Information processing system, information processing method, and program

Abstract

An information processing system, method, and program are provided that can efficiently match employers and job seekers. [Solution] In an information processing system in which a server device, a plurality of employer terminals, and a plurality of job seeker terminals can communicate with each other via communication lines, a control unit of the server device has an employer agent construction unit that constructs, for each employer, an employer task execution model capable of executing tasks handling information about the employer through machine learning using employer learning data about the employer, and a job seeker agent construction unit that constructs, for each job seeker, a job seeker task execution model capable of executing tasks handling information about the job seeker through machine learning using job seeker learning data about the job seeker, and a task execution unit that causes the employer task execution model or the job seeker task execution model to execute the task based on a task execution instruction for the employer task execution model or the job seeker task execution model.

Description

[Technical field]

[0001] The present invention relates to an information processing system, an information processing method, and a program. [Background technology]

[0002] Patent Document 1 discloses a technology for matching job seeker information with recruitment requirements. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2007-79900 A Summary of the Invention [Problem to be solved by the invention]

[0004] There is a demand for technology that can more efficiently match employers and job seekers.

[0005] In view of the above circumstances, the present invention provides an information processing system etc. capable of efficiently matching employers and job seekers. [Means for solving the problem]

[0006] According to one aspect of the present invention, there is provided an information processing system comprising at least one processor, the processor being configured to execute each of the following steps by reading a program: in an employer agent construction step, a employer task execution model capable of executing tasks handling information of the employer is constructed for each employer through machine learning using employer learning data related to the employer, the employer learning data including information of the employer registered in the employer database, information provided by the employer, or public information related to the employer on the network; in a job seeker agent construction step, a job seeker task execution model capable of executing tasks handling information of the job seeker is constructed for each job seeker through machine learning using job seeker learning data related to the job seeker, the job seeker learning data including information of the job seeker registered in the job seeker database, information provided by the job seeker, or public information related to the job seeker on the network; and in a task execution step, an information processing system is provided whereby the employer task execution model or the job seeker task execution model is caused to execute a task based on a task execution instruction for the employer task execution model or the job seeker task execution model.

[0007] According to this embodiment, it is possible to efficiently match recruiters and job seekers using task execution models that are individually constructed for each recruiter and job seeker. [Brief description of the drawings]

[0008] [Figure 1] 1 is a configuration diagram illustrating an information processing system 1. [Diagram 2] 2 is a block diagram showing a hardware configuration of the server device 10. FIG. [Diagram 3] FIG. 2 is a block diagram showing the hardware configuration of a recruiter terminal 20 and a job seeker terminal 30. [Figure 4] FIG. 2 is a block diagram showing functions realized by a server device 10 (control unit 11), a recruiter terminal 20 (control unit 21), and a job seeker terminal 30 (control unit 31). [Diagram 5]FIG. 13 is a diagram illustrating an image of matching between employers and job seekers using a job employer task execution model and a job seeker task execution model. [Figure 6] 1 is an activity diagram showing an example of the flow of information processing (task execution processing) executed by the information processing system 1. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. Various characteristic features shown in the following embodiments can be combined with each other.

[0010] Incidentally, a program for realizing the software appearing in one embodiment may be provided as a non-transitory computer-readable recording medium, or may be provided so as to be downloadable from an external server, or may be provided so that the program is launched on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0011] In addition, in various information processing according to an embodiment, an input and an output according to the input can be realized. Here, as long as an output is obtained as a result of the input, the form of information referenced in such information processing (hereinafter referred to as reference information) is not limited. The reference information may be, for example, rule-based information such as a database, a lookup table, or a predetermined function (including a judgment formula such as a regression formula constructed by a statistical method), or may be a trained model that has previously trained the correlation between the input and the output, or may be a large-scale language model that can output a desired result by inputting a prompt.

[0012] In one embodiment, the term "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In one embodiment, various information is handled, and this information is represented by, for example, physical values ​​of signal values ​​representing voltage and current, high and low signal values ​​as a binary bit collection consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculation can be performed on the circuit in the broad sense.

[0013] Furthermore, a circuit in the broad sense is a circuit realized by at least appropriately combining a circuit, circuitry, a processor, and a memory. The processor may be a general-purpose processor or a dedicated circuit. In other words, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0014] 1. Hardware Configuration This section describes the hardware configuration.

[0015] <Information Processing System 1> 1 is a configuration diagram showing an information processing system 1. The information processing system 1 comprises a communication line 2, a server device 10, a plurality of recruiter terminals 20, and a plurality of job seeker terminals 30. The server device 10, the recruiter terminals 20, and the job seeker terminals 30 are configured to be able to communicate with each other via the communication line 2. The connection between the server device 10, the recruiter terminals 20, and the job seeker terminals 30 may be wired or wireless.

[0016] The information processing system 1 constitutes at least a part of a recruitment and job search system used, for example, by multiple recruiters (a first recruiter U1 and a second recruiter U2) and multiple job seekers (a first job seeker U3 and a second job seeker U4). The information processing system 1 mainly performs functions such as searches for job seekers by recruiters and searches for job offers by job seekers, and mediates communication between recruiters and job seekers. In one embodiment, the information processing system 1 is made up of one or more devices or components. These components will be described below.

[0017] <Server device 10> Fig. 2 is a block diagram showing a hardware configuration of the server device 10. As shown in Fig. 2, the server device 10 includes a control unit 11, a storage unit 12, a communication unit 13, and a communication bus 14. The control unit 11, the storage unit 12, and the communication unit 13 are electrically connected to each other inside the server device 10 via the communication bus 14.

[0018] <Control unit 11> The control unit 11 performs processing and control of the overall operation related to the server device 10. The control unit 11 is, for example, a central processing unit (CPU). The control unit 11 realizes various functions related to the server device 10 by reading out a predetermined program stored in the storage unit 12. That is, information processing by software stored in the storage unit 12 can be specifically realized by the control unit 11, which is an example of hardware, and executed as each functional unit included in the control unit 11. These will be described in more detail in the next section. Note that the control unit 11 is not limited to being a single unit, and the server device 10 may have multiple control units 11 for each function. Also, the server device 10 may be configured with a combination of these.

[0019] <Storage section 12> The storage unit 12 stores various information defined by the above description. This can be implemented, for example, as a storage device such as a solid state drive (SSD) that stores various programs and the like related to the server device 10 executed by the control unit 11, or as a memory such as a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to the program calculations. The storage unit 12 stores various programs, variables, etc. related to the server device 10 executed by the control unit 11.

[0020] <Communications Division 13> The communication unit 13 is preferably a wired communication means such as USB, IEEE1394, Thunderbolt (registered trademark), wired LAN network communication, etc., but may also include wireless LAN network communication, mobile communication such as LTE / 5G, BLUETOOTH (registered trademark) communication, etc. as necessary. In other words, it is more preferable to implement it as a collection of multiple communication means. In other words, the server device 10 may communicate various information from the outside via the communication unit 13 and the network.

[0021] The server device 10 may be in an on-premise form or in a cloud form. The server device 10 in a cloud form may provide the above-mentioned functions and processes in the form of, for example, SaaS (Software as a Service) or cloud computing.

[0022] <Employer Terminal 20> 3 is a block diagram showing the hardware configuration of the recruiter terminal 20 and the job seeker terminal 30. The recruiter terminal 20 is an information processing terminal used by the recruiter. Recruiters include organizations such as profit-making corporations (e.g., companies, etc.), non-profit organizations (e.g., cooperatives, incorporated foundations, etc.), and public corporations (e.g., local governments, etc.) or their personnel. Recruiters also include human resource agencies that act as agents of organizations and act as intermediaries between job seekers and organizations. Human resource agencies are also called headhunters, agents, etc.

[0023] 3A, the recruiter terminal 20 comprises a control unit 21, a memory unit 22, a communication unit 23, an input unit 24, an output unit 25, and a communication bus 26. The control unit 21, the memory unit 22, the communication unit 23, the input unit 24, and the output unit 25 are electrically connected within the recruiter terminal 20 via the communication bus 26. Descriptions of the control unit 21, the memory unit 22, and the communication unit 23 will be omitted as they are similar to the descriptions of the respective units in the server device 10.

[0024] <Input section 24> The input unit 24 accepts operation inputs made by the user. The operation inputs are transferred as command signals to the control unit 21 via the communication bus 26. The control unit 21 may execute predetermined control or calculations based on the transferred command signals as necessary. The input unit 24 may be included in the housing of the recruiter terminal 20, or may be externally attached. For example, the input unit 24 may be implemented as a touch panel integrated with the output unit 25. When the input unit 24 is implemented as a touch panel, the user can input tap operations, swipe operations, and the like to the input unit 24. As the input unit 24, a switch button, a mouse, a track pad, a QWERTY keyboard, and the like can be adopted instead of a touch panel.

[0025] <Output section 25> The output unit 25 displays a screen of a graphical user interface (GUI) that can be operated by the user. The output unit 25 may be included in the housing of the recruiter terminal 20, or may be attached externally. Specifically, the output unit 25 may be implemented as a display device such as a CRT display, a liquid crystal display, an organic EL display, a plasma display, or the like. It is preferable that these display devices are implemented by using different devices depending on the type of recruiter terminal 20.

[0026] <Job Seeker Terminal 30> The job seeker terminal 30 is an information processing terminal used by a job seeker. "Job seekers" include, for example, employed people (those looking to change jobs), prospective new graduates (job-hunting students), students, and the like.

[0027] 3B, the job seeker terminal 30 includes a control unit 31, a memory unit 32, a communication unit 33, an input unit 34, an output unit 35, and a communication bus 36. The control unit 31, the memory unit 32, the communication unit 33, the input unit 34, and the output unit 35 are electrically connected within the job seeker terminal 30 via the communication bus 36. Descriptions of the control unit 31, the memory unit 32, the communication unit 33, the input unit 34, and the output unit 35 will be omitted as they are the same as the descriptions of the respective units in the recruiter terminal 20.

[0028] 2. Functional configuration In this section, the functional configuration of the present embodiment will be described. Information processing by software stored in the storage unit 12 is specifically realized by the control unit 11, which is an example of hardware, and can be executed as each functional unit included in the control unit 11 (at least one processor included in the information processing system 1).

[0029] FIG. 4 is a block diagram showing functions realized by the server device 10 (control unit 11), the recruiter terminal 20 (control unit 21), and the job seeker terminal 30 (control unit 31).

[0030] As shown in FIG. 4A, the server device 10 (control unit 11) includes a basic display control unit 111, an employer agent construction unit 112, an employer label creation unit 113, a job seeker agent construction unit 114, a job seeker label creation unit 115, a question registration unit 116, a platform agent construction unit 117, a task execution unit 118, and an artificial intelligence unit 120.

[0031] 4B, the recruiting party terminal 20 (control unit 21) includes a display unit 211 and an operation acquisition unit 212. As shown in FIG. 4C, the job seeker terminal 30 (control unit 31) includes a display unit 311 and an operation acquisition unit 312.

[0032] <Basic display control unit 111> The basic display control unit 111 is configured to display various information on the recruiter terminal 20 or the job seeker terminal 30. For example, the basic display control unit 111 displays information on job seekers or recruiters registered in the database on the display unit 211 of the recruiter terminal 20 or the display unit 311 of the job seeker terminal 30 in response to a request from each user (recruiters U1, U2 or job seekers U3, U4).

[0033] <Employer Agent Construction Division 112> The recruiter agent construction unit 112 is configured to construct, for each recruiter registered in the recruiter database, a recruiter task execution model capable of executing tasks that handle recruiter information, by machine learning using recruiter learning data related to recruiters. Note that in this specification, "constructing" a model is not limited to creating a new model, but also includes updating an existing model.

