Talent Profile Creation Support Program

Abstract

To provide a talent profile creation support program that generates fair, relative profiles that are easy to compare. [Solution] The personnel profile creation program is a personnel profile creation support program that operates a computer equipped with a memory unit and a processor, and inputs a first prompt to the generating AI, causing it to refer to a RAG constructed based on skill sheets entered by multiple members belonging to at least two or more hierarchical organizational structures stored in the memory unit, and to generate a first natural sentence indicating the strengths of one member compared to multiple members within the organization, inputs a second prompt to the generating AI, causing it to refer to the RAG, and to generate a second natural sentence indicating the strengths of one member compared to multiple members in a higher hierarchical level than the organization, and inputs a third prompt to the generating AI, causing it to synthesize the first natural sentence and the second natural sentence to generate profile information for one member.

Description

[Technical Field]

[0001] The present invention relates to a personnel profile creation support program for creating member profiles. [Background Art]

[0002] When selecting appropriate personnel as members to implement business or development projects in companies or organizations, most cases have conventionally relied heavily on the intuition and experience of project managers and the like, resulting in variation among individuals and a heavy burden, and technical solutions to this problem have been demanded. To address such issues, for example, Patent Document 1 discloses a technique that, when creating an assignment plan by matching recruiters and job seekers, can automatically create a tentative plan through a simple method, wherein the technique compares recruiter-side response information with job seeker-side response information, calculates a first evaluation point serving as an evaluation point of the job seeker from the recruiter's perspective and a second evaluation point serving as an evaluation point of the recruiter from the job seeker's perspective, and performs matching. [Prior Art Documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Unexamined Patent Publication No. 2004-62270 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] However, with the technique described in Patent Document 1, it is inherently difficult to create the recruiter-side response information that is selected by the recruiter. For this reason, it is difficult for the recruiter to understand why a matched job seeker is matched. Therefore, there has been a demand for creating fairly relatively evaluated profiles that are easy for project managers and the like to compare.

[0005] The present invention aims to provide a talent profile creation support program that creates fair and relative profiles that are easy to compare. [Means for solving the problem]

[0006] (1) A human resource profile creation program according to one aspect of the present invention is a human resource profile creation support program that operates a computer (e.g., a human resource profile creation device 1 described later) comprising a memory unit (e.g., a memory unit 12 described later) and a processor (e.g., a control unit 11 described later), wherein the program inputs a first prompt to a generation AI (e.g., a generation AI 3 described later), causes it to refer to a RAG constructed based on skill sheets input from multiple members belonging to at least two or more hierarchical organizations stored in the memory unit, and generates a first natural sentence indicating the strengths of one member compared to multiple members within its own organization; inputs a second prompt to the generation AI, causes it to refer to the RAG, and generates a second natural sentence indicating the strengths of one member compared to multiple members in a higher hierarchical level than its own organization; and inputs a third prompt to the generation AI, causes it to synthesize the first natural sentence and the second natural sentence to generate profile information of one member.

[0007] According to (1) above, it becomes possible to create a fair and relative profile that is easy for project managers and others to compare.

[0008] (2) In the talent profile creation program described in (1) above, the information in the skill sheet used to generate the first natural sentence is professional skills, and the information in the skill sheet used to generate the second natural sentence is personal skills.

[0009] According to (2) above, it becomes possible to create a fair and relative profile that is easy for project managers and others to compare. [Effects of the Invention]

[0010] According to the present invention, it is possible to create a fair and relative profile that is easy to compare. [Brief explanation of the drawing]

[0011] [Figure 1] This is a schematic diagram of a personnel profile creation system, which is an embodiment of the present invention. [Figure 2] This is a block diagram showing the basic configuration of a traffic information provision device, which is an embodiment of the present invention. [Figure 3] This figure shows an example of a first prompt, which is an embodiment of the present invention. [Figure 4] This figure shows an example of a first natural sentence, which is an embodiment of the present invention. [Figure 5] This figure shows an example of a second prompt, which is an embodiment of the present invention. [Figure 6] This figure shows an example of profile information, which is an embodiment of the present invention. [Figure 7] This is a flowchart showing the operation of a talent profile creation system, which is an embodiment of the present invention. [Modes for carrying out the invention]

[0012] The present invention's talent profile creation program functions as a talent profile creation device by running a computer equipped with a memory unit and a processor. It inputs a first prompt to a generating AI and causes it to refer to a RAG (Retrieval Augmented Generation) constructed based on skill sheets entered by multiple members belonging to at least two or more hierarchical organizational structures stored in the memory unit, thereby generating a first natural sentence that shows the strengths of one member compared to multiple members within its own organization. The talent profile creation program then inputs a second prompt to the generating AI and causes it to refer to the RAG, thereby generating a second natural sentence that shows the strengths of one member compared to multiple members in a higher hierarchy. The talent profile creation program then inputs a third prompt to the generating AI and generates profile information for one member by combining the first and second natural sentences. This allows the talent profile creation program to generate fair, relative profiles that are easy to compare.

