System

Abstract

A system is provided.SOLUTION: A system comprising: means for a server to receive and store basic information in a database; means for a user to select a template of a job recruiting slip and input a job recruiting condition; means for the server to generate a draft of the job recruiting slip using a natural-language processing engine based on the received information; means for the user to confirm and correct the draft; and means for the server to link and publish a final version of the job recruiting slip to an external job search website.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Creating job postings for mid-career recruitment takes time and effort, and often does not allow for smooth coordination between on-site staff and human resources personnel. Furthermore, many job postings lack understanding and descriptions of the job content, resulting in inefficient recruitment activities. In such situations, the quality of job postings declines, recruitment activities themselves are delayed, and companies find it difficult to secure talent. [Means for solving the problem]

[0005] The present invention provides a system that includes a server receiving basic information and storing the information in a database, a user selecting a job posting template and entering job requirements, a server using a natural language processing engine to generate a draft job posting based on the information received, a user reviewing and revising the draft, and a server linking the final job posting to an external recruitment website for publication. This allows for rapid and accurate creation of job postings, reducing the burden on on-site and human resources personnel and providing high-quality job postings. The system also includes a server receiving feedback and storing the information in a database, and a server analyzing the feedback information and adjusting the job posting generation algorithm, enabling continuous improvement of recruiting activities. Furthermore, the system includes a user inputting the conditions for sending scout emails, a server generating and sending scout emails based on the conditions, and a server tracking the results of the emails sent, improving the efficiency of scouting activities.

[0006] A "server" is a computing device that receives input information from users, stores it in a database, and performs a series of processes such as generating and publishing job postings.

[0007] "Database" is a data storage device that stores and manages basic information, job information, and feedback information received by the server.

[0008] A "job posting template" is a predefined format or style that a user selects when creating a job posting.

[0009] "Job requirements" refers to information such as job type, desired skills, experience, salary range, etc. that the user enters to be included in the job posting.

[0010] A "natural language processing engine" is an algorithm or software that generates natural-sounding sentences based on input data.

[0011] "Job Posting Draft" means an initial version of a job posting generated by the server using a natural language processing engine.

[0012] "Feedback" refers to information collected from users about their reactions and opinions, and used to improve the system and job postings.

[0013] A "scouting email" is an email sent to candidates who meet certain criteria to pique their interest in a job opening at your company.

[0014] "Tracking" refers to the process of tracking and recording responses to scouting emails and job postings sent out. [Brief explanation of the drawings]

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

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

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

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

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

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

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

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

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

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

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

[0036] The embodiment of the present invention is a generative AI service that supports companies' recruitment operations. This system operates in cooperation with three parties: a server, a terminal, and a user.

[0037] 1. Registering and managing basic information

[0038] The server starts up and opens an API endpoint, preparing to establish communication with the device. The user enters basic information through the device, and the server stores that information in a database. This basic information includes the company name, location, industry type, management policy, etc.

[0039] Examples:

[0040] A human resources manager at an IT company logs into the system and enters basic information about the company (e.g., "XYZ Company, Tokyo, software development industry, corporate culture that values ​​innovation"), and the server receives this information and stores it in a database.

[0041] 2. Job posting generation

[0042] Users log in to the system from their terminal and select a job posting template. Templates are available in multiple formats for different job types and departments. They then enter details such as job type, desired skills, years of experience, and salary range according to the template they select.

[0043] The server receives this information and uses a natural language processing engine to generate a draft job posting, which is then provided to the user for review and modification.

[0044] Examples:

[0045] A human resources representative from the same IT company selects the "Java Developer" template and inputs the desired skills (Java, Spring Framework), years of experience (3+ years), salary range (annual salary 6-8 million yen), etc. The server then generates a draft, and the user can review the proposed content and make any necessary revisions.

[0046] 3. Publication of job postings

[0047] The user checks the final version of the job posting and sends a publishing request to the server. The server saves the job posting and publishes it on external job sites or the company's own recruitment page.

[0048] Examples:

[0049] Once the HR person has reviewed the draft and clicked the "Publish" button, the server saves the job posting and it is instantly updated on directly linked external job sites and the company's own recruitment page. This entire process is completed in just a few minutes, enabling rapid recruitment activities.

[0050] 4. Receiving and Analyzing Feedback

[0051] User feedback is received by the server and stored in a database, where it can be analyzed and used to improve the job posting generation algorithm.

[0052] Examples:

[0053] After publishing a job posting, users input reactions from job seekers and internal feedback (e.g., "few applicants," "the description is difficult to understand," etc.) into the system. This information is stored on the server and will be used the next time a job posting is generated.

[0054] 5. Generate and send scout emails (future feature)

[0055] The user inputs the conditions for the person to whom they want to send a scout email. The server generates and sends the scout email based on these conditions. The server tracks the sending results and provides feedback to the user.

[0056] Examples:

[0057] If a human resources officer wants to send a scouting email to a candidate who is a "full-stack engineer" and "able to work remotely," he or she enters those conditions into the system. The server automatically generates and sends scouting emails to candidates who meet the conditions. Later, the server tracks the response data to the scouting email and provides it to the human resources officer.

[0058] In this way, the present invention effectively and efficiently supports the recruitment operations of companies.

[0059] The processing flow will be explained below.

[0060] Step 1:

[0061] The server starts and opens an API endpoint, allowing devices to communicate with the server.

[0062] Step 2:

[0063] The terminal loads the user interface, allowing the user to access the system.

[0064] Step 3:

[0065] A user logs into the system through a terminal and enters a user ID and password.

[0066] Step 4:

[0067] The server checks the user's authentication information against the database, and if authentication is successful, displays a basic information input screen for the user.

[0068] Step 5:

[0069] The user enters basic information about the company (company name, location, industry, management policy, etc.) and clicks the "Submit" button.

[0070] Step 6:

[0071] The server receives the basic information entered and stores it in a database.

[0072] Step 7:

[0073] The user clicks the "Create a new job posting" button on their device and selects an appropriate template from a list of job posting templates.

[0074] Step 8:

[0075] The server receives the user's template selection request and displays the template form on the user interface.

[0076] Step 9:

[0077] The user enters job requirements such as job type, desired skills, experience, and salary range according to the template, and clicks the "Complete input" button.

[0078] Step 10:

[0079] The server receives the entered job requirements and temporarily stores them.

[0080] Step 11:

[0081] The server invokes a natural language processing engine and generates a draft job posting based on the received information.

[0082] Step 12:

[0083] The generated draft job posting is sent to the user's terminal and displayed on the user interface.

[0084] Step 13:

[0085] The user checks the draft content on the terminal and makes corrections as necessary.

[0086] Step 14:

[0087] After making corrections, the user clicks the "Confirm" button to approve the final version of the job posting.

[0088] Step 15:

[0089] The server stores approved job postings in a database and then links to external job sites and the company's own recruitment page via an API, making them publicly available.

[0090] Step 16:

[0091] Users conduct job postings and input responses from job seekers and internal feedback into the system.

[0092] Step 17:

[0093] The server receives the feedback information and stores it in a database.

[0094] Step 18:

[0095] The server analyzes the feedback and uses it to improve the job posting generation algorithm.

[0096] Step 19:

[0097] The user enters the conditions for the person to whom the scout email is to be sent and clicks the "Conditions input complete" button.

[0098] Step 20:

[0099] The server receives the conditions and generates a scout email.

[0100] Step 21:

[0101] The server sends the generated scout email to the candidate and tracks the sending results.

[0102] Step 22:

[0103] The server analyzes the tracking results and provides feedback to the user through a user interface.

[0104] Example 1

[0105] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0106] Traditional recruitment processes have the problem of taking a lot of time and effort, from creating a job posting to publishing it and receiving feedback. Furthermore, if the job posting does not receive a positive response from job seekers due to inappropriate content, it takes a lot of effort to revise the information and republish it. Furthermore, there is no efficient way to generate and send scouting emails, making it difficult to effectively recruit talent.

[0107] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0108] In this invention, the server includes: [means for receiving basic information and saving it in a database;] [means for selecting a job posting template and inputting job requirements;] [means for generating a draft job posting using a natural language processing engine based on the received information;] [means for publishing the final job posting in conjunction with an external job site or the company's own recruitment page;] [means for opening an API endpoint and establishing communication with a terminal;] [means for receiving feedback and saving it in a database;] [means for analyzing feedback information and adjusting the job posting generation algorithm;] [means for inputting conditions for sending scout emails;] [means for generating and sending scout emails based on the conditions; and [means for tracking the sending results.] This makes it possible to effectively and efficiently carry out a series of recruitment tasks, from creating and publishing a job posting, analyzing feedback, and generating and sending scout emails.

[0109] "Basic information" refers to basic data about a company, such as the company name, location, industry type, and management policy.

[0110] "User" refers to a person who uses this system to manage and operate job information, such as a company's recruiter or human resources officer.

[0111] "Server" refers to a computer that forms the core part of the system and performs various processes such as receiving, storing, analyzing, and publishing data.

[0112] "Terminal" refers to a device used by a user, such as a computer, tablet, or smartphone.

[0113] "Database" refers to organized data storage for storing basic information, job postings, feedback information, etc.

[0114] A "template" is a prescribed format used when creating job postings, prepared for each job type and department.

[0115] "Job requirements" refers to the detailed requirements such as the job type, required skills, years of experience, and salary range listed in the job posting.

[0116] A "natural language processing engine" is an AI model that generates and analyzes sentences, and is a technology that is particularly used to draft job postings.

[0117] "External job sites" refer to websites where companies post job openings and which job seekers can access.

[0118] An "API endpoint" is a connection point between a server and a terminal for communication, and refers to the URL or URI used to send and receive data.

[0119] "Feedback" refers to opinions and impressions from users and job seekers regarding the content of and reactions to job postings.

[0120] A "scouting email" refers to an email sent by a company to recruit job seekers who meet certain criteria.

[0121] "Conditions" refers to the requirements such as job type, skills, and working style of the candidate to whom the scouting email is sent.

[0122] "Tracking" refers to the process of tracking and recording the results of sending scouting emails and the responses to them.

[0123] MODE FOR CARRYING OUT THE INVENTION

[0124] This invention is a generative AI service to support companies' recruitment operations. This system operates in cooperation with three parties: a server, a terminal, and a user, and is implemented according to the following procedure.

[0125] Starting the server and opening the API endpoint

[0126] The server is launched on an AWS EC2 instance, running the Apache Web Server, and an API endpoint is set up, ready to establish communication with the device.

[0127] Registering and managing basic information

[0128] Users access the system through a terminal and enter basic information about their company, including the company name, location, industry type, management policy, etc. The terminal then sends the entered information to the server, which stores it in a database.

[0129] Example: A human resources manager at a manufacturing company logs into the system and enters basic information about the company (e.g., "ABC Manufacturing, Aichi Prefecture, manufacturing industry, corporate culture that emphasizes quality"). The server receives this information and stores it in a database.

[0130] Job posting generation

[0131] Users log in to the system from their terminal and select a job posting template prepared by job type or department. They then enter details such as the job type, desired skills, years of experience, salary range, etc. This information is sent to the server, which uses a generative AI model (e.g., OpenAI GPT-3) to generate a draft job posting.

[0132] Example: A human resources professional selects a "Systems Engineer" template and inputs the required skills (e.g., Python, database management), years of experience (e.g., 5+ years), salary range (e.g., annual salary of 5 million to 7 million yen), etc. The server receives this and uses a generative AI model to generate a draft job posting. The user reviews the draft and makes any necessary revisions.

[0133] Example prompt sentence:

[0134] What skills are you looking for in a "Systems Engineer"? Also include the required years of experience, salary range, and desired characteristics of the company.

[0135] Publication of job postings

[0136] The user checks the final version of the job posting and sends a publishing request to the server, which saves the job posting and publishes it via an external job posting site or the company's own recruitment page.

[0137] Example: A human resources representative makes a final check and clicks the "Publish" button. The server saves the final job posting in the database and immediately updates it on external job sites and the company's own recruitment page, enabling rapid recruitment.

[0138] Receiving and analyzing feedback

[0139] After the job posting is published, the server receives feedback from users and stores it in a database. The server can analyze the feedback information and use it to adjust the job posting generation algorithm.

[0140] Example: After publishing a job posting, users enter feedback from job seekers and internal sources (e.g., "few applicants," "the job posting content is unclear") into the system. The server receives this feedback, stores it in the database, and reflects it the next time a job posting is generated.

[0141] Generate and send scout emails (future feature)

[0142] The user inputs the conditions (job type, skills, working style, etc.) of the person to whom they want to send a scout email. The server generates and sends the scout email based on this information. The server tracks the sending results and provides feedback to the user.

[0143] Example: If a human resources manager wants to send a scouting email to a candidate who is a "mobile app developer" and "able to work remotely," he or she enters those conditions into the system. The server uses a generative AI model to automatically generate a scouting email and sends it to the candidate. The server then tracks the response data and provides it to the human resources manager.

[0144] In this way, the present invention can effectively and efficiently support a company's recruitment operations.

[0145] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0146] Specific processing flow of the system

[0147] Step 1: Start the server and open the API endpoint

[0148] The server is launched on an AWS EC2 instance, running Apache Web Server, and an API endpoint is opened to support companies' recruitment processes.

[0149] Input: Server configuration information and startup script

[0150] Output: API endpoint URL (e.g. https: / / example.com / api)

[0151] Specific operation: The server executes the startup script, which starts the web server and API endpoint, allowing communication with the device.

[0152] Step 2: Register basic information

[0153] Users access the system through a terminal and enter basic company information (company name, location, industry type, management policy).

[0154] Input: Basic company information (company name, location, industry type, management policy, etc.)

[0155] Output: Basic information input data

[0156] What happens: A user accesses the login screen using a browser and enters basic information.

[0157] Step 3: Submit and save basic information

[0158] The device sends basic information to the server, which stores the received information in a database.

[0159] Input: Basic information input data

[0160] Output: Status of saving to database

[0161] Specific operation: The device sends the entered basic information to the server, which verifies the data and saves it in the database.

[0162] Step 4: Select a job template

[0163] Users log in to the system from their terminal and select job posting templates prepared for each job type and department.

[0164] Input: Login credentials, Job template list

[0165] Output: Selected template

[0166] What happens: A user logs into the system and selects from multiple templates.

[0167] Step 5: Enter job requirements

[0168] Based on the template you select, you enter details such as job title, desired skills, years of experience, and salary range.

[0169] Input: Selected template, job requirements (job type, skills, years of experience, etc.)

[0170] Output: Job requirements data entered

[0171] Specific behavior: The user fills out a form with detailed job requirements according to the template.

[0172] Step 6: Submit your job requirements

[0173] The terminal sends the job requirements to the server.

[0174] Input: Job requirements data entered

[0175] Output: Status of sending to server

[0176] Specific operation: The terminal sends the entered job requirements to the server.

[0177] Step 7: Generate a draft job posting

[0178] Based on the job requirements received by the server, a draft job posting is generated using a generative AI model.

[0179] Input: Received job requirements data

[0180] Output: Draft job posting

[0181] Specific operation: The server uses a generative AI model (e.g., OpenAI GPT-3) to generate a draft job posting based on the job requirements.

[0182] Step 8: Review and revise your draft

[0183] The user reviews the draft job posting and makes any necessary corrections.

[0184] Input: Draft job posting

[0185] Output: Final revised job posting

[0186] What happens: The user reviews the draft provided by the server and makes any necessary corrections.

[0187] Step 9: Submit a publishing request

[0188] The user reviews the final job posting and sends a publish request to the server.

[0189] Input: Final revised job posting

[0190] Output: Publish request

[0191] Specific behavior: The user clicks the "Publish" button to send a publication request.

[0192] Step 10: Publish your job posting

[0193] The server stores the final job posting in a database and publishes it on external job sites or the company's own recruitment page.

[0194] Input: Publication request, final job posting

[0195] Output: Job posting publication status

[0196] Specific operation: The server saves the final job posting in the database and reflects it on external job sites and the company's own recruitment page via API.

[0197] Step 11: Receiving feedback

[0198] After the job posting is published, feedback from users is received by the server and stored in a database.

[0199] Input: User feedback

[0200] Output: Stored feedback data

[0201] Specific operation: After a job posting is published, the user enters feedback information into the system, and that information is sent to the server.

[0202] Step 12: Analyze feedback

[0203] The server analyzes the feedback information and uses it to adjust the job posting generation algorithm.

[0204] Input: Stored feedback data

[0205] Output: Analysis results and algorithm adjustments

[0206] Specific operation: The server analyzes the received feedback data and adjusts the algorithm to reflect it when generating the next job posting.

[0207] Step 13: Generate and send scout emails (future feature)

[0208] The user inputs the criteria for the person to whom they want to send a scouting email. The server generates a scouting email based on the criteria and sends it to the candidate. The server tracks the sending results and provides feedback to the user.

[0209] Input: Target conditions for scouting emails (job type, skills, working conditions, etc.)

[0210] Output: Scout email sending status and tracking data

[0211] Specific operation: The user inputs the target conditions, and the server uses the generative AI model to generate and send scouting emails. The server tracks the sending results and provides feedback.

[0212] (Application example 1)

[0213] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0214] Corporate recruitment is a complex process that requires a great deal of effort and time. Particularly in factories, quickly and efficiently recruiting the right talent is directly linked to production efficiency. However, delays in creating and publishing job postings and analyzing feedback can lead to the risk of missing out on key talent. Furthermore, on-site talent management is even more difficult, complicating the management work of dedicated personnel. There is a need for a system that can solve these issues and achieve a more efficient and accurate recruitment process.

[0215] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0216] In this invention, the server includes: [means for the server to receive basic information and save it in a database;] [means for the user to select a job posting template and input job requirements;] [means for the server to use a generative AI model based on the information received to generate a draft of the job posting;] [means for the user to check and correct the draft;] [means for the server to link and publish the final version of the job posting to an external recruitment media;] [means for the robot to support employee information management within the factory; and [means for the robot to input job information using a terminal set up and send it to the server. This enables more efficient human resource management within the factory and the generation and publication of accurate job information.

[0217] A "server" is a central system for data management and processing.

[0218] "Basic information" refers to basic information such as the company's name, location, industry, and management policy.

[0219] A "database" is a system for storing and managing data such as basic information and feedback.

[0220] "User" refers to the company personnel and administrators who operate the system.

[0221] A "template" is a predefined format for quickly creating job postings.

[0222] "Job requirements" refer to details such as the desired job type, required skills, years of experience, salary range, etc.

[0223] A "generative AI model" is an artificial intelligence technology that generates a draft job posting based on input information.

[0224] "Draft" refers to the first unfinished job posting generated.

[0225] "External recruitment media" refers to online platforms and sites that disseminate job information to job seekers.

[0226] A "robot" is an automated mechanical device used in factories.

[0227] "Information management" refers to the process of organizing and storing basic employee information and recruitment data.

[0228] A "terminal" is an electronic device that allows a user to input information.

[0229] System Program Overview

[0230] This invention is a "smart robot personnel assistant" system that streamlines personnel management within factories and supports the accurate creation and publication of job information. This system operates in cooperation with a server, terminals, and robots.

[0231] Server Features

[0232] The server is a central system that manages and processes data. It receives basic information (company name, location, industry, management policy, etc.) entered by users through their devices and stores it in a database. It also uses a generative AI model to generate a draft job posting based on the template selected by the user and the entered job requirements (job type, skills, years of experience, salary range, etc.). The generated draft is then reviewed and revised by the user, and the final version is published by the server to external job media.

[0233] Robot Functions

[0234] The robot supports human resource management within the factory. It uses a set-up terminal to input job information and send it to a server, which can then be used directly by on-site personnel within the factory.

[0235] Data processing and calculation

[0236] 1. Register and manage basic information:

[0237] The user enters basic company information from the terminal, and the server receives it and stores it in a database.

[0238] Hardware used: Device (PC, tablet, etc.)

[0239] Software used: Python, Requests library, RESTful API, database management system (MySQL, etc.)

[0240] 2. Generate and publish job posting:

[0241] The user inputs the job posting template and detailed requirements from the terminal, and the server uses a generative AI model (e.g., GPT-3) to generate a draft job posting.

[0242] Hardware used: Server

[0243] Software used: Generative AI models (e.g., GPT-3)

[0244] The generated draft is then checked and revised by the user, and the server publishes the final version to external recruitment media.

[0245] 3. Feedback analysis:

[0246] The server stores the feedback received after publication in a database and analyzes it to help generate the next job posting.

[0247] Software used: Feedback Analysis Engine

[0248] Specific examples

[0249] Basic information registration:

[0250] The factory's human resources staff enters basic information such as "XYZ Manufacturing, Tokyo, manufacturing industry, emphasis on quality and efficiency" into a terminal, and the server stores this information in a database.

[0251] Job listing generation:

[0252] The same person selects the "Line Operator" template and enters the desired skills of "Machine Operation" and "Quality Control," the number of years of experience of "2 years or more," and the salary range of "3 million to 5 million yen per year." Based on this, the server uses a generative AI model to generate the following prompt:

[0253] Please create a job posting for a line operator. The required skills are "machine operation" and "quality control," the required years of experience are "2 years or more," and the salary range is "annual salary of 3 million to 5 million yen."

[0254] The resulting draft is then reviewed by a human resources representative and revised as necessary. The final job posting is then published by the server to external recruitment media.

[0255] This will enable more efficient human resource management within the factory and generate accurate job information.

[0256] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0257] Step 1:

[0258] The user enters basic information through the terminal and sends it to the server. The basic information entered includes the company name, location, industry, management policy, etc. The server receives this information and stores it in the database. The input data is saved in the appropriate table in the database.

[0259] Step 2:

[0260] The user selects a job posting template from their device and enters the job requirements, including the job type, desired skills, years of experience, salary range, etc. This input data is sent to the server, which receives it and prepares it as input prompts for the generative AI model.

[0261] Step 3:

[0262] Based on the job requirements and template information received by the server, a generative AI model (e.g., GPT-3) is used to generate a draft job posting. A prompt is input to the generative AI model, which then generates a draft job posting. The generated draft is saved on the server and provided to the user.

[0263] Step 4:

[0264] The user checks the draft on their device and makes any necessary corrections. The information corrected by the user is sent to the server, which updates the draft again and generates the final version. This final version of the job posting is also saved in the database.

[0265] Step 5:

[0266] The server connects the final job posting to an external recruitment site and publishes it. Specifically, the job information is sent to the external site using methods such as API or FTP. Information on successful publication of the job posting is also saved in the database as tracking information.

