system

Abstract

We provide the system. [Solution] A means for initializing and controlling a generative artificial intelligence engine, A means of generating email creation support content, A means of generating support content regarding how to ask questions, Methods for detecting power harassment and moral harassment, A system that includes means for performing at least one of these means in response to a request from a user.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Newcomer training in the IT industry, supplementation of sales knowledge, and early detection of harassment are very important issues. However, in conventional systems, it has been difficult to provide support and knowledge according to the needs of individual users, and to detect power harassment and moral harassment. As a result, there has been a problem that users cannot receive appropriate support, and it has been difficult to ensure efficient learning and a safe workplace environment.

Means for Solving the Problems

[0005] The present invention solves the above problems with a system that includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, and means for detecting power harassment and moral harassment. Specifically, by executing at least one of these means in response to a user request, the system provides individually tailored training content and support content to each user, creating an appropriate learning environment and enabling early detection and response to harassment. This enhances users' IT literacy and realizes a safe and efficient workplace environment.

[0006] A "generative artificial intelligence engine" is an artificial intelligence system that has the ability to generate new information and content based on existing data.

[0007] "Initialization" is the process of performing the initial settings and preparations necessary for a system or program to function correctly.

[0008] "Means of control" refers to the operation or management methods used to ensure that a particular device or system functions properly.

[0009] "Email creation support content" refers to information such as advice, templates, and recommended expressions to help users create emails effectively.

[0010] "Support regarding how to ask questions" refers to guides and advice that enable users to ask questions or make inquiries to others in an effective and appropriate manner.

[0011] "Power harassment" refers to acts in the workplace where a superior or colleague exerts inappropriate pressure or engages in inappropriate words or actions towards another person, causing mental or physical distress.

[0012] "Moral harassment" refers to acts that inflict emotional distress by attacking another person's character or morals through words or actions.

[0013] "Means of detection" refers to methods or devices for identifying and recognizing specific events or states.

[0014] A "user request" refers to a request from a user to the system for a specific operation or information.

[0015] "Training content" refers to educational materials and lessons aimed at improving users' skills and acquiring knowledge.

[0016] "Support content" refers to information and advice designed to assist users when performing specific tasks or activities. [Brief explanation of the drawing]

[0017] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0018] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0021] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0022] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0023] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0025] [First Embodiment]

[0026] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0027] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0030] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0033] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0037] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0038] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment.

[0039] The server first initializes its generative artificial intelligence engine (AI engine) and sets up HR-related information. Upon receiving a request from a user, the server determines the type of request and executes the appropriate processing accordingly.

[0040] For example, if a user requests training materials for new employees, the server uses an AI engine to generate the most suitable training content for the user and provides it to them. Similarly, if a user requests assistance with email composition, the server uses an email composition assistance module to guide the user on how to compose effective emails.

[0041] If assistance is needed regarding how to ask questions, the server utilizes a question-asking support module to guide users in communicating effectively. Furthermore, if a user reports power harassment or moral harassment, the server evaluates it using a detection module and suggests appropriate countermeasures.

[0042] Specific example

[0043] Provision of training materials for new employees

[0044] When a user sends a request to the server for new employee training materials, the server uses an AI engine to generate training content optimized for that individual user and returns it to the user. The user then reviews the received training materials and acquires the necessary knowledge and skills.

[0045] Providing email creation support

[0046] If a user needs assistance composing an email, the server uses an email composition support module to generate and provide support content that includes specific phrasing and writing style guidance. The user can then use this content to create an effective email.

[0047] Reports of power harassment / moral harassment

[0048] When a user reports harassment (power harassment or moral harassment), the server uses a harassment detection module to evaluate the report based on more detailed information. The server analyzes the report and proposes countermeasures as needed. As a result, users can address problems early, leading to improvements in the workplace environment.

[0049] This system allows users to efficiently acquire necessary knowledge and create a safe and secure work environment. The entire system works together to support users, thereby improving IT literacy and ensuring workplace safety.

[0050] The following describes the processing flow.

[0051] Provision of training materials for new employees

[0052] Step 1:

[0053] A user issues a request for new employee training materials. Specifically, the user sends a request from their terminal to the server.

[0054] Step 2:

[0055] The server receives a request from the user. It analyzes the content of the request and confirms that it is a request of type "training".

[0056] Step 3:

[0057] The server invokes a generative artificial intelligence engine (AI engine) to generate training content based on user information.

[0058] Step 4:

[0059] The AI ​​engine generates the training content the user needs and returns it to the server.

[0060] Step 5:

[0061] The server sends the generated training content to the user's terminal.

[0062] Step 6:

[0063] Users receive training content via their devices and review its contents.

[0064] Providing email creation support

[0065] Step 1:

[0066] The user issues a request for email creation assistance. Specifically, the user sends a request from their terminal to the server.

[0067] Step 2:

[0068] The server receives a request from the user. It parses the request and confirms that it is of type "email_help".

[0069] Step 3:

[0070] The server calls the email creation support module and generates email creation support content based on user information.

[0071] Step 4:

[0072] The email creation support module generates specific email creation advice and templates requested by the user and returns them to the server.

[0073] Step 5:

[0074] The server sends the generated email composition support content to the user's terminal.

[0075] Step 6:

[0076] Users receive email composition support content via their devices, referring to specific expressions and writing styles.

[0077] Reports of power harassment / moral harassment

[0078] Step 1:

[0079] Users request to report instances of power harassment or moral harassment. Specifically, they send requests from their terminals to the server.

[0080] Step 2:

[0081] The server receives a request from the user. It parses the request and confirms that it is of type "check_abuse".

[0082] Step 3:

[0083] The server calls the moral harassment / power harassment detection module and analyzes the reported content.

[0084] Step 4:

[0085] The harassment / power harassment detection module evaluates the reported content and identifies the problem.

[0086] Step 5:

[0087] The server sends the evaluation results and necessary countermeasures to the user's terminal.

[0088] Step 6:

[0089] Users receive evaluation results via their devices and confirm necessary countermeasures.

[0090] In this way, the system operates by calling the appropriate module for each request, providing the user with the most suitable content, support, and problem-solving solutions.

[0091] (Example 1)

[0092] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0093] A system is needed to efficiently conduct new employee training, provide work support, and improve the work environment within companies. Traditional methods often lacked coherence and efficiency, as various content was generated and support provided individually. Furthermore, there is a need for faster early detection and response to power harassment and moral harassment. These challenges must be addressed in an integrated manner.

[0094] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0095] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, means for detecting power harassment and moral harassment, means for generating new employee training materials, and means for executing at least one of these means in response to a user request. This makes it possible to efficiently conduct new employee training and work support within a company, and to comprehensively support the improvement of the work environment.

[0096] A "generative artificial intelligence engine" refers to algorithms and software used to generate new content based on data.

[0097] "Email creation support content" refers to guidelines and templates that provide assistance with email content, layout, and expression.

[0098] "Support regarding how to ask questions" refers to advice on how to ask questions and how to conduct conversations in order to facilitate smooth communication.

[0099] "Methods for detecting power harassment and moral harassment" refer to algorithms and software used to detect and evaluate inappropriate behavior and remarks in the workplace.

[0100] "New employee training materials" refer to the information and educational content necessary for new employees to adapt to their work.

[0101] "Means of execution in response to user requests" refers to modules and algorithms that perform appropriate processing based on requests made by users to the system.

[0102] This invention is a system for efficiently conducting new employee training, providing work support, and improving the workplace environment within a company. The system includes a generative artificial intelligence engine, an email creation support module, a question-asking support module, and power harassment and moral harassment detection modules. These modules are executed in response to requests from the user.

[0103] The server first initializes the generative artificial intelligence engine (hereinafter referred to as the AI ​​engine) and sets up the HR-related information to be used within the company. This process utilizes cloud services (for example, Amazon Web Services or Microsoft Azure®). With this setup complete, the entire system is ready.

[0104] When a user accesses the system and sends a specific request, the server receives the request and uses natural language processing (NLP) techniques to determine its content. Depending on the request content, various processes such as the following are executed:

[0105] Provision of training materials for new employees

[0106] When a user requests "new employee training materials," the server uses an AI engine to generate training content optimized for the user. This generation utilizes natural language generation models such as OpenAI's ChatGPT®. The generated training content is sent to the user's device in PDF format. For example, the prompt might look like this:

[0107] "Please provide the basic materials for new employee training."

[0108] Providing email creation support

[0109] When a user sends a request saying "Please help me compose an email," the server activates the email composition support module and generates an effective email body, layout, and wording based on the user's request. The generated support content is sent to the user's terminal, and the user uses it as a reference to compose the email. The specific prompt text is as follows:

[0110] "Please create a sample sales email for new customers."

[0111] Reports of power harassment / moral harassment

[0112] When a user submits a request to "report harassment," the server activates a harassment detection module and analyzes the user's report in detail. Using techniques such as sentiment analysis and keyword matching, the severity of the report is assessed. As a result, appropriate countermeasures and action plans are proposed and sent to the user's terminal. An example of a prompt message is shown below.

[0113] "I want to report the power harassment I've experienced from a colleague."

[0114] This system allows users to effectively receive new employee training and the necessary support for their work. Furthermore, workplace harassment issues are quickly detected and addressed, leading to improvements in the work environment and overall increased work performance.

[0115] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0116] Provision of training materials for new employees

[0117] Step 1:

[0118] The user submits a request.

[0119] The user sends a request to the system using their terminal, stating "I would like training materials for new employees." The input is a request for "training materials for new employees," and the output is an initial request to the server. Specifically, the user accesses a web form, selects the required materials, and presses the request button.

[0120] Step 2:

[0121] The server receives and processes the request.

[0122] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses natural language processing (NLP) techniques to classify the request content as "new employee training material."

[0123] Step 3:

[0124] The server uses an AI engine to generate training content.

[0125] The server uses a generative artificial intelligence engine (AI engine) to generate optimized training content based on classified requests. The input consists of the request classification result and internal training data, while the output is the generated training content. Specifically, it uses models such as OpenAI's ChatGPT to generate text.

[0126] Step 4:

[0127] The server sends the generated content to the user's device.

[0128] The server sends the generated training content to the user's terminal. The input is the generated content data, and the output is the content sent to the user's terminal. Specifically, the generated training materials are converted to PDF format and sent to the user via email or a download link.

[0129] Step 5:

[0130] Users review training materials.

[0131] The user reviews the training materials received on their device and acquires the necessary knowledge and skills. The input is the PDF document received on the device, and the output is the knowledge and skills the user has acquired. Specifically, the user opens the PDF and reads its contents.

[0132] Providing email creation support

[0133] Step 1:

[0134] The user submits a request.

[0135] The user sends a request to the system using their terminal, stating "Please help me compose an email." The input is a request for "email composition assistance," and the output is an initial request to the server. Specifically, the user fills in the required information in the email composition assistance request form and submits it.

[0136] Step 2:

[0137] The server receives and processes the request.

[0138] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses NLP technology to classify the request content as "email creation support".

[0139] Step 3:

[0140] The server starts the email creation support module.

[0141] The server launches an email creation support module based on the classified request. The input is the request determination result, and the output is the process of generating the support content. Specifically, it reads past email data and standard template messages.

[0142] Step 4:

[0143] The server generates the support information and sends it to the user's terminal.

[0144] The server generates specific email composition advice and templates based on the user's request and sends them to the user's terminal. The input is the user's request and internal data, and the output is the generated support content. Specifically, it creates a template that includes an effective subject line, greeting, body, and closing, and sends it via email.

[0145] Step 5:

[0146] Users refer to the support information when composing an email.

[0147] The user composes the email while referring to the support information. The input is the support information sent from the server, and the output is the completed email. Specifically, the user uses the received template, edits the content as needed, and sends the email.

[0148] Reports of power harassment / moral harassment

[0149] Step 1:

[0150] The user submits a request.

[0151] The user sends a request to the system using their device, stating "I will report workplace harassment." The input is a "workplace harassment report" request, and the output is an initial request to the server. Specifically, the user enters details into the workplace harassment report form and presses the submit button.

[0152] Step 2:

[0153] The server receives and processes the request.

[0154] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses NLP technology to classify the request content as a "harassment report."

[0155] Step 3:

[0156] The server activates the harassment detection module.

[0157] The server activates the harassment detection module based on the classified request. The input is the request classification result, and the output is the analysis data from the detection module. Specifically, it prepares data for sentiment analysis of the reported text.

[0158] Step 4:

[0159] The server analyzes and evaluates the report content.

[0160] The server analyzes the reported content in detail and assesses the possibility of power harassment or moral harassment. The input is user-reported data, and the output is evaluation result data. Specifically, it uses sentiment analysis and keyword matching to assess the severity of the reported content.

[0161] Step 5:

[0162] The server sends the solution to the user's terminal.

[0163] The server proposes appropriate countermeasures based on the evaluation results and sends them to the user's terminal. The input is the evaluation result data, and the output is the proposed countermeasures. Specifically, it creates a document containing contact information for counseling and instructions for internal reporting, and sends it via email.

[0164] Step 6:

[0165] The user reviews and implements the countermeasures.

[0166] The user reviews the proposed countermeasures and implements them as needed. The input is the proposed countermeasure, and the output is the result of the action taken. Specifically, the user reports the problem to their supervisor or compliance officer according to the proposed reporting procedure.

[0167] Through the processing steps described above, the entire system effectively responds to the diverse needs of users, improving the work environment and supporting business operations.

[0168] (Application Example 1)

[0169] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0170] In the food delivery industry, a challenge exists in that new delivery drivers have difficulty acquiring the necessary knowledge and skills effectively and quickly, and while appropriate communication with customers is required, sufficient training and support are not provided. Furthermore, there is a need to address power harassment and moral harassment during deliveries and within the company early on and improve the workplace environment.

[0171] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0172] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for the user, means for providing training materials specifically for food delivery, means for guiding appropriate communication methods at delivery destinations, means for detecting power harassment and moral harassment, and means for executing at least one of these means in response to a request from the user. This system enables new delivery personnel to acquire the necessary knowledge in a short period of time, communicate appropriately with customers, and detect and respond to harassment early.

[0173] A "generative artificial intelligence engine" is an engine that generates information based on user requests and provides appropriate content and support.

[0174] "Email creation support content" refers to content that includes specific expression methods and guidelines for users to create effective emails.

[0175] "Support regarding how to ask questions" refers to providing guidance and instruction on how to communicate effectively.

[0176] "Means for detecting power harassment and moral harassment" refers to methods for identifying harassment based on user reports and suggesting countermeasures.

[0177] "Training content in the food delivery industry" refers to training materials and programs designed to equip new delivery personnel with the knowledge and skills necessary for food delivery operations.

[0178] "New employee training content" refers to materials and content designed to help new employees acquire the basic knowledge and skills necessary to perform their duties.

[0179] "Training materials" are documents that describe specific instructions and procedures for acquiring particular tasks or skills.

[0180] "Appropriate communication methods at delivery destinations" refers to advice and guidelines for delivery personnel to interact smoothly with customers.

[0181] "Means of execution in response to user requests" refers to means of performing appropriate processing or generating content in response to user requests.

[0182] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a training content generation module specifically for the food delivery industry, and a harassment detection module. This system enables training of new delivery personnel, support for communication with customers, and early detection and response to harassment.

[0183] System Configuration

[0184] The server uses a generative artificial intelligence engine to perform the following actions:

[0185] 1. Generating training content for new employees based on user requests.

[0186] 2. Provision of training materials specifically for food delivery.

[0187] 3. A guide to appropriate communication methods at the delivery destination.

[0188] 4. Detection of power harassment and moral harassment, and presentation of countermeasures.

[0189] Program Processing Description

[0190] This system initializes a generative AI engine using OpenAI's GPT model. When the server receives a request from a user, it starts the appropriate module according to the request and performs the necessary generation, support, and detection processes.

[0191] For example, if a user requests training materials for new employees, the server uses an AI engine to generate training content optimized for that individual user and provides it to the user. Similarly, if a user asks about how to communicate with customers, the server utilizes a question-answering module to generate appropriate guidance and provide it to the user.

[0192] When harassment is reported, the detection module evaluates the situation based on detailed information and suggests countermeasures as needed. This allows users to address problems early and improve the workplace environment.

[0193] Hardware and software used

[0194] Hardware: Typical server

[0195] Software: OpenAI GPT model, Python

[0196] Examples of specific cases and prompt statements

[0197] For example, to ensure a new delivery driver receives efficient training, the following prompt can be used:

[0198] Prompt message:

[0199] "Please generate training content for new food delivery drivers. The content should include how to find efficient delivery routes, how to communicate effectively with customers, and tips for safe driving."

[0200] Example of generated content:

[0201] Training program for new food delivery drivers:

[0202] 1. How to find an efficient delivery route

[0203] How to use map apps

[0204] How to avoid rush hour

[0205] 2. Effective methods of communication with customers

[0206] Polite language

[0207] Troubleshooting Basics

[0208] 3. Points for safe driving

[0209] Stopping and braking

[0210] Maintain an appropriate speed

[0211] In this way, new delivery drivers can acquire the necessary knowledge in a short period of time and be well-prepared before starting work.

[0212] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0213] Step 1:

[0214] Users submit requests from their devices. These requests may include new employee training materials, customer communication guides, and harassment reports.

[0215] Input: User request (e.g., Request for new employee training materials)

[0216] Output: Sending a request to the server

[0217] Specific action: The user uses the terminal interface to type "Generate new employee training materials" and clicks the submit button.

[0218] Step 2:

[0219] The server initializes the generative artificial intelligence engine and receives a request from the user. It then selects the appropriate module based on the content of the request.

[0220] Input: User Request

[0221] Output: Selection and startup of the appropriate module

[0222] Specific operation: The server parses the request content and, if it is for new employee training materials, launches the training content generation module.

[0223] Step 3:

[0224] The generative artificial intelligence engine creates prompt sentences for generating training content and sends them to the AI ​​engine.

[0225] Input: Request details, prompt text to be generated

[0226] Output: Prompt statement generation

[0227] Specific operation: The server generates a prompt message saying "Generate training content for new food delivery drivers" and sends it to the AI ​​engine.

[0228] Step 4:

[0229] The AI ​​engine processes the prompt text and generates training content for new employees.

[0230] Input: Prompt message

[0231] Output: Generated new employee training content

[0232] Specific operation: The AI ​​engine uses the OpenAI GPT model to generate training content based on prompt sentences.

[0233] Step 5:

[0234] The server sends the generated training content back to the user.

