Matching method, matching system, and program

Abstract

This matching method for matching human resources with hiring companies comprises: collecting first information about each of a plurality of hiring companies accessible via the Internet; analyzing a first characteristic indicating a culture of each of the plurality of hiring companies by using the first information and a first large language model; acquiring second information provided from a human resource; collecting third information about the human resource accessible via the Internet by using the second information; analyzing a second characteristic indicating the mindset of the human resource by using the third information and a second large language model; and performing first matching processing using the first characteristic and the second characteristic to identify a hiring company the culture of which fits the human resource from among the plurality of hiring companies.

Description

Matching Method, Matching System, and Program

[0001] The present invention relates to a matching method, a matching system, and a program.

[0002] Conventionally, as a service for providing human resource support in job hunting activities and the like, a matching service that matches job seekers and employers has become widespread. In such a matching service, candidates such as job offers from employers are extracted and presented based on information such as the skills, experience, and desired job content of the job seekers.

[0003] Aspects such as the culture, ideology, and atmosphere of a company are important when a job seeker is looking for a company or assuming long-term employment, but it is difficult to measure and judge them. Generally, in terms of such perspectives, job seekers judge whether they are suitable based on information disclosed on the websites provided by employers. On the other hand, technologies for performing human resource matching based on qualitative evaluation criteria such as corporate culture have also been disclosed.

[0004] For example, in Patent Document 1, a system for providing human resource support is disclosed that uses a total score indicating the degree of matching, considering cultural fit based on corporate culture and skill fit based on the skills of job seekers when matching job seekers and employers. Also, in Patent Document 2, a matching system that performs matching considering the atmosphere and policies of an organization is disclosed in order to reduce mismatches between the employer and the job seeker sides.

[0005] Japanese Patent No. 7219981 Japanese Patent No. 7346648

[0006] In conventional technologies like those described above, questions are set for recruiting companies with which the recruitment agency has a business relationship, and the company culture is quantified based on the information gathered from the companies' responses. In other words, in order to define a recruiting company's corporate culture as an evaluation criterion, collaboration between the recruitment agency and the recruiting company is required beforehand. Therefore, the recruitment agency needs to collaborate with various companies, classify the corporate culture of each company, and use it for matching. As a result, identifying a company's corporate culture and using it for talent matching required many steps. Furthermore, the workload increased even more as the number of companies increased.

[0007] Furthermore, job postings from recruiting companies are generally provided to job seekers through recruitment agencies. Each recruitment agency provides job postings from companies with which it has a business relationship, so the job postings available vary from agency to agency. Therefore, in order for job seekers to obtain a variety of job postings, they need to register with multiple recruitment agencies. Moreover, in order to select job postings from companies that are a good match for them, they need to review many of these postings. As a result of the above, job seekers have had to go through a lot of trouble to gather job information.

[0008] In light of the above issues, the present invention aims to provide a method that further improves the convenience of matching various companies with personnel and enables more appropriate matching.

[0009] To solve the above problems, one embodiment of the present invention has the following configuration. That is, a matching method for matching personnel with recruiting companies, comprising: a first collection step of collecting first information about each of a plurality of recruiting companies that are accessible via the Internet; a first analysis step of analyzing first characteristics indicating the culture of each of the plurality of recruiting companies using the first information and a first large-scale language model; an acquisition step of acquiring second information provided by personnel; a second collection step of collecting third information about the personnel that are accessible via the Internet using the second information; a second analysis step of analyzing second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching step of identifying a recruiting company that is a cultural fit for the personnel from among the plurality of recruiting companies by performing a first matching process using the first characteristics and the second characteristics.

[0010] Another embodiment of the present invention has the following configuration: a matching method for matching personnel with companies, comprising: a first collection step of collecting first information about each of a plurality of companies that are accessible via a network; a first analysis step of analyzing first characteristics indicating the culture of each of the plurality of companies using the first information and a first large-scale language model; an acquisition step of acquiring second information provided by personnel; a second collection step of collecting third information about the personnel that are accessible via a network using the second information; a second analysis step of analyzing second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching step of identifying a company that is a cultural fit for the personnel from among the plurality of companies by performing a first matching process using the first characteristics and the second characteristics.

[0011] Another embodiment of the present invention has the following configuration: a matching system for matching personnel with recruiting companies, comprising: a first collection unit that collects first information about each of a plurality of recruiting companies accessible via the Internet; a first analysis unit that uses the first information and a first large-scale language model to analyze first characteristics indicating the culture of each of the plurality of recruiting companies; an acquisition unit that acquires second information provided by personnel; a second collection unit that uses the second information to collect third information about the personnel that is accessible via the Internet; a second analysis unit that uses the third information and a second large-scale language model to analyze second characteristics indicating the mindset of the personnel; and a first matching unit that performs a first matching process using the first characteristics and the second characteristics to identify recruiting companies from among the plurality of recruiting companies that are a cultural fit for the personnel.

[0012] Another embodiment of the present invention has the following configuration: a matching system for matching personnel with companies, comprising: a first collection unit that collects first information about each of a plurality of companies accessible via a network; a first analysis unit that analyzes first characteristics indicating the culture of each of the plurality of companies using the first information and a first large-scale language model; an acquisition unit that acquires second information provided by personnel; a second collection unit that collects third information about the personnel, accessible via a network, using the second information; a second analysis unit that analyzes second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching unit that identifies companies from among the plurality of companies that are a cultural fit for the personnel by performing a first matching process using the first characteristics and the second characteristics.

[0013] Another embodiment of the present invention has the following configuration: a program that causes a computer to perform: a first collection step of collecting first information about each of a plurality of recruiting companies that are accessible via the Internet; a first analysis step of analyzing first characteristics indicating the culture of each of the plurality of recruiting companies using the first information and a first large-scale language model; an acquisition step of acquiring second information provided by the person; a second collection step of collecting third information about the person that is accessible via the Internet using the second information; a second analysis step of analyzing second characteristics indicating the mindset of the person using the third information and a second large-scale language model; and a first matching step of identifying a recruiting company that is a cultural fit for the person from among the plurality of recruiting companies by performing a first matching process using the first characteristics and the second characteristics.

[0014] Another embodiment of the present invention has the following configuration: a program that causes a computer to perform: a first collection step of collecting first information about each of a plurality of companies that are accessible via a network; a first analysis step of analyzing first characteristics indicating the culture of each of the plurality of companies using the first information and a first large-scale language model; an acquisition step of obtaining second information provided by personnel; a second collection step of collecting third information about the personnel that are accessible via a network using the second information; a second analysis step of analyzing second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching step of identifying a company from among the plurality of companies that is a cultural fit for the personnel by performing a first matching process using the first characteristics and the second characteristics.

[0015] According to the present invention, it is possible to improve the convenience of matching various companies with personnel and to achieve more appropriate matching.

[0016] Block diagram showing an example configuration of the matching system according to the first embodiment of the present invention. Functional block diagram showing an example configuration of the matching server according to the first embodiment of the present invention. Functional block diagram showing a user terminal according to the first embodiment of the present invention. Schematic diagram for explaining the processing of the matching server according to the first embodiment of the present invention. Conceptual diagram of LLM for corporate culture analysis according to the first embodiment of the present invention. Conceptual diagram of LLM for personal tendency analysis according to the first embodiment of the present invention. Conceptual diagram of LLM for corporate required skills analysis according to the first embodiment of the present invention. Sequence diagram of processing according to the first embodiment of the present invention. Sequence diagram of processing according to the first embodiment of the present invention. Flowchart of the matching process according to the first embodiment of the present invention. Schematic diagram for explaining the diagnostic process according to a modified example of the present invention. Flowchart of the diagnostic process according to a modified example of the present invention. Schematic diagram for explaining the outline of the business model according to the present invention.

[0017] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings and other documents. The embodiments described below are merely examples for illustrating the present invention and are not intended to be interpreted as limiting the invention. Furthermore, not all configurations described in each embodiment are necessarily essential for solving the problems of the present invention. In each drawing, the same components are given the same reference numeral to indicate their correspondence. To avoid unnecessary redundancy and to facilitate understanding by those skilled in the art, some parts of the description may be omitted or simplified. For example, detailed explanations of already well-known matters or redundant explanations of substantially identical configurations may be omitted.

[0018] <First Embodiment> [System Configuration] Figure 1 is a schematic diagram showing an example of the configuration of the matching system 1 according to the first embodiment of the present invention. The matching system 1 is a system for providing the matching service according to this embodiment, and for example, the matching service is provided by a matching service provider. In addition, service users who use the matching service include, for example, job seekers who collect job information when looking for a new job (hereinafter also referred to as "personnel") and companies that are looking for personnel (hereinafter also referred to as "recruiting companies").

[0019] Matching System 1 consists of a matching server 100, a user terminal 200, a recruiting company system 300, a linkage system 400, and an external online system 500. Each system and device constituting Matching System 1 is configured to communicate via a network NW.

[0020] The matching server 100 provides the matching service described below to job seekers who access it using the user terminal 200. It is also possible to configure the system so that some or all of the functions described below are provided on the user terminal 200 side (for example, an application program installed on the user terminal 200) or on various systems that can be linked via a network NW. The matching server 100 may be configured on-premises using a general-purpose computer such as a workstation or personal computer, or it may be logically implemented through cloud computing. In this embodiment, for the sake of explanation, one matching server 100 is used as an example, but it is not limited to this; multiple servers may be configured, and servers with different roles, such as an authentication server and a database server, may also be included.

[0021] The user terminal 200 is an operating terminal used by users of the matching service, such as job seekers. The user terminal 200 may consist of an information processing device such as a personal computer, tablet terminal, smartphone, or POS terminal. The user terminal 200 provides users with the functions of the matching service provided by the matching server 100 and applications installed and running on the user terminal 200. In the example in Figure 1, one user terminal 200 is shown, but many more user terminals 200 may be used depending on the job seekers.

[0022] The recruiting company system 300 is a system built by a recruiting company that provides information about the company. In this embodiment, examples of company information will be described later, but for example, the recruiting company system 300 may be a system that provides a homepage to introduce the company, or an email system used by the company. Here, various recruiting companies are assumed, and multiple recruiting company systems 300 are shown. The configuration and functions of the multiple recruiting company systems 300 may differ from each other or may be the same. For convenience, they will be described together as the recruiting company system 300. When it is necessary to explain each recruiting company separately, subscripts (a, b, ...) will be added.

[0023] The collaboration system 400 is an external system that functions in cooperation with the matching server 100 via a network NW. The collaboration system 400 may be, for example, a database (hereinafter referred to as "DB") system that provides information on a predetermined number of recruiting companies, or it may be a function server that provides predetermined functions or services.

