the user and converts this information into digital format. The device also recognizes emotions from the user's facial expressions and voice data and collects emotional data.

[1063] Input: User-entered details, facial expressions, and voice data.

[1064] Output: Collected details and sentiment data.

[1065] Step 2: Send data

[1066] The device converts the collected details and emotion data into an appropriate format and transmits it to a server using a secure communication protocol to ensure data integrity.

[1067] Input: Details and sentiment data converted into digital form.

[1068] Output: The formatted data that is sent to the server.

[1069] Step 3: Database Recording

[1070] The server receives the data sent from the device and records it in a database, including the details and emotion data entered by the user.

[1071] Input: Formatted and submitted user details and sentiment data.

[1072] Output: Information recorded in the database.

[1073] Step 4: Run the generative AI model

[1074] The server runs a generative AI model based on the recorded data. The generative AI model has learned from past court decisions and similar cases, and analyzes the input data and emotional information to score the likelihood of power harassment.

[1075] Input: Details and emotion data recorded in the database.

[1076] Output: Scoring results indicating the possibility of power harassment.

[1077] Step 5: Generate judgment results

[1078] The server generates a judgment result indicating the possibility of power harassment based on the results of the generative AI model, and also considers emotional information to formulate countermeasures that the user should take.

[1079] Input: Scoring results and sentiment data.

[1080] Output: Verification result and countermeasures.

[1081] Step 6: Provide results and solutions

[1082] The server transmits the generated judgment result and countermeasure to the user's terminal.

[1083] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to understand the possibility of power harassment and the actions they should take.

[1084] Input: Verification result and countermeasures.

[1085] Output: Verification results and countermeasures displayed on the user's terminal.

[1086] Step 7: Generate video simulation

[1087] The server generates a video simulation using a generative AI model based on the user's input data and emotional information. This video is highly realistic and reproduces actual scenes. The server stores the generated video on its server and transmits it to the user's device.

[1088] Input: User details and emotion data.

[1089] Output: Video simulation.

[1090] Step 8: Image simulation display

[1091] The terminal receives the video simulation sent from the server and displays it in a form that the user can visually confirm, thereby enabling the user to visually understand the situation.

[1092] Input: Transmitted video simulation.

[1093] Output: The video simulation that is displayed to the user.

[1094] Step 9: Immediate Escalation

[1095] If the result is deemed to be extremely malicious, the server will immediately escalate the issue to the relevant department. The escalation notification will include a detailed report and require urgent action.

[1096] Input: Verdict and detailed report.

[1097] Output: Escalation notification sent to the responsible department.

[1098] (Application example 2)

[1099] 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."

[1100] Conventional power harassment detection systems require users to manually input detailed information, which makes it difficult to recognize emotions and make accurate judgments. Power harassment in factories is particularly difficult to detect, often negatively impacting work efficiency and the work environment. This has created a need for early problem detection and countermeasures.

[1101] 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.

[1102] In this invention, the server includes: means for acquiring multiple data sets including incidents and comments entered by the user; means for transmitting the acquired data to the server; means for receiving the transmitted data and recording it in a database; means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment; means for generating the determination results and corresponding countermeasures and preparing display data; means for transmitting the generated results and countermeasures to the user's terminal; means for generating a video simulation and transmitting highly realistic video to the user's terminal; means for immediately escalating the case to the relevant department if the determination results are deemed malicious; and means for a robot in a factory to monitor interactions between workers using surveillance cameras and voice recognition functions and detect possible acts of power harassment using an emotion engine and a generative AI model. This enables early detection and rapid response to power harassment in the work environment.

[1103] A "user terminal" is a device that a user operates to input information and check results.

[1104] A "server" is a computer system that receives, transmits, and processes data.

[1105] A "database" is a system that records acquired data and allows it to be searched and updated as needed.

[1106] A "generative AI model" is an artificial intelligence model that analyzes data and makes evaluations and predictions based on past cases and patterns.

[1107] The "emotion engine" is a system for recognizing emotions from user input data and audio / visual data.

[1108] A "surveillance camera" is a device that captures and records video in real time.

[1109] The "voice recognition function" is a function that converts voice into text and analyzes it.

[1110] "Power harassment" is the abuse of power in the workplace to cause mental or physical pain to others.

[1111] "Video simulation" is a technology that generates highly realistic images that reproduce real-world situations.

[1112] "Escalation" is the process of quickly notifying higher management or responsible departments when a problem occurs.

[1113] This invention is a system for detecting power harassment problems in factories and providing countermeasures, and is constructed by combining the following hardware and software.

[1114] Hardware configuration:

[1115] Surveillance camera (e.g., Logitech C920): Captures interactions between workers in real time and collects them as video data.

[1116] Microphone (e.g. RODE NT-USB): Collects audio while working and saves it as audio data.

[1117] Factory robots (e.g., Universal Robots UR series): Perform tasks within the factory and centrally process data from surveillance cameras and microphones.

[1118] Software configuration:

[1119] Speech recognition library (e.g., Google Cloud Speech-to-Text API): Converts collected voice data into text and analyzes it.

[1120] Emotion recognition engine (e.g., Affectiva SDK): Recognizes the emotions of workers from collected video and audio data and stores them as data.

[1121] Generative AI model (e.g., OpenAI GPT-4): Analyzes collected data based on past cases and precedents to assess the possibility of power harassment.

[1122] System operation description:

[1123] 1. Data Collection:

[1124] Surveillance cameras and microphones collect real-time images of work taking place inside the factory and store them as surveillance data.

[1125] The robot centrally manages this data and periodically sends it to a server.

[1126] 2. Emotion recognition:

[1127] The server sends the transmitted data to an emotion recognition engine to detect changes in the worker's emotions (e.g., "anger" or "fear").

[1128] The detected emotion data is input into a generative AI model.

[1129] 3. Power harassment judgment:

[1130] The generative AI model compares the input data with past cases and precedents to score the likelihood of power harassment.

[1131] Along with the judgment results, we will suggest countermeasures that the user should take.

[1132] 4. Providing solutions and escalation:

[1133] The results of the assessment and countermeasures are communicated to the worker via the robot's display and voice.

[1134] In serious cases, the server will automatically escalate the situation to the management department and prompt an emergency response.

[1135] Examples:

[1136] 1. Example: In a factory, a scene occurs in which worker A yells at worker B, saying, "Can't you even do something so simple?" A robot monitors this and automatically analyzes whether it is power harassment.

[1137] The surveillance camera captures the footage and recognizes Worker A's emotion as "anger."

[1138] The voice recognition library converts the voice data into text, and the emotion engine recognizes the emotion of Worker B, who felt "fear."

[1139] The generative AI model evaluates this and scores it as having an 80% chance of being power harassment.

[1140] Suggestion of countermeasures: The robot suggests "report to the HR department" and conveys this to Worker B. At the same time, the issue is escalated to the management department.

[1141] Example prompt sentence:

[1142] "A worker was scolded by his boss in a very harsh tone, saying, 'Can't you even do something as simple as this?' Please tell me if this situation is likely to be power harassment and what measures should be taken."

[1143] In this way, power harassment in the work environment can be detected early and dealt with promptly.

[1144] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1145] Step 1:

[1146] Data collection

[1147] The robot collects data entered by the user, including incidents and statements, as well as real-time video and audio data from surveillance cameras and microphones. The data entered by the user includes the incident, trigger, location, behavior, statements, intonation, the relationship between the perpetrator and victim, work history and duration, and past content. The collected data is converted into an appropriate format and sent to the server.

[1148] Input: User input data, surveillance camera video data, microphone audio data

[1149] Output: Sending collected data to a server

[1150] Step 2:

[1151] emotion recognition

[1152] The server receives the transmitted data and sends it to an emotion recognition engine to detect changes in the worker's emotions. The emotion recognition engine identifies emotions from the video and audio data and generates emotion data such as "anger" or "fear." This adds detailed information about the emotional changes.

[1153] Input: Collected video and audio data

[1154] Output: Emotion data generated by the emotion recognition engine

[1155] Step 3:

[1156] Power harassment judgment

[1157] The server inputs emotion data from the emotion recognition engine and detailed data entered by the user into a generative AI model. The generative AI model compares the data with past cases and legal precedents to determine the likelihood of power harassment. This score indicates the likelihood of power harassment and is used by users and the system to determine how seriously it should be taken.

[1158] Input: User details, emotion data

[1159] Output: Possibility score of power harassment

[1160] Step 4:

[1161] Presentation of judgment results and countermeasures

[1162] The server generates countermeasures based on the power harassment likelihood score obtained by the generative AI model. The results and countermeasures are formatted as display data and sent to the user's device. The countermeasures include specific actions such as "consulting with a higher-level manager" or "reporting to the human resources department."

[1163] Input: Possibility score of power harassment

[1164] Output: Sending countermeasures and judgment results to the user's device

[1165] Step 5:

[1166] Generate video simulations

[1167] The server runs a model that generates a video simulation based on the user's input data and emotional information. The model reproduces the real-world scene with high realism and transmits it to the user's device. The user can visually check the generated simulation video to gain a deeper understanding of the situation.

[1168] Input: User input data, emotional information

[1169] Output: Sending the video simulation to the user's device

[1170] Step 6:

[1171] escalation

[1172] If the server judges the result to be extremely malicious, it will immediately escalate the issue. When escalation occurs, a notification is sent to the responsible department requesting an emergency response, along with the judgment result and a detailed report. This allows the management department to take action quickly to resolve the problem.

[1173] Input: Verification results and detailed report

[1174] Output: Escalation notification to the responsible department

[1175] 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.

[1176] 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.

[1177] 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.

[1178] [Third embodiment]

[1179] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1180] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1181] 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).

[1182] 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.

[1183] 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.

[1184] 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).

[1185] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1186] 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.

[1187] 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.

[1188] 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.

[1189] 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.

[1190] 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."

[1191] The present invention relates to a system that acquires detailed data related to power harassment input by users, and uses a generative AI model to determine power harassment and provide countermeasures. This system is composed of a user terminal, a server, a database, a generative AI model, and a video generation function.

[1192] System Program

[1193] User Input Processing

[1194] User

[1195] The user uses a terminal to input detailed information about the suspected power harassment behavior. Specifically, the user inputs the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "When reporting on the progress of the project"), the location (e.g., "Conference room"), the behavior (e.g., "Scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "Very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "3 years"), and past experiences (e.g., "I have received similar scolding three times in the past")).

[1196] Terminal

[1197] The terminal receives data input by the user, converts it into an appropriate format, and transmits it to the server.

[1198] Data transmission and AI judgment

[1199] server

[1200] The server receives the data sent from the terminal and records it in a database.

[1201] A generative AI model is run based on the recorded data, and the possibility of power harassment is judged by comparing it with past cases. The AI ​​model has learned from past court decisions and cases from other companies and the company, and evaluates the input data to generate a score for the possibility of power harassment.

[1202] Providing judgment results and countermeasures

[1203] server

[1204] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model and generates a judgment result. It also creates countermeasures that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the human resources department").

[1205] The judgment result and countermeasures are sent to the user's terminal.

[1206] Terminal

[1207] The terminal displays the judgment result and countermeasures received from the server to the user.

[1208] Generate video simulations

[1209] server

[1210] The server runs a model to generate a video simulation based on user input data. The generated simulation video is highly realistic, allowing users to visually confirm the scene.

[1211] The generated video is stored on the server and sent to the user's terminal.

[1212] Terminal

[1213] The terminal displays the video simulation sent from the server in a viewable form for the user.

[1214] Immediate Escalation

[1215] server

[1216] If the result is deemed to be extremely egregious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[1217] Specific examples

[1218] Example 1: Victim input

[1219] 1. User Input

[1220] The user input a scene in which their boss suddenly raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[1221] Relationship: "Boss"

[1222] Work experience: 3 years

[1223] Location: "Conference Room"

[1224] 2. Data Transmission

[1225] The terminal formats the data and sends it to the server.

