Information processing device, information processing method, and program
The information processing device addresses the limitation of conventional job matching systems by considering reskilling possibilities, providing reliable and detailed skill matching for HR decisions.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- BENESSE CORPORATION
- Filing Date
- 2025-09-16
- Publication Date
- 2026-06-18
Smart Images
Figure 0007876046000001_ABST
Abstract
Description
【Technical Field】 【0001】 The present invention relates to an information processing apparatus, an information processing method, and a program. 【Background Art】 【0002】 Conventionally, systems for assisting in matching job-seeking companies and job seekers have been developed. 【0003】 For example, in Citation 1, there is disclosed a system aimed at matching job seekers affiliated with registered dispatch agencies and job-seeking companies (hereinafter referred to as companies) using terminals connected via a communication network. 【0004】 Also, companies have a desire to identify more optimal employees for the human resources required for their business from the perspective of overall company optimization across businesses and departments. 【Prior Art Documents】 【Patent Documents】 【0005】 【Patent Document 1】 Japanese Unexamined Patent Application Publication No. 2007 - 79900 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0006】 In conventional systems, the matching degree is determined only by comparing the skills specifically required for each job type and business within a company with the skills such as qualifications possessed by employees, and the possibility of reskilling cannot be considered. 【0007】 Therefore, an object of the present invention is to provide an information processing apparatus, an information processing method, and a program capable of performing matching between the human resources required for business and a user in consideration of the possibility of reskilling. 【Means for Solving the Problems】 【0008】 An information processing device according to one aspect of the present disclosure includes: a first extraction unit that extracts JD skills, which are skills included in a first document including job description data; a second extraction unit that extracts CV skills, which are skills included in a second document including career data; an abstraction unit that abstracts the JD skills extracted from the first document and the CV skills extracted from the second document using basic ability data defined for each basic ability related to the skills; a matching unit that compares the JD basic ability, which is an abstraction of the JD skills, with the CV basic ability, which is an abstraction of the CV skills, and calculates a first degree of match between the job description data and the career data; and an output unit that outputs a result including the first degree of match. 【0009】 According to this embodiment, the information processing device can match the personnel required for the job with the user by comparing JD basic abilities and CV basic abilities, which are abstractions of skills extracted from a first document containing job description data and a second document containing career data, respectively, and by considering the possibility of reskilling that can be inferred from the comparison of basic abilities. 【0010】 In the above-described information processing device, the abstraction unit may send an abstraction instruction containing information for abstracting JD skills to the AI server and receive JD basic proficiency as a response. Alternatively, the abstraction unit may send an abstraction instruction containing information for abstracting CV skills to the AI server and receive CV basic proficiency as a response. According to this embodiment, the information processing device can perform highly reliable abstraction processing derived from a vast amount of data by abstracting skills using the AI server. 【0011】 In the above-described information processing device, the abstraction instruction may include an instruction to select one or more basic abilities from the basic ability data to abstract each skill, and to estimate the proportion of each selected basic ability to each skill. According to this embodiment, the information processing device can perform fine-grained matching that takes into account the distribution of the basic abilities that constitute each skill. 【0012】 In the above-described information processing device, the matching unit may compare JD skills and CV skills to calculate a second degree of match between job description data and career data. According to this embodiment, the information processing device can present the user with both a degree of match as an immediate asset, which can be read from the second degree of match, and a degree of match as a potential asset that can be expected to perform well in the future, which can be read from the first degree of match. 【0013】 In the above-described information processing device, the output unit may output a scatter plot on which the vertical axis and horizontal axis represent the second and first match degrees, respectively, plotting the match degrees of multiple career data to a desired job description data, or the match degrees of multiple job description data to a desired career data. According to this embodiment, the information processing device can present the user with a visually understandable representation of whether an individual is immediately productive and has potential, thereby supporting HR personnel in considering transfers and recruitment, and employees in considering reskilling and changing jobs. 【0014】 In the above-described information processing device, the basic abilities may be general abilities not limited to a specific industry or occupation. According to this embodiment, the information processing device can abstract each skill into general abilities, thereby going beyond skills that can be directly grasped from career data and delving into the basic abilities possessed by the user into general abilities that can serve as a common foundation covering various occupations across businesses and departments, and then perform matching. 【0015】 In the above-described information processing device, the basic abilities may be defined as academic disciplines. According to this embodiment, the information processing device can abstract each skill into academic disciplines, thereby enabling matching that goes beyond skills directly obtainable from career data, and delves into academic disciplines that can serve as a common foundation covering various occupations across businesses and departments, based on the basic abilities possessed by the user. 【0016】 In the above-described information processing device, the matching unit may, in addition to / instead of the first degree of match between the job description data and the career data, compare the basic JD skills and the basic CV skills to calculate a first degree of match for each of the JD skills included in the first document. According to this embodiment, by calculating a first degree of match for at least one of the JD skills, the information processing device can present the user with a subdivided degree of match between the job description data and the career data. 【0017】 A method relating to another aspect of this disclosure includes an information processing device extracting JD skills, which are skills contained in a first document containing job description data; extracting CV skills, which are skills contained in a second document containing career data; abstracting the JD skills extracted from the first document and the CV skills extracted from the second document using basic competency data defined for each basic competency related to the skills; calculating a first degree of match between the job description data and the career data by comparing the JD basic competency, which is an abstraction of the JD skills, with the CV basic competency, which is an abstraction of the CV skills; and outputting a result including the first degree of match. 