Talent marketplace system using generative ai and talent marketplace program using generative ai

Abstract

The present invention assists decision-making of a recruiter with more comprehensive and multilateral evaluation and recommendation of human resources based on various input data. Provided is a TMP system 1 for assisting matching between a recruiter and an applicant, said TMP system 1 comprising: an atypical information formatting unit 20 which outputs, via a generative AI 2, history and performance data 22 of each human resource from atypical human resource data 21; an evaluation item setting unit 30 which extracts, via the generative AI 2, evaluation items of each human resource on the basis of job offer data 31 and outputs the evaluation items as a skill definition 32; and a recommendation process unit 40 which selects, via the generative AI 2, recommendation items from evaluation items of the skill definition 32, evaluates, via the generative AI 2, each human resource on the basis of history and performance data 22 in accordance with an evaluation criterion that is set for each of the selected recommendation items, generates, via the generative AI 2, a reason for recommendation for a human resource to be recommended, and outputs the reason together with a result of the evaluation.

Description

Talent marketplace system using generative AI and talent marketplace program using generative AI

[0001] The present invention relates to a technology that contributes to human capital management in a company, and in particular to a technology that is effective when applied to a talent marketplace system and a talent marketplace program that realizes a talent marketplace.

[0002] With a declining productive population, it is essential for Japanese companies to utilize their human capital with maximum efficiency. In other words, it is important for management to maximize the value of "human resources" and increase corporate value over the medium to long term by linking the human resources strategy, in which the HR department procures and develops the human resources necessary to execute the business strategy, with the business strategy, in which each business division formulates a business strategy that utilizes the strengths of its human resources, while maintaining a company-wide overview (management strategy).

[0003] However, in today's VUCA (Volatility, Uncertainty, Complexity, Ambiguity) world, where it is difficult to predict the future, business strategies are changing and the mobility and diversity of human resources is increasing, making it extremely difficult for the traditional centralized decision-making structure led by management and the human resources department to respond to both types of changes.

[0004] Therefore, companies are considering creating an "internal talent market" system (talent marketplace) that uses market principles to match and connect the tasks necessary to carry out business strategies with the abilities, achievements, and will of each individual employee.This system will allow for a flexible link between talent strategies and business strategies, thereby improving on-site engagement and productivity.

[0005] As a technology for determining assignments by matching a recruiter (employer) with an applicant (job seeker), for example, Japanese Patent Laid-Open No. 2004-62270 (Patent Document 1) describes a technique for comparing the recruiter's response information selected by the recruiter from a set of multiple options with the job seeker's response information selected by the job seeker for each of a number of categories common to both the recruiter and the job seeker, calculating the evaluation points of the job seeker from the recruiter's perspective and the evaluation points of the recruiter from the job seeker's perspective, and matching the recruiter and the job seeker based on these points.

[0006] Japanese Patent Application Laid-Open No. 2004-62270

[0007] According to the conventional technology described in Patent Document 1, it is possible to match applicants with job seekers and create assignment plans, but this is a simple matching method that assists the human resources department in creating a tentative assignment plan, and there are limitations to determining the appropriateness and effectiveness of the matching results.

[0008] To enable more appropriate and effective matching in the talent marketplace, it is necessary to support recruiters' decision-making with more comprehensive and multifaceted talent evaluations and recommendations. To achieve this, it is necessary to diversify the data input to the talent marketplace and to establish a system that standardizes evaluations and recommendations by setting multifaceted evaluation criteria.

[0009] Therefore, an object of the present invention is to provide a talent marketplace system and a talent marketplace program that support recruiters' decision-making by providing more comprehensive and multifaceted evaluations and recommendations of talent based on a variety of input data.

[0010] The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

[0011] Among the inventions disclosed in this application, the outline of representative inventions will be briefly explained as follows.

[0012] A talent marketplace system that is a representative embodiment of the present invention is a talent marketplace system that supports matching between recruiters and applicants, and includes: an unstructured information formatting unit that uses a generation AI to extract information on the career and achievements of each applicant from unstructured information related to each of the applicants and output it as career and achievement data; an evaluation item setting unit that uses a generation AI to extract skill information that will be used as evaluation items for each of the applicants based on job information and output it as a skill definition; and a recommendation processing unit that uses the generation AI to select recommended items to be evaluated with priority for the job related to the recruiter from the evaluation items of the skill definition, evaluates each of the selected recommended items based on the career and achievement data in accordance with evaluation criteria set for each of the selected recommended items, and generates a reason for the recommendation for the recommended talent using the generation AI and outputs it together with the evaluation results.

[0013] The present invention can also be applied to a program that causes a computer to operate as the above-mentioned talent marketplace system.

