Information processing method, program, and information processing device

Abstract

On a job posting platform, the matching rate between job postings and workers is appropriately calculated, or appropriate job postings are suggested to workers based on their suitability using the matching rate. [Solution] The information processing method involves an information processing device acquiring worker location information, inputting job information, number of positions available, and number of applicants for each currently recruiting job into a learning model that estimates the matching rate of each job, which is trained using learning data that includes at least the matching rate of each past job and job information, estimating the matching rate of each currently recruiting job on a predetermined date using the data output from the learning model, and recommending each job to the worker, which is set using the estimated matching rate of each job on the predetermined date and the distance between the worker's location based on location information and the work location of each currently recruiting job.

Description

Technical Field

[0001] The present disclosure relates to an information processing method, a program, and an information processing apparatus.

Background Art

[0002] In recent years, there has been known a technology for reducing the mismatch between job offers and job seekers and improving the matching rate by considering information other than resume information, and a technology for distributing job offer information that improves the matching rate between job offers and job seekers (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the prior art, regarding the matching rate, only the matching rate between a specific job offer and an applicant for this job offer (also referred to as a "worker") is considered, and there is room for improvement and variation in the use of the matching rate.

[0005] Therefore, one of the objectives of the disclosed technology is to provide an information processing method, a program, and an information processing apparatus that enable appropriate calculation of the matching rate of workers for a job offer in a platform that provides job offer information, or enable appropriate job offers to be proposed according to workers by using the matching rate.

Means for Solving the Problems

[0006] An information processing method according to one embodiment of the present disclosure includes the following steps: an information processing device acquires worker location information; inputs job information, number of positions available, and number of applicants for each currently recruiting job into a learning model that estimates the matching rate of each job, which is trained using learning data that includes at least the matching rate of each past job and job information; estimates the matching rate of each currently recruiting job on a predetermined date using the data output from the learning model; and recommends each job to the worker, which is set using the estimated matching rate of each job on the predetermined date and the distance between the worker's location based on the location information and the work location of each currently recruiting job. [Effects of the Invention]

[0007] According to the disclosed technology, a job posting platform can appropriately calculate the worker matching rate for job postings, or use the matching rate to suggest appropriate job postings to workers. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows examples of various configurations of information processing systems in the disclosed technology. [Figure 2] This is a block diagram showing an example of an information processing device according to the first embodiment. [Figure 3] This is a block diagram showing an example of a server according to the first embodiment. [Figure 4] This figure shows an example of the correlation between the number of applicants three days prior to application according to the first embodiment and the number of people who have completed their final work. [Figure 5] This figure shows an example of user information according to the first embodiment. [Figure 6] This figure shows an example of job posting information according to the first embodiment. [Figure 7] This figure shows an example of usage history information according to the first embodiment. [Figure 8] This flowchart shows an example of the process for estimating the matching rate of a predetermined area according to the first embodiment. [Figure 9] It is a diagram showing an example of the display of the calculated matching rate in the first embodiment. [Figure 10] It is a block diagram showing an example of a server according to the second embodiment. [Figure 11] It is a flowchart showing an example of processing related to job recommendation according to the second embodiment. [Figure 12] It is a diagram showing an example of a job narrowing-down screen on the worker side according to the second embodiment. [Figure 13] It is a diagram showing an example of a job list screen including recommended jobs according to the second embodiment. [Figure 14] It is a diagram showing an example of specific job information transitioning from the job list according to the second embodiment. [Figure 15] It is a diagram showing an example of a proposal screen to a business operator according to the second embodiment. [Figure 16] It is a diagram showing an example of a setting screen for a business operator according to the second embodiment. [Figure 17] It is a block diagram showing an example of a server according to the third embodiment. [Figure 18] It is a diagram showing the current status of each job according to the third embodiment. [Figure 19] It is a diagram showing an example of traffic information for each job according to the third embodiment. [Figure 20] It is a flowchart showing an example of processing related to job recommendation according to the third embodiment. [Figure 21] It is a diagram showing an example of a push notification according to the third embodiment. [Figure 22] It is a diagram showing an example of a job list screen according to the third embodiment. [Figure 23] It is a diagram showing an example of a job list screen according to the third embodiment. [Figure 24] It is a diagram showing an example of a setting screen for a business operator according to the third embodiment. [Figure 25] It is a diagram showing a flowchart showing an example of each process in the third embodiment.