【0018】 A program relating to another aspect of this disclosure causes one or more computers to perform the following processes: extracting JD skills, which are skills contained in a first document containing job description data; extracting CV skills, which are skills contained in a second document containing career data; abstracting the JD skills extracted from the first document and the CV skills extracted from the second document using basic competency data defined for each basic competency related to the skills; comparing the JD basic competency, which is an abstraction of the JD skills, with the CV basic competency, which is an abstraction of the CV skills, to calculate a first degree of match between the job description data and the career data; and outputting the results, including the first degree of match. [Effects of the Invention] 【0019】 According to the present invention, it is possible to provide an information processing apparatus, an information processing method, and a program capable of matching the human resources required for business with users in consideration of the possibility of reskilling. 【Brief Description of the Drawings】 【0020】 [Figure 1] It is a configuration diagram of an information processing system according to an embodiment of the present invention. [Figure 2] It is a block diagram of an information processing apparatus according to an embodiment of the present invention. [Figure 3] It is a diagram illustrating the content of a skill storage unit according to an embodiment of the present invention. [Figure 4] It is a diagram showing an example of a job description according to an embodiment of the present invention. [Figure 5] It is a diagram illustrating the content of a JD storage unit according to an embodiment of the present invention. [Figure 6] It is a diagram illustrating the content of a CV storage unit according to an embodiment of the present invention. [Figure 7] It is a flowchart showing the processing of an information processing apparatus according to an embodiment of the present invention. [Figure 8] It is a diagram illustrating a scatter diagram according to an embodiment of the present invention. [Figure 9] It is a diagram illustrating a radar chart according to an embodiment of the present invention. 【Modes for Carrying Out the Invention】 【0021】 Embodiments of the present invention will be described with reference to the accompanying drawings. Note that the following embodiments are for facilitating the understanding of the present invention and are not for limiting the interpretation of the present invention. Further, the present invention can be variously modified without departing from its gist. Furthermore, those skilled in the art can adopt embodiments in which each element described below is replaced with an equivalent one, and such embodiments are also included in the scope of the present invention. 【0022】 (System Configuration) The outline of the present invention will be explained using Figure 1. Figure 1 is a configuration diagram of an information processing system according to one embodiment of the present invention. The information processing system comprises an information processing device 100 and a plurality of client terminals 200a, 200b, ..., 200n (hereinafter referred to as client terminals 200), and each client terminal 200 is connected to the information processing device 100 so as to be able to communicate with each other via a communication network N. The communication network N may be a wired communication network or a wireless communication network composed of wired or wireless lines, and may be the Internet or a Local Area Network (LAN). 【0023】 The information processing device 100 is a device that performs matching based on a job description (JD) defined by a company (an example of a first document) and an employee's resume (CV) (an example of a second document). A job description is a document that describes the job content in order to clarify the skills and experience required to perform a particular job. In order to match the job description and the resume, the information processing device 100 uses skill data defined for each skill to extract the skills included in the job description, and similarly extracts the skills included in the resume. The information processing device 100 can compare the JD skills extracted from the job description with the CV skills extracted from the resume and output the matching results. 【0024】 Furthermore, the information processing device 100 abstracts JD skills and CV skills using basic ability data defined for each basic ability related to the skills. The information processing device 100 can then compare the JD basic ability, which is an abstraction of JD skills, with the CV basic ability, which is an abstraction of CV skills, and output a matching result. In this way, in addition to matching required skills with possessed skills, the information processing device 100 can output a matching result using not only the matching result of currently possessed skills, but also the potential ability to acquire skills, by matching the required skills with the basic abilities that lead to the acquisition of those skills. 【0025】 Here, "skills" refer to the abilities required of an employee to perform their job. For example, skills include project management, database management, data analysis, crisis management, communication, and personnel planning. 【0026】 Basic abilities are abilities related to a particular skill. For example, in one embodiment, academic disciplines can be used as basic abilities, and in another embodiment, experience can be used as basic abilities. For example, when academic disciplines are used as basic abilities, the information processing device 100 can abstract the "data analysis" skill into the academic disciplines of "statistics" and "sociology" (an example of basic abilities) using basic ability data. In another example, the information processing device 100 can abstract the "communication" skill into the academic disciplines of "organizational behavior," "language," and "psychology" using basic ability data. The above examples describe cases where one skill is abstracted into multiple academic disciplines, but there are also cases where one skill is abstracted into one academic discipline. Basic abilities are not limited to academic disciplines, but may also be general abilities not limited to a specific industry or occupation, such as "data and information management," "English conversation," and "hospitality." 【0027】 In this embodiment, when the information processing device 100 abstracts one skill into multiple basic abilities, it can obtain the proportion occupied by each basic ability, as described above. 【0028】 Furthermore, in this embodiment, when extracting JD skills and CV skills, the information processing device 100 uses an LLM server (not shown). The LLM server is a device that provides services using LLM and is an example of an AI server. The LLM may be a deep learning model with hundreds of millions of parameters and trained on hundreds of gigabytes or more of natural language data. An example of an LLM is gpt-4o. In one example, the LLM server provides services using LLM via an API (Application Programming Interface). 