[0024] The external online system 500 is a variety of systems configured on a network NW, and is used by the matching server 100 to collect various information about recruiting companies and job seekers (personnel). The external online system 500 may include various types of services, such as SNS (Social Network Service), news sites, social networking sites, and review sites. Therefore, the type of external online system 500 used for information gathering and the data format collected are not particularly limited. Although only one linkage system 400 and one external online system 500 are shown, more systems may be included, and they may be composed of one or more devices depending on the functions and services provided.

[0025] A network (NW) is an internal / external network composed of the Internet, intranet, wireless LAN (Local Area Network), WAN (Wide Area Network), etc. The communication standards and wired / wireless connections used in a network (NW) are not particularly limited, and a network (NW) may be composed of a combination of multiple communication standards.

[0026] Figure 2 is a block diagram showing an example of the functional configuration of the matching server 100 according to this embodiment. The matching server 100 is composed of a control unit 110, a storage unit 130, and a communication unit 140. Each part is configured to communicate with each other via an internal bus or the like.

[0027] The control unit 110 is responsible for controlling the operation of the matching server 100. The control unit 110 is composed of, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an NPU (Neural Network Processing Unit), etc., and provides various functions by reading and executing various programs and data stored in the storage unit 130. The control unit 110 functions as a data management unit 111, a display control unit 112, a recruiting company information acquisition unit 113, a personnel information acquisition unit 114, an information collection unit 115, an LLM management unit 116, a question management unit 117, a culture analysis unit 118, a culture matching unit 119, a skill analysis unit 120, a skill matching unit 121, a diagnostic unit 122, and a communication control unit 123.

[0028] The data management unit 111 works in conjunction with the storage unit 130 to manage various types of data and provides, registers, updates, etc., data as requested. The display control unit 112 controls the display of the UI (User Interface) screen shown on the user terminal 200 based on user operations performed via the user terminal 200.

[0029] The recruiting company information acquisition unit 113 acquires information about recruiting companies (hereinafter referred to as "recruiting company information"). Examples of recruiting company information will be described later. The human resources information acquisition unit 114 acquires information about human resources (hereinafter referred to as "human resources information"). Details of human resources information will be described later. The information gathering unit 115 uses the acquired recruiting company information, human resources information, etc., to collect various information from external sources.

[0030] The LLM management unit 116 manages various LLMs (Large Language Models) according to this embodiment. The LLMs used in this embodiment are assumed to have already undergone training using deep learning technology or the like and be built in a usable state. The degree of training of the LLM is not particularly limited and may be updated as appropriate. Furthermore, the algorithm, type, training data, etc. of the language model for building the LLM are not particularly limited and should be able to provide the functions described later. In this embodiment, the LLM also includes the setting of prompts to enable predetermined outputs. Therefore, the LLM management unit 116 also manages prompts for the LLM in order to realize the functions described later. The content of the prompt description is not particularly limited, but it is described in a way that realizes the output shown in Figures 5A to 5C described later and is input to the LLM in combination with the input data described later.

[0031] The Question Management Unit 117 manages questions regarding job requirements and other conditions for job applicants. The content of the questions is not particularly limited, but it includes items necessary for the mindset analysis described later.

[0032] The Culture Analysis Department 118 analyzes the characteristics of recruiting companies, such as their culture, philosophy, and atmosphere (hereinafter simply referred to as "culture"), using LLM and other methods as described below. Similarly, the Culture Analysis Department 118 analyzes the characteristics of individuals, such as their culture, philosophy, thoughts, and preferences (hereinafter referred to as "mindset"), using LLM and other methods as described below. Examples of culture and mindset analysis will be described later. The term "culture" is not intended to be interpreted restrictively and may include all qualitative characteristics and tendencies possessed by recruiting companies. Furthermore, there are no particular limitations on the granularity of "culture." For example, for a single recruiting company, it may be defined at various levels, such as the entire company, organizational units within the company, company locations, or project units conducting recruitment. Similarly, the term "mindset" is not intended to be interpreted restrictively and may include all qualitative characteristics and tendencies possessed by individuals. The Culture Matching Department 119 matches recruiting companies with individuals based on the analysis results of the Culture Analysis Department 118. In this embodiment, culture and mindset are primarily analyzed using LLM, but this is not the only method. For example, various preprocessing methods, as described later, may be combined, or known statistical methods may be used in combination.

[0033] The Skill Analysis Unit 120 analyzes the skills required by the recruiting company (hereinafter referred to as "company-required skills") based on the recruiting company's job information, using LLM. An example of the analysis of company-required skills will be described later. The Skill Matching Unit 121 matches recruiting companies with candidates based on the analysis results of the Skill Analysis Unit 120.

[0034] The diagnostic unit 122 performs a process to diagnose a person by matching them with a company that has a culture that matches that person. Details of the process of the diagnostic unit 122 will be described in the second embodiment. The communication control unit 123 controls communication with an external device (for example, a user terminal 200) and transmits and receives data.

[0035] The functional configuration of the control unit 110 is an example; the multiple functional blocks shown in Figure 2 may be combined into one, or one functional block may be divided into multiple components. Furthermore, the functions provided by the control unit 110 are not limited to those shown in Figure 2; other functions may be provided as needed.

[0036] The memory unit 130 is a storage device for storing programs, data, and the like for executing various control processes and functions within the control unit 110. The memory unit 130 is composed of volatile / non-volatile storage devices such as RAM (Random Access Memory), ROM (Read Only Memory), HDD (Hard Disk Drive), and flash memory. The memory unit 130 further includes a database for managing data used in accordance with the functions described later.

[0037] Company Information DB 131 manages various information about recruiting companies. Human Resources Information DB 132 manages various information about human resources. Related Information DB 133 manages various related information about recruiting companies and human resources collected via the internet, etc.

[0038] Classification Information DB134 manages information about the types of culture, mindset, and skill classifications and their results used in LLM output, etc. Question Information DB135 manages the content of questions that job seekers are required to input prior to the matching process. Job Posting Information DB136 manages job postings provided by recruiting companies.

[0039] Each database is configured to allow referencing, updating, and registration through the processes described later. Note that the database configuration is just an example; multiple databases as shown in Figure 2 may be combined into one, or one database may be divided into multiple databases.

[0040] The communication unit 140 is a communication interface for communicating with external devices via a network NW. The communication unit 140 may be configured to support multiple communication standards depending on the configuration of the network NW.

[0041] FIG. 3 is a block diagram showing an example of the functional configuration of the user terminal 200 according to the present embodiment. The user terminal 200 includes a control unit 210, a storage unit 220, a communication unit 230, an operation unit 240, a display unit 250, and an external IF (Interface) 260.

[0042] The control unit 210 controls the operation of the user terminal 200. The control unit 210 is composed of, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an NPU (Neural network Processing Unit), etc., and provides various functions by reading and executing various programs and data stored in the storage unit 220.

[0043] The storage unit 220 is a storage device for storing programs, data, etc. for executing various control processes and each function of the control unit 210. The storage unit 220 is composed of volatile / non-volatile storage devices such as a RAM (Random Access Memory), a ROM (Read Only Memory), an HDD (Hard disk Drive), a flash memory, etc.

[0044] The communication unit 230 is a communication interface for communicating with an external device via the network NW. The communication unit 230 may be configured to support a plurality of communication standards according to the configuration of the network NW.

[0045] The operation unit 240 is an interface for receiving operations from the user of the user terminal 200. The operation unit 240 may be composed of a mouse, a keyboard, etc. The display unit 250 is an interface for displaying various screens and is composed of a display, etc. A touch panel display integrating the operation unit 240 and the display unit 250 may be used. The external IF 260 is an interface for connecting to various devices and may be, for example, a connection interface to an imaging unit (not shown) for taking images and sensors for acquiring predetermined information.

[0046] [Matching Process] Using FIG. 4, the matching process by the matching server 100 according to the present embodiment will be described. Here, a matching business operator 410 that manages the matching server 100, an employer company 430 that posts job offers, and a talent 450 who is a job seeker are shown respectively.

[0047] In the present embodiment, as information regarding one employer company 430, employer company information 431 is handled. The employer company information 431 includes job offer information 432, company public information 433, and company-related information 434. The job offer information 432 includes the conditions for job offers by the employer company (work location, job content, required qualifications, age, required skills, etc.). The company public information 433 is information publicly disclosed by the employer company itself and can be directly obtained from the employer company. The company-related information 434 is information related to the employer company collected from an external online system 500 constructed on a network NW such as the Internet using information indicating the employer company (name, account of a predetermined service, location, representative name, business content, etc.). Note that the company-related information 434 may include information indirectly obtained or estimated from publicly disclosed information in addition to information directly publicly disclosed by the employer company itself.

[0048] More specifically, the company public information 433 and the company-related information 434 may include the mission, vision, and values within the official website, representative messages, employee interviews, employment policies, business content, projects, related companies, sponsored projects, affiliated talents, business achievements, various articles from other sites, information on the official SNS account, information on the SNS accounts of employees (including affiliation, position, job content, etc.) (where the company name is included in the profile), advertising information, word-of-mouth information, etc. The company public information 433 and the company-related information 434 may partially include non-public information or information with access restrictions. When handling such cases, it may be made available after filtering a part of the information or performing processing such that sensitive information is not included.

[0049] Furthermore, it is conceivable that some of the collectible job postings 432 may have information that is not publicly available. For example, the job description may be public, but the name of the recruiting company may be kept private. When such job postings are collected, the recruiting company corresponding to the job posting whose name is not public may be identified based on additionally collected information or past job postings. In this case, the recruiting company may be identified by LLM using a prompt that estimates the job posting and the corresponding recruiting company. Alternatively, the recruiting company may be identified by processing with a predetermined matching algorithm that matches recruiting companies based on past job postings. Although company names have been used as an example here, processing may be performed to estimate or identify other undisclosed information items.

[0050] Furthermore, when collecting information from social media, etc., it is acceptable to identify which specific recruiting company, organization, branch, team, or project the user who posted the information (e.g., an employee of the recruiting company) belongs to. If the affiliation is directly indicated, that information may be used. If there is no direct indication, the LLM may be used to identify the user's affiliation, etc., using information from social media and prompts that estimate the corresponding affiliation. In addition, if the estimation can be made with a certain level of confidence, the user's mindset may be added not only as an element of the recruiting company as a whole, but also as an element of each organization (branch, team, etc.) to which the user belongs, and not as an element of other organizations (other teams, etc.). Furthermore, depending on information such as the level of confidence and the size of the target organization, the appropriate influence of the user on the organization may be reflected by assigning weights or degrees to the mindsets that are added.