[0267] Step 6:

[0268] The server receives external feedback and stores it in a database. The feedback is processed by the analytics engine and used to adjust the job posting generation algorithm for the next job posting, thereby improving the job posting process for future jobs.

[0269] Step 7:

[0270] When a user wants to send a scout email based on specific criteria, they input the criteria from their device and send it to the server. The server searches for information that matches the criteria, then uses a generative AI model to generate and send the scout email. The server tracks the results of the email and provides feedback to the user.

[0271] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0272] An embodiment of the present invention is a system that combines a generative AI service that supports corporate recruitment operations with an emotion engine. This system operates in cooperation with a server, a terminal, a user, and the emotion engine.

[0273] 1. Registering and managing basic information

[0274] The server starts up, opens an API endpoint, and prepares to communicate with the device. The user enters basic information through the device, and the server stores that information in a database. This basic information includes the company name, location, industry type, management policy, etc.

[0275] Examples:

[0276] A human resources manager at an IT company logs into the system and enters basic information about the company (for example, "ABC Company, Tokyo, software development industry, corporate culture that values ​​innovation"). The server receives this information and stores it in a database.

[0277] 2. Job posting generation

[0278] Users log in to the system from their terminal and select a job posting template. Templates are available in multiple formats for different job types and departments. They then enter details such as job type, desired skills, years of experience, and salary range according to the template they select.

[0279] The server receives this information and uses a natural language processing engine to generate a draft job posting, which is then provided to the user for review and modification.

[0280] Examples:

[0281] A human resources representative from the same IT company selects the "Java Developer" template and inputs the desired skills (Java, Spring Framework), years of experience (3+ years), salary range (annual salary 6-8 million yen), etc. The server then generates a draft, and the user can review the proposed content and make any necessary revisions.

[0282] 3. Publication of job postings

[0283] The user checks the final version of the job posting and sends a publishing request to the server. The server saves the job posting and publishes it on external job sites or the company's own recruitment page.

[0284] Examples:

[0285] Once the HR person has reviewed the draft and clicked the "Publish" button, the server saves the job posting and it is instantly updated on linked external job sites and the company's own recruitment page. This entire process is completed in just a few minutes, enabling quick recruitment activities.

[0286] 4. Receiving and Analyzing Feedback

[0287] User feedback is received by the server and stored in a database. The server analyzes the feedback and can use it to improve the job posting generation algorithm. An emotion engine is used to analyze the emotional tone of the feedback and identify areas for further improvement.

[0288] Examples:

[0289] After publishing a job posting, users input reactions from job seekers and internal feedback (for example, "few applicants" or "the description is difficult to understand") into the system. This information is stored on the server and analyzed using an emotion engine. The results of the analysis are reflected the next time a job posting is generated.

[0290] 5. Generate and send scout emails (future feature)

[0291] The user inputs the criteria for the person to whom they want to send a scout email. The server generates and sends the scout email based on this information. The emotion engine adjusts the content of the scout email appropriately. The server tracks the sending results and provides feedback to the user.

[0292] Examples:

[0293] If a human resources manager wants to send a scouting email to a candidate who is a "full-stack engineer" and "able to work remotely," he or she can enter those criteria into the system. The server automatically generates a scouting email to candidates who fit the criteria, and the emotion engine adjusts the content before sending it. Later, the server tracks the response data to the scouting email and provides it to the human resources manager.

[0294] 6. Adjusting the emotional tone when generating job postings

[0295] When users create a job posting, the emotion engine analyzes their input in real time and suggests appropriate emotional tones (e.g., proactive, friendly, professional, etc.), making job postings more engaging and effective.

[0296] Examples:

[0297] As users enter their skills, years of experience, and other information into a job posting template, the emotion engine analyzes the information and displays suggestions such as "use a more positive tone." This makes the job posting more likely to catch the attention of job seekers.

[0298] In this way, the present invention effectively and efficiently supports companies' recruitment operations, and realizes the generation of sophisticated job postings and scouting emails that take into account the user's emotions.

[0299] The processing flow will be explained below.

[0300] Step 1:

[0301] The server starts and opens an API endpoint, allowing devices to communicate with the server.

[0302] Step 2:

[0303] The terminal loads the user interface, allowing the user to access the system.

[0304] Step 3:

[0305] A user logs into the system through a terminal and enters a user ID and password.

[0306] Step 4:

[0307] The server checks the user's authentication information against the database, and if authentication is successful, displays a basic information input screen for the user.

[0308] Step 5:

[0309] The user enters basic information about the company (company name, location, industry, management policy, etc.) and clicks the "Submit" button.

[0310] Step 6:

[0311] The server receives the basic information entered and stores it in a database.

[0312] Step 7:

[0313] The user clicks the "Create a new job posting" button on their device and selects an appropriate template from a list of job posting templates.

[0314] Step 8:

[0315] The server receives the user's template selection request and displays the template form on the user interface.

[0316] Step 9:

[0317] The user enters job requirements such as job type, desired skills, experience, and salary range according to the template, and clicks the "Complete input" button.

[0318] Step 10:

[0319] The server receives the entered job requirements and temporarily stores them.

[0320] Step 11:

[0321] The server invokes a natural language processing engine and generates a draft job posting based on the received information.

[0322] Step 12:

[0323] The generated draft job posting is sent to the user's terminal and displayed on the user interface.

[0324] Step 13:

[0325] The user checks the draft content on the terminal and makes corrections as necessary.

[0326] Step 14:

[0327] The emotion engine analyzes the user's input and suggests an appropriate emotional tone (e.g., proactive, friendly, professional, etc.).

[0328] Step 15:

[0329] The user considers the emotional tone suggestions and modifies the content of the job posting.

[0330] Step 16:

[0331] After making corrections, the user clicks the "Confirm" button to approve the final version of the job posting.

[0332] Step 17:

[0333] The server stores approved job postings in a database and then links to external job sites and the company's own recruitment page via an API, making them publicly available.

[0334] Step 18:

[0335] Users conduct job postings and input responses from job seekers and internal feedback into the system.

[0336] Step 19:

[0337] The server receives the feedback information and stores it in a database.

[0338] Step 20:

[0339] The server analyzes the feedback and uses it to improve the job posting generation algorithm.

[0340] Step 21:

[0341] An emotional engine analyzes the emotional tone of the feedback and identifies areas for improvement.

[0342] Step 22:

[0343] The user enters the conditions for the person to whom the scout email is to be sent and clicks the "Conditions input complete" button.

[0344] Step 23:

[0345] The server receives the conditions and generates a scout email.

[0346] Step 24:

[0347] The emotional engine analyzes the content of the scouting email and adjusts the appropriate emotional tone.

[0348] Step 25:

[0349] The server sends the generated scout email to the candidate.

[0350] Step 26:

[0351] The server tracks the results of the transmission and provides feedback to the user through the user interface.

[0352] Example 2

[0353] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0354] In corporate recruitment processes, there are issues with the inefficiency of creating and publishing job postings, receiving and analyzing feedback, and creating and sending scouting emails. In particular, job postings and scouting emails are created only in a standardized format, and there is a lack of a system for collecting and analyzing applicant responses to improve the content. Furthermore, there is a lack of emotional tone adjustment to create job postings and scouting emails that are appealing to job seekers.

[0355] The specification process by the specification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for receiving basic information and saving it in a database, a means for a user to select a job posting template and input job requirements, and a means for using a natural language processing engine to generate a draft of a job posting based on the received information. This enables companies to carry out their recruitment operations efficiently.

[0356] It also includes a means for publishing the final job posting on external job sites or the company's own recruitment page, and a means for the emotion engine to analyze user input in real time and suggest an appropriate emotional tone, making the job posting more attractive and effective for job seekers.

[0357] Furthermore, the system includes a means for the server to receive feedback and store it in a database, a means for analyzing the feedback information and adjusting the job posting generation algorithm, and a means for the emotion engine to analyze the emotional tone of the feedback and reflect it in the next job posting generation, thereby enabling the content of the job posting to be dynamically improved and optimized.

[0358] It also includes a means for users to input conditions for sending scout emails, a means for generating scout emails based on the conditions and sending them after adjustments by the emotion engine, and a means for tracking the sending results and providing feedback to the user, thereby enabling the generation of effective scout emails that match the target, and the confirmation and improvement of their effectiveness.

[0359] A "server" is a computer system that can be accessed remotely over a network and that stores, processes, and provides services to data.

[0360] A "terminal" is a device that is directly operated by a user and provides a user interface, such as a PC or smartphone.

[0361] A "user" is someone who operates the system and inputs data, such as a company's recruiter.

[0362] "Basic information" refers to basic data about a company, such as the company name, location, industry type, and management policy.

[0363] A "template" is a format used when creating a job posting, and is a pre-designed framework for each job type or department.

[0364] "Employment conditions" are employment conditions listed in a job posting, such as job type, required skills, years of experience, and salary range.

[0365] A "natural language processing engine" is an algorithm for analyzing and generating text data, and is a system that creates text using a generative AI model.

[0366] "Draft" is the first draft of the job posting that is generated, and is text that contains content that the user needs to correct and confirm.

[0367] An "emotion engine" is an algorithm that analyzes the emotional tone of text data and suggests appropriate expressions.

[0368] "Feedback" refers to opinions and reactions provided by users and job seekers, as well as data based on them.

[0369] An "algorithm" is a defined set of procedures or computational steps for solving a problem.

[0370] A "scout email" is a recruiting message sent to candidates who meet certain hiring criteria.

[0371] "Tracking" is the process of tracking and recording the results and responses to emails sent.

[0372] An embodiment of the present invention is a system for supporting a company's recruitment operations, in which a server, a terminal, a user, and an emotion engine work together. This system mainly provides a series of functions for receiving, storing, analyzing, generating, and publishing data.

[0373] First, when the server starts up, it opens an API endpoint and prepares for communication with the terminal. The hardware used here is a typical server computer, and the software includes a database management system (DBMS) and API management tools. The user enters basic company information (e.g., company name, location, industry type, management policy, etc.) through the terminal, and the server stores that information in a database.

[0374] Next, the user logs into the system and selects a job posting template. Templates are available in multiple formats and are categorized by job type and department. The user enters details such as the job type, desired skills, years of experience, and salary range according to the template they select. The server receives this information and generates a draft job posting using a natural language processing engine (e.g., GPT-3). This process utilizes a generative AI model. The generated draft is then provided to the user, who can review and modify it.

[0375] The user reviews the final version of the job posting and sends a publication request to the server. The server saves the job posting in a database, links it to external job sites and the company's own recruitment page, and publishes the information. The software used at this stage is an external API integration tool and web server software.

[0376] Receiving and analyzing feedback is also an important function. User feedback is received by the server and stored in a database. The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional tone of the feedback and use it to improve the job posting generation algorithm. The analyzed data is reflected in the next job posting generation.

[0377] In the future, the system will also provide the ability to generate and send scout emails. Users can enter the conditions for sending a scout email, and the server will generate the email based on the conditions. The emotion engine will then adjust the content appropriately. The server will then send the email, track the results, and provide feedback to the user.

[0378] Additionally, when users create a job posting, the emotion engine analyzes their input in real time and suggests appropriate emotional tones (e.g., proactive, friendly, professional, etc.), making job postings more engaging and effective.

[0379] Examples:

[0380] A human resources representative from a company logs into the system and enters basic company information, which the server then saves in a database. Next, they select the "Java Developer" template and enter their details. The server then uses a natural language processing engine to generate a draft job posting. The representative then reviews the final job posting and clicks the publish button, which instantly updates both the company's recruitment page and external job sites. Feedback is also received, analyzed by an emotion engine, and reflected in the next job posting, allowing the company to provide better job content.

[0381] Example prompts using a generative AI model:

[0382] "I'm looking to create a job posting for a Java developer. The required skills are Java and Spring Framework, with at least three years of experience and an annual salary range of ¥6,000,000 to ¥8,000,000."

[0383] With the above configuration, the present invention provides an efficient recruitment support system and improves the recruitment process of a company.

[0384] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0385] Step 1:

[0386] The server starts, opens an API endpoint, and prepares to communicate with the device.

[0387] Input: System startup command

[0388] Output: API endpoint status

[0389] Specific operation: The server accepts the start command and opens an API endpoint (URL) for network communication. The API endpoint waits for requests from the device.

[0390] Step 2:

[0391] The user inputs basic information through the terminal, and the server receives the information.

[0392] Input: Basic company information (company name, location, industry type, management policy, etc.)

[0393] Output: Acknowledgement message

[0394] Specific operation: A user fills in the basic information of their company in the input form on their device and clicks the submit button. The server receives this request, analyzes the data, and then generates and executes an SQL query to save it in the database.

[0395] Step 3:

[0396] The server stores the basic information in a database.

[0397] Input: Basic information data

[0398] Output: Message that saving to database is complete

[0399] Specific operation: The server saves the received basic information in the database management system. If the save is successful, it returns a message to the user indicating that the save was successful.

[0400] Step 4:

[0401] A user logs into the system from a terminal and selects a job posting template.

[0402] Input: Login information (username, password), template to select

[0403] Output: Template selection completion message

[0404] Specific behavior: A user enters their credentials on the login screen to access the system. A list of job templates is displayed, and the user selects the appropriate template.

[0405] Step 5:

[0406] Users enter details such as job title, desired skills, years of experience, and salary range.

[0407] Input: Job type, desired skills, years of experience, salary range

[0408] Output: Input completion message

[0409] Specific operation: The user fills in various details in the input form according to the template selected by the user. After completion, the input contents are sent to the server.

[0410] Step 6:

[0411] The server passes the received information to a natural language processing engine to generate a draft job posting.

[0412] Input: Job details

[0413] Output: Draft job posting

[0414] How it works: The server receives the job details and passes them to a generative AI model (e.g., GPT-3) in the form of a prompt. The natural language processing engine analyzes and generates the prompt to create a draft job posting.

[0415] Example prompt sentence:

[0416] "I'm looking to create a job posting for a Java developer. The required skills are Java and Spring Framework, with at least three years of experience and an annual salary range of ¥6,000,000 to ¥8,000,000."

[0417] Step 7:

[0418] The server returns the generated draft to the user for confirmation.

[0419] Input: Draft job posting

[0420] Output: Draft displayed to user

[0421] Specific operation: The server sends the generated draft to the terminal and displays it to the user, allowing the user to check the contents of the draft and make any necessary corrections.

[0422] Step 8:

[0423] The user reviews the final job posting and sends a publish request to the server.

[0424] Input: Last confirmed job posting

[0425] Output: Publishing request sent successfully message

[0426] Specific behavior: The user reviews the final job posting and clicks the "Publish" button to send a request to the server.

[0427] Step 9:

[0428] The server stores the job postings in a database and links them to external job sites and the company's own recruitment page for publication.

[0429] Input: Publishing request

[0430] Output: Publishing complete message

[0431] Specific operation: The server saves the job posting data in a database, and at the same time, links the job information to external job sites and the company's own recruitment page via API and publishes it immediately.

[0432] Step 10:

[0433] The user enters the feedback and the server receives it.

[0434] Input: Feedback information

[0435] Output: Feedback acknowledgement message

[0436] Specific operation: The user inputs responses and internal feedback from job seekers into the terminal and sends them to the server, which receives them and stores them in a database.

[0437] Step 11:

[0438] The server analyzes the feedback and uses an emotion engine to reflect it in the next job posting generation.

[0439] Input: Feedback information

[0440] Output: Analysis results and algorithm adjustment data

[0441] What it does: The server analyzes the feedback information, evaluates the emotional tone using an emotion engine, and adjusts the job posting generation algorithm based on the analysis results to reflect them in the next generation.

[0442] Step 12:

[0443] The user inputs the conditions for sending the scout mail, and the server generates and sends the scout mail.

[0444] Input: Scout mail conditions

[0445] Output: Generated scout email

[0446] Specific operation: The user inputs the target conditions for the scout email and sends it to the server. The server generates the scout email based on the conditions, adjusts the content using the emotion engine, and sends it.

[0447] Step 13:

[0448] The server tracks the results of the submission and provides feedback to the user.

[0449] Input: Scout email sending conditions

[0450] Output: Sending results and tracking data

[0451] Specific operation: The server tracks the results of sending scout emails, collects reactions and response data, analyzes it, and provides it to the user as feedback.

[0452] (Application example 2)

[0453] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0454] Conventional ad generation systems rely on manual creation of ad copy, which makes it difficult to properly appeal to target users, especially due to a lack of emotional tone adjustment. Furthermore, the process of receiving feedback and incorporating it into the ad generation algorithm is manual and inefficient. Furthermore, individual ads are not properly sent to target users, making it difficult to maximize advertising effectiveness.

[0455] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0456] In this invention, the server includes: [means for the server to receive basic information and store it in a database;] [means for the user to select an advertisement template and input advertisement conditions;] [means for the server to generate a draft advertisement using a natural language processing engine based on the information received;] [means for the user to check and revise the draft;] [means for the server to link and publish the final advertisement to an external advertising platform; and [means for analyzing the emotional tone of the advertisement content using an emotion engine and adjusting it as necessary.] This automates the entire process from advertisement generation to publication and feedback analysis, making it possible to appropriately adjust the emotional tone of the advertisement content. In addition, advertisements can be individually generated and sent based on the recipients of targeted advertisements, maximizing the effectiveness of advertising.

[0457] A "server" is a computer system that has the function of receiving data from client terminals and storing it in a database.

[0458] "Basic information" refers to information such as company name, location, industry type, and management policy that is necessary for creating advertisements and job offers.

[0459] A "database" is a system for systematically storing and managing received basic information and feedback information.

[0460] A "template" is an advertisement or job posting format based on a specific format that can be selected and used by a user.

[0461] "Advertising conditions" are information input as the content of the advertisement to be generated, and include, for example, the title, features, price, and the like.

[0462] A "natural language processing engine" is software that automatically generates drafts of advertisements and job postings based on received information.

[0463] A "draft" is an early version of an advertisement or job posting that can be reviewed and modified by the user.

[0464] "Advertising Platform" means an external service for publishing generated advertisements, including the Internet and mobile apps.

[0465] An "emotion engine" is software that analyzes the emotional tone of advertisements and job postings and adjusts it as needed.

[0466] "Feedback" refers to opinions and evaluations from users and viewers regarding advertisements and job postings.

[0467] "Targeted advertising" means advertising that is generated and delivered to an individual user based on specific criteria.

[0468] "Tracking" refers to tracking the results of sending advertisements and the status of receiving feedback.

[0469] This embodiment relates to an advertisement generation system, and in particular provides a system that combines a generation AI and an emotion engine. To implement the invention, the following main components are included:

[0470] 1. Server

[0471] 2. Client Terminal

[0472] 3. Emotion Engine

[0473] 4. Natural Language Processing Engine

[0474] 5. Database

[0475] 6. Advertising Platform

[0476] The server receives basic information sent from the client terminal and stores it in a database. Basic information includes the company name, location, industry type, management policy, etc. Based on this information, an ad template is selected and ad conditions are entered.

[0477] Using a client terminal, a user selects an ad template and enters the ad conditions, including detailed information such as the title, features, and price. Once the server receives this information, it uses a natural language processing engine to generate a draft ad. This draft is an initial version that the user can review and modify.

[0478] The emotional engine analyzes the emotional tone of the generated draft ad and adjusts it as necessary, ensuring that the ad content appeals appropriately to the target user. Once the user reviews the final version of the ad and sends a publishing request to the server, the server connects the final version of the ad to an external advertising platform and publishes it.

[0479] Furthermore, the server has the function of receiving feedback and storing it in a database. The feedback information is analyzed using an emotion engine and reflected in the ad generation algorithm. This process continuously improves the effectiveness of the ads.

[0480] Users can also input specific conditions when sending targeted ads, and the server will generate targeted ads based on these conditions, adjust them using the emotion engine, and send them. The results of these ads are tracked by the server and fed back to the user.

[0481] For example, if a user wants to generate an ad with the title "An exciting new product!" and the features "High-performance battery, easy-to-use UI, and 1-year warranty," they would use the following prompt:

[0482] prompt:

[0483] Use the company information below to generate copy that strikes a positive and friendly tone.

[0484] Company information: A company, large city, e-commerce industry, customer satisfaction focused

[0485] Advertisement details: Title: "An exciting new product!" Features: "High-performance battery, easy-to-use UI, 1-year warranty"

[0486] Desired tone: Positive and friendly

[0487] This allows users to easily generate ads using generative AI models and obtain ad copy that is tailored to the optimal emotional tone using an emotional engine.

[0488] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0489] Step 1:

[0490] The server receives the basic information and stores it in the database. This is the process of receiving information such as the company name, location, industry type, and management policy sent from the client terminal. The received basic information is stored in the database as is in JSON format. This process ensures that the company's basic information is properly stored in the system.

[0491] Input: Basic information such as company name, location, industry type, management policy, etc. (JSON format)

[0492] Data processing: Save basic information to a database

[0493] Output: Basic information stored in the database

[0494] Step 2:

[0495] The user selects an ad template and enters the ad conditions. The user logs in to the system using a client terminal and selects the desired template from the presented templates. Then, the user enters the necessary ad conditions (title, features, price, etc.).

[0496] Input: Template selection information, advertising conditions (title, features, price, etc.)

[0497] Data processing: converting user input into the appropriate format and sending it to the server

[0498] Output: Template selection information and ad conditions sent to the server

[0499] Step 3:

[0500] The server uses a natural language processing engine based on the received information to generate a draft ad. The server runs the natural language processing engine based on the entered ad conditions and the selected template. This automatically generates an initial draft.

[0501] Input: Template selection information, advertising conditions

[0502] Data processing: Draft generation using a natural language processing engine

[0503] Output: Initial Draft

[0504] Step 4:

[0505] The user checks and modifies the draft. The generated draft is checked on the client terminal and modified as necessary. The modified draft is then sent back to the server.

[0506] Input: Initial Draft

[0507] Data processing: User confirmation and correction

[0508] Output: Revised draft

[0509] Step 5:

[0510] The server uses an emotion engine to analyze the emotional tone of the ad content and adjust it accordingly. The emotion engine analyzes the emotional tone of the revised draft and adjusts it to the specified emotional tone (e.g., positive, friendly).

[0511] Input: Revised draft

[0512] Data processing: Emotional tone analysis and adjustment

[0513] Output: Final draft with emotional tone applied

[0514] Step 6:

[0515] The server links the final version of the advertisement to the advertising platform and publishes it. Upon receiving the publishing request, the server links the final draft to the specified advertising platform and publishes the advertisement.