[0034] The recruiter task execution model may be an output device that outputs the execution results of an input task, or may be a so-called AI (artificial intelligence) agent (recruiter agent) that autonomously executes tasks such as data collection and conversation required to execute the task or goal input by the recruiter on behalf of the recruiter, using natural language processing, a processing model constructed by machine learning, a general-purpose large-scale language model, etc. An AI agent may also be called an autonomous agent. Conversation with the recruiter task execution model is conducted, for example, in the form of a chat using voice or text.

[0035] An "AI agent" is a model that, when a target (goal, objective, etc.) such as "teach me XX" or a task such as "output XX" is input, breaks down the processing required to reach the goal or accomplish the task into subtasks, actions, etc., and collects and analyzes necessary data, generates and executes programs, etc. An AI agent autonomously selects and executes tasks, actions, etc., and does not require user intervention (operation input). In addition, an AI agent may autonomously learn to achieve a goal by autonomously making and executing plans and evaluating the results of the execution. For example, an AI agent may be autonomously updated based on the results of the execution of subtasks (e.g., collected information, analysis results of information, etc.). An AI agent may include a large-scale language model including a generative AI that has been trained to be able to perform these processes.

[0036] Specifically, the AI ​​agent may execute a task, for example, in the following procedure. First, the AI ​​agent receives specific instructions, including a goal or a task, from a user. Next, the AI ​​agent divides the goal or task into multiple subtasks or actions necessary for achieving the goal or task, and generates the subtasks or actions. After dividing the goal or task, the AI ​​agent collects information (e.g., employer registration information, job seeker registration information, etc.) necessary to execute the multiple subtasks and actions. For example, the AI ​​agent collects information by accessing a database, the Internet, etc., web search, API integration with external services, conversation with other AI agents, referring to an employer memory or job seeker memory described later, etc. After collecting the information, the AI ​​agent executes the multiple subtasks and actions based on the collected information, for example, by using an external tool, etc. (e.g., generating and executing code), and outputs a task corresponding to the goal or the execution result of the input task.

[0037] In addition, the AI ​​agent may verify or evaluate the results of the subtask or action by itself, and if it determines that review is necessary, it may correct or improve the subtask or action and re-execute it. For example, after generating code to execute a specific subtask, the AI ​​agent may verify the code by itself and correct the code by providing feedback.

[0038] The tasks executed by the recruiter task execution model (hereinafter, "recruiter tasks") may include, for example, answers to questions about recruiters (who are the providers of recruiter learning data) to which the recruiter task execution model corresponds. In other words, the recruiter task execution model may be capable of answering questions about recruiters as a recruiter task. The recruiter task execution model may also be capable of asking questions to job seekers or other AI agents (job seeker task execution model or platform task execution model) as recruiter tasks or subtasks thereof. This allows the recruiter task execution model to have conversations or interviews with job seekers or other AI agents. As a result, for example, the recruiter task execution model can conduct pseudo-interviews with popular recruiters who are difficult to contact.

[0039] "Questions about the employer" include, for example, questions about the organization's performance, organizational goals, organizational structure, job regulations, organizational culture or atmosphere, job content (requirements for the job), the desired personnel profile, the hiring process, etc. In addition, questions that the employer task execution model itself asks to job seekers or other AI agents include, for example, questions to confirm the suitability (match degree) of the organization's goals, organizational culture, job requirements, etc. with the person asking the question (job seeker).

[0040] The recruiter task may include output of information about the recruiter specified by a task instructor (e.g., the task execution unit 118 described below). The recruiter task may also be information indicating a specific objective (goal, purpose, etc.) regarding the recruiter specified by (e.g., the task execution unit 118 described below). Such a recruiter task is, for example, an instruction such as "present job seekers recommended to the recruiter" or information indicating an objective such as "I would like to know recommended job seekers", and the information output by the recruiter task execution model includes, for example, information about job seekers recommended to the recruiter.

[0041] A "job seeker recommended to an employer" is, for example, a job seeker who is a high match for the employer or the employment information held by the employer, a job seeker who is estimated to be highly likely to reply to a scout email sent by the employer, or a job seeker who is estimated to be highly likely to be hired. The employer task execution model may determine a job seeker to be recommended to an employer by, for example, comparing the registered information of the job seeker to be judged, a job seeker memory described below, and the like with learned information about employers (for example, the employer memory described below) or information about employment (for example, converting each piece of information into a vector and judging the similarity based on the cosine similarity between the vectors, etc.).

[0042] The recruiter learning data used by the recruiter agent construction unit 112 includes information about recruiters registered in the recruiter database (registered information), information provided by recruiters (provided information), or public information about recruiters on the network.

[0043] The "employer database" is a database used in a platform that matches employers and job seekers and is provided by the information processing system 1. In other words, the employer learning data used by the employer agent construction unit 112 may include information on employers registered in a single employer database in a platform.

[0044] The recruiter's registration information (hereinafter "Recruiter Registration Information") includes information about the organization, such as information indicating the business content, industry, organizational size, performance, merchandise handled (products or services), financial situation, and business plan, for the recruiter itself (if the recruiter is an organization), the organization to which the recruiter belongs (if the recruiter is a person in charge of an organization), or the organization the recruiter represents (if the recruiter is a human resources agency). In addition, the recruiter registration information may include information about the entire organization (for example, the business content, industry, and performance of the entire organization), or may include information about a part of the organization (for example, a business division, department, etc.) (for example, the business content, industry, and performance of one business division).

[0045] Furthermore, the employer registration information may include job data (job postings) created by the organization, data on job seekers approached by the employer (e.g., actions such as sending scouting emails, job interviews, and job offers), applicant management system (ATS) data (e.g., attributes or characteristics of personnel employed by the organization), evaluation data on personnel within the organization (e.g., attributes or characteristics of personnel highly rated within the organization), etc.

[0046] In this way, the recruiter learning data may include at least one of the recruiter's recruitment-related data used in the recruitment management system and the recruiter's evaluation of personnel within the organization used in the personnel system as information on the recruiters registered in the recruiter database. This makes it possible to construct a recruiter task execution model that can execute tasks based on the attributes or characteristics of personnel working in the organization.

[0047] The information provided by recruiters includes, for example, information created or acquired by recruiters in the course of recruiting activities (for example, videos explaining the organization, scripts for information sessions, audio information from recruiters at information sessions, meetings or interviews, items sought in meetings or interviews, questions asked in interviews, requirements verbally explained by recruiters, verbal responses from job seekers, etc.). The recruiter agent construction unit 112 acquires the provided information by, for example, accepting input of provided information from recruiter terminal 20, uploading, designation of a storage location on the network, etc. The recruiter agent construction unit 112 may provide incentives to recruiters for inputting provided information, such as the granting of special benefits that can be used on the platform.

[0048] "Public information" is information that is publicly available on websites accessible via the Internet. The types of websites from which public information about employers is obtained are not limited as long as they contain at least information about the employer (organization), and include official websites, unofficial websites, and other websites.

[0049] An official website is a website set up by the organization associated with the recruiter, and examples of such websites include corporate websites, investor relations websites, commercial sites (product sites or service sites), brand sites, promotion sites, owned media sites, campaign sites, e-commerce sites, sites for members, recruitment sites, landing pages, etc. An unofficial website is a website set up by an organization (external institution) different from the organization associated with the recruiter, and examples of such websites include industry information introduction sites, news sites, economic information sites, commercial database sites, etc. Unofficial websites include both free and paid viewing pages. The recruiter's public information may include, for example, interviews with employees of the organization.

[0050] The recruiter agent construction unit 112 may automatically collect public information by keyword searching the Internet using organization names and the like, crawling or web scraping processing, etc. The recruiter agent construction unit 112 may also accept input of access information (e.g., URL, IP address, etc.) indicating the location of the public information on the network from the recruiter terminal 20, and acquire the public information based on the input access information.

[0051] The recruiter task execution model is, for example, a learning model included in the artificial intelligence unit 120 that has been machine-learned so as to be able to take a recruiter task as input and output the execution result of the recruiter task. The recruiter task execution model is, for example, a learning model that has been machine-learned using recruiter learning data and data required for executing the recruiter task (for example, a combination of input examples and output examples of the recruiter task) as training data.

[0052] The training data for the recruiter task execution model may further include statistical information on job seekers in the platform provided by the information processing system 1 (for example, the number of applications to job openings by industry, job type, etc.).

[0053] The employer task execution model may include a large-scale language model including a generation AI. When the employer task execution model is a large-scale language model, the employer agent construction unit 112 may create, for each employer, from the employer learning data, at least one employer memory that the employer task execution model refers to when generating output. This makes it possible to construct an employer task execution model using an existing large-scale language model. It also makes it easier to update the employer task execution model.

[0054] Even when the recruiter task execution model is an AI agent, the recruiter agent construction unit 112 may create, for each recruiter, from the recruiter learning data, at least one recruiter memory that the recruiter task execution model refers to when generating output. This makes it possible to use a pre-constructed AI agent to construct a recruiter task execution model that is suited to the characteristics of each individual recruiter.

[0055] The employer memory is information that represents the characteristics or attributes of an employer, and is a parameter that customizes (determines the characteristics of) the employer task execution model. The employer task execution model executes employer tasks as an employer agent that has learned the characteristics represented in the employer memory. The employer memory is registered, for example, in a employer database, linked to employer registration information.

[0056] The recruiter memory may include at least information about the culture of the organization performing the recruitment, which allows for the construction of a recruiter task execution model capable of responding in a way that takes into account the matching with the recruiter's organizational culture.

[0057] The employer memory includes, for example, information regarding the degree of growth of the organization, the fields in which it is growing, the policies of the organization, the advantages of the business, the type of personnel the organization is looking for (for example, personality and characteristics), etc. Specific examples of the contents of the employer memory include, for example, "a rapidly growing HR company," "a venture company," and "wanting to hire honest and diligent people." The employer memory may be information written in a natural language, or may be code information such as a computer language, a programming language, or a machine code, as long as it can be referenced by the employer task execution model. The number of employer memories provided for one employer task execution model (one employer) may be, for example, on the order of 1,000.

[0058] The recruiter agent construction unit 112, for example, inputs the recruiter learning data into the memory creation model and causes the memory creation model to output a recruiter memory. The memory creation model is, for example, a learning model included in the artificial intelligence unit 120 that has been machine-learned so as to be able to take the recruiter learning data as input and output a recruiter memory. The memory creation model is, for example, a learning model that has been machine-learned using the recruiter learning data and the corresponding recruiter memory as training data.

[0059] The memory creation model may be a large-scale language model including a generation AI. In this case, the recruiter agent construction unit 112 inputs recruiter learning data, inputs a prompt including an instruction to output a recruiter memory corresponding to the recruiter learning data to the memory creation model, and causes the memory creation model to output the recruiter memory. The recruiter agent construction unit 112 may generate a prompt that gives an instruction to the memory creation model to output a recruiter memory corresponding to the recruiter learning data, and inputs the prompt to the memory creation model. Furthermore, the recruiter agent construction unit 112 may input, in addition to the recruiter memory generation / output instruction and the recruiter learning data, a prompt into which, for example, one or more samples of recruiter learning data and one or more corresponding samples of recruiter memory are inserted as examples, samples, or teacher data of input and output pairs to the memory creation model.

[0060] The recruiter agent construction unit 112 may cause the memory generation model, which is a large-scale language model, to output a recruiter memory by inputting input information including an input to the recruiter task execution model, an output of the recruiter task execution model in response to the input, or a combination of these, and an instruction to create a recruiter memory representing the characteristics or attributes of recruiters based on the input information, into the memory generation model, which is a large-scale language model. This makes it possible to add or update the recruiter memory based on the execution results of tasks in the recruiter task execution model, thereby improving the execution accuracy of tasks in the recruiter task execution model.