【0029】 In one embodiment, the LLM server receives an input of instructions (which may also be called a prompt) from another device and returns a response to the other device in accordance with those instructions. In one example, the instructions and responses are at least one of text and an image. 【0030】 The client terminal 200 is a terminal used by users. These users include company employees who enter their resumes via the client terminal 200, as well as human resources personnel involved in the company's personnel allocation. 【0031】 Employees of companies entering their resumes can do so via client terminal 200, allowing them to obtain a list of jobs that match their existing skills or their potential abilities. 【0032】 Human resources personnel can obtain a list of employees matching a given job description by entering the job description via client terminal 200. The job description is not limited to those defined by the company; any description can be used, such as the occupational competency evaluation standards published by the Ministry of Health, Labour and Welfare, job descriptions defined by consulting firms, or job descriptions widely used overseas. 【0033】 While security aspects are not directly related to the gist of this application and are therefore not considered here, it will be understood by those skilled in the art that each user is controlled to be permitted to perform only specific operations on specific data. 【0034】 (Functional configuration: Information processing device) Figure 2 is a block diagram of an information processing device according to one embodiment of the present invention. In Figure 2, only the necessary functional configuration is shown, assuming a single information processing device 100. However, the information processing device 100 can also be configured as part of a multifunctional distributed system consisting of multiple computer systems. 【0035】 The information processing device 100 includes a communication unit 110, a control unit 120, and a storage unit 130. 【0036】 The communication unit 110 is configured to connect the information processing device 100 to a network. For example, the communication unit 110 can be implemented from a LAN card, an analog modem, an ISDN modem, etc., and an interface for connecting these to the processing unit via a transmission path such as a system bus. Through the communication unit 110, the information processing device 100 can receive data from client terminals 200 and other servers (not shown). The other servers can be any server that stores data such as an LLM server, employee data that stores data about employees, or department data that stores information about departments to which job descriptions are associated. 【0037】 The control unit 120 includes an arithmetic processing unit 121, such as a CPU or MPU, which corresponds to a processor, and a memory 122, such as RAM. The arithmetic processing unit 121 (processor) realizes the functions and processes described later in the arithmetic processing unit 121 by loading the program recorded in the storage unit 130 into the memory 122 and executing it based on various inputs. This program may be stored on a computer-readable non-temporary recording medium such as a CD-ROM, or distributed via a network and installed on a computer. The memory 122 functions as work memory necessary for program execution by the arithmetic processing unit 121 (processor). 【0038】 The storage unit 130 is composed of a storage device such as a hard disk and stores various programs necessary for executing processing in the control unit 120, as well as data necessary for executing various programs. In this embodiment, it is desirable that the storage unit 130 has a skill storage unit 131, a JD storage unit 132, and a CV storage unit 133. 【0039】 The skill storage unit 131 stores information related to skills. In one embodiment, it is desirable that the skill storage unit 131 includes a skill ID, skill name, department, description, etc., as shown in Figure 3. The skill ID is information that can uniquely identify a skill. The department is information about the department associated with the skill. For example, the department may store information for one or more departments associated with the skill, or if the skill in question is not limited to a specific department, it may store null or information indicating that it is common to the company. The description stores information that describes the skill. 【0040】 Multiple job descriptions are stored in the JD storage unit 132. In one example, a job description like the one shown in Figure 4 is stored. Each job description is associated with at least the job title and the department to which it belongs. In this embodiment, as shown in Figure 5, the JD storage unit 132 stores job descriptions associated with the JDID, which is information that uniquely identifies the job description, the job title, and the department to which it belongs. It is also desirable that for each job description, the JD storage unit 132 stores multiple JD skills extracted from the job description, as well as the required level for each JD skill. Furthermore, it is desirable that for each JD skill in a job description, the JD storage unit 132 stores multiple JD basic competencies that are abstractions of the JD skill, as well as the required level for each JD basic competency. 【0041】 The requirement level is information indicating the degree of requirement for a skill, and in one embodiment, the requirement level is stored as a numerical value between 0 and 100. Among the multiple extracted JD skills, skills with higher requirement levels are more necessary for the target job description and carry more weight in the matching process compared to skills with lower requirement levels. The requirement level is not limited to expressions represented by numbers, for example, but also includes rankings and other expressions represented by letters and symbols such as A, B, C or A, B, C. 【0042】 Multiple resumes are stored in the CV storage unit 133. In one embodiment, as shown in Figure 6, the resumes are stored in the CV storage unit 133 associated with employee IDs. It is also desirable that for each resume, the CV storage unit 133 stores multiple CV skills extracted from the resume, as well as the level at which each CV skill is possessed. Furthermore, it is desirable that for each CV skill in the resume, the CV storage unit 133 stores multiple CV basic abilities that are abstractions of the said CV skill, as well as the level at which each CV basic ability is possessed. 【0043】 The proficiency level is information indicating the degree to which an employee possesses a skill. In one embodiment, the proficiency level is stored as a numerical value between 0 and 100. A higher proficiency level indicates a higher level of competence compared to an employee with a lower proficiency level. The proficiency level is not limited to numerical representations, but also includes rankings and other representations using letters and symbols, such as A, B, C, or A, B, C. 【0044】 Furthermore, as shown in Figure 2, the arithmetic processing unit 121 includes, as functional units, a skill registration unit 1211, a JD registration unit 1212, a CV registration unit 1213, a JD skill extraction unit 1214, a CV skill extraction unit 1215, an abstraction unit 1216, a request reception unit 1217, a matching unit 1218, and an output unit 1219. 