Talent matching program
The human resources matching support program uses AI to objectively evaluate talent suitability by summarizing skill and project information, reducing user input and subjectivity in talent matching processes.
Patent Information
- Application Number
- JP2024088292
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-05-30
- Publication Date
- 2025-12-11
AI Technical Summary
Existing human resources matching systems require users to input information in a predetermined format, which can be cumbersome and subjective, leading to inefficiencies in talent matching.
A human resources matching support program that utilizes a generation AI to extract and summarize skill and project information, generating evaluation statements to objectively assess talent suitability for projects.
Reduces user burden and personal subjectivity by providing data-driven talent matching, enabling efficient and accurate recruitment of suitable personnel based on objective evaluation results.
Smart Images

Figure 2025180753000001_ABST
Abstract
Description
[Technical Field]
[0001] The present invention relates to a human resources matching program that supports human resources matching. [Background technology]
[0002] Conventionally, there has been known a device that assists recruiters seeking personnel in their search for personnel (see, for example, Patent Document 1). The device described in Patent Document 1 acquires recruiting response information including the specialty and personality of the prospective hire to be assigned by the assignment department (the recruiting side) and job-seeker response information including the specialty and personality of the prospective hire (the job-seeker side), compares the acquired recruiting response information with the job-seeker response information, calculates evaluation points from the recruiting side and evaluation points from the job-seeker side, and matches the assignment department with the prospective hire based on the calculated evaluation points. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2004-062270 Summary of the Invention [Problem to be solved by the invention]
[0004] However, the device described in Patent Document 1 requires users to input answers in a predetermined format to a plurality of items set for matching, which may be cumbersome for the user. [Means for solving the problem]
[0005] A human resources matching support program, which is one embodiment of the present invention, causes a computer to execute the following steps: a first process of inputting a first prompt to a generation AI instructing the AI to extract skill information from human resources information, including skill information related to the skills possessed by the human resources, and generate a first summary sentence that summarizes the extracted skill information; a second process of inputting a second prompt to the generation AI instructing the AI to extract content information from project information, including content information related to the content of the project, and generate a second summary sentence that summarizes the extracted content information; and a third process of inputting a third prompt to the generation AI instructing the AI to generate an evaluation sentence that evaluates the suitability of the human resources as project personnel from the first summary sentence and the second summary sentence. [Effects of the Invention]
[0006] According to the present invention, it is possible to perform talent matching based on data, not on personal subjectivity, while reducing the burden on the user. [Brief explanation of the drawings]
[0007] [Figure 1] 1 is a diagram for explaining an organization to which a human resources matching support program according to an embodiment of the present invention is applied; [Figure 2] FIG. 2 is a diagram for explaining human resource information created based on reports from each of the human resources in FIG. 1; [Figure 3] FIG. 2 is a diagram for explaining project information created based on the declaration from the project organizer in FIG. 1. [Figure 4] 1 is a block diagram showing an example of the overall configuration of a talent matching support system according to an embodiment of the present invention. [Figure 5] 5 is a flowchart showing an example of processing by a human resources matching support program executed by the computer of FIG. 4. [Figure 6] FIG. 5 is a diagram for explaining an example of the first summary sentence of FIG. 4. [Figure 7] FIG. 5 is a diagram for explaining an example of the second summary sentence of FIG. 4. [Figure 8A] 5 is a diagram for explaining an example of the evaluation statement and evaluation result of FIG. 4. [Figure 8B] 5 is a diagram for explaining another example of the evaluation statement and evaluation result of FIG. 4. [Figure 9] FIG. 5 is a diagram for explaining an example of a prompt input from the computer in FIG. 4 to the generation AI. [Figure 10] A diagram for explaining another example of a prompt input from the computer in Figure 4 to the generation AI. DETAILED DESCRIPTION OF THE INVENTION
[0008] Hereinafter, an embodiment of the present invention will be described with reference to Figures 1 to 10. A talent matching support program according to an embodiment of the present invention is a program for supporting talent matching carried out within an organization such as a company using a computer.
[0009] Fig. 1 is a diagram for explaining an organization Gr1 to which a talent matching support program according to an embodiment of the present invention is applied. As shown in Fig. 1, the organization Gr1 is divided into a plurality of groups Gr2 according to the work content, and each group Gr2 includes a plurality of employees (talent) P. Each talent P acquires new skills and improves the level of skills they already possess while performing their work.
