Employment management systems, employment management programs, and employment management methods

Abstract

We provide an employment management system that appropriately recommends job openings in areas far from the job seeker's address. [Solution] A job recommendation model, trained using machine learning to output the probability that a job seeker will be interested in a job posting as output data, is provided with input data including a job seeker identifier representing an actual job seeker and job postings including a work location, thereby outputting the probability that the job seeker will be interested in the job posting.

Description

Technical Field

[0001] The present invention relates to an employment management system, an employment management program, and an employment management method.

Background Art

[0002] There is known a service of a job matching system in which a company (client) posts a short-term job opening for several hours, and a worker selects a job at a time desired by the worker among various companies, and performs short-term work such as part-time work instead of a continuous employment form with a specific company (Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in the conventional service, when a worker receives a job offer, only job offers in a work area close to the address registered by the worker are proposed. However, there is a problem that job offers in the worker's school commute area, commuting area, and other living areas outside the address are not proposed, and the matching with the desired location of the worker's workplace is not appropriately performed.

Means for Solving the Problems

[0005] One aspect of the present invention is an employment management system for recommending job information to job seekers, characterized in that it outputs the probability that a job seeker will be interested in a job information by inputting input data including a job seeker identifier and job information including a work location to a job recommendation model that has been machine-learned to output the probability that a job seeker will be interested in a job information as output data, using training data which includes job seeker information indicating a job seeker, job information, features that identify places of work where the job seeker tends to work, and features that indicate whether or not the job seeker has shown interest in the job information.

[0006] Here, the feature quantity included in the training data that indicates whether or not the job seeker showed interest is preferably information indicating whether or not the job seeker added the job posting to their favorites.

[0007] Furthermore, it is preferable that the feature quantity included in the training data that indicates whether or not the job seeker showed interest is information indicating whether or not the job seeker actually worked for the job posting.

[0008] Furthermore, it is preferable that the feature quantity included in the training data that indicates whether or not the job seeker showed interest is information indicating whether or not the job seeker viewed the job posting.

[0009] Furthermore, it is preferable to prioritize the display of the job posting information on the job seeker's terminal, depending on the probability that the job seeker is interested in the job posting information.

[0010] Furthermore, it is preferable that the training data further includes at least one of the following features: a feature indicating whether or not the job posting is for an industry or job that the job seeker tends to work in; a feature indicating whether or not the job posting is for a day of the week and time of day that the job seeker tends to work in; a feature indicating whether or not the job posting is for a salary range that the job seeker tends to work in; a feature indicating whether or not the job posting is for working hours that the job seeker tends to work in; a feature indicating whether or not the job posting is for additional information that the job seeker tends to work in; a feature indicating that people of the same gender as the gender registered by the job seeker tend to work in the job posting; and a feature indicating that people of the same age group as the age group registered by the job seeker tend to work in the job posting.

[0011] Another aspect of the present invention is an employment management program that recommends job information to job seekers, characterized in that a computer functions as a job recommendation model that has been machine-trained to output the probability that a job seeker will be interested in a job information as output data, using training data that includes job seeker information indicating a job seeker, job information, features that identify places of work where the job seeker is likely to work, and features that indicate whether or not the job seeker has shown interest in the job information, and input data including a job seeker identifier indicating an actual job seeker and job information including a place of work is input to the job recommendation model to output the probability that the job seeker will be interested in the job information.

[0012] Another aspect of the present invention is an employment management method for recommending job information to job seekers, characterized in that a job recommendation model, which has been trained to output the probability that a job seeker will be interested in a job information as output data using training data that includes job seeker information indicating a job seeker, job information, features that identify places of work where the job seeker is likely to work, and features that indicate whether or not the job seeker has shown interest in the job information, is input data that includes a job seeker identifier indicating an actual job seeker and job information including a place of work, thereby causing the model to output the probability that the job seeker will be interested in the job information. [Effects of the Invention]

[0013] According to the present invention, job information in an area far from the applicant's address can also be appropriately recommended.