【0045】 The skill registration unit 1211 receives information about skills and registers it in the skill storage unit 131. In this embodiment, the skill registration unit 1211 receives information about skills from the client terminal 200 and registers it in the skill storage unit 131. In this embodiment, the HR person creates a skill file that defines the relevant skills and their descriptions for each department, and sends the created skill file to the information processing device 100. The skill file is associated with at least a department ID that uniquely identifies the department, and the skill registration unit 1211 registers multiple skills associated with the department ID. As mentioned above, for skills that are not limited to a specific department, the HR person creates a skill file associated with department ID "0: Common" and sends it to the information processing device 100. 【0046】 The JD registration unit 1212 receives the description and stores it in the JD storage unit 132. In this embodiment, the JD registration unit 1212 receives the job description from the client terminal 200 and stores it in the JD storage unit 132. In this embodiment, the HR person transmits the job description associated with the job title and department to the information processing device 100 via the client terminal 200, and the JD registration unit 1212 stores the job description associated with the job title and department. 【0047】 The CV registration unit 1213 receives the resume and stores it in the CV storage unit 133. In this embodiment, the CV registration unit 1213 receives the resume from the client terminal 200 and stores it in the CV storage unit 133. In this embodiment, the employee sends the resume associated with the employee ID to the information processing device 100 via the client terminal 200, and the CV registration unit 1213 stores the resume associated with the employee ID. 【0048】 The JD skill extraction unit 1214 extracts skills included in the job description based on the skill storage unit 131 defined for each skill. In this embodiment, the JD skill extraction unit 1214 retrieves skills associated with departments that match the department to which the job description belongs, as well as skills registered as company-wide skills, from the skill storage unit 131, and uses the acquired skill data to extract skills included in the job description. 【0049】 Furthermore, in this embodiment, the JD skill extraction unit 1214 extracts skills contained in a job description by inputting the job description into a machine learning model. In one example, the machine learning model is configured by learning training data that includes a job description and a set of skills and requirement levels contained in the job description. The machine learning model may be an LLM hosted on an LLM server, or it may be a machine learning model that is not an LLM hosted on the information processing device 100. 【0050】 In this embodiment, the JD skill extraction unit 1214 extracts skills by inputting instructions (hereinafter referred to as "JD skill extraction instructions") containing information for extracting skills within the job description to the LLM. In one example, the JD skill extraction instructions include instructions to extract skills contained within the job description from skill data and to estimate the requirement level of each extracted skill. 【0051】 The JD skill extraction unit 1214 stores the extracted JD skills and the required level of each JD skill in the JD storage unit 132. 【0052】 The CV skill extraction unit 1215 extracts the skills included in the resume based on the skill storage unit 131 defined for each skill. In this embodiment, the CV skill extraction unit 1215 retrieves all skills from the skill storage unit 131 and uses the acquired skill data to extract the skills included in the resume. 【0053】 Furthermore, in this embodiment, the CV skill extraction unit 1215 extracts skills contained in a resume by inputting the resume into a machine learning model. In one example, the machine learning model is configured to learn training data that includes a resume and a set of skills and skill levels contained within the resume. The machine learning model may be an LLM hosted on an LLM server, or it may be a non-LLM machine learning model hosted on the information processing device 100. 【0054】 In this embodiment, the CV skill extraction unit 1215 extracts skills by inputting instructions (hereinafter referred to as "CV skill extraction instructions") containing information for extracting skills from the resume into the LLM. In one example, the CV skill extraction instructions include instructions to extract skills contained in the resume from the skill data and to estimate the level of possession of each extracted skill. 【0055】 The CV skill extraction unit 1215 stores the extracted CV skills and the level at which each CV skill is held in the CV storage unit 133. 【0056】 The abstraction unit 1216 abstracts JD skills and CV skills using the basic ability data defined for each basic ability related to the skills. In this embodiment, for multiple JD skills associated with each job description in the JD storage unit 132, the abstraction unit 1216 abstracts each JD skill into one or more basic abilities and converts the required level of the JD skill to the required level of each basic ability. Similarly, for multiple CV skills associated with each resume in the CV storage unit 133, the abstraction unit 1216 abstracts each CV skill into one or more basic abilities and converts the possession level of the CV skill to the possession level of each basic ability. 【0057】 In this embodiment, the abstraction unit 1216 uses basic ability data that uses academic disciplines as basic abilities. For example, the basic ability data may include science, law, political science, education, social welfare, economics, accounting, distribution, finance, commerce, life sciences, organizational behavior, statistics, languages, psychology, etc. 【0058】 Furthermore, in this embodiment, the abstraction unit 1216 abstracts JD skills or CV skills into one or more basic abilities by inputting them into a machine learning model. In one example, the machine learning model is configured to learn training data that includes a combination of a skill and a set of one or more basic abilities related to that skill and the proportion of each basic ability. The machine learning model may be an LLM hosted on an LLM server, or it may be a machine learning model that is not an LLM hosted on the information processing device 100. 【0059】 In this embodiment, the abstraction unit 1216 abstracts skills by inputting instructions (hereinafter referred to as "abstraction instructions") containing information for abstracting skills into the LLM. In one example, the abstraction instructions include instructions to select one or more basic abilities from basic ability data to abstract the skills, and to estimate the proportion that each selected basic ability accounts for in the skills. 【0060】 Once the basic abilities and the percentages of each basic ability are obtained, the abstraction unit 1216 converts the required level of the JD skills (or the level of the CV skills) into the required level (or level of the CV skills) of each basic ability. In one embodiment, the abstraction unit 1216 converts the required level of the JD skills (or the level of the CV skills) into the required level (or level of the CV skills) of each basic ability based on the following formula. Required level of basic skills = Required level of JD skills × Percentage of the target basic skills / Sum of percentages of all basic skills that abstract JD skills Basic ability level = CV skill level × percentage of target basic ability / sum of percentages of all basic abilities abstracted from CV skills 【0061】 For example, if JD Skill 1 (required level 50), Basic Ability 1 (percentage 70), and Basic Ability 2 (percentage 30), then the required level for Basic Ability 1 = JD Skill required level 50 × Basic Ability 1 percentage 70 / (Basic Ability 1 percentage 70 + Basic Ability 2 percentage 30), which equals 35. 