[0014] The effects obtained by the representative inventions disclosed in this application can be briefly explained as follows.

[0015] In other words, according to a representative embodiment of the present invention, it becomes possible to support the decision-making of recruiters by providing more comprehensive and multifaceted evaluations and recommendations of personnel based on a variety of input data.

[0016] The present invention relates to a talent marketplace system, a talent management system, and a talent recommendation system.

[0017] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In all drawings used to explain the embodiments, the same parts are generally designated by the same reference numerals, and repeated explanations will be omitted. However, parts that have been designated and explained in one drawing may be referred to by the same reference numerals in the explanation of other drawings, although they will not be shown again.

[0018] <Overview> The Talent Marketplace (hereinafter sometimes referred to as "TMP") system, which is one embodiment of the present invention, is an information processing system that matches, for example, multiple internal job postings within a company (the departments, divisions, groups, etc. that are the recruiters (hereinafter sometimes collectively referred to as "recruiters")) with applicants for these positions.

[0019] In this embodiment, in addition to standard information such as the applicant's job title, qualifications, and skills, various non-standard information (text, images, audio, and other information; in this embodiment, it will be described as text information) that has previously been unused and exists within the company is used as input data during matching, making it possible to utilize a variety of input data. Non-standard information includes, for example, matching and personnel-related information such as the applicant's career history and self-promotional statements, and the recruiter's job requirements, as well as various types of text information such as plans, proposals, reports, and chats created by the applicant in the course of their work.

[0020] In order to handle the input of such non-standard information (text information), this embodiment utilizes generative AI (Artificial Intelligence), which has been increasingly used in recent years, such as GhatGPT (registered trademark), to evaluate and select promising talent that matches the recruitment requirements based on the non-standard information using a variety of items. Then, along with quantitative evaluation, qualitative evaluation such as the reasons for recommendation is output as text, thereby supporting the recruiter's decision-making through more comprehensive and multifaceted talent evaluation and recommendation.

[0021] Figure 2 is a diagram outlining an example of talent evaluation and recommendation in a talent marketplace (TMP) system, which is an embodiment of the present invention. In this example, in response to the recruiting requirements for a specific job, the talent applicants (shown on the left side of the diagram) first mechanically filter a list of talent candidates based on standard information such as job title, qualifications, and skills to narrow down the list to a certain number of candidates.

[0022] Then, evaluation items for evaluating personnel from multiple angles are set using as input various standard and non-standard information such as the career history of these personnel, evaluations by superiors, and proposals and reports created in the course of work, as well as standard and non-standard information such as recruitment requirements and job data from the recruiting side. These evaluation items may be set manually by a user such as a recruiting staff member or a human resources staff member, but in this embodiment, the generation AI2 sets them automatically by using non-standard information. The evaluation items proposed by the generation AI2 may also be manually revised to improve practicality and effectiveness.

[0023] Then, for each evaluation item that has been set, the generation AI 2 evaluates, for example, on a five-point scale based on the content described in the standard and non-standard information, and selects promising candidates based on the evaluation. For each of these candidates, the AI ​​outputs the evaluation along with the reasons and grounds for the evaluation as a written statement. For example, promising candidates are recommended to the recruiting party in the form of an "evaluation / recommendation letter" on the right side of the figure.

[0024] <System Configuration> Figure 1 is a diagram showing an overview of an example configuration of a talent marketplace (TMP) system according to one embodiment of the present invention. The TMP system 1 may be implemented, for example, by an information processing terminal such as a personal computer (PC) or a tablet terminal, or by a server device or a virtual server built on a cloud computing service. While the diagram shows a single logical information processing system, the functions may be physically distributed across multiple information processing terminals, servers, etc.

[0025] The TMP system 1 realizes various functions related to TMP construction by, for example, using a CPU (Central Processing Unit) not shown in the figure to execute middleware such as an OS (Operating System), DBMS (DataBase Management System), and web server program that are deployed onto memory from a recording device such as an HDD (Hard Disk Drive) or SSD (Solid State Drive), as well as software that runs on top of them.

[0026] This TMP system 1 has, for example, various units implemented as software, such as a filtering unit 10, an unstructured information shaping unit 20, an evaluation item setting unit 30, and a recommendation processing unit 40. These units may be implemented as a single application program as a whole, or each unit may be implemented as an independent program that works together.

[0027] The filtering unit 10 has the function of inputting information from a talent list 11 of applicants obtained from personnel information and application information along with the recruitment requirements for the target job, filtering and narrowing down the candidates in advance based on standard information such as job title and qualifications, and outputting a list of candidate talents 12.