[1226] 3. AI Judgment and Result Display

[1227] The server receives the data and uses an AI model to score the likelihood of power harassment (e.g., 80%).

[1228] The server generates the judgment result and countermeasures and sends them to the terminal.

[1229] The terminal displays the judgment result and countermeasures to the user.

[1230] 4. Video Simulation Generation

[1231] The server generates a video simulation and transmits it to the terminal.

[1232] The terminal displays the simulation video in a viewable form for the user.

[1233] 5. Escalation

[1234] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[1235] This system contributes to preventing power harassment and improving the work environment by fairly and objectively assessing the possibility of power harassment, providing quick countermeasures, and using highly educational video simulations.

[1236] The processing flow will be explained below.

[1237] Step 1:

[1238] The user enters detailed information about suspected power harassment on the device, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[1239] Step 2:

[1240] The terminal takes the data entered by the user, converts it into the appropriate format, and sends it to the server, ensuring that the data is processed in a consistent format.

[1241] Step 3:

[1242] The server receives the data sent from the device, records it in a database, and stores it for further processing.

[1243] Step 4:

[1244] The server runs a generative AI model based on the recorded data. The generative AI model has learned from past court decisions and cases from other companies and the company, and evaluates the input data to generate a score for the possibility of power harassment.

[1245] Step 5:

[1246] Based on the results of the AI ​​model, the server generates a judgment result indicating the possibility of power harassment, and then creates countermeasures for the user to take (e.g., "consult with a higher-level manager" or "file a complaint with the human resources department") based on the judgment result.

[1247] Step 6:

[1248] The server sends the generated results and countermeasures data to the user's device, allowing the user to quickly check the results and countermeasures.

[1249] Step 7:

[1250] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to understand the possibility of power harassment and the actions they should take.

[1251] Step 8:

[1252] The server runs a model to generate a video simulation based on user input data. The generated simulation video is highly realistic and reproduces the actual scene.

[1253] Step 9:

[1254] The server stores the generated video simulation on the server and transmits a link or file to the user's terminal.

[1255] Step 10:

[1256] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user, thereby enabling the user to visually understand the situation.

[1257] Step 11:

[1258] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[1259] Through the specific processing flow described above, this system responds to power harassment issues fairly and efficiently, and provides prompt countermeasures. In addition, by using highly effective video simulations, it also contributes to preventing power harassment.

[1260] Example 1

[1261] 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."

[1262] Power harassment (pawahara) has become a serious social problem in modern workplaces, and there is a growing demand for systems that can fairly and quickly evaluate and provide appropriate countermeasures. However, current technology lacks the accuracy to make judgments based on detailed incident data or a method to provide visual evidence. This can lead to subjective judgments of power harassment and insufficient countermeasures. There is also a need for a system that can quickly escalate egregious cases of power harassment.

[1263] 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.

[1264] In this invention, the server includes: means for acquiring multiple data sets, including incidents and comments entered by users; means for transmitting the acquired data to the server; means for receiving the transmitted data and recording it in a database; means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment; means for generating a determination result and corresponding countermeasures and preparing display data; means for transmitting the generated results and countermeasures to the user's terminal; means for generating a video simulation based on detailed input data and transmitting highly realistic video to the user's terminal; and means for immediately escalating the generated result to the responsible department if the generated determination result is deemed to be extremely severe. This enables objective and rapid evaluation of cases of power harassment and provides appropriate countermeasures. Furthermore, providing visual evidence can deepen users' understanding and contribute to improving the work environment. Furthermore, severe cases can be quickly escalated to promote appropriate responses.

[1265] A "user" is a person who uses a terminal to input detailed information about a suspected case of power harassment.

[1266] "Occurrence of an incident" refers to a specific act or occurrence of power harassment.

[1267] "Content of remarks" refers to the specific words and actions that were spoken or taken during the power harassment situation.

[1268] "Data" refers to information entered by the user regarding the event, the content of the statement, the trigger, the location, the action, the intonation of the statement, the relationship, the work history and period, and past content.

[1269] "Server" means a computer system that receives, processes, and stores data sent from a user terminal.

[1270] "Database" means a system for recording and storing data received by the server.

[1271] A "generative AI model" is an artificial intelligence algorithm that assesses the possibility of power harassment based on past cases and court precedents.

[1272] "Assessment" is the process by which the generative AI model scores the possibility of power harassment based on the data.

[1273] "Countermeasures" are specific instructions that suggest actions or measures that the user should take based on the judgment results.

[1274] "Video simulation" refers to highly realistic videos and animations that are generated based on data entered by the user.

[1275] "Escalation" is the process of contacting the department in charge of emergency response when a serious case of power harassment is determined.

[1276] The present invention relates to a system that acquires detailed data related to power harassment input by users, and uses a generative AI model to determine power harassment and provide countermeasures. This system is composed of a user terminal, a server, a database, a generative AI model, and a video generation function.

[1277] User Input Processing

[1278] User

[1279] The user uses a terminal to input detailed information about suspected power harassment. For example, the incident ("My boss raised his voice and scolded me during a meeting"), the trigger ("When reporting on the progress of a project"), the location ("Conference room"), the nature of the behavior ("Scolding"), what was said ("Don't you understand that?"), the intonation of the remarks ("Very harsh tone"), the relationship ("Boss and subordinate"), work history or length of time spent with the same organization ("3 years"), and past experiences ("I have received similar scolding three times in the past").

[1280] Terminal

[1281] The device takes the data entered by the user, converts it into the appropriate format, and sends it to the server, either using a standard web browser or a mobile application.

[1282] Data transmission and AI judgment

[1283] server

[1284] The server receives the data sent from the device and records it in a database, which uses SQL-based or NoSQL-based technologies (e.g., MySQL, MongoDB).

[1285] The server runs a generative AI model based on the recorded data, which learns from past court decisions and similar cases, evaluates the input data, and assigns a score to indicate the likelihood of power harassment.

[1286] Providing judgment results and countermeasures

[1287] server

[1288] The server determines the possibility of power harassment based on the results of the AI ​​model and generates specific countermeasures, such as advice to "consult with a higher-ranking manager" or "file a complaint with the human resources department."

[1289] The generated results and countermeasures are sent to the user's terminal.

[1290] Terminal

[1291] The terminal displays the judgment result and countermeasures received from the server to the user using a general message box or notification function.

[1292] Generate video simulations

[1293] server

[1294] The server runs a model to generate a visual simulation based on user input data, using widely used 3D modeling tools (e.g., Unity, Unreal Engine).

[1295] The generated simulation images are highly realistic, allowing users to visually confirm the scene.

[1296] The generated video is stored on the server and sent to the user's terminal.

[1297] Terminal

[1298] The device displays the video simulation sent from the server in a viewable format for the user, using a video player or a dedicated viewer within the application.

[1299] Immediate Escalation

[1300] server

[1301] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the relevant department. A detailed report will be sent along with a notification urging immediate action. Notifications will be sent via email or a dedicated notification system (e.g., Slack, Microsoft Teams).

[1302] Specific examples

[1303] Here is an example prompt:

[1304] Please rate and score whether this case constitutes power harassment based on the following details:

[1305] Incident: The boss raised his voice and scolded the employee during a meeting.

[1306] Trigger: Project progress report

[1307] Location: Conference room

[1308] Action: Scolding

[1309] What he said: Don't you know that?

[1310] Speech intonation: Very strong tone

[1311] Relationship: Boss and subordinate

[1312] Work experience: 3 years

[1313] Past experiences: I have received similar reprimands three times in the past.

[1314] This system not only fairly and objectively assesses the possibility of power harassment and provides prompt countermeasures, but also contributes to preventing power harassment and improving the work environment by using highly educational video simulations.

[1315] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1316] Processing flow

[1317] Step 1: Getting User Input

[1318] Step 2: Send data

[1319] Step 3: Record the data

[1320] Step 4: Generative AI model decision

[1321] Step 5: Generate results and countermeasures

[1322] Step 6: Displaying the results and countermeasures

[1323] Step 7: Generate the video simulation

[1324] Step 8: Viewing the video simulation

[1325] Step 9: Immediate Escalation

[1326] Processing step details

[1327] Step 1: Getting User Input

[1328] User

[1329] The user uses the terminal to input detailed information about the behavior suspected of being power harassment.

[1330] Input: Occurrence, trigger, location, action, statement, intonation, relationship, work history, past content

[1331] Output: Detailed data collected from the user

[1332] What happens: A user enters data into an input field on a web form or mobile application and presses the submit button.

[1333] Step 2: Send data

[1334] Terminal

[1335] The terminal converts the data entered by the user into an appropriate format (such as JSON or XML) and sends it to the server.

[1336] Input: Detailed data entered by the user

[1337] Output: The data format sent to the server

[1338] What it does: Creates an HTTP POST request and sends data to the specified endpoint on the server.

[1339] Step 3: Record the data

[1340] server

[1341] The server receives the data sent from the terminal and records it in a database.

[1342] Input: Formatted data sent from the terminal

[1343] Output: Data recorded in a database

[1344] What it does: Parses the received data and inserts it appropriately into a SQL or NoSQL database.

[1345] Step 4: Generative AI model decision

[1346] server

[1347] The server runs a generative AI model based on the data recorded in the database and compares it with past cases to determine the possibility of power harassment.

[1348] Input: User data recorded in the database

[1349] Output: Power harassment likelihood score

[1350] Specific operation: Convert the recorded data into prompt sentences, input them into the generative AI model, and obtain the scoring results.

[1351] Example prompt:

[1352] Please rate and score whether this case constitutes power harassment based on the following details:

[1353] Incident: The boss raised his voice and scolded the employee during a meeting.

[1354] Trigger: Project progress report

[1355] Location: Conference room

[1356] Action: Scolding

[1357] What he said: Don't you know that?

[1358] Speech intonation: Very strong tone

[1359] Relationship: Boss and subordinate

[1360] Work experience: 3 years

[1361] Past experiences: I have received similar reprimands three times in the past.

[1362] Step 5: Generate results and countermeasures

[1363] server

[1364] Based on the results of the AI ​​model's assessment, the server generates specific countermeasures that the user should take.

[1365] Input: Judgment score

[1366] Output: Verification result and countermeasures

[1367] Specific actions: Analyze the scoring results and generate countermeasures based on the results (e.g., "consult with upper management" or "file a complaint with the HR department").

[1368] Step 6: Displaying the results and countermeasures

[1369] Terminal

[1370] The terminal displays the judgment result and countermeasures received from the server to the user.

[1371] Input: Verification result and countermeasures sent from the server

[1372] Output: Verification results and countermeasures displayed to the user

[1373] Specific behavior: The judgment result and countermeasures will be displayed as a pop-up message or notification.

[1374] Step 7: Generate the video simulation

[1375] server

[1376] The server executes a model that generates a visual simulation based on user input data.

[1377] Input: User details

[1378] Output: Video simulation data

[1379] What it does: Uses 3D modeling tools (e.g. Unity, Unreal Engine) to generate highly realistic images based on input data.

[1380] Step 8: Viewing the video simulation

[1381] Terminal

[1382] The terminal displays the video simulation sent from the server to the user.

[1383] Input: Video simulation data

[1384] Output: Simulation video displayed to the user

[1385] Specific operation: Play the video using a video player or a dedicated viewer within the application.

[1386] Step 9: Immediate Escalation

[1387] server

[1388] If the result of the assessment is deemed to be extremely malicious, the server will immediately escalate the matter to the relevant department.

[1389] Input: Verdict (extremely malicious)

[1390] Output: Escalation notification to the responsible department

[1391] Specific actions: Send a detailed report and a notification to the relevant department to prompt immediate action via email or a dedicated notification system (e.g., Slack, Microsoft Teams).

[1392] (Application example 1)

[1393] 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."

[1394] Power harassment (pa-hara) has become a serious problem in modern workplaces. This problem damages the mental and physical health of employees, reduces work efficiency, and lowers workplace morale. However, the determination of power harassment is subjective and requires specific criteria and evidence, making it difficult to address. There is also the problem of a lack of information provided to victims to take appropriate action. There is a need for a system that can solve these issues and enable swift and objective responses.