【0062】 The abstraction unit 1216 stores the JD basic abilities, which are abstractions of JD skills, and the required level of each JD basic ability in the JD storage unit 132, and the CV basic abilities, which are abstractions of CV skills, and the level of each CV basic ability possessed in the CV storage unit 133. 【0063】 The request receiving unit 1217 receives requests from the user. In this embodiment, the request receiving unit 1217 receives either a request to output an employee list that matches a desired job description, or a request to output a job list that matches a desired resume. 【0064】 The matching unit 1218 compares the JD skills and required levels extracted from the job description with the CV skills and levels held extracted from the resume to calculate a direct match score (an example of a second match score). In this embodiment, the matching unit 1218 calculates a value between 0 and 100 as the direct match score. In one embodiment, the matching unit 1218 can calculate the direct match score based on the following formula. Direct Match Score = Sum of CV skill levels (maximum required level) that match the JD skill / Sum of required JD skill levels × 100 【0065】 For example, the degree of direct match between a job description with JD Skill 1 (required level 50), JD Skill 2 (required level 30), and JD Skill 4 (required level 70) and a resume with CV Skill 1 (held level 70), CV Skill 2 (held level 25), and CV Skill 3 (held level 50) is calculated as (held level 50 of CV Skill 1 + held level 25 of CV Skill 2) / (required level 50 of JD Skill 1 + required level 30 of JD Skill 2 + required level 70 of JD Skill) × 100, which equals 50. Here, we assume that JD Skill 1 = CV Skill 1 and JD Skill 2 = CV Skill 2. 【0066】 In this embodiment, an example is described in which the matching unit 1218 directly calculates the degree of match using the level of possession of CV skills that match the JD skills. However, in another embodiment, the matching unit 1218 may directly calculate the degree of match using the level of possession of CV skills that are similar to the JD skills. The degree of similarity between JD skills and CV skills can be determined by any method. 【0067】 Furthermore, the matching unit 1218 compares the basic skills and required levels of the job description (JD) with the basic skills and levels held in the resume (CV) to calculate a potential match score (an example of a first match score). In this embodiment, the matching unit 1218 calculates a value between 0 and 100 as the potential match score. In one embodiment, the matching unit 1218 can calculate the potential match score based on the following formula. Potential match score = Sum of CV basic ability levels that match JD basic ability levels (maximum required level) / Sum of required JD basic ability levels × 100 【0068】 The output unit 1219 outputs the results, including the match score calculated by the matching unit 1218. In this embodiment, based on the request received by the request receiving unit 1217, the output unit 1219 outputs a scatter plot of the desired job description, with the vertical axis and horizontal axis representing the direct match score and the latent match score, plotting the employees. 【0069】 Furthermore, the output unit 1219 outputs a scatter plot of the desired resume, with the vertical axis representing the degree of direct match and the horizontal axis representing the degree of potential match, based on the request received by the request receiving unit 1217. 【0070】 Furthermore, the output unit 1219 may output a radar chart showing the skill gap between a specific job and a specific user, using the JD skills and required levels extracted from a specific job description and the CV skills and levels obtained from a specific resume. 【0071】 (Functional configuration: Client terminal) The client terminal 200 is an information processing device equipped with the function to communicate with the information processing device 100 via the network N. Specifically, it can be a mobile phone, smartphone, PC, PDA, tablet, etc., but is not limited to these. The client terminal 200 includes an input unit 210 such as a keyboard, mouse, or touch panel that accepts user input, a control unit 220 including an arithmetic processing unit and memory, a communication unit 230 that connects the client terminal 200 to the network, and a display unit 240 such as a display that shows various images processed under the control of the arithmetic processing unit. 【0072】 The control unit 220 includes a skill file transmission unit 221 that transmits a skill file defining relevant skills and their descriptions for each department to the information processing device 100; a JD transmission unit 222 that transmits the job title and job description associated with the department to the information processing device 100; a CV transmission unit 223 that transmits a resume associated with the employee ID to the information processing device 100; and a request transmission unit 224 that transmits requests from users to the information processing device 100. 【0073】 (operation) Next, the operation of the information processing system will be explained. Figure 7 is a flowchart showing an example of processing performed by an information processing device according to one embodiment of the present invention. 【0074】 In step S101, the skill registration unit 1211 of the information processing device 100 receives information about skills and registers it in the skill storage unit 131. In this embodiment, the skill registration unit 1211 is assumed to have received a skill file from the client terminal 200 that defines the skills related to each department and their descriptions. The skill file is associated with at least a department ID that uniquely identifies the department, and the skill registration unit 1211 registers multiple skills associated with the department ID. Here, it is assumed that the skill registration unit 1211 has received multiple skill files and registered the data shown in Figure 4. 【0075】 In step S102, the JD registration unit 1212 of the information processing device 100 receives the description and stores it in the JD storage unit 132. In this embodiment, the JD registration unit 1212 receives job descriptions associated with job titles and departments from the client terminal 200 and stores the job descriptions associated with job titles and departments. Here, it is assumed that the JD registration unit 1212 receives multiple job descriptions associated with each department of the company and stores them in the JD storage unit 132. 【0076】 In step S103, the CV registration unit 1213 of the information processing device 100 receives the resume and stores it in the CV storage unit 133. In this embodiment, the CV registration unit 1213 receives the resume associated with the employee ID from the client terminal 200 and stores the resume associated with the employee ID. Here, it is assumed that the CV registration unit 1213 receives multiple resumes associated with each employee ID and stores them in the CV storage unit 133. 【0077】 In step S104, the JD skill extraction unit 1214 of the information processing device 100 extracts the skills included in the job description based on the skill storage unit 131 defined for each skill. In this embodiment, the JD skill extraction unit 1214 retrieves from the skill storage unit 131 the skills associated with the department to which the target job description belongs, and skills registered as company-wide skills, and uses the acquired skill data to extract the skills included in the job description. 