[0010] Each employee P is evaluated at every predetermined evaluation period by a personnel evaluator M, such as their immediate superior. In the personnel evaluation, each employee P declares their performance during the evaluation period, and an evaluation interview is held between each employee P and the personnel evaluator M based on the declared results. For the evaluation interview, an interview sheet prepared based on the declaration of each employee P, for example, an interview sheet prepared by each employee P for the evaluation interview, is used. The interview sheet prepared for the evaluation interview is managed by organization Gr1 as the personnel information of each employee P.
[0011] Fig. 2 is a diagram for explaining the human resource information 1. As shown in Fig. 2, the human resource information 1 includes information such as the name and other identification information of each human resource P, affiliation, job content, specific achievements, etc. The achievements information includes information such as the specific work performance content undertaken during the evaluation period, self-evaluation of that, good actions as specific achievements, and actions to be improved in the future.
[0012] Such performance information reflects the skills and levels possessed by each talent P, such as the skills utilized by each talent P in carrying out their work, newly acquired skills, and skills whose levels have been improved. In other words, specific skills are directly described, or specific skills and their levels can be naturally inferred from the description. Therefore, it can be said that talent information 1 includes skill information regarding the skills possessed by each talent P and their levels.
[0013] By comparing the skill information of multiple personnel P, it is possible to identify skills possessed by each personnel P that others do not possess or skills that are relatively superior (higher level) than others, i.e., the strengths of each personnel P. Therefore, it can be said that the skill information contains strengths information regarding the strengths of each personnel P.
[0014] The human resource information 1 may include information in checklist format as skill information. In this case, the burden on each human resource P when creating the human resource information 1 (interview sheet) is reduced. Furthermore, by multiple human resources P creating the human resource information 1 (skill information) using a checklist with a standardized format, the strengths of each human resource P can be accurately identified. For example, a checklist that complies with indicators such as the IT Skills Standard (ITSS) (registered trademark) can be used.
[0015] In organization Gr1 in Figure 1, when a time-limited task or new task (hereinafter referred to as a "project") is undertaken for a specific purpose separate from regular work, it is necessary to search for a suitable talent P for the project from among multiple talents P. For example, when a specific project is proposed by project organizer H, such as a manager who runs organization Gr1, and specific objectives, duration, etc. are set as project content at a kick-off meeting or similar, a search is made for suitable talent P according to the project content. Information about the project content, such as minutes of the kick-off meeting, created based on declarations from project organizer H, is managed by organization Gr1 as project information for each project.
[0016] Fig. 3 is a diagram for explaining the project information 2. As shown in Fig. 3, the project information 2 includes information for identifying the project, such as the name, and information about the content of the project, such as the specific objectives and period. The project information 2 may be, for example, minutes of a kick-off meeting, but may be in any other format as long as it is prepared based on a declaration from the project organizer H and includes the objectives of the project, etc.
[0017] When a project organizer H who has knowledge of project information 2 consults with a personnel evaluator M who has knowledge of the personnel information 1 of each personnel P to search for personnel P, bias in personnel selection may occur depending on the relationship between the project organizer H and the personnel evaluator M, or the burden of searching for personnel P may become excessive. Therefore, in this embodiment, a personnel matching support system is configured as follows, by utilizing the personnel information 1 and project information 2 managed by organization Gr1, so that the burden of searching for personnel P can be reduced while appropriate personnel P for the project can be found.
[0018] Fig. 4 is a block diagram showing an example of the overall configuration of a talent matching support system 100 according to an embodiment of the present invention. As shown in Fig. 4, the talent matching support system 100 includes a computer 10 and a generation AI 20. The computer 10 and the generation AI 20 are communicatively connected to each other via a communication network including a LAN (local area network) or a WAN (wide area network) such as the Internet.
[0019] The computer 10 is, for example, a personal computer managed by the organization Gr1, and includes a processor such as a CPU, memories such as ROM and RAM, and other peripheral circuits. A talent matching support program according to an embodiment of the present invention is stored in the computer 10, and the talent matching support program is executed by the computer 10. The talent information 1 and project information 2 may be stored and managed in the computer 10, or may be stored and managed in a server or the like managed by the organization Gr1, and acquired by the computer 10 when the talent matching support program is executed.
[0020] The generation AI 20 is a large-scale language model (LLM), and when instructions or questions (prompts) are input in natural language, it outputs a response in natural language. The generation AI 20 can be, for example, a conversational AI such as ChatGPT.