MODE FOR CARRYING OUT THE INVENTION

[0009] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. The same elements are denoted by the same reference numerals, and redundant descriptions are omitted.

[0010] [Disclosed Technology] In the disclosed technology, a service (hereinafter also referred to as "job seeking service") that enables viewing, application, etc. of job information is provided on a platform that provides job information (also referred to as "job information providing platform"). Further, in the disclosed technology, not only the matching rate between a worker and a specific job is calculated, but also the matching rate for each job within a predetermined area is calculated, or appropriate jobs are proposed to the worker based on the position of the worker and the matching rate of each job. By using the job information providing platform to recruit workers for jobs, a business operator aims to eliminate labor shortages.

[0011] The "area" in the disclosed technology may mean a predetermined range based on the position of the worker. For example, GPS (Global Positioning System) position information from the processing terminal used by the worker is acquired, and a range within a predetermined distance based on this position information may be set as the area. Thereby, it becomes possible to identify job information for which the workplace is in an area including the position where the worker is currently located, rather than an area such as a general municipality or village.

[0012] According to the above processing, for example, in the job information providing platform, by appropriately calculating the matching rate of the worker for each job according to the predetermined area, it becomes possible to utilize it for setting job information or performing analysis of job information. Also, in the job information providing platform, not only within a predetermined area, but also by using the position of the worker and the matching rate of each job, it becomes possible to propose appropriate jobs according to the worker or adaptively change the salary, transportation expenses, etc.

[0013] [Example of System Configuration] Figure 1 shows examples of the configurations of information processing system 1 in the disclosure technology. Hereafter, workers and businesses may be collectively referred to as "users". In the example shown in Figure 1, each information processing device 10A, 10B, and 10C used by each user, an information processing device 20 that constitutes the system providing job posting services, and a database 30 that manages user information, job posting information, and usage history information of each worker are connected via network N.

[0014] Furthermore, if any number of information processing devices 10A, 10B, and 10C are connected to the network N and are not distinguished individually, they may also be referred to as information processing device 10. Also, if there are multiple devices, they may be referred to as the nth device (where n is an integer) to distinguish them.

[0015] The information processing device 10 is, for example, a smartphone, a computer, or a tablet device. The information processing device 10 can provide the job search service to users by installing an application (also called a "native app") that runs the job search service in the disclosure technology. Furthermore, if the job search service is implemented on a web page, the information processing device 10 can also use the job search service using a web application (also called a "web app") on a web browser. Hereafter, these applications will be collectively referred to as "job search apps."

[0016] The information processing device 20 is, for example, a server and may consist of one or more devices. The information processing device 20 also manages the job information provision platform, provides job information, registers users, and processes job viewing, applications, etc. Hereinafter, the information processing device 20 will also be referred to as the server 20.

[0017] Database 30 has a storage unit that stores or manages user information of each worker registered with the job placement service, job information of each business operator, usage history information of each worker, learning model information, etc. The following describes the various embodiments that realize the disclosed technology.

[0018] [First Embodiment] In the first embodiment, in addition to using the matching rate of a specific job, the matching rate for a given area is calculated from the number of people to be recruited and the number of applicants for each job within that area. In the first embodiment, the platform performs various processes to improve user convenience, such as suggesting the content of job information to businesses that set up job postings within a given area based on the area's matching rate, changing how workers view job information, and providing job information.

[0019] For example, in the first embodiment, instead of using pre-made master data such as general prefectures and municipalities, it is possible to identify an area based on the worker's location for each worker and adaptively set this area. In this case, by using an area appropriate to the worker, it becomes possible to show or suggest appropriate job information to the worker using the matching rate of the area appropriate to the worker.

[0020] <Example of user-side device configuration> Figure 2 is a block diagram showing an example of an information processing device 10 according to the first embodiment. The information processing device 10 includes one or more processing units (processors: CPUs) 110, one or more network communication interfaces 120, memory 130, user interface 150, and one or more communication buses 170 for interconnecting these components.

[0021] The user interface 150 is, for example, a user interface comprising a display 151 and an input device (such as a keyboard and / or mouse or some other pointing device) 152. The user interface 150 may also be a touch panel.

[0022] The memory 130 may be, for example, a high-speed random-access memory such as DRAM, SRAM, DDR RAM, or other random-access solid-state memory, or it may be a non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state memory devices. The memory 130 may also be a computer-readable non-temporary recording medium.