Brief Description of the Drawings

[0014] [Figure 1] It is a diagram showing the configuration of an employment management system in an embodiment of the present invention. [Figure 2] It is a diagram showing the configuration of a management server in an embodiment of the present invention. [Figure 3] It is a diagram showing the configuration of an employer terminal in an embodiment of the present invention. [Figure 4] It is a diagram showing the configuration of an applicant terminal in an embodiment of the present invention. [Figure 5] It is a functional block diagram showing the configuration of a management server in an embodiment of the present invention. [Figure 6] It is a diagram showing an example of an applicant information database in an embodiment of the present invention. [Figure 7] It is a diagram showing an example of a job information database in an embodiment of the present invention. [Figure 8] [[ID=3*]]It is a diagram showing an example of a grid system in an embodiment of the present invention. [Figure 9] It is a diagram showing an example of a performance database in an embodiment of the present invention. [Figure 10] It is a schematic diagram showing the configuration of a job recommendation model in an embodiment of the present invention. [Figure 11] It is a diagram showing an example of learning data in an embodiment of the present invention. [Figure 12] [[ID=*3]]It is a diagram showing an example of output data in an embodiment of the present invention. [Figure 13] It is a diagram showing an example of input data in an embodiment of the present invention. [Figure 14] It is a diagram showing an example of output data in an embodiment of the present invention. [Figure 15] It is a diagram showing an example of the recommendation result of job information in a conventional employment management system. [Figure 16] It is a diagram showing an example of a recommendation result of job information in an employment management system in an embodiment of the present invention.

Embodiments for Carrying Out the Invention

[0015] As shown in FIG. 1, an employment management system 100 in an embodiment of the present invention includes a management server 102, an employer terminal 104, and a job seeker terminal 106. The employer terminal 104 and the job seeker terminal 106 may be singular or plural. The management server 102, the employer terminal 104, and the job seeker terminal 106 are connected to each other via an information communication network 108 such as the Internet so as to be able to exchange information with each other.

[0016] Note that the information communication network 108 is not limited to the Internet, and any network that can communicatively connect the management server 102, the employer terminal 104, and the job seeker terminal 106 to each other may be used. For example, it may be a dedicated line, a public line (such as a telephone line, a mobile communication line, etc.), a wired LAN (Local Area Network), a wireless LAN, or a combination of the Internet and these.

[0017] The management server 102 is a server that performs processes such as conclusion of an employment contract between an employer and a job seeker (worker), attendance management of the job seeker (worker), and payment of salary from the employer to the job seeker (worker) via the employer terminal 104 and the job seeker terminal 106. Further, the management server 102 is also a server that performs a process of proposing job openings from the employer to the job seeker.

[0018] As shown in Figure 2, the management server 102 is composed of a processing unit 10, a storage unit 12, an input unit 14, an output unit 16, and a communication unit 18. The processing unit 10 includes means such as a processor that performs calculations, such as a CPU or GPU. The processing unit 10 executes the management server program stored in the storage unit 12 to realize the function of matching companies using employee terminals 104 with workers using job seeker terminals 106 in the employment management system 100 of this embodiment, and to perform processing such as employment contracts, attendance management, and salary payments. The storage unit 12 includes storage means such as semiconductor memory, hard disk, and memory card. The storage unit 12 is connected to the processing unit 10 in an accessible manner and stores the management server program and information necessary for its processing. The input unit 14 includes means for inputting information. The input unit 14 includes, for example, a keyboard, touch panel, buttons, etc., for receiving input from an administrator. The output unit 16 includes means for outputting the processing results of the management server 102, such as a user interface screen (UI) for receiving input information from an administrator. The output unit 16 includes, for example, a display for presenting images to the administrator. The communication unit 18 is configured to include an interface for communicating information with the employer terminal 104 and the job seeker terminal 106 via the information and communication network 108. Communication by the communication unit 18 may be wired or wireless.