[0021] 5 is a flowchart showing an example of processing by the talent matching support program executed by the computer 10. When the talent matching support program is launched, in step S1, the computer 10 inputs a first prompt to the generation AI 20 to instruct it to generate a first summary 1A that extracts and summarizes skill information from talent information 1 (first processing). Next, in step S2, the computer 10 obtains the first summary 1A generated by the generation AI 20.
[0022] FIG. 6 is a diagram illustrating an example of a first summary 1A. In the example of FIG. 6, the first summary 1A includes, along with the identification information and affiliation of each employee P, strengths information regarding the skills and their levels possessed by each employee P, which has been extracted, summarized, and itemized from employee information 1, such as an interview sheet. The first summary 1A also includes comments on the skills identified as strengths of each employee P among the itemized skills, and comments summarizing the characteristics of each employee P. In this way, the first summary 1A contains content similar to the introduction that a personnel evaluator M would write when introducing employee P, such as his or her subordinate, to a project organizer H as a candidate for a project.
[0023] 5, the computer 10 inputs a second prompt to the generation AI 20 (second process) to instruct the generation AI 20 to generate a second summary sentence 2A that extracts and summarizes content information from the project information 2. Then, in step S4, the computer 10 obtains the second summary sentence 2A generated by the generation AI 20.
[0024] Fig. 7 is a diagram illustrating an example of the second summary 2A. As shown in Fig. 7, the second summary 2A includes information for identifying the project as well as summary information about the outline of the project content that is extracted and summarized from project information 2 such as minutes of meetings.
[0025] In step S5 of Figure 5, the computer 10 inputs a third prompt to the generation AI 20 to instruct it to generate an evaluation statement 3 that evaluates the suitability of each person P as a project member from the first summary statement 1A and the second summary statement 2A (third process). The computer 10 also inputs a fourth prompt to the generation AI 20 to instruct it to generate an evaluation result that indicates the degree of suitability from the generated evaluation statement 3 (fourth process). Next, in step S6, the computer 10 acquires the evaluation statement 3 and the evaluation result generated by the generation AI 20.
[0026] 8A and 8B are diagrams illustrating an example of the evaluation statement 3 and the evaluation result. As shown in Fig. 8A and 8B, the evaluation statement 3 includes information for identifying the project, the identification information and affiliation of each human resource P, comments evaluating the suitability of each human resource P as a member of the project, and an evaluation result indicating the degree of suitability.
[0027] The comments evaluating suitability include comments evaluating at least one of the degree of match with the project objectives and the degree of contribution to the project that are expected for each talent P. More specifically, comments are included from the perspective of whether or not the skills of each talent P match the skills required to achieve the project objectives and whether or not they are at a level required to promote the project.
[0028] In evaluation statement 3 in Figure 8A, among the skills of A (Human Resource P), the skills necessary to achieve the project's objectives and to promote the project are listed, and positive comments are added. The evaluation result states that A (Human Resource P) is highly suitable as a project member. In addition, the comment states that the skills required to achieve the project's objectives (skills related to generative AI in the example in Figure 8A) should be confirmed to ensure that they are at the level necessary to promote the project.
[0029] In evaluation statement 3 in Figure 8B, negative comments are added pointing out that A (Personnel)'s skills do not match the skills required to achieve the project's objectives, and the evaluation result states that A (Personnel) is not suitable as a project member.
[0030] In this way, by executing the talent matching support program (FIG. 5) according to an embodiment of the present invention, it is possible to generate a first summary 1A, a second summary 2A, and an evaluation 3 from talent information 1 (FIG. 2) and project information 2 (FIG. 3) managed by organization Gr1. In this case, when searching for talent P to be a member of a project, the talent P, their talent evaluator M, and the project organizer H do not incur the burden of inputting new information.
[0031] By viewing the evaluation statement 3 and the evaluation results, the project organizer H can easily and accurately judge the suitability of each talent P as a member of the project, and can recruit the appropriate talent P as a member of the project. In this case, for example, for talent P with high evaluation results, the project organizer H can refer to the first summary statement 1A or interview the talent P's affiliated personnel evaluator M. Furthermore, by viewing the evaluation statement 3 and the first summary statement 1A and second summary statement 2A used in the evaluation, the talent P who has been recruited as a member of the project can easily and accurately understand the reason for their recruitment and then decide whether or not to accept the recruitment.