[0061] The recruiter agent construction unit 112 may present the recruiter with questions for creating the recruiter memory, and create the recruiter memory based on the answers entered by the recruiter to the questions. For example, the recruiter agent construction unit 112 may input the answers of the recruiter to a memory generation model, and cause the memory generation model to output the recruiter memory. Furthermore, the recruiter agent construction unit 112 may use the results of creating the recruiter memory, which is a subtask executed by a recruiter task execution model, which is an AI agent, when executing a recruiter task such as extracting job seekers to recommend to recruiters.

[0062] The recruiter agent construction unit 112 may create a recruiter memory representing the characteristics or attributes of the recruiter from the recruiter learning data for each recruiter, and may construct a recruiter task execution model by machine learning using the recruiter memory. This allows a recruiter task execution model dedicated to each individual recruiter based on the recruiter memory. Note that the recruiter task execution model constructed using the recruiter memory may be a dedicated model machine-learned to take a task as input and output the execution result of the task, or may be a large-scale language model (specifically, for example, a learning model in which a general-purpose large-scale language model is subjected to reinforcement learning, fine tuning, etc. using the recruiter learning data and the recruiter memory).

[0063] The recruiter agent construction unit 112 may construct a recruiter task execution model by machine learning using recruiter labels created by the recruiter label creation unit 113 described below and edited by the recruiter. This allows the recruiter to customize the recruiter task execution model and modify its characteristics, thereby improving the accuracy of matching between recruiters and job seekers using the recruiter task execution model.

[0064] The recruiter agent construction unit 112 may construct a recruiter task execution model by machine learning that further uses recorded data of interviews between a human resources intermediary (headhunter) and a recruiter or job seeker. This allows the recruiter task execution model to learn know-how about interviews, matching, etc., possessed by the human resources intermediary, thereby improving the accuracy of matching between recruiters and job seekers using the recruiter task execution model.

[0065] An interview between a recruitment agency and a job seeker includes, for example, an exchange of information about the type of employee they are looking for. An interview between a recruitment agency and a job seeker may include, in addition to an interview about employment, a consultation about the job seeker's career, life, comparison with other job seekers (relative standing), etc. In addition, "recorded data of the interview" includes, for example, video, audio data, text data, etc.

[0066] The recruiter agent construction unit 112 may prepare learning data for each department of the recruiter, each job type of the recruiter, or each job handled by the recruiter, and construct a recruiter task execution model for each department, job type, or job. This allows a recruiter task execution model to be constructed for each department, job type, or job in a large-scale organization, enabling the recruiter task execution model to execute more detailed recruiter tasks.

[0067] <Recruiter Label Creation Department 113> The recruiter label creation unit 113 is configured to cause the memory generation model to output at least one recruiter label by inputting a recruiter memory and an instruction to generate a recruiter label expressing the contents of the recruiter memory in natural language to the memory generation model.

[0068] A recruiter label expresses the contents of each recruiter memory in natural language (words or sentences) that the recruiter can understand. Note that multiple recruiter labels may be generated from one recruiter memory. The recruiter label is registered, for example, in a recruiter database in association with recruiter registration information.

[0069] The recruiter label creation unit 113, for example, takes an employer memory as input, inputs a prompt including an instruction to output an employer label corresponding to the employer memory to the memory creation model, and causes the memory creation model to output the employer label. The recruiter label creation unit 113 may generate a prompt that gives an instruction to the memory creation model to output an employer label corresponding to the employer memory, and inputs the prompt to the memory creation model. Furthermore, the recruiter label creation unit 113 may input, in addition to the employer memory and the instruction to generate and output the employer label, a prompt into which, for example, one or more employer memory samples and one or more corresponding employer label samples are inserted as examples, samples, or training data of input and output pairs to the memory creation model.

[0070] The recruiter label creation unit 113 may obtain a recruiter label created by a recruiter task execution model, which is an AI agent. The AI ​​agent may, for example, execute generation of a recruiter label as a subtask when executing a recruiter task such as extracting job seekers to recommend to recruiters. The AI ​​agent may also execute generation of a recruiter memory and a recruiter label as a subtask. The recruiter label creation unit 113 may also cause the recruiter task execution model to create a prompt for the memory creation model to create a recruiter label, and input the prompt to the memory creation model.

[0071] The recruiter label creation unit 113 presents all or part of the multiple recruiter labels generated for each recruiter to the corresponding recruiter (recruiter terminal 20). This allows the recruiter to check the components (features or attributes) of his or her own recruiter task execution model (AI agent).

[0072] For example, the recruiter label creation unit 113 may determine whether a plurality of recruiter labels have positive content or negative content, and present only the recruiter labels determined to have positive content to the recruiter terminal 20. The recruiter label creation unit 113 may cause a large-scale language model including a generation AI to determine whether the recruiter label has positive content or negative content. The recruiter label creation unit 113 may also use the result of determining whether the recruiter label has positive content or negative content, which is a subtask executed by a recruiter task execution model, which is an AI agent, when executing a recruiter task such as extracting job seekers to be recommended to recruiters.

[0073] Furthermore, the recruiter label creation unit 113 may present to the recruiter terminal 20, among the multiple recruiter labels, recruiter labels that have a high degree of relevance with items that are emphasized in matching with job seekers, or a high degree of relevance with items that recruiters want to appeal to job seekers (for example, recruiter labels with a high degree of relevance within a specified rank order). These items may be extracted from recruiter learning data (recruiter registration information), for example, or may be input by the recruiter. Furthermore, the recruiter label creation unit 113 may present to the recruiter terminal 20 a recruiter label summary that summarizes combinations of several recruiter labels.

[0074] The recruiter label creation unit 113 may accept edits of the recruiter label from the corresponding recruiter (recruiter terminal 20). This makes it possible to customize the recruiter task execution model based on the edits of the recruiter label by the recruiter. Note that editing also includes the deletion of the recruiter label.

[0075] <Job Seeker Agent Construction Division 114> The job seeker agent construction unit 114 is configured to construct, for each job seeker registered in the job seeker database, a job seeker task execution model capable of executing tasks handling information about the job seeker, through machine learning using job seeker learning data related to the job seeker.

[0076] The job seeker task execution model may be an output device that outputs the execution results of an input task, or may be a so-called AI agent (job seeker agent) that autonomously executes subtasks such as data collection and conversation required to execute the task or goal input by the job seeker on behalf of the job seeker, using natural language processing, a processing model constructed by machine learning, a general-purpose large-scale language model, etc. Conversation with the job seeker task execution model is conducted, for example, in the form of a chat using voice or text.

[0077] The task executed by the job seeker task execution model (hereinafter, "job seeker task") may include, for example, an answer to a question about a job seeker to which the job seeker task execution model corresponds (the source of the job seeker learning data). In other words, the job seeker task execution model may be capable of executing an answer to a question about a job seeker as a job seeker task. The job seeker task execution model may also be capable of executing a question to an employer or another AI agent (employer task execution model or platform task execution model) as a job seeker task or a subtask thereof. This enables the job seeker task execution model to have a conversation or interview with an employer or another AI agent. As a result, for example, the job seeker task execution model can conduct a simulated interview with a popular job seeker who is difficult to contact.

[0078] "Questions about job seekers" include, for example, questions about desired conditions (industry, job type, duties, annual salary, working conditions, etc.), past experience (industry, job type, duties, position, skills, qualifications, etc.), selling points, career plans, personality, values, etc. In addition, questions that the job seeker task execution model itself asks employers or other AI agents include, for example, questions to confirm the suitability (degree of match) of the job seeker's desired conditions, personality, values, etc. with the person asking the questions (employer).

[0079] The job seeker task may include, for example, a correction or a suggestion for a correction of a sentence related to the work history of the job seeker to which the job seeker task execution model corresponds (the source of the job seeker learning data). In other words, the job seeker task execution model may be capable of executing a correction or a suggestion for a correction of a sentence related to the work history of the job seeker as a job seeker task.

[0080] The job seeker task may include output of information about the job seeker designated by a task instructor (e.g., the task execution unit 118 described below). The job seeker task may also be information indicating a specific objective (goal, purpose, etc.) regarding the job seeker designated by (e.g., the task execution unit 118 described below). Such a job seeker task may be, for example, an instruction such as "present recommended job offers to the job seeker" or information indicating an objective such as "I would like to know recommended job offers," and the information output by the job seeker task execution model may include, for example, information about job offers recommended to the job seeker.

[0081] An "offering recommended to a job seeker" is, for example, an offer from an employer that is highly matched to the job seeker, an offer that is estimated to have a high probability of passing the screening, or an offer that is estimated to have a high probability of closing. The job seeker task execution model may determine an offer to be recommended to a job seeker by, for example, comparing information on the job to be judged, the employer memory of the employer who is hiring the job, and information on the job seeker that has already been learned (for example, a job seeker memory, etc.) (for example, converting each piece of information into a vector and judging the similarity based on the cosine similarity between the vectors, etc.).

[0082] The job seeker learning data used by the job seeker agent construction unit 114 includes information about the job seeker registered in the job seeker database (registered information), information provided by the job seeker (provided information), or public information about the job seeker on the network.

[0083] The "job seeker database" is a database used together with the employer database in a platform provided by the information processing system 1. In other words, the job seeker learning data used by the job seeker agent construction unit 114 may include information on job seekers registered in a job seeker database in one platform.

[0084] The registration information of a job seeker (hereinafter referred to as "job seeker registration information") includes, for example, the job seeker's resume, work history, and other profile information. A "resume" is a document that mainly describes the job seeker's profile, current situation, educational background, employment history, desired working conditions, etc. A "work history," also known as a résumé, is a document in which a job seeker conveys to an employer the job seeker's work history, experience, skills, qualifications, etc. related to his or her past work. The job seeker registration information may also include the job seeker's desired industry, job type, and other conditions.

[0085] The job seeker registration information may also include a history of the job seeker's actions (e.g., viewing a job offer, applying for a job offer, opening a scouting email, replying to a scouting email, conducting an interview, closing a job offer, etc.). Furthermore, the job seeker registration information may also include a record of the content of an interview or meeting with a recruiter.

[0086] In this way, the job seeker learning data may include at least one of the job offers that the job seeker viewed or applied for, and the scout emails that the job seeker opened or replied to, as information on the job seeker registered in the job seeker database. This makes it possible to build a job seeker task execution model that can execute tasks based on the job seeker's behavioral history regarding job seeking.

[0087] The information provided by the job seeker includes, for example, the results of tests such as aptitude tests and personality tests, and documents in which the job seeker describes his or her own values. The job seeker agent construction unit 114 acquires the provided information by, for example, accepting the input of the provided information from the job seeker terminal 30, uploading the information, or designating a storage location on the network. The job seeker agent construction unit 114 may provide incentives to the job seeker in return for inputting the provided information, such as granting special benefits that can be used on the platform.

[0088] The type of website from which publicly available information about a job seeker is obtained is not limited as long as it contains at least some information about the job seeker, and may include the job seeker's own blog site or SNS site, a third-party website that contains information about the job seeker (e.g., the contents of an interview), etc.

[0089] The job seeker agent construction unit 114 may automatically collect public information by keyword search using the job seeker's name or the like on the Internet, crawling or web scraping processing, etc. The job seeker agent construction unit 114 may also accept input of access information indicating the location of the public information on the network from the job seeker terminal 30, and acquire the public information based on the input access information.

[0090] The job seeker task execution model is, for example, a learning model included in the artificial intelligence unit 120 that has been machine-learned so as to be able to input a job seeker task and output the execution result of the job seeker task. The job seeker task execution model is, for example, a learning model that has been machine-learned using job seeker learning data and data required for executing a job seeker task (for example, a combination of input examples and output examples of a job seeker task) as training data.