[0235] Input: Generated training content

[0236] Output: Return to user

[0237] Specific operation: The server sends the generated training content (e.g., how to find efficient routes, tips for safe driving, etc.) to the user's device.

[0238] Step 6:

[0239] Users receive and review training content generated on their devices. If necessary, they can save or print the training content.

[0240] Input: Training content returned from the server

[0241] Output: Display and save training content

[0242] Specific actions: The user reviews the training content received on their device and, if necessary, saves it to the device or prints it out to use as reference material.

[0243] Through the processing steps described above, this system can efficiently provide training content in response to user requests and support the skill development of new delivery drivers in the food delivery industry.

[0244] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0245] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a power harassment and moral harassment detection module, and an emotion engine. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment. Furthermore, by providing support that takes the user's emotions into consideration, it realizes more appropriate and effective support.

[0246] Overall system configuration

[0247] The server comprises a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules operate to generate and provide appropriate support content in response to user requests.

[0248] Program execution flow (overview)

[0249] 1. Server initialization

[0250] The server initializes each engine and module upon startup. In particular, the generative AI engine and the emotion engine load detailed settings and training data.

[0251] 2. Receiving requests from users

[0252] Users send requests to the server through their devices. These requests include new employee training materials, email composition assistance, assistance with asking questions, and reports of power harassment / moral harassment.

[0253] 3. Emotion recognition by an emotion engine

[0254] The server uses an emotion engine to analyze user emotion data (e.g., facial expressions, voice tone, etc.) sent with the request. The analysis results are then fed back to other modules.

[0255] 4. Processing of each module

[0256] Based on the request type and sentiment analysis results, the appropriate module (e.g., generative AI engine, email creation support module, question-asking support module, power harassment / moral harassment detection module) is selected. This allows each module to generate the most suitable support content.

[0257] 5. Responding to the user

[0258] The generated support content and evaluation results are returned to the user. If the results of the emotion engine are incorporated, the content will be optimized for the user's situation and emotions.

[0259] Specific example

[0260] Provision of training materials for new employees

[0261] A user sends a request for new employee training materials to the server. Along with this request, the user's facial expression data and voice are also sent. The server analyzes this data using an emotion engine to determine the user's emotional state. For example, if the user is nervous, training materials written in a relaxing tone are generated. A generative artificial intelligence engine generates specific training content, and the server provides these materials to the user.

[0262] Providing email creation support

[0263] The user submits a request for email composition assistance. The emotion engine generates an email composition guide that includes advice to alleviate the user's feelings of anxiety or impatience if they are experiencing such emotions. The email composition assistance module provides specific expressions and templates, which the server returns to the user.

[0264] Reports of power harassment / moral harassment

[0265] Users report instances of power harassment or moral harassment. The emotion engine evaluates the user's mental state, and the server uses this information to perform a detailed analysis using a moral harassment / power harassment detection module. Along with the evaluation results, appropriate counseling and countermeasures are provided to the user.

[0266] This system allows users to efficiently receive the support and information they need, and enables appropriate responses based on their emotional state. By combining it with an emotional engine, a more human-centered approach is added, which can increase user satisfaction and effectiveness.

[0267] The following describes the processing flow.

[0268] Provision of training materials for new employees

[0269] Step 1:

[0270] A user issues a request for new employee training materials. The user sends the request from their terminal to the server.

[0271] Step 2:

[0272] The server receives the request. It verifies that the request type is "training".

[0273] Step 3:

[0274] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[0275] Step 4:

[0276] The server uses an emotion engine to analyze the user's emotional data and determine the user's emotional state.

[0277] Step 5:

[0278] The server calls the generative artificial intelligence engine to generate optimal training content based on user information and emotional state.

[0279] Step 6:

[0280] The generative artificial intelligence engine generates training content and returns it to the server.

[0281] Step 7:

[0282] The server sends the generated training content to the user terminal. The content is configured taking into account the user's emotional state.

[0283] Step 8:

[0284] The user receives the training content through the terminal and checks the content.

[0285] Provision of email creation support

[0286] Step 1:

[0287] The user issues a request for email creation support. The user sends a request from the user terminal to the server.

[0288] Step 2:

[0289] The server receives the request. It is confirmed that the type of the request is "email_help".

[0290] Step 3:

[0291] The server receives the user's emotional data (such as facial expressions, voice tones, etc.) sent from the user terminal.

[0292] Step 4:

[0293] The server analyzes the user's emotional data using the emotion engine to determine the user's emotional state.

[0294] Step 5:

[0295] The server calls the email creation support module and generates optimal email creation support content based on user information and emotional state.

[0296] Step 6:

[0297] The email creation support module generates email creation support content and returns it to the server.

[0298] Step 7:

[0299] The server sends the generated email creation support content to the user terminal. The content is adjusted taking into account the user's emotional state.

[0300] Step 8:

[0301] The user receives the email creation support content through the terminal and checks the specific expression methods and writing styles.

[0302] Report on power harassment / mora harassment

[0303] Step 1:

[0304] The user requests a report on power harassment or mora harassment. A request is sent from the user terminal to the server.

[0305] Step 2:

[0306] The server receives the request. It is confirmed that the type of the request is "check_abuse".

[0307] Step 3:

[0308] The server receives the user's emotional data (such as facial expressions, voice tones, etc.) sent from the user terminal.

[0309] Step 4:

[0310] The server uses an emotion engine to analyze the user's emotional data and evaluate their mental state.

[0311] Step 5:

[0312] The server invokes a power harassment / moral harassment detection module and evaluates the reported content.

[0313] Step 6:

[0314] The power harassment / moral harassment detection module analyzes the reported content and identifies the problem.

[0315] Step 7:

[0316] The server generates the analysis results and necessary countermeasures and sends them to the user's terminal.

[0317] Step 8:

[0318] Users receive evaluation results and countermeasures via their devices and review the countermeasures as needed.

[0319] Through the above processes, users can efficiently receive the necessary support and information, enabling them to respond appropriately. Furthermore, by combining this with an emotion engine, appropriate support content is provided according to the user's emotional state.

[0320] (Example 2)

[0321] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0322] In conventional systems, AI-powered email creation support, new employee training support, and detection and response to power harassment and moral harassment were handled separately, making centralized management and response difficult. Furthermore, appropriate support that considered the user's emotional state was not adequately provided. Therefore, there is a need to improve the user experience and establish an efficient support system.

[0323] In Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, means for detecting power harassment and moral harassment, means for analyzing the user's emotional data and feeding the results back to other modules for emotion recognition, and means for executing at least one of these means in response to a request from the user. This makes it possible to efficiently and centrally provide support that meets the diverse needs of the user. Furthermore, it is possible to provide appropriate content that takes into account the user's emotional state, and an improvement in the user experience can be expected.

[0324] A "generative artificial intelligence engine" refers to artificial intelligence that automatically generates various types of support content based on user requests.

[0325] "Email creation support content" refers to the guidelines, templates, and specific examples of emails that users need when creating emails for business or personal use.

[0326] "Support regarding how to ask questions" refers to guidelines and sample sentences provided to help users ask appropriate questions and phrases depending on the situation.

[0327] "Methods for detecting power harassment and moral harassment" refers to technologies that analyze user reports and collected data to identify the presence of power harassment and moral harassment.

[0328] "User sentiment data" refers to information about emotions extracted from the user's facial expressions, voice tone, text content, etc.

[0329] "Emotion recognition means" refers to technology that analyzes collected user emotion data and feeds the results back to various support modules.

[0330] "Means of execution in response to user requests" refers to the technology for selecting the appropriate module and executing its function in response to user requests.

[0331] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a power harassment and moral harassment detection module, and an emotion engine. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment. Furthermore, by providing support that takes the user's emotions into consideration, it realizes more appropriate and effective support.

[0332] Overall system configuration

[0333] The server comprises a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules operate to generate and provide appropriate support content in response to user requests.

[0334] Program execution flow (overview)

[0335] The server initializes each engine and module upon startup. In particular, the generative AI engine and the emotion engine load detailed settings and training data.

[0336] Users send requests to the server through their devices. These requests include new employee training materials, email composition assistance, assistance with asking questions, and reports of power harassment / moral harassment. The server analyzes the user's emotional data (e.g., facial expressions, voice tone, etc.) sent with the requests using an emotion engine. The analysis results are then fed back to other modules.

[0337] Based on the request type and sentiment analysis results, the appropriate module (e.g., generative AI engine, email creation support module, question-asking support module, power harassment / moral harassment detection module) is selected. Each module then generates the optimal support content. The generated support content and evaluation results are returned to the user. If the sentiment engine results are incorporated, the content will be optimized for the user's situation and emotions.

[0338] Specific example

[0339] Provision of training materials for new employees

[0340] A user sends a request for new employee training materials to the server. Along with this request, the user's facial expression data and voice are also sent. The server analyzes this data using an emotion engine to determine the user's emotional state. For example, if the user is nervous, training materials written in a relaxing tone are generated. A generative artificial intelligence engine generates specific training content, and the server provides these materials to the user.

[0341] Specific example:

[0342] A user requests, "Please tell me about your new employee training methods."

[0343] Example prompt: "Please provide relaxing training materials for new employees who are feeling nervous."

[0344] Providing email creation support

[0345] The user submits a request for email composition assistance. The emotion engine generates an email composition guide that includes advice to alleviate the user's feelings of anxiety or impatience if they are experiencing such emotions. The email composition assistance module provides specific expressions and templates, which the server returns to the user.

[0346] Specific example:

[0347] The user requests help creating a formal work email.

[0348] Example prompt: "Please provide a formal email template for users who are feeling anxious."

[0349] Reports of power harassment / moral harassment

[0350] Users report instances of power harassment or moral harassment. The emotion engine evaluates the user's mental state, and the server uses this information to perform a detailed analysis using a moral harassment / power harassment detection module. Along with the evaluation results, appropriate counseling and countermeasures are provided to the user.

[0351] Specific example:

[0352] A user requests to report workplace harassment from their superior.

[0353] Example prompt: "Please provide appropriate counseling for users who are experiencing harassment."

[0354] Hardware and software to be used

[0355] The server runs on a computer equipped with a high-performance processor and large memory capacity. Generative artificial intelligence engines and emotion engines are implemented using machine learning libraries such as Python, Tensorflow®, and PyTorch. Databases such as MySQL® and PostgreSQL are used. User terminals include personal computers, smartphones, and tablets, which communicate with the server via the internet.

[0356] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0357] Step 1: Server Initialization

[0358] The server initializes each engine and module upon startup. Specifically, it initializes the generative AI engine and the emotion engine, loading their respective configuration parameters and training data. It reads configuration files (e.g., config.yaml) and sets the connection settings and initial values ​​for each module. The input for this step is the configuration files and training data, and the output is a state where each module is ready to operate.

[0359] Step 2: Receiving requests from users

[0360] The user sends a request to the server through their device. For example, if a user sends a request for new employee training materials, the request data will include text-based instructions and user sentiment data (facial expressions, voice tone, etc.). The server receives the HTTP request and parses the data in JSON format. The input to this step is the user's request data, and the output is the parsed request content and sentiment data.

[0361] Step 3: Emotion recognition by the emotion engine

[0362] The server analyzes the emotion data received from the user using an emotion engine. Facial expression data is analyzed using a facial recognition algorithm, and voice data is identified using a voice analysis algorithm to determine the emotional state. For example, if the user is tense, it will be tagged as tense. The input for this step is the user's facial expression data and voice data, and the output is the identified emotional state (e.g., tense, relaxed).

[0363] Step 4: Processing each module

[0364] The server selects the appropriate module based on the request type and sentiment analysis results. For example, if a user requests email composition assistance and sentiment data indicates anxiety, the server generates assistance content combining the email composition assistance module and the sentiment recognition results. The generative AI engine receives a prompt to generate specific content and produces the optimal response. The input for this step is the request type and sentiment analysis results, and the output is the generated assistance content.

[0365] Step 5: Responding to the user

[0366] The server returns the generated support content and evaluation results to the user. The generated content may include, for example, training materials for new employees, email creation guides, or appropriate counseling content for reports of power harassment / moral harassment. This step also incorporates the results of the emotion engine, ensuring that content is tailored to the user's situation and emotions. The server generates response data in JSON format and sends it to the user's terminal as an HTTP response. The input for this step is the generated support content and emotion analysis results, while the output is the response message to the user.

[0367] (Application Example 2)

[0368] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0369] In modern factories and manufacturing sites, the proper and efficient implementation of new employee training and problem reporting, as well as the early detection and response to harassment, are critical issues. Furthermore, customized support tailored to the user's emotional state is required, but conventional systems have struggled to integrate these aspects. This invention aims to solve these problems and improve the working environment and increase efficiency.

[0370] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0371] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for analyzing user emotion data and generating support content corresponding to the emotional state, and means for displaying and communicating the support content to the user via voice. This enables efficient training of new employees, early detection and countermeasures for problem reporting and harassment, and customized support tailored to the user's emotional state.

[0372] A "generative artificial intelligence engine" is an artificial intelligence that generates appropriate content from text and data. This makes it possible to answer complex questions and create appropriate emails, among other things.

[0373] "Email creation support content" refers to support information and templates provided to help users create emails efficiently and appropriately.

[0374] "Support regarding how to ask questions" refers to support information that provides appropriate expressions and phrases to effectively ask questions or make requests.

[0375] "Methods for detecting power harassment and moral harassment" refer to functions that analyze user input and emotional data to determine whether or not harassment has occurred.

[0376] "User emotional data" refers to data about a user's emotional state, obtained from things like their facial expressions and voice tone.

[0377] "Support tailored to emotional state" refers to support information provided in the format and tone most appropriate to the user, based on the user's emotional data.

[0378] "New employee training content" refers to educational materials and guidelines designed to support the learning of new employees in factories, manufacturing sites, and other similar settings.

[0379] "Counseling" refers to the act or process of providing mental support to victims of power harassment or moral harassment.

[0380] "Means for communicating support content visually and audibly" refers to devices such as displays and speakers that provide the generated support content to the user visually and audibly.

[0381] The present invention relates to a robotic system for supporting new employee training, which is installed in factory robots and includes a generative artificial intelligence engine, an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. Embodiments of this system are described in detail below.

[0382] Hardware and software to be used

[0383] 1. Touchscreen interface

[0384] Example: Use a Raspberry Pi touchscreen.

[0385] 2. Speech Recognition and Synthesis Module

[0386] Example: Use Google® Cloud Speech-to-Text or Text-to-Speech.

[0387] 3. Emotion Recognition Camera

[0388] Example: Use Intel RealSense.

[0389] System program description

[0390] The server includes a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules generate and provide appropriate support content in response to requests from users (factory workers).

[0391] Specifically, the server performs data processing and calculations using the following methods:

[0392] 1. Generative artificial intelligence engine

[0393] When a user requests training materials for new employees, a generative artificial intelligence engine (e.g., GPT-4®) generates the training content.

[0394] The generated training materials will include factory safety guidelines, operating procedures for key equipment, and disaster response procedures.

[0395] 2. Emotional Engine

[0396] An emotion engine (e.g., Affectiva Emotion AI) analyzes data from the user's facial expressions and voice tone to evaluate the user's emotional state.

[0397] For example, if a user is feeling anxious, provide training materials that include relaxation techniques.

[0398] 3. Provided via display and audio.

[0399] Actual display and audio output are performed through the robot's touchscreen and speakers.

[0400] The generated training materials and support content are displayed and communicated via audio in a way that is optimized for the user's emotional state.

[0401] Specific examples

[0402] For example, when a new factory worker requests training materials, the following prompt message is input to the generative artificial intelligence engine:

[0403] Please generate training materials for new employees. Include the following: factory safety guidelines, operating procedures for key equipment, and disaster response procedures. Users are nervous.

[0404] The emotion engine analyzes the user's facial expressions and voice tone at that time and obtains the following analysis data:

[0405] The system analyzes facial expression data and voice tone in real time to evaluate the user's emotional state. The analysis result is [tension, anxiety].

[0406] The generative artificial intelligence engine generates the following based on this data:

[0407] Generate training materials that include relaxation techniques. The proposed training materials will include calm language and relaxation methods.

[0408] The robot then displays the training materials via a touchscreen and uses voice prompts to encourage relaxation.

[0409] This system not only enables efficient training of new employees, but also facilitates problem reporting, early detection and countermeasures against harassment, and customized support tailored to the user's emotional state.

[0410] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0411] Step 1:

[0412] Server initialization and configuration

[0413] Upon startup, the server initializes each engine (generative AI engine, emotion engine) and module (email creation support module, question-asking support module, power harassment / moral harassment detection module) and performs the necessary configurations. This includes loading existing training data and the configuration values ​​for each module. This prepares the server to respond quickly to user requests.

[0414] Step 2:

[0415] Received a request from the user.

[0416] Through a terminal, users (factory workers) send requests for new employee training materials, problem reports, and other information to the server. These requests also include the user's facial expression data and voice. The server receives this information and begins a process to analyze the relevant data. Input data includes the request content, facial expression data, and voice data. Output data provides initial analysis results to be passed on to the next analysis step.

[0417] Step 3:

[0418] Analysis of emotional data using an emotion engine

[0419] The server inputs facial expression data and voice data received from the user into an emotion engine (e.g., Affectiva Emotion AI) to analyze the user's emotional state. The emotion engine converts facial expressions and voice tone into digital data in real time and evaluates emotional states such as tension and anxiety. The input data consists of facial expression data and voice data, and the output data is the evaluation result of the emotional state. The server retrieves this evaluation result and proceeds to the next processing step.

[0420] Step 4:

[0421] Content generation using generative artificial intelligence engines

[0422] Based on the request content and the analysis results of the emotion engine, the server inputs a prompt message into a generative artificial intelligence engine (e.g., GPT-4) to generate corresponding support content (new employee training materials, email content, etc.). For example, it inputs a prompt message like the following:

[0423] Please generate training materials for new employees. Include the following: factory safety guidelines, operating procedures for key equipment, and disaster response procedures. Users are nervous.

[0424] The generative artificial intelligence engine analyzes this prompt and generates the necessary content. The input data consists of the request and the sentiment analysis results, while the output data is the generated training material.

[0425] Step 5:

[0426] Provision of generated support content

[0427] The server displays and delivers supportive content obtained from a generative artificial intelligence engine to the user in an appropriate format. Specifically, it displays text using the robot's touchscreen and provides explanations via a speaker. The input data is the generated supportive content, and the output data is the visual and auditory information provided to the user. This allows the user to efficiently obtain the necessary information.