[1395] 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.

[1396] In this invention, the server includes means for scoring cases using a generative AI model based on information entered by the user and proposing appropriate countermeasures, means for conducting a detailed analysis of the incident with reference to past court cases and related cases, and means for providing the user with the results of the assessment and countermeasures. This makes it possible to quickly and objectively assess the possibility of power harassment and provide appropriate countermeasures to the victim.

[1397] "The incidents and comments entered by the user" refers to detailed information about specific incidents and comments related to power harassment that the user encountered in the workplace.

[1398] "Multiple pieces of information" refer to multiple data points or attributes related to power harassment, including the events that occurred and the content of statements.

[1399] "Means of acquisition" refers to the methods and functions for collecting information entered by the user and converting it into an appropriate format.

[1400] "Means for sending to the server side" refers to the communication means or protocol for sending the acquired information to the server via the network.

[1401] "Means for receiving transmitted information and recording it on a recording medium" refers to a method or function for receiving transmitted information and storing it in a database or storage device.

[1402] "Means for running a generative AI model" refers to the method or function of using a trained artificial intelligence model to assess potential power harassment in light of past cases.

[1403] "Means for generating judgment results and corresponding countermeasures, and preparing information for display" refers to the method or function for creating appropriate countermeasures based on the judgment results obtained by the AI ​​model and converting them into the format required for display to the user.

[1404] "Means for sending the generated results and countermeasures to the user's terminal" refers to the communication means or protocol for appropriately delivering the generated results and countermeasures to the user's terminal.

[1405] "Means for generating video simulations and transmitting highly realistic video to user terminals" refers to methods and functions for creating video that realistically reproduces actual power harassment scenes based on information provided by users and transmitting the video to user terminals.

[1406] "Means for immediate escalation to the relevant department if the assessment result is deemed to be serious" refers to a method or function for immediately sending a notification to the appropriate organization or department in charge urging them to take emergency action if the case of power harassment is deemed to be serious.

[1407] "Means for scoring cases using a generative AI model based on information entered by the user and proposing appropriate countermeasures" refers to methods or functions for using an AI model to assess the risk of power harassment based on information provided by the user and proposing appropriate countermeasures.

[1408] "Means for conducting detailed analysis of events by referring to past court cases and related cases" refers to methods and functions for conducting detailed analysis of current cases based on related court cases and cases that occurred in the past.

[1409] This invention is a system in which a user inputs detailed information related to power harassment in the workplace, and a generative AI model determines the possibility of power harassment based on that information and proposes appropriate countermeasures. This system is configured to include a user terminal, a server, a database, a generative AI model, and an image generation function.

[1410] System Program

[1411] User Input Processing

[1412] User

[1413] Using a smartphone app, users input detailed information about suspected power harassment, including the incident, trigger, location, details of the behavior, remarks, intonation of remarks, relationships, work history, duration, and similar past cases.

[1414] Smartphone app

[1415] The smartphone app receives information entered by the user, converts it into an appropriate format, and sends it to the server. The application is developed using Swift (iOS) and Kotlin (Android).

[1416] Data transmission and AI judgment

[1417] server

[1418] The server receives the information sent from the smartphone app and records it in a database using PostgreSQL.

[1419] Based on the recorded information, a generative AI model (using TensorFlow) is run to compare it with past court cases and other cases to determine the possibility of power harassment. The generative AI model evaluates the input information and assigns a score to the possibility of power harassment.

[1420] Providing judgment results and countermeasures

[1421] server

[1422] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model, generates a judgment result, and also suggests countermeasures that the user should take.

[1423] The assessment results and countermeasures are sent to a smartphone app.

[1424] Smartphone app

[1425] The smartphone app displays the judgment results and countermeasures received from the server to the user.

[1426] Generate video simulations

[1427] server

[1428] The server generates a video simulation based on the user's input data, using Unity for video generation.

[1429] The generated simulation video is stored on a server and sent to a smartphone app.

[1430] Smartphone app

[1431] The smartphone app can visually display the video simulation sent from the server to the user.

[1432] Immediate Escalation

[1433] server

[1434] If the result is deemed malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting immediate action.

[1435] Specific examples

[1436] Example 1: Victim input

[1437] 1. User Input

[1438] The user input a scene in which their boss raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[1439] Relationship: Boss and subordinate

[1440] Work experience: 3 years

[1441] Location: Conference room

[1442] Previous similar incidents: Received similar reprimands three times

[1443] 2. Data Transmission

[1444] The smartphone app formats the data and sends it to the server.

[1445] 3. AI Judgment and Result Display

[1446] The server receives the data and uses an AI model to score the likelihood of power harassment. Example: 80%

[1447] The system generates a judgment result (high possibility of power harassment) and countermeasures (report to supervisor) and sends them to a smartphone.

[1448] 4. Video Simulation Generation

[1449] The server generates a video simulation and sends it to the smartphone.

[1450] The smartphone displays the simulation video in a viewable format for the user.

[1451] 5. Escalation

[1452] The server determines the case is serious and sends a notification to the HR department requesting immediate action.

[1453] Prompt Sentence Examples

[1454] Enter the following prompt into the generative AI model:

[1455] A user has submitted the following case of power harassment:

[1456] Incident: Your boss raises his voice and scolds you during a meeting.

[1457] What he said: Don't you know that?

[1458] Intonation: Very strong

[1459] Relationship: Boss and subordinate

[1460] Work experience: 3 years

[1461] Location: Conference room

[1462] Previous cases: Received similar reprimands three times

[1463] Based on this case, please assess and score the possibility of power harassment. Also, please suggest appropriate countermeasures.

[1464] In this way, the system can quickly provide an AI decision and appropriate response based on user input.

[1465] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1466] Step 1: User Input Processing

[1467] Users use a smartphone app to input information about suspected power harassment, including the incident, trigger, location, behavior, remarks, intonation of remarks, relationships, work history, duration, and similar past cases. The input information is then converted into an appropriate format within the smartphone app.

[1468] Input: Detailed information related to power harassment provided by the user

[1469] Output: Formatted information

[1470] Step 2: Submit your information

[1471] The device (smartphone app) sends the formatted information entered by the user to the server. HTTPS is used as the communication protocol.

[1472] Input: Formatted information

[1473] Output: Data sent to the server

[1474] Step 3: Record in the database

[1475] The server receives the information sent from the terminal and records it in a database. PostgreSQL is used for the database. The received data is properly organized by information item and saved in the database.

[1476] Input: Information sent from the device

[1477] Output: Information recorded in the database

[1478] Step 4: AI model decision

[1479] The server runs a generative AI model (using TensorFlow) based on the recorded information, which compares it with past court cases and other relevant cases to determine whether the person is engaging in power harassment. This process involves data analysis using neural networks.

[1480] Input: Information recorded in the database

[1481] Output: Possibility score of power harassment

[1482] Step 5: Generate results and countermeasures

[1483] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model and generates a judgment result. It also suggests countermeasures that the user should take (e.g., reporting to a supervisor, contacting human resources, etc.). The series of countermeasures are generated based on a template.

[1484] Input: Possibility score of power harassment, judgment result template

[1485] Output: Judgment results and countermeasures

[1486] Step 6: Submitting results and action plans

[1487] The server then sends the generated judgment results and countermeasures to the smartphone app, again using HTTPS as the communication protocol.

[1488] Input: Judgment result and countermeasures

[1489] Output: Data sent to smartphone app

[1490] Step 7: Viewing results and solutions

[1491] The device (smartphone app) displays the assessment results and countermeasures received from the server to the user. To display the results on an interface that is easy for the user to see, alerts and notifications are used in particular.

[1492] Input: Judgment result and countermeasures

[1493] Output: Displayed on the user's smartphone screen

[1494] Step 8: Generate the video simulation

[1495] The server generates a video simulation based on the user's input data. Unity is used for video generation, and the generated video is saved on the server.

[1496] Input: User-entered data

[1497] Output: Generated video simulation

[1498] Step 9: Send the video simulation

[1499] The server then sends the generated video simulation to a smartphone app, which also uses HTTPS as the communication protocol.

[1500] Input: Generated video simulation

[1501] Output: Data sent to smartphone app

[1502] Step 10: Viewing the video simulation

[1503] The terminal (smartphone app) displays the video simulation sent from the server in a viewable format for the user. An interactive UI is used to improve the user experience.

[1504] Input: Video simulation data

[1505] Output: Playback on the user's smartphone screen

[1506] Step 11: Immediate Escalation

[1507] If the server determines that the result is malicious, it immediately escalates it to the appropriate department, a process that automatically sends a detailed report along with a notification requesting immediate action.

[1508] Input: Malicious judgment result

[1509] Output: Urgent notification to the relevant department

[1510] 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.

[1511] This invention relates to a system that uses a generative AI model to determine power harassment and provide countermeasures by combining detailed data related to power harassment entered by the user with an emotion engine that recognizes the emotions at the time. This system is composed of a user terminal, a server, a database, a generative AI model, an emotion engine, and a video generation function.

[1512] System Program

[1513] User Input Processing

[1514] User

[1515] Using a terminal, users input detailed information about suspected power harassment, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[1516] Terminal

[1517] The device receives data entered by the user, converts it into an appropriate format, and sends it to the server. At the same time, it also recognizes emotions from the user's facial expressions and voice, and sends this information to the server.

[1518] Data transmission and emotion recognition by emotion engine

[1519] server

[1520] The server receives the data sent from the device and records it in a database, including the user's emotional information recognized by the emotion engine.

[1521] The server runs a generative AI model based on the recorded data and emotional information. The generative AI model learns from past court cases and cases from other companies and the company, and evaluates the input data and emotional information to score the possibility of power harassment.

[1522] Providing judgment results and countermeasures

[1523] server

[1524] Based on the results of the AI ​​model, the server generates a judgment result indicating the possibility of power harassment. It also takes into account emotional information and creates countermeasures that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department").

[1525] The judgment result and countermeasures are sent to the user's terminal.

[1526] Terminal

[1527] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to comprehensively understand the possibility of power harassment and the actions they should take, including emotional information.

[1528] Generate video simulations

[1529] server

[1530] The server executes a model to generate a video simulation based on the user's input data and emotional information. The generated simulation video is highly realistic and reproduces the actual scene.

[1531] The generated video is stored on the server and sent to the user's terminal.

[1532] Terminal

[1533] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user, thereby enabling the user to visually understand the situation.

[1534] Immediate Escalation

[1535] server

[1536] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[1537] Specific examples

[1538] Example 1: Victim input

[1539] 1. User Input

[1540] The user input a scenario in which their boss suddenly raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[1541] Relationship: "Boss"

[1542] Work experience: 3 years

[1543] Location: "Conference Room"

[1544] Recognizes "fear" from the user's facial expression when inputting.

[1545] 2. Data Transmission

[1546] The device formats the data and sends it to the server, including the emotional information (fear).

[1547] 3. AI Judgment and Result Display

[1548] The server receives the data and uses an AI model to score the likelihood of power harassment (e.g., 80%).

[1549] The system generates a judgment result and countermeasures, taking into account emotional information, and sends them to the device.

[1550] The terminal displays the judgment result and countermeasures to the user.

[1551] 4. Video Simulation Generation

[1552] The server generates a video simulation and transmits it to the terminal.

[1553] The terminal displays the simulation video in a viewable form for the user.

[1554] 5. Escalation

[1555] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[1556] This system fairly and objectively judges the possibility of power harassment and provides prompt countermeasures by taking into account the user's emotional information. It also contributes to preventing power harassment by using highly educational video simulations.

[1557] The processing flow will be explained below.

[1558] Step 1:

[1559] The user inputs detailed information about the suspected power harassment on the terminal, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "When reporting on the progress of the project"), the location (e.g., "Conference room"), the nature of the behavior (e.g., "Scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "Very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "3 years"), and past experiences (e.g., "I have received similar scolding three times in the past")

[1560] Step 2:

[1561] When a user types, the device collects data such as the user's facial expression, voice, and typing speed in real time, and the emotion engine analyzes this data to recognize the user's emotional state (e.g., "fear," "anger," etc.).