[0091] The job seeker task execution model may include a large-scale language model including a generation AI. When the job seeker task execution model is a large-scale language model, the job seeker agent construction unit 114 may create, for each job seeker, from the job seeker learning data, at least one job seeker memory that the job seeker task execution model refers to when generating an output. This allows the job seeker task execution model to be constructed using an existing large-scale language model (for example, a large-scale language model common to the employer task execution model). This also makes it easier to update the job seeker task execution model.

[0092] Even when the job seeker task execution model is an AI agent, the job seeker agent construction unit 114 may create, for each job seeker, at least one job seeker memory, from the job seeker learning data, which the job seeker task execution model refers to when generating an output. This makes it possible to use a pre-constructed AI agent to construct a job seeker task execution model that is suited to the individuality of each job seeker.

[0093] The job seeker memory is information that represents the characteristics or attributes of a job seeker, and is a parameter that customizes (determines the characteristics of) the job seeker task execution model. The job seeker task execution model executes a job seeker task as a job seeker agent that has learned the characteristics represented in the job seeker memory. The job seeker memory is registered in a job seeker database, for example, in association with job seeker registration information.

[0094] The job seeker memory may include at least information about the personality of the job seeker. This makes it possible to construct a job seeker task execution model that is capable of responding in consideration of matching with the personality of the job seeker. In particular, if the employer memory includes information about the culture of an organization as described above, it becomes possible to match the culture of the job seeker's organization with the personality of the job seeker by combining (dialogue with) the employer task execution model and the job seeker task execution model.

[0095] The job seeker memory includes, for example, information about the job seeker's values, values ​​the job seeker does not like, the job seeker's life goals, essential working conditions, preferred working conditions, etc. Specific examples of the contents of the job seeker memory include, for example, "valuing personal charm," "valuing work-life balance," and "wanting to make an impact on the world." The job seeker memory may be information written in a natural language, or may be code information such as a computer language, a programming language, or a machine language, as long as the job seeker task execution model can refer to it. The number of job seeker memories prepared for one job seeker task execution model (one job seeker) may be, for example, on the order of 1,000.

[0096] The job seeker agent construction unit 114, for example, inputs job seeker learning data into a memory creation model and causes the memory creation model to output a job seeker memory. The memory creation model is a learning model similar to that used by the employer agent construction unit 112. When the memory creation model is a large-scale language model including a generation AI, the job seeker agent construction unit 114 inputs the job seeker learning data, inputs a prompt including an instruction to output a job seeker memory corresponding to the job seeker learning data into the memory creation model, and causes the memory creation model to output the job seeker memory. A specific procedure for outputting the job seeker memory (an example of a prompt) is the same as the procedure for outputting the employer memory.

[0097] The job seeker agent construction unit 114 may cause the memory generation model, which is a large-scale language model, to output a job seeker memory by inputting input information including an input to the job seeker task execution model, an output of the job seeker task execution model in response to the input, or a combination of these, and an instruction to create a previous job seeker memory representing the characteristics or attributes of a job seeker based on the input information, to the memory generation model, which is a large-scale language model. This makes it possible to add or update the job seeker memory based on the execution results of the tasks of the job seeker task execution model, thereby improving the execution accuracy of the tasks of the job seeker task execution model.

[0098] The job seeker agent construction unit 114 may present the job seeker with questions for creating a job seeker memory, and create the job seeker memory based on the answers entered by the job seeker to the questions. For example, the job seeker agent construction unit 114 may input the answers of the job seeker to a memory generation model and cause the memory generation model to output the job seeker memory. In addition, the job seeker agent construction unit 114 may use the results of the creation of the job seeker memory, which is a subtask executed by a job seeker task execution model, which is an AI agent, when executing a job seeker task such as extracting job offers to recommend to a job seeker.

[0099] The job seeker agent construction unit 114 may create a job seeker memory representing the characteristics or attributes of the job seeker from the job seeker learning data for each job seeker, and may construct a job seeker task execution model by machine learning using the job seeker memory. This makes it possible to construct a job seeker task execution model dedicated to each individual job seeker based on the job seeker memory. Note that the job seeker task execution model constructed using the job seeker memory may be a dedicated model that has been machine-learned to take a task as input and output the execution result of the task, or may be a large-scale language model (specifically, for example, a learning model in which a general-purpose large-scale language model is subjected to reinforcement learning, fine tuning, etc. using the job seeker learning data and the job seeker memory).

[0100] The job seeker agent construction unit 114 may construct a job seeker task execution model by machine learning using a job seeker label created by the job seeker label creation unit 115 described below and edited by the job seeker. This allows the job seeker to customize the job seeker task execution model and modify its characteristics, thereby improving the accuracy of matching between employers and job seekers using the job seeker task execution model.

[0101] The job seeker agent construction unit 114 may construct a job seeker task execution model by machine learning that further uses recorded data of interviews between the human resources agency and the recruiter or job seeker. This allows the job seeker task execution model to learn questions from the human resources agency and answers to those questions, thereby improving the accuracy of matching between recruiters and job seekers using the job seeker task execution model.

[0102] <Job Seeker Label Creation Division 115> The job applicant label creation unit 115 is configured to cause the memory generation model to output at least one job applicant label by inputting a job applicant memory and an instruction to generate a job applicant label expressing the contents of the job applicant memory in natural language to the memory generation model.

[0103] A job applicant label expresses the contents of each job applicant memory in a natural language (words or sentences) that the job applicant can understand. Note that multiple job applicant labels may be generated from one job applicant memory. The job applicant label is registered, for example, in a job applicant database in association with the job applicant registration information.

[0104] The job applicant label creation unit 115, for example, inputs a job applicant memory, inputs a prompt including an instruction to output a job applicant label corresponding to the job applicant memory to the memory creation model, and causes the memory creation model to output the job applicant label. A specific procedure for outputting the job applicant label (an example of a prompt) is the same as the procedure for outputting the employer label.

[0105] The job seeker label creation unit 115 may obtain a job seeker label created by a job seeker task execution model, which is an AI agent. The AI ​​agent may execute generation of a job seeker label as a subtask when executing a job seeker task, such as extracting job offers to be recommended to a job seeker. The AI ​​agent may also execute generation of a job seeker memory and a job seeker label as a subtask. The job seeker label creation unit 115 may also cause the job seeker task execution model to create a prompt for the memory creation model to create a job seeker label, and input the prompt to the memory creation model.

[0106] The job seeker label creation unit 115 presents all or a part of the multiple job seeker labels generated for each job seeker to the corresponding job seeker (job seeker terminal 30). This enables the job seeker to check the components (features or attributes) of his or her own job seeker task execution model (AI agent).

[0107] For example, the job seeker label creation unit 115 may determine whether a plurality of job seeker labels have positive content or negative content, and present only the job seeker labels determined to have positive content to the job seeker terminal 30. The job seeker label creation unit 115 may cause a large-scale language model including a generation AI to determine whether a job seeker label has positive content or negative content. The job seeker label creation unit 115 may also use the result of determining whether a job seeker label has positive content or negative content, which is a subtask executed by a job seeker task execution model, which is an AI agent, when executing a job seeker task such as extracting job offers to be recommended to a job seeker.

[0108] Furthermore, the job seeker label creation unit 115 may present to the job seeker terminal 30, among the multiple job seeker labels, job seeker labels that have a high degree of relevance with items that are important in matching with employers, or a high degree of relevance with items that the job seeker wants to appeal to employers (for example, job seeker labels with a high degree of relevance within a specified order). These items may be extracted from the job seeker learning data (job seeker registration information), for example, or may be input by the job seeker. Furthermore, the job seeker label creation unit 115 may present to the job seeker terminal 30 a job seeker label summary that summarizes a combination of several job seeker labels.

[0109] The job seeker label creation unit 115 may accept editing of the job seeker label from the corresponding job seeker (job seeker terminal 30). This makes it possible to customize the job seeker task execution model based on editing of the job seeker label by the job seeker. Note that editing also includes deleting the job seeker label.

[0110] <Question Registration Section 116> The question registration unit 116 is configured to register advance questions for employers received from job seekers. Specifically, the question registration unit 116 receives the selection of an employer and the input of an advance question for the employer from the job seeker terminal 30. The question registration unit 116 links the job seeker who created the advance question, the selected employer (the subject of the advance question), and the advance question, and registers them, for example, in a job seeker database. Note that one advance question (a common question for multiple employers) may be registered for multiple employers.

[0111] <Platform Agent Construction Division 117> The platform agent construction unit 117 is configured to construct a platform task execution model capable of executing tasks handling information of any recruiter or any job seeker, through machine learning using the recruiter learning data and the job seeker learning data.

[0112] The platform task execution model is constructed for each platform provided by the information processing system 1. The platform task execution model may be an output device that outputs the execution result of an input task, or may be a so-called AI agent (platform agent) that autonomously executes tasks such as data collection and conversation.

[0113] Tasks executed by the platform task execution model (hereinafter, platform tasks) include recruiter tasks executed by the recruiter task execution model and job seeker tasks executed by the job seeker task execution model. Platform tasks may also include tasks that are not targeted at individual recruiters or job seekers, but are targeted at multiple recruiters or job seekers registered on the platform (for example, tasks related to the compilation of statistical data on recruiters or job seekers).

[0114] The platform task execution model may be capable of answering questions about recruiters or job seekers as a platform task. In other words, the platform task execution model may be capable of conversing or interviewing with recruiters, job seekers, or other AI agents.

[0115] When the platform task execution model is an AI agent, the platform task execution model may generate subtasks based on the goal or task input input by the recruiter or job seeker, and select an optimal model (e.g., a job seeker task execution model or a recruiter task execution model) for each subtask to execute the process. The platform task execution model may generate a prompt required for executing the subtask, and input the generated prompt to the selected model to cause the model to execute the process.

[0116] The employer learning data used by the platform agent construction unit 117 may include information on a plurality of employers registered in the employer database used by the employer agent construction unit 112. Furthermore, the job seeker learning data used by the platform agent construction unit 117 may include information on a plurality of job seekers registered in the job seeker database used by the job seeker agent construction unit 114. As a result, the employer task execution model and the job seeker task execution model, and the platform task execution model are constructed based on common employer learning data and job seeker learning data, and therefore, by having the employer task execution model or the job seeker task execution model cooperate with the platform task execution model, it becomes possible to efficiently match employers and job seekers based on information on the entire platform.

[0117] The employer learning data used by the platform agent construction unit 117 may include information on employers for whom an employer task execution model has not been constructed. Similarly, the job seeker learning data used by the platform agent construction unit 117 may include information on job seekers for whom an employee task execution model has not been constructed. This makes it possible to provide the user (employer or job seeker) with the execution results of platform tasks based on information on employers for whom an employer task execution model has not been constructed, or on information on job seekers for whom an employee task execution model has not been constructed.

[0118] The platform task execution model is, for example, a learning model included in the artificial intelligence unit 120 that has been machine-learned to be able to take a platform task as an input and output the execution result of the platform task. The platform task execution model is, for example, a learning model that has been machine-learned using recruiter learning data, job seeker learning data, and data required for executing a platform task (for example, a combination of input examples and output examples of a platform task) as training data.

[0119] The training data of the platform task execution model includes employer learning data including information on multiple employers and job seeker learning data including information on multiple job seekers, and therefore includes statistical information on employers and statistical information on job seekers in the platform provided by the information processing system 1.