[0051] When estimating the affiliation of recruiting companies or users as described above using LLM, the confidence level of the estimation may also be obtained. The obtained confidence level may be stored as company-related information 434 in association with the estimation results, and may be used in subsequent processing such as matching processing, or it may be presented as information to job seekers.

[0052] It should be noted that the publicly available company information 433 and the company-related information 434 are not strictly separate and may be managed as a single entity. Furthermore, the publicly available job posting information 433 and the job-related information 434 may be managed using a predetermined hierarchy or classification. For example, within a single recruiting company, information may be collected and managed in accordance with classifications and hierarchies such as the entire company, internal organizational units, branch units, and project units. In other words, each unit may have different characteristics, and the hierarchy and classification may be carried out with these in mind. Some of the publicly available company information 433 may be obtained, for example, via the internet.

[0053] Furthermore, information about one individual 450 is handled as personnel information 451. Personnel information 451 includes job seeker information 452, aptitude information 453, and personnel-related information 454. Job seeker information 452 includes job seeker conditions (desired work location, desired job content, desired treatment, etc.), information on work experience, and qualifications held. Aptitude information 453 includes information on responses to questions predefined by the matching service provider (matching server 100). The content and format of the questions are defined in the question management DB 135 of the matching server 100, but are not particularly limited. For example, it may be a questionnaire format including multiple multiple-choice questions with several options, or it may be a free-response format. The questions may include items to identify the characteristics and personality of the individual. For example, they may include questions about individual inclinations (stable, challenging, etc.) and tendencies for maintaining motivation. There are also no particular limitations on the types or number of questions. Personnel-related information 454 is information about a person that is collected from an external online system 500 built on a network NW such as the Internet, using information that identifies the person (such as their name or account for a designated service). Personnel-related information 454 may include, for example, posts on social media, public announcements, related news, and comments from third parties on the Internet. Personnel-related information 454 may also include the person's affiliation, position, work history, and job description.

[0054] On the matching service provider 410 side, a network and search functions are configured to enable the collection and acquisition of various types of information as described above. In this embodiment, three types of LLMs (LLM 411, 413, 417) are used. These are assumed to be pre-configured and ready for use. Examples of each LLM will be explained using Figures 5A to 5C.

[0055] On the matching server 100 of the matching service provider 410, the company's public information 433 and company-related information 434 of the recruiting company 430 are input into the LLM 411. The LLM 411 then outputs information indicating which of the predefined culture classifications the recruiting company 430 belongs to as the recruiting company 430's culture classification information 412. The matching server 100 also inputs the personnel's aptitude information 453 and personnel-related information 454 into the LLM 413. Note that the information to be input is not limited to these; for example, the personnel's job-seeking information 452 may also be used. The LLM 413 then outputs information indicating which of the predefined mindset classifications the personnel 450 belongs to as the personnel 450's mindset classification information 414.

[0056] Furthermore, the matching server 100 performs culture matching using the culture classification information 412 of the recruiting company 430 and the mindset classification information 414 of the personnel 450. This culture matching is performed using a ratio or score indicating how well the recruiting company 430 and the personnel 450 match in terms of culture / mindset evaluation criteria. The matching method based on the two classification information is not particularly limited. For example, a score may be calculated using a weighted sum based on the conversion formulas of each element of the culture classification information 412 and the mindset classification information 414, or the distance between feature points in a multidimensional space composed of each element may be calculated. Here, only one recruiting company 430 is shown, but in reality, culture matching is performed between multiple recruiting companies and personnel 450, and the recruiting companies 430 are ranked according to the matching results and output as a culture fit result 416. The culture fit result 416 may include, for example, the confidence level of the estimation when using LLM 411 or LLM 413 (corresponding to the "element strength" described later). In other words, the system may be configured to present job seekers with a confidence level indicating the accuracy of the estimated results regarding the culture at the recruiting company 430.

[0057] As mentioned above, it is possible to collect information on the recruiting company 430 as publicly available company information 433 and company-related information 434, including not only the overall company culture but also information at the organizational level and location level. Therefore, culture matching 415 is not limited to matching the overall company culture with personnel, but may be structured to perform matching at predetermined levels within the company. In particular, for companies above a certain size, it is assumed that the culture is diverse depending on the team or project. Therefore, the units of culture classification and matching may be switched depending on the size of the recruiting company, the characteristics of the project or work, etc.

[0058] Based on the culture fit results 416, the matching server 100 provides the recruiting company 430 with information on the candidates 450 who have scored above a predetermined value. Similarly, based on the culture fit results 416, the matching server 100 provides the candidates 450 with information on the recruiting company 430 that has scored above a predetermined value, as well as recruiting company 430 that have ranked highly.

[0059] Furthermore, the matching server 100 inputs the job information 432 of the recruiting company 430 into the LLM 417. The LLM 417 then outputs the skills that the recruiting company 430 requires from among several predetermined skill classifications as skill classification information 418 for the recruiting company 430. The matching server 100 then performs skill matching 420 between the skill classification information 418 of the recruiting company 430 that ranks highly in the culture fit result 416 and the skill information 419 shown in the job seeker information 452 of the person 450. This skill matching is performed using a ratio or score indicating how well the recruiting company 430 and the person 450 match in terms of the evaluation criteria for the company's required skills. The matching method based on the person 450's skill information and the recruiting company 430's skill classification information is not particularly limited. Also, if there are essential company-required skills, these may be prioritized and reflected in the score. Skill matching is performed between multiple top-ranked recruiting companies and 450 candidates. Based on the matching results, 430 recruiting companies are ranked and output as skill fit results 421.

[0060] In the configuration shown in Figure 4, job information 432 is used to obtain skill classification information 418, but the configuration is not limited to this. For example, skills required by the entire recruiting company 430, or skills required by an organization within the recruiting company 430, may also be considered. In such cases, job posting information 433 and job-related information 434 may be used to further obtain skill classification information 418. In this case, the information items entered into LLM 411 and LLM 417 of the job posting information 433 and job-related information 434 may be configured to be different. Alternatively, the information items entered into LLM 417 may be collected separately from the internet or other sources. Furthermore, experience / skills that are essential for the industry, business, and job type related to the recruiting company and the job content may be used as input to LLM 417.

[0061] Furthermore, although not shown in the configuration of Figure 4, the skill information 419 of personnel 450 may be obtained using personnel information 451 and LLM. Then, by performing skill matching using the derived skill information 419 and skill classification information 418, companies with a skill fit may be identified.

[0062] Based on the skill fit results 421, the matching server 100 provides the recruiting company 430 with information about the individuals 450 who have scored above a predetermined value. Similarly, based on the skill fit results 421, the matching server 100 provides the individuals 450 with information about the recruiting companies 430 that have scored above a predetermined value, as well as recruiting companies 430 that are ranked highly.

[0063] Figure 5A is a conceptual diagram illustrating LLM411, which is used to analyze corporate culture. LLM411 takes publicly available company information 433 and company-related information 434 of a recruiting company as input and outputs culture classification information 412 that shows the cultural tendencies of the recruiting company. In this embodiment, the culture classification information 412 for a recruiting company is represented by a predetermined number of elements (elements 1 to N) and element strengths (element strengths 1 to N) that indicate the degree of suitability to those elements. Therefore, LLM411 outputs how strong (tendency) a particular recruiting company is to each element (culture classification). The number of elements and the types of elements here are not particularly limited and may be adjusted as appropriate depending on the industry, job type, technology field, time period, etc. of the recruiting company.

[0064] For example, LLM411 derives one or more elements (which may be the same as or different from N) corresponding to a culture based on the input information, and clusters them into the aforementioned multiple elements (=N). Then, LLM411 may derive the element strength of each element based on the distribution of each clustered element.

[0065] Figure 5B is a conceptual diagram illustrating LLM413, which is used to analyze the mindset tendencies of individual personnel. LLM413 takes personnel aptitude information 453 and personnel-related information 454 as input and outputs mindset classification information 414 that shows the mindset tendencies of the personnel. Aptitude information 453 may be the personnel's answers to questions in a predetermined aptitude test, etc., used as is, or it may be used after predetermined preprocessing. In this embodiment, the personnel mindset classification information 414 is represented by a plurality of predetermined elements (elements 1 to N) and element strengths (element strengths 1 to N) that indicate the degree of suitability to them. Therefore, LLM413 outputs how strong (tendency) a person is for each element (mindset classification). The number of elements and the types of elements here are not particularly limited and may be adjusted as appropriate according to gender, age, experience, roots, etc.

[0066] For example, LLM413 derives one or more elements (which may be the same as or different from N) corresponding to culture based on the input information, and clusters them into the aforementioned multiple elements (=N). Then, LLM413 may derive the element strength of each element based on the distribution of each clustered element.

[0067] Figure 5C is a conceptual diagram illustrating the LLM 417 used to analyze the skills required by recruiting companies. The LLM 417 takes job information 432 provided by recruiting companies as input and outputs skill classification information 418 that shows the trends in the company-required skills required by said recruiting company. The job information 432 may be a predetermined number of job postings from the recruiting company as input, or it may be all of the job postings acquired from said recruiting company as input. In addition, the job information 432 may be classified based on the organizational structure and positions within the recruiting company, and skill classifications may be derived corresponding to each criterion. In this embodiment, the skill classification information 418 for recruiting companies is represented by a predetermined number of elements (elements 1 to N) and element importance scores (element importance scores 1 to N) that indicate the degree of importance of those elements. Therefore, the LLM 417 outputs how important each element (skill classification) is to a given recruiting company. The number of elements and the types of elements here are not particularly limited and may be adjusted as appropriate depending on the industry, technology field, etc. of the recruiting company. Furthermore, skills are not limited to qualifications, but may also include experience, work history, abilities, and technical skills.

[0068] Furthermore, depending on the recruiting company, it may not be possible to adequately collect job posting information 433 and company-related information 434 via social media or the internet. Alternatively, it is conceivable that only job posting information 432 may be available, and information about the company's culture may not be obtainable. In such cases, information may be collected individually from the relevant company, a culture analysis may be conducted, and then used in the transition matching process.

[0069] Furthermore, regarding the recruiting companies 430, there may be cases where different companies share the same name, or where different names actually refer to the same company. The system may be configured to handle such cases, including name matching and association. Such processing may be carried out, for example, based on publicly available company information 433 and related company information 434 collected for each recruiting company 430. More specifically, the determination may be based on information such as the head office location and business operations.