[0023] Another example of memory 130 may be one or more storage devices located remotely from the CPU 110. In one embodiment, memory 130 stores programs, modules, and data structures, or subsets thereof, relating to the following job application.

[0024] The operating system 131 includes, for example, procedures for handling various basic system services and for executing tasks using hardware.

[0025] The network communication module 132 is used, for example, to connect the information processing device 10 to another computer via one or more network communication interfaces 120 and one or more communication networks such as the Internet, other wide area networks, local area networks, and metropolitan area networks.

[0026] App data 133 includes data processed when a user uses the job search service. For example, app data 133 includes user information and information obtained from server 20a. Specifically, job search information related to job postings is included in app data 133.

[0027] The service processing module 134 executes various processes on the job information provision platform provided by the server 20a. For example, the service processing module 134 includes an acquisition module 135, an output module 136, and a processing module 137, which will be described later. The service processing module 134 may also change its processing content depending on whether the user logged into the job information provision platform is a business operator or a worker. Furthermore, the service processing module 134 corresponds to the job application described above.

[0028] For example, let's assume that information processing device 10A is a device used by users on the business side, and information processing device 10B is a device used by users on the worker side. In the following, the business side information processing device 10A will be denoted with the letter A, and the worker side information processing device 10B will be denoted with the letter B. Furthermore, in the following, we will describe an example in which processing is separated according to the logged-in user for a job application, but it is also possible to implement separate applications for businesses and workers. Note that information processing device 10C is also assumed to be a device used by users on the worker side, but since it is the same as information processing device 10B, the explanation will be omitted.

[0029] ≪Business side≫ First, let's explain each process in the information processing device 10A on the business side. The acquisition module 135A acquires user operations on buttons (an example of UI components) on the display screen shown on the display 151A. For example, the acquisition module 135A acquires commands corresponding to click operations on the job posting settings button on a web page or application screen (job posting settings screen) via the user interface 150A. The job posting settings button is a button that businesses use to set recruitment conditions, etc., when posting job openings.

[0030] Furthermore, the acquisition module 135A may acquire commands corresponding to click operations on the publishing settings button within a web page or application screen (publishing settings screen) via the user interface 150A. The publishing settings button is used by businesses when they want to set and publish job postings.

[0031] Output module 136A outputs job information and job setting requests set by the user on the business side to server 20.

[0032] Processing module 137A configures the job information as described above and, if necessary, performs verification and / or approval of workers who have applied for the job (also referred to as "applicant workers"). For example, processing module 137A accepts operations from the business operator, cooperates with server 20a, configures the conditions information for individual job postings included in the job information, and, if necessary, performs verification and / or approval of applicant workers. Processing module 137 may also perform approval processing as part of the automatic matching process for workers matched by automatic matching. Automatic matching includes automatically determining the compatibility between job information and worker information and automatically associating suitable workers with job information.

[0033] The display control module 138A controls the display of a single screen in the job application. For example, the display control module 138A controls the display of the job settings screen, the publication settings screen, and, if necessary, the applicant worker confirmation screen and the applicant worker work history information display screen on the display 151A.

[0034] ≪Worker's perspective≫ Next, we will describe the processes in the worker-side information processing device 10B. Based on the worker's operation, the operating system 131B downloads and installs the job application program from a designated website or the like to its information processing device 10B. This makes the service processing module 134B executable.

[0035] The service processing module 134B accesses a URL that provides a job information provision platform, retrieves screen information related to the job service from the server 20a via the job application, and sends information to the server 20a.

[0036] The acquisition module 135B acquires user operations on each UI component (buttons, input fields, setting items, etc.) within the display screen shown on the display 151B. For example, the acquisition module 135B acquires registration information set by the worker's user operations on the new user registration screen, or acquires job application information set by the worker's user operations on the job application information setting screen.

[0037] When new user registration information is set based on user operations, output module 136B sends the registration information and a new registration request to server 20a. The registration information includes, for example, the user's name, address, telephone number, etc., entered by the worker. When server 20a receives the new registration request, it processes the new user registration based on the registration information. When user job application information is set based on user operations, output module 136B sends the job application information and a setting request to server 20a. When server 20a receives the setting request, it processes the user's job application information based on the job application information.

[0038] Furthermore, if a worker performs an operation that requests to view job information (including viewing a list of job information), output module 136B outputs a viewing request to server 20a. Also, if a worker performs an operation to apply for a job from the job information viewing screen, output module 136B outputs an application request to server 20a.