[1562] Step 3:

[1563] The terminal acquires detailed information entered by the user and emotion data recognized by the emotion engine, and converts them into an appropriate format.

[1564] Step 4:

[1565] The terminal transmits the formatted detailed information and emotion data to the server.

[1566] Step 5:

[1567] The server receives the data sent from the device and records it in a database, where both detailed information and emotional data are stored.

[1568] Step 6:

[1569] The server runs a generative AI model based on the recorded data. The generative AI model learns from past court decisions and cases from other companies and the company, and evaluates the input data and emotional data to score the likelihood of power harassment.

[1570] Step 7:

[1571] The server generates a judgment result indicating the possibility of power harassment based on the results of the AI ​​model, and also takes into account the user's emotional information to create a response that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department").

[1572] Step 8:

[1573] The server transmits the generated judgment results, countermeasures, and emotion data to the user's terminal.

[1574] Step 9:

[1575] The terminal displays the judgment result, countermeasures, and emotion data received from the server to the user.

[1576] Step 10:

[1577] The server runs a model to generate a video simulation based on the user's input information and emotional data. The simulation video is highly realistic and reproduces specific scenes.

[1578] Step 11:

[1579] The server stores the generated video simulation and transmits it to the user's device.

[1580] Step 12:

[1581] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user.

[1582] Step 13:

[1583] If the result is deemed to be extremely egregious, the server will immediately escalate the issue to the appropriate department, which will include a detailed report and a notification requesting immediate action.

[1584] This processing flow allows the system to combine user input information and emotional data to accurately determine the possibility of power harassment and provide specific countermeasures. Furthermore, through video simulations, the system can contribute to education and raising awareness of power harassment prevention.

[1585] Example 2

[1586] 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."

[1587] Conventional power harassment response systems are primarily based solely on information entered by the user, and lack objectivity and consideration of emotional information. As a result, the determination of power harassment is subjective, and the countermeasures offered are limited. Furthermore, there is no simulation function for visually understanding the actual situation, which can make it difficult for users to select appropriate countermeasures. Furthermore, in extremely severe cases of power harassment, there is a lack of a rapid escalation system. The purpose of this system is to solve these problems.

[1588] 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.

[1589] In this invention, the server includes a means for acquiring emotional data from the user's facial expressions and voice, a means for executing an emotion engine based on the emotional data to evaluate the impact of power harassment, and a means for generating a video simulation and transmitting the highly realistic video to the user's terminal. This enables a fair and objective assessment of power harassment that takes into account the user's emotional information, provides appropriate countermeasures, and enables a visual understanding of the situation. Furthermore, in extremely severe cases, rapid escalation is possible.

[1590] "User" means an individual or end user who inputs information about power harassment into the system.

[1591] A "terminal" is a device used by a user to input information and receive analysis results and countermeasures.

[1592] A "server" is a computer system that receives data from a user and performs analysis and necessary processing.

[1593] "Database" means a data storage system for storing user-entered information and system-generated analysis results.

[1594] A "generative AI model" is an artificial intelligence algorithm that analyzes acts of power harassment and scores the possibility of power harassment based on past cases and court precedents.

[1595] The "emotion engine" is a system component that recognizes emotions from the user's facial expressions and voice, and generates emotion data based on that.

[1596] "Video simulation" is a visual simulation that recreates realistic scenes based on information and emotional data entered by the user.

[1597] "Escalation" is the process of notifying the relevant department of an emergency response to extremely serious cases of power harassment.

[1598] "Emotion data" is information about emotions recognized from the user's facial expressions and voice.

[1599] The "judgment result" is the evaluation score regarding the possibility of power harassment derived by the generative AI model and the countermeasures based on that score.

[1600] "Countermeasures" are action plans or recommendations that users should take if it is determined that there is a possibility of power harassment.

[1601] This invention relates to a system that combines detailed data related to power harassment entered by users with an emotion engine that recognizes the emotions at the time, and uses a generative AI model to determine power harassment and provide countermeasures. This system is configured to include a user terminal, a server, a database, a generative AI model, an emotion engine, and an image generation function.

[1602] User Input Processing

[1603] The user uses a device to input detailed information about the power harassment. Specifically, the user inputs details such as the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "reprimand"), the remarks (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history (e.g., "3 years"), and past experiences (e.g., "I've received similar reprimands three times in the past"). This input information is acquired along with the user's facial expressions and voice data, which are analyzed by the emotion engine. The device converts this data into an appropriate format and sends it to the server.

[1604] Data transmission and emotion recognition by emotion engine

[1605] The server receives the data sent from the device and records it in a database, including the user's emotional information recognized by the emotion engine. The server then runs a generative AI model based on the recorded data and emotional information. The generative AI model has learned from past court decisions and other cases, and evaluates the input data and emotional information to score the likelihood of power harassment.

[1606] Providing judgment results and countermeasures

[1607] The server generates a judgment result indicating the possibility of power harassment based on the results of the generative AI model. It also takes emotional information into consideration and creates countermeasures for the user to take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department"), and sends the judgment result and countermeasures to the user's device. The device displays the judgment result and countermeasures received from the server to the user, allowing the user to comprehensively understand the possibility of power harassment and the actions they should take.

[1608] Generate video simulations

[1609] The server generates a video simulation based on the user's input data and emotional information. This simulation video is highly realistic and designed to reproduce actual situations. The generated video is stored on the server and sent to the user's device. The device displays the video simulation sent from the server, allowing the user to visually confirm the situation.

[1610] Immediate Escalation

[1611] If the result is deemed to be extremely malicious, the server will immediately escalate the issue to the relevant department. The escalation notification will include a detailed report, and the relevant department will take immediate action.

[1612] Specific examples

[1613] As a specific example, the following case can be considered.

[1614] (Example 1: Victim input)

[1615] 1. User Input

[1616] The user enters, "My boss suddenly raised his voice during a meeting and scolded me, saying, 'Don't you understand that?'" The intonation of the remarks is "very strong," and the user also includes the detail that they felt pressured.

[1617] Select "Superior" as the relationship

[1618] Select "3 years" as work experience

[1619] Select "Meeting Room" as the location

[1620] Recognizes "fear" from the user's facial expression when inputting.

[1621] 2. Data Transmission

[1622] The device formats the data and sends it to the server, including the emotional information (fear).

[1623] 3. AI Judgment and Result Display

[1624] The server receives the data and uses a generative AI model to score the likelihood of power harassment (e.g., 80%).

[1625] The server also takes emotional information into consideration, generates a judgment result and countermeasures, and sends them to the terminal.

[1626] The terminal displays the judgment result and countermeasures to the user.

[1627] 4. Video Simulation Generation

[1628] The server generates a video simulation and transmits it to the terminal.

[1629] The terminal displays the simulation video in a viewable form for the user.

[1630] 5. Escalation

[1631] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[1632] Prompt Sentence Examples

[1633] Prompt example 1: Input of power harassment cases

[1634] "The boss raised his voice and scolded the employee during a meeting. It happened in a conference room during a project progress report. What was said: "Don't you understand that?", intonation: "Very strong tone", relationship: "Boss and subordinate", work history: "3 years", past experiences: "I've been scolded in a similar way three times in the past." The user felt fear while typing.

[1635] Based on this prompt, the system determines whether there is suspicion of power harassment and, if necessary, generates a video simulation or immediately escalates the situation.

[1636] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1637] Step 1: User Input Processing

[1638] The user inputs detailed information about the power harassment. This information includes the incident, trigger, location, behavior, remarks, intonation of remarks, relationships, work history, and past details. The device collects the data entered by the user and converts this information into digital format. The device also recognizes emotions from the user's facial expressions and voice data and collects emotional data.

[1639] Input: User-entered details, facial expressions, and voice data.

[1640] Output: Collected details and sentiment data.

[1641] Step 2: Send data

[1642] The device converts the collected details and emotion data into an appropriate format and transmits it to a server using a secure communication protocol to ensure data integrity.

[1643] Input: Details and sentiment data converted into digital form.

[1644] Output: The formatted data that is sent to the server.

[1645] Step 3: Database Recording

[1646] The server receives the data sent from the device and records it in a database, including the details and emotion data entered by the user.

[1647] Input: Formatted and submitted user details and sentiment data.

[1648] Output: Information recorded in the database.

[1649] Step 4: Run the generative AI model

[1650] The server runs a generative AI model based on the recorded data. The generative AI model has learned from past court decisions and similar cases, and analyzes the input data and emotional information to score the likelihood of power harassment.

[1651] Input: Details and emotion data recorded in the database.

[1652] Output: Scoring results indicating the possibility of power harassment.

[1653] Step 5: Generate judgment results

[1654] The server generates a judgment result indicating the possibility of power harassment based on the results of the generative AI model, and also considers emotional information to formulate countermeasures that the user should take.

[1655] Input: Scoring results and sentiment data.

[1656] Output: Verification result and countermeasures.

[1657] Step 6: Provide results and solutions

[1658] The server transmits the generated judgment result and countermeasure to the user's terminal.

[1659] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to understand the possibility of power harassment and the actions they should take.

[1660] Input: Verification result and countermeasures.

[1661] Output: Verification results and countermeasures displayed on the user's terminal.

[1662] Step 7: Generate video simulation

[1663] The server generates a video simulation using a generative AI model based on the user's input data and emotional information. This video is highly realistic and reproduces actual scenes. The server stores the generated video on its server and transmits it to the user's device.

[1664] Input: User details and emotion data.

[1665] Output: Video simulation.

[1666] Step 8: Image simulation display

[1667] The terminal receives the video simulation sent from the server and displays it in a form that the user can visually confirm, thereby enabling the user to visually understand the situation.

[1668] Input: Transmitted video simulation.

[1669] Output: The video simulation that is displayed to the user.

[1670] Step 9: Immediate Escalation

[1671] If the result is deemed to be extremely malicious, the server will immediately escalate the issue to the relevant department. The escalation notification will include a detailed report and require urgent action.

[1672] Input: Verdict and detailed report.

[1673] Output: Escalation notification sent to the responsible department.

[1674] (Application example 2)

[1675] 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."

[1676] Conventional power harassment detection systems require users to manually input detailed information, which makes it difficult to recognize emotions and make accurate judgments. Power harassment in factories is particularly difficult to detect, often negatively impacting work efficiency and the work environment. This has created a need for early problem detection and countermeasures.

[1677] 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.

[1678] In this invention, the server includes: means for acquiring multiple data sets including incidents and comments entered by the user; means for transmitting the acquired data to the server; means for receiving the transmitted data and recording it in a database; means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment; means for generating the determination results and corresponding countermeasures and preparing display data; means for transmitting the generated results and countermeasures to the user's terminal; means for generating a video simulation and transmitting highly realistic video to the user's terminal; means for immediately escalating the case to the relevant department if the determination results are deemed malicious; and means for a robot in a factory to monitor interactions between workers using surveillance cameras and voice recognition functions and detect possible acts of power harassment using an emotion engine and a generative AI model. This enables early detection and rapid response to power harassment in the work environment.

[1679] A "user terminal" is a device that a user operates to input information and check results.

[1680] A "server" is a computer system that receives, transmits, and processes data.

[1681] A "database" is a system that records acquired data and allows it to be searched and updated as needed.

[1682] A "generative AI model" is an artificial intelligence model that analyzes data and makes evaluations and predictions based on past cases and patterns.

[1683] The "emotion engine" is a system for recognizing emotions from user input data and audio / visual data.

[1684] A "surveillance camera" is a device that captures and records video in real time.

[1685] The "voice recognition function" is a function that converts voice into text and analyzes it.

[1686] "Power harassment" is the abuse of power in the workplace to cause mental or physical pain to others.

[1687] "Video simulation" is a technology that generates highly realistic images that reproduce real-world situations.