[0516] Input: Final draft with emotional tone applied

[0517] Data processing: Collaboration with advertising platforms

[0518] Output: Published ad

[0519] Step 7:

[0520] The server receives the feedback and stores it in a database. It then analyzes the feedback information using an emotion engine and reflects it in the ad generation algorithm, which will result in more effective ad generation in the future.

[0521] Input: Feedback information

[0522] Data processing: storing and analyzing feedback information

[0523] Output: Improved ad generation algorithm

[0524] Step 8:

[0525] The user inputs the conditions for targeting the advertisement to be sent. The user inputs the conditions for targeting the advertisement to be sent using the client terminal.

[0526] Input: Criteria for sending targeted ads

[0527] Data processing: Send condition information to the server

[0528] Output: Condition information sent to the server

[0529] Step 9:

[0530] The server generates and sends targeted ads based on the conditions. The emotion engine adjusts the content of the ads and automatically sends them to the specified targets.

[0531] Input: Criteria for sending targeted ads

[0532] Data processing: Emotion engine adjustment, ad generation and delivery

[0533] Output: Targeted ads sent

[0534] Step 10:

[0535] The server tracks the results of the advertisements sent, collecting response data (number of views, number of clicks, etc.) and generating feedback based on that data.

[0536] Input: Targeted ads sent

[0537] Data processing: Reaction data collection and analysis

[0538] Output: Tracking data, feedback

[0539] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0540] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0541] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0542] [Second embodiment]

[0543] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0544] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0545] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0546] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0547] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0548] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0549] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0550] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0551] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0552] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0553] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0554] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0555] The embodiment of the present invention is a generative AI service that supports companies' recruitment operations. This system operates in cooperation with three parties: a server, a terminal, and a user.

[0556] 1. Registering and managing basic information

[0557] The server starts up and opens an API endpoint, preparing to establish communication with the device. The user enters basic information through the device, and the server stores that information in a database. This basic information includes the company name, location, industry type, management policy, etc.

[0558] Examples:

[0559] A human resources manager at an IT company logs into the system and enters basic information about the company (e.g., "XYZ Company, Tokyo, software development industry, corporate culture that values ​​innovation"), and the server receives this information and stores it in a database.

[0560] 2. Job posting generation

[0561] Users log in to the system from their terminal and select a job posting template. Templates are available in multiple formats for different job types and departments. They then enter details such as job type, desired skills, years of experience, and salary range according to the template they select.

[0562] The server receives this information and uses a natural language processing engine to generate a draft job posting, which is then provided to the user for review and modification.

[0563] Examples:

[0564] A human resources representative from the same IT company selects the "Java Developer" template and inputs the desired skills (Java, Spring Framework), years of experience (3+ years), salary range (annual salary 6-8 million yen), etc. The server then generates a draft, and the user can review the proposed content and make any necessary revisions.

[0565] 3. Publication of job postings

[0566] The user checks the final version of the job posting and sends a publishing request to the server. The server saves the job posting and publishes it on external job sites or the company's own recruitment page.

[0567] Examples:

[0568] Once the HR person has reviewed the draft and clicked the "Publish" button, the server saves the job posting and it is instantly updated on directly linked external job sites and the company's own recruitment page. This entire process is completed in just a few minutes, enabling rapid recruitment activities.

[0569] 4. Receiving and Analyzing Feedback

[0570] User feedback is received by the server and stored in a database, where it can be analyzed and used to improve the job posting generation algorithm.

[0571] Examples:

[0572] After publishing a job posting, users input reactions from job seekers and internal feedback (e.g., "few applicants," "the description is difficult to understand," etc.) into the system. This information is stored on the server and will be used the next time a job posting is generated.

[0573] 5. Generate and send scout emails (future feature)

[0574] The user inputs the conditions for the person to whom they want to send a scout email. The server generates and sends the scout email based on these conditions. The server tracks the sending results and provides feedback to the user.

[0575] Examples:

[0576] If a human resources officer wants to send a scouting email to a candidate who is a "full-stack engineer" and "able to work remotely," he or she enters those conditions into the system. The server automatically generates and sends scouting emails to candidates who meet the conditions. Later, the server tracks the response data to the scouting email and provides it to the human resources officer.

[0577] In this way, the present invention effectively and efficiently supports the recruitment operations of companies.

[0578] The processing flow will be explained below.

[0579] Step 1:

[0580] The server starts and opens an API endpoint, allowing devices to communicate with the server.

[0581] Step 2:

[0582] The terminal loads the user interface, allowing the user to access the system.

[0583] Step 3:

[0584] A user logs into the system through a terminal and enters a user ID and password.

[0585] Step 4:

[0586] The server checks the user's authentication information against the database, and if authentication is successful, displays a basic information input screen for the user.

[0587] Step 5:

[0588] The user enters basic information about the company (company name, location, industry, management policy, etc.) and clicks the "Submit" button.

[0589] Step 6:

[0590] The server receives the basic information entered and stores it in a database.

[0591] Step 7:

[0592] The user clicks the "Create a new job posting" button on their device and selects an appropriate template from a list of job posting templates.

[0593] Step 8:

[0594] The server receives the user's template selection request and displays the template form on the user interface.

[0595] Step 9:

[0596] The user enters job requirements such as job type, desired skills, experience, and salary range according to the template, and clicks the "Complete input" button.

[0597] Step 10:

[0598] The server receives the entered job requirements and temporarily stores them.

[0599] Step 11:

[0600] The server invokes a natural language processing engine and generates a draft job posting based on the received information.

[0601] Step 12:

[0602] The generated draft job posting is sent to the user's terminal and displayed on the user interface.

[0603] Step 13:

[0604] The user checks the draft content on the terminal and makes corrections as necessary.

[0605] Step 14:

[0606] After making corrections, the user clicks the "Confirm" button to approve the final version of the job posting.

[0607] Step 15:

[0608] The server stores approved job postings in a database and then links to external job sites and the company's own recruitment page via an API, making them publicly available.

[0609] Step 16:

[0610] Users conduct job postings and input responses from job seekers and internal feedback into the system.

[0611] Step 17:

[0612] The server receives the feedback information and stores it in a database.

[0613] Step 18:

[0614] The server analyzes the feedback and uses it to improve the job posting generation algorithm.

[0615] Step 19:

[0616] The user enters the conditions for the person to whom the scout email is to be sent and clicks the "Conditions input complete" button.

[0617] Step 20:

[0618] The server receives the conditions and generates a scout email.

[0619] Step 21:

[0620] The server sends the generated scout email to the candidate and tracks the sending results.

[0621] Step 22:

[0622] The server analyzes the tracking results and provides feedback to the user through a user interface.

[0623] Example 1

[0624] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0625] Traditional recruitment processes have the problem of taking a lot of time and effort, from creating a job posting to publishing it and receiving feedback. Furthermore, if the job posting does not receive a positive response from job seekers due to inappropriate content, it takes a lot of effort to revise the information and republish it. Furthermore, there is no efficient way to generate and send scouting emails, making it difficult to effectively recruit talent.

[0626] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0627] In this invention, the server includes: [means for receiving basic information and saving it in a database;] [means for selecting a job posting template and inputting job requirements;] [means for generating a draft job posting using a natural language processing engine based on the received information;] [means for publishing the final job posting in conjunction with an external job site or the company's own recruitment page;] [means for opening an API endpoint and establishing communication with a terminal;] [means for receiving feedback and saving it in a database;] [means for analyzing feedback information and adjusting the job posting generation algorithm;] [means for inputting conditions for sending scout emails;] [means for generating and sending scout emails based on the conditions; and [means for tracking the sending results.] This makes it possible to effectively and efficiently carry out a series of recruitment tasks, from creating and publishing a job posting, analyzing feedback, and generating and sending scout emails.

[0628] "Basic information" refers to basic data about a company, such as the company name, location, industry type, and management policy.

[0629] "User" refers to a person who uses this system to manage and operate job information, such as a company's recruiter or human resources officer.

[0630] "Server" refers to a computer that forms the core part of the system and performs various processes such as receiving, storing, analyzing, and publishing data.

[0631] "Terminal" refers to a device used by a user, such as a computer, tablet, or smartphone.

[0632] "Database" refers to organized data storage for storing basic information, job postings, feedback information, etc.

[0633] A "template" is a prescribed format used when creating job postings, prepared for each job type and department.

[0634] "Job requirements" refers to the detailed requirements such as the job type, required skills, years of experience, and salary range listed in the job posting.

[0635] A "natural language processing engine" is an AI model that generates and analyzes sentences, and is a technology that is particularly used to draft job postings.

[0636] "External job sites" refer to websites where companies post job openings and which job seekers can access.

[0637] An "API endpoint" is a connection point between a server and a terminal for communication, and refers to the URL or URI used to send and receive data.

[0638] "Feedback" refers to opinions and impressions from users and job seekers regarding the content of and reactions to job postings.

[0639] A "scouting email" refers to an email sent by a company to recruit job seekers who meet certain criteria.

[0640] "Conditions" refers to the requirements such as job type, skills, and working style of the candidate to whom the scouting email is sent.

[0641] "Tracking" refers to the process of tracking and recording the results of sending scouting emails and the responses to them.

[0642] MODE FOR CARRYING OUT THE INVENTION

[0643] This invention is a generative AI service to support companies' recruitment operations. This system operates in cooperation with three parties: a server, a terminal, and a user, and is implemented according to the following procedure.

[0644] Starting the server and opening the API endpoint

[0645] The server is launched on an AWS EC2 instance, running the Apache Web Server, and an API endpoint is set up, ready to establish communication with the device.

[0646] Registering and managing basic information

[0647] Users access the system through a terminal and enter basic information about their company, including the company name, location, industry type, management policy, etc. The terminal then sends the entered information to the server, which stores it in a database.

[0648] Example: A human resources manager at a manufacturing company logs into the system and enters basic information about the company (e.g., "ABC Manufacturing, Aichi Prefecture, manufacturing industry, corporate culture that emphasizes quality"). The server receives this information and stores it in a database.

[0649] Job posting generation

[0650] Users log in to the system from their terminal and select a job posting template prepared by job type or department. They then enter details such as the job type, desired skills, years of experience, salary range, etc. This information is sent to the server, which uses a generative AI model (e.g., OpenAI GPT-3) to generate a draft job posting.

[0651] Example: A human resources professional selects a "Systems Engineer" template and inputs the required skills (e.g., Python, database management), years of experience (e.g., 5+ years), salary range (e.g., annual salary of 5 million to 7 million yen), etc. The server receives this and uses a generative AI model to generate a draft job posting. The user reviews the draft and makes any necessary revisions.

[0652] Example prompt sentence:

[0653] What skills are you looking for in a "Systems Engineer"? Also include the required years of experience, salary range, and desired characteristics of the company.

[0654] Publication of job postings

[0655] The user checks the final version of the job posting and sends a publishing request to the server, which saves the job posting and publishes it via an external job posting site or the company's own recruitment page.

[0656] Example: A human resources representative makes a final check and clicks the "Publish" button. The server saves the final job posting in the database and immediately updates it on external job sites and the company's own recruitment page, enabling rapid recruitment.

[0657] Receiving and analyzing feedback

[0658] After the job posting is published, the server receives feedback from users and stores it in a database. The server can analyze the feedback information and use it to adjust the job posting generation algorithm.

[0659] Example: After publishing a job posting, users enter feedback from job seekers and internal sources (e.g., "few applicants," "the job posting content is unclear") into the system. The server receives this feedback, stores it in the database, and reflects it the next time a job posting is generated.

[0660] Generate and send scout emails (future feature)

[0661] The user inputs the conditions (job type, skills, working style, etc.) of the person to whom they want to send a scout email. The server generates and sends the scout email based on this information. The server tracks the sending results and provides feedback to the user.

[0662] Example: If a human resources manager wants to send a scouting email to a candidate who is a "mobile app developer" and "able to work remotely," he or she enters those conditions into the system. The server uses a generative AI model to automatically generate a scouting email and sends it to the candidate. The server then tracks the response data and provides it to the human resources manager.

[0663] In this way, the present invention can effectively and efficiently support a company's recruitment operations.

[0664] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0665] Specific processing flow of the system

[0666] Step 1: Start the server and open the API endpoint

[0667] The server is launched on an AWS EC2 instance, running Apache Web Server, and an API endpoint is opened to support companies' recruitment processes.

[0668] Input: Server configuration information and startup script

[0669] Output: API endpoint URL (e.g. https: / / example.com / api)

[0670] Specific operation: The server executes the startup script, which starts the web server and API endpoint, allowing communication with the device.

[0671] Step 2: Register basic information

[0672] Users access the system through a terminal and enter basic company information (company name, location, industry type, management policy).

[0673] Input: Basic company information (company name, location, industry type, management policy, etc.)

[0674] Output: Basic information input data

[0675] What happens: A user accesses the login screen using a browser and enters basic information.

[0676] Step 3: Submit and save basic information

[0677] The device sends basic information to the server, which stores the received information in a database.

[0678] Input: Basic information input data

[0679] Output: Status of saving to database

[0680] Specific operation: The device sends the entered basic information to the server, which verifies the data and saves it in the database.

[0681] Step 4: Select a job template

[0682] Users log in to the system from their terminal and select job posting templates prepared for each job type and department.

[0683] Input: Login credentials, Job template list

[0684] Output: Selected template

[0685] What happens: A user logs into the system and selects from multiple templates.

[0686] Step 5: Enter job requirements

[0687] Based on the template you select, you enter details such as job title, desired skills, years of experience, and salary range.

[0688] Input: Selected template, job requirements (job type, skills, years of experience, etc.)

[0689] Output: Job requirements data entered

[0690] Specific behavior: The user fills out a form with detailed job requirements according to the template.

[0691] Step 6: Submit your job requirements

[0692] The terminal sends the job requirements to the server.

[0693] Input: Job requirements data entered

[0694] Output: Status of sending to server

[0695] Specific operation: The terminal sends the entered job requirements to the server.

[0696] Step 7: Generate a draft job posting

[0697] Based on the job requirements received by the server, a draft job posting is generated using a generative AI model.

[0698] Input: Received job requirements data

[0699] Output: Draft job posting

[0700] Specific operation: The server uses a generative AI model (e.g., OpenAI GPT-3) to generate a draft job posting based on the job requirements.

[0701] Step 8: Review and revise your draft

[0702] The user reviews the draft job posting and makes any necessary corrections.

[0703] Input: Draft job posting

[0704] Output: Final revised job posting

[0705] What happens: The user reviews the draft provided by the server and makes any necessary corrections.

[0706] Step 9: Submit a publishing request

[0707] The user reviews the final job posting and sends a publish request to the server.

[0708] Input: Final revised job posting

[0709] Output: Publish request

[0710] Specific behavior: The user clicks the "Publish" button to send a publication request.

[0711] Step 10: Publish your job posting

[0712] The server stores the final job posting in a database and publishes it on external job sites or the company's own recruitment page.

[0713] Input: Publication request, final job posting

[0714] Output: Job posting publication status

[0715] Specific operation: The server saves the final job posting in the database and reflects it on external job sites and the company's own recruitment page via API.

[0716] Step 11: Receiving feedback

[0717] After the job posting is published, feedback from users is received by the server and stored in a database.

[0718] Input: User feedback

[0719] Output: Stored feedback data

[0720] Specific operation: After a job posting is published, the user enters feedback information into the system, and that information is sent to the server.

[0721] Step 12: Analyze feedback

[0722] The server analyzes the feedback information and uses it to adjust the job posting generation algorithm.

[0723] Input: Stored feedback data

[0724] Output: Analysis results and algorithm adjustments

[0725] Specific operation: The server analyzes the received feedback data and adjusts the algorithm to reflect it when generating the next job posting.

[0726] Step 13: Generate and send scout emails (future feature)

[0727] The user inputs the criteria for the person to whom they want to send a scouting email. The server generates a scouting email based on the criteria and sends it to the candidate. The server tracks the sending results and provides feedback to the user.

[0728] Input: Target conditions for scouting emails (job type, skills, working conditions, etc.)

[0729] Output: Scout email sending status and tracking data

[0730] Specific operation: The user inputs the target conditions, and the server uses the generative AI model to generate and send scouting emails. The server tracks the sending results and provides feedback.

[0731] (Application example 1)

[0732] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0733] Corporate recruitment is a complex process that requires a great deal of effort and time. Particularly in factories, quickly and efficiently recruiting the right talent is directly linked to production efficiency. However, delays in creating and publishing job postings and analyzing feedback can lead to the risk of missing out on key talent. Furthermore, on-site talent management is even more difficult, complicating the management work of dedicated personnel. There is a need for a system that can solve these issues and achieve a more efficient and accurate recruitment process.

[0734] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0735] In this invention, the server includes: [means for the server to receive basic information and save it in a database;] [means for the user to select a job posting template and input job requirements;] [means for the server to use a generative AI model based on the information received to generate a draft of the job posting;] [means for the user to check and correct the draft;] [means for the server to link and publish the final version of the job posting to an external recruitment media;] [means for the robot to support employee information management within the factory; and [means for the robot to input job information using a terminal set up and send it to the server. This enables more efficient human resource management within the factory and the generation and publication of accurate job information.

[0736] A "server" is a central system for data management and processing.

[0737] "Basic information" refers to basic information such as the company's name, location, industry, and management policy.

[0738] A "database" is a system for storing and managing data such as basic information and feedback.

[0739] "User" refers to the company personnel and administrators who operate the system.

[0740] A "template" is a predefined format for quickly creating job postings.

[0741] "Job requirements" refer to details such as the desired job type, required skills, years of experience, salary range, etc.

[0742] A "generative AI model" is an artificial intelligence technology that generates a draft job posting based on input information.

[0743] "Draft" refers to the first unfinished job posting generated.

[0744] "External recruitment media" refers to online platforms and sites that disseminate job information to job seekers.

[0745] A "robot" is an automated mechanical device used in factories.

[0746] "Information management" refers to the process of organizing and storing basic employee information and recruitment data.

[0747] A "terminal" is an electronic device that allows a user to input information.

[0748] System Program Overview

[0749] This invention is a "smart robot personnel assistant" system that streamlines personnel management within factories and supports the accurate creation and publication of job information. This system operates in cooperation with a server, terminals, and robots.

[0750] Server Features

[0751] The server is a central system that manages and processes data. It receives basic information (company name, location, industry, management policy, etc.) entered by users through their devices and stores it in a database. It also uses a generative AI model to generate a draft job posting based on the template selected by the user and the entered job requirements (job type, skills, years of experience, salary range, etc.). The generated draft is then reviewed and revised by the user, and the final version is published by the server to external job media.

[0752] Robot Functions

[0753] The robot supports human resource management within the factory. It uses a set-up terminal to input job information and send it to a server, which can then be used directly by on-site personnel within the factory.

[0754] Data processing and calculation

[0755] 1. Register and manage basic information:

[0756] The user enters basic company information from the terminal, and the server receives it and stores it in a database.

[0757] Hardware used: Device (PC, tablet, etc.)

[0758] Software used: Python, Requests library, RESTful API, database management system (MySQL, etc.)

[0759] 2. Generate and publish job posting:

[0760] The user inputs the job posting template and detailed requirements from the terminal, and the server uses a generative AI model (e.g., GPT-3) to generate a draft job posting.

[0761] Hardware used: Server

[0762] Software used: Generative AI models (e.g., GPT-3)

[0763] The generated draft is then checked and revised by the user, and the server publishes the final version to external recruitment media.

[0764] 3. Feedback analysis:

[0765] The server stores the feedback received after publication in a database and analyzes it to help generate the next job posting.

[0766] Software used: Feedback Analysis Engine

[0767] Specific examples

[0768] Basic information registration:

[0769] The factory's human resources staff enters basic information such as "XYZ Manufacturing, Tokyo, manufacturing industry, emphasis on quality and efficiency" into a terminal, and the server stores this information in a database.

[0770] Job listing generation:

[0771] The same person selects the "Line Operator" template and enters the desired skills of "Machine Operation" and "Quality Control," the number of years of experience of "2 years or more," and the salary range of "3 million to 5 million yen per year." Based on this, the server uses a generative AI model to generate the following prompt:

[0772] Please create a job posting for a line operator. The required skills are "machine operation" and "quality control," the required years of experience are "2 years or more," and the salary range is "annual salary of 3 million to 5 million yen."

[0773] The resulting draft is then reviewed by a human resources representative and revised as necessary. The final job posting is then published by the server to external recruitment media.

[0774] This will enable more efficient human resource management within the factory and generate accurate job information.

[0775] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0776] Step 1:

[0777] The user enters basic information through the terminal and sends it to the server. The basic information entered includes the company name, location, industry, management policy, etc. The server receives this information and stores it in the database. The input data is saved in the appropriate table in the database.

[0778] Step 2:

[0779] The user selects a job posting template from their device and enters the job requirements, including the job type, desired skills, years of experience, salary range, etc. This input data is sent to the server, which receives it and prepares it as input prompts for the generative AI model.

[0780] Step 3:

[0781] Based on the job requirements and template information received by the server, a generative AI model (e.g., GPT-3) is used to generate a draft job posting. A prompt is input to the generative AI model, which then generates a draft job posting. The generated draft is saved on the server and provided to the user.

[0782] Step 4:

[0783] The user checks the draft on their device and makes any necessary corrections. The information corrected by the user is sent to the server, which updates the draft again and generates the final version. This final version of the job posting is also saved in the database.

[0784] Step 5:

[0785] The server connects the final job posting to an external recruitment site and publishes it. Specifically, the job information is sent to the external site using methods such as API or FTP. Information on successful publication of the job posting is also saved in the database as tracking information.

[0786] Step 6:

[0787] The server receives external feedback and stores it in a database. The feedback is processed by the analytics engine and used to adjust the job posting generation algorithm for the next job posting, thereby improving the job posting process for future jobs.

[0788] Step 7:

[0789] When a user wants to send a scout email based on specific criteria, they input the criteria from their device and send it to the server. The server searches for information that matches the criteria, then uses a generative AI model to generate and send the scout email. The server tracks the results of the email and provides feedback to the user.

[0790] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0791] An embodiment of the present invention is a system that combines a generative AI service that supports corporate recruitment operations with an emotion engine. This system operates in cooperation with a server, a terminal, a user, and the emotion engine.