[0428] Step 6:

[0429] Gathering feedback and optimizing the system

[0430] Users provide feedback on the content they are given. The server collects this feedback information and uses it to generate future support content and perform sentiment analysis. The feedback includes aspects such as understanding and satisfaction with the content. The input data is user feedback information, and the output data is optimized system settings. This allows the system to continuously improve.

[0431] The above describes the processing steps of this system.

[0432] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0433] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0434] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0435] [Second Embodiment]

[0436] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0437] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0438] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0439] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0440] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0441] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0442] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0443] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0444] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0445] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0446] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0447] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0448] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment.

[0449] The server first initializes its generative artificial intelligence engine (AI engine) and sets up HR-related information. Upon receiving a request from a user, the server determines the type of request and executes the appropriate processing accordingly.

[0450] For example, if a user requests training materials for new employees, the server uses an AI engine to generate the most suitable training content for the user and provides it to them. Similarly, if a user requests assistance with email composition, the server uses an email composition assistance module to guide the user on how to compose effective emails.

[0451] If assistance is needed regarding how to ask questions, the server utilizes a question-asking support module to guide users in communicating effectively. Furthermore, if a user reports power harassment or moral harassment, the server evaluates it using a detection module and suggests appropriate countermeasures.

[0452] Specific example

[0453] Provision of training materials for new employees

[0454] When a user sends a request to the server for new employee training materials, the server uses an AI engine to generate training content optimized for that individual user and returns it to the user. The user then reviews the received training materials and acquires the necessary knowledge and skills.

[0455] Providing email creation support

[0456] If a user needs assistance composing an email, the server uses an email composition support module to generate and provide support content that includes specific phrasing and writing style guidance. The user can then use this content to create an effective email.

[0457] Reports of power harassment / moral harassment

[0458] When a user reports harassment (power harassment or moral harassment), the server uses a harassment detection module to evaluate the report based on more detailed information. The server analyzes the report and proposes countermeasures as needed. As a result, users can address problems early, leading to improvements in the workplace environment.

[0459] This system allows users to efficiently acquire necessary knowledge and create a safe and secure work environment. The entire system works together to support users, thereby improving IT literacy and ensuring workplace safety.

[0460] The following describes the processing flow.

[0461] Provision of training materials for new employees

[0462] Step 1:

[0463] A user issues a request for new employee training materials. Specifically, the user sends a request from their terminal to the server.

[0464] Step 2:

[0465] The server receives a request from the user. It analyzes the content of the request and confirms that it is a request of type "training".

[0466] Step 3:

[0467] The server invokes a generative artificial intelligence engine (AI engine) to generate training content based on user information.

[0468] Step 4:

[0469] The AI ​​engine generates the training content the user needs and returns it to the server.

[0470] Step 5:

[0471] The server sends the generated training content to the user's terminal.

[0472] Step 6:

[0473] Users receive training content via their devices and review its contents.

[0474] Providing email creation support

[0475] Step 1:

[0476] The user issues a request for email creation assistance. Specifically, the user sends a request from their terminal to the server.

[0477] Step 2:

[0478] The server receives a request from the user. It parses the request and confirms that it is of type "email_help".

[0479] Step 3:

[0480] The server calls the email creation support module and generates email creation support content based on user information.

[0481] Step 4:

[0482] The email creation support module generates specific email creation advice and templates requested by the user and returns them to the server.

[0483] Step 5:

[0484] The server sends the generated email composition support content to the user's terminal.

[0485] Step 6:

[0486] Users receive email composition support content via their devices, referring to specific expressions and writing styles.

[0487] Reports of power harassment / moral harassment

[0488] Step 1:

[0489] Users request to report instances of power harassment or moral harassment. Specifically, they send requests from their terminals to the server.

[0490] Step 2:

[0491] The server receives a request from the user. It parses the request and confirms that it is of type "check_abuse".

[0492] Step 3:

[0493] The server calls the moral harassment / power harassment detection module and analyzes the reported content.

[0494] Step 4:

[0495] The harassment / power harassment detection module evaluates the reported content and identifies the problem.

[0496] Step 5:

[0497] The server sends the evaluation results and necessary countermeasures to the user's terminal.

[0498] Step 6:

[0499] Users receive evaluation results via their devices and confirm necessary countermeasures.

[0500] In this way, the system operates by calling the appropriate module for each request, providing the user with the most suitable content, support, and problem-solving solutions.

[0501] (Example 1)

[0502] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0503] A system is needed to efficiently conduct new employee training, provide work support, and improve the work environment within companies. Traditional methods often lacked coherence and efficiency, as various content was generated and support provided individually. Furthermore, there is a need for faster early detection and response to power harassment and moral harassment. These challenges must be addressed in an integrated manner.

[0504] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0505] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, means for detecting power harassment and moral harassment, means for generating new employee training materials, and means for executing at least one of these means in response to a user request. This makes it possible to efficiently conduct new employee training and work support within a company, and to comprehensively support the improvement of the work environment.

[0506] A "generative artificial intelligence engine" refers to algorithms and software used to generate new content based on data.

[0507] "Email creation support content" refers to guidelines and templates that provide assistance with email content, layout, and expression.

[0508] "Support regarding how to ask questions" refers to advice on how to ask questions and how to conduct conversations in order to facilitate smooth communication.

[0509] "Methods for detecting power harassment and moral harassment" refer to algorithms and software used to detect and evaluate inappropriate behavior and remarks in the workplace.

[0510] "New employee training materials" refer to the information and educational content necessary for new employees to adapt to their work.

[0511] "Means of execution in response to user requests" refers to modules and algorithms that perform appropriate processing based on requests made by users to the system.

[0512] This invention is a system for efficiently conducting new employee training, providing work support, and improving the workplace environment within a company. The system includes a generative artificial intelligence engine, an email creation support module, a question-asking support module, and power harassment and moral harassment detection modules. These modules are executed in response to requests from the user.

[0513] The server first initializes the generative artificial intelligence engine (hereinafter referred to as the AI ​​engine) and sets up the HR-related information to be used within the company. This process utilizes cloud services (for example, Amazon Web Services or Microsoft Azure). With this setup complete, the entire system is ready.

[0514] When a user accesses the system and sends a specific request, the server receives the request and uses natural language processing (NLP) techniques to determine its content. Depending on the request content, various processes such as the following are executed:

[0515] Provision of training materials for new employees

[0516] When a user requests "new employee training materials," the server uses an AI engine to generate training content optimized for the user. This generation utilizes natural language generation models such as OpenAI's ChatGPT. The generated training content is sent to the user's device in PDF format. For example, the prompt might look like this:

[0517] "Please provide the basic materials for new employee training."

[0518] Providing email creation support

[0519] When a user sends a request saying "Please help me compose an email," the server activates the email composition support module and generates an effective email body, layout, and wording based on the user's request. The generated support content is sent to the user's terminal, and the user uses it as a reference to compose the email. The specific prompt text is as follows:

[0520] "Please create a sample sales email for new customers."

[0521] Reports of power harassment / moral harassment

[0522] When a user submits a request to "report harassment," the server activates a harassment detection module and analyzes the user's report in detail. Using techniques such as sentiment analysis and keyword matching, the severity of the report is assessed. As a result, appropriate countermeasures and action plans are proposed and sent to the user's terminal. An example of a prompt message is shown below.

[0523] "I want to report the power harassment I've experienced from a colleague."

[0524] This system allows users to effectively receive new employee training and the necessary support for their work. Furthermore, workplace harassment issues are quickly detected and addressed, leading to improvements in the work environment and overall increased work performance.

[0525] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0526] Provision of training materials for new employees

[0527] Step 1:

[0528] The user submits a request.

[0529] The user sends a request to the system using their terminal, stating "I would like training materials for new employees." The input is a request for "training materials for new employees," and the output is an initial request to the server. Specifically, the user accesses a web form, selects the required materials, and presses the request button.

[0530] Step 2:

[0531] The server receives and processes the request.

[0532] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses natural language processing (NLP) techniques to classify the request content as "new employee training material."

[0533] Step 3:

[0534] The server uses an AI engine to generate training content.

[0535] The server uses a generative artificial intelligence engine (AI engine) to generate optimized training content based on classified requests. The input consists of the request classification result and internal training data, while the output is the generated training content. Specifically, it uses models such as OpenAI's ChatGPT to generate text.

[0536] Step 4:

[0537] The server sends the generated content to the user's device.

[0538] The server sends the generated training content to the user's terminal. The input is the generated content data, and the output is the content sent to the user's terminal. Specifically, the generated training materials are converted to PDF format and sent to the user via email or a download link.

[0539] Step 5:

[0540] Users review training materials.

[0541] The user reviews the training materials received on their device and acquires the necessary knowledge and skills. The input is the PDF document received on the device, and the output is the knowledge and skills the user has acquired. Specifically, the user opens the PDF and reads its contents.

[0542] Providing email creation support

[0543] Step 1:

[0544] The user submits a request.

[0545] The user sends a request to the system using their terminal, stating "Please help me compose an email." The input is a request for "email composition assistance," and the output is an initial request to the server. Specifically, the user fills in the required information in the email composition assistance request form and submits it.

[0546] Step 2:

[0547] The server receives and processes the request.

[0548] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses NLP technology to classify the request content as "email creation support".

[0549] Step 3:

[0550] The server starts the email creation support module.

[0551] The server launches an email creation support module based on the classified request. The input is the request determination result, and the output is the process of generating the support content. Specifically, it reads past email data and standard template messages.

[0552] Step 4:

[0553] The server generates the support information and sends it to the user's terminal.

[0554] The server generates specific email composition advice and templates based on the user's request and sends them to the user's terminal. The input is the user's request and internal data, and the output is the generated support content. Specifically, it creates a template that includes an effective subject line, greeting, body, and closing, and sends it via email.

[0555] Step 5:

[0556] Users refer to the support information when composing an email.

[0557] The user composes the email while referring to the support information. The input is the support information sent from the server, and the output is the completed email. Specifically, the user uses the received template, edits the content as needed, and sends the email.

[0558] Reports of power harassment / moral harassment

[0559] Step 1:

[0560] The user submits a request.

[0561] The user sends a request to the system using their device, stating "I will report workplace harassment." The input is a "workplace harassment report" request, and the output is an initial request to the server. Specifically, the user enters details into the workplace harassment report form and presses the submit button.

[0562] Step 2:

[0563] The server receives and processes the request.

[0564] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses NLP technology to classify the request content as a "harassment report."

[0565] Step 3:

[0566] The server activates the harassment detection module.

[0567] The server activates the harassment detection module based on the classified request. The input is the request classification result, and the output is the analysis data from the detection module. Specifically, it prepares data for sentiment analysis of the reported text.

[0568] Step 4:

[0569] The server analyzes and evaluates the report content.

[0570] The server analyzes the reported content in detail and assesses the possibility of power harassment or moral harassment. The input is user-reported data, and the output is evaluation result data. Specifically, it uses sentiment analysis and keyword matching to assess the severity of the reported content.

[0571] Step 5:

[0572] The server sends the solution to the user's terminal.

[0573] The server proposes appropriate countermeasures based on the evaluation results and sends them to the user's terminal. The input is the evaluation result data, and the output is the proposed countermeasures. Specifically, it creates a document containing contact information for counseling and instructions for internal reporting, and sends it via email.

[0574] Step 6:

[0575] The user reviews and implements the countermeasures.

[0576] The user reviews the proposed countermeasures and implements them as needed. The input is the proposed countermeasure, and the output is the result of the action taken. Specifically, the user reports the problem to their supervisor or compliance officer according to the proposed reporting procedure.

[0577] Through the processing steps described above, the entire system effectively responds to the diverse needs of users, improving the work environment and supporting business operations.

[0578] (Application Example 1)

[0579] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0580] In the food delivery industry, a challenge exists in that new delivery drivers have difficulty acquiring the necessary knowledge and skills effectively and quickly, and while appropriate communication with customers is required, sufficient training and support are not provided. Furthermore, there is a need to address power harassment and moral harassment during deliveries and within the company early on and improve the workplace environment.

[0581] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0582] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for the user, means for providing training materials specifically for food delivery, means for guiding appropriate communication methods at delivery destinations, means for detecting power harassment and moral harassment, and means for executing at least one of these means in response to a request from the user. This system enables new delivery personnel to acquire the necessary knowledge in a short period of time, communicate appropriately with customers, and detect and respond to harassment early.

[0583] A "generative artificial intelligence engine" is an engine that generates information based on user requests and provides appropriate content and support.

[0584] "Email creation support content" refers to content that includes specific expression methods and guidelines for users to create effective emails.

[0585] "Support regarding how to ask questions" refers to providing guidance and instruction on how to communicate effectively.

[0586] "Means for detecting power harassment and moral harassment" refers to methods for identifying harassment based on user reports and suggesting countermeasures.

[0587] "Training content in the food delivery industry" refers to training materials and programs designed to equip new delivery personnel with the knowledge and skills necessary for food delivery operations.

[0588] "New employee training content" refers to materials and content designed to help new employees acquire the basic knowledge and skills necessary to perform their duties.

[0589] "Training materials" are documents that describe specific instructions and procedures for acquiring particular tasks or skills.

[0590] "Appropriate communication methods at delivery destinations" refers to advice and guidelines for delivery personnel to interact smoothly with customers.

[0591] "Means of execution in response to user requests" refers to means of performing appropriate processing or generating content in response to user requests.

[0592] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a training content generation module specifically for the food delivery industry, and a harassment detection module. This system enables training of new delivery personnel, support for communication with customers, and early detection and response to harassment.

[0593] System Configuration

[0594] The server uses a generative artificial intelligence engine to perform the following actions:

[0595] 1. Generating training content for new employees based on user requests.

[0596] 2. Provision of training materials specifically for food delivery.

[0597] 3. A guide to appropriate communication methods at the delivery destination.

[0598] 4. Detection of power harassment and moral harassment, and presentation of countermeasures.

[0599] Program Processing Description

[0600] This system initializes a generative AI engine using OpenAI's GPT model. When the server receives a request from a user, it starts the appropriate module according to the request and performs the necessary generation, support, and detection processes.

[0601] For example, if a user requests training materials for new employees, the server uses an AI engine to generate training content optimized for that individual user and provides it to the user. Similarly, if a user asks about how to communicate with customers, the server utilizes a question-answering module to generate appropriate guidance and provide it to the user.

[0602] When harassment is reported, the detection module evaluates the situation based on detailed information and suggests countermeasures as needed. This allows users to address problems early and improve the workplace environment.

[0603] Hardware and software used

[0604] Hardware: Typical server

[0605] Software: OpenAI GPT model, Python

[0606] Examples of specific cases and prompt statements

[0607] For example, to ensure a new delivery driver receives efficient training, the following prompt can be used:

[0608] Prompt message:

[0609] "Please generate training content for new food delivery drivers. The content should include how to find efficient delivery routes, how to communicate effectively with customers, and tips for safe driving."

[0610] Example of generated content:

[0611] Training program for new food delivery drivers:

[0612] 1. How to find an efficient delivery route

[0613] How to use map apps

[0614] How to avoid rush hour

[0615] 2. Effective methods of communication with customers

[0616] Polite language

[0617] Troubleshooting Basics

[0618] 3. Points for safe driving

[0619] Stopping and braking

[0620] Maintain an appropriate speed

[0621] In this way, new delivery drivers can acquire the necessary knowledge in a short period of time and be well-prepared before starting work.

[0622] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0623] Step 1:

[0624] Users submit requests from their devices. These requests may include new employee training materials, customer communication guides, and harassment reports.

[0625] Input: User request (e.g., Request for new employee training materials)

[0626] Output: Sending a request to the server

[0627] Specific action: The user uses the terminal interface to type "Generate new employee training materials" and clicks the submit button.

[0628] Step 2:

[0629] The server initializes the generative artificial intelligence engine and receives a request from the user. It then selects the appropriate module based on the content of the request.

[0630] Input: User Request

[0631] Output: Selection and startup of the appropriate module

[0632] Specific operation: The server parses the request content and, if it is for new employee training materials, launches the training content generation module.

[0633] Step 3:

[0634] The generative artificial intelligence engine creates prompt sentences for generating training content and sends them to the AI ​​engine.

[0635] Input: Request details, prompt text to be generated

[0636] Output: Prompt statement generation

[0637] Specific operation: The server generates a prompt message saying "Generate training content for new food delivery drivers" and sends it to the AI ​​engine.

[0638] Step 4:

[0639] The AI ​​engine processes the prompt text and generates training content for new employees.

[0640] Input: Prompt message

[0641] Output: Generated new employee training content

[0642] Specific operation: The AI ​​engine uses the OpenAI GPT model to generate training content based on prompt sentences.

[0643] Step 5:

[0644] The server sends the generated training content back to the user.

[0645] Input: Generated training content

[0646] Output: Return to user

[0647] Specific operation: The server sends the generated training content (e.g., how to find efficient routes, tips for safe driving, etc.) to the user's device.

[0648] Step 6:

[0649] Users receive and review training content generated on their devices. If necessary, they can save or print the training content.

[0650] Input: Training content returned from the server

[0651] Output: Display and save training content

[0652] Specific actions: The user reviews the training content received on their device and, if necessary, saves it to the device or prints it out to use as reference material.

[0653] Through the processing steps described above, this system can efficiently provide training content in response to user requests and support the skill development of new delivery drivers in the food delivery industry.

[0654] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0655] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a power harassment and moral harassment detection module, and an emotion engine. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment. Furthermore, by providing support that takes the user's emotions into consideration, it realizes more appropriate and effective support.

[0656] Overall system configuration

[0657] The server comprises a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules operate to generate and provide appropriate support content in response to user requests.

[0658] Program execution flow (overview)

[0659] 1. Server initialization

[0660] The server initializes each engine and module upon startup. In particular, the generative AI engine and the emotion engine load detailed settings and training data.

[0661] 2. Receiving requests from users

[0662] Users send requests to the server through their devices. These requests include new employee training materials, email composition assistance, assistance with asking questions, and reports of power harassment / moral harassment.

[0663] 3. Emotion recognition by an emotion engine

[0664] The server uses an emotion engine to analyze user emotion data (e.g., facial expressions, voice tone, etc.) sent with the request. The analysis results are then fed back to other modules.

[0665] 4. Processing of each module

[0666] Based on the request type and sentiment analysis results, the appropriate module (e.g., generative AI engine, email creation support module, question-asking support module, power harassment / moral harassment detection module) is selected. This allows each module to generate the most suitable support content.