[0013] A preferred embodiment of the personnel profile creation device of the present invention will be described below with reference to the figures. Figure 1 shows the basic configuration of the personnel profile creation system 100. As shown in Figure 1, the personnel profile creation system 100 comprises a personnel profile creation device 1, a communication network 60, client terminals 2 (information processing devices such as mobile terminals, tablet terminals, and PCs), and a generation AI 3.

[0014] Client terminal 2 is an information processing device such as a PC, mobile device, or tablet device, and is used by each member, such as an employee, belonging to an organization such as a company that has at least two or more hierarchical structures. For example, when client terminal 2 receives a personnel search instruction from a member such as a project manager belonging to a higher level of the organization, it outputs the personnel search instruction to personnel profile creation device 1.

[0015] The talent profile creation apparatus 1 is a computer, a server, or the like, and operates as a talent profile creation apparatus by executing a talent profile creation program. The talent profile creation apparatus 1 previously acquires and stores skill sheets including professional skills (e.g., program development, accounting and taxation, legal affairs, etc.) and personal skills (e.g., hobbies, languages, place of origin, etc.) of members belonging to an organization via each member's client terminal 2. The talent profile creation apparatus 1 inputs a first prompt to the generative AI 3, causes the generative AI 3 to refer to a RAG constructed based on skill sheets respectively input by a plurality of members belonging to the stored organization, and causes the generative AI 3 to generate and store a first natural sentence indicating the strengths of one said member obtained by comparing the member with a plurality of members within the organization, and stores the first natural sentence. Furthermore, the talent profile creation apparatus 1 inputs a second prompt to the generative AI 3, causes the generative AI 3 to refer to the RAG, and causes the generative AI 3 to generate and store a second natural sentence indicating the strengths of one member obtained by comparing the member with a plurality of members in an upper hierarchy such as project managers, and stores the second natural sentence. Then, when the talent profile creation apparatus 1 receives a person search instruction from the client terminal 2, it inputs a third prompt to the generative AI 3, and causes the generative AI 3 to combine the first natural sentence and the second natural sentence to generate profile information of one member. The generative AI 3 outputs the generated profile information to the client terminal 2. By doing so, the talent profile creation apparatus 1 can create fairly and relatively evaluated profiles that are easy to compare. The outline of the system configuration in the present embodiment has been described above.

[0016] Next, the generative AI 3 will be described. The generative AI 3 is, for example, a computer, web server, or the like arranged on a network that is known to those skilled in the art. The generative AI 3 receives a prompt of a request sentence from the talent profile creation apparatus 1, which will be described later, and inputs the prompt into a trained language model (for example, a Large Language Model (LLM) such as ChatGPT (registered trademark)). The generative AI 3 refers to a RAG constructed in advance based on skill sheets stored in the talent profile creation apparatus 1, and generates data conforming to the request for the prompt (for example, a first natural sentence indicating the strengths of one member obtained by comparing a plurality of members within the own organization, a second natural sentence indicating the strengths of one member obtained by comparing a plurality of members in an upper hierarchy, profile information of one member obtained by combining the first natural sentence and the second natural sentence, etc.). The generative AI 3 outputs the generated text data to the talent profile creation apparatus 1 and the client terminal 2. Note that the trained language model (for example, a large language model) may be, for example, a learning model trained using a large amount of data on the Web as learning data, and in particular, the learning model may be a learning model provided from the outside. Further, the trained language model may be a learning model obtained by performing additional learning related to a specialized field such as a processing program on the provided learning model, for example, learning by prompt and / or fine tuning, transfer learning, or the like.

[0017] Next, the talent profile creation apparatus 1 will be described. As shown in FIG. 2, the talent profile creation apparatus 1 is configured to include a control unit 11, a storage unit 12, a communication unit 13, a display unit 14, and an input unit 15.