[0792] 1. Registering and managing basic information

[0793] The server starts up, opens an API endpoint, and prepares to communicate with the device. The user enters basic information through the device, and the server stores that information in a database. This basic information includes the company name, location, industry type, management policy, etc.

[0794] Examples:

[0795] A human resources manager at an IT company logs into the system and enters basic information about the company (for example, "ABC Company, Tokyo, software development industry, corporate culture that values ​​innovation"). The server receives this information and stores it in a database.

[0796] 2. Job posting generation

[0797] Users log in to the system from their terminal and select a job posting template. Templates are available in multiple formats for different job types and departments. They then enter details such as job type, desired skills, years of experience, and salary range according to the template they select.

[0798] The server receives this information and uses a natural language processing engine to generate a draft job posting, which is then provided to the user for review and modification.

[0799] Examples:

[0800] A human resources representative from the same IT company selects the "Java Developer" template and inputs the desired skills (Java, Spring Framework), years of experience (3+ years), salary range (annual salary 6-8 million yen), etc. The server then generates a draft, and the user can review the proposed content and make any necessary revisions.

[0801] 3. Publication of job postings

[0802] The user checks the final version of the job posting and sends a publishing request to the server. The server saves the job posting and publishes it on external job sites or the company's own recruitment page.

[0803] Examples:

[0804] Once the HR person has reviewed the draft and clicked the "Publish" button, the server saves the job posting and it is instantly updated on linked external job sites and the company's own recruitment page. This entire process is completed in just a few minutes, enabling quick recruitment activities.

[0805] 4. Receiving and Analyzing Feedback

[0806] User feedback is received by the server and stored in a database. The server analyzes the feedback and can use it to improve the job posting generation algorithm. An emotion engine is used to analyze the emotional tone of the feedback and identify areas for further improvement.

[0807] Examples:

[0808] After publishing a job posting, users input reactions from job seekers and internal feedback (for example, "few applicants" or "the description is difficult to understand") into the system. This information is stored on the server and analyzed using an emotion engine. The results of the analysis are reflected the next time a job posting is generated.

[0809] 5. Generate and send scout emails (future feature)

[0810] The user inputs the criteria for the person to whom they want to send a scout email. The server generates and sends the scout email based on this information. The emotion engine adjusts the content of the scout email appropriately. The server tracks the sending results and provides feedback to the user.

[0811] Examples:

[0812] If a human resources manager wants to send a scouting email to a candidate who is a "full-stack engineer" and "able to work remotely," he or she can enter those criteria into the system. The server automatically generates a scouting email to candidates who fit the criteria, and the emotion engine adjusts the content before sending it. Later, the server tracks the response data to the scouting email and provides it to the human resources manager.

[0813] 6. Adjusting the emotional tone when generating job postings

[0814] When users create a job posting, the emotion engine analyzes their input in real time and suggests appropriate emotional tones (e.g., proactive, friendly, professional, etc.), making job postings more engaging and effective.

[0815] Examples:

[0816] As users enter their skills, years of experience, and other information into a job posting template, the emotion engine analyzes the information and displays suggestions such as "use a more positive tone." This makes the job posting more likely to catch the attention of job seekers.

[0817] In this way, the present invention effectively and efficiently supports companies' recruitment operations, and realizes the generation of sophisticated job postings and scouting emails that take into account the user's emotions.

[0818] The processing flow will be explained below.

[0819] Step 1:

[0820] The server starts and opens an API endpoint, allowing devices to communicate with the server.

[0821] Step 2:

[0822] The terminal loads the user interface, allowing the user to access the system.

[0823] Step 3:

[0824] A user logs into the system through a terminal and enters a user ID and password.

[0825] Step 4:

[0826] The server checks the user's authentication information against the database, and if authentication is successful, displays a basic information input screen for the user.

[0827] Step 5:

[0828] The user enters basic information about the company (company name, location, industry, management policy, etc.) and clicks the "Submit" button.

[0829] Step 6:

[0830] The server receives the basic information entered and stores it in a database.

[0831] Step 7:

[0832] The user clicks the "Create a new job posting" button on their device and selects an appropriate template from a list of job posting templates.

[0833] Step 8:

[0834] The server receives the user's template selection request and displays the template form on the user interface.

[0835] Step 9:

[0836] The user enters job requirements such as job type, desired skills, experience, and salary range according to the template, and clicks the "Complete input" button.

[0837] Step 10:

[0838] The server receives the entered job requirements and temporarily stores them.

[0839] Step 11:

[0840] The server invokes a natural language processing engine and generates a draft job posting based on the received information.

[0841] Step 12:

[0842] The generated draft job posting is sent to the user's terminal and displayed on the user interface.

[0843] Step 13:

[0844] The user checks the draft content on the terminal and makes corrections as necessary.

[0845] Step 14:

[0846] The emotion engine analyzes the user's input and suggests an appropriate emotional tone (e.g., proactive, friendly, professional, etc.).

[0847] Step 15:

[0848] The user considers the emotional tone suggestions and modifies the content of the job posting.

[0849] Step 16:

[0850] After making corrections, the user clicks the "Confirm" button to approve the final version of the job posting.

[0851] Step 17:

[0852] The server stores approved job postings in a database and then links to external job sites and the company's own recruitment page via an API, making them publicly available.

[0853] Step 18:

[0854] Users conduct job postings and input responses from job seekers and internal feedback into the system.

[0855] Step 19:

[0856] The server receives the feedback information and stores it in a database.

[0857] Step 20:

[0858] The server analyzes the feedback and uses it to improve the job posting generation algorithm.

[0859] Step 21:

[0860] An emotional engine analyzes the emotional tone of the feedback and identifies areas for improvement.

[0861] Step 22:

[0862] The user enters the conditions for the person to whom the scout email is to be sent and clicks the "Conditions input complete" button.

[0863] Step 23:

[0864] The server receives the conditions and generates a scout email.

[0865] Step 24:

[0866] The emotional engine analyzes the content of the scouting email and adjusts the appropriate emotional tone.

[0867] Step 25:

[0868] The server sends the generated scout email to the candidate.

[0869] Step 26:

[0870] The server tracks the results of the transmission and provides feedback to the user through the user interface.

[0871] Example 2

[0872] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0873] In corporate recruitment processes, there are issues with the inefficiency of creating and publishing job postings, receiving and analyzing feedback, and creating and sending scouting emails. In particular, job postings and scouting emails are created only in a standardized format, and there is a lack of a system for collecting and analyzing applicant responses to improve the content. Furthermore, there is a lack of emotional tone adjustment to create job postings and scouting emails that are appealing to job seekers.

[0874] The specification process by the specification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for receiving basic information and saving it in a database, a means for a user to select a job posting template and input job requirements, and a means for using a natural language processing engine to generate a draft of a job posting based on the received information. This enables companies to carry out their recruitment operations efficiently.

[0875] It also includes a means for publishing the final job posting on external job sites or the company's own recruitment page, and a means for the emotion engine to analyze user input in real time and suggest an appropriate emotional tone, making the job posting more attractive and effective for job seekers.

[0876] Furthermore, the system includes a means for the server to receive feedback and store it in a database, a means for analyzing the feedback information and adjusting the job posting generation algorithm, and a means for the emotion engine to analyze the emotional tone of the feedback and reflect it in the next job posting generation, thereby enabling the content of the job posting to be dynamically improved and optimized.

[0877] It also includes a means for users to input conditions for sending scout emails, a means for generating scout emails based on the conditions and sending them after adjustments by the emotion engine, and a means for tracking the sending results and providing feedback to the user, thereby enabling the generation of effective scout emails that match the target, and the confirmation and improvement of their effectiveness.

[0878] A "server" is a computer system that can be accessed remotely over a network and that stores, processes, and provides services to data.

[0879] A "terminal" is a device that is directly operated by a user and provides a user interface, such as a PC or smartphone.

[0880] A "user" is someone who operates the system and inputs data, such as a company's recruiter.

[0881] "Basic information" refers to basic data about a company, such as the company name, location, industry type, and management policy.

[0882] A "template" is a format used when creating a job posting, and is a pre-designed framework for each job type or department.

[0883] "Employment conditions" are employment conditions listed in a job posting, such as job type, required skills, years of experience, and salary range.

[0884] A "natural language processing engine" is an algorithm for analyzing and generating text data, and is a system that creates text using a generative AI model.

[0885] "Draft" is the first draft of the job posting that is generated, and is text that contains content that the user needs to correct and confirm.

[0886] An "emotion engine" is an algorithm that analyzes the emotional tone of text data and suggests appropriate expressions.

[0887] "Feedback" refers to opinions and reactions provided by users and job seekers, as well as data based on them.

[0888] An "algorithm" is a defined set of procedures or computational steps for solving a problem.

[0889] A "scout email" is a recruiting message sent to candidates who meet certain hiring criteria.

[0890] "Tracking" is the process of tracking and recording the results and responses to emails sent.

[0891] An embodiment of the present invention is a system for supporting a company's recruitment operations, in which a server, a terminal, a user, and an emotion engine work together. This system mainly provides a series of functions for receiving, storing, analyzing, generating, and publishing data.

[0892] First, when the server starts up, it opens an API endpoint and prepares for communication with the terminal. The hardware used here is a typical server computer, and the software includes a database management system (DBMS) and API management tools. The user enters basic company information (e.g., company name, location, industry type, management policy, etc.) through the terminal, and the server stores that information in a database.

[0893] Next, the user logs into the system and selects a job posting template. Templates are available in multiple formats and are categorized by job type and department. The user enters details such as the job type, desired skills, years of experience, and salary range according to the template they select. The server receives this information and generates a draft job posting using a natural language processing engine (e.g., GPT-3). This process utilizes a generative AI model. The generated draft is then provided to the user, who can review and modify it.

[0894] The user reviews the final version of the job posting and sends a publication request to the server. The server saves the job posting in a database, links it to external job sites and the company's own recruitment page, and publishes the information. The software used at this stage is an external API integration tool and web server software.

[0895] Receiving and analyzing feedback is also an important function. User feedback is received by the server and stored in a database. The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional tone of the feedback and use it to improve the job posting generation algorithm. The analyzed data is reflected in the next job posting generation.

[0896] In the future, the system will also provide the ability to generate and send scout emails. Users can enter the conditions for sending a scout email, and the server will generate the email based on the conditions. The emotion engine will then adjust the content appropriately. The server will then send the email, track the results, and provide feedback to the user.

[0897] Additionally, when users create a job posting, the emotion engine analyzes their input in real time and suggests appropriate emotional tones (e.g., proactive, friendly, professional, etc.), making job postings more engaging and effective.

[0898] Examples:

[0899] A human resources representative from a company logs into the system and enters basic company information, which the server then saves in a database. Next, they select the "Java Developer" template and enter their details. The server then uses a natural language processing engine to generate a draft job posting. The representative then reviews the final job posting and clicks the publish button, which instantly updates both the company's recruitment page and external job sites. Feedback is also received, analyzed by an emotion engine, and reflected in the next job posting, allowing the company to provide better job content.

[0900] Example prompts using a generative AI model:

[0901] "I'm looking to create a job posting for a Java developer. The required skills are Java and Spring Framework, with at least three years of experience and an annual salary range of ¥6,000,000 to ¥8,000,000."

[0902] With the above configuration, the present invention provides an efficient recruitment support system and improves the recruitment process of a company.

[0903] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0904] Step 1:

[0905] The server starts, opens an API endpoint, and prepares to communicate with the device.

[0906] Input: System startup command

[0907] Output: API endpoint status

[0908] Specific operation: The server accepts the start command and opens an API endpoint (URL) for network communication. The API endpoint waits for requests from the device.

[0909] Step 2:

[0910] The user inputs basic information through the terminal, and the server receives the information.

[0911] Input: Basic company information (company name, location, industry type, management policy, etc.)

[0912] Output: Acknowledgement message

[0913] Specific operation: A user fills in the basic information of their company in the input form on their device and clicks the submit button. The server receives this request, analyzes the data, and then generates and executes an SQL query to save it in the database.

[0914] Step 3:

[0915] The server stores the basic information in a database.

[0916] Input: Basic information data

[0917] Output: Message that saving to database is complete

[0918] Specific operation: The server saves the received basic information in the database management system. If the save is successful, it returns a message to the user indicating that the save was successful.

[0919] Step 4:

[0920] A user logs into the system from a terminal and selects a job posting template.

[0921] Input: Login information (username, password), template to select

[0922] Output: Template selection completion message

[0923] Specific behavior: A user enters their credentials on the login screen to access the system. A list of job templates is displayed, and the user selects the appropriate template.

[0924] Step 5:

[0925] Users enter details such as job title, desired skills, years of experience, and salary range.

[0926] Input: Job type, desired skills, years of experience, salary range

[0927] Output: Input completion message

[0928] Specific operation: The user fills in various details in the input form according to the template selected by the user. After completion, the input contents are sent to the server.

[0929] Step 6:

[0930] The server passes the received information to a natural language processing engine to generate a draft job posting.

[0931] Input: Job details

[0932] Output: Draft job posting

[0933] How it works: The server receives the job details and passes them to a generative AI model (e.g., GPT-3) in the form of a prompt. The natural language processing engine analyzes and generates the prompt to create a draft job posting.

[0934] Example prompt sentence:

[0935] "I'm looking to create a job posting for a Java developer. The required skills are Java and Spring Framework, with at least three years of experience and an annual salary range of ¥6,000,000 to ¥8,000,000."

[0936] Step 7:

[0937] The server returns the generated draft to the user for confirmation.

[0938] Input: Draft job posting

[0939] Output: Draft displayed to user

[0940] Specific operation: The server sends the generated draft to the terminal and displays it to the user, allowing the user to check the contents of the draft and make any necessary corrections.

[0941] Step 8:

[0942] The user reviews the final job posting and sends a publish request to the server.

[0943] Input: Last confirmed job posting

[0944] Output: Publishing request sent successfully message

[0945] Specific behavior: The user reviews the final job posting and clicks the "Publish" button to send a request to the server.

[0946] Step 9:

[0947] The server stores the job postings in a database and links them to external job sites and the company's own recruitment page for publication.

[0948] Input: Publishing request

[0949] Output: Publishing complete message

[0950] Specific operation: The server saves the job posting data in a database, and at the same time, links the job information to external job sites and the company's own recruitment page via API and publishes it immediately.

[0951] Step 10:

[0952] The user enters the feedback and the server receives it.

[0953] Input: Feedback information

[0954] Output: Feedback acknowledgement message

[0955] Specific operation: The user inputs responses and internal feedback from job seekers into the terminal and sends them to the server, which receives them and stores them in a database.

[0956] Step 11:

[0957] The server analyzes the feedback and uses an emotion engine to reflect it in the next job posting generation.

[0958] Input: Feedback information

[0959] Output: Analysis results and algorithm adjustment data

[0960] What it does: The server analyzes the feedback information, evaluates the emotional tone using an emotion engine, and adjusts the job posting generation algorithm based on the analysis results to reflect them in the next generation.

[0961] Step 12:

[0962] The user inputs the conditions for sending the scout mail, and the server generates and sends the scout mail.

[0963] Input: Scout mail conditions

[0964] Output: Generated scout email

[0965] Specific operation: The user inputs the target conditions for the scout email and sends it to the server. The server generates the scout email based on the conditions, adjusts the content using the emotion engine, and sends it.

[0966] Step 13:

[0967] The server tracks the results of the submission and provides feedback to the user.

[0968] Input: Scout email sending conditions

[0969] Output: Sending results and tracking data

[0970] Specific operation: The server tracks the results of sending scout emails, collects reactions and response data, analyzes it, and provides it to the user as feedback.

[0971] (Application example 2)

[0972] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0973] Conventional ad generation systems rely on manual creation of ad copy, which makes it difficult to properly appeal to target users, especially due to a lack of emotional tone adjustment. Furthermore, the process of receiving feedback and incorporating it into the ad generation algorithm is manual and inefficient. Furthermore, individual ads are not properly sent to target users, making it difficult to maximize advertising effectiveness.

[0974] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0975] In this invention, the server includes: [means for the server to receive basic information and store it in a database;] [means for the user to select an advertisement template and input advertisement conditions;] [means for the server to generate a draft advertisement using a natural language processing engine based on the information received;] [means for the user to check and revise the draft;] [means for the server to link and publish the final advertisement to an external advertising platform; and [means for analyzing the emotional tone of the advertisement content using an emotion engine and adjusting it as necessary.] This automates the entire process from advertisement generation to publication and feedback analysis, making it possible to appropriately adjust the emotional tone of the advertisement content. In addition, advertisements can be individually generated and sent based on the recipients of targeted advertisements, maximizing the effectiveness of advertising.

[0976] A "server" is a computer system that has the function of receiving data from client terminals and storing it in a database.

[0977] "Basic information" refers to information such as company name, location, industry type, and management policy that is necessary for creating advertisements and job offers.

[0978] A "database" is a system for systematically storing and managing received basic information and feedback information.

[0979] A "template" is an advertisement or job posting format based on a specific format that can be selected and used by a user.

[0980] "Advertising conditions" are information input as the content of the advertisement to be generated, and include, for example, the title, features, price, and the like.

[0981] A "natural language processing engine" is software that automatically generates drafts of advertisements and job postings based on received information.

[0982] A "draft" is an early version of an advertisement or job posting that can be reviewed and modified by the user.

[0983] "Advertising Platform" means an external service for publishing generated advertisements, including the Internet and mobile apps.

[0984] An "emotion engine" is software that analyzes the emotional tone of advertisements and job postings and adjusts it as needed.

[0985] "Feedback" refers to opinions and evaluations from users and viewers regarding advertisements and job postings.

[0986] "Targeted advertising" means advertising that is generated and delivered to an individual user based on specific criteria.

[0987] "Tracking" refers to tracking the results of sending advertisements and the status of receiving feedback.

[0988] This embodiment relates to an advertisement generation system, and in particular provides a system that combines a generation AI and an emotion engine. To implement the invention, the following main components are included:

[0989] 1. Server

[0990] 2. Client Terminal

[0991] 3. Emotion Engine

[0992] 4. Natural Language Processing Engine

[0993] 5. Database

[0994] 6. Advertising Platform

[0995] The server receives basic information sent from the client terminal and stores it in a database. Basic information includes the company name, location, industry type, management policy, etc. Based on this information, an ad template is selected and ad conditions are entered.

[0996] Using a client terminal, a user selects an ad template and enters the ad conditions, including detailed information such as the title, features, and price. Once the server receives this information, it uses a natural language processing engine to generate a draft ad. This draft is an initial version that the user can review and modify.

[0997] The emotional engine analyzes the emotional tone of the generated draft ad and adjusts it as necessary, ensuring that the ad content appeals appropriately to the target user. Once the user reviews the final version of the ad and sends a publishing request to the server, the server connects the final version of the ad to an external advertising platform and publishes it.

[0998] Furthermore, the server has the function of receiving feedback and storing it in a database. The feedback information is analyzed using an emotion engine and reflected in the ad generation algorithm. This process continuously improves the effectiveness of the ads.

[0999] Users can also input specific conditions when sending targeted ads, and the server will generate targeted ads based on these conditions, adjust them using the emotion engine, and send them. The results of these ads are tracked by the server and fed back to the user.

[1000] For example, if a user wants to generate an ad with the title "An exciting new product!" and the features "High-performance battery, easy-to-use UI, and 1-year warranty," they would use the following prompt:

[1001] prompt:

[1002] Use the company information below to generate copy that strikes a positive and friendly tone.

[1003] Company information: A company, large city, e-commerce industry, customer satisfaction focused

[1004] Advertisement details: Title: "An exciting new product!" Features: "High-performance battery, easy-to-use UI, 1-year warranty"

[1005] Desired tone: Positive and friendly

[1006] This allows users to easily generate ads using generative AI models and obtain ad copy that is tailored to the optimal emotional tone using an emotional engine.

[1007] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1008] Step 1:

[1009] The server receives the basic information and stores it in the database. This is the process of receiving information such as the company name, location, industry type, and management policy sent from the client terminal. The received basic information is stored in the database as is in JSON format. This process ensures that the company's basic information is properly stored in the system.

[1010] Input: Basic information such as company name, location, industry type, management policy, etc. (JSON format)

[1011] Data processing: Save basic information to a database

[1012] Output: Basic information stored in the database

[1013] Step 2:

[1014] The user selects an ad template and enters the ad conditions. The user logs in to the system using a client terminal and selects the desired template from the presented templates. Then, the user enters the necessary ad conditions (title, features, price, etc.).

[1015] Input: Template selection information, advertising conditions (title, features, price, etc.)

[1016] Data processing: converting user input into the appropriate format and sending it to the server

[1017] Output: Template selection information and ad conditions sent to the server

[1018] Step 3:

[1019] The server uses a natural language processing engine based on the received information to generate a draft ad. The server runs the natural language processing engine based on the entered ad conditions and the selected template. This automatically generates an initial draft.

[1020] Input: Template selection information, advertising conditions

[1021] Data processing: Draft generation using a natural language processing engine

[1022] Output: Initial Draft

[1023] Step 4:

[1024] The user checks and modifies the draft. The generated draft is checked on the client terminal and modified as necessary. The modified draft is then sent back to the server.

[1025] Input: Initial Draft

[1026] Data processing: User confirmation and correction

[1027] Output: Revised draft

[1028] Step 5:

[1029] The server uses an emotion engine to analyze the emotional tone of the ad content and adjust it accordingly. The emotion engine analyzes the emotional tone of the revised draft and adjusts it to the specified emotional tone (e.g., positive, friendly).

[1030] Input: Revised draft

[1031] Data processing: Emotional tone analysis and adjustment

[1032] Output: Final draft with emotional tone applied

[1033] Step 6:

[1034] The server links the final version of the advertisement to the advertising platform and publishes it. Upon receiving the publishing request, the server links the final draft to the specified advertising platform and publishes the advertisement.

[1035] Input: Final draft with emotional tone applied

[1036] Data processing: Collaboration with advertising platforms

[1037] Output: Published ad

[1038] Step 7:

[1039] The server receives the feedback and stores it in a database. It then analyzes the feedback information using an emotion engine and reflects it in the ad generation algorithm, which will result in more effective ad generation in the future.

[1040] Input: Feedback information

[1041] Data processing: storing and analyzing feedback information

[1042] Output: Improved ad generation algorithm

[1043] Step 8:

[1044] The user inputs the conditions for targeting the advertisement to be sent. The user inputs the conditions for targeting the advertisement to be sent using the client terminal.