[0667] 5. Responding to the user

[0668] The generated support content and evaluation results are returned to the user. If the results of the emotion engine are incorporated, the content will be optimized for the user's situation and emotions.

[0669] Specific example

[0670] Provision of training materials for new employees

[0671] A user sends a request for new employee training materials to the server. Along with this request, the user's facial expression data and voice are also sent. The server analyzes this data using an emotion engine to determine the user's emotional state. For example, if the user is nervous, training materials written in a relaxing tone are generated. A generative artificial intelligence engine generates specific training content, and the server provides these materials to the user.

[0672] Providing email creation support

[0673] The user submits a request for email composition assistance. The emotion engine generates an email composition guide that includes advice to alleviate the user's feelings of anxiety or impatience if they are experiencing such emotions. The email composition assistance module provides specific expressions and templates, which the server returns to the user.

[0674] Reports of power harassment / moral harassment

[0675] Users report instances of power harassment or moral harassment. The emotion engine evaluates the user's mental state, and the server uses this information to perform a detailed analysis using a moral harassment / power harassment detection module. Along with the evaluation results, appropriate counseling and countermeasures are provided to the user.

[0676] This system allows users to efficiently receive the support and information they need, and enables appropriate responses based on their emotional state. By combining it with an emotional engine, a more human-centered approach is added, which can increase user satisfaction and effectiveness.

[0677] The following describes the processing flow.

[0678] Provision of training materials for new employees

[0679] Step 1:

[0680] A user issues a request for new employee training materials. The user sends the request from their terminal to the server.

[0681] Step 2:

[0682] The server receives the request. It verifies that the request type is "training".

[0683] Step 3:

[0684] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[0685] Step 4:

[0686] The server uses an emotion engine to analyze the user's emotional data and determine the user's emotional state.

[0687] Step 5:

[0688] The server invokes a generative artificial intelligence engine to generate optimal training content based on user information and emotional state.

[0689] Step 6:

[0690] A generative artificial intelligence engine generates training content and returns it to the server.

[0691] Step 7:

[0692] The server sends the generated training content to the user's terminal. The content is structured with the user's emotional state in mind.

[0693] Step 8:

[0694] Users receive training content via their devices and review its contents.

[0695] Providing email creation support

[0696] Step 1:

[0697] The user issues a request for email creation assistance. The request is sent from the user's terminal to the server.

[0698] Step 2:

[0699] The server receives the request. It verifies that the request type is "email_help".

[0700] Step 3:

[0701] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[0702] Step 4:

[0703] The server uses an emotion engine to analyze the user's emotional data and determine the user's emotional state.

[0704] Step 5:

[0705] The server calls an email composition support module and generates optimal email composition support content based on user information and emotional state.

[0706] Step 6:

[0707] The email composition support module generates the email composition support content and returns it to the server.

[0708] Step 7:

[0709] The server sends the generated email composition support content to the user's terminal. The content is adjusted to take the user's emotional state into consideration.

[0710] Step 8:

[0711] Users receive email composition support content via their devices and check specific expressions and writing styles.

[0712] Reports of power harassment / moral harassment

[0713] Step 1:

[0714] A user requests to report instances of power harassment or moral harassment. The user sends the request from their terminal to the server.

[0715] Step 2:

[0716] The server receives the request. It verifies that the request type is "check_abuse".

[0717] Step 3:

[0718] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[0719] Step 4:

[0720] The server uses an emotion engine to analyze the user's emotional data and evaluate their mental state.

[0721] Step 5:

[0722] The server invokes a power harassment / moral harassment detection module and evaluates the reported content.

[0723] Step 6:

[0724] The power harassment / moral harassment detection module analyzes the reported content and identifies the problem.

[0725] Step 7:

[0726] The server generates the analysis results and necessary countermeasures and sends them to the user's terminal.

[0727] Step 8:

[0728] Users receive evaluation results and countermeasures via their devices and review the countermeasures as needed.

[0729] Through the above processes, users can efficiently receive the necessary support and information, enabling them to respond appropriately. Furthermore, by combining this with an emotion engine, appropriate support content is provided according to the user's emotional state.

[0730] (Example 2)

[0731] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0732] In conventional systems, AI-powered email creation support, new employee training support, and detection and response to power harassment and moral harassment were handled separately, making centralized management and response difficult. Furthermore, appropriate support that considered the user's emotional state was not adequately provided. Therefore, there is a need to improve the user experience and establish an efficient support system.

[0733] In Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, means for detecting power harassment and moral harassment, means for analyzing the user's emotional data and feeding the results back to other modules for emotion recognition, and means for executing at least one of these means in response to a request from the user. This makes it possible to efficiently and centrally provide support that meets the diverse needs of the user. Furthermore, it is possible to provide appropriate content that takes into account the user's emotional state, and an improvement in the user experience can be expected.

[0734] A "generative artificial intelligence engine" refers to artificial intelligence that automatically generates various types of support content based on user requests.

[0735] "Email creation support content" refers to the guidelines, templates, and specific examples of emails that users need when creating emails for business or personal use.

[0736] "Support regarding how to ask questions" refers to guidelines and sample sentences provided to help users ask appropriate questions and phrases depending on the situation.

[0737] "Methods for detecting power harassment and moral harassment" refers to technologies that analyze user reports and collected data to identify the presence of power harassment and moral harassment.

[0738] "User sentiment data" refers to information about emotions extracted from the user's facial expressions, voice tone, text content, etc.

[0739] "Emotion recognition means" refers to technology that analyzes collected user emotion data and feeds the results back to various support modules.

[0740] "Means of execution in response to user requests" refers to the technology for selecting the appropriate module and executing its function in response to user requests.

[0741] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a power harassment and moral harassment detection module, and an emotion engine. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment. Furthermore, by providing support that takes the user's emotions into consideration, it realizes more appropriate and effective support.

[0742] Overall system configuration

[0743] The server comprises a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules operate to generate and provide appropriate support content in response to user requests.

[0744] Program execution flow (overview)

[0745] The server initializes each engine and module upon startup. In particular, the generative AI engine and the emotion engine load detailed settings and training data.

[0746] Users send requests to the server through their devices. These requests include new employee training materials, email composition assistance, assistance with asking questions, and reports of power harassment / moral harassment. The server analyzes the user's emotional data (e.g., facial expressions, voice tone, etc.) sent with the requests using an emotion engine. The analysis results are then fed back to other modules.

[0747] Based on the request type and sentiment analysis results, the appropriate module (e.g., generative AI engine, email creation support module, question-asking support module, power harassment / moral harassment detection module) is selected. Each module then generates the optimal support content. The generated support content and evaluation results are returned to the user. If the sentiment engine results are incorporated, the content will be optimized for the user's situation and emotions.

[0748] Specific example

[0749] Provision of training materials for new employees

[0750] A user sends a request for new employee training materials to the server. Along with this request, the user's facial expression data and voice are also sent. The server analyzes this data using an emotion engine to determine the user's emotional state. For example, if the user is nervous, training materials written in a relaxing tone are generated. A generative artificial intelligence engine generates specific training content, and the server provides these materials to the user.

[0751] Specific example:

[0752] A user requests, "Please tell me about your new employee training methods."

[0753] Example prompt: "Please provide relaxing training materials for new employees who are feeling nervous."

[0754] Providing email creation support

[0755] The user submits a request for email composition assistance. The emotion engine generates an email composition guide that includes advice to alleviate the user's feelings of anxiety or impatience if they are experiencing such emotions. The email composition assistance module provides specific expressions and templates, which the server returns to the user.

[0756] Specific example:

[0757] The user requests help creating a formal work email.

[0758] Example prompt: "Please provide a formal email template for users who are feeling anxious."

[0759] Reports of power harassment / moral harassment

[0760] Users report instances of power harassment or moral harassment. The emotion engine evaluates the user's mental state, and the server uses this information to perform a detailed analysis using a moral harassment / power harassment detection module. Along with the evaluation results, appropriate counseling and countermeasures are provided to the user.

[0761] Specific example:

[0762] A user requests to report workplace harassment from their superior.

[0763] Example prompt: "Please provide appropriate counseling for users who are experiencing harassment."

[0764] Hardware and software to be used

[0765] The server runs on a computer equipped with a high-performance processor and large memory capacity. Generative artificial intelligence engines and emotion engines are implemented using machine learning libraries such as Python, TensorFlow, and PyTorch. RDBMS such as MySQL and PostgreSQL are used for the database. User terminals include personal computers, smartphones, and tablets, which communicate with the server via the internet.

[0766] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0767] Step 1: Server Initialization

[0768] The server initializes each engine and module upon startup. Specifically, it initializes the generative AI engine and the emotion engine, loading their respective configuration parameters and training data. It reads configuration files (e.g., config.yaml) and sets the connection settings and initial values ​​for each module. The input for this step is the configuration files and training data, and the output is a state where each module is ready to operate.

[0769] Step 2: Receiving requests from users

[0770] The user sends a request to the server through their device. For example, if a user sends a request for new employee training materials, the request data will include text-based instructions and user sentiment data (facial expressions, voice tone, etc.). The server receives the HTTP request and parses the data in JSON format. The input to this step is the user's request data, and the output is the parsed request content and sentiment data.

[0771] Step 3: Emotion recognition by the emotion engine

[0772] The server analyzes the emotion data received from the user using an emotion engine. Facial expression data is analyzed using a facial recognition algorithm, and voice data is identified using a voice analysis algorithm to determine the emotional state. For example, if the user is tense, it will be tagged as tense. The input for this step is the user's facial expression data and voice data, and the output is the identified emotional state (e.g., tense, relaxed).

[0773] Step 4: Processing each module

[0774] The server selects the appropriate module based on the request type and sentiment analysis results. For example, if a user requests email composition assistance and sentiment data indicates anxiety, the server generates assistance content combining the email composition assistance module and the sentiment recognition results. The generative AI engine receives a prompt to generate specific content and produces the optimal response. The input for this step is the request type and sentiment analysis results, and the output is the generated assistance content.

[0775] Step 5: Responding to the user

[0776] The server returns the generated support content and evaluation results to the user. The generated content may include, for example, training materials for new employees, email creation guides, or appropriate counseling content for reports of power harassment / moral harassment. This step also incorporates the results of the emotion engine, ensuring that content is tailored to the user's situation and emotions. The server generates response data in JSON format and sends it to the user's terminal as an HTTP response. The input for this step is the generated support content and emotion analysis results, while the output is the response message to the user.

[0777] (Application Example 2)

[0778] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0779] In modern factories and manufacturing sites, the proper and efficient implementation of new employee training and problem reporting, as well as the early detection and response to harassment, are critical issues. Furthermore, customized support tailored to the user's emotional state is required, but conventional systems have struggled to integrate these aspects. This invention aims to solve these problems and improve the working environment and increase efficiency.

[0780] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0781] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for analyzing user emotion data and generating support content corresponding to the emotional state, and means for displaying and communicating the support content to the user via voice. This enables efficient training of new employees, early detection and countermeasures for problem reporting and harassment, and customized support tailored to the user's emotional state.

[0782] A "generative artificial intelligence engine" is an artificial intelligence that generates appropriate content from text and data. This makes it possible to answer complex questions and create appropriate emails, among other things.

[0783] "Email creation support content" refers to support information and templates provided to help users create emails efficiently and appropriately.

[0784] "Support regarding how to ask questions" refers to support information that provides appropriate expressions and phrases to effectively ask questions or make requests.

[0785] "Methods for detecting power harassment and moral harassment" refer to functions that analyze user input and emotional data to determine whether or not harassment has occurred.

[0786] "User emotional data" refers to data about a user's emotional state, obtained from things like their facial expressions and voice tone.

[0787] "Support tailored to emotional state" refers to support information provided in the format and tone most appropriate to the user, based on the user's emotional data.

[0788] "New employee training content" refers to educational materials and guidelines designed to support the learning of new employees in factories, manufacturing sites, and other similar settings.

[0789] "Counseling" refers to the act or process of providing mental support to victims of power harassment or moral harassment.

[0790] "Means for communicating support content visually and audibly" refers to devices such as displays and speakers that provide the generated support content to the user visually and audibly.

[0791] The present invention relates to a robotic system for supporting new employee training, which is installed in factory robots and includes a generative artificial intelligence engine, an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. Embodiments of this system are described in detail below.

[0792] Hardware and software to be used

[0793] 1. Touchscreen interface

[0794] Example: Use a Raspberry Pi touchscreen.

[0795] 2. Speech Recognition and Synthesis Module

[0796] Example: Use Google Cloud Speech-to-Text or Text-to-Speech.

[0797] 3. Emotion Recognition Camera

[0798] Example: Use Intel RealSense.

[0799] System program description

[0800] The server includes a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules generate and provide appropriate support content in response to requests from users (factory workers).

[0801] Specifically, the server performs data processing and calculations using the following methods:

[0802] 1. Generative artificial intelligence engine

[0803] When a user requests training materials for new employees, a generative artificial intelligence engine (e.g., GPT-4) generates the training content.

[0804] The generated training materials will include factory safety guidelines, operating procedures for key equipment, and disaster response procedures.

[0805] 2. Emotional Engine

[0806] An emotion engine (e.g., Affectiva Emotion AI) analyzes data from the user's facial expressions and voice tone to evaluate the user's emotional state.

[0807] For example, if a user is feeling anxious, provide training materials that include relaxation techniques.

[0808] 3. Provided via display and audio.

[0809] Actual display and audio output are performed through the robot's touchscreen and speakers.

[0810] The generated training materials and support content are displayed and communicated via audio in a way that is optimized for the user's emotional state.

[0811] Specific examples

[0812] For example, when a new factory worker requests training materials, the following prompt message is input to the generative artificial intelligence engine:

[0813] Please generate training materials for new employees. Include the following: factory safety guidelines, operating procedures for key equipment, and disaster response procedures. Users are nervous.

[0814] The emotion engine analyzes the user's facial expressions and voice tone at that time and obtains the following analysis data:

[0815] The system analyzes facial expression data and voice tone in real time to evaluate the user's emotional state. The analysis result is [tension, anxiety].

[0816] The generative artificial intelligence engine generates the following based on this data:

[0817] Generate training materials that include relaxation techniques. The proposed training materials will include calm language and relaxation methods.

[0818] The robot then displays the training materials via a touchscreen and uses voice prompts to encourage relaxation.

[0819] This system not only enables efficient training of new employees, but also facilitates problem reporting, early detection and countermeasures against harassment, and customized support tailored to the user's emotional state.

[0820] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0821] Step 1:

[0822] Server initialization and configuration

[0823] Upon startup, the server initializes each engine (generative AI engine, emotion engine) and module (email creation support module, question-asking support module, power harassment / moral harassment detection module) and performs the necessary configurations. This includes loading existing training data and the configuration values ​​for each module. This prepares the server to respond quickly to user requests.

[0824] Step 2:

[0825] Received a request from the user.

[0826] Through a terminal, users (factory workers) send requests for new employee training materials, problem reports, and other information to the server. These requests also include the user's facial expression data and voice. The server receives this information and begins a process to analyze the relevant data. Input data includes the request content, facial expression data, and voice data. Output data provides initial analysis results to be passed on to the next analysis step.

[0827] Step 3:

[0828] Analysis of emotional data using an emotion engine

[0829] The server inputs facial expression data and voice data received from the user into an emotion engine (e.g., Affectiva Emotion AI) to analyze the user's emotional state. The emotion engine converts facial expressions and voice tone into digital data in real time and evaluates emotional states such as tension and anxiety. The input data consists of facial expression data and voice data, and the output data is the evaluation result of the emotional state. The server retrieves this evaluation result and proceeds to the next processing step.

[0830] Step 4:

[0831] Content generation using generative artificial intelligence engines

[0832] Based on the request content and the analysis results of the emotion engine, the server inputs a prompt message into a generative artificial intelligence engine (e.g., GPT-4) to generate corresponding support content (new employee training materials, email content, etc.). For example, it inputs a prompt message like the following:

[0833] Please generate training materials for new employees. Include the following: factory safety guidelines, operating procedures for key equipment, and disaster response procedures. Users are nervous.

[0834] The generative artificial intelligence engine analyzes this prompt and generates the necessary content. The input data consists of the request and the sentiment analysis results, while the output data is the generated training material.

[0835] Step 5:

[0836] Provision of generated support content

[0837] The server displays and delivers supportive content obtained from a generative artificial intelligence engine to the user in an appropriate format. Specifically, it displays text using the robot's touchscreen and provides explanations via a speaker. The input data is the generated supportive content, and the output data is the visual and auditory information provided to the user. This allows the user to efficiently obtain the necessary information.

[0838] Step 6:

[0839] Gathering feedback and optimizing the system

[0840] Users provide feedback on the content they are given. The server collects this feedback information and uses it to generate future support content and perform sentiment analysis. The feedback includes aspects such as understanding and satisfaction with the content. The input data is user feedback information, and the output data is optimized system settings. This allows the system to continuously improve.

[0841] The above describes the processing steps of this system.

[0842] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0843] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0844] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0845] [Third Embodiment]

[0846] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0847] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0848] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0849] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0850] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0851] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0852] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0853] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0854] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0855] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0856] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0857] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0858] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment.

[0859] The server first initializes its generative artificial intelligence engine (AI engine) and sets up HR-related information. Upon receiving a request from a user, the server determines the type of request and executes the appropriate processing accordingly.

[0860] For example, if a user requests training materials for new employees, the server uses an AI engine to generate the most suitable training content for the user and provides it to them. Similarly, if a user requests assistance with email composition, the server uses an email composition assistance module to guide the user on how to compose effective emails.

[0861] If assistance is needed regarding how to ask questions, the server utilizes a question-asking support module to guide users in communicating effectively. Furthermore, if a user reports power harassment or moral harassment, the server evaluates it using a detection module and suggests appropriate countermeasures.

[0862] Specific example

[0863] Provision of training materials for new employees

[0864] When a user sends a request to the server for new employee training materials, the server uses an AI engine to generate training content optimized for that individual user and returns it to the user. The user then reviews the received training materials and acquires the necessary knowledge and skills.

[0865] Providing email creation support

[0866] If a user needs assistance composing an email, the server uses an email composition support module to generate and provide support content that includes specific phrasing and writing style guidance. The user can then use this content to create an effective email.