[0120] Furthermore, the training data for the platform task execution model may include external information relating to the employment of workers that is publicly available outside the platform. Examples of external information include labor statistics and policies of public institutions such as the government. This enables the platform task execution model to output (answer) taking into consideration the correction of matching bias and responses to policies (e.g., an increase in the number of female executives).

[0121] The platform task execution model may be a large-scale language model that is a generative AI. When the platform task execution model is a large-scale language model, the platform agent construction unit 117 may create at least one platform memory that the platform task execution model refers to when generating an output from the employer learning data and the job seeker learning data.

[0122] The platform memory is information that represents the statistical characteristics of the entire platform, and is a parameter that customizes (determines the characteristics of) the platform task execution model. The platform memory includes information such as statistics on employers, such as the number of job openings, the number of applications, and the number of successful jobs (success rate) for each job attribute (industry type, job type, job function, requirements, etc.), and statistics on job seekers, such as the number of scout emails received, the number of replies to scout emails (reply rate), the number of job applications, and the number of successful job offers (success rate) for each job seeker attribute (career history, experience, skills, age, etc.).

[0123] The platform agent construction unit 117, for example, inputs the employer learning data and the job seeker learning data to the memory creation model and causes the memory creation model to output a platform memory. The memory creation model is a learning model similar to that used by the employer agent construction unit 112. When the memory creation model is a large-scale language model, the platform agent construction unit 117 inputs the employer learning data and the job seeker learning data, inputs a prompt including an instruction to output a platform memory corresponding to the employer learning data and the job seeker learning data to the memory creation model, and causes the memory creation model to output the platform memory. A specific procedure for outputting the platform memory (an example of a prompt) is the same as the procedure for outputting the employer memory.

[0124] The platform agent construction unit 117 may create a platform memory representing the characteristics of the entire platform from the employer learning data and the job seeker learning data, and may construct a platform task execution model by machine learning using the platform memory. Note that the platform task execution model constructed using the platform memory may be a dedicated model machine-learned to input a task and output the execution result of the task, or may be a large-scale language model (specifically, for example, a learning model in which a general-purpose large-scale language model is fine-tuned by the platform memory, etc.).

[0125] <Task Execution Unit 118> The task execution unit 118 is configured to cause the recruiter task execution model or the job seeker task execution model to execute a task (recruiter task or job seeker task) based on an execution instruction for the recruiter task execution model or the job seeker task execution model.

[0126] The execution instruction for the recruiter task or the job seeker task may be information indicating the task (e.g., a question to the AI ​​agent) that is directly input from the recruiter terminal 20 or the job seeker terminal 30, or may be information (e.g., a prompt) generated by the task executing unit 118 based on the content of the task input from the recruiter terminal 20 or the job seeker terminal 30. Moreover, the execution instruction for the recruiter task or the job seeker task may be information (e.g., an instruction to execute a periodic task) that is automatically generated by the task executing unit 118, not based on an input (instruction) from the recruiter terminal 20 or the job seeker terminal 30.

[0127] Furthermore, the recruiter task or job seeker task may be generated autonomously by the recruiter task execution model or job seeker task execution model (AI agent) in response to a target (purpose, goal, etc.) input by a user. In this case, prompts, which are instructions to the recruiter task execution model or job seeker task execution model, are also generated by the recruiter task execution model or job seeker task execution model itself.

[0128] For example, if the recruiter task execution model is an AI agent, the task execution unit 118 inputs the task or goal included in the execution instruction input by the recruiter to the AI ​​agent constructed by learning the recruiter learning data. The AI ​​agent generates the required subtask based on the input. The AI ​​agent may generate a prompt for executing the subtask and input the prompt to an external large-scale language model to obtain the execution result of the subtask from the large-scale language model, or may execute the generated prompt by itself. In addition, the AI ​​agent may create input information for causing an external learning model that executes a specific process to execute a specific process, and input the input information to the learning model to obtain the execution result of the subtask from the learning model. The AI ​​agent presents information corresponding to the input task or goal to the recruiter.

[0129] If the employer task execution model and the job seeker task execution model are large-scale language models, the task execution unit 118 may input the employer memory or the job seeker memory together with the task (employer task or job seeker task) as reference information to the employer task execution model or the job seeker task execution model that is the target of the task execution instruction, and cause the employer task execution model or the job seeker task execution model to execute the task.

[0130] Specifically, the task execution unit 118 receives a task and an employer memory or a job seeker memory as input, inputs a prompt including an instruction to output an execution result of the task (e.g., an answer to a question) to the employer task execution model or the job seeker task execution model, and causes the execution result to be output to the employer task execution model or the job seeker task execution model. The task execution unit 118 may generate a prompt that gives an instruction to the employer task execution model or the job seeker task execution model to output the execution result of the task, and inputs the prompt to the employer task execution model or the job seeker task execution model. Furthermore, the task execution unit 118 may input to the employer task execution model or the job seeker task execution model a prompt in which, for example, one or more sample combinations of a task and an employer memory or a job seeker memory and one or more corresponding sample execution results are inserted as examples, samples, or teacher data of input and output pairs, in addition to the execution instruction for the task, the task, and the employer memory or the job seeker memory.

[0131] If the employer task or job seeker task is a response to a question about the employer or job seeker, the task execution unit 118 may accept a question for the employer task execution model or the job seeker task execution model as an instruction to execute the task, and cause the employer task execution model or the job seeker task execution model to output the response to the question.

[0132] The task execution unit 118 may request a job recruiter corresponding to the job recruiter task execution model or a job seeker corresponding to the job seeker task execution model to have a conversation with the user, depending on the content of the conversation between the user (job recruiter or job seeker) and the job recruiter task execution model or the job seeker task execution model. This makes it possible to switch the user's conversation partner from the job recruiter task execution model or the job seeker task execution model (AI agent) to the person in question when a question is asked that the job recruiter task execution model or the job seeker task execution model cannot answer, or in a situation where a direct conversation with the person in question should be conducted, etc.

[0133] For example, when the employer task execution model or the job seeker task execution model cannot generate an answer (when it issues an answer indicating that an answer cannot be generated), the task execution unit 118 outputs a notification to the employer terminal 20 of the employer corresponding to the employer task execution model or the job seeker terminal 30 of the job seeker corresponding to the job seeker task execution model, requesting them to join a conversation with the user. When the employer or job seeker accepts this request, communication is initiated between the employer terminal 20 or the job seeker terminal 30 and the user's terminal for conversation (for example, video conference, voice call, etc.).

[0134] Furthermore, the task execution unit 118 may request the recruiter or the job seeker to join the conversation if a specific important keyword (including similar keywords) set in advance is included in either the user's question or the answer of the recruiter task execution model or the job seeker task execution model. Important keywords include, for example, keywords related to applying for a job, hiring procedures, etc.

[0135] In addition, if the employer or job seeker does not accept a request to converse with the user, or if there is no response from the employer or job seeker, the task execution unit 118 may present a question from the user to the employer or job seeker and accept a response from the user afterwards.

[0136] The task execution unit 118 may perform a job seeker presentation process of extracting a job seeker recommended to a job seeker corresponding to the job seeker task execution model by having the job seeker task execution model execute a task, or a job offer presentation process of extracting a job offer recommended to a job seeker corresponding to the job seeker task execution model by having the job seeker task execution model execute a task. This makes it possible to present recommended job seekers based on information about the job seeker (job seeker learning data), or present recommended job offers based on information about the job seeker (job seeker learning data), thereby improving the efficiency of matching between job seekers and job seekers. Note that the recommended job offers may be job offers offered by a job seeker who is recommended to the job seeker. In other words, the task execution unit 118 may extract a job seeker recommended to the job seeker using the job seeker task execution model.

[0137] The task execution unit 118 causes at least one recommended job seeker extracted by the job seeker presentation process to be displayed on the recruiter terminal 20. In addition, the task execution unit 118 causes at least one recommended job seeker extracted by the job seeker presentation process to be displayed on the job seeker terminal 30.

[0138] If the recruiter task execution model is capable of answering questions about the recruiter and the job seeker task execution model is capable of answering questions about the job seeker, the task execution unit 118 may have the recruiter task execution model and the job seeker task execution model have a conversation and determine a job seeker to be recommended to the recruiter or a job offer to be recommended to the job seeker based on the content of the conversation. This makes it possible to present recommended job seekers or recommended job offers through a simulated interview between the recruiter and the job seeker (an interview between AI agents).

[0139] A conversation between the employer task execution model and the job seeker task execution model is started, for example, by the task execution unit 118 having one model ask a question to the other model. For example, when determining a recommended job seeker, the task execution unit 118 has the employer task execution model ask a question to the job seeker task execution model. Also, for example, when determining a recommended job offer, the task execution unit 118 has the job seeker task execution model ask a question to the employer task execution model.

[0140] In the case where the recruiter task execution model and the job seeker task execution model are each an AI agent, the task execution unit 118 may receive, for example, a task execution instruction from the recruiter to "find (list) matching job seekers," and input the task execution instruction to the recruiter task execution model corresponding to the recruiter as a goal. The recruiter task execution model sequentially asks questions to multiple job seeker task execution models (opponent AI agents) as subtasks for the input goal, extracts necessary job seeker information, vectorizes the information, calculates similarity, and determines job seekers with a high degree of match, and outputs or visualizes the results. Here, the recruiter task execution model and the job seeker task execution model each autonomously generate necessary subtasks (for example, questions to measure compatibility), and further generate and execute codes required to execute the subtasks. The subtasks include, for example, generating and presenting questions to the opponent AI agent required for matching, and analyzing answers from the AI ​​agent to the questions. The task execution unit 118 presents recommended job seekers determined by the recruiter task execution model to the user.

[0141] Furthermore, when the recruiter task execution model and the job seeker task execution model are each an AI agent, the task execution unit 118 may receive, for example, a task execution instruction from a job seeker to "find (list) matching recruiters or job offers," and input the task execution instruction as a goal to the job seeker task execution model corresponding to the job seeker. As a subtask of the input goal, the job seeker task execution model sequentially asks questions to multiple recruiter task execution models (opponent AI agents), extracts necessary recruiter information, vectorizes it, calculates similarity, and so on, to determine recruiters or job offers with a high degree of match, and outputs or visualizes the results. The task execution unit 118 presents the recommended recruiters or recommended jobs determined by the recruiter task execution model to the user.

[0142] In response to an instruction to execute a task "search for a matching job seeker", the task execution unit 118 may input a task "calculate the degree of match with each of a plurality of job seeker task execution models" to the employer task execution model, and cause the employer task execution model to calculate the degree of match with each job seeker. Furthermore, the task execution unit 118 may present to the user job seekers whose degree of match is equal to or higher than a predetermined level as recommended job seekers. The same applies to the presentation of recommended employers or recommended job offers.

[0143] The task execution unit 118, for example, records a conversation log in a job recruiter task execution model and a job seeker task execution model and analyzes the log to determine whether the job seeker corresponding to the job seeker task execution model is a recommended job seeker by the job recruiter corresponding to the job recruiter task execution model, or whether the job offer by the job recruiter corresponding to the job recruiter task execution model is a recommended job offer for the job seeker corresponding to the job seeker task execution model.

[0144] The task execution unit 118 may determine a recommended job seeker or a recommended job offer based on the number of times of agreement in the conversation log. Here, "agreement" may be interpreted as an affirmative answer or a positive response to a question. For example, if the number of times of agreement is equal to or greater than a predetermined number, it is determined that the person is a recommended job seeker or a recommended job offer. The task execution unit 118 may also cause the employer task execution model or the job seeker task execution model to ask a specific question, and determine a recommended job seeker or a recommended job offer based on the answer to the question. The specific question is, for example, a question that confirms the employer's willingness to hire (employment possibility) or the job seeker's willingness to apply (application possibility).