[1045] Input: Criteria for sending targeted ads

[1046] Data processing: Send condition information to the server

[1047] Output: Condition information sent to the server

[1048] Step 9:

[1049] The server generates and sends targeted ads based on the conditions. The emotion engine adjusts the content of the ads and automatically sends them to the specified targets.

[1050] Input: Criteria for sending targeted ads

[1051] Data processing: Emotion engine adjustment, ad generation and delivery

[1052] Output: Targeted ads sent

[1053] Step 10:

[1054] The server tracks the results of the advertisements sent, collecting response data (number of views, number of clicks, etc.) and generating feedback based on that data.

[1055] Input: Targeted ads sent

[1056] Data processing: Reaction data collection and analysis

[1057] Output: Tracking data, feedback

[1058] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1059] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1060] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1061] [Third embodiment]

[1062] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1063] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1064] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1065] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1066] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1067] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1068] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1069] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1070] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1071] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1072] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1073] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1074] The embodiment of the present invention is a generative AI service that supports companies' recruitment operations. This system operates in cooperation with three parties: a server, a terminal, and a user.

[1075] 1. Registering and managing basic information

[1076] The server starts up and opens an API endpoint, preparing to establish communication with the device. The user enters basic information through the device, and the server stores that information in a database. This basic information includes the company name, location, industry type, management policy, etc.

[1077] Examples:

[1078] A human resources manager at an IT company logs into the system and enters basic information about the company (e.g., "XYZ Company, Tokyo, software development industry, corporate culture that values ​​innovation"), and the server receives this information and stores it in a database.

[1079] 2. Job posting generation

[1080] Users log in to the system from their terminal and select a job posting template. Templates are available in multiple formats for different job types and departments. They then enter details such as job type, desired skills, years of experience, and salary range according to the template they select.

[1081] The server receives this information and uses a natural language processing engine to generate a draft job posting, which is then provided to the user for review and modification.

[1082] Examples:

[1083] A human resources representative from the same IT company selects the "Java Developer" template and inputs the desired skills (Java, Spring Framework), years of experience (3+ years), salary range (annual salary 6-8 million yen), etc. The server then generates a draft, and the user can review the proposed content and make any necessary revisions.

[1084] 3. Publication of job postings

[1085] The user checks the final version of the job posting and sends a publishing request to the server. The server saves the job posting and publishes it on external job sites or the company's own recruitment page.

[1086] Examples:

[1087] Once the HR person has reviewed the draft and clicked the "Publish" button, the server saves the job posting and it is instantly updated on directly linked external job sites and the company's own recruitment page. This entire process is completed in just a few minutes, enabling rapid recruitment activities.

[1088] 4. Receiving and Analyzing Feedback

[1089] User feedback is received by the server and stored in a database, where it can be analyzed and used to improve the job posting generation algorithm.

[1090] Examples:

[1091] After publishing a job posting, users input reactions from job seekers and internal feedback (e.g., "few applicants," "the description is difficult to understand," etc.) into the system. This information is stored on the server and will be used the next time a job posting is generated.

[1092] 5. Generate and send scout emails (future feature)

[1093] The user inputs the conditions for the person to whom they want to send a scout email. The server generates and sends the scout email based on these conditions. The server tracks the sending results and provides feedback to the user.

[1094] Examples:

[1095] If a human resources officer wants to send a scouting email to a candidate who is a "full-stack engineer" and "able to work remotely," he or she enters those conditions into the system. The server automatically generates and sends scouting emails to candidates who meet the conditions. Later, the server tracks the response data to the scouting email and provides it to the human resources officer.

[1096] In this way, the present invention effectively and efficiently supports the recruitment operations of companies.

[1097] The processing flow will be explained below.

[1098] Step 1:

[1099] The server starts and opens an API endpoint, allowing devices to communicate with the server.

[1100] Step 2:

[1101] The terminal loads the user interface, allowing the user to access the system.

[1102] Step 3:

[1103] A user logs into the system through a terminal and enters a user ID and password.

[1104] Step 4:

[1105] The server checks the user's authentication information against the database, and if authentication is successful, displays a basic information input screen for the user.

[1106] Step 5:

[1107] The user enters basic information about the company (company name, location, industry, management policy, etc.) and clicks the "Submit" button.

[1108] Step 6:

[1109] The server receives the basic information entered and stores it in a database.

[1110] Step 7:

[1111] The user clicks the "Create a new job posting" button on their device and selects an appropriate template from a list of job posting templates.

[1112] Step 8:

[1113] The server receives the user's template selection request and displays the template form on the user interface.

[1114] Step 9:

[1115] The user enters job requirements such as job type, desired skills, experience, and salary range according to the template, and clicks the "Complete input" button.

[1116] Step 10:

[1117] The server receives the entered job requirements and temporarily stores them.

[1118] Step 11:

[1119] The server invokes a natural language processing engine and generates a draft job posting based on the received information.

[1120] Step 12:

[1121] The generated draft job posting is sent to the user's terminal and displayed on the user interface.

[1122] Step 13:

[1123] The user checks the draft content on the terminal and makes corrections as necessary.

[1124] Step 14:

[1125] After making corrections, the user clicks the "Confirm" button to approve the final version of the job posting.

[1126] Step 15:

[1127] The server stores approved job postings in a database and then links to external job sites and the company's own recruitment page via an API, making them publicly available.

[1128] Step 16:

[1129] Users conduct job postings and input responses from job seekers and internal feedback into the system.

[1130] Step 17:

[1131] The server receives the feedback information and stores it in a database.

[1132] Step 18:

[1133] The server analyzes the feedback and uses it to improve the job posting generation algorithm.

[1134] Step 19:

[1135] The user enters the conditions for the person to whom the scout email is to be sent and clicks the "Conditions input complete" button.

[1136] Step 20:

[1137] The server receives the conditions and generates a scout email.

[1138] Step 21:

[1139] The server sends the generated scout email to the candidate and tracks the sending results.

[1140] Step 22:

[1141] The server analyzes the tracking results and provides feedback to the user through a user interface.

[1142] Example 1

[1143] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1144] Traditional recruitment processes have the problem of taking a lot of time and effort, from creating a job posting to publishing it and receiving feedback. Furthermore, if the job posting does not receive a positive response from job seekers due to inappropriate content, it takes a lot of effort to revise the information and republish it. Furthermore, there is no efficient way to generate and send scouting emails, making it difficult to effectively recruit talent.

[1145] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1146] In this invention, the server includes: [means for receiving basic information and saving it in a database;] [means for selecting a job posting template and inputting job requirements;] [means for generating a draft job posting using a natural language processing engine based on the received information;] [means for publishing the final job posting in conjunction with an external job site or the company's own recruitment page;] [means for opening an API endpoint and establishing communication with a terminal;] [means for receiving feedback and saving it in a database;] [means for analyzing feedback information and adjusting the job posting generation algorithm;] [means for inputting conditions for sending scout emails;] [means for generating and sending scout emails based on the conditions; and [means for tracking the sending results.] This makes it possible to effectively and efficiently carry out a series of recruitment tasks, from creating and publishing a job posting, analyzing feedback, and generating and sending scout emails.

[1147] "Basic information" refers to basic data about a company, such as the company name, location, industry type, and management policy.

[1148] "User" refers to a person who uses this system to manage and operate job information, such as a company's recruiter or human resources officer.

[1149] "Server" refers to a computer that forms the core part of the system and performs various processes such as receiving, storing, analyzing, and publishing data.

[1150] "Terminal" refers to a device used by a user, such as a computer, tablet, or smartphone.

[1151] "Database" refers to organized data storage for storing basic information, job postings, feedback information, etc.

[1152] A "template" is a prescribed format used when creating job postings, prepared for each job type and department.

[1153] "Job requirements" refers to the detailed requirements such as the job type, required skills, years of experience, and salary range listed in the job posting.

[1154] A "natural language processing engine" is an AI model that generates and analyzes sentences, and is a technology that is particularly used to draft job postings.

[1155] "External job sites" refer to websites where companies post job openings and which job seekers can access.

[1156] An "API endpoint" is a connection point between a server and a terminal for communication, and refers to the URL or URI used to send and receive data.

[1157] "Feedback" refers to opinions and impressions from users and job seekers regarding the content of and reactions to job postings.

[1158] A "scouting email" refers to an email sent by a company to recruit job seekers who meet certain criteria.

[1159] "Conditions" refers to the requirements such as job type, skills, and working style of the candidate to whom the scouting email is sent.

[1160] "Tracking" refers to the process of tracking and recording the results of sending scouting emails and the responses to them.

[1161] MODE FOR CARRYING OUT THE INVENTION

[1162] This invention is a generative AI service to support companies' recruitment operations. This system operates in cooperation with three parties: a server, a terminal, and a user, and is implemented according to the following procedure.

[1163] Starting the server and opening the API endpoint

[1164] The server is launched on an AWS EC2 instance, running the Apache Web Server, and an API endpoint is set up, ready to establish communication with the device.

[1165] Registering and managing basic information

[1166] Users access the system through a terminal and enter basic information about their company, including the company name, location, industry type, management policy, etc. The terminal then sends the entered information to the server, which stores it in a database.

[1167] Example: A human resources manager at a manufacturing company logs into the system and enters basic information about the company (e.g., "ABC Manufacturing, Aichi Prefecture, manufacturing industry, corporate culture that emphasizes quality"). The server receives this information and stores it in a database.

[1168] Job posting generation

[1169] Users log in to the system from their terminal and select a job posting template prepared by job type or department. They then enter details such as the job type, desired skills, years of experience, salary range, etc. This information is sent to the server, which uses a generative AI model (e.g., OpenAI GPT-3) to generate a draft job posting.

[1170] Example: A human resources professional selects a "Systems Engineer" template and inputs the required skills (e.g., Python, database management), years of experience (e.g., 5+ years), salary range (e.g., annual salary of 5 million to 7 million yen), etc. The server receives this and uses a generative AI model to generate a draft job posting. The user reviews the draft and makes any necessary revisions.

[1171] Example prompt sentence:

[1172] What skills are you looking for in a "Systems Engineer"? Also include the required years of experience, salary range, and desired characteristics of the company.

[1173] Publication of job postings

[1174] The user checks the final version of the job posting and sends a publishing request to the server, which saves the job posting and publishes it via an external job posting site or the company's own recruitment page.

[1175] Example: A human resources representative makes a final check and clicks the "Publish" button. The server saves the final job posting in the database and immediately updates it on external job sites and the company's own recruitment page, enabling rapid recruitment.

[1176] Receiving and analyzing feedback

[1177] After the job posting is published, the server receives feedback from users and stores it in a database. The server can analyze the feedback information and use it to adjust the job posting generation algorithm.

[1178] Example: After publishing a job posting, users enter feedback from job seekers and internal sources (e.g., "few applicants," "the job posting content is unclear") into the system. The server receives this feedback, stores it in the database, and reflects it the next time a job posting is generated.

[1179] Generate and send scout emails (future feature)

[1180] The user inputs the conditions (job type, skills, working style, etc.) of the person to whom they want to send a scout email. The server generates and sends the scout email based on this information. The server tracks the sending results and provides feedback to the user.

[1181] Example: If a human resources manager wants to send a scouting email to a candidate who is a "mobile app developer" and "able to work remotely," he or she enters those conditions into the system. The server uses a generative AI model to automatically generate a scouting email and sends it to the candidate. The server then tracks the response data and provides it to the human resources manager.

[1182] In this way, the present invention can effectively and efficiently support a company's recruitment operations.

[1183] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1184] Specific processing flow of the system

[1185] Step 1: Start the server and open the API endpoint

[1186] The server is launched on an AWS EC2 instance, running Apache Web Server, and an API endpoint is opened to support companies' recruitment processes.

[1187] Input: Server configuration information and startup script

[1188] Output: API endpoint URL (e.g. https: / / example.com / api)

[1189] Specific operation: The server executes the startup script, which starts the web server and API endpoint, allowing communication with the device.

[1190] Step 2: Register basic information

[1191] Users access the system through a terminal and enter basic company information (company name, location, industry type, management policy).

[1192] Input: Basic company information (company name, location, industry type, management policy, etc.)

[1193] Output: Basic information input data

[1194] What happens: A user accesses the login screen using a browser and enters basic information.

[1195] Step 3: Submit and save basic information

[1196] The device sends basic information to the server, which stores the received information in a database.

[1197] Input: Basic information input data

[1198] Output: Status of saving to database

[1199] Specific operation: The device sends the entered basic information to the server, which verifies the data and saves it in the database.

[1200] Step 4: Select a job template

[1201] Users log in to the system from their terminal and select job posting templates prepared for each job type and department.

[1202] Input: Login credentials, Job template list

[1203] Output: Selected template

[1204] What happens: A user logs into the system and selects from multiple templates.

[1205] Step 5: Enter job requirements

[1206] Based on the template you select, you enter details such as job title, desired skills, years of experience, and salary range.

[1207] Input: Selected template, job requirements (job type, skills, years of experience, etc.)

[1208] Output: Job requirements data entered

[1209] Specific behavior: The user fills out a form with detailed job requirements according to the template.

[1210] Step 6: Submit your job requirements

[1211] The terminal sends the job requirements to the server.

[1212] Input: Job requirements data entered

[1213] Output: Status of sending to server

[1214] Specific operation: The terminal sends the entered job requirements to the server.

[1215] Step 7: Generate a draft job posting

[1216] Based on the job requirements received by the server, a draft job posting is generated using a generative AI model.

[1217] Input: Received job requirements data

[1218] Output: Draft job posting

[1219] Specific operation: The server uses a generative AI model (e.g., OpenAI GPT-3) to generate a draft job posting based on the job requirements.

[1220] Step 8: Review and revise your draft

[1221] The user reviews the draft job posting and makes any necessary corrections.

[1222] Input: Draft job posting

[1223] Output: Final revised job posting

[1224] What happens: The user reviews the draft provided by the server and makes any necessary corrections.

[1225] Step 9: Submit a publishing request

[1226] The user reviews the final job posting and sends a publish request to the server.

[1227] Input: Final revised job posting

[1228] Output: Publish request

[1229] Specific behavior: The user clicks the "Publish" button to send a publication request.

[1230] Step 10: Publish your job posting

[1231] The server stores the final job posting in a database and publishes it on external job sites or the company's own recruitment page.

[1232] Input: Publication request, final job posting

[1233] Output: Job posting publication status

[1234] Specific operation: The server saves the final job posting in the database and reflects it on external job sites and the company's own recruitment page via API.

[1235] Step 11: Receiving feedback

[1236] After the job posting is published, feedback from users is received by the server and stored in a database.

[1237] Input: User feedback

[1238] Output: Stored feedback data

[1239] Specific operation: After a job posting is published, the user enters feedback information into the system, and that information is sent to the server.

[1240] Step 12: Analyze feedback

[1241] The server analyzes the feedback information and uses it to adjust the job posting generation algorithm.

[1242] Input: Stored feedback data

[1243] Output: Analysis results and algorithm adjustments

[1244] Specific operation: The server analyzes the received feedback data and adjusts the algorithm to reflect it when generating the next job posting.

[1245] Step 13: Generate and send scout emails (future feature)

[1246] The user inputs the criteria for the person to whom they want to send a scouting email. The server generates a scouting email based on the criteria and sends it to the candidate. The server tracks the sending results and provides feedback to the user.

[1247] Input: Target conditions for scouting emails (job type, skills, working conditions, etc.)

[1248] Output: Scout email sending status and tracking data

[1249] Specific operation: The user inputs the target conditions, and the server uses the generative AI model to generate and send scouting emails. The server tracks the sending results and provides feedback.

[1250] (Application example 1)

[1251] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1252] Corporate recruitment is a complex process that requires a great deal of effort and time. Particularly in factories, quickly and efficiently recruiting the right talent is directly linked to production efficiency. However, delays in creating and publishing job postings and analyzing feedback can lead to the risk of missing out on key talent. Furthermore, on-site talent management is even more difficult, complicating the management work of dedicated personnel. There is a need for a system that can solve these issues and achieve a more efficient and accurate recruitment process.

[1253] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1254] In this invention, the server includes: [means for the server to receive basic information and save it in a database;] [means for the user to select a job posting template and input job requirements;] [means for the server to use a generative AI model based on the information received to generate a draft of the job posting;] [means for the user to check and correct the draft;] [means for the server to link and publish the final version of the job posting to an external recruitment media;] [means for the robot to support employee information management within the factory; and [means for the robot to input job information using a terminal set up and send it to the server. This enables more efficient human resource management within the factory and the generation and publication of accurate job information.

[1255] A "server" is a central system for data management and processing.

[1256] "Basic information" refers to basic information such as the company's name, location, industry, and management policy.

[1257] A "database" is a system for storing and managing data such as basic information and feedback.

[1258] "User" refers to the company personnel and administrators who operate the system.

[1259] A "template" is a predefined format for quickly creating job postings.

[1260] "Job requirements" refer to details such as the desired job type, required skills, years of experience, salary range, etc.

[1261] A "generative AI model" is an artificial intelligence technology that generates a draft job posting based on input information.

[1262] "Draft" refers to the first unfinished job posting generated.

[1263] "External recruitment media" refers to online platforms and sites that disseminate job information to job seekers.

[1264] A "robot" is an automated mechanical device used in factories.

[1265] "Information management" refers to the process of organizing and storing basic employee information and recruitment data.

[1266] A "terminal" is an electronic device that allows a user to input information.

[1267] System Program Overview

[1268] This invention is a "smart robot personnel assistant" system that streamlines personnel management within factories and supports the accurate creation and publication of job information. This system operates in cooperation with a server, terminals, and robots.

[1269] Server Features

[1270] The server is a central system that manages and processes data. It receives basic information (company name, location, industry, management policy, etc.) entered by users through their devices and stores it in a database. It also uses a generative AI model to generate a draft job posting based on the template selected by the user and the entered job requirements (job type, skills, years of experience, salary range, etc.). The generated draft is then reviewed and revised by the user, and the final version is published by the server to external job media.

[1271] Robot Functions

[1272] The robot supports human resource management within the factory. It uses a set-up terminal to input job information and send it to a server, which can then be used directly by on-site personnel within the factory.

[1273] Data processing and calculation

[1274] 1. Register and manage basic information:

[1275] The user enters basic company information from the terminal, and the server receives it and stores it in a database.

[1276] Hardware used: Device (PC, tablet, etc.)

[1277] Software used: Python, Requests library, RESTful API, database management system (MySQL, etc.)

[1278] 2. Generate and publish job posting:

[1279] The user inputs the job posting template and detailed requirements from the terminal, and the server uses a generative AI model (e.g., GPT-3) to generate a draft job posting.

[1280] Hardware used: Server

[1281] Software used: Generative AI models (e.g., GPT-3)

[1282] The generated draft is then checked and revised by the user, and the server publishes the final version to external recruitment media.

[1283] 3. Feedback analysis:

[1284] The server stores the feedback received after publication in a database and analyzes it to help generate the next job posting.

[1285] Software used: Feedback Analysis Engine

[1286] Specific examples

[1287] Basic information registration:

[1288] The factory's human resources staff enters basic information such as "XYZ Manufacturing, Tokyo, manufacturing industry, emphasis on quality and efficiency" into a terminal, and the server stores this information in a database.

[1289] Job listing generation:

[1290] The same person selects the "Line Operator" template and enters the desired skills of "Machine Operation" and "Quality Control," the number of years of experience of "2 years or more," and the salary range of "3 million to 5 million yen per year." Based on this, the server uses a generative AI model to generate the following prompt:

[1291] Please create a job posting for a line operator. The required skills are "machine operation" and "quality control," the required years of experience are "2 years or more," and the salary range is "annual salary of 3 million to 5 million yen."

[1292] The resulting draft is then reviewed by a human resources representative and revised as necessary. The final job posting is then published by the server to external recruitment media.

[1293] This will enable more efficient human resource management within the factory and generate accurate job information.

[1294] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1295] Step 1:

[1296] The user enters basic information through the terminal and sends it to the server. The basic information entered includes the company name, location, industry, management policy, etc. The server receives this information and stores it in the database. The input data is saved in the appropriate table in the database.

[1297] Step 2:

[1298] The user selects a job posting template from their device and enters the job requirements, including the job type, desired skills, years of experience, salary range, etc. This input data is sent to the server, which receives it and prepares it as input prompts for the generative AI model.

[1299] Step 3:

[1300] Based on the job requirements and template information received by the server, a generative AI model (e.g., GPT-3) is used to generate a draft job posting. A prompt is input to the generative AI model, which then generates a draft job posting. The generated draft is saved on the server and provided to the user.

[1301] Step 4:

[1302] The user checks the draft on their device and makes any necessary corrections. The information corrected by the user is sent to the server, which updates the draft again and generates the final version. This final version of the job posting is also saved in the database.

[1303] Step 5:

[1304] The server connects the final job posting to an external recruitment site and publishes it. Specifically, the job information is sent to the external site using methods such as API or FTP. Information on successful publication of the job posting is also saved in the database as tracking information.

[1305] Step 6:

[1306] The server receives external feedback and stores it in a database. The feedback is processed by the analytics engine and used to adjust the job posting generation algorithm for the next job posting, thereby improving the job posting process for future jobs.

[1307] Step 7:

[1308] When a user wants to send a scout email based on specific criteria, they input the criteria from their device and send it to the server. The server searches for information that matches the criteria, then uses a generative AI model to generate and send the scout email. The server tracks the results of the email and provides feedback to the user.

[1309] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1310] An embodiment of the present invention is a system that combines a generative AI service that supports corporate recruitment operations with an emotion engine. This system operates in cooperation with a server, a terminal, a user, and the emotion engine.

[1311] 1. Registering and managing basic information

[1312] The server starts up, opens an API endpoint, and prepares to communicate with the device. The user enters basic information through the device, and the server stores that information in a database. This basic information includes the company name, location, industry type, management policy, etc.

[1313] Examples:

[1314] A human resources manager at an IT company logs into the system and enters basic information about the company (for example, "ABC Company, Tokyo, software development industry, corporate culture that values ​​innovation"). The server receives this information and stores it in a database.

[1315] 2. Job posting generation

[1316] Users log in to the system from their terminal and select a job posting template. Templates are available in multiple formats for different job types and departments. They then enter details such as job type, desired skills, years of experience, and salary range according to the template they select.

[1317] The server receives this information and uses a natural language processing engine to generate a draft job posting, which is then provided to the user for review and modification.