[0867] Reports of power harassment / moral harassment

[0868] When a user reports harassment (power harassment or moral harassment), the server uses a harassment detection module to evaluate the report based on more detailed information. The server analyzes the report and proposes countermeasures as needed. As a result, users can address problems early, leading to improvements in the workplace environment.

[0869] This system allows users to efficiently acquire necessary knowledge and create a safe and secure work environment. The entire system works together to support users, thereby improving IT literacy and ensuring workplace safety.

[0870] The following describes the processing flow.

[0871] Provision of training materials for new employees

[0872] Step 1:

[0873] A user issues a request for new employee training materials. Specifically, the user sends a request from their terminal to the server.

[0874] Step 2:

[0875] The server receives a request from the user. It analyzes the content of the request and confirms that it is a request of type "training".

[0876] Step 3:

[0877] The server invokes a generative artificial intelligence engine (AI engine) to generate training content based on user information.

[0878] Step 4:

[0879] The AI ​​engine generates the training content the user needs and returns it to the server.

[0880] Step 5:

[0881] The server sends the generated training content to the user's terminal.

[0882] Step 6:

[0883] Users receive training content via their devices and review its contents.

[0884] Providing email creation support

[0885] Step 1:

[0886] The user issues a request for email creation assistance. Specifically, the user sends a request from their terminal to the server.

[0887] Step 2:

[0888] The server receives a request from the user. It parses the request and confirms that it is of type "email_help".

[0889] Step 3:

[0890] The server calls the email creation support module and generates email creation support content based on user information.

[0891] Step 4:

[0892] The email creation support module generates specific email creation advice and templates requested by the user and returns them to the server.

[0893] Step 5:

[0894] The server sends the generated email composition support content to the user's terminal.

[0895] Step 6:

[0896] Users receive email composition support content via their devices, referring to specific expressions and writing styles.

[0897] Reports of power harassment / moral harassment

[0898] Step 1:

[0899] Users request to report instances of power harassment or moral harassment. Specifically, they send requests from their terminals to the server.

[0900] Step 2:

[0901] The server receives a request from the user. It parses the request and confirms that it is of type "check_abuse".

[0902] Step 3:

[0903] The server calls the moral harassment / power harassment detection module and analyzes the reported content.

[0904] Step 4:

[0905] The harassment / power harassment detection module evaluates the reported content and identifies the problem.

[0906] Step 5:

[0907] The server sends the evaluation results and necessary countermeasures to the user's terminal.

[0908] Step 6:

[0909] Users receive evaluation results via their devices and confirm necessary countermeasures.

[0910] In this way, the system operates by calling the appropriate module for each request, providing the user with the most suitable content, support, and problem-solving solutions.

[0911] (Example 1)

[0912] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0913] A system is needed to efficiently conduct new employee training, provide work support, and improve the work environment within companies. Traditional methods often lacked coherence and efficiency, as various content was generated and support provided individually. Furthermore, there is a need for faster early detection and response to power harassment and moral harassment. These challenges must be addressed in an integrated manner.

[0914] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0915] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, means for detecting power harassment and moral harassment, means for generating new employee training materials, and means for executing at least one of these means in response to a user request. This makes it possible to efficiently conduct new employee training and work support within a company, and to comprehensively support the improvement of the work environment.

[0916] A "generative artificial intelligence engine" refers to algorithms and software used to generate new content based on data.

[0917] "Email creation support content" refers to guidelines and templates that provide assistance with email content, layout, and expression.

[0918] "Support regarding how to ask questions" refers to advice on how to ask questions and how to conduct conversations in order to facilitate smooth communication.

[0919] "Methods for detecting power harassment and moral harassment" refer to algorithms and software used to detect and evaluate inappropriate behavior and remarks in the workplace.

[0920] "New employee training materials" refer to the information and educational content necessary for new employees to adapt to their work.

[0921] "Means of execution in response to user requests" refers to modules and algorithms that perform appropriate processing based on requests made by users to the system.

[0922] This invention is a system for efficiently conducting new employee training, providing work support, and improving the workplace environment within a company. The system includes a generative artificial intelligence engine, an email creation support module, a question-asking support module, and power harassment and moral harassment detection modules. These modules are executed in response to requests from the user.

[0923] The server first initializes the generative artificial intelligence engine (hereinafter referred to as the AI ​​engine) and sets up the HR-related information to be used within the company. This process utilizes cloud services (for example, Amazon Web Services or Microsoft Azure). With this setup complete, the entire system is ready.

[0924] When a user accesses the system and sends a specific request, the server receives the request and uses natural language processing (NLP) techniques to determine its content. Depending on the request content, various processes such as the following are executed:

[0925] Provision of training materials for new employees

[0926] When a user requests "new employee training materials," the server uses an AI engine to generate training content optimized for the user. This generation utilizes natural language generation models such as OpenAI's ChatGPT. The generated training content is sent to the user's device in PDF format. For example, the prompt might look like this:

[0927] "Please provide the basic materials for new employee training."

[0928] Providing email creation support

[0929] When a user sends a request saying "Please help me compose an email," the server activates the email composition support module and generates an effective email body, layout, and wording based on the user's request. The generated support content is sent to the user's terminal, and the user uses it as a reference to compose the email. The specific prompt text is as follows:

[0930] "Please create a sample sales email for new customers."

[0931] Reports of power harassment / moral harassment

[0932] When a user submits a request to "report harassment," the server activates a harassment detection module and analyzes the user's report in detail. Using techniques such as sentiment analysis and keyword matching, the severity of the report is assessed. As a result, appropriate countermeasures and action plans are proposed and sent to the user's terminal. An example of a prompt message is shown below.

[0933] "I want to report the power harassment I've experienced from a colleague."

[0934] This system allows users to effectively receive new employee training and the necessary support for their work. Furthermore, workplace harassment issues are quickly detected and addressed, leading to improvements in the work environment and overall increased work performance.

[0935] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0936] Provision of training materials for new employees

[0937] Step 1:

[0938] The user submits a request.

[0939] The user sends a request to the system using their terminal, stating "I would like training materials for new employees." The input is a request for "training materials for new employees," and the output is an initial request to the server. Specifically, the user accesses a web form, selects the required materials, and presses the request button.

[0940] Step 2:

[0941] The server receives and processes the request.

[0942] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses natural language processing (NLP) techniques to classify the request content as "new employee training material."

[0943] Step 3:

[0944] The server uses an AI engine to generate training content.

[0945] The server uses a generative artificial intelligence engine (AI engine) to generate optimized training content based on classified requests. The input consists of the request classification result and internal training data, while the output is the generated training content. Specifically, it uses models such as OpenAI's ChatGPT to generate text.

[0946] Step 4:

[0947] The server sends the generated content to the user's device.

[0948] The server sends the generated training content to the user's terminal. The input is the generated content data, and the output is the content sent to the user's terminal. Specifically, the generated training materials are converted to PDF format and sent to the user via email or a download link.

[0949] Step 5:

[0950] Users review training materials.

[0951] The user reviews the training materials received on their device and acquires the necessary knowledge and skills. The input is the PDF document received on the device, and the output is the knowledge and skills the user has acquired. Specifically, the user opens the PDF and reads its contents.

[0952] Providing email creation support

[0953] Step 1:

[0954] The user submits a request.

[0955] The user sends a request to the system using their terminal, stating "Please help me compose an email." The input is a request for "email composition assistance," and the output is an initial request to the server. Specifically, the user fills in the required information in the email composition assistance request form and submits it.

[0956] Step 2:

[0957] The server receives and processes the request.

[0958] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses NLP technology to classify the request content as "email creation support".

[0959] Step 3:

[0960] The server starts the email creation support module.

[0961] The server launches an email creation support module based on the classified request. The input is the request determination result, and the output is the process of generating the support content. Specifically, it reads past email data and standard template messages.

[0962] Step 4:

[0963] The server generates the support information and sends it to the user's terminal.

[0964] The server generates specific email composition advice and templates based on the user's request and sends them to the user's terminal. The input is the user's request and internal data, and the output is the generated support content. Specifically, it creates a template that includes an effective subject line, greeting, body, and closing, and sends it via email.

[0965] Step 5:

[0966] Users refer to the support information when composing an email.

[0967] The user composes the email while referring to the support information. The input is the support information sent from the server, and the output is the completed email. Specifically, the user uses the received template, edits the content as needed, and sends the email.

[0968] Reports of power harassment / moral harassment

[0969] Step 1:

[0970] The user submits a request.

[0971] The user sends a request to the system using their device, stating "I will report workplace harassment." The input is a "workplace harassment report" request, and the output is an initial request to the server. Specifically, the user enters details into the workplace harassment report form and presses the submit button.

[0972] Step 2:

[0973] The server receives and processes the request.

[0974] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses NLP technology to classify the request content as a "harassment report."

[0975] Step 3:

[0976] The server activates the harassment detection module.

[0977] The server activates the harassment detection module based on the classified request. The input is the request classification result, and the output is the analysis data from the detection module. Specifically, it prepares data for sentiment analysis of the reported text.

[0978] Step 4:

[0979] The server analyzes and evaluates the report content.

[0980] The server analyzes the reported content in detail and assesses the possibility of power harassment or moral harassment. The input is user-reported data, and the output is evaluation result data. Specifically, it uses sentiment analysis and keyword matching to assess the severity of the reported content.

[0981] Step 5:

[0982] The server sends the solution to the user's terminal.

[0983] The server proposes appropriate countermeasures based on the evaluation results and sends them to the user's terminal. The input is the evaluation result data, and the output is the proposed countermeasures. Specifically, it creates a document containing contact information for counseling and instructions for internal reporting, and sends it via email.

[0984] Step 6:

[0985] The user reviews and implements the countermeasures.

[0986] The user reviews the proposed countermeasures and implements them as needed. The input is the proposed countermeasure, and the output is the result of the action taken. Specifically, the user reports the problem to their supervisor or compliance officer according to the proposed reporting procedure.

[0987] Through the processing steps described above, the entire system effectively responds to the diverse needs of users, improving the work environment and supporting business operations.

[0988] (Application Example 1)

[0989] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0990] In the food delivery industry, a challenge exists in that new delivery drivers have difficulty acquiring the necessary knowledge and skills effectively and quickly, and while appropriate communication with customers is required, sufficient training and support are not provided. Furthermore, there is a need to address power harassment and moral harassment during deliveries and within the company early on and improve the workplace environment.

[0991] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0992] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for the user, means for providing training materials specifically for food delivery, means for guiding appropriate communication methods at delivery destinations, means for detecting power harassment and moral harassment, and means for executing at least one of these means in response to a request from the user. This system enables new delivery personnel to acquire the necessary knowledge in a short period of time, communicate appropriately with customers, and detect and respond to harassment early.

[0993] A "generative artificial intelligence engine" is an engine that generates information based on user requests and provides appropriate content and support.

[0994] "Email creation support content" refers to content that includes specific expression methods and guidelines for users to create effective emails.

[0995] "Support regarding how to ask questions" refers to providing guidance and instruction on how to communicate effectively.

[0996] "Means for detecting power harassment and moral harassment" refers to methods for identifying harassment based on user reports and suggesting countermeasures.

[0997] "Training content in the food delivery industry" refers to training materials and programs designed to equip new delivery personnel with the knowledge and skills necessary for food delivery operations.

[0998] "New employee training content" refers to materials and content designed to help new employees acquire the basic knowledge and skills necessary to perform their duties.

[0999] "Training materials" are documents that describe specific instructions and procedures for acquiring particular tasks or skills.

[1000] "Appropriate communication methods at delivery destinations" refers to advice and guidelines for delivery personnel to interact smoothly with customers.

[1001] "Means of execution in response to user requests" refers to means of performing appropriate processing or generating content in response to user requests.

[1002] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a training content generation module specifically for the food delivery industry, and a harassment detection module. This system enables training of new delivery personnel, support for communication with customers, and early detection and response to harassment.

[1003] System Configuration

[1004] The server uses a generative artificial intelligence engine to perform the following actions:

[1005] 1. Generating training content for new employees based on user requests.

[1006] 2. Provision of training materials specifically for food delivery.

[1007] 3. A guide to appropriate communication methods at the delivery destination.

[1008] 4. Detection of power harassment and moral harassment, and presentation of countermeasures.

[1009] Program Processing Description

[1010] This system initializes a generative AI engine using OpenAI's GPT model. When the server receives a request from a user, it starts the appropriate module according to the request and performs the necessary generation, support, and detection processes.

[1011] For example, if a user requests training materials for new employees, the server uses an AI engine to generate training content optimized for that individual user and provides it to the user. Similarly, if a user asks about how to communicate with customers, the server utilizes a question-answering module to generate appropriate guidance and provide it to the user.

[1012] When harassment is reported, the detection module evaluates the situation based on detailed information and suggests countermeasures as needed. This allows users to address problems early and improve the workplace environment.

[1013] Hardware and software used

[1014] Hardware: Typical server

[1015] Software: OpenAI GPT model, Python

[1016] Examples of specific cases and prompt statements

[1017] For example, to ensure a new delivery driver receives efficient training, the following prompt can be used:

[1018] Prompt message:

[1019] "Please generate training content for new food delivery drivers. The content should include how to find efficient delivery routes, how to communicate effectively with customers, and tips for safe driving."

[1020] Example of generated content:

[1021] Training program for new food delivery drivers:

[1022] 1. How to find an efficient delivery route

[1023] How to use map apps

[1024] How to avoid rush hour

[1025] 2. Effective methods of communication with customers

[1026] Polite language

[1027] Troubleshooting Basics

[1028] 3. Points for safe driving

[1029] Stopping and braking

[1030] Maintain an appropriate speed

[1031] In this way, new delivery drivers can acquire the necessary knowledge in a short period of time and be well-prepared before starting work.

[1032] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1033] Step 1:

[1034] Users submit requests from their devices. These requests may include new employee training materials, customer communication guides, and harassment reports.

[1035] Input: User request (e.g., Request for new employee training materials)

[1036] Output: Sending a request to the server

[1037] Specific action: The user uses the terminal interface to type "Generate new employee training materials" and clicks the submit button.

[1038] Step 2:

[1039] The server initializes the generative artificial intelligence engine and receives a request from the user. It then selects the appropriate module based on the content of the request.

[1040] Input: User Request

[1041] Output: Selection and startup of the appropriate module

[1042] Specific operation: The server parses the request content and, if it is for new employee training materials, launches the training content generation module.

[1043] Step 3:

[1044] The generative artificial intelligence engine creates prompt sentences for generating training content and sends them to the AI ​​engine.

[1045] Input: Request details, prompt text to be generated

[1046] Output: Prompt statement generation

[1047] Specific operation: The server generates a prompt message saying "Generate training content for new food delivery drivers" and sends it to the AI ​​engine.

[1048] Step 4:

[1049] The AI ​​engine processes the prompt text and generates training content for new employees.

[1050] Input: Prompt message

[1051] Output: Generated new employee training content

[1052] Specific operation: The AI ​​engine uses the OpenAI GPT model to generate training content based on prompt sentences.

[1053] Step 5:

[1054] The server sends the generated training content back to the user.

[1055] Input: Generated training content

[1056] Output: Return to user

[1057] Specific operation: The server sends the generated training content (e.g., how to find efficient routes, tips for safe driving, etc.) to the user's device.

[1058] Step 6:

[1059] Users receive and review training content generated on their devices. If necessary, they can save or print the training content.

[1060] Input: Training content returned from the server

[1061] Output: Display and save training content

[1062] Specific actions: The user reviews the training content received on their device and, if necessary, saves it to the device or prints it out to use as reference material.

[1063] Through the processing steps described above, this system can efficiently provide training content in response to user requests and support the skill development of new delivery drivers in the food delivery industry.

[1064] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1065] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a power harassment and moral harassment detection module, and an emotion engine. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment. Furthermore, by providing support that takes the user's emotions into consideration, it realizes more appropriate and effective support.

[1066] Overall system configuration

[1067] The server comprises a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules operate to generate and provide appropriate support content in response to user requests.

[1068] Program execution flow (overview)

[1069] 1. Server initialization

[1070] The server initializes each engine and module upon startup. In particular, the generative AI engine and the emotion engine load detailed settings and training data.

[1071] 2. Receiving requests from users

[1072] Users send requests to the server through their devices. These requests include new employee training materials, email composition assistance, assistance with asking questions, and reports of power harassment / moral harassment.

[1073] 3. Emotion recognition by an emotion engine

[1074] The server uses an emotion engine to analyze user emotion data (e.g., facial expressions, voice tone, etc.) sent with the request. The analysis results are then fed back to other modules.

[1075] 4. Processing of each module

[1076] Based on the request type and sentiment analysis results, the appropriate module (e.g., generative AI engine, email creation support module, question-asking support module, power harassment / moral harassment detection module) is selected. This allows each module to generate the most suitable support content.

[1077] 5. Responding to the user

[1078] The generated support content and evaluation results are returned to the user. If the results of the emotion engine are incorporated, the content will be optimized for the user's situation and emotions.

[1079] Specific example

[1080] Provision of training materials for new employees

[1081] A user sends a request for new employee training materials to the server. Along with this request, the user's facial expression data and voice are also sent. The server analyzes this data using an emotion engine to determine the user's emotional state. For example, if the user is nervous, training materials written in a relaxing tone are generated. A generative artificial intelligence engine generates specific training content, and the server provides these materials to the user.

[1082] Providing email creation support

[1083] The user submits a request for email composition assistance. The emotion engine generates an email composition guide that includes advice to alleviate the user's feelings of anxiety or impatience if they are experiencing such emotions. The email composition assistance module provides specific expressions and templates, which the server returns to the user.

[1084] Reports of power harassment / moral harassment

[1085] Users report instances of power harassment or moral harassment. The emotion engine evaluates the user's mental state, and the server uses this information to perform a detailed analysis using a moral harassment / power harassment detection module. Along with the evaluation results, appropriate counseling and countermeasures are provided to the user.

[1086] This system allows users to efficiently receive the support and information they need, and enables appropriate responses based on their emotional state. By combining it with an emotional engine, a more human-centered approach is added, which can increase user satisfaction and effectiveness.

[1087] The following describes the processing flow.

[1088] Provision of training materials for new employees

[1089] Step 1:

[1090] A user issues a request for new employee training materials. The user sends the request from their terminal to the server.

[1091] Step 2:

[1092] The server receives the request. It verifies that the request type is "training".

[1093] Step 3:

[1094] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[1095] Step 4:

[1096] The server uses an emotion engine to analyze the user's emotional data and determine the user's emotional state.