[0070] [Processing Sequence] The processing sequence according to this embodiment will be explained using Figures 6 to 8. The entity that executes each process shown in the processing sequence below executes the process in response to an external request or at a predetermined timing. Each process may be realized by the operation of the control unit of the device shown in Figure 2 as each functional unit, but here, for the sake of simplicity, the processing entity will be described comprehensively.

[0071] (Information Gathering and Analysis Processing) Figure 6 shows the processing sequence from information gathering on recruiting companies to each analysis process. This processing sequence may be executed at predetermined times or repeated periodically. In addition, if each analysis process for recruiting companies can be performed, some processes may be omitted or added, or the order of the processes may be changed.

[0072] In step S601, the collaborative system 400 provides the matching server 100 with a list of companies offering job openings. This provision may be made based on a request from the matching server 100, or it may be configured to be provided periodically. The composition of the list of companies offering job openings is not particularly limited; it may consist only of the names of the companies, or it may include company information published on company websites, as well as job information at that time. Alternatively, a list of multiple companies offering job openings may be generated and obtained by crawling publicly available websites. Examples of websites in this case include press releases regarding fundraising. Information may be collected through crawling or scraping. Furthermore, the collaborative system 400 may be configured so that the crawling targets and scope can be set in advance by an administrator or other person.

[0073] In step S602, the matching server 100 obtains a list of recruiting companies provided by the linked system 400. Note that the list of recruiting companies is not limited to being obtained from the linked system 400; it may also be set by the matching business operator managing the matching server 100 specifying any companies. In this case, the step in S601 may be omitted.

[0074] In step S603, the matching server 100 performs a search for recruiting companies based on the list of recruiting companies obtained in step S602. For example, it searches for and accesses company websites provided by the recruiting company's recruitment system 300 via the internet.

[0075] In step S604, the recruiting company system 300 provides publicly available company information. In this provision, in response to an access from the matching server 100, the system may provide information on the entire web page requested by the matching server 100, or it may provide predetermined string information or information contained in predetermined fields within the web page.

[0076] In step S605, the matching server 100 sequentially performs crawling and other operations on the information obtained as a result of accessing the network information provided by the recruiting company system 300 for each recruiting company shown in the list of recruiting companies. In this way, the matching server 100 collects the publicly available company information of each recruiting company.

[0077] In step S606, the matching server 100 accesses the external online system 500 by searching for information publicly available on the internet using the list of recruiting companies obtained in step S602. The external online system 500 may be various types of information disclosure systems, such as social networking services (SNS), summary sites, social networking sites, or review sites.

[0078] In step S607, the external online system 500 provides company-related information of the recruiting company. This provision may include the information of the entire web page requested by the matching server 100, or it may include predetermined string information written on the web page, or information contained in predetermined fields.

[0079] In step S608, the matching server 100 accesses network information provided by various external online systems 500 and collects company-related information of recruiting companies from the information obtained through this access.

[0080] The series of processes in steps S603 to S608 may be performed periodically or when a predetermined event occurs. Furthermore, steps S603 to S605 and steps S606 to S608 may be executed at different times. Each step is performed for all recruiting companies listed in the recruiting company list.

[0081] In step S609, the matching server 100 uses the collected information to analyze the corporate culture of each recruiting company shown in the list of recruiting companies. As explained using Figure 5A, the analysis of corporate culture is performed by obtaining the culture classification information of the recruiting company using LLM, etc., as described above.

[0082] In step S610, the matching server 100 performs a search for job information provided by recruiting companies based on the list of recruiting companies obtained in step S602. For example, it searches for and accesses company websites provided by the recruiting company's recruitment system 300 via the internet.

[0083] In step S611, the recruiting company system 300 provides job information. In this provision, in response to an access from the matching server 100, the system may provide information on the entire web page requested by the matching server 100, or it may provide predetermined string information or information contained in predetermined fields within the web page.

[0084] In step S612, the matching server 100 collects job information by accessing network information provided by recruiting companies. In addition to obtaining job information from the recruiting company system 300, job information may also be obtained by searching for job information published on the internet and collecting job information from the target recruiting company. Furthermore, if the list of recruiting companies obtained in step S602 contains job information, this may also be used.

[0085] The series of processes in steps S610 to S612 may be performed periodically or when a predetermined event occurs. Furthermore, each step is executed for all recruiting companies listed in the recruiting company list.

[0086] In step S613, the matching server 100 uses the collected job information to analyze the required skills of each recruiting company listed in the list of recruiting companies. As explained using Figure 5C, the analysis of required skills is performed by using LLM to obtain the skill classification information of the recruiting company.

[0087] (Matching Process) Figure 7 shows the processing sequence for matching recruiting companies with job seekers. This processing sequence is initiated, for example, when the matching server 100 receives a matching request from a job seeker. The processing sequence may also be initiated based on a request from the company. In this case, it may be executed when the recruiting company accesses the matching server 100 and makes a matching request. In this case, it is preferable that the recruiting company provides information about itself that can be used for purposes such as culture analysis in advance.

[0088] In step S701, the user terminal 200 sends a matching request to the matching server 100 based on the instructions of the operator (personnel). This matching request may be executed, for example, by accessing a web screen (not shown) provided by the matching server 100 and performing a predetermined screen operation.

[0089] In step S702, the matching server 100 responds to the matching request and requests the user terminal 200 to input various information required for the matching process. Note that the input of various information related to personnel does not need to be done all at once, and may be done as appropriate depending on the timing of processing by the matching server 100 and the content of the processing requested by the personnel. The matching server 100 displays a UI screen (not shown) for inputting various information on the user terminal 200.

[0090] In step S703, the user terminal 200 receives input of various information from the operator via the UI screen (not shown) displayed on the display unit 250. The information input here is the information included in the personnel information 451 shown in Figure 4. As described above, the input information also includes answers to questions such as aptitude tests, which are used as aptitude information 453 in the matching process. The user terminal 200 transmits the received information to the matching server 100.

[0091] In step S704, the matching server 100 acquires the information transmitted from the user terminal 200 as personnel information.

[0092] In step S705, the matching server 100 accesses the external online system 500 by searching for publicly available information on the internet using the personnel information obtained in step S704. The external online system 500 may be various types of information disclosure systems, such as social networking services (SNS), summary sites, social networking sites, or review sites.

[0093] In step S706, the external online system 500 provides personnel-related information. This provision may include the information of the entire web page requested by the matching server 100, or it may include predetermined string information or information contained in predetermined fields within the web page.

[0094] In step S707, the matching server 100 accesses network information provided by various external online systems 500 and collects personnel-related information from the information obtained through this access.

[0095] In step S708, the matching server 100 performs preprocessing on the responses to the aptitude test and other information included in the personnel information obtained in step S704. The content of this preprocessing is not particularly limited. For example, it may determine tendencies that can be identified in advance based on the answers to the questions, or it may derive information to be used to filter candidate recruiting companies. Alternatively, if the answers to the questions are used as they are in the subsequent LLM processing, this step may be omitted.

[0096] In step S709, the matching server 100 uses the collected information to perform a mindset analysis of individual employees. As explained using Figure 5B, the mindset analysis of employees is performed by obtaining mindset classification information of the employees using LLM.

[0097] In step S710, the matching server 100 performs a matching process using the culture classification information of each of the multiple recruiting companies obtained in the corporate culture analysis process in step S609 of Figure 6, and the mindset classification information of the personnel obtained in step S709. Through this process, recruiting companies that are a cultural fit for the personnel are identified and ranked.

[0098] In step S711, the matching server 100 outputs the matching results obtained in step S710 as a response to the matching request. The output targets here may be the top recruiting companies by a predetermined number, or recruiting companies with a matching degree equal to or greater than a predetermined value, or they may be specified by the operator of the user terminal 200.

[0099] In step S712, the user terminal 200 displays the matching results output from the matching server 100, which show companies that are a good cultural fit for the user.

[0100] In step S713, the user terminal 200 sends a matching request to the matching server 100 to identify a job-seeking company that is a good skill fit from among companies that are a good cultural fit, based on the operator's instructions. This matching request may be executed, for example, by performing a predetermined screen operation on a web screen (not shown) displaying job-seeking companies that are a good cultural fit. Alternatively, the operator may specify a desired job-seeking company from among several job-seeking companies that are a good cultural fit, and specify whether or not there are job openings or positions that are a good skill fit for that company.

[0101] In step S714, the matching server 100 uses the corporate skills requirements of the recruiting companies obtained in the corporate skills analysis process in step S613 of Figure 6, along with the job seeker information of the individual, to perform a matching process to identify recruiting companies that have job openings or positions that are a skill fit for the individual from among the recruiting companies shown as a result of the matching process obtained in step S711.

[0102] In step S715, the matching server 100 outputs the matching results obtained in step S714 as a response to the skill fit matching request. The output targets here may be the top recruiting companies by a predetermined number, or recruiting companies with a matching degree above a predetermined value, or they may be specified by the operator of the user terminal 200. Furthermore, the matching server 100 outputs job information that matches the skills from among the job information provided by recruiting companies that match the culture fit. If there is no job information with a high skill fit matching degree among the recruiting companies that match the culture fit, the server may output a notification to that effect.

[0103] In step S716, the user terminal 200 displays the matching results output from the matching server 100, which show companies with a culture fit and job information that matches the skills of those companies.

[0104] In step S717, the matching server 100 records the matching results, associating the highly matched individuals with the recruiting companies.

[0105] In step S718, the matching server 100 provides the information recorded in step S718 to a predetermined linkage system 400 and a recruiting company system 300 established by recruiting companies with a high degree of matching. The recipients of this information are not particularly limited, but are predetermined. Furthermore, the scope of information provided to external parties may be predetermined, for example, with the consent of the personnel.

[0106] In step S719, the collaboration system 400 records the information provided by the matching server 100. The manner in which the collaboration system 400 records the information is not particularly limited. Therefore, the collaboration system 400 may select which of the provided information to record.

[0107] In step S720, the recruiting company system 300 records the information provided by the matching server 100. The manner in which the recruiting company system 300 records the information is not particularly limited. Therefore, the recruiting company system 300 may select which of the provided information to record.

[0108] (Job Information Provision Processing) Figure 8 shows the processing sequence when providing job information from recruiting companies to a candidate. This processing sequence may be performed for a candidate, for example, when the candidate has made the desired settings and is provided with job information from recruiting companies that match the culture fit and skill fit as a result of the processing in Figure 7.

[0109] In step S801, the recruiting company system 300 provides new job information to the matching server 100. For example, a recruiting company can determine that there are candidates who are a cultural fit and skill fit for the company, based on processes such as steps S718 and S720 in the processing sequence shown in Figure 7. Based on this information, the recruiting company can then issue new job information targeting such candidates or deploy already issued job information that matches the candidates.