[0019] The employee terminal 104 is a terminal used by an employer who wants to hire workers. As shown in Figure 3, the employee terminal 104 is composed of a processing unit 20, a storage unit 22, an input unit 24, an output unit 26, and a communication unit 28. The processing unit 20 includes means such as a processor that performs calculations such as a CPU or GPU. The processing unit 20 realizes the function of an employee terminal in the employment management system 100 in this embodiment by executing the employee terminal program stored in the storage unit 22. The storage unit 22 includes storage means such as a semiconductor memory, a hard disk, or a memory card. The storage unit 22 is connected to the processing unit 20 in an accessible manner and stores the employee terminal program and information necessary for its processing. The input unit 24 includes means for inputting information. The input unit 24 includes, for example, a keyboard, a touch panel, buttons, a camera, etc., for receiving input from the employer on the company side. The output unit 26 includes means for outputting information necessary for processing on the employee terminal 104, such as a display that shows image information such as a screen for receiving input information from the employer or a user interface screen (UI). The communication unit 28 is configured to include an interface for communicating information with the management server 102 via the information communication network 108. Communication by the communication unit 28 may be wired or wireless.

[0020] The job seeker terminal 106 is a terminal used by job seekers who wish to work, such as part-time jobs. As shown in Figure 4, the job seeker terminal 106 is composed of a processing unit 30, a storage unit 32, an input unit 34, an output unit 36, and a communication unit 38. The processing unit 30 includes means such as a processor that performs calculations, such as a CPU or GPU. The processing unit 30 realizes the function of a job seeker terminal in the employment management system 100 in this embodiment by executing the job seeker terminal program stored in the storage unit 32. The storage unit 32 includes storage means such as a semiconductor memory, a hard disk, or a memory card. The storage unit 32 is connected to the processing unit 30 in an accessible manner and stores the job seeker terminal program and information necessary for its processing. The input unit 34 includes means for inputting information. The input unit 34 includes, for example, a keyboard, a touch panel, buttons, a camera, etc., which receive input from the user who will be a worker. The output unit 36 ​​includes means for outputting information necessary for processing on the job seeker terminal 106, such as a display that shows image information such as a screen for receiving input information from the user or a user interface screen (UI). The communication unit 38 is configured to include an interface for communicating information with the management server 102 via the information communication network 108. Communication by the communication unit 38 may be wired or wireless.

[0021] The employer terminal 104 and the job seeker terminal 106 can be any information processing device capable of executing a program to provide services for the employment management system 100. For example, the employer terminal 104 and the job seeker terminal 106 can be dedicated stationary or portable devices, personal computers (PCs), tablet computers, smartphones, mobile phones, PHS (Personal Handy-phone System) terminals, personal digital assistants (PDAs), or multi-functional television receivers with information processing capabilities (so-called smart TVs).

[0022] The employer terminal 104 and the job seeker terminal 106 can be implemented by installing an application (for employers or job seekers). Job seekers operate the job seeker terminal 106 to launch the application and register for use with the management server 102. Similarly, employers operate the employer terminal 104 to launch the application and register for use with the management server 102.

[0023] Furthermore, the above program may be provided as a program product. That is, it may be provided as a program stored on a tangible information medium, or as a program that can be downloaded via an information and communication network such as the Internet.

[0024] Furthermore, the processing in this embodiment is performed by processors included in the management server 102, the employee terminal 104, and the job seeker terminal 106. At this time, the processing of the management server 102, the employee terminal 104, and the job seeker terminal 106 may be combined as appropriate, or may be divided and performed by multiple devices as appropriate.

[0025] [Service Overview] The employment management system 100 according to this embodiment provides services such as concluding employment contracts, managing attendance, and paying salaries between job seekers who wish to work part-time or similar jobs and employers who wish to hire people.