[0032] The first summary 1A, the second summary 2A, and the evaluation 3 are generated as natural sentences that resemble the exchanges that occur when the project organizer H consults with the personnel evaluator M to search for talent P (Figs. 6 to 8B). This makes them easy to read and can be used in a natural manner by the project organizer H and each talent P.
[0033] 9 and 10 are diagrams illustrating an example of a prompt input from the computer 10 to the generation AI 20. FIG. 9 is an example of a prompt corresponding to the first prompt. The prompt illustrated in FIG. 9 is written to include predetermined items such as "Prerequisites and Roles," "Details of the Current Task," "Description of the Data Source," and "Notes and Points to Note." Personnel information 1 for each person P is inserted in the "Data Source" field. FIG. 10 is an example of a prompt corresponding to the third and fourth prompts. The prompt illustrated in FIG. 10 is written to include predetermined items such as "Prerequisites and Roles," "Details of the Current Task," "Description of the Data Source," "Matching Rate Judgment Criteria," and "Notes and Points to Note." A second summary sentence 2A is inserted in the "Project Data" field, and a first summary sentence 1A is inserted in the "Personal Profile Data" field. As shown in Figures 9 and 10, by using systematic prompts containing predetermined items, it is possible to accurately obtain answers containing the necessary information (first summary sentence 1A, second summary sentence 2A, evaluation sentence 3) regardless of the format of the human resources information 1 or project information 2.
[0034] According to this embodiment, the following effects can be achieved. (1) The human resources matching support program causes the computer 10 to execute the following steps (Figures 4 and 5): a first process (step S1) of inputting a first prompt to the generation AI 20 instructing the AI 20 to generate a first summary 1A by extracting and summarizing skill information from human resources information 1, which includes skill information on the skills possessed by each human resource P; a second process (step S3) of inputting a second prompt to the generation AI 20 instructing the AI 20 to generate a second summary 2A by extracting and summarizing content information from project information 2, which includes content information on the content of the project; and a third process (step S5) of inputting a third prompt to the generation AI 20 instructing the AI 20 to generate an evaluation 3 that evaluates the suitability of each human resource P as a project member from the first summary 1A and the second summary 2A.
[0035] In this way, first summary 1A, second summary 2A, and evaluation 3 are generated from personnel information 1 (Figure 2) and project information 2 (Figure 3) managed by organization Gr1, thereby reducing the burden on each personnel P, personnel evaluator M, and project organizer H and enabling personnel matching based on data rather than on personal subjectivity. Furthermore, by viewing evaluation 3, project organizer H can easily and accurately determine the suitability of each personnel P as project personnel and recruit appropriate personnel P as project personnel. Furthermore, by viewing evaluation 3 and the first summary 1A and second summary 2A used in the evaluation, personnel P who have been recruited as project personnel can easily and accurately understand the reason for their recruitment and then decide whether or not to accept the recruitment.
[0036] (2) Human Resources Information 1 is information created based on reports from each individual P (Figure 2). Project Information 2 is information created based on reports from Project Organizer H (Figure 3). Each individual P, such as an employee of Organization Gr1, undergoes a personnel evaluation at each designated evaluation period. In the personnel evaluation, each individual P reports their performance during the evaluation period, and personnel information 1, such as an interview sheet, is created based on the reports and managed by Organization Gr1. Using such personnel information 1 when searching for project personnel reduces the burden on each individual P and the personnel evaluator M, and enables the search for appropriate personnel P from an impartial perspective, regardless of the relationship between Project Organizer H and the personnel evaluator M. Furthermore, project information 2 in any format, such as minutes of the kick-off meeting, created based on reports from Project Organizer H is also managed by Organization Gr1. Using such project information 2 when searching for project personnel reduces the burden on Project Organizer H.
[0037] (3) The talent matching support program further causes the computer 10 to execute a fourth process (step S5) in which a fourth prompt is input to the generation AI 20 to instruct the generation AI 20 to generate an evaluation result indicating the degree of suitability from the evaluation statement 3 (FIG. 5). In this case, the project organizer H can grasp the suitability of each talent P as a project member from the evaluation result. Furthermore, by prioritizing talent P indicated by the evaluation result as having a high degree of suitability, the burden on the project organizer H to search for talent P can be reduced even when searching for project members from among many talents P.
[0038] (4) The skill information includes information in a checklist format. In this case, the burden on each talent P when creating talent information 1 such as an interview sheet can be reduced. Also, by having multiple talents P create talent information 1 (skill information) using a checklist in a unified format, the suitability of each talent P as a project member can be evaluated from a more fair perspective.