[1688] "Escalation" is the process of quickly notifying higher management or responsible departments when a problem occurs.

[1689] This invention is a system for detecting power harassment problems in factories and providing countermeasures, and is constructed by combining the following hardware and software.

[1690] Hardware configuration:

[1691] Surveillance camera (e.g., Logitech C920): Captures interactions between workers in real time and collects them as video data.

[1692] Microphone (e.g. RODE NT-USB): Collects audio while working and saves it as audio data.

[1693] Factory robots (e.g., Universal Robots UR series): Perform tasks within the factory and centrally process data from surveillance cameras and microphones.

[1694] Software configuration:

[1695] Speech recognition library (e.g., Google Cloud Speech-to-Text API): Converts collected voice data into text and analyzes it.

[1696] Emotion recognition engine (e.g., Affectiva SDK): Recognizes the emotions of workers from collected video and audio data and stores them as data.

[1697] Generative AI model (e.g., OpenAI GPT-4): Analyzes collected data based on past cases and precedents to assess the possibility of power harassment.

[1698] System operation description:

[1699] 1. Data Collection:

[1700] Surveillance cameras and microphones collect real-time images of work taking place inside the factory and store them as surveillance data.

[1701] The robot centrally manages this data and periodically sends it to a server.

[1702] 2. Emotion recognition:

[1703] The server sends the transmitted data to an emotion recognition engine to detect changes in the worker's emotions (e.g., "anger" or "fear").

[1704] The detected emotion data is input into a generative AI model.

[1705] 3. Power harassment judgment:

[1706] The generative AI model compares the input data with past cases and precedents to score the likelihood of power harassment.

[1707] Along with the judgment results, we will suggest countermeasures that the user should take.

[1708] 4. Providing solutions and escalation:

[1709] The results of the assessment and countermeasures are communicated to the worker via the robot's display and voice.

[1710] In serious cases, the server will automatically escalate the situation to the management department and prompt an emergency response.

[1711] Examples:

[1712] 1. Example: In a factory, a scene occurs in which worker A yells at worker B, saying, "Can't you even do something so simple?" A robot monitors this and automatically analyzes whether it is power harassment.

[1713] The surveillance camera captures the footage and recognizes Worker A's emotion as "anger."

[1714] The voice recognition library converts the voice data into text, and the emotion engine recognizes the emotion of Worker B, who felt "fear."

[1715] The generative AI model evaluates this and scores it as having an 80% chance of being power harassment.

[1716] Suggestion of countermeasures: The robot suggests "report to the HR department" and conveys this to Worker B. At the same time, the issue is escalated to the management department.

[1717] Example prompt sentence:

[1718] "A worker was scolded by his boss in a very harsh tone, saying, 'Can't you even do something as simple as this?' Please tell me if this situation is likely to be power harassment and what measures should be taken."

[1719] In this way, power harassment in the work environment can be detected early and dealt with promptly.

[1720] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1721] Step 1:

[1722] Data collection

[1723] The robot collects data entered by the user, including incidents and statements, as well as real-time video and audio data from surveillance cameras and microphones. The data entered by the user includes the incident, trigger, location, behavior, statements, intonation, the relationship between the perpetrator and victim, work history and duration, and past content. The collected data is converted into an appropriate format and sent to the server.

[1724] Input: User input data, surveillance camera video data, microphone audio data

[1725] Output: Sending collected data to a server

[1726] Step 2:

[1727] emotion recognition

[1728] The server receives the transmitted data and sends it to an emotion recognition engine to detect changes in the worker's emotions. The emotion recognition engine identifies emotions from the video and audio data and generates emotion data such as "anger" or "fear." This adds detailed information about the emotional changes.

[1729] Input: Collected video and audio data

[1730] Output: Emotion data generated by the emotion recognition engine

[1731] Step 3:

[1732] Power harassment judgment

[1733] The server inputs emotion data from the emotion recognition engine and detailed data entered by the user into a generative AI model. The generative AI model compares the data with past cases and legal precedents to determine the likelihood of power harassment. This score indicates the likelihood of power harassment and is used by users and the system to determine how seriously it should be taken.

[1734] Input: User details, emotion data

[1735] Output: Possibility score of power harassment

[1736] Step 4:

[1737] Presentation of judgment results and countermeasures

[1738] The server generates countermeasures based on the power harassment likelihood score obtained by the generative AI model. The results and countermeasures are formatted as display data and sent to the user's device. The countermeasures include specific actions such as "consulting with a higher-level manager" or "reporting to the human resources department."

[1739] Input: Possibility score of power harassment

[1740] Output: Sending countermeasures and judgment results to the user's device

[1741] Step 5:

[1742] Generate video simulations

[1743] The server runs a model that generates a video simulation based on the user's input data and emotional information. The model reproduces the real-world scene with high realism and transmits it to the user's device. The user can visually check the generated simulation video to gain a deeper understanding of the situation.

[1744] Input: User input data, emotional information

[1745] Output: Sending the video simulation to the user's device

[1746] Step 6:

[1747] escalation

[1748] If the server judges the result to be extremely malicious, it will immediately escalate the issue. When escalation occurs, a notification is sent to the responsible department requesting an emergency response, along with the judgment result and a detailed report. This allows the management department to take action quickly to resolve the problem.

[1749] Input: Verification results and detailed report

[1750] Output: Escalation notification to the responsible department

[1751] 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.

[1752] 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.

[1753] 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.

[1754] [Fourth embodiment]

[1755] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1756] 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.

[1757] 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).

[1758] 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.

[1759] 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.

[1760] 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).

[1761] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1762] 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.

[1763] 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.

[1764] 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.

[1765] 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.

[1766] 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.

[1767] 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."

[1768] The present invention relates to a system that acquires detailed data related to power harassment input by users, and uses a generative AI model to determine power harassment and provide countermeasures. This system is composed of a user terminal, a server, a database, a generative AI model, and a video generation function.

[1769] System Program

[1770] User Input Processing

[1771] User

[1772] The user uses a terminal to input detailed information about the suspected power harassment behavior. Specifically, the user inputs the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "When reporting on the progress of the project"), the location (e.g., "Conference room"), the behavior (e.g., "Scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "Very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "3 years"), and past experiences (e.g., "I have received similar scolding three times in the past")).

[1773] Terminal

[1774] The terminal receives data input by the user, converts it into an appropriate format, and transmits it to the server.

[1775] Data transmission and AI judgment

[1776] server

[1777] The server receives the data sent from the terminal and records it in a database.

[1778] A generative AI model is run based on the recorded data, and the possibility of power harassment is judged by comparing it with past cases. The AI ​​model has learned from past court decisions and cases from other companies and the company, and evaluates the input data to generate a score for the possibility of power harassment.

[1779] Providing judgment results and countermeasures

[1780] server

[1781] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model and generates a judgment result. It also creates countermeasures that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the human resources department").

[1782] The judgment result and countermeasures are sent to the user's terminal.

[1783] Terminal

[1784] The terminal displays the judgment result and countermeasures received from the server to the user.

[1785] Generate video simulations

[1786] server

[1787] The server runs a model to generate a video simulation based on user input data. The generated simulation video is highly realistic, allowing users to visually confirm the scene.

[1788] The generated video is stored on the server and sent to the user's terminal.

[1789] Terminal

[1790] The terminal displays the video simulation sent from the server in a viewable form for the user.

[1791] Immediate Escalation

[1792] server

[1793] If the result is deemed to be extremely egregious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[1794] Specific examples

[1795] Example 1: Victim input

[1796] 1. User Input

[1797] The user input a scene in which their boss suddenly raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[1798] Relationship: "Boss"

[1799] Work experience: 3 years

[1800] Location: "Conference Room"

[1801] 2. Data Transmission

[1802] The terminal formats the data and sends it to the server.

[1803] 3. AI Judgment and Result Display

[1804] The server receives the data and uses an AI model to score the likelihood of power harassment (e.g., 80%).

[1805] The server generates the judgment result and countermeasures and sends them to the terminal.

[1806] The terminal displays the judgment result and countermeasures to the user.

[1807] 4. Video Simulation Generation

[1808] The server generates a video simulation and transmits it to the terminal.

[1809] The terminal displays the simulation video in a viewable form for the user.

[1810] 5. Escalation

[1811] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[1812] This system contributes to preventing power harassment and improving the work environment by fairly and objectively assessing the possibility of power harassment, providing quick countermeasures, and using highly educational video simulations.

[1813] The processing flow will be explained below.

[1814] Step 1:

[1815] The user enters detailed information about suspected power harassment on the device, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[1816] Step 2:

[1817] The terminal takes the data entered by the user, converts it into the appropriate format, and sends it to the server, ensuring that the data is processed in a consistent format.

[1818] Step 3:

[1819] The server receives the data sent from the device, records it in a database, and stores it for further processing.

[1820] Step 4:

[1821] The server runs a generative AI model based on the recorded data. The generative AI model has learned from past court decisions and cases from other companies and the company, and evaluates the input data to generate a score for the possibility of power harassment.

[1822] Step 5:

[1823] Based on the results of the AI ​​model, the server generates a judgment result indicating the possibility of power harassment, and then creates countermeasures for the user to take (e.g., "consult with a higher-level manager" or "file a complaint with the human resources department") based on the judgment result.

[1824] Step 6:

[1825] The server sends the generated results and countermeasures data to the user's device, allowing the user to quickly check the results and countermeasures.

[1826] Step 7:

[1827] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to understand the possibility of power harassment and the actions they should take.

[1828] Step 8:

[1829] The server runs a model to generate a video simulation based on user input data. The generated simulation video is highly realistic and reproduces the actual scene.

[1830] Step 9:

[1831] The server stores the generated video simulation on the server and transmits a link or file to the user's terminal.

[1832] Step 10:

[1833] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user, thereby enabling the user to visually understand the situation.

[1834] Step 11:

[1835] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[1836] Through the specific processing flow described above, this system responds to power harassment issues fairly and efficiently, and provides prompt countermeasures. In addition, by using highly effective video simulations, it also contributes to preventing power harassment.

[1837] Example 1

[1838] 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."

[1839] Power harassment (pawahara) has become a serious social problem in modern workplaces, and there is a growing demand for systems that can fairly and quickly evaluate and provide appropriate countermeasures. However, current technology lacks the accuracy to make judgments based on detailed incident data or a method to provide visual evidence. This can lead to subjective judgments of power harassment and insufficient countermeasures. There is also a need for a system that can quickly escalate egregious cases of power harassment.

[1840] 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.

[1841] In this invention, the server includes: means for acquiring multiple data sets, including incidents and comments entered by users; means for transmitting the acquired data to the server; means for receiving the transmitted data and recording it in a database; means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment; means for generating a determination result and corresponding countermeasures and preparing display data; means for transmitting the generated results and countermeasures to the user's terminal; means for generating a video simulation based on detailed input data and transmitting highly realistic video to the user's terminal; and means for immediately escalating the generated result to the responsible department if the generated determination result is deemed to be extremely severe. This enables objective and rapid evaluation of cases of power harassment and provides appropriate countermeasures. Furthermore, providing visual evidence can deepen users' understanding and contribute to improving the work environment. Furthermore, severe cases can be quickly escalated to promote appropriate responses.

[1842] A "user" is a person who uses a terminal to input detailed information about a suspected case of power harassment.

[1843] "Occurrence of an incident" refers to a specific act or occurrence of power harassment.

[1844] "Content of remarks" refers to the specific words and actions that were spoken or taken during the power harassment situation.

[1845] "Data" refers to information entered by the user regarding the event, the content of the statement, the trigger, the location, the action, the intonation of the statement, the relationship, the work history and period, and past content.

[1846] "Server" means a computer system that receives, processes, and stores data sent from a user terminal.

[1847] "Database" means a system for recording and storing data received by the server.

[1848] A "generative AI model" is an artificial intelligence algorithm that assesses the possibility of power harassment based on past cases and court precedents.

[1849] "Assessment" is the process by which the generative AI model scores the possibility of power harassment based on the data.