[0026] Specifically, job seekers transmit their own information, namely job seeker information, to the management server 102 via the job seeker terminal 106. Employers transmit job information, including employment conditions such as working hours and wages, to the management server 102 via the employer terminal 104, presenting the job information to job seekers. Job seekers view the job information presented by the employer via the job seeker terminal 106, and if they wish to apply, they transmit a request for employment to the management server 102. Employers view the job seekers' applications via the employer terminal 104, and if they approve the application, they receive a code from the management server 102 linking the employment conditions with the job seeker who wishes to work. When a job seeker arrives at work, they use the job seeker terminal 106 to read the code shown by the employer via the employer terminal 104. As a result, the management server 102 stores that the job seeker has agreed to the employment conditions and that a contract has been concluded, and also stores the time the code was read as the arrival time. Furthermore, the job seeker reads the code again when leaving work. The management server 102 then stores the time the code was read as the departure time.

[0027] Furthermore, the management server 102 deposits wages, which are the compensation for labor, into an account pre-associated with the job seeker, according to the job seeker's arrival and departure times and the wage payment conditions, such as hourly wages, included in the employment conditions. It also collects wages and fees, such as commissions, paid to the job seeker from an account pre-associated with the employer. In addition, it may perform accounting processing related to these wage payments.

[0028] In this way, managing contracts upon arrival at work simplifies the cumbersome procedures for both job seekers and employers. Furthermore, by managing arrival and departure times, working hours can be accurately tracked, enabling accurate payment of wages and other services.

[0029] [Processing on management server 102] In this embodiment, the employment management system 100 performs a job recommendation process that recommends job information to job seekers.

[0030] Figure 5 is a block diagram showing the functional configuration of the management server 102. By executing the management server program, the management server 102 functions as a job seeker information receiving unit 40, a job information receiving unit 42, a job recommendation learning unit 44, and a job recommendation processing unit 46.

[0031] Furthermore, the storage unit 12 of the management server 102 functions as a job seeker information database 60, a job posting information database 62, and a performance database 64.

[0032] As shown in Figure 6, the job seeker information database 60 stores information about job seekers, such as an identifier for identifying the job seeker (hereinafter referred to as the job seeker ID), the job seeker's name, address, email address, telephone number, date of birth, age, gender, image (photo) of the job seeker, Good rating rate, cancellation rate, and last-minute cancellation rate, in association with each other. The information stored in the job seeker information database 60 is not limited to this information, and any information necessary about the job seeker is acceptable.

[0033] As shown in Figure 7, the job information database 62 stores associated information about the employer and employment conditions, such as an identifier for identifying job information (hereinafter referred to as the job ID), an identifier for identifying the employer providing the job information (hereinafter referred to as the employer ID), the employer's name (e.g., store name), address, contact telephone number, industry, and images (photographs) related to the employer. The information stored in the job information database 62 is not limited to this information, but may include any information necessary about the employer and information related to employment conditions, such as information necessary to display in the job information described later.

[0034] Furthermore, location information such as the job seeker's address and the workplace in the job posting is identified by a grid system that represents geospace as a mesh of cells. For example, as shown in Figure 8, a grid system that divides geospace into hexagonal cells, such as H3 provided by Uber, can be applied. Alternatively, a grid system that divides geospace into square cells, such as S2 Geometry provided by Google, may be applied. Each cell in the grid system is assigned unique address information. For example, in H3, as shown in Figure 8, each cell is assigned address information that combines unique numbers along the i, j, and k axes. Location information such as the job seeker's address and workplace used in the employment management system 100 is represented by the address information of the cell corresponding to that location.

[0035] As shown in Figure 9, the performance database 64 stores information such as job seeker ID, job ID, employer ID, employment conditions presented by the employer to the job seeker, whether the job seeker agreed to the employment conditions, arrival time, departure time, and salary to be paid, all linked together. The performance database 64 is registered based on the employment conditions presented by the employer when an application for a job and acceptance of the application are established between the job seeker and the employer. In addition, arrival time and departure time are registered when the job seeker actually arrives at and leaves work.

[0036] Furthermore, the job seeker information database 60 also stores employer IDs that indicate employers (stores) previously registered as "favorites" by job seekers. To register as a "favorite," for example, a "Add to Favorites" button could be displayed on a webpage that shows job postings by employer, and job seekers could register by clicking or tapping this button. However, the method of registering as a "favorite" is not limited to this; any method that allows job seekers to identify and register employers is acceptable.