[0039] Processing module 137B may perform other processing related to the job posting service. For example, it may have worker-side functions such as a notification function for favorite job postings and a messaging function with employers.

[0040] The display control module 138B controls the display of the job application screen. For example, the display control module 138B controls the display to show the new user registration screen in response to a new registration request, the job seeker information settings screen in response to a settings request, the job information viewing screen (including the job information list screen) in response to a viewing request, the job application screen in response to an application request, etc., on the display 151B.

[0041] One or more processing units (CPUs) 110 read and execute each module from the memory 130 as needed. For example, one or more processing units (CPUs) 110 may constitute a communication unit by executing a network communication module 132 stored in the memory 130. Alternatively, one or more processing units (CPUs) 110 may constitute a service processing unit, an acquisition unit, an output unit, a processing unit, and a display control unit by executing a service processing module 134, an acquisition module 135, an output module 136, a processing module 137, and a display control module 138, respectively, which are stored in the memory 130. Furthermore, the processing of each of the service processing module 134, the acquisition module 135, the output module 136, the processing module 137, and the display control module 138 may be executed by one or more processing units (CPUs) 110.

[0042] In other embodiments, the service processing module 134, acquisition module 135, output module 136, processing module 137, and display control module 138 may be standalone applications stored in the memory 130 of the information processing device 10. Standalone applications include, but are not limited to, acquisition applications, output applications, processing applications, and display control applications. In yet another embodiment, the service processing module 134, acquisition module 135, output module 136, processing module 137, and display control module 138 may be add-ons or plug-ins to other applications.

[0043] Each of the elements described above may be stored in one or more of the aforementioned storage devices. Each of the modules described above corresponds to a set of instructions for performing the functions described above. The modules or programs (i.e., sets of instructions) described above do not need to be implemented as separate software programs, procedures, or modules, and therefore various subsets of these modules may be combined or reconfigured in various embodiments. In one embodiment, memory 130 may store a subset of the modules and data structures described above. Furthermore, memory 130 may store additional modules and data structures not described above.

[0044] As mentioned above, the application for the business operator and the application for the worker may be implemented as separate applications. In this case, the worker can use the job application by downloading and installing the worker's job application program from a designated website or the like to their information processing device 10B, and the business operator can use the business's job application program by downloading and installing the business operator's job application program from a designated website or the like to their information processing device 10A.

[0045] <Example of server-side device configuration> Figure 3 is a block diagram showing an example of a server 20a according to the first embodiment. The server 20a includes one or more processing units (CPUs) 210a, one or more network communication interfaces 220a, memory 230a, and one or more communication buses 270a for interconnecting these components.

[0046] The server 20a may optionally include a user interface 250a, which may include a display device (not shown) and a keyboard and / or mouse (or some other pointing device or other input device; not shown).

[0047] Memory 230a is, for example, a high-speed random-access memory such as DRAM, SRAM, DDR RAM, or other random-access solid-state memory, and may also be a non-volatile memory such as one or more magnetic disk memory devices, optical disk memory devices, flash memory devices, or other non-volatile solid-state memory devices. Memory 230a may also be a computer-readable non-temporary recording medium.

[0048] Another example of memory 230a is one or more storage devices located remotely from the CPU 210a. In one embodiment, memory 230a stores the following programs, modules, and data structures, or subsets thereof.

[0049] The operating system 231a includes, for example, procedures for handling various basic system services and for executing tasks using hardware.

[0050] The network communication module 232a is used, for example, to connect the server 20a to another computer via one or more network communication interfaces 220a and one or more communication networks such as the Internet, other wide area networks, local area networks, and metropolitan area networks.

[0051] User information 233a includes information about users who use the job posting platform. For example, user information 233a includes the user's name, address, telephone number, available working hours, and evaluation, etc., associated with each user ID. User information 233a will be described later using Figure 5.

[0052] Job posting 234a includes one or more job postings registered on a job posting platform (or job service). For example, job posting 234a includes information such as the job posting date, working hours or days, employer, work location, hourly wage, and job description, associated with the job ID. Job posting 234a will be described later using Figure 6.

[0053] Usage history information 235a includes job ID, location, work days, work hours, salary, etc., associated with the user ID. Usage history information 235a may also include evaluations from the employer regarding the work performed. Usage history information 235a will be described later using Figure 7.