[0110] In step S802, the matching server 100 obtains job information provided by the recruiting company system 300.

[0111] In step S803, the matching server 100 identifies companies that have newly provided job information and individuals who are a good cultural fit and skill fit, based on the matching results obtained through the processing sequence shown in Figure 7. If there are multiple suitable individuals, multiple individuals may be targeted.

[0112] In step S804, the matching server 100 outputs the job information to the user terminal 200 used by the person identified in step S803.

[0113] In step S805, the user terminal 200 displays the job information output from the matching server 100. This job information may be displayed, for example, when the user terminal 200 accesses the matching server 100 based on the operator's instructions.

[0114] In step S806, the user terminal 200 sends an inquiry to the matching server 100 regarding the desired job information, based on the operator's instructions.

[0115] In step S807, the matching server 100 executes processing in response to inquiries from the user terminal 200. This processing may include, for example, exclusive control over inquiries from other user terminals regarding the same job posting, or control the status of inquiries.

[0116] In step S808, the matching server 100 notifies the recruiting company system 300 of the recruiting company providing the job information that it has received an inquiry regarding the job information. This notification may include information about the person who made the inquiry.

[0117] In step S809, the recruiting company system 300 receives a notification from the matching server 100.

[0118] In step S810, the recruiting company system 300 performs response processing for inquiries regarding job information. This response processing may include, for example, notifying the user terminal used by the person who made the inquiry, presenting more detailed job information, or introducing the company.

[0119] In step S811, the user terminal 200 displays the response from the recruiting company system 300 to the inquiry.

[0120] Furthermore, the inquiry regarding job information in step S806 may be configured to be made directly to the recruiting company system 300 without going through the matching server 100. Also, the response to the inquiry in step S810 may be configured to be made to the user terminal 200 via the matching server 100.

[0121] As shown in Figures 6 to 8, the entire matching process sequence according to this embodiment is executed through the cooperation of each system, with the matching server 100 at the center. Note that some or all of the processes shown in Figures 6 to 8 may be executed simultaneously in parallel, assuming multiple individuals or multiple recruiting companies.

[0122] It should be noted that the processing sequence up to Figure 7 can be executed even without a contract for personnel placement between the matching service provider that provides the matching server 100 and the recruiting company that provides the recruiting company system 300. Similarly, the processing flow shown in Figure 8 can be partially executed without such a contract, but for example, from step S808 onwards, such a contract may be required. Therefore, if there is no contract for personnel placement between the matching service provider that provides the matching server 100 and the recruiting company that provides the recruiting company system 300 at step S807, the processing may be temporarily suspended at this point, and human action may be taken to conclude such a contract. After the conclusion of the contract is completed, the processing from step S808 onwards may be performed.

[0123] [Processing Flow] Figure 9 is a flowchart showing a part of the matching process by the matching server 100 according to this embodiment, and corresponds to the processing of the matching server 100 among the processes shown in Figure 7. This processing flow is realized, for example, by the control unit 110 of the matching server 100 reading and executing programs and various data stored in the storage unit 130. The UI screen (not shown) provided by the matching server 100 is displayed on the display unit 250 of the user terminal 200, and user operations are accepted via the UI screen and instructed to the matching server 100. Here, for the sake of simplicity, the processing entity is comprehensively described as the matching server 100, instead of the various functional units of the control unit 110 shown in Figure 2.

[0124] Before this processing flow is initiated, various types of information are registered in each database in the storage unit 130 and configured to be accessible by the matching server 100. Furthermore, it is assumed that various types of information can be registered, edited, deleted, and accessed in each database.

[0125] In step S901, the matching server 100 displays a predetermined UI screen (not shown) to the user terminal 200. This UI screen may be displayed in response to a matching request from the user terminal 200. The UI screen may also display predetermined questions, questionnaires, aptitude tests, etc., and allow the operator to input them.

[0126] In step S902, the matching server 100 obtains the personnel information entered via the UI screen displayed in step S901.

[0127] In step S903, the matching server 100 uses the personnel information obtained in step S902 to search for publicly available information on the internet and other sources to collect personnel-related information. The collected personnel-related information may be information published in various forms, such as on social networking sites, summary sites, social networking sites, and review sites.

[0128] In step S904, the matching server 100 performs preprocessing on the personnel information. In this embodiment, the matching server 100 performs preprocessing using the answers to questions entered via the UI screen displayed in step S901. This processing corresponds to preprocessing for the subsequent LLM-based processing, and this step may be omitted if the same aptitude test is used. Preprocessing may involve extracting predetermined keywords from the answers, or it may involve converting the answer content into numerical values ​​for aptitude evaluation criteria based on predefined conditions.

[0129] In step S905, the matching server 100 uses the acquired information to perform a mindset analysis of individual personnel. As explained using Figure 5B, the mindset analysis of personnel is performed by obtaining mindset classification information of personnel using LLM.

[0130] In step S906, the matching server 100 performs matching based on the culture classification information of each recruiting company and the mindset classification information of the personnel. The elements included in the culture classification information of recruiting companies and the mindset classification information of personnel, as shown in Figures 5A and 5B, do not necessarily match. In such cases, for example, criteria or conversion formulas that show the correlation or similarity between the elements included in the culture classification information of recruiting companies and the elements included in the mindset classification information of personnel are defined in advance, and these are used to perform matching. Then, based on the degree of matching, the matching server 100 ranks multiple companies for the personnel and extracts recruiting companies that are a cultural fit.

[0131] In step S907, the matching server 100 displays information on job-seeking companies that are a cultural fit on the user terminal 200 as a result of the matching in step S906. If many job-seeking companies are a good fit, the server may output the top job-seeking companies by a predetermined number or job-seeking companies with a matching degree equal to or greater than a predetermined value.

[0132] In step S908, the matching server 100 determines whether it has received a matching request from the user terminal 200 for a job opening that matches the user's skills. If a matching request has been received (step S908: YES), the matching server 100 proceeds to step S909. On the other hand, if a matching request has not been received (step S908: NO), this processing flow is terminated.

[0133] In step S909, the matching server 100 targets recruiting companies that are a cultural fit for the individual, and uses the recruiting skills of the recruiting companies obtained through the recruiting skills analysis process and the individual's job-seeking information to perform a matching process to identify recruiting companies that are a skill fit for the individual from among the recruiting companies shown as a result of the matching process obtained in step S907.

[0134] In step S910, the matching server 100 displays information on companies that are a cultural fit and job postings from those companies that are a skill fit, as matching results from step S909, on the user terminal 200. If many companies are a good fit, the server may output the top companies by a predetermined number or companies with a matching degree above a predetermined value. If there are no job postings with a high skill fit matching degree among the companies that are a cultural fit, the server may output a notification to that effect. The process flow then ends.

[0135] In the above embodiment, the process described involved determining cultural fit followed by skill fit. However, the process is not limited to this; it may also be configured to determine job postings that are a skill fit for each individual, and then, based on those results, identify companies that are a cultural fit. Furthermore, the configuration may allow individuals to selectively decide which process to follow for matching.

[0136] In summary, this embodiment enhances the convenience of matching various companies with talent, enabling more appropriate matching. For example, matching service providers can analyze the culture of recruiting companies using network information, without needing to enter into prior contracts or collaborations with recruiting companies, and then match them with talent. Furthermore, it becomes possible to connect talent with recruiting companies across a large number of companies in the world, without being bound by contracts or other agreements. In addition, talent can obtain information on recruiting companies that are a cultural match for them, rather than being limited to companies managed by the registered recruitment service provider.

[0137] <Second Embodiment> A second embodiment of the present invention will now be described. Note that the same configuration as the first embodiment will not be described, and the explanation will focus on the differences. In the first embodiment, an embodiment primarily aimed at matching personnel with recruiting companies was described. In this embodiment, an embodiment primarily aimed at diagnosing what kind of corporate culture a person fits into will be described. Note that the configurations of the first and second embodiments are not mutually exclusive, and both functions may be incorporated into the matching system 1.

[0138] [Outline of Diagnostic Process] The diagnostic process by the matching server 100 according to this embodiment will be explained using Figure 10. Here, the matching business operator 1000 that manages the matching server 100, the predetermined company 1010 that serves as the diagnostic indicator, and the person to be diagnosed 1020 are shown, respectively. The predetermined company 1010 may be, for example, a well-known company that already has a certain level of recognition. In this embodiment, it is assumed that multiple companies are predetermined as the predetermined company 1010. The person to be diagnosed 1020 is a user who wishes to use the diagnostic service according to this embodiment.

[0139] In this embodiment, information relating to a single predetermined company 1010 includes publicly available company information 1011 and related company information 1012. Publicly available company information 1011 is information obtainable from the company website of the predetermined company, etc. Related company information 1012 is information collected from an external online system 500 built on a network NW such as the Internet, using company information (name, etc.). The scope of publicly available company information 1011 and related company information 1012 may be adjusted according to the settings of the predetermined company 1010.

[0140] For example, even within a single company, corporate culture can change depending on its stage and size. More specifically, a well-known company's corporate culture may have changed between its early stages as a startup and its later stages as a global company due to changes in the workforce and environment. Therefore, the publicly available company information 1011 and company-related information 1012 may be limited to information concerning a particular well-known company's early stages. This structure makes it possible to assess whether the person being assessed fits the corporate culture of a particular well-known company in its early stages. In addition, since some companies may have undergone business transformation or expansion, it may be appropriate to present information including their original business and current businesses.

[0141] Furthermore, the system handles aptitude information 1021 and personal information 1022 as information about one person to be diagnosed 1020. Aptitude information 1021 consists of answers to questions defined in advance by the matching service provider (matching server 100). The content and format of the questions are not particularly limited. Personal information 1022 is information about the person to be diagnosed 1020 that is collected from an external online system 500 built on a network NW such as the Internet. Personal information 1022 may be the same as the personnel information described in the first embodiment.

[0142] The matching service provider 1000 is configured to collect and acquire the various types of information described above. In this embodiment, two types of LLMs (LLM1001, 1003) are used. These are assumed to be pre-built and ready for use.

[0143] On the matching server 100 of the matching service provider 1000, the company's public information 1011 and company-related information 1012 of a designated company 1010 are input to the LLM 1001. The LLM 1001 then outputs the designated company's culture classification information 1002, indicating which of a predetermined group of culture classifications the designated company 1010 belongs to. The LLM 1001 in this embodiment may have the same configuration as the LLM 411 described with reference to Figure 5A in the first embodiment, or it may be different.