[1850] "Countermeasures" are specific instructions that suggest actions or measures that the user should take based on the judgment results.

[1851] "Video simulation" refers to highly realistic videos and animations that are generated based on data entered by the user.

[1852] "Escalation" is the process of contacting the department in charge of emergency response when a serious case of power harassment is determined.

[1853] The present invention relates to a system that acquires detailed data related to power harassment input by users, and uses a generative AI model to determine power harassment and provide countermeasures. This system is composed of a user terminal, a server, a database, a generative AI model, and a video generation function.

[1854] User Input Processing

[1855] User

[1856] The user uses a terminal to input detailed information about suspected power harassment. For example, the incident ("My boss raised his voice and scolded me during a meeting"), the trigger ("When reporting on the progress of a project"), the location ("Conference room"), the nature of the behavior ("Scolding"), what was said ("Don't you understand that?"), the intonation of the remarks ("Very harsh tone"), the relationship ("Boss and subordinate"), work history or length of time spent with the same organization ("3 years"), and past experiences ("I have received similar scolding three times in the past").

[1857] Terminal

[1858] The device takes the data entered by the user, converts it into the appropriate format, and sends it to the server, either using a standard web browser or a mobile application.

[1859] Data transmission and AI judgment

[1860] server

[1861] The server receives the data sent from the device and records it in a database, which uses SQL-based or NoSQL-based technologies (e.g., MySQL, MongoDB).

[1862] The server runs a generative AI model based on the recorded data, which learns from past court decisions and similar cases, evaluates the input data, and assigns a score to indicate the likelihood of power harassment.

[1863] Providing judgment results and countermeasures

[1864] server

[1865] The server determines the possibility of power harassment based on the results of the AI ​​model and generates specific countermeasures, such as advice to "consult with a higher-ranking manager" or "file a complaint with the human resources department."

[1866] The generated results and countermeasures are sent to the user's terminal.

[1867] Terminal

[1868] The terminal displays the judgment result and countermeasures received from the server to the user using a general message box or notification function.

[1869] Generate video simulations

[1870] server

[1871] The server runs a model to generate a visual simulation based on user input data, using widely used 3D modeling tools (e.g., Unity, Unreal Engine).

[1872] The generated simulation images are highly realistic, allowing users to visually confirm the scene.

[1873] The generated video is stored on the server and sent to the user's terminal.

[1874] Terminal

[1875] The device displays the video simulation sent from the server in a viewable format for the user, using a video player or a dedicated viewer within the application.

[1876] Immediate Escalation

[1877] server

[1878] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the relevant department. A detailed report will be sent along with a notification urging immediate action. Notifications will be sent via email or a dedicated notification system (e.g., Slack, Microsoft Teams).

[1879] Specific examples

[1880] Here is an example prompt:

[1881] Please rate and score whether this case constitutes power harassment based on the following details:

[1882] Incident: The boss raised his voice and scolded the employee during a meeting.

[1883] Trigger: Project progress report

[1884] Location: Conference room

[1885] Action: Scolding

[1886] What he said: Don't you know that?

[1887] Speech intonation: Very strong tone

[1888] Relationship: Boss and subordinate

[1889] Work experience: 3 years

[1890] Past experiences: I have received similar reprimands three times in the past.

[1891] This system not only fairly and objectively assesses the possibility of power harassment and provides prompt countermeasures, but also contributes to preventing power harassment and improving the work environment by using highly educational video simulations.

[1892] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1893] Processing flow

[1894] Step 1: Getting User Input

[1895] Step 2: Send data

[1896] Step 3: Record the data

[1897] Step 4: Generative AI model decision

[1898] Step 5: Generate results and countermeasures

[1899] Step 6: Displaying the results and countermeasures

[1900] Step 7: Generate the video simulation

[1901] Step 8: Viewing the video simulation

[1902] Step 9: Immediate Escalation

[1903] Processing step details

[1904] Step 1: Getting User Input

[1905] User

[1906] The user uses the terminal to input detailed information about the behavior suspected of being power harassment.

[1907] Input: Occurrence, trigger, location, action, statement, intonation, relationship, work history, past content

[1908] Output: Detailed data collected from the user

[1909] What happens: A user enters data into an input field on a web form or mobile application and presses the submit button.

[1910] Step 2: Send data

[1911] Terminal

[1912] The terminal converts the data entered by the user into an appropriate format (such as JSON or XML) and sends it to the server.

[1913] Input: Detailed data entered by the user

[1914] Output: The data format sent to the server

[1915] What it does: Creates an HTTP POST request and sends data to the specified endpoint on the server.

[1916] Step 3: Record the data

[1917] server

[1918] The server receives the data sent from the terminal and records it in a database.

[1919] Input: Formatted data sent from the terminal

[1920] Output: Data recorded in a database

[1921] What it does: Parses the received data and inserts it appropriately into a SQL or NoSQL database.

[1922] Step 4: Generative AI model decision

[1923] server

[1924] The server runs a generative AI model based on the data recorded in the database and compares it with past cases to determine the possibility of power harassment.

[1925] Input: User data recorded in the database

[1926] Output: Power harassment likelihood score

[1927] Specific operation: Convert the recorded data into prompt sentences, input them into the generative AI model, and obtain the scoring results.

[1928] Example prompt:

[1929] Please rate and score whether this case constitutes power harassment based on the following details:

[1930] Incident: The boss raised his voice and scolded the employee during a meeting.

[1931] Trigger: Project progress report

[1932] Location: Conference room

[1933] Action: Scolding

[1934] What he said: Don't you know that?

[1935] Speech intonation: Very strong tone

[1936] Relationship: Boss and subordinate

[1937] Work experience: 3 years

[1938] Past experiences: I have received similar reprimands three times in the past.

[1939] Step 5: Generate results and countermeasures

[1940] server

[1941] Based on the results of the AI ​​model's assessment, the server generates specific countermeasures that the user should take.

[1942] Input: Judgment score

[1943] Output: Verification result and countermeasures

[1944] Specific actions: Analyze the scoring results and generate countermeasures based on the results (e.g., "consult with upper management" or "file a complaint with the HR department").

[1945] Step 6: Displaying the results and countermeasures

[1946] Terminal

[1947] The terminal displays the judgment result and countermeasures received from the server to the user.

[1948] Input: Verification result and countermeasures sent from the server

[1949] Output: Verification results and countermeasures displayed to the user

[1950] Specific behavior: The judgment result and countermeasures will be displayed as a pop-up message or notification.

[1951] Step 7: Generate the video simulation

[1952] server

[1953] The server executes a model that generates a visual simulation based on user input data.

[1954] Input: User details

[1955] Output: Video simulation data

[1956] What it does: Uses 3D modeling tools (e.g. Unity, Unreal Engine) to generate highly realistic images based on input data.

[1957] Step 8: Viewing the video simulation

[1958] Terminal

[1959] The terminal displays the video simulation sent from the server to the user.

[1960] Input: Video simulation data

[1961] Output: Simulation video displayed to the user

[1962] Specific operation: Play the video using a video player or a dedicated viewer within the application.

[1963] Step 9: Immediate Escalation

[1964] server

[1965] If the result of the assessment is deemed to be extremely malicious, the server will immediately escalate the matter to the relevant department.

[1966] Input: Verdict (extremely malicious)

[1967] Output: Escalation notification to the responsible department

[1968] Specific actions: Send a detailed report and a notification to the relevant department to prompt immediate action via email or a dedicated notification system (e.g., Slack, Microsoft Teams).

[1969] (Application example 1)

[1970] 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."

[1971] Power harassment (pa-hara) has become a serious problem in modern workplaces. This problem damages the mental and physical health of employees, reduces work efficiency, and lowers workplace morale. However, the determination of power harassment is subjective and requires specific criteria and evidence, making it difficult to address. There is also the problem of a lack of information provided to victims to take appropriate action. There is a need for a system that can solve these issues and enable swift and objective responses.

[1972] 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.

[1973] In this invention, the server includes means for scoring cases using a generative AI model based on information entered by the user and proposing appropriate countermeasures, means for conducting a detailed analysis of the incident with reference to past court cases and related cases, and means for providing the user with the results of the assessment and countermeasures. This makes it possible to quickly and objectively assess the possibility of power harassment and provide appropriate countermeasures to the victim.

[1974] "The incidents and comments entered by the user" refers to detailed information about specific incidents and comments related to power harassment that the user encountered in the workplace.

[1975] "Multiple pieces of information" refer to multiple data points or attributes related to power harassment, including the events that occurred and the content of statements.

[1976] "Means of acquisition" refers to the methods and functions for collecting information entered by the user and converting it into an appropriate format.

[1977] "Means for sending to the server side" refers to the communication means or protocol for sending the acquired information to the server via the network.

[1978] "Means for receiving transmitted information and recording it on a recording medium" refers to a method or function for receiving transmitted information and storing it in a database or storage device.

[1979] "Means for running a generative AI model" refers to the method or function of using a trained artificial intelligence model to assess potential power harassment in light of past cases.

[1980] "Means for generating judgment results and corresponding countermeasures, and preparing information for display" refers to the method or function for creating appropriate countermeasures based on the judgment results obtained by the AI ​​model and converting them into the format required for display to the user.

[1981] "Means for sending the generated results and countermeasures to the user's terminal" refers to the communication means or protocol for appropriately delivering the generated results and countermeasures to the user's terminal.

[1982] "Means for generating video simulations and transmitting highly realistic video to user terminals" refers to methods and functions for creating video that realistically reproduces actual power harassment scenes based on information provided by users and transmitting the video to user terminals.

[1983] "Means for immediate escalation to the relevant department if the assessment result is deemed to be serious" refers to a method or function for immediately sending a notification to the appropriate organization or department in charge urging them to take emergency action if the case of power harassment is deemed to be serious.

[1984] "Means for scoring cases using a generative AI model based on information entered by the user and proposing appropriate countermeasures" refers to methods or functions for using an AI model to assess the risk of power harassment based on information provided by the user and proposing appropriate countermeasures.

[1985] "Means for conducting detailed analysis of events by referring to past court cases and related cases" refers to methods and functions for conducting detailed analysis of current cases based on related court cases and cases that occurred in the past.

[1986] This invention is a system in which a user inputs detailed information related to power harassment in the workplace, and a generative AI model determines the possibility of power harassment based on that information and proposes appropriate countermeasures. This system is configured to include a user terminal, a server, a database, a generative AI model, and an image generation function.

[1987] System Program

[1988] User Input Processing

[1989] User

[1990] Using a smartphone app, users input detailed information about suspected power harassment, including the incident, trigger, location, details of the behavior, remarks, intonation of remarks, relationships, work history, duration, and similar past cases.

[1991] Smartphone app

[1992] The smartphone app receives information entered by the user, converts it into an appropriate format, and sends it to the server. The application is developed using Swift (iOS) and Kotlin (Android).

[1993] Data transmission and AI judgment

[1994] server

[1995] The server receives the information sent from the smartphone app and records it in a database using PostgreSQL.

[1996] Based on the recorded information, a generative AI model (using TensorFlow) is run to compare it with past court cases and other cases to determine the possibility of power harassment. The generative AI model evaluates the input information and assigns a score to the possibility of power harassment.

[1997] Providing judgment results and countermeasures

[1998] server

[1999] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model, generates a judgment result, and also suggests countermeasures that the user should take.

[2000] The assessment results and countermeasures are sent to a smartphone app.

[2001] Smartphone app

[2002] The smartphone app displays the judgment results and countermeasures received from the server to the user.

[2003] Generate video simulations

[2004] server

[2005] The server generates a video simulation based on the user's input data, using Unity for video generation.

[2006] The generated simulation video is stored on a server and sent to a smartphone app.

[2007] Smartphone app

[2008] The smartphone app can visually display the video simulation sent from the server to the user.

[2009] Immediate Escalation

[2010] server

[2011] If the result is deemed malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting immediate action.