[0054] The learning model information 236a includes learning model information for estimating the matching rate of an area, which is learned using learning data that includes at least the matching rate and statistical data of each past job posting within an area identified based on predetermined location information. For example, the learning data is learning data that includes at least the matching rate of the number of completed jobs relative to the number of people recruited and statistical data on work for each past job posting. For example, the learning model information 236a includes the structure of the neural network for estimating the matching rate mentioned above, as well as hyperparameters. At least one of the user information 233a, job posting information 234a, usage history information 235a, and learning model information 236a may be stored in the database 30.

[0055] The service control module 237a provides a job posting service that makes job postings available to workers and manages processes related to job posting settings, publication, and matching on the job information provision platform. For example, the service control module 237a has an acquisition module 238a, a learning module 239a, an estimation module 240a, and an output module 241a for processes related to job posting settings, publication, and matching. The following mainly describes the process of determining the matching rate for each area. In the first embodiment, a learning model that estimates the matching rate is used to determine the matching rate for an area.

[0056] Estimation of area matching rate The acquisition module 238a acquires location information for predetermined locations. For example, the acquisition module 238a acquires locations such as the address of each worker, a location where they stay for a predetermined number of times or for a predetermined amount of time, a station used a predetermined number of times or more, the work location of a job posting that has been matched more than a predetermined number of times, or the worker's current location as a predetermined location. The acquisition module 238a may acquire the above-mentioned location information by analyzing data from past job applications, or it may acquire location information if such information is set for the worker, or it may acquire location information from a satellite positioning system.

[0057] The learning module 239a inputs predetermined statistical data into a learning model that estimates the matching rate of an area (predetermined range) identified based on location information. The learning model is a learning model trained using learning data that includes at least the matching rate of the number of people who completed work relative to the number of people recruited (also called the "work completion matching rate") and statistical data on the work days for each past job posting within the identified area. The area may not be a conventional division of a city, town, or village, but rather a range within a predetermined distance from a predetermined location using a map with a known regional mesh method (a map that divides a region on a map into a grid for statistical purposes), or a map using Hex (a map that uses a coordinate system with hexagons as a grid).

[0058] For example, the training data includes historical information about each job posting for each area, as well as data on employment (statistical data). (1) Example of statistical data • Date (month and day) of each workday, time of work, day of the week, whether it is a public holiday or not, and weather (actual weather, weather forecast data) (2) An example of data related to job postings • Number of cancellations (number of cancellations from the workday to the day before the scheduled date) • The matching rate between the number of job openings and the number of completed jobs. By learning the above training data, the learning model can be programmed with statistical data to determine the number of cancellations and the matching rate for a given workday in a specified area.

[0059] The learning module 239a may obtain regional weather forecast data, including location information for a predetermined location, from a service that publishes weather forecast data via an API, as weather forecast data included in predetermined statistical data to be input to the learning model. Examples of services that publish weather forecast data include the Japan Meteorological Agency's weather forecast website. The learning model may be able to specify any future day beyond the present for which the area matching rate is to be estimated.

[0060] The estimation module 240a uses data output from the learning model to estimate the matching rate for job postings on a given day in an area. For example, the estimation module 240a may estimate the matching rate for a given workday by inputting a predetermined calculation formula, which uses the area-specific matching rate output from the learning model as a parameter, into a calculation formula for calculating the matching rate for a given workday. The predetermined calculation formula may include a calculation formula that uses parameters such as the estimated number of people who will complete their work (number of job postings × estimated matching rate), the estimated increase in the number of job postings from the present to the given workday, and the estimated increase in the number of applicants.

[0061] Through the above process, it becomes possible to calculate the matching rate for a given area, and by specifying future work days, it becomes possible to calculate the matching rate for each area on those days. As a result, it becomes possible to make suggestions to businesses that set up job postings within a given area, based on the area's matching rate, to change how workers view job postings, and to perform various processes that improve user convenience as a job posting platform.

[0062] In addition to the processing described above, the acquisition module 238a may also include acquiring location information indicating the worker's position from a processing device used by the worker (e.g., information processing device 10B). For example, the location information indicating the worker's position may be GPS location information acquired by the worker's processing device, or location information for a location specified by the worker.

[0063] Through the above process, it becomes possible to adaptively set a predetermined area that includes the worker's current location and the location where the worker wishes to work, thereby meeting the worker's work location needs while appropriately estimating the area matching rate.