[1318] Examples:

[1319] A human resources representative from the same IT company selects the "Java Developer" template and inputs the desired skills (Java, Spring Framework), years of experience (3+ years), salary range (annual salary 6-8 million yen), etc. The server then generates a draft, and the user can review the proposed content and make any necessary revisions.

[1320] 3. Publication of job postings

[1321] The user checks the final version of the job posting and sends a publishing request to the server. The server saves the job posting and publishes it on external job sites or the company's own recruitment page.

[1322] Examples:

[1323] Once the HR person has reviewed the draft and clicked the "Publish" button, the server saves the job posting and it is instantly updated on linked external job sites and the company's own recruitment page. This entire process is completed in just a few minutes, enabling quick recruitment activities.

[1324] 4. Receiving and Analyzing Feedback

[1325] User feedback is received by the server and stored in a database. The server analyzes the feedback and can use it to improve the job posting generation algorithm. An emotion engine is used to analyze the emotional tone of the feedback and identify areas for further improvement.

[1326] Examples:

[1327] After publishing a job posting, users input reactions from job seekers and internal feedback (for example, "few applicants" or "the description is difficult to understand") into the system. This information is stored on the server and analyzed using an emotion engine. The results of the analysis are reflected the next time a job posting is generated.

[1328] 5. Generate and send scout emails (future feature)

[1329] The user inputs the criteria for the person to whom they want to send a scout email. The server generates and sends the scout email based on this information. The emotion engine adjusts the content of the scout email appropriately. The server tracks the sending results and provides feedback to the user.

[1330] Examples:

[1331] If a human resources manager wants to send a scouting email to a candidate who is a "full-stack engineer" and "able to work remotely," he or she can enter those criteria into the system. The server automatically generates a scouting email to candidates who fit the criteria, and the emotion engine adjusts the content before sending it. Later, the server tracks the response data to the scouting email and provides it to the human resources manager.

[1332] 6. Adjusting the emotional tone when generating job postings

[1333] When users create a job posting, the emotion engine analyzes their input in real time and suggests appropriate emotional tones (e.g., proactive, friendly, professional, etc.), making job postings more engaging and effective.

[1334] Examples:

[1335] As users enter their skills, years of experience, and other information into a job posting template, the emotion engine analyzes the information and displays suggestions such as "use a more positive tone." This makes the job posting more likely to catch the attention of job seekers.

[1336] In this way, the present invention effectively and efficiently supports companies' recruitment operations, and realizes the generation of sophisticated job postings and scouting emails that take into account the user's emotions.

[1337] The processing flow will be explained below.

[1338] Step 1:

[1339] The server starts and opens an API endpoint, allowing devices to communicate with the server.

[1340] Step 2:

[1341] The terminal loads the user interface, allowing the user to access the system.

[1342] Step 3:

[1343] A user logs into the system through a terminal and enters a user ID and password.

[1344] Step 4:

[1345] The server checks the user's authentication information against the database, and if authentication is successful, displays a basic information input screen for the user.

[1346] Step 5:

[1347] The user enters basic information about the company (company name, location, industry, management policy, etc.) and clicks the "Submit" button.

[1348] Step 6:

[1349] The server receives the basic information entered and stores it in a database.

[1350] Step 7:

[1351] The user clicks the "Create a new job posting" button on their device and selects an appropriate template from a list of job posting templates.

[1352] Step 8:

[1353] The server receives the user's template selection request and displays the template form on the user interface.

[1354] Step 9:

[1355] The user enters job requirements such as job type, desired skills, experience, and salary range according to the template, and clicks the "Complete input" button.

[1356] Step 10:

[1357] The server receives the entered job requirements and temporarily stores them.

[1358] Step 11:

[1359] The server invokes a natural language processing engine and generates a draft job posting based on the received information.

[1360] Step 12:

[1361] The generated draft job posting is sent to the user's terminal and displayed on the user interface.

[1362] Step 13:

[1363] The user checks the draft content on the terminal and makes corrections as necessary.

[1364] Step 14:

[1365] The emotion engine analyzes the user's input and suggests an appropriate emotional tone (e.g., proactive, friendly, professional, etc.).

[1366] Step 15:

[1367] The user considers the emotional tone suggestions and modifies the content of the job posting.

[1368] Step 16:

[1369] After making corrections, the user clicks the "Confirm" button to approve the final version of the job posting.

[1370] Step 17:

[1371] The server stores approved job postings in a database and then links to external job sites and the company's own recruitment page via an API, making them publicly available.

[1372] Step 18:

[1373] Users conduct job postings and input responses from job seekers and internal feedback into the system.

[1374] Step 19:

[1375] The server receives the feedback information and stores it in a database.

[1376] Step 20:

[1377] The server analyzes the feedback and uses it to improve the job posting generation algorithm.

[1378] Step 21:

[1379] An emotional engine analyzes the emotional tone of the feedback and identifies areas for improvement.

[1380] Step 22:

[1381] The user enters the conditions for the person to whom the scout email is to be sent and clicks the "Conditions input complete" button.

[1382] Step 23:

[1383] The server receives the conditions and generates a scout email.

[1384] Step 24:

[1385] The emotional engine analyzes the content of the scouting email and adjusts the appropriate emotional tone.

[1386] Step 25:

[1387] The server sends the generated scout email to the candidate.

[1388] Step 26:

[1389] The server tracks the results of the transmission and provides feedback to the user through the user interface.

[1390] Example 2

[1391] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1392] In corporate recruitment processes, there are issues with the inefficiency of creating and publishing job postings, receiving and analyzing feedback, and creating and sending scouting emails. In particular, job postings and scouting emails are created only in a standardized format, and there is a lack of a system for collecting and analyzing applicant responses to improve the content. Furthermore, there is a lack of emotional tone adjustment to create job postings and scouting emails that are appealing to job seekers.

[1393] The specification process by the specification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for receiving basic information and saving it in a database, a means for a user to select a job posting template and input job requirements, and a means for using a natural language processing engine to generate a draft of a job posting based on the received information. This enables companies to carry out their recruitment operations efficiently.

[1394] It also includes a means for publishing the final job posting on external job sites or the company's own recruitment page, and a means for the emotion engine to analyze user input in real time and suggest an appropriate emotional tone, making the job posting more attractive and effective for job seekers.

[1395] Furthermore, the system includes a means for the server to receive feedback and store it in a database, a means for analyzing the feedback information and adjusting the job posting generation algorithm, and a means for the emotion engine to analyze the emotional tone of the feedback and reflect it in the next job posting generation, thereby enabling the content of the job posting to be dynamically improved and optimized.

[1396] It also includes a means for users to input conditions for sending scout emails, a means for generating scout emails based on the conditions and sending them after adjustments by the emotion engine, and a means for tracking the sending results and providing feedback to the user, thereby enabling the generation of effective scout emails that match the target, and the confirmation and improvement of their effectiveness.

[1397] A "server" is a computer system that can be accessed remotely over a network and that stores, processes, and provides services to data.

[1398] A "terminal" is a device that is directly operated by a user and provides a user interface, such as a PC or smartphone.

[1399] A "user" is someone who operates the system and inputs data, such as a company's recruiter.

[1400] "Basic information" refers to basic data about a company, such as the company name, location, industry type, and management policy.

[1401] A "template" is a format used when creating a job posting, and is a pre-designed framework for each job type or department.

[1402] "Employment conditions" are employment conditions listed in a job posting, such as job type, required skills, years of experience, and salary range.

[1403] A "natural language processing engine" is an algorithm for analyzing and generating text data, and is a system that creates text using a generative AI model.

[1404] "Draft" is the first draft of the job posting that is generated, and is text that contains content that the user needs to correct and confirm.

[1405] An "emotion engine" is an algorithm that analyzes the emotional tone of text data and suggests appropriate expressions.

[1406] "Feedback" refers to opinions and reactions provided by users and job seekers, as well as data based on them.

[1407] An "algorithm" is a defined set of procedures or computational steps for solving a problem.

[1408] A "scout email" is a recruiting message sent to candidates who meet certain hiring criteria.

[1409] "Tracking" is the process of tracking and recording the results and responses to emails sent.

[1410] An embodiment of the present invention is a system for supporting a company's recruitment operations, in which a server, a terminal, a user, and an emotion engine work together. This system mainly provides a series of functions for receiving, storing, analyzing, generating, and publishing data.

[1411] First, when the server starts up, it opens an API endpoint and prepares for communication with the terminal. The hardware used here is a typical server computer, and the software includes a database management system (DBMS) and API management tools. The user enters basic company information (e.g., company name, location, industry type, management policy, etc.) through the terminal, and the server stores that information in a database.

[1412] Next, the user logs into the system and selects a job posting template. Templates are available in multiple formats and are categorized by job type and department. The user enters details such as the job type, desired skills, years of experience, and salary range according to the template they select. The server receives this information and generates a draft job posting using a natural language processing engine (e.g., GPT-3). This process utilizes a generative AI model. The generated draft is then provided to the user, who can review and modify it.

[1413] The user reviews the final version of the job posting and sends a publication request to the server. The server saves the job posting in a database, links it to external job sites and the company's own recruitment page, and publishes the information. The software used at this stage is an external API integration tool and web server software.

[1414] Receiving and analyzing feedback is also an important function. User feedback is received by the server and stored in a database. The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional tone of the feedback and use it to improve the job posting generation algorithm. The analyzed data is reflected in the next job posting generation.

[1415] In the future, the system will also provide the ability to generate and send scout emails. Users can enter the conditions for sending a scout email, and the server will generate the email based on the conditions. The emotion engine will then adjust the content appropriately. The server will then send the email, track the results, and provide feedback to the user.

[1416] Additionally, when users create a job posting, the emotion engine analyzes their input in real time and suggests appropriate emotional tones (e.g., proactive, friendly, professional, etc.), making job postings more engaging and effective.

[1417] Examples:

[1418] A human resources representative from a company logs into the system and enters basic company information, which the server then saves in a database. Next, they select the "Java Developer" template and enter their details. The server then uses a natural language processing engine to generate a draft job posting. The representative then reviews the final job posting and clicks the publish button, which instantly updates both the company's recruitment page and external job sites. Feedback is also received, analyzed by an emotion engine, and reflected in the next job posting, allowing the company to provide better job content.

[1419] Example prompts using a generative AI model:

[1420] "I'm looking to create a job posting for a Java developer. The required skills are Java and Spring Framework, with at least three years of experience and an annual salary range of ¥6,000,000 to ¥8,000,000."

[1421] With the above configuration, the present invention provides an efficient recruitment support system and improves the recruitment process of a company.

[1422] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1423] Step 1:

[1424] The server starts, opens an API endpoint, and prepares to communicate with the device.

[1425] Input: System startup command

[1426] Output: API endpoint status

[1427] Specific operation: The server accepts the start command and opens an API endpoint (URL) for network communication. The API endpoint waits for requests from the device.

[1428] Step 2:

[1429] The user inputs basic information through the terminal, and the server receives the information.

[1430] Input: Basic company information (company name, location, industry type, management policy, etc.)

[1431] Output: Acknowledgement message

[1432] Specific operation: A user fills in the basic information of their company in the input form on their device and clicks the submit button. The server receives this request, analyzes the data, and then generates and executes an SQL query to save it in the database.

[1433] Step 3:

[1434] The server stores the basic information in a database.

[1435] Input: Basic information data

[1436] Output: Message that saving to database is complete

[1437] Specific operation: The server saves the received basic information in the database management system. If the save is successful, it returns a message to the user indicating that the save was successful.

[1438] Step 4:

[1439] A user logs into the system from a terminal and selects a job posting template.

[1440] Input: Login information (username, password), template to select

[1441] Output: Template selection completion message

[1442] Specific behavior: A user enters their credentials on the login screen to access the system. A list of job templates is displayed, and the user selects the appropriate template.

[1443] Step 5:

[1444] Users enter details such as job title, desired skills, years of experience, and salary range.

[1445] Input: Job type, desired skills, years of experience, salary range

[1446] Output: Input completion message

[1447] Specific operation: The user fills in various details in the input form according to the template selected by the user. After completion, the input contents are sent to the server.

[1448] Step 6:

[1449] The server passes the received information to a natural language processing engine to generate a draft job posting.

[1450] Input: Job details

[1451] Output: Draft job posting

[1452] How it works: The server receives the job details and passes them to a generative AI model (e.g., GPT-3) in the form of a prompt. The natural language processing engine analyzes and generates the prompt to create a draft job posting.

[1453] Example prompt sentence:

[1454] "I'm looking to create a job posting for a Java developer. The required skills are Java and Spring Framework, with at least three years of experience and an annual salary range of ¥6,000,000 to ¥8,000,000."

[1455] Step 7:

[1456] The server returns the generated draft to the user for confirmation.

[1457] Input: Draft job posting

[1458] Output: Draft displayed to user

[1459] Specific operation: The server sends the generated draft to the terminal and displays it to the user, allowing the user to check the contents of the draft and make any necessary corrections.

[1460] Step 8:

[1461] The user reviews the final job posting and sends a publish request to the server.

[1462] Input: Last confirmed job posting

[1463] Output: Publishing request sent successfully message

[1464] Specific behavior: The user reviews the final job posting and clicks the "Publish" button to send a request to the server.

[1465] Step 9:

[1466] The server stores the job postings in a database and links them to external job sites and the company's own recruitment page for publication.

[1467] Input: Publishing request

[1468] Output: Publishing complete message

[1469] Specific operation: The server saves the job posting data in a database, and at the same time, links the job information to external job sites and the company's own recruitment page via API and publishes it immediately.

[1470] Step 10:

[1471] The user enters the feedback and the server receives it.

[1472] Input: Feedback information

[1473] Output: Feedback acknowledgement message

[1474] Specific operation: The user inputs responses and internal feedback from job seekers into the terminal and sends them to the server, which receives them and stores them in a database.

[1475] Step 11:

[1476] The server analyzes the feedback and uses an emotion engine to reflect it in the next job posting generation.

[1477] Input: Feedback information

[1478] Output: Analysis results and algorithm adjustment data

[1479] What it does: The server analyzes the feedback information, evaluates the emotional tone using an emotion engine, and adjusts the job posting generation algorithm based on the analysis results to reflect them in the next generation.

[1480] Step 12:

[1481] The user inputs the conditions for sending the scout mail, and the server generates and sends the scout mail.

[1482] Input: Scout mail conditions

[1483] Output: Generated scout email

[1484] Specific operation: The user inputs the target conditions for the scout email and sends it to the server. The server generates the scout email based on the conditions, adjusts the content using the emotion engine, and sends it.

[1485] Step 13:

[1486] The server tracks the results of the submission and provides feedback to the user.

[1487] Input: Scout email sending conditions

[1488] Output: Sending results and tracking data

[1489] Specific operation: The server tracks the results of sending scout emails, collects reactions and response data, analyzes it, and provides it to the user as feedback.

[1490] (Application example 2)

[1491] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1492] Conventional ad generation systems rely on manual creation of ad copy, which makes it difficult to properly appeal to target users, especially due to a lack of emotional tone adjustment. Furthermore, the process of receiving feedback and incorporating it into the ad generation algorithm is manual and inefficient. Furthermore, individual ads are not properly sent to target users, making it difficult to maximize advertising effectiveness.

[1493] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1494] In this invention, the server includes: [means for the server to receive basic information and store it in a database;] [means for the user to select an advertisement template and input advertisement conditions;] [means for the server to generate a draft advertisement using a natural language processing engine based on the information received;] [means for the user to check and revise the draft;] [means for the server to link and publish the final advertisement to an external advertising platform; and [means for analyzing the emotional tone of the advertisement content using an emotion engine and adjusting it as necessary.] This automates the entire process from advertisement generation to publication and feedback analysis, making it possible to appropriately adjust the emotional tone of the advertisement content. In addition, advertisements can be individually generated and sent based on the recipients of targeted advertisements, maximizing the effectiveness of advertising.

[1495] A "server" is a computer system that has the function of receiving data from client terminals and storing it in a database.

[1496] "Basic information" refers to information such as company name, location, industry type, and management policy that is necessary for creating advertisements and job offers.

[1497] A "database" is a system for systematically storing and managing received basic information and feedback information.

[1498] A "template" is an advertisement or job posting format based on a specific format that can be selected and used by a user.

[1499] "Advertising conditions" are information input as the content of the advertisement to be generated, and include, for example, the title, features, price, and the like.

[1500] A "natural language processing engine" is software that automatically generates drafts of advertisements and job postings based on received information.

[1501] A "draft" is an early version of an advertisement or job posting that can be reviewed and modified by the user.

[1502] "Advertising Platform" means an external service for publishing generated advertisements, including the Internet and mobile apps.

[1503] An "emotion engine" is software that analyzes the emotional tone of advertisements and job postings and adjusts it as needed.

[1504] "Feedback" refers to opinions and evaluations from users and viewers regarding advertisements and job postings.

[1505] "Targeted advertising" means advertising that is generated and delivered to an individual user based on specific criteria.

[1506] "Tracking" refers to tracking the results of sending advertisements and the status of receiving feedback.

[1507] This embodiment relates to an advertisement generation system, and in particular provides a system that combines a generation AI and an emotion engine. To implement the invention, the following main components are included:

[1508] 1. Server

[1509] 2. Client Terminal

[1510] 3. Emotion Engine

[1511] 4. Natural Language Processing Engine

[1512] 5. Database

[1513] 6. Advertising Platform

[1514] The server receives basic information sent from the client terminal and stores it in a database. Basic information includes the company name, location, industry type, management policy, etc. Based on this information, an ad template is selected and ad conditions are entered.

[1515] Using a client terminal, a user selects an ad template and enters the ad conditions, including detailed information such as the title, features, and price. Once the server receives this information, it uses a natural language processing engine to generate a draft ad. This draft is an initial version that the user can review and modify.

[1516] The emotional engine analyzes the emotional tone of the generated draft ad and adjusts it as necessary, ensuring that the ad content appeals appropriately to the target user. Once the user reviews the final version of the ad and sends a publishing request to the server, the server connects the final version of the ad to an external advertising platform and publishes it.

[1517] Furthermore, the server has the function of receiving feedback and storing it in a database. The feedback information is analyzed using an emotion engine and reflected in the ad generation algorithm. This process continuously improves the effectiveness of the ads.

[1518] Users can also input specific conditions when sending targeted ads, and the server will generate targeted ads based on these conditions, adjust them using the emotion engine, and send them. The results of these ads are tracked by the server and fed back to the user.

[1519] For example, if a user wants to generate an ad with the title "An exciting new product!" and the features "High-performance battery, easy-to-use UI, and 1-year warranty," they would use the following prompt:

[1520] prompt:

[1521] Use the company information below to generate copy that strikes a positive and friendly tone.

[1522] Company information: A company, large city, e-commerce industry, customer satisfaction focused

[1523] Advertisement details: Title: "An exciting new product!" Features: "High-performance battery, easy-to-use UI, 1-year warranty"

[1524] Desired tone: Positive and friendly

[1525] This allows users to easily generate ads using generative AI models and obtain ad copy that is tailored to the optimal emotional tone using an emotional engine.

[1526] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1527] Step 1:

[1528] The server receives the basic information and stores it in the database. This is the process of receiving information such as the company name, location, industry type, and management policy sent from the client terminal. The received basic information is stored in the database as is in JSON format. This process ensures that the company's basic information is properly stored in the system.

[1529] Input: Basic information such as company name, location, industry type, management policy, etc. (JSON format)

[1530] Data processing: Save basic information to a database

[1531] Output: Basic information stored in the database

[1532] Step 2:

[1533] The user selects an ad template and enters the ad conditions. The user logs in to the system using a client terminal and selects the desired template from the presented templates. Then, the user enters the necessary ad conditions (title, features, price, etc.).

[1534] Input: Template selection information, advertising conditions (title, features, price, etc.)

[1535] Data processing: converting user input into the appropriate format and sending it to the server

[1536] Output: Template selection information and ad conditions sent to the server

[1537] Step 3:

[1538] The server uses a natural language processing engine based on the received information to generate a draft ad. The server runs the natural language processing engine based on the entered ad conditions and the selected template. This automatically generates an initial draft.

[1539] Input: Template selection information, advertising conditions

[1540] Data processing: Draft generation using a natural language processing engine

[1541] Output: Initial Draft

[1542] Step 4:

[1543] The user checks and modifies the draft. The generated draft is checked on the client terminal and modified as necessary. The modified draft is then sent back to the server.

[1544] Input: Initial Draft

[1545] Data processing: User confirmation and correction

[1546] Output: Revised draft

[1547] Step 5:

[1548] The server uses an emotion engine to analyze the emotional tone of the ad content and adjust it accordingly. The emotion engine analyzes the emotional tone of the revised draft and adjusts it to the specified emotional tone (e.g., positive, friendly).

[1549] Input: Revised draft

[1550] Data processing: Emotional tone analysis and adjustment

[1551] Output: Final draft with emotional tone applied

[1552] Step 6:

[1553] The server links the final version of the advertisement to the advertising platform and publishes it. Upon receiving the publishing request, the server links the final draft to the specified advertising platform and publishes the advertisement.

[1554] Input: Final draft with emotional tone applied

[1555] Data processing: Collaboration with advertising platforms

[1556] Output: Published ad

[1557] Step 7:

[1558] The server receives the feedback and stores it in a database. It then analyzes the feedback information using an emotion engine and reflects it in the ad generation algorithm, which will result in more effective ad generation in the future.

[1559] Input: Feedback information

[1560] Data processing: storing and analyzing feedback information

[1561] Output: Improved ad generation algorithm

[1562] Step 8:

[1563] The user inputs the conditions for targeting the advertisement to be sent. The user inputs the conditions for targeting the advertisement to be sent using the client terminal.

[1564] Input: Criteria for sending targeted ads

[1565] Data processing: Send condition information to the server

[1566] Output: Condition information sent to the server

[1567] Step 9:

[1568] The server generates and sends targeted ads based on the conditions. The emotion engine adjusts the content of the ads and automatically sends them to the specified targets.

[1569] Input: Criteria for sending targeted ads

[1570] Data processing: Emotion engine adjustment, ad generation and delivery

[1571] Output: Targeted ads sent

[1572] Step 10:

[1573] The server tracks the results of the advertisements sent, collecting response data (number of views, number of clicks, etc.) and generating feedback based on that data.