【0078】 In this embodiment, the JD skill extraction unit 1214 extracts skills by inputting an instruction (hereinafter referred to as "JD skill extraction instruction") containing information for extracting skills within the job description to the LLM. In one example, the JD skill extraction instruction includes instructions to extract skills contained within the job description from the skill data and to estimate the requirement level of each extracted skill. Here, for the job description of JDID "0001" in the JD storage unit 132, the JD skill extraction unit 1214 extracts JD skill 1 (required level 50), JD skill 2 (required level 30), and JD skill 4 (required level 70). 【0079】 The JD skill extraction unit 1214 stores the extracted JD skills and the required level for each JD skill in the JD storage unit 132. The JD skill extraction unit 1214 then considers that it has extracted skills for all job descriptions in the JD storage unit 132. 【0080】 In step S105, the CV skill extraction unit 1215 of the information processing device 100 extracts the skills included in the resume based on the skill storage unit 131 defined for each skill. In this embodiment, the CV skill extraction unit 1215 retrieves all skills from the skill storage unit 131 and uses the acquired skill data to extract the skills included in the resume. 【0081】 In this embodiment, the CV skill extraction unit 1215 extracts skills by inputting instructions (hereinafter referred to as "CV skill extraction instructions") containing information for extracting skills from the resume into the LLM. In one example, the CV skill extraction instructions include instructions to extract skills contained in the resume from the skill data and to estimate the level of possession of each extracted skill. Here, for the resume of employee ID "A0001" in the CV storage unit 133, the CV skill extraction unit 1215 extracts CV skill 1 (level 70), CV skill 2 (level 25), and CV skill 3 (level 50). 【0082】 The CV skill extraction unit 1215 stores the extracted CV skills and the level of each CV skill in the CV storage unit 133. The CV skill extraction unit 1215 then considers that it has extracted skills from all the resumes in the CV storage unit 133. 【0083】 In step S106, the abstraction unit 1216 of the information processing device 100 abstracts the JD skills using the basic ability data defined for each basic ability related to the skills. In this embodiment, the abstraction unit 1216 abstracts each of the multiple JD skills associated with each job description in the JD storage unit 132 into one or more basic abilities, and converts the requirement level of the JD skills to the requirement level of each basic ability. 【0084】 In this embodiment, the abstraction unit 1216 uses basic ability data that uses academic disciplines as basic abilities. For example, the basic ability data may include science, law, political science, education, social welfare, economics, accounting, distribution, finance, commerce, life sciences, organizational behavior, statistics, languages, psychology, etc. 【0085】 In this embodiment, the abstraction unit 1216 abstracts skills by inputting instructions (hereinafter referred to as "abstraction instructions") containing information for abstracting skills into the LLM. In one example, the abstraction instructions include instructions to select one or more basic abilities from basic ability data to abstract the skills, and to estimate the proportion of each selected basic ability to the skills. Here, it is assumed that the abstraction unit 1216 inputs instructions regarding JD skill 1 into the LLM and obtains a response including basic ability 1 (proportion 70) and basic ability 2 (proportion 30). 【0086】 Once the basic abilities and the proportions of each basic ability are obtained, the abstraction unit 1216 converts the required level of the JD skills into the required level of each basic ability. In one embodiment, the abstraction unit 1216 converts the required level of the JD skills into the required level of each basic ability based on the following formula. 【0087】 Here, the abstraction unit 1216 assumes that, in the case of JD Skill 1 (required level 50), Basic Ability 1 (percentage 70), and Basic Ability 2 (percentage 30), the required level of Basic Ability 1 is converted to 35 by calculating the required level of JD Skill 50 × percentage of Basic Ability 1 70 / (percentage of Basic Ability 1 70 + percentage of Basic Ability 2 30). 【0088】 The abstraction unit 1216 stores the JD basic abilities, which are abstractions of JD skills, and the required level of each JD basic ability in the JD storage unit 132. Here, for the job description with JDID "0001" in the JD storage unit 132, the abstraction unit 1216 abstracts the JD skills as shown in Figure 5. Similarly, the abstraction unit 1216 abstracts the JD skills for all job descriptions in the JD storage unit 132. 【0089】 In step S107, the abstraction unit 1216 abstracts the CV skills using the basic ability data defined for each basic ability related to the skills. In this embodiment, the abstraction unit 1216 abstracts each of the multiple CV skills associated with each resume in the CV storage unit 133 into one or more basic abilities, and converts the level of possession of the CV skills to the level of possession of each basic ability. 【0090】 The abstraction process for CV skills is the same as in step S106, so a detailed explanation is omitted. Here, the abstraction unit 1216 abstracts the CV skills for the resume of employee ID "A0001" in the CV storage unit 133, as shown in Figure 6. Similarly, the abstraction unit 1216 abstracts the CV skills for all resumes in the CV storage unit 133. 【0091】 In step S108, when the request receiving unit 1217 of the information processing device 100 receives a request from the client terminal 200, in step S109, the request receiving unit 1217 determines the type of request. In this embodiment, the request receiving unit 1217 receives either a request to output an employee list that matches a desired job description, or a request to output a job list that matches a desired resume. 【0092】 If it is determined that a request has been received to output a list of employees that match the desired job description (S109: JD match request), in step S110, the matching unit 1218 of the information processing device 100 compares the JD skills and required levels extracted from the desired job description with the CV skills and possessed levels extracted from each resume in the CV storage unit 133 to calculate the direct match degree. In this embodiment, the matching unit 1218 calculates the direct match degree as a number between 0 and 100. In this embodiment, the matching unit 1218 calculates the direct match degree based on the following formula. Direct Match Score = Sum of CV skill levels (maximum required level) that match the JD skill / Sum of required JD skill levels × 100 【0093】 Here, the matching unit 1218 calculates a direct match score of "50" by comparing the JD Skill 1 (required level 50), JD Skill 2 (required level 30), and JD Skill 4 (required level 70) extracted from the job description of JDID "0001" with the CV Skill 1 (held level 70), CV Skill 2 (held level 25), and CV Skill 3 (held level 50) extracted from the resume of employee ID "A0001". Subsequently, the matching unit 1218 compares all resumes stored in the CV storage unit 133 with the desired job description and calculates a direct match score. 