[0028] The non-standard information formatting unit 20 has the function of inputting non-standard personnel data 21, which is various non-standard information (text) such as proposals and reports related to the candidate personnel 12, and automatically extracting the career history, achievements, etc. using the generation AI 2, formatting the content of the career history and achievements data 22 into a predetermined format, and outputting it as career / achievement data 22. By automatically writing the content of the career / achievement data 22 using the generation AI 2, applicants can save time and effort, and by standardizing the non-standard information to a certain extent, it can be made easier to use.

[0029] Furthermore, by defining the skill definition 32 described later in advance and automatically extracting the career history, achievements, etc. in accordance with this, it is possible to extract the information in a form that matches the information that the recruiter is looking for. This helps to address the problem that applicants have when trying to highlight their career history and achievements in an application, such as "I don't know what to write about," and the problem that recruiters have when evaluating the applicant's career history and achievements, such as "I can't judge what they didn't write about."

[0030] The evaluation item setting unit 30 has the function of dynamically creating and outputting a skill definition 32 that lists all the skills (evaluation items) required in each job data 31, using job data 31 from multiple recruiters as input. Skills required in a large number of job data 31 can be said to be skills in high demand. For example, the skill definition 32 may be automatically created by the generation AI 2 using the job data 31 as input, taking into account the purpose and content of the recruitment, and the content may then be manually adjusted and customized by a human resources department, etc. Characteristics and positioning, such as "skills that can be learned quickly" and "skills that take time to learn," or weighting, such as importance, may be set. When creating the skill definition 32, existing skill definitions may be used as initial values ​​to reflect existing initiatives, management policies, etc.

[0031] The recommendation processing unit 40 has a function of inputting the skill definition 32 and career and performance data 22 related to each candidate 12, selecting an evaluation axis consisting of skills to be analyzed with a focus on the recruitment requirements related to the target job from the skills (evaluation items) in the skill definition 32 using the generation AI 2, evaluating each candidate 12 based on the selected evaluation axis using the generation AI 2, and outputting promising candidates based on the evaluation results as recommended candidates 43. When outputting the recommended candidates 43, for example, the generation AI 2 may generate the strengths and weaknesses of the target candidate and the support that the department to which the candidate belongs should provide, and outputting this as a sentence such as the "evaluation / recommendation letter" shown in FIG. 2 in a form that is understandable to the recruiter, thereby encouraging a soft landing of the match.

[0032] When selecting evaluation axes, for example, the recruiter may subjectively adjust or revise the evaluation axes objectively selected by the generation AI 2 according to the requirements of the business or project. For each selected evaluation axis, the recruiter can set the evaluation criteria (the criteria used to read the personnel data) according to the circumstances of the business or project. For example, a five-point evaluation could be set, such as "5: Deep knowledge and experience that can lead, 4: Ready to contribute immediately, 3: Experience and can put into practice with training, 2: Knowledge, 1: No particular aptitude."

[0033] When evaluating each candidate 12, the generation AI 2 objectively and thoroughly reads a large amount of standardized and unstandardized information about each candidate 12 according to the evaluation axes and evaluation criteria. From the evaluation results for each candidate 12, a skill map 42 can be obtained that shows how many candidates have what level of each skill (evaluation item) in the skill definition 32, allowing for skill-based talent inventory.

[0034] When outputting recommended talent 43 based on the evaluation results of each candidate talent 12, adding not only quantitative numerical values ​​but also text explaining the reason for the recommendation (i.e., a summary of the candidate talent 12's career based on the evaluation criteria) can help the recruiter to have a qualitative understanding when evaluating and judging the recommended talent 43 and to explain it to others (without having to go and obtain raw data).

[0035] 3 is a flowchart outlining an example of the process flow for recommending talent in one embodiment of the present invention. First, the filtering unit 10 inputs information from a talent list 11 of applicants along with the recruitment requirements for the target job, filters it based on standard information such as job titles and qualifications, and outputs candidate talent 12 (step S01). Next, the unstructured information formatting unit 20 inputs proposals, reports, and other unstructured talent data 21 related to the candidate talent 12, automatically extracts career history and achievements using the generation AI 2, formats them into a predetermined format, and outputs the career history and achievements data 22 (step S02).

[0036] Thereafter, the evaluation item setting unit 30 inputs job data 31 from multiple recruiters, and dynamically creates or updates a skill definition 32 that lists all the skills (evaluation items) required in each job data 31 using the generation AI 2 and manually, and outputs it (S03).