[0144] Furthermore, the matching server 100 inputs the aptitude information 1021 and person-related information 1022 of the person to be diagnosed 1020 into the LLM 1003. The LLM 1003 then outputs the mindset classification information 1004 of the person to be diagnosed 1020, indicating which of the predetermined set of mindset classifications the person 450 belongs to. The LLM 1003 according to this embodiment may have the same configuration as the LLM 413 described using Figure 5B in the first embodiment, or it may be different.

[0145] Furthermore, the matching server 100 performs culture matching using the culture classification information 1002 of a designated company 1010 and the mindset classification information 1004 of the person being diagnosed 1020. This culture matching is performed using a ratio or score indicating the degree to which the designated company 1010 and the person being diagnosed 1020 match in terms of cultural evaluation criteria. The matching method based on the two classification information is not particularly limited. Here, one designated company 1010 is shown, but in reality, culture matching is performed between multiple designated companies and the person being diagnosed 1020, and the designated companies 1010 are ranked according to the matching results and output as the diagnosis result 1006.

[0146] Based on the diagnostic result 1006, the matching server 100 provides the person being diagnosed 1020 with information on designated companies 1010 that have a score above a predetermined value, or designated companies 1010 that have ranked highly. The diagnostic result 1006 provided here may indicate which elements are strong and therefore a good cultural fit with the designated company 1010. For example, the original source used to determine those elements (e.g., the company's publicly available information on mission, vision, and values, or information disseminated on social media) may also be presented. Alternatively, a ranking of designated companies with an overall good cultural fit may be presented, or designated companies diagnosed as having a high degree of fit for each element included in the classification information may be presented.

[0147] It should be noted that, depending on the types and number of elements included in the classification information, biases may occur in the corporate culture of each company. For example, if we consider the corporate culture of a company in its early stages, the classification of each company's corporate culture may concentrate on similar elements such as "highly proactive" and "long working hours." As a result, differences in corporate culture between multiple companies may become less apparent.

[0148] Assuming such a scenario, this embodiment deals with an algorithm and diagnosis that separates common features from differing features in a densely clustered group of companies, and identifies the differing features. In this embodiment, the following procedure is followed for each company using the collected information.

[0149] [1] For each company, a list of expressions representing its culture is combined into a single sentence (culture-related sentence) and embedded (created as an embedded expression). Embedding itself is a well-known process, so a detailed explanation is omitted, but it is used for efficiency through conversion to a lower dimension and for extracting semantic relationships. If weighting is not applied, the expressions are simply listed. For example, strings are combined as follows to create a related sentence. Example of combination: "People who can speak their minds flatly, respect others' opinions, do work pragmatically, spend holidays alone, prioritize performance over familiarity, people"

[0150] When adding weight to a characteristic, prefixes such as "very strongly," "strongly," "somewhat," "slightly," or "slightly" may be added. In this case, it is preferable to use a prefix that sounds natural. Example combination: "People who can speak their minds strongly and objectively, people who can respect others' opinions very strongly, people who can compartmentalize their work to some extent, people who strongly prefer to spend their holidays alone, people who strongly prioritize performance over familiarity."

[0151] [2] Embed the culture-related sentences obtained above into the feature space. Note that if all the text included in the information collected as company-related information (official website, articles, SNS, word of mouth, data on various platforms, etc.) is simply combined and embedded, business descriptions and other information will be included. This is information other than information related to the company culture and will become noise when analyzing the culture. Therefore, by extracting only the information related to culture and embedding it into the feature space, a culture-specific embedding representation can be obtained. The extraction of the necessary information here may be performed by filtering based on predetermined conditions.

[0152] [3] Since Embedding represents each company as a high-dimensional vector, Principal Component Analysis (PCA) is applied to the set of vectors for all companies. Principal Component Analysis is a well-known method, and a detailed explanation is omitted here. This creates a new axis that represents all companies in a way that maximizes variability. These axes are ranked in order of their contribution rate, from the highest to the lowest, as the first principal component, second principal component, ..., the Xth principal component. Then, axes with low contribution rates are eliminated, and the data is converted to a lower-dimensional space.

[0153] [4] In the transformed low-dimensional space, clustering is performed for each company. This allows for classification that focuses most on the differences between each company. By performing this process for each of the multiple companies, the characteristics of each company can be represented in the low-dimensional space.

[0154] If we want to classify the person being assessed into one of Y clusters (corresponding to Y companies), we use the same method as described above to represent the company in the same low-dimensional space as the company, based on aptitude information and personal information. In this case, the representative point of the cluster to which each company belongs is treated as the company itself, and we determine which of the Y clusters the person being assessed belongs to, that is, which company's cluster they belong to. By representing an individual's mindset in a low-dimensional space and identifying which cluster they belong to, it becomes possible to classify each individual into the well-known company that best fits them. Using a similar method, it is also possible to classify a large number of companies into a predetermined number of clusters.

[0155] Furthermore, in order to obtain personal information about the person being diagnosed, content (text) that the individual has posted or that has been mentioned in relation to the individual, such as SNS information, may be entered into LLM and expressed as a mindset.

[0156] Furthermore, when conducting aptitude tests on individuals for diagnostic purposes, questions may be set up that relate to the axis (≒ principal component) that best facilitates the identification of each cluster. This can improve the effectiveness of increasing the efficiency of the test. In other words, the questions in aptitude tests used to obtain aptitude information may be adjusted to improve the accuracy of the diagnosis.

[0157] Examples of question design methods to improve the efficiency of the assessment include the following: 1. For each cluster, obtain a set of cultural / mindset elements common to the group of companies within that cluster. This set of cultural / mindset elements will be considered the representative set of features for that cluster. 2. Review the representative set of features for all clusters and aggregate the set of features that are included only in a specific cluster (or a small number of clusters) from all clusters. 3. Since this set of features is considered to be effective in identifying all clusters, design questions to identify them.

[0158] In addition, when extracting expressions (string information) related to culture and mindset from text information, prompts may be added to LLM to ensure that each expression is as "independent" as possible. If expressions are not independent, even similar expressions will be recognized as unrelated. Therefore, even if similar cultural and mindset expressions are included, the similarity cannot be recognized (a problem with methods such as Bag of Words), but this can be resolved. Specifically, cultural expressions such as "open communication" and "anyone can freely express their opinions" are similar to each other, so it is preferable to avoid having such similar expressions.

[0159] [Processing Flow] Figure 11 is a flowchart showing the diagnostic processing performed by the matching server 100 according to this embodiment. This processing flow is realized, for example, by the control unit 110 of the matching server 100 reading and executing programs and various data stored in the storage unit 130. The UI screen (not shown) provided by the matching server 100 is displayed on the display unit 250 of the user terminal 200, and user operations are accepted via the UI screen and instructed to the matching server 100. For the sake of simplicity, here, instead of the various functional units of the control unit 110 shown in Figure 2, the processing entity is comprehensively described as the matching server 100.

[0160] Before this processing flow is initiated, various types of information are registered in each database in the storage unit 130 and configured to be accessible by the matching server 100. Furthermore, it is assumed that various types of information can be registered, edited, deleted, and accessed in each database.

[0161] In step S1101, the matching server 100 displays a predetermined UI screen (not shown) to the user terminal 200. This UI screen may be displayed in response to a diagnostic request from the user terminal 200. The UI screen may also display predetermined questions, questionnaires, aptitude tests, etc., and allow the operator to input them.

[0162] In step S1102, the matching server 100 obtains the information entered via the UI screen displayed in step S1101.

[0163] In step S1103, the matching server 100 uses the information obtained in step S1102 to search for publicly available information on the internet and other sources to collect information about the person to be diagnosed. The information collected may be from various sources, such as social networking sites, summary sites, social networking sites, and review sites.

[0164] In step S1104, the matching server 100 performs aptitude determination processing for the person to be diagnosed. In this embodiment, the matching server 100 performs preprocessing using the answers to questions, etc., entered via the UI screen displayed in step S1101. This preprocessing step may be omitted if the answers to the aptitude test are used directly in the subsequent LLM processing. Preprocessing may involve extracting predetermined keywords from the answers, or it may involve converting the answer content into numerical values ​​of aptitude evaluation criteria based on predefined conditions.

[0165] In step S1105, the matching server 100 uses the acquired information to perform a mindset analysis of the person being diagnosed. Similar to the process described with reference to Figure 5B in the first embodiment, the mindset analysis is performed by obtaining the mindset classification information of the person being diagnosed using LLM.

[0166] In step S1106, the matching server 100 performs matching based on the culture classification information of a predetermined company and the mindset classification information of the person being diagnosed. In the first embodiment, the elements included in the culture classification information of a predetermined company and the mindset classification information of the person being diagnosed do not necessarily match, as shown in Figures 5A and 5B. Therefore, criteria and conversion formulas that show the correlation and similarity between the elements included in the culture classification information of a predetermined company and the elements included in the mindset classification information of the person being diagnosed are defined in advance, and the matching process is performed using these. Then, based on the degree of matching, the matching server 100 ranks multiple companies for the person being diagnosed and extracts and identifies predetermined companies that are a cultural fit.

[0167] In step S1107, the matching server 100 displays the matching result from step S1206 as a diagnostic result on the user terminal 200. The items displayed here are not particularly limited, but only the specific company that best matches overall based on the degree of matching of each element of the classification information may be displayed, or specific companies with a high degree of matching for each element of the classification information may be displayed. Then, this processing flow ends.

[0168] In summary, this embodiment enables individuals to identify companies that align with their mindset and obtain useful information for job searching. Furthermore, individuals can utilize the matching of their own characteristics with company cultures for further self-analysis. By also presenting well-known companies that represent each cluster, individuals and those around them can intuitively understand what kind of companies they are most likely to fit into. This increases individuals' motivation to share their diagnostic results, making it easier for the diagnostic tool to spread to those around them. As a result, a wider range of talent data can be obtained, enabling the provision of more comprehensive and optimal matching.

[0169] [Business Model Outline] Figure 12 is a schematic diagram illustrating the business model configuration based on an embodiment of the present invention. It shows a higher-level conceptual representation of the processes described in Figures 4 and 6 to 8 in the first embodiment.

[0170] The main entities involved in the process are matching service providers, recruiting companies, individuals (job seekers), and various systems on the network (SNS, etc.). Matching service providers correspond to the matching server 100 in Figure 1. Individuals correspond to the user terminals 200 in Figure 1. Recruiting companies correspond to the recruiting company systems 300 (300a to 300c) in Figure 1. The network corresponds to the collaborative system 400 and external online systems 500, which are configured to communicate via the network NW in Figure 1.

[0171] In accordance with the processing flow and sequence described above, each business process proceeds as follows. Note that the parts and processes where monetary or other incentives are given to each entity are omitted here.