[1097] Step 5:

[1098] The server invokes a generative artificial intelligence engine to generate optimal training content based on user information and emotional state.

[1099] Step 6:

[1100] A generative artificial intelligence engine generates training content and returns it to the server.

[1101] Step 7:

[1102] The server sends the generated training content to the user's terminal. The content is structured with the user's emotional state in mind.

[1103] Step 8:

[1104] Users receive training content via their devices and review its contents.

[1105] Providing email creation support

[1106] Step 1:

[1107] The user issues a request for email creation assistance. The request is sent from the user's terminal to the server.

[1108] Step 2:

[1109] The server receives the request. It verifies that the request type is "email_help".

[1110] Step 3:

[1111] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[1112] Step 4:

[1113] The server uses an emotion engine to analyze the user's emotional data and determine the user's emotional state.

[1114] Step 5:

[1115] The server calls an email composition support module and generates optimal email composition support content based on user information and emotional state.

[1116] Step 6:

[1117] The email composition support module generates the email composition support content and returns it to the server.

[1118] Step 7:

[1119] The server sends the generated email composition support content to the user's terminal. The content is adjusted to take the user's emotional state into consideration.

[1120] Step 8:

[1121] Users receive email composition support content via their devices and check specific expressions and writing styles.

[1122] Reports of power harassment / moral harassment

[1123] Step 1:

[1124] A user requests to report instances of power harassment or moral harassment. The user sends the request from their terminal to the server.

[1125] Step 2:

[1126] The server receives the request. It verifies that the request type is "check_abuse".

[1127] Step 3:

[1128] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[1129] Step 4:

[1130] The server uses an emotion engine to analyze the user's emotional data and evaluate their mental state.

[1131] Step 5:

[1132] The server invokes a power harassment / moral harassment detection module and evaluates the reported content.

[1133] Step 6:

[1134] The power harassment / moral harassment detection module analyzes the reported content and identifies the problem.

[1135] Step 7:

[1136] The server generates the analysis results and necessary countermeasures and sends them to the user's terminal.

[1137] Step 8:

[1138] Users receive evaluation results and countermeasures via their devices and review the countermeasures as needed.

[1139] Through the above processes, users can efficiently receive the necessary support and information, enabling them to respond appropriately. Furthermore, by combining this with an emotion engine, appropriate support content is provided according to the user's emotional state.

[1140] (Example 2)

[1141] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1142] In conventional systems, AI-powered email creation support, new employee training support, and detection and response to power harassment and moral harassment were handled separately, making centralized management and response difficult. Furthermore, appropriate support that considered the user's emotional state was not adequately provided. Therefore, there is a need to improve the user experience and establish an efficient support system.

[1143] In Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, means for detecting power harassment and moral harassment, means for analyzing the user's emotional data and feeding the results back to other modules for emotion recognition, and means for executing at least one of these means in response to a request from the user. This makes it possible to efficiently and centrally provide support that meets the diverse needs of the user. Furthermore, it is possible to provide appropriate content that takes into account the user's emotional state, and an improvement in the user experience can be expected.

[1144] A "generative artificial intelligence engine" refers to artificial intelligence that automatically generates various types of support content based on user requests.

[1145] "Email creation support content" refers to the guidelines, templates, and specific examples of emails that users need when creating emails for business or personal use.

[1146] "Support regarding how to ask questions" refers to guidelines and sample sentences provided to help users ask appropriate questions and phrases depending on the situation.

[1147] "Methods for detecting power harassment and moral harassment" refers to technologies that analyze user reports and collected data to identify the presence of power harassment and moral harassment.

[1148] "User sentiment data" refers to information about emotions extracted from the user's facial expressions, voice tone, text content, etc.

[1149] "Emotion recognition means" refers to technology that analyzes collected user emotion data and feeds the results back to various support modules.

[1150] "Means of execution in response to user requests" refers to the technology for selecting the appropriate module and executing its function in response to user requests.

[1151] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a power harassment and moral harassment detection module, and an emotion engine. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment. Furthermore, by providing support that takes the user's emotions into consideration, it realizes more appropriate and effective support.

[1152] Overall system configuration

[1153] The server comprises a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules operate to generate and provide appropriate support content in response to user requests.

[1154] Program execution flow (overview)

[1155] The server initializes each engine and module upon startup. In particular, the generative AI engine and the emotion engine load detailed settings and training data.

[1156] Users send requests to the server through their devices. These requests include new employee training materials, email composition assistance, assistance with asking questions, and reports of power harassment / moral harassment. The server analyzes the user's emotional data (e.g., facial expressions, voice tone, etc.) sent with the requests using an emotion engine. The analysis results are then fed back to other modules.

[1157] Based on the request type and sentiment analysis results, the appropriate module (e.g., generative AI engine, email creation support module, question-asking support module, power harassment / moral harassment detection module) is selected. Each module then generates the optimal support content. The generated support content and evaluation results are returned to the user. If the sentiment engine results are incorporated, the content will be optimized for the user's situation and emotions.

[1158] Specific example

[1159] Provision of training materials for new employees

[1160] A user sends a request for new employee training materials to the server. Along with this request, the user's facial expression data and voice are also sent. The server analyzes this data using an emotion engine to determine the user's emotional state. For example, if the user is nervous, training materials written in a relaxing tone are generated. A generative artificial intelligence engine generates specific training content, and the server provides these materials to the user.

[1161] Specific example:

[1162] A user requests, "Please tell me about your new employee training methods."

[1163] Example prompt: "Please provide relaxing training materials for new employees who are feeling nervous."

[1164] Providing email creation support

[1165] The user submits a request for email composition assistance. The emotion engine generates an email composition guide that includes advice to alleviate the user's feelings of anxiety or impatience if they are experiencing such emotions. The email composition assistance module provides specific expressions and templates, which the server returns to the user.

[1166] Specific example:

[1167] The user requests help creating a formal work email.

[1168] Example prompt: "Please provide a formal email template for users who are feeling anxious."

[1169] Reports of power harassment / moral harassment

[1170] Users report instances of power harassment or moral harassment. The emotion engine evaluates the user's mental state, and the server uses this information to perform a detailed analysis using a moral harassment / power harassment detection module. Along with the evaluation results, appropriate counseling and countermeasures are provided to the user.

[1171] Specific example:

[1172] A user requests to report workplace harassment from their superior.

[1173] Example prompt: "Please provide appropriate counseling for users who are experiencing harassment."

[1174] Hardware and software to be used

[1175] The server runs on a computer equipped with a high-performance processor and large memory capacity. Generative artificial intelligence engines and emotion engines are implemented using machine learning libraries such as Python, TensorFlow, and PyTorch. RDBMS such as MySQL and PostgreSQL are used for the database. User terminals include personal computers, smartphones, and tablets, which communicate with the server via the internet.

[1176] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1177] Step 1: Server Initialization

[1178] The server initializes each engine and module upon startup. Specifically, it initializes the generative AI engine and the emotion engine, loading their respective configuration parameters and training data. It reads configuration files (e.g., config.yaml) and sets the connection settings and initial values ​​for each module. The input for this step is the configuration files and training data, and the output is a state where each module is ready to operate.

[1179] Step 2: Receiving requests from users

[1180] The user sends a request to the server through their device. For example, if a user sends a request for new employee training materials, the request data will include text-based instructions and user sentiment data (facial expressions, voice tone, etc.). The server receives the HTTP request and parses the data in JSON format. The input to this step is the user's request data, and the output is the parsed request content and sentiment data.

[1181] Step 3: Emotion recognition by the emotion engine

[1182] The server analyzes the emotion data received from the user using an emotion engine. Facial expression data is analyzed using a facial recognition algorithm, and voice data is identified using a voice analysis algorithm to determine the emotional state. For example, if the user is tense, it will be tagged as tense. The input for this step is the user's facial expression data and voice data, and the output is the identified emotional state (e.g., tense, relaxed).

[1183] Step 4: Processing each module

[1184] The server selects the appropriate module based on the request type and sentiment analysis results. For example, if a user requests email composition assistance and sentiment data indicates anxiety, the server generates assistance content combining the email composition assistance module and the sentiment recognition results. The generative AI engine receives a prompt to generate specific content and produces the optimal response. The input for this step is the request type and sentiment analysis results, and the output is the generated assistance content.

[1185] Step 5: Responding to the user

[1186] The server returns the generated support content and evaluation results to the user. The generated content may include, for example, training materials for new employees, email creation guides, or appropriate counseling content for reports of power harassment / moral harassment. This step also incorporates the results of the emotion engine, ensuring that content is tailored to the user's situation and emotions. The server generates response data in JSON format and sends it to the user's terminal as an HTTP response. The input for this step is the generated support content and emotion analysis results, while the output is the response message to the user.

[1187] (Application Example 2)

[1188] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1189] In modern factories and manufacturing sites, the proper and efficient implementation of new employee training and problem reporting, as well as the early detection and response to harassment, are critical issues. Furthermore, customized support tailored to the user's emotional state is required, but conventional systems have struggled to integrate these aspects. This invention aims to solve these problems and improve the working environment and increase efficiency.

[1190] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1191] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for analyzing user emotion data and generating support content corresponding to the emotional state, and means for displaying and communicating the support content to the user via voice. This enables efficient training of new employees, early detection and countermeasures for problem reporting and harassment, and customized support tailored to the user's emotional state.

[1192] A "generative artificial intelligence engine" is an artificial intelligence that generates appropriate content from text and data. This makes it possible to answer complex questions and create appropriate emails, among other things.

[1193] "Email creation support content" refers to support information and templates provided to help users create emails efficiently and appropriately.

[1194] "Support regarding how to ask questions" refers to support information that provides appropriate expressions and phrases to effectively ask questions or make requests.

[1195] "Methods for detecting power harassment and moral harassment" refer to functions that analyze user input and emotional data to determine whether or not harassment has occurred.

[1196] "User emotional data" refers to data about a user's emotional state, obtained from things like their facial expressions and voice tone.

[1197] "Support tailored to emotional state" refers to support information provided in the format and tone most appropriate to the user, based on the user's emotional data.

[1198] "New employee training content" refers to educational materials and guidelines designed to support the learning of new employees in factories, manufacturing sites, and other similar settings.

[1199] "Counseling" refers to the act or process of providing mental support to victims of power harassment or moral harassment.

[1200] "Means for communicating support content visually and audibly" refers to devices such as displays and speakers that provide the generated support content to the user visually and audibly.

[1201] The present invention relates to a robotic system for supporting new employee training, which is installed in factory robots and includes a generative artificial intelligence engine, an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. Embodiments of this system are described in detail below.

[1202] Hardware and software to be used

[1203] 1. Touchscreen interface

[1204] Example: Use a Raspberry Pi touchscreen.

[1205] 2. Speech Recognition and Synthesis Module

[1206] Example: Use Google Cloud Speech-to-Text or Text-to-Speech.

[1207] 3. Emotion Recognition Camera

[1208] Example: Use Intel RealSense.

[1209] System program description

[1210] The server includes a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules generate and provide appropriate support content in response to requests from users (factory workers).

[1211] Specifically, the server performs data processing and calculations using the following methods:

[1212] 1. Generative artificial intelligence engine

[1213] When a user requests training materials for new employees, a generative artificial intelligence engine (e.g., GPT-4) generates the training content.

[1214] The generated training materials will include factory safety guidelines, operating procedures for key equipment, and disaster response procedures.

[1215] 2. Emotional Engine

[1216] An emotion engine (e.g., Affectiva Emotion AI) analyzes data from the user's facial expressions and voice tone to evaluate the user's emotional state.

[1217] For example, if a user is feeling anxious, provide training materials that include relaxation techniques.

[1218] 3. Provided via display and audio.

[1219] Actual display and audio output are performed through the robot's touchscreen and speakers.

[1220] The generated training materials and support content are displayed and communicated via audio in a way that is optimized for the user's emotional state.

[1221] Specific examples

[1222] For example, when a new factory worker requests training materials, the following prompt message is input to the generative artificial intelligence engine:

[1223] Please generate training materials for new employees. Include the following: factory safety guidelines, operating procedures for key equipment, and disaster response procedures. Users are nervous.

[1224] The emotion engine analyzes the user's facial expressions and voice tone at that time and obtains the following analysis data:

[1225] The system analyzes facial expression data and voice tone in real time to evaluate the user's emotional state. The analysis result is [tension, anxiety].

[1226] The generative artificial intelligence engine generates the following based on this data:

[1227] Generate training materials that include relaxation techniques. The proposed training materials will include calm language and relaxation methods.

[1228] The robot then displays the training materials via a touchscreen and uses voice prompts to encourage relaxation.

[1229] This system not only enables efficient training of new employees, but also facilitates problem reporting, early detection and countermeasures against harassment, and customized support tailored to the user's emotional state.

[1230] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1231] Step 1:

[1232] Server initialization and configuration

[1233] Upon startup, the server initializes each engine (generative AI engine, emotion engine) and module (email creation support module, question-asking support module, power harassment / moral harassment detection module) and performs the necessary configurations. This includes loading existing training data and the configuration values ​​for each module. This prepares the server to respond quickly to user requests.

[1234] Step 2:

[1235] Received a request from the user.

[1236] Through a terminal, users (factory workers) send requests for new employee training materials, problem reports, and other information to the server. These requests also include the user's facial expression data and voice. The server receives this information and begins a process to analyze the relevant data. Input data includes the request content, facial expression data, and voice data. Output data provides initial analysis results to be passed on to the next analysis step.

[1237] Step 3:

[1238] Analysis of emotional data using an emotion engine

[1239] The server inputs facial expression data and voice data received from the user into an emotion engine (e.g., Affectiva Emotion AI) to analyze the user's emotional state. The emotion engine converts facial expressions and voice tone into digital data in real time and evaluates emotional states such as tension and anxiety. The input data consists of facial expression data and voice data, and the output data is the evaluation result of the emotional state. The server retrieves this evaluation result and proceeds to the next processing step.

[1240] Step 4:

[1241] Content generation using generative artificial intelligence engines

[1242] Based on the request content and the analysis results of the emotion engine, the server inputs a prompt message into a generative artificial intelligence engine (e.g., GPT-4) to generate corresponding support content (new employee training materials, email content, etc.). For example, it inputs a prompt message like the following:

[1243] Please generate training materials for new employees. Include the following: factory safety guidelines, operating procedures for key equipment, and disaster response procedures. Users are nervous.

[1244] The generative artificial intelligence engine analyzes this prompt and generates the necessary content. The input data consists of the request and the sentiment analysis results, while the output data is the generated training material.

[1245] Step 5:

[1246] Provision of generated support content

[1247] The server displays and delivers supportive content obtained from a generative artificial intelligence engine to the user in an appropriate format. Specifically, it displays text using the robot's touchscreen and provides explanations via a speaker. The input data is the generated supportive content, and the output data is the visual and auditory information provided to the user. This allows the user to efficiently obtain the necessary information.

[1248] Step 6:

[1249] Gathering feedback and optimizing the system

[1250] Users provide feedback on the content they are given. The server collects this feedback information and uses it to generate future support content and perform sentiment analysis. The feedback includes aspects such as understanding and satisfaction with the content. The input data is user feedback information, and the output data is optimized system settings. This allows the system to continuously improve.

[1251] The above describes the processing steps of this system.

[1252] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1253] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1254] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1255] [Fourth Embodiment]

[1256] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1257] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1258] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1259] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1260] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1261] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1262] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1263] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1264] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1265] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1266] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1267] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1268] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1269] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment.

[1270] The server first initializes its generative artificial intelligence engine (AI engine) and sets up HR-related information. Upon receiving a request from a user, the server determines the type of request and executes the appropriate processing accordingly.

[1271] For example, if a user requests training materials for new employees, the server uses an AI engine to generate the most suitable training content for the user and provides it to them. Similarly, if a user requests assistance with email composition, the server uses an email composition assistance module to guide the user on how to compose effective emails.

[1272] If assistance is needed regarding how to ask questions, the server utilizes a question-asking support module to guide users in communicating effectively. Furthermore, if a user reports power harassment or moral harassment, the server evaluates it using a detection module and suggests appropriate countermeasures.

[1273] Specific example

[1274] Provision of training materials for new employees

[1275] When a user sends a request to the server for new employee training materials, the server uses an AI engine to generate training content optimized for that individual user and returns it to the user. The user then reviews the received training materials and acquires the necessary knowledge and skills.

[1276] Providing email creation support

[1277] If a user needs assistance composing an email, the server uses an email composition support module to generate and provide support content that includes specific phrasing and writing style guidance. The user can then use this content to create an effective email.

[1278] Reports of power harassment / moral harassment

[1279] When a user reports harassment (power harassment or moral harassment), the server uses a harassment detection module to evaluate the report based on more detailed information. The server analyzes the report and proposes countermeasures as needed. As a result, users can address problems early, leading to improvements in the workplace environment.

[1280] This system allows users to efficiently acquire necessary knowledge and create a safe and secure work environment. The entire system works together to support users, thereby improving IT literacy and ensuring workplace safety.

[1281] The following describes the processing flow.

[1282] Provision of training materials for new employees

[1283] Step 1:

[1284] A user issues a request for new employee training materials. Specifically, the user sends a request from their terminal to the server.

[1285] Step 2:

[1286] The server receives a request from the user. It analyzes the content of the request and confirms that it is a request of type "training".

[1287] Step 3:

[1288] The server invokes a generative artificial intelligence engine (AI engine) to generate training content based on user information.

[1289] Step 4:

[1290] The AI ​​engine generates the training content the user needs and returns it to the server.

[1291] Step 5:

[1292] The server sends the generated training content to the user's terminal.

[1293] Step 6:

[1294] Users receive training content via their devices and review its contents.

[1295] Providing email creation support

[1296] Step 1:

[1297] The user issues a request for email creation assistance. Specifically, the user sends a request from their terminal to the server.

[1298] Step 2:

[1299] The server receives a request from the user. It parses the request and confirms that it is of type "email_help".

[1300] Step 3:

[1301] The server calls the email creation support module and generates email creation support content based on user information.

[1302] Step 4:

[1303] The email creation support module generates specific email creation advice and templates requested by the user and returns them to the server.

[1304] Step 5:

[1305] The server sends the generated email composition support content to the user's terminal.

[1306] Step 6:

[1307] Users receive email composition support content via their devices, referring to specific expressions and writing styles.

[1308] Reports of power harassment / moral harassment

[1309] Step 1:

[1310] Users request to report instances of power harassment or moral harassment. Specifically, they send requests from their terminals to the server.