[2012] Specific examples

[2013] Example 1: Victim input

[2014] 1. User Input

[2015] The user input a scene in which their boss raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[2016] Relationship: Boss and subordinate

[2017] Work experience: 3 years

[2018] Location: Conference room

[2019] Previous similar incidents: Received similar reprimands three times

[2020] 2. Data Transmission

[2021] The smartphone app formats the data and sends it to the server.

[2022] 3. AI Judgment and Result Display

[2023] The server receives the data and uses an AI model to score the likelihood of power harassment. Example: 80%

[2024] The system generates a judgment result (high possibility of power harassment) and countermeasures (report to supervisor) and sends them to a smartphone.

[2025] 4. Video Simulation Generation

[2026] The server generates a video simulation and sends it to the smartphone.

[2027] The smartphone displays the simulation video in a viewable format for the user.

[2028] 5. Escalation

[2029] The server determines the case is serious and sends a notification to the HR department requesting immediate action.

[2030] Prompt Sentence Examples

[2031] Enter the following prompt into the generative AI model:

[2032] A user has submitted the following case of power harassment:

[2033] Incident: Your boss raises his voice and scolds you during a meeting.

[2034] What he said: Don't you know that?

[2035] Intonation: Very strong

[2036] Relationship: Boss and subordinate

[2037] Work experience: 3 years

[2038] Location: Conference room

[2039] Previous cases: Received similar reprimands three times

[2040] Based on this case, please assess and score the possibility of power harassment. Also, please suggest appropriate countermeasures.

[2041] In this way, the system can quickly provide an AI decision and appropriate response based on user input.

[2042] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2043] Step 1: User Input Processing

[2044] Users use a smartphone app to input information about suspected power harassment, including the incident, trigger, location, behavior, remarks, intonation of remarks, relationships, work history, duration, and similar past cases. The input information is then converted into an appropriate format within the smartphone app.

[2045] Input: Detailed information related to power harassment provided by the user

[2046] Output: Formatted information

[2047] Step 2: Submit your information

[2048] The device (smartphone app) sends the formatted information entered by the user to the server. HTTPS is used as the communication protocol.

[2049] Input: Formatted information

[2050] Output: Data sent to the server

[2051] Step 3: Record in the database

[2052] The server receives the information sent from the terminal and records it in a database. PostgreSQL is used for the database. The received data is properly organized by information item and saved in the database.

[2053] Input: Information sent from the device

[2054] Output: Information recorded in the database

[2055] Step 4: AI model decision

[2056] The server runs a generative AI model (using TensorFlow) based on the recorded information, which compares it with past court cases and other relevant cases to determine whether the person is engaging in power harassment. This process involves data analysis using neural networks.

[2057] Input: Information recorded in the database

[2058] Output: Possibility score of power harassment

[2059] Step 5: Generate results and countermeasures

[2060] The server determines whether there is a high possibility of power harassment based on the results of the AI ​​model and generates a judgment result. It also suggests countermeasures that the user should take (e.g., reporting to a supervisor, contacting human resources, etc.). The series of countermeasures are generated based on a template.

[2061] Input: Possibility score of power harassment, judgment result template

[2062] Output: Judgment results and countermeasures

[2063] Step 6: Submitting results and action plans

[2064] The server then sends the generated judgment results and countermeasures to the smartphone app, again using HTTPS as the communication protocol.

[2065] Input: Judgment result and countermeasures

[2066] Output: Data sent to smartphone app

[2067] Step 7: Viewing results and solutions

[2068] The device (smartphone app) displays the assessment results and countermeasures received from the server to the user. To display the results on an interface that is easy for the user to see, alerts and notifications are used in particular.

[2069] Input: Judgment result and countermeasures

[2070] Output: Displayed on the user's smartphone screen

[2071] Step 8: Generate the video simulation

[2072] The server generates a video simulation based on the user's input data. Unity is used for video generation, and the generated video is saved on the server.

[2073] Input: User-entered data

[2074] Output: Generated video simulation

[2075] Step 9: Send the video simulation

[2076] The server then sends the generated video simulation to a smartphone app, which also uses HTTPS as the communication protocol.

[2077] Input: Generated video simulation

[2078] Output: Data sent to smartphone app

[2079] Step 10: Viewing the video simulation

[2080] The terminal (smartphone app) displays the video simulation sent from the server in a viewable format for the user. An interactive UI is used to improve the user experience.

[2081] Input: Video simulation data

[2082] Output: Playback on the user's smartphone screen

[2083] Step 11: Immediate Escalation

[2084] If the server determines that the result is malicious, it immediately escalates it to the appropriate department, a process that automatically sends a detailed report along with a notification requesting immediate action.

[2085] Input: Malicious judgment result

[2086] Output: Urgent notification to the relevant department

[2087] 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.

[2088] This invention relates to a system that uses a generative AI model to determine power harassment and provide countermeasures by combining detailed data related to power harassment entered by the user with an emotion engine that recognizes the emotions at the time. This system is composed of a user terminal, a server, a database, a generative AI model, an emotion engine, and a video generation function.

[2089] System Program

[2090] User Input Processing

[2091] User

[2092] Using a terminal, users input detailed information about suspected power harassment, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "three years"), and past experiences (e.g., "I have received similar scolding three times in the past").

[2093] Terminal

[2094] The device receives data entered by the user, converts it into an appropriate format, and sends it to the server. At the same time, it also recognizes emotions from the user's facial expressions and voice, and sends this information to the server.

[2095] Data transmission and emotion recognition by emotion engine

[2096] server

[2097] The server receives the data sent from the device and records it in a database, including the user's emotional information recognized by the emotion engine.

[2098] The server runs a generative AI model based on the recorded data and emotional information. The generative AI model learns from past court cases and cases from other companies and the company, and evaluates the input data and emotional information to score the possibility of power harassment.

[2099] Providing judgment results and countermeasures

[2100] server

[2101] Based on the results of the AI ​​model, the server generates a judgment result indicating the possibility of power harassment. It also takes into account emotional information and creates countermeasures that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department").

[2102] The judgment result and countermeasures are sent to the user's terminal.

[2103] Terminal

[2104] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to comprehensively understand the possibility of power harassment and the actions they should take, including emotional information.

[2105] Generate video simulations

[2106] server

[2107] The server executes a model to generate a video simulation based on the user's input data and emotional information. The generated simulation video is highly realistic and reproduces the actual scene.

[2108] The generated video is stored on the server and sent to the user's terminal.

[2109] Terminal

[2110] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user, thereby enabling the user to visually understand the situation.

[2111] Immediate Escalation

[2112] server

[2113] If the result is deemed to be extremely malicious, the server will immediately escalate the incident to the appropriate department, sending a detailed report and a notification requesting urgent action.

[2114] Specific examples

[2115] Example 1: Victim input

[2116] 1. User Input

[2117] The user input a scenario in which their boss suddenly raised his voice during a meeting and scolded them, saying, "Don't you understand that?" The intonation of the remark was very strong and the user felt pressured.

[2118] Relationship: "Boss"

[2119] Work experience: 3 years

[2120] Location: "Conference Room"

[2121] Recognizes "fear" from the user's facial expression when inputting.

[2122] 2. Data Transmission

[2123] The device formats the data and sends it to the server, including the emotional information (fear).

[2124] 3. AI Judgment and Result Display

[2125] The server receives the data and uses an AI model to score the likelihood of power harassment (e.g., 80%).

[2126] The system generates a judgment result and countermeasures, taking into account emotional information, and sends them to the device.

[2127] The terminal displays the judgment result and countermeasures to the user.

[2128] 4. Video Simulation Generation

[2129] The server generates a video simulation and transmits it to the terminal.

[2130] The terminal displays the simulation video in a viewable form for the user.

[2131] 5. Escalation

[2132] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[2133] This system fairly and objectively judges the possibility of power harassment and provides prompt countermeasures by taking into account the user's emotional information. It also contributes to preventing power harassment by using highly educational video simulations.

[2134] The processing flow will be explained below.

[2135] Step 1:

[2136] The user inputs detailed information about the suspected power harassment on the terminal, including the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "When reporting on the progress of the project"), the location (e.g., "Conference room"), the nature of the behavior (e.g., "Scolding"), what was said (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "Very harsh tone"), the relationship (e.g., "boss and subordinate"), work history or length of time spent with the same organization (e.g., "3 years"), and past experiences (e.g., "I have received similar scolding three times in the past")

[2137] Step 2:

[2138] When a user types, the device collects data such as the user's facial expression, voice, and typing speed in real time, and the emotion engine analyzes this data to recognize the user's emotional state (e.g., "fear," "anger," etc.).

[2139] Step 3:

[2140] The terminal acquires detailed information entered by the user and emotion data recognized by the emotion engine, and converts them into an appropriate format.

[2141] Step 4:

[2142] The terminal transmits the formatted detailed information and emotion data to the server.

[2143] Step 5:

[2144] The server receives the data sent from the device and records it in a database, where both detailed information and emotional data are stored.

[2145] Step 6:

[2146] The server runs a generative AI model based on the recorded data. The generative AI model learns from past court decisions and cases from other companies and the company, and evaluates the input data and emotional data to score the likelihood of power harassment.

[2147] Step 7:

[2148] The server generates a judgment result indicating the possibility of power harassment based on the results of the AI ​​model, and also takes into account the user's emotional information to create a response that the user should take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department").

[2149] Step 8:

[2150] The server transmits the generated judgment results, countermeasures, and emotion data to the user's terminal.

[2151] Step 9:

[2152] The terminal displays the judgment result, countermeasures, and emotion data received from the server to the user.

[2153] Step 10:

[2154] The server runs a model to generate a video simulation based on the user's input information and emotional data. The simulation video is highly realistic and reproduces specific scenes.

[2155] Step 11:

[2156] The server stores the generated video simulation and transmits it to the user's device.

[2157] Step 12:

[2158] The terminal receives the video simulation sent from the server and displays it in a form that can be visually confirmed by the user.

[2159] Step 13:

[2160] If the result is deemed to be extremely egregious, the server will immediately escalate the issue to the appropriate department, which will include a detailed report and a notification requesting immediate action.

[2161] This processing flow allows the system to combine user input information and emotional data to accurately determine the possibility of power harassment and provide specific countermeasures. Furthermore, through video simulations, the system can contribute to education and raising awareness of power harassment prevention.

[2162] Example 2

[2163] 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."

[2164] Conventional power harassment response systems are primarily based solely on information entered by the user, and lack objectivity and consideration of emotional information. As a result, the determination of power harassment is subjective, and the countermeasures offered are limited. Furthermore, there is no simulation function for visually understanding the actual situation, which can make it difficult for users to select appropriate countermeasures. Furthermore, in extremely severe cases of power harassment, there is a lack of a rapid escalation system. The purpose of this system is to solve these problems.

[2165] 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.

[2166] In this invention, the server includes a means for acquiring emotional data from the user's facial expressions and voice, a means for executing an emotion engine based on the emotional data to evaluate the impact of power harassment, and a means for generating a video simulation and transmitting the highly realistic video to the user's terminal. This enables a fair and objective assessment of power harassment that takes into account the user's emotional information, provides appropriate countermeasures, and enables a visual understanding of the situation. Furthermore, in extremely severe cases, rapid escalation is possible.

[2167] "User" means an individual or end user who inputs information about power harassment into the system.

[2168] A "terminal" is a device used by a user to input information and receive analysis results and countermeasures.

[2169] A "server" is a computer system that receives data from a user and performs analysis and necessary processing.

[2170] "Database" means a data storage system for storing user-entered information and system-generated analysis results.

[2171] A "generative AI model" is an artificial intelligence algorithm that analyzes acts of power harassment and scores the possibility of power harassment based on past cases and court precedents.

[2172] The "emotion engine" is a system component that recognizes emotions from the user's facial expressions and voice, and generates emotion data based on that.

[2173] "Video simulation" is a visual simulation that recreates realistic scenes based on information and emotional data entered by the user.

[2174] "Escalation" is the process of notifying the relevant department of an emergency response to extremely serious cases of power harassment.