[1574] Input: Targeted ads sent

[1575] Data processing: Reaction data collection and analysis

[1576] Output: Tracking data, feedback

[1577] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1578] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1579] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1580] [Fourth embodiment]

[1581] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1582] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1583] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1584] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1585] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1586] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1587] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1588] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1589] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1590] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1591] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1592] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1593] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1594] The embodiment of the present invention is a generative AI service that supports companies' recruitment operations. This system operates in cooperation with three parties: a server, a terminal, and a user.

[1595] 1. Registering and managing basic information

[1596] The server starts up and opens an API endpoint, preparing to establish communication with the device. The user enters basic information through the device, and the server stores that information in a database. This basic information includes the company name, location, industry type, management policy, etc.

[1597] Examples:

[1598] A human resources manager at an IT company logs into the system and enters basic information about the company (e.g., "XYZ Company, Tokyo, software development industry, corporate culture that values ​​innovation"), and the server receives this information and stores it in a database.

[1599] 2. Job posting generation

[1600] Users log in to the system from their terminal and select a job posting template. Templates are available in multiple formats for different job types and departments. They then enter details such as job type, desired skills, years of experience, and salary range according to the template they select.

[1601] The server receives this information and uses a natural language processing engine to generate a draft job posting, which is then provided to the user for review and modification.

[1602] Examples:

[1603] A human resources representative from the same IT company selects the "Java Developer" template and inputs the desired skills (Java, Spring Framework), years of experience (3+ years), salary range (annual salary 6-8 million yen), etc. The server then generates a draft, and the user can review the proposed content and make any necessary revisions.

[1604] 3. Publication of job postings

[1605] The user checks the final version of the job posting and sends a publishing request to the server. The server saves the job posting and publishes it on external job sites or the company's own recruitment page.

[1606] Examples:

[1607] Once the HR person has reviewed the draft and clicked the "Publish" button, the server saves the job posting and it is instantly updated on directly linked external job sites and the company's own recruitment page. This entire process is completed in just a few minutes, enabling rapid recruitment activities.

[1608] 4. Receiving and Analyzing Feedback

[1609] User feedback is received by the server and stored in a database, where it can be analyzed and used to improve the job posting generation algorithm.

[1610] Examples:

[1611] After publishing a job posting, users input reactions from job seekers and internal feedback (e.g., "few applicants," "the description is difficult to understand," etc.) into the system. This information is stored on the server and will be used the next time a job posting is generated.

[1612] 5. Generate and send scout emails (future feature)

[1613] The user inputs the conditions for the person to whom they want to send a scout email. The server generates and sends the scout email based on these conditions. The server tracks the sending results and provides feedback to the user.

[1614] Examples:

[1615] If a human resources officer wants to send a scouting email to a candidate who is a "full-stack engineer" and "able to work remotely," he or she enters those conditions into the system. The server automatically generates and sends scouting emails to candidates who meet the conditions. Later, the server tracks the response data to the scouting email and provides it to the human resources officer.

[1616] In this way, the present invention effectively and efficiently supports the recruitment operations of companies.

[1617] The processing flow will be explained below.

[1618] Step 1:

[1619] The server starts and opens an API endpoint, allowing devices to communicate with the server.

[1620] Step 2:

[1621] The terminal loads the user interface, allowing the user to access the system.

[1622] Step 3:

[1623] A user logs into the system through a terminal and enters a user ID and password.

[1624] Step 4:

[1625] The server checks the user's authentication information against the database, and if authentication is successful, displays a basic information input screen for the user.

[1626] Step 5:

[1627] The user enters basic information about the company (company name, location, industry, management policy, etc.) and clicks the "Submit" button.

[1628] Step 6:

[1629] The server receives the basic information entered and stores it in a database.

[1630] Step 7:

[1631] The user clicks the "Create a new job posting" button on their device and selects an appropriate template from a list of job posting templates.

[1632] Step 8:

[1633] The server receives the user's template selection request and displays the template form on the user interface.

[1634] Step 9:

[1635] The user enters job requirements such as job type, desired skills, experience, and salary range according to the template, and clicks the "Complete input" button.

[1636] Step 10:

[1637] The server receives the entered job requirements and temporarily stores them.

[1638] Step 11:

[1639] The server invokes a natural language processing engine and generates a draft job posting based on the received information.

[1640] Step 12:

[1641] The generated draft job posting is sent to the user's terminal and displayed on the user interface.

[1642] Step 13:

[1643] The user checks the draft content on the terminal and makes corrections as necessary.

[1644] Step 14:

[1645] After making corrections, the user clicks the "Confirm" button to approve the final version of the job posting.

[1646] Step 15:

[1647] The server stores approved job postings in a database and then links to external job sites and the company's own recruitment page via an API, making them publicly available.

[1648] Step 16:

[1649] Users conduct job postings and input responses from job seekers and internal feedback into the system.

[1650] Step 17:

[1651] The server receives the feedback information and stores it in a database.

[1652] Step 18:

[1653] The server analyzes the feedback and uses it to improve the job posting generation algorithm.

[1654] Step 19:

[1655] The user enters the conditions for the person to whom the scout email is to be sent and clicks the "Conditions input complete" button.

[1656] Step 20:

[1657] The server receives the conditions and generates a scout email.

[1658] Step 21:

[1659] The server sends the generated scout email to the candidate and tracks the sending results.

[1660] Step 22:

[1661] The server analyzes the tracking results and provides feedback to the user through a user interface.

[1662] Example 1

[1663] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1664] Traditional recruitment processes have the problem of taking a lot of time and effort, from creating a job posting to publishing it and receiving feedback. Furthermore, if the job posting does not receive a positive response from job seekers due to inappropriate content, it takes a lot of effort to revise the information and republish it. Furthermore, there is no efficient way to generate and send scouting emails, making it difficult to effectively recruit talent.

[1665] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1666] In this invention, the server includes: [means for receiving basic information and saving it in a database;] [means for selecting a job posting template and inputting job requirements;] [means for generating a draft job posting using a natural language processing engine based on the received information;] [means for publishing the final job posting in conjunction with an external job site or the company's own recruitment page;] [means for opening an API endpoint and establishing communication with a terminal;] [means for receiving feedback and saving it in a database;] [means for analyzing feedback information and adjusting the job posting generation algorithm;] [means for inputting conditions for sending scout emails;] [means for generating and sending scout emails based on the conditions; and [means for tracking the sending results.] This makes it possible to effectively and efficiently carry out a series of recruitment tasks, from creating and publishing a job posting, analyzing feedback, and generating and sending scout emails.

[1667] "Basic information" refers to basic data about a company, such as the company name, location, industry type, and management policy.

[1668] "User" refers to a person who uses this system to manage and operate job information, such as a company's recruiter or human resources officer.

[1669] "Server" refers to a computer that forms the core part of the system and performs various processes such as receiving, storing, analyzing, and publishing data.

[1670] "Terminal" refers to a device used by a user, such as a computer, tablet, or smartphone.

[1671] "Database" refers to organized data storage for storing basic information, job postings, feedback information, etc.

[1672] A "template" is a prescribed format used when creating job postings, prepared for each job type and department.

[1673] "Job requirements" refers to the detailed requirements such as the job type, required skills, years of experience, and salary range listed in the job posting.

[1674] A "natural language processing engine" is an AI model that generates and analyzes sentences, and is a technology that is particularly used to draft job postings.

[1675] "External job sites" refer to websites where companies post job openings and which job seekers can access.

[1676] An "API endpoint" is a connection point between a server and a terminal for communication, and refers to the URL or URI used to send and receive data.

[1677] "Feedback" refers to opinions and impressions from users and job seekers regarding the content of and reactions to job postings.

[1678] A "scouting email" refers to an email sent by a company to recruit job seekers who meet certain criteria.

[1679] "Conditions" refers to the requirements such as job type, skills, and working style of the candidate to whom the scouting email is sent.

[1680] "Tracking" refers to the process of tracking and recording the results of sending scouting emails and the responses to them.

[1681] MODE FOR CARRYING OUT THE INVENTION

[1682] This invention is a generative AI service to support companies' recruitment operations. This system operates in cooperation with three parties: a server, a terminal, and a user, and is implemented according to the following procedure.

[1683] Starting the server and opening the API endpoint

[1684] The server is launched on an AWS EC2 instance, running the Apache Web Server, and an API endpoint is set up, ready to establish communication with the device.

[1685] Registering and managing basic information

[1686] Users access the system through a terminal and enter basic information about their company, including the company name, location, industry type, management policy, etc. The terminal then sends the entered information to the server, which stores it in a database.

[1687] Example: A human resources manager at a manufacturing company logs into the system and enters basic information about the company (e.g., "ABC Manufacturing, Aichi Prefecture, manufacturing industry, corporate culture that emphasizes quality"). The server receives this information and stores it in a database.

[1688] Job posting generation

[1689] Users log in to the system from their terminal and select a job posting template prepared by job type or department. They then enter details such as the job type, desired skills, years of experience, salary range, etc. This information is sent to the server, which uses a generative AI model (e.g., OpenAI GPT-3) to generate a draft job posting.

[1690] Example: A human resources professional selects a "Systems Engineer" template and inputs the required skills (e.g., Python, database management), years of experience (e.g., 5+ years), salary range (e.g., annual salary of 5 million to 7 million yen), etc. The server receives this and uses a generative AI model to generate a draft job posting. The user reviews the draft and makes any necessary revisions.

[1691] Example prompt sentence:

[1692] What skills are you looking for in a "Systems Engineer"? Also include the required years of experience, salary range, and desired characteristics of the company.

[1693] Publication of job postings

[1694] The user checks the final version of the job posting and sends a publishing request to the server, which saves the job posting and publishes it via an external job posting site or the company's own recruitment page.

[1695] Example: A human resources representative makes a final check and clicks the "Publish" button. The server saves the final job posting in the database and immediately updates it on external job sites and the company's own recruitment page, enabling rapid recruitment.

[1696] Receiving and analyzing feedback

[1697] After the job posting is published, the server receives feedback from users and stores it in a database. The server can analyze the feedback information and use it to adjust the job posting generation algorithm.

[1698] Example: After publishing a job posting, users enter feedback from job seekers and internal sources (e.g., "few applicants," "the job posting content is unclear") into the system. The server receives this feedback, stores it in the database, and reflects it the next time a job posting is generated.

[1699] Generate and send scout emails (future feature)

[1700] The user inputs the conditions (job type, skills, working style, etc.) of the person to whom they want to send a scout email. The server generates and sends the scout email based on this information. The server tracks the sending results and provides feedback to the user.

[1701] Example: If a human resources manager wants to send a scouting email to a candidate who is a "mobile app developer" and "able to work remotely," he or she enters those conditions into the system. The server uses a generative AI model to automatically generate a scouting email and sends it to the candidate. The server then tracks the response data and provides it to the human resources manager.

[1702] In this way, the present invention can effectively and efficiently support a company's recruitment operations.

[1703] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1704] Specific processing flow of the system

[1705] Step 1: Start the server and open the API endpoint

[1706] The server is launched on an AWS EC2 instance, running Apache Web Server, and an API endpoint is opened to support companies' recruitment processes.

[1707] Input: Server configuration information and startup script

[1708] Output: API endpoint URL (e.g. https: / / example.com / api)

[1709] Specific operation: The server executes the startup script, which starts the web server and API endpoint, allowing communication with the device.

[1710] Step 2: Register basic information

[1711] Users access the system through a terminal and enter basic company information (company name, location, industry type, management policy).

[1712] Input: Basic company information (company name, location, industry type, management policy, etc.)

[1713] Output: Basic information input data

[1714] What happens: A user accesses the login screen using a browser and enters basic information.

[1715] Step 3: Submit and save basic information

[1716] The device sends basic information to the server, which stores the received information in a database.

[1717] Input: Basic information input data

[1718] Output: Status of saving to database

[1719] Specific operation: The device sends the entered basic information to the server, which verifies the data and saves it in the database.

[1720] Step 4: Select a job template

[1721] Users log in to the system from their terminal and select job posting templates prepared for each job type and department.

[1722] Input: Login credentials, Job template list

[1723] Output: Selected template

[1724] What happens: A user logs into the system and selects from multiple templates.

[1725] Step 5: Enter job requirements

[1726] Based on the template you select, you enter details such as job title, desired skills, years of experience, and salary range.

[1727] Input: Selected template, job requirements (job type, skills, years of experience, etc.)

[1728] Output: Job requirements data entered

[1729] Specific behavior: The user fills out a form with detailed job requirements according to the template.

[1730] Step 6: Submit your job requirements

[1731] The terminal sends the job requirements to the server.

[1732] Input: Job requirements data entered

[1733] Output: Status of sending to server

[1734] Specific operation: The terminal sends the entered job requirements to the server.

[1735] Step 7: Generate a draft job posting

[1736] Based on the job requirements received by the server, a draft job posting is generated using a generative AI model.

[1737] Input: Received job requirements data

[1738] Output: Draft job posting

[1739] Specific operation: The server uses a generative AI model (e.g., OpenAI GPT-3) to generate a draft job posting based on the job requirements.

[1740] Step 8: Review and revise your draft

[1741] The user reviews the draft job posting and makes any necessary corrections.

[1742] Input: Draft job posting

[1743] Output: Final revised job posting

[1744] What happens: The user reviews the draft provided by the server and makes any necessary corrections.

[1745] Step 9: Submit a publishing request

[1746] The user reviews the final job posting and sends a publish request to the server.

[1747] Input: Final revised job posting

[1748] Output: Publish request

[1749] Specific behavior: The user clicks the "Publish" button to send a publication request.

[1750] Step 10: Publish your job posting

[1751] The server stores the final job posting in a database and publishes it on external job sites or the company's own recruitment page.

[1752] Input: Publication request, final job posting

[1753] Output: Job posting publication status

[1754] Specific operation: The server saves the final job posting in the database and reflects it on external job sites and the company's own recruitment page via API.

[1755] Step 11: Receiving feedback

[1756] After the job posting is published, feedback from users is received by the server and stored in a database.

[1757] Input: User feedback

[1758] Output: Stored feedback data

[1759] Specific operation: After a job posting is published, the user enters feedback information into the system, and that information is sent to the server.

[1760] Step 12: Analyze feedback

[1761] The server analyzes the feedback information and uses it to adjust the job posting generation algorithm.

[1762] Input: Stored feedback data

[1763] Output: Analysis results and algorithm adjustments

[1764] Specific operation: The server analyzes the received feedback data and adjusts the algorithm to reflect it when generating the next job posting.

[1765] Step 13: Generate and send scout emails (future feature)

[1766] The user inputs the criteria for the person to whom they want to send a scouting email. The server generates a scouting email based on the criteria and sends it to the candidate. The server tracks the sending results and provides feedback to the user.

[1767] Input: Target conditions for scouting emails (job type, skills, working conditions, etc.)

[1768] Output: Scout email sending status and tracking data

[1769] Specific operation: The user inputs the target conditions, and the server uses the generative AI model to generate and send scouting emails. The server tracks the sending results and provides feedback.

[1770] (Application example 1)

[1771] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1772] Corporate recruitment is a complex process that requires a great deal of effort and time. Particularly in factories, quickly and efficiently recruiting the right talent is directly linked to production efficiency. However, delays in creating and publishing job postings and analyzing feedback can lead to the risk of missing out on key talent. Furthermore, on-site talent management is even more difficult, complicating the management work of dedicated personnel. There is a need for a system that can solve these issues and achieve a more efficient and accurate recruitment process.

[1773] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1774] In this invention, the server includes: [means for the server to receive basic information and save it in a database;] [means for the user to select a job posting template and input job requirements;] [means for the server to use a generative AI model based on the information received to generate a draft of the job posting;] [means for the user to check and correct the draft;] [means for the server to link and publish the final version of the job posting to an external recruitment media;] [means for the robot to support employee information management within the factory; and [means for the robot to input job information using a terminal set up and send it to the server. This enables more efficient human resource management within the factory and the generation and publication of accurate job information.

[1775] A "server" is a central system for data management and processing.

[1776] "Basic information" refers to basic information such as the company's name, location, industry, and management policy.

[1777] A "database" is a system for storing and managing data such as basic information and feedback.

[1778] "User" refers to the company personnel and administrators who operate the system.

[1779] A "template" is a predefined format for quickly creating job postings.

[1780] "Job requirements" refer to details such as the desired job type, required skills, years of experience, salary range, etc.

[1781] A "generative AI model" is an artificial intelligence technology that generates a draft job posting based on input information.

[1782] "Draft" refers to the first unfinished job posting generated.

[1783] "External recruitment media" refers to online platforms and sites that disseminate job information to job seekers.

[1784] A "robot" is an automated mechanical device used in factories.

[1785] "Information management" refers to the process of organizing and storing basic employee information and recruitment data.

[1786] A "terminal" is an electronic device that allows a user to input information.

[1787] System Program Overview

[1788] This invention is a "smart robot personnel assistant" system that streamlines personnel management within factories and supports the accurate creation and publication of job information. This system operates in cooperation with a server, terminals, and robots.

[1789] Server Features

[1790] The server is a central system that manages and processes data. It receives basic information (company name, location, industry, management policy, etc.) entered by users through their devices and stores it in a database. It also uses a generative AI model to generate a draft job posting based on the template selected by the user and the entered job requirements (job type, skills, years of experience, salary range, etc.). The generated draft is then reviewed and revised by the user, and the final version is published by the server to external job media.

[1791] Robot Functions

[1792] The robot supports human resource management within the factory. It uses a set-up terminal to input job information and send it to a server, which can then be used directly by on-site personnel within the factory.

[1793] Data processing and calculation

[1794] 1. Register and manage basic information:

[1795] The user enters basic company information from the terminal, and the server receives it and stores it in a database.

[1796] Hardware used: Device (PC, tablet, etc.)

[1797] Software used: Python, Requests library, RESTful API, database management system (MySQL, etc.)

[1798] 2. Generate and publish job posting:

[1799] The user inputs the job posting template and detailed requirements from the terminal, and the server uses a generative AI model (e.g., GPT-3) to generate a draft job posting.

[1800] Hardware used: Server

[1801] Software used: Generative AI models (e.g., GPT-3)

[1802] The generated draft is then checked and revised by the user, and the server publishes the final version to external recruitment media.

[1803] 3. Feedback analysis:

[1804] The server stores the feedback received after publication in a database and analyzes it to help generate the next job posting.

[1805] Software used: Feedback Analysis Engine

[1806] Specific examples

[1807] Basic information registration:

[1808] The factory's human resources staff enters basic information such as "XYZ Manufacturing, Tokyo, manufacturing industry, emphasis on quality and efficiency" into a terminal, and the server stores this information in a database.

[1809] Job listing generation:

[1810] The same person selects the "Line Operator" template and enters the desired skills of "Machine Operation" and "Quality Control," the number of years of experience of "2 years or more," and the salary range of "3 million to 5 million yen per year." Based on this, the server uses a generative AI model to generate the following prompt:

[1811] Please create a job posting for a line operator. The required skills are "machine operation" and "quality control," the required years of experience are "2 years or more," and the salary range is "annual salary of 3 million to 5 million yen."

[1812] The resulting draft is then reviewed by a human resources representative and revised as necessary. The final job posting is then published by the server to external recruitment media.

[1813] This will enable more efficient human resource management within the factory and generate accurate job information.

[1814] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1815] Step 1:

[1816] The user enters basic information through the terminal and sends it to the server. The basic information entered includes the company name, location, industry, management policy, etc. The server receives this information and stores it in the database. The input data is saved in the appropriate table in the database.

[1817] Step 2:

[1818] The user selects a job posting template from their device and enters the job requirements, including the job type, desired skills, years of experience, salary range, etc. This input data is sent to the server, which receives it and prepares it as input prompts for the generative AI model.

[1819] Step 3:

[1820] Based on the job requirements and template information received by the server, a generative AI model (e.g., GPT-3) is used to generate a draft job posting. A prompt is input to the generative AI model, which then generates a draft job posting. The generated draft is saved on the server and provided to the user.

[1821] Step 4:

[1822] The user checks the draft on their device and makes any necessary corrections. The information corrected by the user is sent to the server, which updates the draft again and generates the final version. This final version of the job posting is also saved in the database.

[1823] Step 5:

[1824] The server connects the final job posting to an external recruitment site and publishes it. Specifically, the job information is sent to the external site using methods such as API or FTP. Information on successful publication of the job posting is also saved in the database as tracking information.

[1825] Step 6:

[1826] The server receives external feedback and stores it in a database. The feedback is processed by the analytics engine and used to adjust the job posting generation algorithm for the next job posting, thereby improving the job posting process for future jobs.

[1827] Step 7:

[1828] When a user wants to send a scout email based on specific criteria, they input the criteria from their device and send it to the server. The server searches for information that matches the criteria, then uses a generative AI model to generate and send the scout email. The server tracks the results of the email and provides feedback to the user.

[1829] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1830] An embodiment of the present invention is a system that combines a generative AI service that supports corporate recruitment operations with an emotion engine. This system operates in cooperation with a server, a terminal, a user, and the emotion engine.

[1831] 1. Registering and managing basic information

[1832] The server starts up, opens an API endpoint, and prepares to communicate with the device. The user enters basic information through the device, and the server stores that information in a database. This basic information includes the company name, location, industry type, management policy, etc.

[1833] Examples:

[1834] A human resources manager at an IT company logs into the system and enters basic information about the company (for example, "ABC Company, Tokyo, software development industry, corporate culture that values ​​innovation"). The server receives this information and stores it in a database.

[1835] 2. Job posting generation

[1836] Users log in to the system from their terminal and select a job posting template. Templates are available in multiple formats for different job types and departments. They then enter details such as job type, desired skills, years of experience, and salary range according to the template they select.

[1837] The server receives this information and uses a natural language processing engine to generate a draft job posting, which is then provided to the user for review and modification.

[1838] Examples:

[1839] A human resources representative from the same IT company selects the "Java Developer" template and inputs the desired skills (Java, Spring Framework), years of experience (3+ years), salary range (annual salary 6-8 million yen), etc. The server then generates a draft, and the user can review the proposed content and make any necessary revisions.

[1840] 3. Publication of job postings

[1841] The user checks the final version of the job posting and sends a publishing request to the server. The server saves the job posting and publishes it on external job sites or the company's own recruitment page.

[1842] Examples:

[1843] Once the HR person has reviewed the draft and clicked the "Publish" button, the server saves the job posting and it is instantly updated on linked external job sites and the company's own recruitment page. This entire process is completed in just a few minutes, enabling quick recruitment activities.