[0018] The control unit 11 is constituted by an arithmetic processing device such as a microprocessor, and controls each part constituting the talent profile creation apparatus 1. Details of the control unit 11 will be described later.

[0019] The storage unit 12 is constituted by a semiconductor memory or the like, and includes a program storage unit 120, an organization information storage unit 121, and a skill sheet storage unit 122. The program storage unit 120 includes control programs called firmware and operating systems, as well as a personnel profile creation program. By executing the personnel profile creation program on the control unit 11, the control unit 11 functions as the first natural language generation unit 110, the second natural language generation unit 111, and the profile generation unit 112. The organizational information storage unit 121 stores organizational information, including information about organizations such as organization names, the types of work each organization is responsible for, and the relationships (business connections) between organizations. The skill sheet storage unit 122 stores skill sheets entered by each member of the organization, for example, via the client terminal 2. The skill sheet includes, for example, information about the organization to which the member belongs, the member's professional skills (e.g., program development, accounting and tax, legal affairs, etc.), and personal skills (e.g., hobbies, languages, place of origin, etc.). Furthermore, the memory unit 12 may store a RAG, which serves as the knowledge base for the generated AI 3, that has been pre-constructed based on the skill sheets stored in the skill sheet memory unit 122.

[0020] The memory unit 12 has been described above, but for example, the organizational information memory unit 121 and the skill sheet memory unit 122 may be provided by a server device (not shown) connected to the communication network 60.

[0021] The communication unit 13 has a DSP, etc., and complies with standards such as LTE (Long Term Evolution), 4G (4th Generation), 5G (5th Generation), and Wi-Fi (registered trademark), enabling wireless and wired communication with external systems via the communication network 60. The communication network 60 is realized by networks such as the Internet or a mobile phone network, or a network combining these. A LAN (Local Area Network) may also be included as part of the network.

[0022] The display unit 14 is composed of a display device such as a liquid crystal display or an organic electroluminescent panel. The display unit 14 displays an image in response to instructions from the control unit 11. The input unit 15 consists of input devices (not shown) such as physical switches called numeric keypads and a touch panel mounted on top of the display surface of the display unit 14.

[0023] Next, the details of the control unit 11 will be explained. The control unit 11 is composed of a microprocessor having a CPU, RAM, ROM, and I / O. The CPU executes the personnel profile creation program read from the ROM or storage unit 12. During execution, it reads information from the RAM, ROM, and storage unit 12, writes information to the RAM and storage unit 12, and exchanges signals with the communication unit 13. In this way, the personnel profile creation process in this embodiment is realized through the cooperation of hardware and software (program). As mentioned above, if a server device (not shown) connected to the communication network 60 is equipped with an organizational information storage unit 121 and a skill sheet storage unit 122, the personnel profile creation device 1 may perform input / output with the storage unit of the server device (not shown) via the communication network 60.

[0024] As shown in Figure 2, the control unit 11 includes a first natural language generation unit 110, a second natural language generation unit 111, and a profile generation unit 112 as functional blocks.

[0025] The first natural language generation unit 110 inputs a first prompt to the generation AI 3 and causes it to refer to the RAG constructed based on skill sheets entered by multiple members belonging to at least two or more hierarchical organizational structures stored in the skill sheet storage unit 122, thereby generating a first natural language sentence that shows the strengths of one member compared to multiple members within its own organization.

[0026] Figure 3 shows an example of the first prompt. The first prompt may, for example, be pre-set and stored in the memory unit 12. As shown in Figure 3, the first prompt has the following items: "#Prerequisites / Role", "#Details of this task", "#Description of the data source", "#Notes / Points to note", and "#Data source below". The item "#Prerequisites / Role" describes the prerequisites (e.g., HR personnel) and role (e.g., skill classification, introductory text creation) set for generating AI3 to generate the first natural language sentence based on the first prompt. The item "#Task Details" will state that, based on the first prompt, the first natural sentence to be generated will be a personal profile of each member belonging to an organization such as a company, based on their professional skills, highlighting their strengths. The item "#Data Source Description" contains a description of each member's skill sheet stored in the skill sheet storage unit 122, which is necessary for the generating AI3 to generate the first natural language sentence of the individual profile. The section "#Notes and Precautions" contains notes and precautions regarding the handling of data sources and the content of the first natural sentence. The item "#Data Source Below" is appended with a RAG constructed based on the skill sheets stored in the skill sheet storage unit 122.