[0037] The Job Seeker Information Reception Unit 40 receives information from job seekers who wish to apply for a job, via the job seeker terminal 106. The Job Seeker Information Reception Unit 40 receives information about job seekers entered from the job seeker terminal 106 used by the job seekers. The information about job seekers received by the Job Seeker Information Reception Unit 40 is stored in the Job Seeker Information Database 60. The job seeker information received by the Job Seeker Information Reception Unit 40 may also be transmitted to the employer terminal 104 via the information and communication network 108.

[0038] The job information receiving unit 42 receives job information from the employer terminal 104, including employment conditions such as working hours, break times, place of work, wages (hourly wage, daily wage, etc.), overtime pay, various allowances, wage payment date, job description, and grounds for dismissal. The job information receiving unit 42 receives job information from the employer terminal 104 via the information and communication network 108. The job information receiving unit 42 also receives disclosure conditions from the employer terminal 104 via the information and communication network 108 that limit where the job postings are published. The job information receiving unit 42 may also transmit the received job information to the job seeker terminal 106 via the information and communication network 108 according to the disclosure conditions.

[0039] The job recommendation learning unit 44 constructs a job recommendation model using machine learning to extract information about jobs that are recommended to job seekers. Specifically, as shown in Figure 10, the job recommendation learning unit 44 constructs a job recommendation model that outputs predicted values ​​as output data 74 indicating whether or not a job seeker is interested in job information from employers by inputting training data 72 into LightGBM 70 and applying machine learning. LightGBM 70 is an analysis algorithm based on the decision tree algorithm in machine learning, and is a data analysis method in the field of "supervised learning" that expresses target variables from given data. Note that the job recommendation model in this embodiment can also be constructed using a neural network instead of LightGBM 70.

[0040] As shown in Figure 11, the training data 72 is a dataset that associates job seeker IDs, job posting IDs, and information indicating features related to the employment of job seekers. Note that the training data 72 may also include additional information such as employer IDs in addition to the features.

[0041] Features are information that represents the characteristics of employment for each job seeker. Feature 1 indicates whether or not the workplace is one that the job seeker tends to work in. The location is indicated by address information in a grid system. Training data 72 preferably contains at least Feature 1.

[0042] Feature 2 is information indicating whether the job posting is for an industry or job that the job seeker tends to work in. The industry can be any industry related to employment provided in the employment management system 100. Examples of industries include food service, accommodation, retail, and delivery. The job (position) is set for each industry, and for example, in the food service industry it can be waitstaff, dishwashing, and cooking; in the accommodation industry it can be banquet staff and front desk staff; in the retail industry it can be cleaning, cashiering, and stocking; and in the delivery industry it can be inspection, sorting, picking, packing, and loading / unloading. However, the industry and job (position) are not limited to these, and can be any industry and job (position) for which an employment contract can be made between the job seeker and the employer in the employment management system 100.

[0043] Feature 3 indicates whether the job posting is for a day and time that the job seeker tends to work. Feature 4 indicates whether the job posting is for a salary range that the job seeker tends to work in. Feature 5 indicates whether the job posting has working hours that the job seeker tends to work in. Feature 6 indicates whether the job posting has additional information that the job seeker tends to work in. Additional information can include, for example, information on working conditions such as "no experience necessary" or "flexible hairstyle." Feature 7 indicates whether the job posting is for a gender that the job seeker has registered as being the same gender as the person who posted the job. Feature 8 indicates whether the job posting is for a gender that the job seeker has registered as being the same age as the person who posted the job.

[0044] Feature 9 is information indicating whether or not a job seeker showed interest in the job posting. For example, feature 9 could indicate whether or not a job seeker added the job posting to their favorites. Alternatively, feature 9 could indicate whether or not a job seeker actually worked for the job posting. Alternatively, feature 9 could indicate whether or not a job seeker viewed the job posting. Whether or not a job seeker showed interest in the job posting can be represented as 1 if they added it to their favorites, and 0 if they did not. Alternatively, it can be represented as 1 if they actually worked for the job posting and 0 if they did not. Alternatively, it can be represented as 1 if they actually viewed the job posting and 0 if they did not.