[0064] In addition to the processing described above, the learning module 239a may also train the learning model with training data that includes statistical data containing at least one data point for the date, day of the week, holiday, and weather of past working days. For example, the learning module 239a may use past days as working days and use the date, day of the week, whether or not it is a holiday, and weather (weather forecast and actual weather) data for each day as training data.

[0065] The above process makes it possible to improve the accuracy of the matching rate for a given area. For example, the matching rate can be improved by training the model with past data such as whether a day of the week was a weekend or a public holiday, and whether the weather was sunny or not. The learning model can calculate the matching rate for similar statistical data from past statistical data. Specifically, it becomes possible to determine the matching rate for a desired workday while taking into account past data such as the matching rate decreasing on rainy days and increasing on public holidays.

[0066] In addition to any of the above processing, learning module 239a may also train the learning model with learning data that includes at least the number of cancellations on working days. For example, learning module 239a may use the cancellation rate for each past day as learning data to train the matching rate. For example, it becomes possible to learn past data such as the fact that the cancellation rate is higher when rain is forecast.

[0067] By performing the above process, it becomes possible to improve the estimation accuracy of the matching rate for a given area by considering the cancellation rate of that area.

[0068] In addition to any of the above processing, estimation module 240a may also obtain estimated values ​​for the increase in the number of recruits and the increase in the number of applicants up to future work dates, which are estimated by a learning model based on current statistical data, the number of recruits, and the number of applicants. For example, the learning model can estimate the increase in the number of recruits and the increase in the number of applicants on a work day in the same area from past data. The learning model can also estimate the increase in the number of recruits and the increase in the number of applicants for sunny, cloudy, and rainy weather based on past weather data. The number of applicants may also be the number of people who actually completed their work (number of completed workers).

[0069] For example, the estimation module 240a inputs current statistical data, the number of people to be recruited, and the number of applicants in a predetermined area into a learning model, and obtains estimated values ​​from the learning model for the increase in the number of people to be recruited and the increase in the number of applicants until a future work date.

[0070] Estimation module 240a may include estimating the matching rate for future workdays using the current number of positions available, the number of applicants, and estimated values ​​for the increase in positions available and the increase in applicants. For example, estimation module 240a may use at least one function to estimate the matching rate for future workdays, with the current number of positions available, the number of applicants, and estimated values ​​for the increase in positions available and the increase in applicants as parameters.

[0071] Specifically, estimation module 240a may estimate the matching rate for future workdays by inputting the current number of applicants, the number of applicants, and the estimated values ​​for the increase in the number of applicants and the increase in the number of applicants into one or more of the functions described above. In addition to functions, estimation module 240a may also use a machine learning model that has learned the relationship between the current number of applicants, the number of applicants, the estimated values ​​for the increase in the number of applicants and the increase in the number of applicants, and the matching rate for future workdays as a method for estimating the matching rate for future workdays.

[0072] By performing the above process, it becomes possible to improve the accuracy of estimating the matching rate for a given area by considering the number of people to be recruited and the increase in the number of applicants up to the future work date in that area.

[0073] In addition to any of the above processes, estimation module 240a may also include estimating the matching rate for future workdays based on the number of cancellations for those future workdays. For example, estimation module 240a estimates the number of people who will cancel on the day by multiplying the number of applicants for that day by the cancellation rate for the same day in the past that corresponds to the future workday. In this case, estimation module 240a estimates the matching rate that takes into account the number of people who will cancel on the day.

[0074] By performing the above process, it becomes possible to improve the accuracy of estimating the matching rate for a given area by taking into account the cancellation rate on the day of work in that area.

[0075] Estimation module 240a may include estimating the increase in the number of available workers from the present to a future work date, and estimating the matching rate for that future work date based on the increase in workers. For example, estimation module 240a estimates the daily increase in the number of new workers using the job application app (new workers). For instance, if 5 new workers are added per day, and there are 3 days until the work date, it estimates that 15 (5 x 3) new workers will be added from the present to the work date. Estimation module 240a may also estimate the matching rate by taking this increase in new workers into account when calculating the number of applicants.

[0076] The increase in the number of available workers may be calculated by area, using the number of new workers on the job posting app on the same day in past data, or based on the average increase in new workers per day.

[0077] By performing the above process, it becomes possible to improve the accuracy of estimating the matching rate for a given area by taking into account the increase in the number of new workers in the job posting app.