[1311] Step 2:

[1312] The server receives a request from the user. It parses the request and confirms that it is of type "check_abuse".

[1313] Step 3:

[1314] The server calls the moral harassment / power harassment detection module and analyzes the reported content.

[1315] Step 4:

[1316] The harassment / power harassment detection module evaluates the reported content and identifies the problem.

[1317] Step 5:

[1318] The server sends the evaluation results and necessary countermeasures to the user's terminal.

[1319] Step 6:

[1320] Users receive evaluation results via their devices and confirm necessary countermeasures.

[1321] In this way, the system operates by calling the appropriate module for each request, providing the user with the most suitable content, support, and problem-solving solutions.

[1322] (Example 1)

[1323] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1324] A system is needed to efficiently conduct new employee training, provide work support, and improve the work environment within companies. Traditional methods often lacked coherence and efficiency, as various content was generated and support provided individually. Furthermore, there is a need for faster early detection and response to power harassment and moral harassment. These challenges must be addressed in an integrated manner.

[1325] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1326] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, means for detecting power harassment and moral harassment, means for generating new employee training materials, and means for executing at least one of these means in response to a user request. This makes it possible to efficiently conduct new employee training and work support within a company, and to comprehensively support the improvement of the work environment.

[1327] A "generative artificial intelligence engine" refers to algorithms and software used to generate new content based on data.

[1328] "Email creation support content" refers to guidelines and templates that provide assistance with email content, layout, and expression.

[1329] "Support regarding how to ask questions" refers to advice on how to ask questions and how to conduct conversations in order to facilitate smooth communication.

[1330] "Methods for detecting power harassment and moral harassment" refer to algorithms and software used to detect and evaluate inappropriate behavior and remarks in the workplace.

[1331] "New employee training materials" refer to the information and educational content necessary for new employees to adapt to their work.

[1332] "Means of execution in response to user requests" refers to modules and algorithms that perform appropriate processing based on requests made by users to the system.

[1333] This invention is a system for efficiently conducting new employee training, providing work support, and improving the workplace environment within a company. The system includes a generative artificial intelligence engine, an email creation support module, a question-asking support module, and power harassment and moral harassment detection modules. These modules are executed in response to requests from the user.

[1334] The server first initializes the generative artificial intelligence engine (hereinafter referred to as the AI ​​engine) and sets up the HR-related information to be used within the company. This process utilizes cloud services (for example, Amazon Web Services or Microsoft Azure). With this setup complete, the entire system is ready.

[1335] When a user accesses the system and sends a specific request, the server receives the request and uses natural language processing (NLP) techniques to determine its content. Depending on the request content, various processes such as the following are executed:

[1336] Provision of training materials for new employees

[1337] When a user requests "new employee training materials," the server uses an AI engine to generate training content optimized for the user. This generation utilizes natural language generation models such as OpenAI's ChatGPT. The generated training content is sent to the user's device in PDF format. For example, the prompt might look like this:

[1338] "Please provide the basic materials for new employee training."

[1339] Providing email creation support

[1340] When a user sends a request saying "Please help me compose an email," the server activates the email composition support module and generates an effective email body, layout, and wording based on the user's request. The generated support content is sent to the user's terminal, and the user uses it as a reference to compose the email. The specific prompt text is as follows:

[1341] "Please create a sample sales email for new customers."

[1342] Reports of power harassment / moral harassment

[1343] When a user submits a request to "report harassment," the server activates a harassment detection module and analyzes the user's report in detail. Using techniques such as sentiment analysis and keyword matching, the severity of the report is assessed. As a result, appropriate countermeasures and action plans are proposed and sent to the user's terminal. An example of a prompt message is shown below.

[1344] "I want to report the power harassment I've experienced from a colleague."

[1345] This system allows users to effectively receive new employee training and the necessary support for their work. Furthermore, workplace harassment issues are quickly detected and addressed, leading to improvements in the work environment and overall increased work performance.

[1346] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1347] Provision of training materials for new employees

[1348] Step 1:

[1349] The user submits a request.

[1350] The user sends a request to the system using their terminal, stating "I would like training materials for new employees." The input is a request for "training materials for new employees," and the output is an initial request to the server. Specifically, the user accesses a web form, selects the required materials, and presses the request button.

[1351] Step 2:

[1352] The server receives and processes the request.

[1353] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses natural language processing (NLP) techniques to classify the request content as "new employee training material."

[1354] Step 3:

[1355] The server uses an AI engine to generate training content.

[1356] The server uses a generative artificial intelligence engine (AI engine) to generate optimized training content based on classified requests. The input consists of the request classification result and internal training data, while the output is the generated training content. Specifically, it uses models such as OpenAI's ChatGPT to generate text.

[1357] Step 4:

[1358] The server sends the generated content to the user's device.

[1359] The server sends the generated training content to the user's terminal. The input is the generated content data, and the output is the content sent to the user's terminal. Specifically, the generated training materials are converted to PDF format and sent to the user via email or a download link.

[1360] Step 5:

[1361] Users review training materials.

[1362] The user reviews the training materials received on their device and acquires the necessary knowledge and skills. The input is the PDF document received on the device, and the output is the knowledge and skills the user has acquired. Specifically, the user opens the PDF and reads its contents.

[1363] Providing email creation support

[1364] Step 1:

[1365] The user submits a request.

[1366] The user sends a request to the system using their terminal, stating "Please help me compose an email." The input is a request for "email composition assistance," and the output is an initial request to the server. Specifically, the user fills in the required information in the email composition assistance request form and submits it.

[1367] Step 2:

[1368] The server receives and processes the request.

[1369] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses NLP technology to classify the request content as "email creation support".

[1370] Step 3:

[1371] The server starts the email creation support module.

[1372] The server launches an email creation support module based on the classified request. The input is the request determination result, and the output is the process of generating the support content. Specifically, it reads past email data and standard template messages.

[1373] Step 4:

[1374] The server generates the support information and sends it to the user's terminal.

[1375] The server generates specific email composition advice and templates based on the user's request and sends them to the user's terminal. The input is the user's request and internal data, and the output is the generated support content. Specifically, it creates a template that includes an effective subject line, greeting, body, and closing, and sends it via email.

[1376] Step 5:

[1377] Users refer to the support information when composing an email.

[1378] The user composes the email while referring to the support information. The input is the support information sent from the server, and the output is the completed email. Specifically, the user uses the received template, edits the content as needed, and sends the email.

[1379] Reports of power harassment / moral harassment

[1380] Step 1:

[1381] The user submits a request.

[1382] The user sends a request to the system using their device, stating "I will report workplace harassment." The input is a "workplace harassment report" request, and the output is an initial request to the server. Specifically, the user enters details into the workplace harassment report form and presses the submit button.

[1383] Step 2:

[1384] The server receives and processes the request.

[1385] The server analyzes the request received from the user and determines its content. The input is the user's request data, and the output is the data of the determination result. Specifically, the server uses NLP technology to classify the request content as a "harassment report."

[1386] Step 3:

[1387] The server activates the harassment detection module.

[1388] The server activates the harassment detection module based on the classified request. The input is the request classification result, and the output is the analysis data from the detection module. Specifically, it prepares data for sentiment analysis of the reported text.

[1389] Step 4:

[1390] The server analyzes and evaluates the report content.

[1391] The server analyzes the reported content in detail and assesses the possibility of power harassment or moral harassment. The input is user-reported data, and the output is evaluation result data. Specifically, it uses sentiment analysis and keyword matching to assess the severity of the reported content.

[1392] Step 5:

[1393] The server sends the solution to the user's terminal.

[1394] The server proposes appropriate countermeasures based on the evaluation results and sends them to the user's terminal. The input is the evaluation result data, and the output is the proposed countermeasures. Specifically, it creates a document containing contact information for counseling and instructions for internal reporting, and sends it via email.

[1395] Step 6:

[1396] The user reviews and implements the countermeasures.

[1397] The user reviews the proposed countermeasures and implements them as needed. The input is the proposed countermeasure, and the output is the result of the action taken. Specifically, the user reports the problem to their supervisor or compliance officer according to the proposed reporting procedure.

[1398] Through the processing steps described above, the entire system effectively responds to the diverse needs of users, improving the work environment and supporting business operations.

[1399] (Application Example 1)

[1400] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1401] In the food delivery industry, a challenge exists in that new delivery drivers have difficulty acquiring the necessary knowledge and skills effectively and quickly, and while appropriate communication with customers is required, sufficient training and support are not provided. Furthermore, there is a need to address power harassment and moral harassment during deliveries and within the company early on and improve the workplace environment.

[1402] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1403] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for the user, means for providing training materials specifically for food delivery, means for guiding appropriate communication methods at delivery destinations, means for detecting power harassment and moral harassment, and means for executing at least one of these means in response to a request from the user. This system enables new delivery personnel to acquire the necessary knowledge in a short period of time, communicate appropriately with customers, and detect and respond to harassment early.

[1404] A "generative artificial intelligence engine" is an engine that generates information based on user requests and provides appropriate content and support.

[1405] "Email creation support content" refers to content that includes specific expression methods and guidelines for users to create effective emails.

[1406] "Support regarding how to ask questions" refers to providing guidance and instruction on how to communicate effectively.

[1407] "Means for detecting power harassment and moral harassment" refers to methods for identifying harassment based on user reports and suggesting countermeasures.

[1408] "Training content in the food delivery industry" refers to training materials and programs designed to equip new delivery personnel with the knowledge and skills necessary for food delivery operations.

[1409] "New employee training content" refers to materials and content designed to help new employees acquire the basic knowledge and skills necessary to perform their duties.

[1410] "Training materials" are documents that describe specific instructions and procedures for acquiring particular tasks or skills.

[1411] "Appropriate communication methods at delivery destinations" refers to advice and guidelines for delivery personnel to interact smoothly with customers.

[1412] "Means of execution in response to user requests" refers to means of performing appropriate processing or generating content in response to user requests.

[1413] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a training content generation module specifically for the food delivery industry, and a harassment detection module. This system enables training of new delivery personnel, support for communication with customers, and early detection and response to harassment.

[1414] System Configuration

[1415] The server uses a generative artificial intelligence engine to perform the following actions:

[1416] 1. Generating training content for new employees based on user requests.

[1417] 2. Provision of training materials specifically for food delivery.

[1418] 3. A guide to appropriate communication methods at the delivery destination.

[1419] 4. Detection of power harassment and moral harassment, and presentation of countermeasures.

[1420] Program Processing Description

[1421] This system initializes a generative AI engine using OpenAI's GPT model. When the server receives a request from a user, it starts the appropriate module according to the request and performs the necessary generation, support, and detection processes.

[1422] For example, if a user requests training materials for new employees, the server uses an AI engine to generate training content optimized for that individual user and provides it to the user. Similarly, if a user asks about how to communicate with customers, the server utilizes a question-answering module to generate appropriate guidance and provide it to the user.

[1423] When harassment is reported, the detection module evaluates the situation based on detailed information and suggests countermeasures as needed. This allows users to address problems early and improve the workplace environment.

[1424] Hardware and software used

[1425] Hardware: Typical server

[1426] Software: OpenAI GPT model, Python

[1427] Examples of specific cases and prompt statements

[1428] For example, to ensure a new delivery driver receives efficient training, the following prompt can be used:

[1429] Prompt message:

[1430] "Please generate training content for new food delivery drivers. The content should include how to find efficient delivery routes, how to communicate effectively with customers, and tips for safe driving."

[1431] Example of generated content:

[1432] Training program for new food delivery drivers:

[1433] 1. How to find an efficient delivery route

[1434] How to use map apps

[1435] How to avoid rush hour

[1436] 2. Effective methods of communication with customers

[1437] Polite language

[1438] Troubleshooting Basics

[1439] 3. Points for safe driving

[1440] Stopping and braking

[1441] Maintain an appropriate speed

[1442] In this way, new delivery drivers can acquire the necessary knowledge in a short period of time and be well-prepared before starting work.

[1443] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1444] Step 1:

[1445] Users submit requests from their devices. These requests may include new employee training materials, customer communication guides, and harassment reports.

[1446] Input: User request (e.g., Request for new employee training materials)

[1447] Output: Sending a request to the server

[1448] Specific action: The user uses the terminal interface to type "Generate new employee training materials" and clicks the submit button.

[1449] Step 2:

[1450] The server initializes the generative artificial intelligence engine and receives a request from the user. It then selects the appropriate module based on the content of the request.

[1451] Input: User Request

[1452] Output: Selection and startup of the appropriate module

[1453] Specific operation: The server parses the request content and, if it is for new employee training materials, launches the training content generation module.

[1454] Step 3:

[1455] The generative artificial intelligence engine creates prompt sentences for generating training content and sends them to the AI ​​engine.

[1456] Input: Request details, prompt text to be generated

[1457] Output: Prompt statement generation

[1458] Specific operation: The server generates a prompt message saying "Generate training content for new food delivery drivers" and sends it to the AI ​​engine.

[1459] Step 4:

[1460] The AI ​​engine processes the prompt text and generates training content for new employees.

[1461] Input: Prompt message

[1462] Output: Generated new employee training content

[1463] Specific operation: The AI ​​engine uses the OpenAI GPT model to generate training content based on prompt sentences.

[1464] Step 5:

[1465] The server sends the generated training content back to the user.

[1466] Input: Generated training content

[1467] Output: Return to user

[1468] Specific operation: The server sends the generated training content (e.g., how to find efficient routes, tips for safe driving, etc.) to the user's device.

[1469] Step 6:

[1470] Users receive and review training content generated on their devices. If necessary, they can save or print the training content.

[1471] Input: Training content returned from the server

[1472] Output: Display and save training content

[1473] Specific actions: The user reviews the training content received on their device and, if necessary, saves it to the device or prints it out to use as reference material.

[1474] Through the processing steps described above, this system can efficiently provide training content in response to user requests and support the skill development of new delivery drivers in the food delivery industry.

[1475] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1476] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a power harassment and moral harassment detection module, and an emotion engine. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment. Furthermore, by providing support that takes the user's emotions into consideration, it realizes more appropriate and effective support.

[1477] Overall system configuration

[1478] The server comprises a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules operate to generate and provide appropriate support content in response to user requests.

[1479] Program execution flow (overview)

[1480] 1. Server initialization

[1481] The server initializes each engine and module upon startup. In particular, the generative AI engine and the emotion engine load detailed settings and training data.

[1482] 2. Receiving requests from users

[1483] Users send requests to the server through their devices. These requests include new employee training materials, email composition assistance, assistance with asking questions, and reports of power harassment / moral harassment.

[1484] 3. Emotion recognition by an emotion engine

[1485] The server uses an emotion engine to analyze user emotion data (e.g., facial expressions, voice tone, etc.) sent with the request. The analysis results are then fed back to other modules.

[1486] 4. Processing of each module

[1487] Based on the request type and sentiment analysis results, the appropriate module (e.g., generative AI engine, email creation support module, question-asking support module, power harassment / moral harassment detection module) is selected. This allows each module to generate the most suitable support content.

[1488] 5. Responding to the user

[1489] The generated support content and evaluation results are returned to the user. If the results of the emotion engine are incorporated, the content will be optimized for the user's situation and emotions.

[1490] Specific example

[1491] Provision of training materials for new employees

[1492] A user sends a request for new employee training materials to the server. Along with this request, the user's facial expression data and voice are also sent. The server analyzes this data using an emotion engine to determine the user's emotional state. For example, if the user is nervous, training materials written in a relaxing tone are generated. A generative artificial intelligence engine generates specific training content, and the server provides these materials to the user.

[1493] Providing email creation support

[1494] The user submits a request for email composition assistance. The emotion engine generates an email composition guide that includes advice to alleviate the user's feelings of anxiety or impatience if they are experiencing such emotions. The email composition assistance module provides specific expressions and templates, which the server returns to the user.

[1495] Reports of power harassment / moral harassment

[1496] Users report instances of power harassment or moral harassment. The emotion engine evaluates the user's mental state, and the server uses this information to perform a detailed analysis using a moral harassment / power harassment detection module. Along with the evaluation results, appropriate counseling and countermeasures are provided to the user.

[1497] This system allows users to efficiently receive the support and information they need, and enables appropriate responses based on their emotional state. By combining it with an emotional engine, a more human-centered approach is added, which can increase user satisfaction and effectiveness.

[1498] The following describes the processing flow.

[1499] Provision of training materials for new employees

[1500] Step 1:

[1501] A user issues a request for new employee training materials. The user sends the request from their terminal to the server.

[1502] Step 2:

[1503] The server receives the request. It verifies that the request type is "training".

[1504] Step 3:

[1505] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[1506] Step 4:

[1507] The server uses an emotion engine to analyze the user's emotional data and determine the user's emotional state.

[1508] Step 5:

[1509] The server invokes a generative artificial intelligence engine to generate optimal training content based on user information and emotional state.

[1510] Step 6:

[1511] A generative artificial intelligence engine generates training content and returns it to the server.

[1512] Step 7:

[1513] The server sends the generated training content to the user's terminal. The content is structured with the user's emotional state in mind.

[1514] Step 8:

[1515] Users receive training content via their devices and review its contents.

[1516] Providing email creation support

[1517] Step 1:

[1518] The user issues a request for email creation assistance. The request is sent from the user's terminal to the server.

[1519] Step 2:

[1520] The server receives the request. It verifies that the request type is "email_help".

[1521] Step 3:

[1522] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[1523] Step 4:

[1524] The server uses an emotion engine to analyze the user's emotional data and determine the user's emotional state.

[1525] Step 5:

[1526] The server calls an email composition support module and generates optimal email composition support content based on user information and emotional state.

[1527] Step 6:

[1528] The email composition support module generates the email composition support content and returns it to the server.

[1529] Step 7:

[1530] The server sends the generated email composition support content to the user's terminal. The content is adjusted to take the user's emotional state into consideration.

[1531] Step 8:

[1532] Users receive email composition support content via their devices and check specific expressions and writing styles.

[1533] Reports of power harassment / moral harassment

[1534] Step 1:

[1535] A user requests to report instances of power harassment or moral harassment. The user sends the request from their terminal to the server.

[1536] Step 2:

[1537] The server receives the request. It verifies that the request type is "check_abuse".

[1538] Step 3:

[1539] The server receives user emotion data (facial expressions, voice tone, etc.) transmitted from the user's terminal.

[1540] Step 4:

[1541] The server uses an emotion engine to analyze the user's emotional data and evaluate their mental state.