[0045] The training data 72 preferably includes features 2 through 9. However, it is not mandatory for the training data 72 to include features 2 through 9. For example, the training data 72 may include any combination of features 2 through 9.

[0046] Furthermore, in order to avoid the information of job seekers' addresses affecting the output data, it is preferable, but not limited to, that the training data 72 does not include features indicating the job seekers' addresses.

[0047] The training data 72 preferably utilizes information from past employment of job seekers in the employment management system 100. Alternatively, the training data may utilize information for which each feature has been assumed for a hypothetical job seeker.

[0048] The training data 72 is input into LightGBM 70, and machine learning is performed so that LightGBM 70 outputs a predicted value as output data 74 indicating whether or not a job seeker is interested in a job posting from an employer. Specifically, for example, when the training data 72 is input to LightGBM 70, the job recommendation model is trained so that the probability that a job seeker has favorited a recommended job posting is correctly output as output data 74. Also, for example, when the training data 72 is input to LightGBM 70, the job recommendation model is trained so that the probability that a job seeker has actually worked for a recommended job posting is correctly output as output data 74. As a result, as shown in Figure 12, a job recommendation model is constructed that outputs the probability that a job seeker is interested in a job posting as output data 74, associated with a job seeker ID indicating a job seeker and a job posting ID indicating a job posting.

[0049] When actually recommending job information from employers to job seekers, the job recommendation processing unit 46 of the management server 102, as shown in Figure 13, associates the information provided as job information from the employer with the job seeker ID representing the job seeker and the job ID representing the job information, and inputs this information into the trained job recommendation model.

[0050] The features entered as input data preferably include information indicating the work location. The location is indicated by address information in the grid system. It is also preferable to include information on the industry and duties being advertised. The industry and duties are those entered by the employer as job information. It is also preferable to include information indicating the days of the week and hours of work included in the job information. It is also preferable to include information indicating the compensation for employment. It is also preferable to include information indicating the working hours of the job. It is also preferable to include additional information regarding the job. Additional information may include, for example, information on working conditions such as "no experience necessary" or "flexible hairstyle." It is also preferable to include information indicating gender and age requirements for the job.

[0051] By inputting this type of data into the job recommendation model, as shown in Figure 14, the probability that a job seeker will be interested in a job offered by a particular employer is output as output data for each combination of a job seeker ID (representing a job seeker) and a job ID (representing a job offer).

[0052] Figure 15 illustrates an example of a conventional employment management system that recommended job postings based solely on the job seeker's address, without considering the location where the job seeker is likely to work. Despite the fact that the actual places where job seekers worked were concentrated in areas far from their address, the employment management system recommended job postings primarily in the vicinity of the job seeker's address. In other words, the system failed to appropriately recommend job postings in areas where the job seeker was most likely to actually work.

[0053] Figure 16 shows an example in the employment management system 100 of this embodiment, where job postings are recommended considering the work locations where job seekers tend to work. The employment management system 100 is able to appropriately recommend job postings not only in the vicinity of the job seeker's address, but also in areas far from the address where the job seeker tends to work.

[0054] When recommending job openings to job seekers, the output data should be sorted in descending order of the probability of interest for each job seeker ID, and a predetermined number of job postings should be recommended from the top. The job recommendation processing unit 46 of the management server 102 preferably transmits information to the job seeker terminal 106 used by each job seeker so that when job postings are viewed on the job seeker terminal 106, the job postings are displayed in descending order of the probability of interest for that job seeker. This allows job seekers to preferentially view job postings that they are likely to be interested in using the job seeker terminal 106. Therefore, matching between job seekers and employers can be performed more effectively.