[2175] "Emotion data" is information about emotions recognized from the user's facial expressions and voice.

[2176] The "judgment result" is the evaluation score regarding the possibility of power harassment derived by the generative AI model and the countermeasures based on that score.

[2177] "Countermeasures" are action plans or recommendations that users should take if it is determined that there is a possibility of power harassment.

[2178] This invention relates to a system that combines detailed data related to power harassment entered by users with an emotion engine that recognizes the emotions at the time, and uses a generative AI model to determine power harassment and provide countermeasures. This system is configured to include a user terminal, a server, a database, a generative AI model, an emotion engine, and an image generation function.

[2179] User Input Processing

[2180] The user uses a device to input detailed information about the power harassment. Specifically, the user inputs details such as the incident (e.g., "My boss raised his voice and scolded me during a meeting"), the trigger (e.g., "during a project progress report"), the location (e.g., "conference room"), the nature of the behavior (e.g., "reprimand"), the remarks (e.g., "Don't you understand that?"), the intonation of the remarks (e.g., "very harsh tone"), the relationship (e.g., "boss and subordinate"), work history (e.g., "3 years"), and past experiences (e.g., "I've received similar reprimands three times in the past"). This input information is acquired along with the user's facial expressions and voice data, which are analyzed by the emotion engine. The device converts this data into an appropriate format and sends it to the server.

[2181] Data transmission and emotion recognition by emotion engine

[2182] The server receives the data sent from the device and records it in a database, including the user's emotional information recognized by the emotion engine. The server then runs a generative AI model based on the recorded data and emotional information. The generative AI model has learned from past court decisions and other cases, and evaluates the input data and emotional information to score the likelihood of power harassment.

[2183] Providing judgment results and countermeasures

[2184] The server generates a judgment result indicating the possibility of power harassment based on the results of the generative AI model. It also takes emotional information into consideration and creates countermeasures for the user to take (e.g., "consult with a higher-level manager" or "file a complaint with the HR department"), and sends the judgment result and countermeasures to the user's device. The device displays the judgment result and countermeasures received from the server to the user, allowing the user to comprehensively understand the possibility of power harassment and the actions they should take.

[2185] Generate video simulations

[2186] The server generates a video simulation based on the user's input data and emotional information. This simulation video is highly realistic and designed to reproduce actual situations. The generated video is stored on the server and sent to the user's device. The device displays the video simulation sent from the server, allowing the user to visually confirm the situation.

[2187] Immediate Escalation

[2188] If the result is deemed to be extremely malicious, the server will immediately escalate the issue to the relevant department. The escalation notification will include a detailed report, and the relevant department will take immediate action.

[2189] Specific examples

[2190] As a specific example, the following case can be considered.

[2191] (Example 1: Victim input)

[2192] 1. User Input

[2193] The user enters, "My boss suddenly raised his voice during a meeting and scolded me, saying, 'Don't you understand that?'" The intonation of the remarks is "very strong," and the user also includes the detail that they felt pressured.

[2194] Select "Superior" as the relationship

[2195] Select "3 years" as work experience

[2196] Select "Meeting Room" as the location

[2197] Recognizes "fear" from the user's facial expression when inputting.

[2198] 2. Data Transmission

[2199] The device formats the data and sends it to the server, including the emotional information (fear).

[2200] 3. AI Judgment and Result Display

[2201] The server receives the data and uses a generative AI model to score the likelihood of power harassment (e.g., 80%).

[2202] The server also takes emotional information into consideration, generates a judgment result and countermeasures, and sends them to the terminal.

[2203] The terminal displays the judgment result and countermeasures to the user.

[2204] 4. Video Simulation Generation

[2205] The server generates a video simulation and transmits it to the terminal.

[2206] The terminal displays the simulation video in a viewable form for the user.

[2207] 5. Escalation

[2208] In extremely serious cases, the server will immediately escalate the matter to the relevant department and request a response.

[2209] Prompt Sentence Examples

[2210] Prompt example 1: Input of power harassment cases

[2211] "The boss raised his voice and scolded the employee during a meeting. It happened in a conference room during a project progress report. What was said: "Don't you understand that?", intonation: "Very strong tone", relationship: "Boss and subordinate", work history: "3 years", past experiences: "I've been scolded in a similar way three times in the past." The user felt fear while typing.

[2212] Based on this prompt, the system determines whether there is suspicion of power harassment and, if necessary, generates a video simulation or immediately escalates the situation.

[2213] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2214] Step 1: User Input Processing

[2215] The user inputs detailed information about the power harassment. This information includes the incident, trigger, location, behavior, remarks, intonation of remarks, relationships, work history, and past details. The device collects the data entered by the user and converts this information into digital format. The device also recognizes emotions from the user's facial expressions and voice data and collects emotional data.

[2216] Input: User-entered details, facial expressions, and voice data.

[2217] Output: Collected details and sentiment data.

[2218] Step 2: Send data

[2219] The device converts the collected details and emotion data into an appropriate format and transmits it to a server using a secure communication protocol to ensure data integrity.

[2220] Input: Details and sentiment data converted into digital form.

[2221] Output: The formatted data that is sent to the server.

[2222] Step 3: Database Recording

[2223] The server receives the data sent from the device and records it in a database, including the details and emotion data entered by the user.

[2224] Input: Formatted and submitted user details and sentiment data.

[2225] Output: Information recorded in the database.

[2226] Step 4: Run the generative AI model

[2227] The server runs a generative AI model based on the recorded data. The generative AI model has learned from past court decisions and similar cases, and analyzes the input data and emotional information to score the likelihood of power harassment.

[2228] Input: Details and emotion data recorded in the database.

[2229] Output: Scoring results indicating the possibility of power harassment.

[2230] Step 5: Generate judgment results

[2231] The server generates a judgment result indicating the possibility of power harassment based on the results of the generative AI model, and also considers emotional information to formulate countermeasures that the user should take.

[2232] Input: Scoring results and sentiment data.

[2233] Output: Verification result and countermeasures.

[2234] Step 6: Provide results and solutions

[2235] The server transmits the generated judgment result and countermeasure to the user's terminal.

[2236] The device receives the results of the assessment and the countermeasures from the server and displays them to the user, allowing the user to understand the possibility of power harassment and the actions they should take.

[2237] Input: Verification result and countermeasures.

[2238] Output: Verification results and countermeasures displayed on the user's terminal.

[2239] Step 7: Generate video simulation

[2240] The server generates a video simulation using a generative AI model based on the user's input data and emotional information. This video is highly realistic and reproduces actual scenes. The server stores the generated video on its server and transmits it to the user's device.

[2241] Input: User details and emotion data.

[2242] Output: Video simulation.

[2243] Step 8: Image simulation display

[2244] The terminal receives the video simulation sent from the server and displays it in a form that the user can visually confirm, thereby enabling the user to visually understand the situation.

[2245] Input: Transmitted video simulation.

[2246] Output: The video simulation that is displayed to the user.

[2247] Step 9: Immediate Escalation

[2248] If the result is deemed to be extremely malicious, the server will immediately escalate the issue to the relevant department. The escalation notification will include a detailed report and require urgent action.

[2249] Input: Verdict and detailed report.

[2250] Output: Escalation notification sent to the responsible department.

[2251] (Application example 2)

[2252] 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."

[2253] Conventional power harassment detection systems require users to manually input detailed information, which makes it difficult to recognize emotions and make accurate judgments. Power harassment in factories is particularly difficult to detect, often negatively impacting work efficiency and the work environment. This has created a need for early problem detection and countermeasures.

[2254] 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.

[2255] In this invention, the server includes: means for acquiring multiple data sets including incidents and comments entered by the user; means for transmitting the acquired data to the server; means for receiving the transmitted data and recording it in a database; means for executing a generative AI model based on the recorded data and comparing it with past cases to determine the possibility of power harassment; means for generating the determination results and corresponding countermeasures and preparing display data; means for transmitting the generated results and countermeasures to the user's terminal; means for generating a video simulation and transmitting highly realistic video to the user's terminal; means for immediately escalating the case to the relevant department if the determination results are deemed malicious; and means for a robot in a factory to monitor interactions between workers using surveillance cameras and voice recognition functions and detect possible acts of power harassment using an emotion engine and a generative AI model. This enables early detection and rapid response to power harassment in the work environment.

[2256] A "user terminal" is a device that a user operates to input information and check results.

[2257] A "server" is a computer system that receives, transmits, and processes data.

[2258] A "database" is a system that records acquired data and allows it to be searched and updated as needed.

[2259] A "generative AI model" is an artificial intelligence model that analyzes data and makes evaluations and predictions based on past cases and patterns.

[2260] The "emotion engine" is a system for recognizing emotions from user input data and audio / visual data.

[2261] A "surveillance camera" is a device that captures and records video in real time.

[2262] The "voice recognition function" is a function that converts voice into text and analyzes it.

[2263] "Power harassment" is the abuse of power in the workplace to cause mental or physical pain to others.

[2264] "Video simulation" is a technology that generates highly realistic images that reproduce real-world situations.

[2265] "Escalation" is the process of quickly notifying higher management or responsible departments when a problem occurs.

[2266] This invention is a system for detecting power harassment problems in factories and providing countermeasures, and is constructed by combining the following hardware and software.

[2267] Hardware configuration:

[2268] Surveillance camera (e.g., Logitech C920): Captures interactions between workers in real time and collects them as video data.

[2269] Microphone (e.g. RODE NT-USB): Collects audio while working and saves it as audio data.

[2270] Factory robots (e.g., Universal Robots UR series): Perform tasks within the factory and centrally process data from surveillance cameras and microphones.

[2271] Software configuration:

[2272] Speech recognition library (e.g., Google Cloud Speech-to-Text API): Converts collected voice data into text and analyzes it.

[2273] Emotion recognition engine (e.g., Affectiva SDK): Recognizes the emotions of workers from collected video and audio data and stores them as data.

[2274] Generative AI model (e.g., OpenAI GPT-4): Analyzes collected data based on past cases and precedents to assess the possibility of power harassment.

[2275] System operation description:

[2276] 1. Data Collection:

[2277] Surveillance cameras and microphones collect real-time images of work taking place inside the factory and store them as surveillance data.

[2278] The robot centrally manages this data and periodically sends it to a server.

[2279] 2. Emotion recognition:

[2280] The server sends the transmitted data to an emotion recognition engine to detect changes in the worker's emotions (e.g., "anger," "fear").

[2281] The detected emotion data is input into a generative AI model.

[2282] 3. Power harassment judgment:

[2283] The generative AI model compares the input data with past cases and precedents to score the likelihood of power harassment.

[2284] Along with the judgment results, we will suggest countermeasures that the user should take.

[2285] 4. Providing solutions and escalation:

[2286] The results of the assessment and countermeasures are communicated to the worker via the robot's display and voice.

[2287] In serious cases, the server will automatically escalate the situation to the management department and prompt an emergency response.

[2288] Examples:

[2289] 1. Example: In a factory, a scene occurs in which worker A yells at worker B, saying, "Can't you even do something so simple?" A robot monitors this and automatically analyzes whether it is power harassment.

[2290] The surveillance camera captur...

Claims

1. A means for acquiring a plurality of data including occurrence events and speech contents input by a user; A means for transmitting the acquired data to the server side; means for receiving the transmitted data and recording it in a database; A method to run a generative AI model based on recorded data and compare it with past cases to determine the possibility of power harassment, and means for generating a determination result and a corresponding countermeasure and preparing data for display; means for transmitting the generated results and countermeasures to a user's terminal; means for generating a video simulation and transmitting a highly realistic video to a user terminal; If the result of the assessment is deemed malicious, there is a means to immediately escalate the matter to the relevant department. A system including:

2. The system of claim 1 is provided with a means for acquiring data entered by the user regarding the incident, trigger, location, action, statement, intonation of statement, relationship between perpetrator and victim, work history and period, and past content.

3. 2. The system according to claim 1, further comprising means for transmitting the generated video simulation to a user terminal and displaying it in a form that can be viewed by the user.

Images