[0027] Specifically, the first natural language generation unit 110 inputs, for example, a first prompt to the generation AI 3. The generation AI 3 refers to the RAG attached to the item "#Data Source Below" in the first prompt and generates a first natural language sentence (personal profile) that shows the strengths of each member by comparing multiple members within its own organization. The first natural language generation unit 110 retrieves the generated first natural language sentence (personal profile) from the generation AI 3. Figure 4 shows an example of a first natural sentence. The first natural sentence (personal profile) shown in Figure 4 describes the skills (strengths) that characterize an individual member when compared with other members belonging to the same organization, and the reasons for this. The first natural language generation unit 110 stores the first natural language text (personal profile) of each member that it has acquired in the storage unit 12.

[0028] The second natural language generation unit 111 inputs a second prompt to the generation AI 3, causes it to refer to the RAG, and generates a second natural language sentence that shows the strengths of one member when comparing multiple members in a higher hierarchy than its own organization. Figure 5 shows an example of a second prompt. The second prompt may, for example, be pre-set and stored in the memory unit 12. As shown in Figure 5, the second prompt, like the first prompt in Figure 3, includes the following items: "#Prerequisites / Roles", "#Details of this Task", "#Description of Data Source", "#Notes / Points to Note", and "#Data Source Below". The item "#Prerequisites / Role" contains the prerequisites (e.g., HR person) and role (e.g., skill sorting, introductory text creation, etc.) that are set for generating AI3 to generate the second natural language sentence based on the second prompt, similar to the first prompt in Figure 3. The item "#Task Details" will be described as a second natural language statement generated based on the second prompt, which will show the strengths of each member in the upper ranks, such as project managers, within an organization such as a company, based on their personal skills. The item "#Data Source Description" contains a description of each member's skill sheet stored in the skill sheet storage unit 122, which is necessary for the generating AI 3 to generate the second natural language sentence of the personal profile, similar to the first prompt in Figure 3. The item "#Notes and Precautions" contains notes and precautions regarding the handling of the data source and the content of the second natural language sentence, similar to the first prompt in Figure 3. The item "#Data Source Below" is appended with a RAG constructed based on the skill sheet stored in the skill sheet storage unit 122, similar to the first prompt in Figure 3.

[0029] Specifically, the second natural language generation unit 111 inputs, for example, the second prompt shown in Figure 5 to the generation AI 3. The generation AI 3 refers to the RAG attached to the item "#Data Source Below" in the second prompt and generates a second natural language (personal profile) that shows the strengths of each member by comparing multiple members in the higher hierarchy. Here, the second natural sentence (personal profile) describes the skills (strengths) and reasons that characterize a single member by comparing them with each member belonging to a different higher-level organization than in the case of the first natural sentence in Figure 4. The second natural language generation unit 111 retrieves the generated second natural language text (personal profile) from the generation AI 3. The second natural language generation unit 111 stores the retrieved second natural language text (personal profile) for each member in the storage unit 12.

[0030] For example, when the profile generation unit 112 receives a person search instruction from the client terminal 2, it inputs a third prompt to the generation AI 3 and generates profile information for one member by combining the first natural language sentence and the second natural language sentence. Specifically, when the profile generation unit 112 receives a person search instruction from a client terminal 2, such as a project manager, to search for a member suitable for the project, it reads a pre-configured third prompt from the storage unit 12 that instructs the generation of profile information for one member by combining the first natural language sentence and the second natural language sentence. The profile generation unit 112 adds the first natural language sentence and the second natural language sentence to the read third prompt and inputs it to the generation AI 3. The generating AI3 synthesizes the first and second natural language sentences based on the third prompt to generate profile information for one member. The generating AI3 outputs the generated profile information to client terminal 2, such as the project manager. Client terminal 2 displays the received profile information on a display device (not shown) that it has. Figure 6 shows an example of profile information. As shown in Figure 6, it is preferable that the generated profile information also includes text that describes the person's character. By doing so, project managers and others can easily determine from multiple perspectives whether or not a candidate is suitable for the project. In Figure 6, the profile generation unit 112 caused the generation AI 3 to generate profile information for one member. However, it is also possible to have the generation AI 3 generate a list of profile information for multiple members and output it to the client terminal 2. In this case, it is preferable that the profile information in the list be arranged in order of members who possess skills suitable for the project. The configuration of each functional unit of the computer used to run the personnel profile creation program exemplified in this embodiment has been described above.

[0031] Next, the operation of this embodiment will be described with reference to the flowchart shown in Figure 7. Figure 7 is a flowchart showing the operation of the personnel profile creation system 100.