[0145] The task execution unit 118 may determine a job seeker to be recommended to a recruiter or a job offer to be recommended to a job seeker based on the similarity calculated from the feature amount of the recruiter learning data and the feature amount of the job seeker learning data. This makes it possible to extract recommended job seekers or recommended job offers based on quantitative judgment based on the information of the job seeker and the recruiter, respectively.

[0146] As the features of the learning data, vector data obtained by vectorizing individual information for each category included in the learning data (for example, industry, job type, duties, skills, qualifications, etc. included in the job posting or job seeker registration information) or the entire learning data is used. Vectorization is performed, for example, by natural language processing using morphological analysis or quantification using known techniques such as encoding. The similarity between the employer learning data and the job seeker learning data is represented by the difference between the features of the same category or the overall features (for example, vector distance). For example, the task execution unit 118 determines that a job offer or a job seeker of an employer whose similarity (for example, cosine similarity) between the employer learning data and the job seeker learning data is equal to or greater than a threshold value is a recommended job offer or a recommended job seeker.

[0147] The task execution unit 118 may determine a job seeker to be recommended to a recruiter or a job offer to be recommended to a job seeker based on the similarity calculated from the feature amount of the recruiter memory as the feature amount of the recruiter learning data and the feature amount of the job seeker memory as the feature amount of the job seeker learning data. This makes it possible to determine a recommended job seeker or a recommended job offer based on the features or attributes of the recruiter and the job seeker, thereby improving the accuracy of the determination.

[0148] The features of the employer memory and the features of the job seeker memory may be vectorized versions of individual employer memories and individual job seeker memories, respectively, or may be vectorized versions of all of a plurality of employer memories for one employer and all of a plurality of job seeker memories for one job seeker.

[0149] The task executing unit 118 may determine the number of combinations of the employer memory and the job seeker memory in which the difference between the individual feature quantities is small as the degree of similarity, or may determine the smallness of the difference between the feature quantities of the entire employer memory and the feature quantities of the entire job seeker memory as the degree of similarity. The task executing unit 118 determines the job offers or job seekers of the employer whose similarity is equal to or greater than a predetermined value as recommended job offers or recommended job seekers.

[0150] 5 is a diagram showing an image of matching between a recruiter and a job seeker using a recruiter task execution model and a job seeker task execution model. The task execution unit 118 estimates the degree of matching between a recruiter and a job seeker by a conversation between the constructed recruiter task execution model and the job seeker task execution model, or by comparing the registered recruiter label (recruiter memory) and the job seeker label (job seeker memory).

[0151] The task execution unit 118 may present the result of the conversation between the employer task execution model and the job seeker task execution model to at least one of the employer corresponding to the employer task execution model and the job seeker corresponding to the job seeker task execution model. This allows the employer or the job seeker to check the response of their agent. Furthermore, depending on the confirmation result, the employer or the job seeker can customize the employer task execution model or the job seeker task execution model (AI agent) by editing the employer label, the job seeker label, etc.

[0152] For example, the task execution unit 118 displays the employer task execution model, the job seeker task execution model, and the conversation log on the employer terminal 20 or the job seeker terminal 30 in response to a request from the employer or the job seeker. Note that if the task execution unit 118 receives a refusal of the presentation from the job seeker (preferably together with the reason for refusing the presentation), it may choose not to present the results of the conversation (log) to the employer.

[0153] The task execution unit 118 may present a conversation log between the user (job seeker) and the job recruiter task execution model to the job recruiter corresponding to the job recruiter task execution model. The task execution unit 118 may also present a conversation log between the user (job seeker) and the job recruiter task execution model to the job seeker corresponding to the job recruiter task execution model.

[0154] The task execution unit 118 may cause the employer task execution model corresponding to the employer specified by the job seeker to execute, as an employer task, a response to the advance question registered by the question registration unit 116. This allows the job seeker to obtain a response from the employer task execution model at any time.

[0155] For example, when the job seeker views information of the employer designated as the subject of the advance question (e.g., organizational information, job posting, etc.), the task execution unit 118 may cause the employer task execution model to execute responses to the advance question. In addition, the task execution unit 118 may cause the job seeker task execution model to execute responses to the advance question when the job seeker performs an action other than viewing the employer's information (e.g., logging in to a platform service, applying for a job, etc.).

[0156] The task execution unit 118 may input, for example, a goal input by a job seeker, such as "I want to revise my resume," to the job seeker task execution model, which is an AI agent. The job seeker task execution model generates and executes subtasks, such as obtaining text related to the job seeker's work history, such as a resume, registered by the job seeker, analyzing the text related to the work history, identifying parts to be revised, generating revision proposals, identifying missing information, generating questions to obtain the missing information, presenting questions to the job seeker, accepting answers, and generating revision proposals based on the answers. This allows the job seeker task execution model to revise the text related to the job seeker's work history while talking or interviewing the job seeker, thereby improving the quality of the job seeker's work history.

[0157] The task execution unit 118 causes the platform task execution model to execute the platform task based on an execution instruction for the platform task to the platform task execution model.

[0158] The platform task execution instruction may be information indicating the task that is directly input from the recruiter terminal 20 or the job seeker terminal 30, or may be information (e.g., a prompt) generated by the task executing unit 118 based on the content of the task input from the recruiter terminal 20 or the job seeker terminal 30. Furthermore, the platform task execution instruction may be information (e.g., an instruction to execute a periodic task) that is automatically generated by the task executing unit 118, regardless of input (instruction) from the recruiter terminal 20 or the job seeker terminal 30.

[0159] When the task execution unit 118 causes the employer task execution model or the job seeker task execution model to execute a task (employer task or job seeker task), it may also cause the platform task execution model to execute a task identical to the task or a task related to the task. This makes it possible to present to the user the execution result of the task by the platform task execution model in addition to the execution result of the task by the individual employer task execution model or the job seeker task execution model. As a result, the user can search for a job seeker or a job using the execution result based on a specific employer or a specific job seeker (e.g., an answer to a question) and the execution result based on the information of the entire platform (e.g., a statistical suggestion).

[0160] "Tasks related to employer tasks or job seeker tasks" include, for example, if the employer task or job seeker task is an answer to a question about employers or job seekers, the presentation of statistical information related to keywords included in the question. Also, "tasks related to employer tasks or job seeker tasks" include, for example, if the employer task or job seeker task is the extraction of recommended job seekers or recommended jobs, the presentation of recommended job seekers or recommended jobs based on statistical information of the entire platform (for example, for the purpose of reducing hiring bias, achieving policies, etc.).

[0161] When each task execution model is an AI agent, the task execution unit 118 may cause the multiple AI agents to mutually input task execution instructions from one AI agent to another AI agent. For example, an AI agent may execute a prompt generated by another AI agent for subtask execution, and may input a prompt that it has generated to the other AI agent and cause the other AI agent to execute it. In this way, the task execution unit 118 may cause the multiple task execution models to mutually input prompts and exchange their respective outputs.

[0162] <Artificial Intelligence Department 120> The artificial intelligence unit 120 is configured to receive input from each functional unit and return the instructed output. The artificial intelligence used by the server device 10 in each functional unit may be a common one, or may be prepared individually for each functional unit.

[0163] The artificial intelligence unit 120 may be an AI (Artificial Intelligence) equipped with a learning model such as a transformer including GPT (Generative Pretrained Transformer, including GPT-1, GPT-2, GPT-3, and GPT-4), BERT (Bidirectional Encoder Representations from Transformers), BART (Bidirectional and Auto-regressive Transformer), or a language model such as a recurrent neural network (RNN), and may include a generative AI including a large-scale language model. The large-scale language model is a type of generative AI and includes models provided by services such as OpenAI's GPT, Google's Gemini, and Microsoft's Azure AI Studio. In addition, the artificial intelligence unit 120 may include any machine learning model, deep learning model, artificial intelligence model, etc.

[0164] The language model is an example of a learning model based on a machine learning algorithm. Specific examples of machine learning algorithms include nearest neighbor methods, naive Bayes methods, decision trees, support vector machines, and deep learning using neural networks. The artificial intelligence unit 120 can apply the above algorithms as appropriate.

[0165] The artificial intelligence unit 120 may have a trained model constructed by a learning method such as supervised learning, unsupervised learning, or self-supervised learning. In supervised learning, machine learning is performed using training data (training data). The training data is composed of a pair of input data for learning and output data (correct answer data). In addition, the language model may not only be one trained for a specific task, but also a general-purpose model that can be used for a wide range of tasks.

[0166] The artificial intelligence unit 120 may be a general-purpose natural language processing learning model such as a large language model (LLM) that has learned a huge amount of data as an artificial intelligence. The LLM is a learning model that has learned a large amount of large data (for example, (i) web content on the Internet, or (ii) data stored in a specified database) composed of text data and the like in advance, and can execute various language processing tasks by giving tasks. According to a given prompt, it can perform a wide range of natural language processing tasks such as grasping sentence patterns and contexts, answering questions, and generating sentences. Such a general-purpose learning model includes a language model that can handle various tasks without fine tuning by One-shot Learning, Few-shot Learning, etc. In addition, the general-purpose learning model may be configured to be able to handle various tasks by Zero-shot Learning as well. The artificial intelligence used in each functional unit of the control unit 11 may be a separate learning model, or a common general-purpose learning model.

[0167] The learning models included in the artificial intelligence unit 120 (learning models used in each functional unit, such as the employer task execution model and the job seeker task execution model) can undergo additional learning as transfer learning or fine tuning. For example, each time new data is registered, the artificial intelligence unit 120 may perform additional learning and fine tuning using the new data as new teacher data. This improves the accuracy of the information output from the learning models.

[0168] The learning model included in the artificial intelligence unit 120 may be a learning model (distilled model) obtained by knowledge distillation using an original learning model. In knowledge distillation, a learned model such as a large-scale language model is used as a teacher model, and the parameters of the student model are adjusted so that the output loss (Soft Target Loss) of the student model (distilled model) relative to the output (Soft Target) of the teacher model is reduced, thereby learning the student model, and the student model becomes a distilled model. In addition, the student model may be learned so that the output loss (Hard Target Loss) of the student model relative to the correct label (Hard Target) of the teacher data (combination of input data and output data of the learning model) is reduced. Compared to the original learning model (teacher model), the distilled model has a smaller number of parameters and a smaller processing load while having performance close to that of the learning model. Therefore, by using the distilled model, the cost of the information processing system 1 can be reduced.

[0169] For example, the learning model used in each functional unit may be a distilled model trained using a combination of input data and output data in a large-scale language model as training data. Furthermore, when the information processing system 1 is introduced, a large-scale language model may be used as the learning model used in each functional unit, and when training data from the large-scale language model is accumulated, a distilled model obtained by knowledge distillation using the training data may be used as the learning model used in each functional unit.

[0170] <Display section> The display unit 211 of the recruiter terminal 20 and the display unit 311 of the job seeker terminal 30 each display a screen indicated by the screen data transmitted from the server device 10.

[0171] <Operation acquisition section> The operation acquisition unit 212 of the recruiting party terminal 20 accepts operations by a recruiter who uses the recruiting party terminal 20. The operation acquisition unit 312 of the job seeker terminal 30 accepts operations by a job seeker who uses the job seeker terminal 30.

[0172] 3. Information processing method This section describes an information processing method of the server device 10. This information processing method is executed by a computer, with each unit of the server device 10 acting as each step.

[0173] This information processing comprises an employer agent construction step, a job seeker agent construction step, and a task execution step. In the employer agent construction step, a employer task execution model capable of executing tasks handling information about the employer is constructed for each employer by machine learning using employer learning data about the employer. In the job seeker agent construction step, a job seeker task execution model capable of executing tasks handling information about the job seeker is constructed for each job seeker by machine learning using job seeker learning data about the job seeker. In the task execution step, the employer task execution model or the job seeker task execution model is caused to execute the task based on a task execution instruction given to the employer task execution model or the job seeker task execution model.