【0094】 Next, in step S111, the matching unit 1218 compares the JD basic skills and required level of the desired job description with the CV basic skills and possessed level of each resume to calculate the potential match score. In this embodiment, the matching unit 1218 calculates a value between 0 and 100 as the potential match score. In this embodiment, the matching unit 1218 calculates the potential match score based on the following formula. Potential match score = Sum of CV basic ability levels that match JD basic ability levels (maximum required level) / Sum of required JD basic ability levels × 100 【0095】 Here, the matching unit 1218 calculates a potential match score of "73" between the job description for JDID "0001" and the resume for employee ID "A0001". Subsequently, the matching unit 1218 compares all resumes stored in the CV storage unit 133 with the desired job description and calculates the potential match score. 【0096】 Finally, in step S112, the output unit 1219 of the information processing device 100 outputs the results including the match score calculated by the matching unit 1218. In this embodiment, the output unit 1219 outputs a scatter plot for the desired job description, with the vertical axis and horizontal axis representing the direct match score and the latent match score, plotting the employees. Here, it is assumed that the output unit 1219 outputs a scatter plot as shown in Figure 8. 【0097】 On the other hand, if it is determined that a request has been received to output a list of jobs that match the desired resume (S109: CV match request), in step S113, the matching unit 1218 compares the CV skills and levels extracted from the desired resume with the JD skills and required levels extracted from each job description in the JD storage unit 132 to calculate the degree of direct match. The calculation of the degree of direct match is the same as in S110, so a detailed explanation is omitted. 【0098】 Next, in step S114, the matching unit 1218 compares the basic skills and skill levels of the desired CV with the basic skills and required skill levels of the JD for each job description to calculate the potential match. The calculation of the potential match is the same as in S111, so a detailed explanation is omitted. 【0099】 Finally, in step S115, the output unit 1219 outputs the results including the match score calculated by the matching unit 1218. In this embodiment, the output unit 1219 outputs a scatter plot of the desired resume, with the vertical axis and horizontal axis representing the direct match score and the potential match score, plotting the jobs. Alternatively, the output unit 1219 may, for example, in response to a user's operation to select a plot corresponding to a specific job on the scatter plot, output a radar chart showing the skill gap between a specific job and a specific user, as shown in Figure 9, using the JD skills and required levels extracted from the selected specific job description and the CV skills and possessed levels extracted from the desired resume. 【0100】 In this embodiment, an example was described in which the matching unit 1218 calculates the potential match between the entire job description and the entire resume by comparing the basic JD abilities derived from all JD skills extracted from the job description with the basic CV abilities derived from all CV skills extracted from the resume. However, in addition to / instead of the potential match between the entire job description and the entire resume, the matching unit 1218 may also calculate the potential match by comparing the basic JD abilities with the basic CV abilities for each JD skill extracted from the job description. 【0101】 Furthermore, in this embodiment, an example was described in which the abstraction unit 1216 causes the LLM server to abstract each skill into one or more basic abilities and estimate the proportion of each skill that these basic abilities occupy. However, as mentioned above, the information processing device 100 may use a machine learning model other than LLM, hosted on the information processing device 100, to abstract skills and estimate the proportion of each skill that basic abilities occupy, or it may use any other technology, not limited to a machine learning model, to perform these processes. Similarly, for the skill extraction process by the JD skill extraction unit 1214 and the CV skill extraction unit 1215, the information processing device 100 may use a machine learning model other than LLM, hosted on the information processing device 100, or it may use any other technology, not limited to a machine learning model, to extract skills. 【0102】 As described above, according to this embodiment, the information processing device 100 can match the necessary personnel with employees by comparing the JD basic abilities and CV basic abilities, which are abstractions of skills extracted from the job description and resume, respectively, and taking into account the possibility of reskilling. In this way, by abstracting skills and performing matching, users of the information processing device 100, such as HR personnel, employees, and employees' supervisors, can discover employee aptitudes that the users themselves were unaware of. Furthermore, by abstracting skills and performing matching, users of the information processing device 100 can find personnel who can thrive across job types and industries. 【0103】 Furthermore, the information processing device 100 can perform highly reliable abstraction processing derived from a vast amount of data by abstracting skills using the LLM server. 【0104】 Furthermore, the information processing device 100 can perform fine-tuned matching that takes into account the distribution of the basic abilities that constitute each skill by having the LLM server abstract each skill into one or more basic abilities and estimate the proportion of each skill that these basic abilities represent. 【0105】 Furthermore, the information processing device 100 can calculate a direct match score by comparing JD skills and CV skills, and can present the user with both a match score as an immediate asset, which can be read from the direct match score, and a match score as a potential asset, which can be read from the latent match score, indicating potential for future success. 【0106】 Furthermore, the information processing device 100 can output a scatter plot with the vertical and horizontal axes representing direct match degree and potential match degree, providing users with a visually understandable representation of whether an employee is immediately productive or has potential. This supports HR personnel in considering transfers and employees in considering reskilling. For example, as shown in Figure 8, the upper right area of the scatter plot shows a high match degree from both the perspective of immediate productiveness and potential, suggesting that transfer / reskilling is the highest priority. The lower right area of the scatter plot shows a high match degree from the perspective of immediate productiveness, suggesting that from a short-term perspective, transfer / reskilling is the second highest priority after the upper right area. The upper left area shows a high match degree from the perspective of potential, suggesting that from a long-term perspective, transfer / reskilling is the second highest priority after the upper right area. It should be noted that user considerations of transfers and reskilling are not limited to within the company; the information processing device 100 can also support considerations of recruitment and job changes. 【0107】 Furthermore, by abstracting each skill into an academic discipline, the information processing device 100 can not only grasp skills directly from a resume, but also delve into the fundamental abilities possessed by the user into an academic discipline that can serve as a common foundation covering various job types across businesses and departments, enabling matching. 