[0032] In step S10, the first natural language generation unit 110 reads the first prompt from the memory unit 12 and inputs the first prompt to the generation AI 3.

[0033] In step S11, the generating AI 3 refers to the RAG attached to the first prompt input in step S10 and generates a first natural sentence (personal profile) that shows the strengths of each member by comparing multiple members within its own organization. The generating AI 3 outputs the generated first natural sentence (personal profile) to the personnel profile creation device 1, and the personnel profile creation device 1 stores the received first natural sentence (personal profile) in the storage unit 12.

[0034] In step S12, the second natural language generation unit 111 reads the second prompt from the memory unit 12 and inputs the second prompt to the generation AI 3.

[0035] In step S13, the generating AI3 refers to the RAG attached to the second prompt input in step S12 and generates a second natural sentence (personal profile) that shows the strengths of each member by comparing multiple members of the higher hierarchy. The generating AI3 outputs the generated second natural sentence (personal profile) to the personnel profile creation device 1, and the personnel profile creation device 1 stores the received second natural sentence (personal profile) in the storage unit 12.

[0036] In step S14, when the profile generation unit 112 receives a person search instruction from the client terminal 2, it reads the third prompt from the storage unit 12 and inputs the third prompt to the generation AI 3.

[0037] In step S15, the generating AI3 generates profile information for one member by combining the first natural language sentence and the second natural language sentence based on the third prompt input in step S14. The generating AI3 outputs the generated profile information to the client terminal 2.

[0038] In step S16, client terminal 2 receives and displays the profile information generated in step S15.

[0039] As described above, a computer (personnel profile creation device 1) that runs the personnel profile creation program of the present invention on a processor can create a fair and relative profile that is easy to compare.

[0040] Each of the devices can be implemented using hardware, software, or a combination thereof. Furthermore, the method for creating a human resource profile, which is performed through the collaborative efforts of each device included in the human resource profile creation system 100 described above, can also be implemented using hardware, software, or a combination thereof. Here, implementation by software means implementation through a computer reading and executing a program.

[0041] Programs can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memory (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (random access memory)). Programs may also be supplied to a computer using various types of transient computer-readable media. Examples of transient computer-readable media include electrical signals, optical signals, and electromagnetic waves. Transitory computer-readable media can be supplied to a computer via wired communication channels such as electric wires and optical fibers, or via wireless communication channels.

[0042] While the embodiments described above are preferred embodiments of the present invention, the scope of the present invention is not limited to these embodiments, and various modifications can be made to implement the invention without departing from the spirit of the invention.

[0043] <Variation> In the embodiment described above, the personnel profile creation device 1 is implemented using a single computer or the like. However, it may also be implemented as a distributed processing system, where each function of the personnel profile creation device 1 is distributed among multiple server devices as appropriate. Specifically, for example, as mentioned above, the organizational information storage unit 121 and the skill sheet storage unit 122 may be stored and managed on server devices different from the personnel profile creation device 1. Alternatively, the functions of the personnel profile creation device 1 may be implemented using virtual server functionality or similar features on the cloud. [Explanation of Symbols]

[0044] 100 Talent Profile Creation System 1. Personnel Profile Creation Device 10. Talent Profile Creation Program 11 Control Unit 110 1st natural sentence generation part 111 Second natural sentence generation part 112 Profile Generation Unit 12 Storage section 120 Program Storage Unit 121 Organization information storage unit 122 Skill Sheet Storage Unit 13 Communications Department 14 Display section 15 Input section 2. Client terminals (information processing terminals) 60 Communication Networks

Claims

1. A personnel profile creation support program that operates a computer equipped with a memory unit and a processor, The generating AI is given a first prompt and is made to refer to a RAG constructed based on skill sheets entered by multiple members belonging to at least two or more hierarchical organizational structures stored in the memory unit, and is made to generate a first natural language sentence that shows the strengths of one member compared to multiple members within the organization. The generating AI is given a second prompt and made to refer to the RAG to generate a second natural language sentence that shows the strengths of one of the members compared to multiple members in a higher hierarchy than the organization itself. A talent profile creation support program characterized by inputting a third prompt to the generating AI and generating profile information for one of the members by combining the first natural language sentence and the second natural language sentence.

2. The information in the skill sheet used to generate the first natural sentence is professional skills. The talent profile creation support program according to claim 1, characterized in that the information in the skill sheet used to generate the second natural language is personal skills.

Images