[0174] 6 is an activity diagram showing an example of the flow of information processing (task execution processing) executed by the information processing system 1. Below, the information processing will be described along with each activity in this activity diagram.

[0175] The task execution process starts with constructing a recruiter task execution model and a job seeker task execution model. The server device 10 first acquires recruiter learning data and job seeker learning data (activity A101). Next, the server device 10 uses these data to construct a recruiter task execution model and a job seeker task execution model (activity A102). Note that activities A101 and A102 are performed appropriately at any timing.

[0176] The recruiter or job seeker inputs a task (e.g., a question about the recruiter or job seeker) to be executed by the constructed recruiter task execution model or job seeker task execution model at the recruiter terminal 20 or job seeker terminal 30 (activity A201). The server device 10 converts the task execution instruction received from the recruiter terminal 20 or job seeker terminal 30 into the recruiter task execution model or job seeker task execution model (activity A202).

[0177] Next, the server device 10 outputs the execution result of the task according to the recruiter task execution model or the job seeker task execution model (for example, an answer to a question) to the recruiter terminal 20 or the job seeker terminal 30 (activity A203). As a result, the execution result of the task is displayed on the recruiter terminal 20 or the job seeker terminal 30 (activity A204).

[0178] 4. Effect The operation of this embodiment can be summarized as follows: In other words, the job recruiter and the job seeker can be efficiently matched by the task execution model (AI agent) individually constructed for each of the job recruiter and the job seeker.

[0179] Although the embodiment of the present invention has been described above, the present invention is not limited to this, and can be modified as appropriate without departing from the technical concept of the invention.

[0180] 5.Other In the above embodiment, the server device 10 performs various storage and control, but multiple external devices may be used instead of the server device 10. That is, various information and programs may be distributed and stored in multiple external devices using block chain technology or the like. In particular, the artificial intelligence unit 120 may be an external configuration of the server device 10. In that case, the artificial intelligence unit 120, which is an external configuration, is provided by, for example, an artificial intelligence service server, and is configured to receive input from each functional unit of the server device 10, receive a request to execute an artificial intelligence service, and return an instructed output as a processing result to the server device 10. The artificial intelligence service server may be a server that provides a service using a language model as a learning model, or may be a server that executes a language processing task using a language model. The artificial intelligence service server may be constructed by LLM. The artificial intelligence service server receives input of a prompt by text, image, voice, etc., and generates and responds to the prompt.

[0181] The information processing system 1 does not necessarily have to include the platform agent construction unit 117. In other words, the information processing system 1 does not necessarily have to construct a platform task execution model. In addition, the information processing system 1 does not necessarily have to include the employer label creation unit 113, the job seeker label creation unit 115, and the question registration unit 116.

[0182] The aspect of the present embodiment is not limited to the information processing system 1, and may be an information processing method or a program. The information processing method includes each step executed by the information processing system 1. The program causes a computer to execute each step of the information processing system 1.

[0183] It may be provided in any of the following ways:

[0184] (1) An information processing system comprising at least one processor, the processor being configured to execute each of the following steps by reading a program: an employer agent construction step, a employer task execution model capable of executing tasks handling information of employers is constructed for each employer through machine learning using employer learning data related to the employer, wherein the employer learning data includes information of the employer registered in a job seeker database, information provided by the employer, or public information about the employer on a network; a job seeker agent construction step, a job seeker task execution model capable of executing tasks handling information of the job seeker is constructed for each job seeker through machine learning using job seeker learning data related to the job seeker, wherein the job seeker learning data includes information of the job seeker registered in a job seeker database, information provided by the job seeker, or public information about the job seeker on the network; and a task execution step, a job seeker task execution model or a job seeker task execution model is caused to execute a task based on a task execution instruction for the employer task execution model or the job seeker task execution model.

[0185] (2) In the information processing system described in (1) above, in the employing agent construction step, learning data is prepared for each department of the employing agent, each job type of the employing agent, or each job handled by the employing agent, and an employing agent task execution model is constructed for each department, each job type, or each job.

[0186] (3) In the information processing system described in (1) or (2) above, the employer learning data includes, as information about employers registered in the employer database, at least one of data about the employer's recruitment used in a recruitment management system and data about the employer's evaluation of personnel within the organization used in a personnel system.

[0187] (4) In the information processing system described in any one of (1) to (3) above, the job seeker learning data includes, as information on the job seeker registered in the job seeker database, at least one of job offers that the job seeker has viewed or applied for, and scouting emails that the job seeker has opened or replied to.

[0188] (5) An information processing system according to any one of (1) to (4) above, wherein the employer task execution model includes a large-scale language model, and in the employer agent construction step, a employer memory is created for each employer from the employer learning data, which the employer task execution model refers to when generating an output, and the job seeker task execution model includes a large-scale language model, and in the job seeker agent construction step, a job seeker memory is created for each job seeker from the job seeker learning data, which the job seeker task execution model refers to when generating an output, and in the task execution step, the employer memory or the job seeker memory is input as reference information together with a task to the employer task execution model or the job seeker task execution model, and the employer task execution model or the job seeker task execution model is caused to execute the task.

[0189] (6) An information processing system according to any one of (1) to (4) above, wherein in the employer agent construction step, a employer memory representing the characteristics or attributes of the employer is created for each employer from the employer learning data, and the employer task execution model is constructed by machine learning using the employer memory; and in the job seeker agent construction step, a job seeker memory representing the characteristics or attributes of the job seeker is created for each job seeker from the job seeker learning data, and the job seeker task execution model is constructed by machine learning using the job seeker memory.

[0190] (7) An information processing system according to (5) or (6) above, wherein the employer memory includes at least information about the culture of the organization placing the job, and the job seeker memory includes at least information about the personality of the job seeker.

[0191] (8) An information processing system according to any one of (5) to (7) above, wherein the employer agent construction step causes the memory generation model, which is a large-scale language model, to output the employer memory by inputting input information including an input to the employer task execution model, an output of the employer task execution model in response to the input, or a combination thereof, and an instruction to create the employer memory representing the characteristics or attributes of an employer based on the input information, and the job seeker agent construction step causes the memory generation model to output the job seeker memory by inputting input information including an input to the job seeker task execution model, an output of the job seeker task execution model in response to the input, or a combination thereof, and an instruction to create the job seeker memory representing the characteristics or attributes of a job seeker based on the input information, into the memory generation model, which is a large-scale language model,

[0192] (9) In the information processing system described in (8) above, the processor is configured to further execute the following steps: in a recruiter label creation step, the recruiter memory and instructions for generating an recruiter label expressing the contents of the recruiter memory in natural language are input to the memory generation model, thereby causing the memory generation model to output the recruiter label and present all or a part of the recruiter label to the corresponding recruiter; and in a job seeker label creation step, the job seeker memory and instructions for generating a job seeker label expressing the contents of the job seeker memory in natural language are input to the memory generation model, thereby causing the memory generation model to output the job seeker label and present all or a part of the job seeker label to the corresponding job seeker.

[0193] (10) In the information processing system described in (9) above, in the employer label creation step, edits to the employer label are accepted from a corresponding employer, and in the job seeker label creation step, edits to the job seeker label are accepted from a corresponding job seeker.

[0194] (11) In the information processing system described in (10) above, in the employer agent construction step, the employer task execution model is constructed by machine learning using the edited employer label, and in the job seeker agent construction step, the job seeker task execution model is constructed by machine learning using the edited job seeker label.

[0195] (12) In the information processing system described in any one of (1) to (11) above, the employer task execution model is capable of answering questions about employers as tasks, and the job seeker task execution model is capable of answering questions about job seekers as tasks, and in the task execution step, a question for the employer task execution model or the job seeker task execution model is accepted as an instruction to execute a task, and the employer task execution model or the job seeker task execution model is caused to output a response to the question.

[0196] (13) In the information processing system described above in (12), in the task execution step, an information processing system requests a recruiter corresponding to the recruiter task execution model or a job seeker corresponding to the job seeker task execution model to have a conversation with the user depending on the content of the conversation between the user and the recruiter task execution model or the job seeker task execution model.

[0197] (14) In the information processing system described in (12) or (13) above, in the recruiter agent construction step, the recruiter task execution model is constructed by machine learning that further uses recorded data of interviews between the human resources agency and the recruiter or job seeker.

[0198] (15) In the information processing system described in any one of (12) to (14) above, in the job seeker agent construction step, the job seeker task execution model is constructed by machine learning that further uses recorded data of interviews between a human resources agency and a recruiter or a job seeker.

[0199] (16) In the information processing system described in any one of (1) to (15) above, in the task execution step, a job seeker extraction process is performed in which the employer task execution model is made to execute a task, thereby extracting job seekers to be recommended to the employer corresponding to the employer task execution model, or a job offer extraction process is performed in which the job seeker task execution model is made to execute a task, thereby extracting job offers to be recommended to job seekers.

[0200] (17) In the information processing system described in (16) above, the employer task execution model is capable of answering questions about employers as a task, and the job seeker task execution model is capable of answering questions about job seekers as a task, and in the task execution step, the employer task execution model and the job seeker task execution model are caused to have a conversation, and based on the content of the conversation, a job seeker to be recommended to the employer or a job to be recommended to the job seeker is determined.

[0201] (18) In the information processing system described in any one of (1) to (17) above, in the task execution step, a job seeker to be recommended to the employer, or a job to be recommended to the job seeker, is determined based on a similarity calculated from the features of the employer learning data and the features of the job seeker learning data.

[0202] (19) An information processing system according to (18) above, wherein the employer task execution model includes a large-scale language model, and in the employer agent construction step, a employer memory is created for each employer from the employer learning data, which the employer task execution model refers to when generating output; the job seeker task execution model includes a large-scale language model, and in the job seeker agent construction step, a job seeker memory is created for each job seeker from the job seeker learning data, which the job seeker task execution model refers to when generating output; and in the task execution step, a job seeker to be recommended to the employer or a job to be recommended to the job seeker is determined based on a similarity calculated from a feature of the employer memory and a feature of the job seeker memory.

[0203] (20) An information processing system according to any one of (1) to (19) above, wherein the employer task execution model is capable of answering questions about the employer as a task, and the job seeker task execution model is capable of answering questions about the job seeker as a task, and in the task execution step, the result of a conversation between the employer task execution model and the job seeker task execution model is presented to at least one of the employer corresponding to the employer task execution model and the job seeker corresponding to the job seeker task execution model.

[0204] (21) In the information processing system described in any one of (1) to (20) above, the processor is configured to further execute the following steps: in the question registration step, registering advance questions for the employer received from the job seeker; and in the task execution step, causing the employer task execution model corresponding to the employer specified by the job seeker to execute a task to answer the registered advance questions.

[0205] (22) In the information processing system described in any one of (1) to (21) above, the processor is configured to further execute the following steps: in the platform agent construction step, a platform task execution model capable of executing a task handling information of any employer or any job seeker is constructed by machine learning using the employer learning data and the job seeker learning data; and in the task execution step, when the employer task execution model or the job seeker task execution model is made to execute a task, the platform task execution model is also made to execute a task identical to the task or a task related to the task.

[0206] (23) In the information processing system described in (22) above, the employer learning data used in the employer agent building step includes information on employers registered in the employer database in one platform, the job seeker learning data used in the job seeker agent building step includes information on job seekers registered in the job seeker database in the platform, the employer learning data used in the platform agent building step includes information on a plurality of employers registered in the employer database used in the employer agent agent building step, and the job seeker learning data used in the platform agent building step includes information on a plurality of job seekers registered in the job seeker database used in the job seeker agent building step.