【0108】 Furthermore, by abstracting each skill into experience, the information processing device 100 can perform matching not only with skills that can be directly grasped from a resume, but also by delving into experiences that can be measured with a greater degree of freedom from the user's entire career to assess the fundamental abilities the user possesses. 【0109】 Furthermore, the information processing device 100 can calculate the potential match degree for each JD skill, and then present the user with a more detailed match degree that breaks down the potential match degree between the job description and the resume. [Explanation of symbols] 【0110】 100...Information processing unit, 110...Communication unit, 120...Control unit, 121...Calculation processing unit, 122...Memory, 1211...Skill registration unit, 1212...JD registration unit, 1213...CV registration unit, 1214...JD skill extraction unit (first extraction unit), 1215...CV skill extraction unit (second extraction unit), 1216...Abstraction unit, 1217...Request reception unit, 1218...Matching unit, 1219...Output unit, 130...Storage unit, 131...Skill storage unit, 132...JD storage unit, 133...CV storage unit, 200...Client terminal, N...Network
Claims
[Claim 1] A first extraction unit extracts JD skills, which are skills contained within a first document containing job description data, A second extraction unit extracts CV skills, which are skills contained within a second document containing career data, An abstraction unit that inputs the JD skills extracted from the first document into a machine learning model to acquire JD basic abilities abstracted as one or more basic abilities related to the JD skills, and inputs the CV skills extracted from the second document into the machine learning model to acquire CV basic abilities abstracted as one or more basic abilities related to the CV skills, wherein the machine learning model is configured to learn training data including skills and one or more basic abilities related to those skills, A matching unit calculates a first degree of match between the job description data and the career data by comparing the JD basic ability, which is an abstraction of the JD skills, with the CV basic ability, which is an abstraction of the CV skills, by quantifying them. An output unit that outputs a result including the first match degree, An information processing device equipped with the following features. [Claim 2] A first extraction unit that extracts JD skills, which are skills included in a first document containing job description data, A second extraction unit extracts CV skills, which are skills contained within a second document containing career data, An abstraction unit that inputs an abstraction instruction to the LLM containing information for abstracting the JD skills extracted from the first document, thereby obtaining JD basic abilities abstracted as one or more basic abilities related to the JD skills, and inputs an abstraction instruction to the LLM containing information for abstracting the CV skills extracted from the second document, thereby obtaining CV basic abilities abstracted as one or more basic abilities related to the CV skills, A matching unit calculates a first degree of match between the job description data and the career data by comparing the JD basic ability, which is an abstraction of the JD skills, with the CV basic ability, which is an abstraction of the CV skills, by quantifying them. An output unit that outputs a result including the first match degree, An information processing device equipped with the following features. [Claim 3] The abstraction instruction includes instructions to select one or more fundamental abilities to abstract each skill from fundamental ability data defined for each fundamental ability related to the skill, and to estimate the proportion that each selected fundamental ability accounts for each skill. The information processing apparatus according to claim 2. [Claim 4] The matching unit compares the JD skills and the CV skills to calculate a second degree of match between the job description data and the career data. The information processing apparatus according to claim 1 or 2, wherein the output unit outputs a result further including the second match degree. [Claim 5] The information processing apparatus according to claim 4, wherein the output unit outputs a scatter plot on which the vertical axis and horizontal axis represent the second match degree and the first match degree, respectively, plotting the match degree of multiple history data to a desired job description data, or the match degree of multiple job description data to a desired history data. [Claim 6] The information processing apparatus according to claim 1 or 2, wherein the aforementioned basic capabilities are general-purpose capabilities not limited to a specific industry or occupation. [Claim 7] The information processing device according to claim 1 or 2, wherein the aforementioned basic ability is an academic discipline. [Claim 8] The information processing apparatus according to claim 1 or 2, wherein the matching unit, in addition to / instead of the first degree of match between the job description data and the career data, compares the JD basic ability and the CV basic ability to calculate a first degree of match for at least one of the JD skills included in the first document. [Claim 9] Information processing device, Extracting JD skills, which are skills contained within the first document containing job description data, Extracting CV skills, which are skills contained within the second document containing career data, The process involves inputting the JD skills extracted from the first document into a machine learning model to obtain JD basic abilities abstracted as one or more basic abilities related to the JD skills, and inputting the CV skills extracted from the second document into the machine learning model to obtain CV basic abilities abstracted as one or more basic abilities related to the CV skills, wherein the machine learning model is configured to learn training data that includes combinations of skills and one or more basic abilities related to those skills. The first degree of match between the job description data and the career data is calculated by comparing the JD basic ability, which is an abstraction of the JD skills, with the CV basic ability, which is an abstraction of the CV skills, by quantifying them. Output the result including the first match score. A method that includes this. [Claim 10] On one or more computers, The process involves extracting JD skills, which are skills contained within the first document containing job description data, and The process involves extracting CV skills, which are skills contained within a second document containing career data, and The process involves inputting an abstraction instruction into the LLM containing information for abstracting the JD skills extracted from the first document, thereby obtaining JD basic abilities abstracted as one or more basic abilities related to the JD skills, and inputting an abstraction instruction into the LLM containing information for abstracting the CV skills extracted from the second document, thereby obtaining CV basic abilities abstracted as one or more basic abilities related to the CV skills. A process to calculate a first degree of match between the job description data and the career data by comparing the JD basic ability, which is an abstraction of the JD skills, with the CV basic ability, which is an abstraction of the CV skills, by quantifying them. A process to output a result including the first match degree. A program that executes the command.