[1844] 4. Receiving and Analyzing Feedback

[1845] User feedback is received by the server and stored in a database. The server analyzes the feedback and can use it to improve the job posting generation algorithm. An emotion engine is used to analyze the emotional tone of the feedback and identify areas for further improvement.

[1846] Examples:

[1847] After publishing a job posting, users input reactions from job seekers and internal feedback (for example, "few applicants" or "the description is difficult to understand") into the system. This information is stored on the server and analyzed using an emotion engine. The results of the analysis are reflected the next time a job posting is generated.

[1848] 5. Generate and send scout emails (future feature)

[1849] The user inputs the criteria for the person to whom they want to send a scout email. The server generates and sends the scout email based on this information. The emotion engine adjusts the content of the scout email appropriately. The server tracks the sending results and provides feedback to the user.

[1850] Examples:

[1851] If a human resources manager wants to send a scouting email to a candidate who is a "full-stack engineer" and "able to work remotely," he or she can enter those criteria into the system. The server automatically generates a scouting email to candidates who fit the criteria, and the emotion engine adjusts the content before sending it. Later, the server tracks the response data to the scouting email and provides it to the human resources manager.

[1852] 6. Adjusting the emotional tone when generating job postings

[1853] When users create a job posting, the emotion engine analyzes their input in real time and suggests appropriate emotional tones (e.g., proactive, friendly, professional, etc.), making job postings more engaging and effective.

[1854] Examples:

[1855] As users enter their skills, years of experience, and other information into a job posting template, the emotion engine analyzes the information and displays suggestions such as "use a more positive tone." This makes the job posting more likely to catch the attention of job seekers.

[1856] In this way, the present invention effectively and efficiently supports companies' recruitment operations, and realizes the generation of sophisticated job postings and scouting emails that take into account the user's emotions.

[1857] The processing flow will be explained below.

[1858] Step 1:

[1859] The server starts and opens an API endpoint, allowing devices to communicate with the server.

[1860] Step 2:

[1861] The terminal loads the user interface, allowing the user to access the system.

[1862] Step 3:

[1863] A user logs into the system through a terminal and enters a user ID and password.

[1864] Step 4:

[1865] The server checks the user's authentication information against the database, and if authentication is successful, displays a basic information input screen for the user.

[1866] Step 5:

[1867] The user enters basic information about the company (company name, location, industry, management policy, etc.) and clicks the "Submit" button.

[1868] Step 6:

[1869] The server receives the basic information entered and stores it in a database.

[1870] Step 7:

[1871] The user clicks the "Create a new job posting" button on their device and selects an appropriate template from a list of job posting templates.

[1872] Step 8:

[1873] The server receives the user's template selection request and displays the template form on the user interface.

[1874] Step 9:

[1875] The user enters job requirements such as job type, desired skills, experience, and salary range according to the template, and clicks the "Complete input" button.

[1876] Step 10:

[1877] The server receives the entered job requirements and temporarily stores them.

[1878] Step 11:

[1879] The server invokes a natural language processing engine and generates a draft job posting based on the received information.

[1880] Step 12:

[1881] The generated draft job posting is sent to the user's terminal and displayed on the user interface.

[1882] Step 13:

[1883] The user checks the draft content on the terminal and makes corrections as necessary.

[1884] Step 14:

[1885] The emotion engine analyzes the user's input and suggests an appropriate emotional tone (e.g., proactive, friendly, professional, etc.).

[1886] Step 15:

[1887] The user considers the emotional tone suggestions and modifies the content of the job posting.

[1888] Step 16:

[1889] After making corrections, the user clicks the "Confirm" button to approve the final version of the job posting.

[1890] Step 17:

[1891] The server stores approved job postings in a database and then links to external job sites and the company's own recruitment page via an API, making them publicly available.

[1892] Step 18:

[1893] Users conduct job postings and input responses from job seekers and internal feedback into the system.

[1894] Step 19:

[1895] The server receives the feedback information and stores it in a database.

[1896] Step 20:

[1897] The server analyzes the feedback and uses it to improve the job posting generation algorithm.

[1898] Step 21:

[1899] An emotional engine analyzes the emotional tone of the feedback and identifies areas for improvement.

[1900] Step 22:

[1901] The user enters the conditions for the person to whom the scout email is to be sent and clicks the "Conditions input complete" button.

[1902] Step 23:

[1903] The server receives the conditions and generates a scout email.

[1904] Step 24:

[1905] The emotional engine analyzes the content of the scouting email and adjusts the appropriate emotional tone.

[1906] Step 25:

[1907] The server sends the generated scout email to the candidate.

[1908] Step 26:

[1909] The server tracks the results of the transmission and provides feedback to the user through the user interface.

[1910] Example 2

[1911] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1912] In corporate recruitment processes, there are issues with the inefficiency of creating and publishing job postings, receiving and analyzing feedback, and creating and sending scouting emails. In particular, job postings and scouting emails are created only in a standardized format, and there is a lack of a system for collecting and analyzing applicant responses to improve the content. Furthermore, there is a lack of emotional tone adjustment to create job postings and scouting emails that are appealing to job seekers.

[1913] The specification process by the specification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for receiving basic information and saving it in a database, a means for a user to select a job posting template and input job requirements, and a means for using a natural language processing engine to generate a draft of a job posting based on the received information. This enables companies to carry out their recruitment operations efficiently.

[1914] It also includes a means for publishing the final job posting on external job sites or the company's own recruitment page, and a means for the emotion engine to analyze user input in real time and suggest an appropriate emotional tone, making the job posting more attractive and effective for job seekers.

[1915] Furthermore, the system includes a means for the server to receive feedback and store it in a database, a means for analyzing the feedback information and adjusting the job posting generation algorithm, and a means for the emotion engine to analyze the emotional tone of the feedback and reflect it in the next job posting generation, thereby enabling the content of the job posting to be dynamically improved and optimized.

[1916] It also includes a means for users to input conditions for sending scout emails, a means for generating scout emails based on the conditions and sending them after adjustments by the emotion engine, and a means for tracking the sending results and providing feedback to the user, thereby enabling the generation of effective scout emails that match the target, and the confirmation and improvement of their effectiveness.

[1917] A "server" is a computer system that can be accessed remotely over a network and that stores, processes, and provides services to data.

[1918] A "terminal" is a device that is directly operated by a user and provides a user interface, such as a PC or smartphone.

[1919] A "user" is someone who operates the system and inputs data, such as a company's recruiter.

[1920] "Basic information" refers to basic data about a company, such as the company name, location, industry type, and management policy.

[1921] A "template" is a format used when creating a job posting, and is a pre-designed framework for each job type or department.

[1922] "Employment conditions" are employment conditions listed in a job posting, such as job type, required skills, years of experience, and salary range.

[1923] A "natural language processing engine" is an algorithm for analyzing and generating text data, and is a system that creates text using a generative AI model.

[1924] "Draft" is the first draft of the job posting that is generated, and is text that contains content that the user needs to correct and confirm.

[1925] An "emotion engine" is an algorithm that analyzes the emotional tone of text data and suggests appropriate expressions.

[1926] "Feedback" refers to opinions and reactions provided by users and job seekers, as well as data based on them.

[1927] An "algorithm" is a defined set of procedures or computational steps for solving a problem.

[1928] A "scout email" is a recruiting message sent to candidates who meet certain hiring criteria.

[1929] "Tracking" is the process of tracking and recording the results and responses to emails sent.

[1930] An embodiment of the present invention is a system for supporting a company's recruitment operations, in which a server, a terminal, a user, and an emotion engine work together. This system mainly provides a series of functions for receiving, storing, analyzing, generating, and publishing data.

[1931] First, when the server starts up, it opens an API endpoint and prepares for communication with the terminal. The hardware used here is a typical server computer, and the software includes a database management system (DBMS) and API management tools. The user enters basic company information (e.g., company name, location, industry type, management policy, etc.) through the terminal, and the server stores that information in a database.

[1932] Next, the user logs into the system and selects a job posting template. Templates are available in multiple formats and are categorized by job type and department. The user enters details such as the job type, desired skills, years of experience, and salary range according to the template they select. The server receives this information and generates a draft job posting using a natural language processing engine (e.g., GPT-3). This process utilizes a generative AI model. The generated draft is then provided to the user, who can review and modify it.

[1933] The user reviews the final version of the job posting and sends a publication request to the server. The server saves the job posting in a database, links it to external job sites and the company's own recruitment page, and publishes the information. The software used at this stage is an external API integration tool and web server software.

[1934] Receiving and analyzing feedback is also an important function. User feedback is received by the server and stored in a database. The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional tone of the feedback and use it to improve the job posting generation algorithm. The analyzed data is reflected in the next job posting generation.

[1935] In the future, the system will also provide the ability to generate and send scout emails. Users can enter the conditions for sending a scout email, and the server will generate the email based on the conditions. The emotion engine will then adjust the content appropriately. The server will then send the email, track the results, and provide feedback to the user.

[1936] Additionally, when users create a job posting, the emotion engine analyzes their input in real time and suggests appropriate emotional tones (e.g., proactive, friendly, professional, etc.), making job postings more engaging and effective.

[1937] Examples:

[1938] A human resources representative from a company logs into the system and enters basic company information, which the server then saves in a database. Next, they select the "Java Developer" template and enter their details. The server then uses a natural language processing engine to generate a draft job posting. The representative then reviews the final job posting and clicks the publish button, which instantly updates both the company's recruitment page and external job sites. Feedback is also received, analyzed by an emotion engine, and reflected in the next job posting, allowing the company to provide better job content.

[1939] Example prompts using a generative AI model:

[1940] "I'm looking to create a job posting for a Java developer. The required skills are Java and Spring Framework, with at least three years of experience and an annual salary range of ¥6,000,000 to ¥8,000,000."

[1941] With the above configuration, the present invention provides an efficient recruitment support system and improves the recruitment process of a company.

[1942] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1943] Step 1:

[1944] The server starts, opens an API endpoint, and prepares to communicate with the device.

[1945] Input: System startup command

[1946] Output: API endpoint status

[1947] Specific operation: The server accepts the start command and opens an API endpoint (URL) for network communication. The API endpoint waits for requests from the device.

[1948] Step 2:

[1949] The user inputs basic information through the terminal, and the server receives the information.

[1950] Input: Basic company information (company name, location, industry type, management policy, etc.)

[1951] Output: Acknowledgement message

[1952] Specific operation: A user fills in the basic information of their company in the input form on their device and clicks the submit button. The server receives this request, analyzes the data, and then generates and executes an SQL query to save it in the database.

[1953] Step 3:

[1954] The server stores the basic information in a database.

[1955] Input: Basic information data

[1956] Output: Message that saving to database is complete

[1957] Specific operation: The server saves the received basic information in the database management system. If the save is successful, it returns a message to the user indicating that the save was successful.

[1958] Step 4:

[1959] A user logs into the system from a terminal and selects a job posting template.

[1960] Input: Login information (username, password), template to select

[1961] Output: Template selection completion message

[1962] Specific behavior: A user enters their credentials on the login screen to access the system. A list of job templates is displayed, and the user selects the appropriate template.

[1963] Step 5:

[1964] Users enter details such as job title, desired skills, years of experience, and salary range.

[1965] Input: Job type, desired skills, years of experience, salary range

[1966] Output: Input completion message

[1967] Specific operation: The user fills in various details in the input form according to the template selected by the user. After completion, the input contents are sent to the server.

[1968] Step 6:

[1969] The server passes the received information to a natural language processing engine to generate a draft job posting.

[1970] Input: Job details

[1971] Output: Draft job posting

[1972] How it works: The server receives the job details and passes them to a generative AI model (e.g., GPT-3) in the form of a prompt. The natural language processing engine analyzes and generates the prompt to create a draft job posting.

[1973] Example prompt sentence:

[1974] "I'm looking to create a job posting for a Java developer. The required skills are Java and Spring Framework, with at least three years of experience and an annual salary range of ¥6,000,000 to ¥8,000,000."

[1975] Step 7:

[1976] The server returns the generated draft to the user for confirmation.

[1977] Input: Draft job posting

[1978] Output: Draft displayed to user

[1979] Specific operation: The server sends the generated draft to the terminal and displays it to the user, allowing the user to check the contents of the draft and make any necessary corrections.

[1980] Step 8:

[1981] The user reviews the final job posting and sends a publish request to the server.

[1982] Input: Last confirmed job posting

[1983] Output: Publishing request sent successfully message

[1984] Specific behavior: The user reviews the final job posting and clicks the "Publish" button to send a request to the server.

[1985] Step 9:

[1986] The server stores the job postings in a database and links them to external job sites and the company's own recruitment page for publication.

[1987] Input: Publishing request

[1988] Output: Publishing complete message

[1989] Specific operation: The server saves the job posting data in a database, and at the same time, it links the job information to external job sites and the company's own recruitment page via API and publishes it immediately.

[1990] Step 10:

[1991] The user enters the feedback and the server receives it.

[1992] Input: Feedback information

[1993] Output: Feedback acknowledgement message

[1994] Specific operation: The user inputs responses and internal feedback from job seekers into the terminal and sends them to the server, which receives them and stores them in a database.

[1995] Step 11:

[1996] The server analyzes the feedback and uses an emotion engine to reflect it in the next job posting generation.

[1997] Input: Feedback information

[1998] Output: Analysis results and algorithm adjustment data

[1999] What it does: The server analyzes the feedback information, evaluates the emotional tone using an emotion engine, and adjusts the job posting generation algorithm based on the analysis results to reflect them in the next generation.

[2000] Step 12:

[2001] The user inputs the conditions for sending the scout mail, and the server generates and sends the scout mail.

[2002] Input: Scout mail conditions

[2003] Output: Generated scout email

[2004] Specific operation: The user inputs the target conditions for the scout email and sends it to the server. The server generates the scout email based on the conditions, adjusts the content using the emotion engine, and sends it.

[2005] Step 13:

[2006] The server tracks the results of the submission and provides feedback to the user.

[2007] Input: Scout email sending conditions

[2008] Output: Sending results and tracking data

[2009] Specific operation: The server tracks the results of sending scout emails, collects reactions and response data, analyzes it, and provides it to the user as feedback.

[2010] (Application example 2)

[2011] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2012] Conventional ad generation systems rely on manual creation of ad copy, which makes it difficult to properly appeal to target users, especially due to a lack of emotional tone adjustment. Furthermore, the process of receiving feedback and incorporating it into the ad generation algorithm is manual and inefficient. Furthermore, individual ads are not properly sent to target users, making it difficult to maximize advertising effectiveness.

[2013] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[2014] In this invention, the server includes: [means for the server to receive basic information and store it in a database;] [means for the user to select an advertisement template and input advertisement conditions;] [means for the server to generate a draft advertisement using a natural language processing engine based on the information received;] [means for the user to check and revise the draft;] [means for the server to link and publish the final advertisement to an external advertising platform; and [means for analyzing the emotional tone of the advertisement content using an emotion engine and adjusting it as necessary.] This automates the entire process from advertisement generation to publication and feedback analysis, making it possible to appropriately adjust the emotional tone of the advertisement content. In addition, advertisements can be individually generated and sent based on the recipients of targeted advertisements, maximizing the effectiveness of advertising.

[2015] A "server" is a computer system that has the function of receiving data from client terminals and storing it in a database.

[2016] "Basic information" refers to information such as company name, location, industry type, and management policy that is necessary for creating advertisements and job offers.

[2017] A "database" is a system for systematically storing and managing received basic information and feedback information.

[2018] A "template" is an advertisement or job posting format based on a specific format that can be selected and used by a user.

[2019] "Advertising conditions" are information input as the content of the advertisement to be generated, and include, for example, the title, features, price, and the like.

[2020] A "natural language processing engine" is software that automatically generates drafts of advertisements and job postings based on received information.

[2021] A "draft" is an early version of an advertisement or job posting that can be reviewed and modified by the user.

[2022] "Advertising Platform" means an external service for publishing generated advertisements, including the Internet and mobile apps.

[2023] An "emotion engine" is software that analyzes the emotional tone of advertisements and job postings and adjusts it as needed.

[2024] "Feedback" refers to opinions and evaluations from users and viewers regarding advertisements and job postings.

[2025] "Targeted advertising" means advertising that is generated and delivered to an individual user based on specific criteria.

[2026] "Tracking" refers to tracking the results of sending advertisements and the status of receiving feedback.

[2027] This embodiment relates to an advertisement generation system, and in particular provides a system that combines a generation AI and an emotion engine. To implement the invention, the following main components are included:

[2028] 1. Server

[2029] 2. Client Terminal

[2030] 3. Emotion Engine

[2031] 4. Natural Language Processing Engine

[2032] 5. Database

[2033] 6. Advertising Platform

[2034] The server receives basic information sent from the client terminal and stores it in a database. Basic information includes the company name, location, industry type, management policy, etc. Based on this information, an ad template is selected and ad conditions are entered.

[2035] Using a client terminal, a user selects an ad template and enters the ad conditions, including detailed information such as the title, features, and price. Once the server receives this information, it uses a natural language processing engine to generate a draft ad. This draft is an initial version that the user can review and modify.

[2036] The emotional engine analyzes the emotional tone of the generated draft ad and adjusts it as necessary, ensuring that the ad content appeals appropriately to the target user. Once the user reviews the final version of the ad and sends a publishing request to the server, the server connects the final version of the ad to an external advertising platform and publishes it.

[2037] Furthermore, the server has the function of receiving feedback and storing it in a database. The feedback information is analyzed using an emotion engine and reflected in the ad generation algorithm. This process continuously improves the effectiveness of the ads.

[2038] Users can also input specific conditions when sending targeted ads, and the server will generate targeted ads based on these conditions, adjust them using the emotion engine, and send them. The results of these ads are tracked by the server and fed back to the user.

[2039] For example, if a user wants to generate an ad with the title "An exciting new product!" and the features "High-performance battery, easy-to-use UI, and 1-year warranty," they would use the following prompt:

[2040] prompt:

[2041] Use the company information below to generate copy that strikes a positive and friendly tone.

[2042] Company information: A company, large city, e-commerce industry, customer satisfaction focused

[2043] Advertisement details: Title: "An exciting new product!" Features: "High-performance battery, easy-to-use UI, 1-year warranty"

[2044] Desired tone: Positive and friendly

[2045] This allows users to easily generate ads using generative AI models and obtain ad copy that is tailored to the optimal emotional tone using an emotional engine.

[2046] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2047] Step 1:

[2048] The server receives the basic information and stores it in the database. This is the process of receiving information such as the company name, location, industry type, and management policy sent from the client terminal. The received basic information is stored in the database as is in JSON format. This process ensures that the company's basic information is properly stored in the system.

[2049] Input: Basic information such as company name, location, industry type, management policy, etc. (JSON format)

[2050] Data processing: Save basic information to a database

[2051] Output: Basic information stored in the database

[2052] Step 2:

[2053] The user selects an ad template and enters the ad conditions. The user logs in to the system using a client terminal and selects the desired template from the presented templates. Then, the user enters the necessary ad conditions (title, features, price, etc.).

[2054] Input: Template selection information, advertising conditions (title, features, price, etc.)

[2055] Data processing: converting user input into the appropriate format and sending it to the server

[2056] Output: Template selection information and ad conditions sent to the server

[2057] Step 3:

[2058] The server uses a natural language processing engine based on the received information to generate a draft ad. The server runs the natural language processing engine based on the entered ad conditions and the selected template. This automatically generates an initial draft.

[2059] Input: Template selection information, advertising conditions

[2060] Data processing: Draft generation using a natural language processing engine

[2061] Output: Initial Draft

[2062] Step 4:

[2063] The user checks and modifies the draft. The generated draft is checked on the client terminal and modified as necessary. The modified draft is then sent back to the server.

[2064] Input: Initial Draft

[2065] Data processing: User confirmation and correction

[2066] Output: Revised draft

[2067] Step 5:

[2068] The server uses an emotion engine to analyze the emotional tone of the ad content and adjust it accordingly. The emotion engine analyzes the emotional tone of the revised draft and adjusts it to the specified emotional tone (e.g., positive, friendly).

[2069] Input: Revised draft

[2070] Data processing: Emotional tone analysis and adjustment

[2071] Output: Final draft with emotional tone applied

[2072] Step 6:

[2073] The server links the final version of the advertisement to the advertising platform and publishes it. Upon receiving the publishing request, the server links the final draft to the specified advertising platform and publishes the advertisement.

[2074] Input: Final draft with emotional tone applied

[2075] Data processing: Collaboration with advertising platforms

[2076] Output: Published ad

[2077] Step 7:

[2078] The server receives the feedback and stores it in a database. It then analyzes the feedback information using an emotion engine and reflects it in the ad generation algorithm, which will result in more effective ad generation in the future.

[2079] Input: Feedback information

[2080] Data processing: storing and analyzing feedback information

[2081] Output: Improved ad generation algorithm

[2082] Step 8:

[2083] The user inputs the conditions for targeting the advertisement to be sent. The user inputs the conditions for targeting the advertisement to be sent using the client terminal.

[2084] Input: Criteria for sending targeted ads

[2085] Data processing: Send condition information to the server

[2086] Output: Condition information sent to the server

[2087] Step 9:

[2088] The server generates and sends targeted ads based on the conditions. The emotion engine adjusts the content of the ads and automatically sends them to the specified targets.

[2089] Input: Criteria for sending targeted ads

[2090] Data processing: Emotion engine adjustment, ad generation and delivery

[2091] Output: Targeted ads sent

[2092] Step 10:

[2093] The server tracks the results of the advertisements sent, collecting response data (number of views, number of clicks, etc.) and generating feedback based on that data.

[2094] Input: Targeted ads sent

[2095] Data processing: Reaction data collection and analysis

[2096] Output: Tracking data, feedback

[2097] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2098] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2099] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2100] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2101] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2102] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2103] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2104] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2105] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2106] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2107] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2108] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2109] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored ...

Claims

1. A means for the server to receive the basic information and store it in a database; A means for a user to select a job posting template and input job requirements; a means for generating a draft of the job posting using a natural language processing engine based on the information received by the server; a means for users to review and revise the draft; The server will link the final job posting to an external job site and publish it. A system including:

2. a means for the server to receive the feedback and store it in a database; The server further includes means for analyzing the feedback information and adjusting the job posting generation algorithm. The system of claim 1 .

3. A means for a user to input conditions for sending scout emails; A means for the server to generate and send a scout email based on conditions; The server further includes means for tracking the results of the transmission. The system of claim 1 .

Images