[1542] Step 5:

[1543] The server invokes a power harassment / moral harassment detection module and evaluates the reported content.

[1544] Step 6:

[1545] The power harassment / moral harassment detection module analyzes the reported content and identifies the problem.

[1546] Step 7:

[1547] The server generates the analysis results and necessary countermeasures and sends them to the user's terminal.

[1548] Step 8:

[1549] Users receive evaluation results and countermeasures via their devices and review the countermeasures as needed.

[1550] Through the above processes, users can efficiently receive the necessary support and information, enabling them to respond appropriately. Furthermore, by combining this with an emotion engine, appropriate support content is provided according to the user's emotional state.

[1551] (Example 2)

[1552] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1553] In conventional systems, AI-powered email creation support, new employee training support, and detection and response to power harassment and moral harassment were handled separately, making centralized management and response difficult. Furthermore, appropriate support that considered the user's emotional state was not adequately provided. Therefore, there is a need to improve the user experience and establish an efficient support system.

[1554] In Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for generating email creation support content, means for generating support content regarding how to ask questions, means for detecting power harassment and moral harassment, means for analyzing the user's emotional data and feeding the results back to other modules for emotion recognition, and means for executing at least one of these means in response to a request from the user. This makes it possible to efficiently and centrally provide support that meets the diverse needs of the user. Furthermore, it is possible to provide appropriate content that takes into account the user's emotional state, and an improvement in the user experience can be expected.

[1555] A "generative artificial intelligence engine" refers to artificial intelligence that automatically generates various types of support content based on user requests.

[1556] "Email creation support content" refers to the guidelines, templates, and specific examples of emails that users need when creating emails for business or personal use.

[1557] "Support regarding how to ask questions" refers to guidelines and sample sentences provided to help users ask appropriate questions and phrases depending on the situation.

[1558] "Methods for detecting power harassment and moral harassment" refers to technologies that analyze user reports and collected data to identify the presence of power harassment and moral harassment.

[1559] "User sentiment data" refers to information about emotions extracted from the user's facial expressions, voice tone, text content, etc.

[1560] "Emotion recognition means" refers to technology that analyzes collected user emotion data and feeds the results back to various support modules.

[1561] "Means of execution in response to user requests" refers to the technology for selecting the appropriate module and executing its function in response to user requests.

[1562] This invention relates to a system comprising a generative artificial intelligence engine, an email creation support module, a question-asking support module, a power harassment and moral harassment detection module, and an emotion engine. This enables efficient training of new employees, supplementation of sales knowledge, and early detection and response to harassment. Furthermore, by providing support that takes the user's emotions into consideration, it realizes more appropriate and effective support.

[1563] Overall system configuration

[1564] The server comprises a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules operate to generate and provide appropriate support content in response to user requests.

[1565] Program execution flow (overview)

[1566] The server initializes each engine and module upon startup. In particular, the generative AI engine and the emotion engine load detailed settings and training data.

[1567] Users send requests to the server through their devices. These requests include new employee training materials, email composition assistance, assistance with asking questions, and reports of power harassment / moral harassment. The server analyzes the user's emotional data (e.g., facial expressions, voice tone, etc.) sent with the requests using an emotion engine. The analysis results are then fed back to other modules.

[1568] Based on the request type and sentiment analysis results, the appropriate module (e.g., generative AI engine, email creation support module, question-asking support module, power harassment / moral harassment detection module) is selected. Each module then generates the optimal support content. The generated support content and evaluation results are returned to the user. If the sentiment engine results are incorporated, the content will be optimized for the user's situation and emotions.

[1569] Specific example

[1570] Provision of training materials for new employees

[1571] A user sends a request for new employee training materials to the server. Along with this request, the user's facial expression data and voice are also sent. The server analyzes this data using an emotion engine to determine the user's emotional state. For example, if the user is nervous, training materials written in a relaxing tone are generated. A generative artificial intelligence engine generates specific training content, and the server provides these materials to the user.

[1572] Specific example:

[1573] A user requests, "Please tell me about your new employee training methods."

[1574] Example prompt: "Please provide relaxing training materials for new employees who are feeling nervous."

[1575] Providing email creation support

[1576] The user submits a request for email composition assistance. The emotion engine generates an email composition guide that includes advice to alleviate the user's feelings of anxiety or impatience if they are experiencing such emotions. The email composition assistance module provides specific expressions and templates, which the server returns to the user.

[1577] Specific example:

[1578] The user requests help creating a formal work email.

[1579] Example prompt: "Please provide a formal email template for users who are feeling anxious."

[1580] Reports of power harassment / moral harassment

[1581] Users report instances of power harassment or moral harassment. The emotion engine evaluates the user's mental state, and the server uses this information to perform a detailed analysis using a moral harassment / power harassment detection module. Along with the evaluation results, appropriate counseling and countermeasures are provided to the user.

[1582] Specific example:

[1583] A user requests to report workplace harassment from their superior.

[1584] Example prompt: "Please provide appropriate counseling for users who are experiencing harassment."

[1585] Hardware and software to be used

[1586] The server runs on a computer equipped with a high-performance processor and large memory capacity. Generative artificial intelligence engines and emotion engines are implemented using machine learning libraries such as Python, TensorFlow, and PyTorch. RDBMS such as MySQL and PostgreSQL are used for the database. User terminals include personal computers, smartphones, and tablets, which communicate with the server via the internet.

[1587] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1588] Step 1: Server Initialization

[1589] The server initializes each engine and module upon startup. Specifically, it initializes the generative AI engine and the emotion engine, loading their respective configuration parameters and training data. It reads configuration files (e.g., config.yaml) and sets the connection settings and initial values ​​for each module. The input for this step is the configuration files and training data, and the output is a state where each module is ready to operate.

[1590] Step 2: Receiving requests from users

[1591] The user sends a request to the server through their device. For example, if a user sends a request for new employee training materials, the request data will include text-based instructions and user sentiment data (facial expressions, voice tone, etc.). The server receives the HTTP request and parses the data in JSON format. The input to this step is the user's request data, and the output is the parsed request content and sentiment data.

[1592] Step 3: Emotion recognition by the emotion engine

[1593] The server analyzes the emotion data received from the user using an emotion engine. Facial expression data is analyzed using a facial recognition algorithm, and voice data is identified using a voice analysis algorithm to determine the emotional state. For example, if the user is tense, it will be tagged as tense. The input for this step is the user's facial expression data and voice data, and the output is the identified emotional state (e.g., tense, relaxed).

[1594] Step 4: Processing each module

[1595] The server selects the appropriate module based on the request type and sentiment analysis results. For example, if a user requests email composition assistance and sentiment data indicates anxiety, the server generates assistance content combining the email composition assistance module and the sentiment recognition results. The generative AI engine receives a prompt to generate specific content and produces the optimal response. The input for this step is the request type and sentiment analysis results, and the output is the generated assistance content.

[1596] Step 5: Responding to the user

[1597] The server returns the generated support content and evaluation results to the user. The generated content may include, for example, training materials for new employees, email creation guides, or appropriate counseling content for reports of power harassment / moral harassment. This step also incorporates the results of the emotion engine, ensuring that content is tailored to the user's situation and emotions. The server generates response data in JSON format and sends it to the user's terminal as an HTTP response. The input for this step is the generated support content and emotion analysis results, while the output is the response message to the user.

[1598] (Application Example 2)

[1599] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1600] In modern factories and manufacturing sites, the proper and efficient implementation of new employee training and problem reporting, as well as the early detection and response to harassment, are critical issues. Furthermore, customized support tailored to the user's emotional state is required, but conventional systems have struggled to integrate these aspects. This invention aims to solve these problems and improve the working environment and increase efficiency.

[1601] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1602] In this invention, the server includes means for initializing and controlling a generative artificial intelligence engine, means for analyzing user emotion data and generating support content corresponding to the emotional state, and means for displaying and communicating the support content to the user via voice. This enables efficient training of new employees, early detection and countermeasures for problem reporting and harassment, and customized support tailored to the user's emotional state.

[1603] A "generative artificial intelligence engine" is an artificial intelligence that generates appropriate content from text and data. This makes it possible to answer complex questions and create appropriate emails, among other things.

[1604] "Email creation support content" refers to support information and templates provided to help users create emails efficiently and appropriately.

[1605] "Support regarding how to ask questions" refers to support information that provides appropriate expressions and phrases to effectively ask questions or make requests.

[1606] "Methods for detecting power harassment and moral harassment" refer to functions that analyze user input and emotional data to determine whether or not harassment has occurred.

[1607] "User emotional data" refers to data about a user's emotional state, obtained from things like their facial expressions and voice tone.

[1608] "Support tailored to emotional state" refers to support information provided in the format and tone most appropriate to the user, based on the user's emotional data.

[1609] "New employee training content" refers to educational materials and guidelines designed to support the learning of new employees in factories, manufacturing sites, and other similar settings.

[1610] "Counseling" refers to the act or process of providing mental support to victims of power harassment or moral harassment.

[1611] "Means for communicating support content visually and audibly" refers to devices such as displays and speakers that provide the generated support content to the user visually and audibly.

[1612] The present invention relates to a robotic system for supporting new employee training, which is installed in factory robots and includes a generative artificial intelligence engine, an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. Embodiments of this system are described in detail below.

[1613] Hardware and software to be used

[1614] 1. Touchscreen interface

[1615] Example: Use a Raspberry Pi touchscreen.

[1616] 2. Speech Recognition and Synthesis Module

[1617] Example: Use Google Cloud Speech-to-Text or Text-to-Speech.

[1618] 3. Emotion Recognition Camera

[1619] Example: Use Intel RealSense.

[1620] System program description

[1621] The server includes a generative artificial intelligence engine (AI engine), an emotion engine, HR-related information, an email creation support module, a question-asking support module, and a power harassment and moral harassment detection module. These modules generate and provide appropriate support content in response to requests from users (factory workers).

[1622] Specifically, the server performs data processing and calculations using the following methods:

[1623] 1. Generative artificial intelligence engine

[1624] When a user requests training materials for new employees, a generative artificial intelligence engine (e.g., GPT-4) generates the training content.

[1625] The generated training materials will include factory safety guidelines, operating procedures for key equipment, and disaster response procedures.

[1626] 2. Emotional Engine

[1627] An emotion engine (e.g., Affectiva Emotion AI) analyzes data from the user's facial expressions and voice tone to evaluate the user's emotional state.

[1628] For example, if a user is feeling anxious, provide training materials that include relaxation techniques.

[1629] 3. Provided via display and audio.

[1630] Actual display and audio output are performed through the robot's touchscreen and speakers.

[1631] The generated training materials and support content are displayed and communicated via audio in a way that is optimized for the user's emotional state.

[1632] Specific examples

[1633] For example, when a new factory worker requests training materials, the following prompt message is input to the generative artificial intelligence engine:

[1634] Please generate training materials for new employees. Include the following: factory safety guidelines, operating procedures for key equipment, and disaster response procedures. Users are nervous.

[1635] The emotion engine analyzes the user's facial expressions and voice tone at that time and obtains the following analysis data:

[1636] The system analyzes facial expression data and voice tone in real time to evaluate the user's emotional state. The analysis result is [tension, anxiety].

[1637] The generative artificial intelligence engine generates the following based on this data:

[1638] Generate training materials that include relaxation techniques. The proposed training materials will include calm language and relaxation methods.

[1639] The robot then displays the training materials via a touchscreen and uses voice prompts to encourage relaxation.

[1640] This system not only enables efficient training of new employees, but also facilitates problem reporting, early detection and countermeasures against harassment, and customized support tailored to the user's emotional state.

[1641] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1642] Step 1:

[1643] Server initialization and configuration

[1644] Upon startup, the server initializes each engine (generative AI engine, emotion engine) and module (email creation support module, question-asking support module, power harassment / moral harassment detection module) and performs the necessary configurations. This includes loading existing training data and the configuration values ​​for each module. This prepares the server to respond quickly to user requests.

[1645] Step 2:

[1646] Received a request from the user.

[1647] Through a terminal, users (factory workers) send requests for new employee training materials, problem reports, and other information to the server. These requests also include the user's facial expression data and voice. The server receives this information and begins a process to analyze the relevant data. Input data includes the request content, facial expression data, and voice data. Output data provides initial analysis results to be passed on to the next analysis step.

[1648] Step 3:

[1649] Analysis of emotional data using an emotion engine

[1650] The server inputs facial expression data and voice data received from the user into an emotion engine (e.g., Affectiva Emotion AI) to analyze the user's emotional state. The emotion engine converts facial expressions and voice tone into digital data in real time and evaluates emotional states such as tension and anxiety. The input data consists of facial expression data and voice data, and the output data is the evaluation result of the emotional state. The server retrieves this evaluation result and proceeds to the next processing step.

[1651] Step 4:

[1652] Content generation using generative artificial intelligence engines

[1653] Based on the request content and the analysis results of the emotion engine, the server inputs a prompt message into a generative artificial intelligence engine (e.g., GPT-4) to generate corresponding support content (new employee training materials, email content, etc.). For example, it inputs a prompt message like the following:

[1654] Please generate training materials for new employees. Include the following: factory safety guidelines, operating procedures for key equipment, and disaster response procedures. Users are nervous.

[1655] The generative artificial intelligence engine analyzes this prompt and generates the necessary content. The input data consists of the request and the sentiment analysis results, while the output data is the generated training material.

[1656] Step 5:

[1657] Provision of generated support content

[1658] The server displays and delivers supportive content obtained from a generative artificial intelligence engine to the user in an appropriate format. Specifically, it displays text using the robot's touchscreen and provides explanations via a speaker. The input data is the generated supportive content, and the output data is the visual and auditory information provided to the user. This allows the user to efficiently obtain the necessary information.

[1659] Step 6:

[1660] Gathering feedback and optimizing the system

[1661] Users provide feedback on the content they are given. The server collects this feedback information and uses it to generate future support content and perform sentiment analysis. The feedback includes aspects such as understanding and satisfaction with the content. The input data is user feedback information, and the output data is optimized system settings. This allows the system to continuously improve.

[1662] The above describes the processing steps of this system.

[1663] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1664] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1665] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1666] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1667] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1668] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1669] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1670] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1671] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1672] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1673] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1674] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1675] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1676] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1677] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1678] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1679] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1680] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1681] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1682] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1683] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1684] The following is further disclosed regarding the embodiments described above.

[1685] (Claim 1)

[1686] A means for initializing and controlling a generative artificial intelligence engine,

[1687] A means of generating email creation support content,

[1688] A means of generating support content regarding how to ask questions,

[1689] Methods for detecting power harassment and moral harassment,

[1690] A system that includes means for performing at least one of these means in response to a request from a user.

[1691] (Claim 2)

[1692] The system according to claim 1, comprising means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for a user, and means for executing at least one of these means in response to a request from a user.

[1693] (Claim 3)

[1694] The system according to claim 1, comprising means for detecting power harassment and moral harassment and providing the evaluation results to the user.

[1695] "Example 1"

[1696] (Claim 1)

[1697] A means for initializing and controlling a generative artificial intelligence engine,

[1698] A means of generating email creation support content,

[1699] A means of generating support content regarding how to ask questions,

[1700] Methods for detecting power harassment and moral harassment,

[1701] A means of generating training materials for new employees,

[1702] A system that includes means for performing at least one of these means in response to a request from a user.

[1703] (Claim 2)

[1704] The system according to claim 1, comprising means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for the user, and means for generating email creation support content.

[1705] (Claim 3)

[1706] The system according to claim 1, comprising means for detecting power harassment and moral harassment and providing the evaluation results to the user.

[1707] "Application Example 1"

[1708] (Claim 1)

[1709] A means for initializing and controlling a generative artificial intelligence engine,

[1710] A means of generating email creation support content,

[1711] A means of generating support content regarding how to ask questions,

[1712] Methods for detecting power harassment and moral harassment,

[1713] A means of generating training content for the food delivery industry,

[1714] A system that includes means for performing at least one of these means in response to a request from a user.

[1715] (Claim 2)

[1716] The system according to claim 1, comprising means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for a user, and means for providing training materials specialized for food delivery.

[1717] (Claim 3)

[1718] The system according to claim 1, including means for guiding appropriate communication methods at the delivery destination.

[1719] "Example 2 of combining an emotion engine"

[1720] (Claim 1)

[1721] A means for initializing and controlling a generative artificial intelligence engine,

[1722] A means of generating email creation support content,

[1723] A means of generating support content regarding how to ask questions,

[1724] Methods for detecting power harassment and moral harassment,

[1725] An emotion recognition means that analyzes user emotion data and feeds the results back to other modules,

[1726] A system that includes means for performing at least one of these means in response to a request from a user.

[1727] (Claim 2)

[1728] The system according to claim 1, comprising means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for the user, and means for analyzing the user's emotional data and feeding the results back to other modules.

[1729] (Claim 3)

[1730] The system according to claim 1, comprising means for detecting power harassment and moral harassment and providing appropriate counseling content based on the analysis results.

[1731] "Application example 2 when combining with an emotional engine"

[1732] (Claim 1)

[1733] A means for initializing and controlling a generative artificial intelligence engine,

[1734] A means of generating email creation support content,

[1735] A means of generating support content regarding how to ask questions,

[1736] Methods for detecting power harassment and moral harassment,

[1737] A means for analyzing user emotional data and generating support content according to the emotional state,

[1738] A means of communicating the support provided to the user through display and audio,

[1739] A system that includes means for performing at least one of these means in response to a request from a user.

[1740] (Claim 2)

[1741] The system according to claim 1, comprising means for initializing and controlling a generative artificial intelligence engine; means for generating new employee training content suitable for the user; means for analyzing the user's emotional state and providing optimized training materials based on the analysis results; and means for executing at least one of these means in response to a request from the user.

[1742] (Claim 3)

[1743] The system according to claim 1, comprising means for detecting power harassment and moral harassment and providing appropriate counseling and countermeasures based on the user's emotional state. [Explanation of Symbols]

[1744] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for initializing and controlling a generative artificial intelligence engine, A means of generating email creation support content, A means of generating support content regarding how to ask questions, Methods for detecting power harassment and moral harassment, A system that includes means for performing at least one of these means in response to a request from a user.

2. The system according to claim 1, comprising means for initializing and controlling a generative artificial intelligence engine, means for generating new employee training content suitable for a user, and means for executing at least one of these means in response to a request from a user.

3. The system according to claim 1, comprising means for detecting power harassment and moral harassment and providing the evaluation results to the user.

Images