[0172] (1) Matching service providers collect information about various companies (company-related information) using the internet and other means.

[0173] (2) The matching service provider collects information (publicly available company information) from the company's website or other sources.

[0174] (3) The matching service provider will use the collected information and LLM to analyze the culture of each company.

[0175] (4) The matching service provider will use the collected information and LLM to analyze the skills required by each company.

[0176] (5) Matching service providers accept matching requests from individuals to find companies that are a cultural fit for them. At this time, information about the individuals (job seeker information and suitability information) is also collected.

[0177] (6) Matching service providers collect publicly available information about personnel (publicly available personnel information) using the internet or other means.

[0178] (7) The matching service provider will use the collected information and LLM to analyze the mindset of the personnel.

[0179] (8) The matching service provider uses the analyzed mindset of the talent and the analyzed culture of each company to perform a matching process and identify companies that are a cultural fit for the talent.

[0180] (9) Matching service providers present candidates with information on companies that are a good cultural fit.

[0181] (10) Matching service providers accept matching requests to identify companies that are a good skill fit for the individuals.

[0182] (11) The matching service provider uses the skills required by each company and the information on the personnel to perform the matching process and identify companies that are a good skill fit from among companies that are a good cultural fit.

[0183] (12) Matching service providers will present candidates with information on companies that are a good cultural fit and skill fit.

[0184] (13) Matching service providers shall provide companies with information on personnel who are a good cultural fit and a good skill fit.

[0185] (14) The recruiting company shall provide the matching service provider with job information that matches the cultural fit and skill fit of the candidate.

[0186] (15) Matching service providers present job postings provided by companies to individuals who are a cultural fit and skill fit for those companies.

[0187] Traditionally, recruitment agencies (matching agencies) and recruiting companies were expected to collaborate only after a pre-established contract regarding talent placement. This process involved considerable effort and steps, and this effort increased as the number of recruiting companies grew. To secure a contract, it was necessary to conduct sales activities with each recruiting company individually and then conclude a contract. During this process, for example, the recruitment agency would obtain internal information about the recruiting company.

[0188] However, despite significant costs and effort, recruitment agencies often fail to secure contracts with hiring companies. As a result, the number of companies that recruitment agencies can potentially match candidates with remains extremely limited, even though there are potentially many companies that offer a good cultural and skill fit for the individual.

[0189] Furthermore, as is typical in the business practices of talent platforms, recruitment agencies naturally only introduce candidates to companies with which they have already signed a recruitment agreement. This is because they cannot receive incentives such as matching fees if they inform candidates that they would be a good fit for other companies. As a result, candidates are limited to being matched with companies that each agency has connections with, making it impossible for them to be matched with the most suitable company in the world.

[0190] The business model shown in Figure 12 makes it possible to solve the above-mentioned problems and achieve more appropriate talent matching. For example, regarding culture, the conventional problem of difficulty in surveying a large number of companies with which there has been no contact is solved by the above configuration. Therefore, with the above configuration, it becomes possible to analyze the culture and skills of recruiting companies over a wider range, without being bound by the business practices and frameworks of conventional recruitment agencies, and to match them with talent.

[0191] <Other Embodiments> In addition, the present invention can also be realized by supplying a program or application for realizing the functions of one or more embodiments described above to a system or device using a network or storage medium, and having one or more processors in the computer of that system or device read and execute the program.

[0192] Alternatively, one or more functions may be implemented using a circuit (for example, an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array)).

[0193] Furthermore, some or all of the above-mentioned functions may be configured as services (SaaS (Software as a Service)) provided by a system that is configured on an external network such as the internet. Additionally, these services may be configured as cloud-based services or provided by a system built within the service provider's premises.

[0194] Although various embodiments have been described above with reference to the drawings, it goes without saying that this disclosure is not limited to these examples. It will be clear to those skilled in the art that various modifications, alterations, substitutions, additions, deletions, and equivalents can occur within the scope of the claims, and these will naturally fall within the technical scope of this disclosure. Furthermore, the components of the various embodiments described above can be combined arbitrarily without departing from the spirit of the invention.

[0195] Furthermore, in this specification, expressions such as "first" and "second" are used merely for convenience to distinguish them from other elements. Therefore, they are not intended to be interpreted as limiting the scope to specific components. Accordingly, it goes without saying that if there are more components, these expressions may be reinterpreted as appropriate depending on the configuration to which they apply.

[0196] Thus, the present invention is not limited to the embodiments described above. It is also intended and within the scope of protection to be provided for the combination of each configuration of the embodiments, as well as for modifications and applications by those skilled in the art based on the description in the specification and well-known technology.

[0197] As described above, the following matters are disclosed in this specification:

[0198] (Technology 1) A matching method for matching personnel with recruiting companies, comprising: a first collection step (e.g., S603 to S608) of collecting first information about each of several recruiting companies that are accessible via the Internet (e.g., NW); a first analysis step (e.g., S609) of analyzing first characteristics indicating the culture of each of the several recruiting companies using the first information and a first large-scale language model; an acquisition step (e.g., S702 to S704) of acquiring second information provided by personnel; a second collection step (e.g., S705 to S707) of collecting third information about the personnel that are accessible via the Internet using the second information; and a second analysis step (e.g., S709) of analyzing second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model. A matching method comprising: a first matching step (e.g., S710) which identifies a recruiting company that is a cultural fit for the individual from among the multiple recruiting companies by performing a first matching process using the first and second characteristics described above. This configuration makes it possible to improve the convenience of matching various companies with individuals and to achieve more appropriate matching. Specifically, matching service providers can analyze the culture of recruiting companies using network information and then match them with individuals, without needing to enter into contracts or collaborations with recruiting companies in advance. Furthermore, it becomes possible to connect individuals with recruiting companies without being bound by contracts, etc., targeting a large number of companies in the world. In addition, individuals can obtain information on recruiting companies that are a cultural match for them, without being limited to companies managed by the registered recruitment service provider.

[0199] (Technology 2) The matching method according to Technology 1, further comprising: a third collection step (e.g., S610 to S612) for collecting job information from each of the multiple recruiting companies; a third analysis step (e.g., S613) for analyzing a third characteristic indicating the skills required by each of the multiple companies using the job information and a third large-scale language model; and a second matching step (e.g., S714) for further identifying recruiting companies that are a skill fit for the person from among recruiting companies that are a cultural fit for the person by performing a second matching process using the third characteristic and the second information. With this configuration, for example, it becomes possible to identify companies that are a skill fit for the person from among companies that are a cultural fit, based on the skills the person possesses. It also becomes possible to identify the skills required by the company based on publicly available information.

[0200] (Technology 3) The matching method according to Technology 2, further comprising a first provision step (e.g., S711, S715) of providing the person with information including at least one of the results of the first matching process and the results of the second matching process. This configuration makes it possible to provide a person with information on companies that are a cultural fit and skill fit, regardless of whether they are well-known or not.

[0201] (Technical 4) The matching method according to any one of Technical 1 to 3, further comprising a second provision step (e.g., S720) of providing the recruiting company with the personnel information of the personnel when the personnel and the job information are identified as a cultural fit by the first matching process. This configuration makes it possible, for example, to provide companies with information on personnel who are a cultural fit and to effectively secure personnel.

[0202] (Technology 5) A matching method according to any one of Technologies 1 to 4, further comprising: a receiving step (e.g., S801, S802) for receiving new job information from recruiting companies; a identifying step (e.g., S803) for identifying personnel who are a cultural fit for the recruiting company that provided the new job information, based on the results of the first matching process; and a presentation step (e.g., S804) for presenting the new job information to the identified personnel. With this configuration, for example, when new job information is provided by a company, it becomes possible to provide the new job information to personnel who have been determined to be a cultural fit for that company, thereby improving the convenience of sharing job-related information between personnel and companies.

[0203] (Technology 6) The matching method according to any one of Technology 1 to Technology 5, further comprising a company designation process (for example, S601, S602) for receiving designations from multiple recruiting companies. This configuration makes it possible, for example, to perform cultural fit matching with personnel for specific companies.

[0204] (Technology 7) The matching method described in any of Techniques 1 to 6, wherein the first information is information collected by searching the internet from at least one of the following sources: the recruiting company's official website, official SNS account, the SNS account of the recruiting company's employees, advertisements, news sites, summary sites, social networking sites, and review sites. This configuration makes it possible, for example, to utilize publicly available information on the internet for analyzing a company's culture.

[0205] (Technology 8) The matching method described in any of Techniques 1 to 7, wherein the third information is information collected by searching the internet from at least one of the personnel's SNS accounts and publicly available information. This configuration makes it possible to use, for example, publicly available information on the internet to analyze the mindset of personnel.

[0206] (Technology 9) A matching method for matching personnel with companies, comprising: a first collection step (e.g., S603 to S608) of collecting first information about each of a plurality of companies that are accessible via a network (e.g., NW); a first analysis step (e.g., S609) of analyzing first characteristics indicating the culture of each of the plurality of companies using the first information and a first large-scale language model; an acquisition step (e.g., S702 to S704) of acquiring second information provided by personnel; a second collection step (e.g., S705 to S707) of collecting third information about the personnel that are accessible via a network using the second information; a second analysis step (e.g., S709) of analyzing second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching step (e.g., S710) of identifying a company that is a cultural fit for the personnel from among the plurality of companies by performing a first matching process using the first characteristics and the second characteristics. A matching method that includes the following features. This configuration enhances the convenience of matching various companies with talent, for example, and enables more appropriate matching. Specifically, matching service providers can analyze a company's culture using network information and then match it with talent, without needing to enter into contracts or partnerships with companies in advance. Furthermore, it becomes possible to connect talent with a large number of companies in the world without being bound by contracts or other agreements. In addition, talent can obtain information on companies that are a cultural match for them, rather than being limited to companies managed by the registered recruitment service provider.

[0207] (Technology 10) A matching system (e.g., 1,100) for matching personnel with recruiting companies, comprising: a first collection unit (e.g., 113, 115) that collects first information about each of a plurality of recruiting companies accessible via the Internet (e.g., NW); a first analysis unit (e.g., 118) that uses the first information and a first large-scale language model to analyze first characteristics indicating the culture of each of the plurality of recruiting companies; an acquisition unit (e.g., 112, 114) that acquires second information provided by personnel; a second collection unit (e.g., 115) that uses the second information to collect third information about the personnel that is accessible via the Internet; a second analysis unit (e.g., 118) that uses the third information and a second large-scale language model to analyze second characteristics indicating the mindset of the personnel; and a first matching unit (e.g., 119) that performs a first matching process using the first characteristics and the second characteristics to identify a recruiting company from among the plurality of recruiting companies that is a cultural fit for the personnel. This matching system has the following features: This configuration enhances the convenience of matching various companies with talent, enabling more appropriate matching. Specifically, matching service providers can analyze the culture of recruiting companies using network information, without needing to enter into prior contracts or collaborations with recruiting companies, and then match them with talent. Furthermore, it becomes possible to connect talent with recruiting companies without being bound by contracts or other agreements, targeting a large number of companies in the world. In addition, talent can obtain information on recruiting companies that are a cultural match for them, rather than being limited to companies managed by the registered recruitment service provider.