[0037] Then, the recommendation processing unit 40 inputs the skill definition 32 and the career and performance data 22 for each candidate 12, and selects, by the generation AI 2 and manually, evaluation axes (recommended items) from the evaluation items in the skill definition 32 to focus on analyzing the recruitment requirements for the target job (S04). Then, evaluation criteria for scoring are set for each selected evaluation axis (S05), and the generation AI 2 scores and evaluates each candidate 12 based on each evaluation axis (recommended items) (S06). Furthermore, for promising candidates, a landing support (comprehensive review) is performed to generate information such as the reason for the recommendation, the strengths and weaknesses of the candidate, and the support that the candidate's department should provide to the candidate (S07). These are then output as a recommended candidate 43 in the form of an "evaluation / recommendation letter" as shown in the example of Figure 2, and the process ends.

[0038] As described above, the TMP system 1 according to one embodiment of the present invention can evaluate and select promising personnel who match the recruitment requirements based on a variety of non-standard information (text information) using a variety of multifaceted items by utilizing the generation AI 2. Furthermore, along with quantitative evaluations such as scoring, qualitative evaluations such as the reasons for recommending the personnel can be output as text, thereby supporting the decision-making of recruiters through more comprehensive and multifaceted evaluations and recommendations of personnel.

[0039] The invention made by the inventor has been specifically described above based on the embodiments, but it goes without saying that the present invention is not limited to the above embodiments and can be modified in various ways without departing from the spirit of the invention. Furthermore, the above embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those having all of the described configurations. Furthermore, it is possible to add, delete, or replace part of the configuration of the above embodiments with other configurations.

[0040] Note that the above-described configurations, functions, processing units, processing means, etc. may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. Furthermore, the above-described configurations, functions, etc. may be implemented in software by a processor interpreting and executing a program that implements each function. Information such as the program, table, and file that implements each function can be stored in a storage device such as a memory, hard disk, or SSD, or in a storage medium such as an IC card, SD card, or DVD.

[0041] In addition, in the above figures, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily show all the control lines and information lines that are actually implemented. In reality, it can be assumed that almost all components are interconnected.

[0042] The present invention can be used in a talent marketplace system and a talent marketplace program that realizes a talent marketplace.

[0043] 1...Talent Marketplace (TMP) system, 2...Generation AI, 10...Filtering unit, 11...Talent list, 12...Candidate talent, 20...Atypical information shaping unit, 21...Atypical talent data, 22...Career and achievement data, 30...Evaluation item setting unit, 31...Recruitment data, 32...Skill definition, 40...Recommendation processing unit, 42...Skill map, 43...Recommended talent

Claims

1. A talent marketplace system for assisting in matching recruiters and applicants, comprising: an unstructured information shaping unit that extracts information on the work experience and achievements of each talent from the unstructured information related to each talent of the applicants and outputs it as work experience / achievement data by means of a generative AI; an evaluation item setting unit that extracts information on skills that are evaluation items for each talent by means of a generative AI based on job information and outputs it as a skill definition; and a recommendation processing unit that selects, by means of a generative AI, recommendation items to be evaluated with emphasis in the cases related to the recruiters from the evaluation items of the skill definition, evaluates each of the selected recommendation items according to the evaluation criteria set for each, evaluates each talent by means of a generative AI based on the work experience / achievement data, generates, by means of a generative AI, the reasons for the recommendation for the talent to be recommended, and outputs the result together with the evaluation result. A talent marketplace system.

2. The talent marketplace system according to claim 1, wherein the evaluation item setting unit accepts input for revising the skill definition extracted by means of a generative AI. A talent marketplace system.

3. The talent marketplace system according to claim 1, wherein the recommendation processing unit accepts input for revising the recommendation items selected by means of a generative AI. A talent marketplace system.

4. The talent marketplace system according to claim 1, wherein the recommendation processing unit generates, by means of a generative AI, information including the strengths / weaknesses of the talent to be recommended and advice for the recruiter for the talent to be recommended, and outputs the result together with the evaluation result. A talent marketplace system.

5. The talent marketplace system according to claim 1, further comprising a filtering unit that filters each talent of the applicants in advance based on the structured information of each talent. A talent marketplace system.

6. A talent marketplace program for operating a computer as a talent marketplace system that supports matching between recruiters and applicants, the computer performing the steps of: extracting, by a generation AI, information on the experience and achievements of each talent from the unstructured information related to each talent of the applicants and outputting it as experience / achievement data; extracting, by a generation AI, information on skills that are evaluation items for each talent based on job information and outputting it as a skill definition; selecting, by a generation AI, recommended items to be evaluated with emphasis in the case related to the recruiter from the evaluation items of the skill definition; evaluating, by a generation AI, each talent based on the experience / achievement data according to the evaluation criteria set for each of the selected recommended items; and generating, by a generation AI, a reason for recommendation for the talent to be recommended and outputting it together with the result of the evaluation. A talent marketplace program that executes these steps.

Images