[0055] Furthermore, the management server 102 may be configured to send push notifications of job information to the job seeker terminals 106. For example, the management server 102 may send push notifications to the job seeker terminals 106 used by each job seeker for job information that the probability of the job seeker being interested in that job seeker is above a predetermined value. This allows job seekers to quickly learn about job information that they are likely to be interested in using their job seeker terminals 106. In this case as well, matching between job seekers and employers can be made more effective. Furthermore, whether or not to send push notifications of job information may be set according to the wishes of the job seeker or the employer. [Explanation of Symbols]

[0056] 10 Processing unit, 12 Storage unit, 14 Input unit, 16 Output unit, 18 Communication unit, 20 Processing unit, 22 Storage unit, 24 Input unit, 26 Output unit, 28 Communication unit, 30 Processing unit, 32 Storage unit, 34 Input unit, 36 Output unit, 38 Communication unit, 40 Job seeker information reception unit, 42 Job information reception unit, 44 Job recommendation learning unit, 46 Job recommendation processing unit, 60 Job seeker information database, 62 Job information database, 64 Performance database, 70 LightGBM, 72 Learning data, 74 Output data, 100 Employment management system, 102 Management server, 104 Employer terminal, 106 Job seeker terminal, 108 Information communication network.

Claims

1. An employment management system that recommends job information to job seekers, An employment management system characterized by outputting the probability that a job seeker will be interested in a job posting as output data, by inputting input data including a job seeker identifier and job posting information including a work location, to a job recommendation model that has been machine-learned to output the probability that a job seeker will be interested in a job posting, using training data that includes job seeker information indicating a job seeker, job posting information, features that identify the work locations in which the job seeker tends to work, and features that indicate whether or not the job seeker has shown interest in the job posting.

2. An employment management system according to claim 1, An employment management system characterized in that the feature quantity included in the training data that indicates whether or not the job seeker has shown interest is information indicating whether or not the job seeker has added the job posting to their favorites.

3. An employment management system according to claim 1, An employment management system characterized in that the feature quantity included in the training data that indicates whether or not the job seeker showed interest is information indicating whether or not the job seeker actually worked for the job posting.

4. An employment management system according to claim 1, An employment management system characterized in that the feature quantity indicating whether or not the job seeker showed interest in the training data is information indicating whether or not the job seeker viewed the job posting.

5. An employment management system according to any one of claims 1 to 4, An employment management system characterized by prioritizing the display of the job posting information on the job seeker's terminal used by the job seeker, according to the probability that the job seeker shows interest in the job posting information.

6. An employment management system according to claim 1, An employment management system characterized in that the training data further includes at least one of the following features: a feature indicating whether or not the job posting is for an industry or job that the job seeker tends to work in; a feature indicating whether or not the job posting is for a day of the week and time of day that the job seeker tends to work in; a feature indicating whether or not the job posting is for a salary range that the job seeker tends to work in; a feature indicating whether or not the job posting is for working hours that the job seeker tends to work in; a feature indicating whether or not the job posting is for additional information that the job seeker tends to work in; a feature indicating that people of the same gender as the gender registered by the job seeker tend to work in the field; and a feature indicating that people of the same age group as the age group registered by the job seeker tend to work in the field.

7. An employment management program that recommends job information to job seekers, Computers, Using training data that includes job seeker information, job postings, features that identify the work locations where the job seeker tends to work, and features that indicate whether or not the job seeker has shown interest in the job posting, a machine learning job recommendation model is configured to output the probability that the job seeker will show interest in the job posting as output data. An employment management program characterized by inputting input data, including a job seeker identifier indicating an actual job seeker and job information including the place of work, into the job recommendation model, thereby outputting the probability that the job seeker will be interested in the job information.

8. An employment management method that recommends job information to job seekers, On the computer, An employment management method characterized by outputting the probability that a job seeker will be interested in a job posting as output data, by inputting input data including a job seeker identifier and job posting information including a work location, into a job recommendation model that has been machine-learned to output the probability that a job seeker will be interested in a job posting, using training data that includes job seeker information indicating a job seeker, job posting information, features that identify the work locations in which the job seeker tends to work, and features that indicate whether or not the job seeker has shown interest in the job posting.

Images