[0208] (Technology 11) A matching system (e.g., 1,100) for matching personnel with companies, comprising: a first collection unit (e.g., 113, 115) that collects first information about each of a plurality of companies accessible via a network (e.g., NW); a first analysis unit (e.g., 118) that uses the first information and a first large-scale language model to analyze first characteristics indicating the culture of each of the plurality of companies; an acquisition unit (e.g., 112, 114) that acquires second information provided by personnel; a second collection unit (e.g., 115) that uses the second information to collect third information about the personnel that is accessible via a network; a second analysis unit (e.g., 118) that uses the third information and a second large-scale language model to analyze second characteristics indicating the mindset of the personnel; and a first matching unit (e.g., 119) that performs a first matching process using the first characteristics and the second characteristics to identify a company from among the plurality of companies that is a cultural fit for the personnel. This configuration enhances the convenience of matching various companies with talent, enabling more appropriate matches. Specifically, matching service providers can analyze a company's culture using network information and then match them with talent, without needing to enter into prior contracts or partnerships with companies. Furthermore, it becomes possible to connect talent with a large number of companies without being bound by contracts or other agreements. In addition, talent can obtain information on companies that are a cultural match for them, rather than being limited to companies managed by the registered recruitment service provider.

[0209] (Technical 12) A first collection step (e.g., S603 to S608) of collecting first information about each of several recruiting companies that is accessible via the Internet (e.g., NW) to a computer (e.g., 100); a first analysis step (e.g., S609) of analyzing first characteristics that indicate the culture of each of the several recruiting companies using the first information and a first large-scale language model; an acquisition step (e.g., S702 to S704) of acquiring second information provided by the personnel; a second collection step (e.g., S705 to S707) of collecting third information about the personnel that is accessible via the Internet using the second information; and a second analysis step (e.g., S709) of analyzing second characteristics that indicate the mindset of the personnel using the third information and a second large-scale language model. A program for performing a first matching process (e.g., S710) using the first and second characteristics described above to identify a recruiting company that is a cultural fit for the individual from among the multiple recruiting companies. This configuration makes it possible to improve the convenience of matching various companies with individuals and to achieve more appropriate matching. Specifically, matching service providers can analyze the culture of recruiting companies using network information and then match them with individuals, without needing to enter into contracts or collaborations with recruiting companies in advance. Furthermore, it becomes possible to connect individuals with recruiting companies without being bound by contracts, etc., targeting a large number of companies in the world. Individuals can also obtain information on recruiting companies that are a cultural match for them, without being limited to companies managed by the registered recruitment service provider.

[0210] (Technical 13) A first collection step (e.g., S603 to S608) of collecting first information about each of several companies that are accessible to a computer (e.g., 100) via a network (e.g., NW); a first analysis step (e.g., S609) of analyzing first characteristics that represent the culture of each of the several companies using the first information and a first large-scale language model; an acquisition step (e.g., S702 to S704) of acquiring second information provided by personnel; a second collection step (e.g., S705 to S707) of collecting third information about the personnel that are accessible via a network using the second information; a second analysis step (e.g., S709) of analyzing second characteristics that represent the mindset of the personnel using the third information and a second large-scale language model; and a first matching step (e.g., S710) of identifying a company that is a cultural fit for the personnel from among the several companies by performing a first matching process using the first characteristics and the second characteristics. A program to execute this. This configuration makes it possible to improve the convenience of matching various companies with talent, for example, and to achieve more appropriate matches. Specifically, matching service providers can analyze a company's culture using network information and then match it with talent, without needing to enter into contracts or partnerships with companies in advance. It also makes it possible to connect talent with companies from a large number of companies in the world, without being bound by contracts, etc. Furthermore, talent can obtain information on companies that are a cultural match for them, rather than being limited to companies managed by the registered recruitment service provider.

[0211] Although various embodiments have been described above with reference to the drawings, it goes without saying that the present invention is not limited to these examples. It is clear to those skilled in the art that various modifications or alterations can be conceived within the scope of the claims, and these will naturally also fall within the technical scope of the present invention. Furthermore, the components of the above embodiments may be combined in any way without departing from the spirit of the invention.

[0212] This application is based on the Japanese Patent Application No. 2024-189190 filed on October 28, 2024, the contents of which are incorporated by reference within this application.

[0213] The present invention is useful, for example, as an apparatus, system, or method for improving the convenience of matching personnel with recruiting companies and achieving more appropriate matching.

[0214] 1...Matching System 100...Matching Server 110...Control Unit 111...Data Management Unit 112...Display Control Unit 113...Recruiting Company Information Acquisition Unit 114...Personnel Information Acquisition Unit 115...Information Collection Unit 116...LLM Management Unit 117...Question Management Unit 118...Culture Analysis Unit 119...Culture Matching Unit 120...Skill Analysis Unit 121...Skill Matching Unit 122...Diagnosis Unit 123...Communication Control Unit 130...Storage Unit 131...Company Information DB 132...Personnel Information DB 133...Peripheral Information DB 134...Classification Information DB 135...Question Information DB 136...Recruiting Information DB 200...User Terminal 210...Control Unit 220...Storage Unit 230...Communication Unit 240...Operation Unit 250...Display Unit 260...External Interface 300...Recruiting Company System 400...Integrated systems 500...External online systems NW...Network

Claims

1. A matching method for matching personnel with recruiting companies, comprising: a first collection step of collecting first information about each of a plurality of recruiting companies that are accessible via the internet; a first analysis step of analyzing first characteristics indicating the culture of each of the plurality of recruiting companies using the first information and a first large-scale language model; an acquisition step of acquiring second information provided by personnel; a second collection step of collecting third information about the personnel that are accessible via the internet using the second information; a second analysis step of analyzing second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching step of identifying recruiting companies that are a cultural fit for the personnel from among the plurality of recruiting companies by performing a first matching process using the first characteristics and the second characteristics.

2. The matching method according to claim 1, further comprising: a third collection step of collecting job information from each of the multiple recruiting companies; a third analysis step of analyzing a third characteristic indicating the skills required by each of the multiple companies using the job information and a third large-scale language model; and a second matching step of further identifying recruiting companies that are a skill fit for the person from among recruiting companies that are a cultural fit for the person by performing a second matching process using the third characteristic and the second information.

3. The matching method according to claim 2, further comprising a first provision step of providing the personnel with information including at least one of the results of the first matching process and the results of the second matching process.

4. The matching method according to claim 1, further comprising a second provision step of providing the recruiting company with the personnel information of the personnel if the personnel and the job information are identified as a cultural fit through the first matching process.

5. The matching method according to claim 1, further comprising: a receiving step of receiving new job information from recruiting companies; a identifying step of identifying personnel who are a cultural fit for the recruiting company that provided the new job information, based on the results of the first matching process; and a presentation step of presenting the new job information to the identified personnel.

6. The matching method according to claim 1, further comprising a company designation step for receiving designations from the plurality of recruiting companies.

7. The matching method according to claim 1, wherein the first information is information collected by searching the internet from at least one of the recruiting company's official website, official SNS account, the SNS account of the recruiting company's employees, advertisements, and news sites, summary sites, social networking sites, and review sites.

8. The matching method according to claim 1, wherein the third information is information collected by searching the internet from at least one of the personnel's SNS accounts and publicly available information.

9. A matching method for matching personnel with companies, comprising: a first collection step of collecting first information about each of a plurality of companies that are accessible via a network; a first analysis step of analyzing first characteristics indicating the culture of each of the plurality of companies using the first information and a first large-scale language model; an acquisition step of acquiring second information provided by personnel; a second collection step of collecting third information about the personnel that are accessible via a network using the second information; a second analysis step of analyzing second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching step of identifying a company from among the plurality of companies that is a cultural fit for the personnel by performing a first matching process using the first characteristics and the second characteristics.

10. A matching system for matching personnel with recruiting companies, comprising: a first collection unit that collects first information about each of a plurality of recruiting companies accessible via the Internet; a first analysis unit that analyzes first characteristics indicating the culture of each of the plurality of recruiting companies using the first information and a first large-scale language model; an acquisition unit that acquires second information provided by personnel; a second collection unit that collects third information about the personnel that is accessible via the Internet using the second information; a second analysis unit that analyzes second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching unit that identifies a recruiting company from among the plurality of recruiting companies that is a cultural fit for the personnel by performing a first matching process using the first characteristics and the second characteristics.

11. A matching system for matching personnel with companies, comprising: a first collection unit that collects first information about each of a plurality of companies accessible via a network; a first analysis unit that analyzes first characteristics indicating the culture of each of the plurality of companies using the first information and a first large-scale language model; an acquisition unit that acquires second information provided by personnel; a second collection unit that collects third information about the personnel accessible via a network using the second information; a second analysis unit that analyzes second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching unit that identifies companies from among the plurality of companies that are a cultural fit for the personnel by performing a first matching process using the first characteristics and the second characteristics.

12. A program for a computer to perform the following steps: a first collection step of collecting first information about each of several recruiting companies that are accessible via the Internet; a first analysis step of analyzing first characteristics indicating the culture of each of the several recruiting companies using the first information and a first large-scale language model; an acquisition step of acquiring second information provided by a person; a second collection step of collecting third information about the person that is accessible via the Internet using the second information; a second analysis step of analyzing second characteristics indicating the mindset of the person using the third information and a second large-scale language model; and a first matching step of identifying a recruiting company that is a cultural fit for the person from among the several recruiting companies by performing a first matching process using the first characteristics and the second characteristics.

13. A program for causing a computer to perform the following steps: a first collection step of collecting first information about each of several companies that are accessible via a network; a first analysis step of analyzing first characteristics indicating the culture of each of the several companies using the first information and a first large-scale language model; an acquisition step of acquiring second information provided by personnel; a second collection step of collecting third information about the personnel that are accessible via a network using the second information; a second analysis step of analyzing second characteristics indicating the mindset of the personnel using the third information and a second large-scale language model; and a first matching step of identifying a company from among the several companies that is a cultural fit for the personnel by performing a first matching process using the first characteristics and the second characteristics.

Images