[0207] (24) An information processing method comprising steps executed by an information processing system according to any one of (1) to (23) above.

[0208] (25) A program for causing a computer to execute each step of the information processing system described in any one of (1) to (23) above. Of course, this is not the case.

[0209] Finally, although various embodiments according to the present disclosure have been described, these are presented as examples and are not intended to limit the scope of the invention. The novel embodiment can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments and their modifications are included within the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0210] 1: Information processing system 2: Communication lines 10: Server device 11: Control section 12: Storage section 13: Communications Department 14: Communication bus 20: Recruiter terminal 21: Control section 22: Storage section 23: Communications Department 24: Input section 25: Output section 26: Communication bus 30: Job Seeker Terminal 31: Control section 32: Storage section 33: Communications Department 34: Input section 35: Output section 36: Communication bus 111: Basic display control unit 112: Recruiter Agent Construction Department 113: Recruiter Label Creation Department 114: Job Seeker Agent Construction Department 115: Job Seeker Label Creation Department 116: Question Registration Section 117: Platform Agent Construction Department 118: Task execution unit 120: Artificial Intelligence Department 211: Display section 212: Operation acquisition section 311: Display unit 312: Operation acquisition unit

Claims

1. An information processing system, At least one processor; The processor is configured to execute the following steps by reading the program: In the recruiter agent construction step, a recruiter memory representing the characteristics or attributes of the recruiter is generated based on recruiter learning data relating to the recruiter, and a recruiter task execution model capable of executing tasks handling information of the recruiter is constructed for each recruiter by machine learning using the recruiter learning data, wherein the recruiter learning data includes information of the recruiter registered in a recruiter database, information provided by the recruiter, or public information relating to the recruiter on the network; In the job seeker agent construction step, a job seeker memory representing the characteristics or attributes of the job seeker is generated based on job seeker learning data relating to the job seeker, and a job seeker task execution model capable of executing tasks handling information about the job seeker is constructed for each job seeker by machine learning using the job seeker learning data, wherein the job seeker learning data includes information about the job seeker registered in a job seeker database, information provided by the job seeker, or public information about the job seeker on a network; In the task execution step, the job recruiter task execution model or the job seeker task execution model is caused to execute the task based on a task execution instruction for the job recruiter task execution model or the job seeker task execution model, In the recruiter label creation step, a recruiter label expressing the contents of the recruiter memory in natural language is created, and all or a part of the recruiter label is presented to the corresponding recruiter; In the job applicant label creating step, a job applicant label expressing the contents of the job applicant memory in natural language is created, and the information processing system presents all or a part of the job applicant label to the corresponding job applicant.

2. 2. The information processing system according to claim 1, In the employing agent construction step, learning data is prepared for each department of the employing agent, each job type of the employing agent, or each job handled by the employing agent, and an employing agent task execution model is constructed for each department, job type, or job.

3. 2. The information processing system according to claim 1, An information processing system, wherein the employer learning data includes at least one of the following information about employers registered in the employer database: data about the employer's recruitment used in a recruitment management system, and data about the employer's evaluation of personnel within an organization used in a human resources system.

4. 2. The information processing system according to claim 1, An information processing system, wherein the job seeker learning data includes at least one of job offers that the job seeker has viewed or applied for, and scouting emails that the job seeker has opened or replied to, as information on the job seeker registered in the job seeker database.

5. An information processing system, comprising: At least one processor; The processor is configured to execute the following steps by reading the program: In the recruiter agent construction step, a recruiter task execution model capable of executing tasks that handle information of recruiters is constructed for each recruiter by machine learning using recruiter learning data related to the recruiter, wherein the recruiter learning data includes information of the recruiter registered in the recruiter database, information provided by the recruiter, or public information related to the recruiter on the network; In the job seeker agent construction step, a job seeker task execution model capable of executing tasks that handle information about the job seeker is constructed for each job seeker by machine learning using job seeker learning data about the job seeker, wherein the job seeker learning data includes information about the job seeker registered in a job seeker database, information provided by the job seeker, or public information about the job seeker on a network; In the task execution step, the job recruiter task execution model or the job seeker task execution model is caused to execute the task based on a task execution instruction for the job recruiter task execution model or the job seeker task execution model, The recruiter task execution model includes a large-scale language model; In the recruiter agent construction step, a recruiter memory is created for each recruiter from the recruiter learning data, which the recruiter task execution model refers to when generating an output; the job seeker task performance model includes a large-scale language model; In the job seeker agent construction step, a job seeker memory is created for each job seeker from the job seeker learning data, which the job seeker task execution model refers to when generating an output; In the task execution step, the recruiter memory or the job seeker memory is input as reference information together with a task to the recruiter task execution model or the job seeker task execution model, and the recruiter task execution model or the job seeker task execution model is made to execute the task; In the recruiter agent construction step, input information including an input to the recruiter task execution model, an output of the recruiter task execution model in response to the input, or a combination of these, and an instruction to create the recruiter memory representing the characteristics or attributes of recruiters based on the input information is input to a memory generation model, which is a large-scale language model, thereby causing the memory generation model to output the recruiter memory; the job seeker agent construction step includes inputting, into the memory generation model, input information including an input to the job seeker task execution model, an output of the job seeker task execution model in response to the input, or a combination thereof, and an instruction to create the job seeker memory representing characteristics or attributes of a job seeker based on the input information, thereby causing the memory generation model to output the job seeker memory; In the recruiter label creation step, the recruiter memory and an instruction to generate a recruiter label expressing the content of the recruiter memory in natural language are input to the memory generation model, thereby causing the memory generation model to output the recruiter label and presenting all or a part of the recruiter label to a corresponding recruiter; In the job applicant label creation step, the job applicant memory and instructions for generating a job applicant label expressing the contents of the job applicant memory in natural language are input to the memory generation model, thereby causing the memory generation model to output the job applicant label and presenting all or a part of the job applicant label to the corresponding job applicant.

6. 2. The information processing system according to claim 1, In the recruiter agent construction step, the recruiter task execution model is constructed by machine learning using the recruiter memory, an information processing system, wherein in the job seeker agent construction step, the job seeker task execution model is constructed by machine learning using the job seeker memory.

7. 6. The information processing system according to claim 5, the recruiter memory includes at least information about the culture of the recruiting organization; The job seeker memory includes at least information about personality traits of job seekers.

8. 6. The information processing system according to claim 5, In the recruiter label creation step, an edit of the recruiter label is accepted from a corresponding recruiter, an information processing system, in the job seeker label creating step, receiving an edit of the job seeker label from a corresponding job seeker.

9. 9. The information processing system according to claim 8, In the recruiter agent construction step, the recruiter task execution model is constructed by machine learning using the edited recruiter label, an information processing system, wherein in the job seeker agent construction step, the job seeker task execution model is constructed by machine learning using the edited job seeker label.

10. 2. The information processing system according to claim 1, The recruiter task execution model is capable of answering questions about recruiters as a task, The job seeker task execution model is capable of answering questions about the job seeker as a task, The information processing system, in the task execution step, receives a question for the employer task execution model or the job seeker task execution model as an instruction to execute a task, and causes the employer task execution model or the job seeker task execution model to output an answer to the question.

11. 11. The information processing system according to claim 10, In the task execution step, an information processing system requests a recruiter corresponding to the recruiter task execution model or a job seeker corresponding to the job seeker task execution model to have a conversation with the user, depending on the content of the conversation between the user and the recruiter task execution model or the job seeker task execution model.

12. 11. The information processing system according to claim 10, In the recruiter agent construction step, an information processing system constructs the recruiter task execution model by machine learning that further uses recorded data of interviews between the human resources agency and the recruiter or job seeker.

13. 11. The information processing system according to claim 10, An information processing system, wherein in the job seeker agent construction step, the job seeker task execution model is constructed by machine learning further using recorded data of interviews between a human resources agency and a recruiter or a job seeker.

14. 2. The information processing system according to claim 1, The task execution step is an information processing system that performs a job seeker extraction process in which the employer task execution model executes a task to extract job seekers to be recommended to the employer corresponding to the employer task execution model, or a job offer extraction process in which the job seeker task execution model executes a task to extract job offers to be recommended to job seekers.

15. 15. The information processing system according to claim 14, The recruiter task execution model is capable of answering questions about recruiters as a task, The job seeker task execution model is capable of answering questions about the job seeker as a task, In the task execution step, an information processing system has the employer task execution model and the job seeker task execution model converse with each other, and determines a job seeker to be recommended to the employer, or a job offer to be recommended to the job seeker, based on the content of the conversation.

16. 2. The information processing system according to claim 1, In the task execution step, an information processing system determines a job seeker to be recommended to an employer, or a job offer to be recommended to a job seeker, based on a similarity calculated from the features of the employer learning data and the features of the job seeker learning data.

17. 17. The information processing system according to claim 16, In the task execution step, an information processing system determines a job seeker to be recommended to a recruiter, or a job offer to be recommended to a job seeker, based on a similarity calculated from the features of the recruiter memory and the features of the job seeker memory.

18. 2. The information processing system according to claim 1, The recruiter task execution model is capable of answering questions about recruiters as a task, The job seeker task execution model is capable of answering questions about the job seeker as a task, In the task execution step, an information processing system presents the results of a conversation between the employer task execution model and the job seeker task execution model to at least one of the employer corresponding to the employer task execution model and the job seeker corresponding to the job seeker task execution model.

19. 2. The information processing system according to claim 1, The processor is further configured to perform the steps of: In the question registration step, the job seeker registers the advance questions to be asked by the employer. In the task execution step, the information processing system causes the employer task execution model corresponding to the employer specified by the job seeker to execute a task of answering the registered advance questions.

20. An information processing system, comprising: At least one processor; The processor is configured to execute the following steps by reading the program: In the recruiter agent construction step, a recruiter task execution model capable of executing tasks that handle information of recruiters is constructed for each recruiter by machine learning using recruiter learning data related to the recruiter, wherein the recruiter learning data includes information of the recruiter registered in the recruiter database, information provided by the recruiter, or public information related to the recruiter on the network; In the job seeker agent construction step, a job seeker task execution model capable of executing tasks that handle information about the job seeker is constructed for each job seeker by machine learning using job seeker learning data about the job seeker, wherein the job seeker learning data includes information about the job seeker registered in a job seeker database, information provided by the job seeker, or public information about the job seeker on a network; In the task execution step, the job recruiter task execution model or the job seeker task execution model is caused to execute the task based on a task execution instruction for the job recruiter task execution model or the job seeker task execution model, In the platform agent construction step, a platform task execution model capable of executing a task handling information of an arbitrary employer or an arbitrary job seeker is constructed by machine learning using the employer learning data and the job seeker learning data; an information processing system, in which, in the task execution step, when the employer task execution model or the job seeker task execution model is made to execute a task, the platform task execution model is also made to execute a task identical to the task or a task related to the task.

21. 21. The information processing system according to claim 20, The recruiter learning data used in the recruiter agent construction step includes information on recruiters registered in the recruiter database in one platform, the job seeker learning data used in the job seeker agent construction step includes information on job seekers registered in the job seeker database in the platform, The recruiter learning data used in the platform agent construction step includes information on a plurality of recruiters registered in the recruiter database used in the recruiter agent construction step, an information processing system, wherein the job seeker learning data used in the platform agent construction step includes information on a plurality of job seekers registered in the job seeker database used in the job seeker agent construction step.

22. 1. An information processing method, comprising: An information processing method comprising the steps executed by the information processing system according to any one of claims 1 to 21.

23. A program, A program for causing a computer to execute each step of the information processing system according to any one of claims 1 to 21.

Images