[0039] (5) The skill information further includes strength information about skills that each person P possesses that are relatively superior to those of others, and the first summary 1A includes the strength information in itemized form (Figure 6). In this case, the project organizer H and the people P who have been recruited as project members can easily understand the strengths of each person P.
[0040] (6) The second summary 2A includes summary information about the project content (Figure 7). In this case, the project organizer H and the personnel P who have been recruited as project personnel can easily understand the project overview.
[0041] (7) Evaluation statement 3 includes an evaluation statement that evaluates at least one of the expected degree of conformance of each talent P to the project's objectives and the degree of contribution to the project (Figs. 8A and 8B). In this case, the project organizer H can easily and accurately determine the suitability of each talent P as a member of the project. Furthermore, talent P who has been recruited as a member of the project can easily and accurately understand the reason for their recruitment.
[0042] (8) Personnel P is an employee, and personnel information 1 is information that is periodically created for the personnel evaluation of each employee (Figure 2). By utilizing the existing personnel information 1 managed by organization Gr1 when searching for project personnel, the burden on each personnel P and personnel evaluator M can be reduced, and appropriate personnel P can be searched for from an impartial perspective, regardless of the relationship between project organizer H and personnel evaluator M.
[0043] In the above embodiment, an example was described in which a personnel evaluation interview sheet was used as the personnel information 1, but the personnel information is not limited to this. The personnel information may be any information that includes skill information related to the skills possessed by each personnel. For example, the personnel information may be a document in which a personnel P who is willing to participate in a project other than normal work describes his / her skills and the working hours allocated to the project, or a document created to apply for a specific project.
[0044] In the above embodiment, an example was described in which project personnel were searched for from among human resources P belonging to a single organization Gr1, but human resources are not limited to those belonging to a single organization, and projects are not limited to those within a specific organization Gr1.
[0045] The above description is merely an example, and the present invention is not limited to the above-described embodiment and modifications as long as the features of the present invention are not impaired. One or more of the above-described embodiment and modifications can be arbitrarily combined, and modifications can also be combined with each other. [Explanation of symbols]
[0046] 1 Human Resources Information, 1A First Summary, 2 Project Information, 2A Second Summary, 3 Evaluation, 10 Computer, 20 Generative AI, 100 Human Resources Matching Support System, Gr1 Organization, Gr2 Group, H Project Organizer, M Human Resources Evaluator, P Human Resources
Claims
1. a first process of inputting a first prompt to the generation AI to instruct the AI to extract and summarize skill information from human resource information including skill information related to skills possessed by the human resource; a second process of inputting a second prompt to the generation AI instructing the AI to generate a second summary by extracting content information from project information including content information relating to the content of the project and summarizing the content information; a third process of inputting a third prompt to the generation AI to instruct the AI to generate an evaluation statement evaluating the suitability of the person as a member of the project from the first summary statement and the second summary statement; A human resources matching support program characterized by causing a computer to execute the above.
2. 2. The human resources matching support program according to claim 1, The human resource information is information created based on a declaration from the human resource, The human resources matching support program is characterized in that the project information is information created based on a declaration from the organizer of the project.
3. 2. The human resources matching support program according to claim 1, A human resources matching support program characterized in that the computer further executes a fourth process of inputting a fourth prompt to the generation AI instructing it to generate an evaluation result indicating the degree of suitability from the evaluation statement.
4. 3. The human resources matching support program according to claim 2, The human resources matching support program is characterized in that the skill information includes information in a checklist format.
5. 2. The human resources matching support program according to claim 1, The skill information further includes strength information regarding skills that the human resource possesses that are relatively superior to those of others, A human resources matching support program, characterized in that the first summary includes the strengths information in itemized form.
6. 2. The human resources matching support program according to claim 1, The human resources matching support program is characterized in that the second summary further includes summary information regarding an outline of the content of the project.
7. 2. The human resources matching support program according to claim 1, A talent matching support program characterized in that the evaluation statement includes comments evaluating at least one of the degree to which the talent is expected to meet the project's objectives and the degree to which the talent will contribute to the project.
8. 3. The human resources matching support program according to claim 2, The human resources are employees, The human resources matching support program is characterized in that the human resources information is information that is periodically created for the personnel evaluation of the human resources.
Citation Information
Patent Citations
Personnel matching trial method and personnel matching trial system
JP2004062270A