[0078] In addition to any of the above-described processes, learning module 239a may also train a learning model using learning data that includes category information of past job postings. For example, learning module 239a may train a learning model that estimates the matching rate using learning data that includes information about the job duties of the job postings as category information.

[0079] The learning module 239a may include training the learning model with category information, including the type and / or content of the job postings. For example, the learning module 239a may include the job content included in the job postings shown in Figure 6 below as training data.

[0080] In this case, the estimation module 240a may include estimating the matching rate for each job category in a predetermined area. For example, the estimation module 240a estimates the matching rate not only for a predetermined area but also for each job category (such as job content). Furthermore, to enable the identification of specific job postings, the job categories may use information that allows for the identification of job postings.

[0081] By considering the job categories, the accuracy of estimating the matching rate for a given area can be improved. For example, since the matching rates for delivery work and cashier work differ, estimating the matching rate according to the job category allows for a more accurate estimation.

[0082] Output module 241a publishes job postings based on the publication date or period if the required number of applicants or hires have not been completed. For example, output module 241a may publish job postings to workers on the job posting platform when the publication date has arrived. Also, if location information indicating the worker's location is obtained, output module 241a may publish job postings where the work location is within a predetermined area based on the location information to that worker. For example, output module 241a may control the display of the job posting list screen to workers who have made a viewing request.

[0083] The output module 241a may output to a predetermined device the matching rate for a predetermined area obtained by the above-described process, estimated values ​​for the increase in the number of people to be recruited and the increase in the number of applicants until the future work date, and the matching rate for a predetermined work date in a predetermined area. For example, if the business operator uses the matching rate data as a reference when setting up job postings, the output module 241a may output it to the business operator's information processing device 10A, or if the worker uses it as a reference when applying, it may output it to the worker's information processing device 10B.

[0084] Through the above process, the matching rate for a given area can be used to propose to businesses the number of people to recruit in each area and working conditions such as salary, and can also be used to propose to workers the timing of their applications, etc.

[0085] ≪Specific Example 1≫ Next, we will explain, using a specific example, how to estimate the matching rate for designated working days in a designated area. For example, the historical data used as training data includes the following information for each day. Date (Month / Day of the current year), day of the week, public holidays, weather (actual weather and weather forecast data), number of cancellations (e.g., number of cancellations each day from 7 days prior to the work date), work completion matching rate (the ratio of the number of people who completed work to the number of applicants for the job) For example, a mesh method or Hexagonal grid may be used to divide the area. Furthermore, the aforementioned historical data may be obtained for each area. Regarding the areas, predetermined areas may be set based on location information obtained from the worker's terminal.

[0086] The learning module 239a uses the aforementioned training data to perform machine learning to estimate the matching rate of a given area and generates a trained model.

[0087] The learning module 239a inputs the current date, the date of the workday, and the weather on the workday as statistical data into the learning model. For example, let's assume that the learning module 239a has the following data input. • Current date Example: Friday, November 8, 2024 • Date of workday Example: Monday, November 11, 2024 • Weather forecast for the designated area on the day of work. Example: Predicted rainfall for each time period (1:00 0mm, 2:00 5mm, ..., 24:00 10mm) • Number of cancellations in the designated area (daily number of cancellations from 7 days prior to the work day) • Work completion matching rate (percentage of completed work)

[0088] Assume the following data was output from the learning model. • Estimated work completion matching rate: Example 1: 85% (November 11th) • Estimated number of completed jobs: Example: 85 people (estimated number of applicants: 100 x job completion matching rate: 0.85) • Estimated daily increase in the number of applicants: Example: (November 8th (Fri): 4 people, November 9th (Sat): 6 people, November 10th (Sun): 8 people, November 11th (Mon): 10 people) • Estimated daily increase in the number of applicants. Example: (November 8th (Fri): 6 people, November 9th (Sat): 10 people, November 10th (Sun): 16 people, November 11th (Mon): -4 people) November 11th will take into consideration the number of cancellations.

[0089] The following is an example of data showing the current matching status (as of November 8, 2024 (Friday)). Number of people recruited Example: 90 people Number of applicants: Example: 60 people

[0090] The estimation module 240a calculates the matching rate for a given area on a working day based on the data output by the learning model and the data indicating the current matching status, using, for example, the following formula. • Estimated number of people to be recruited on the day: 90 + 4 + 6 + 8 + 10 = 118 people • Estimated number of people who completed work on the day: 60 + 6 + 10 + 16 - 4 = 88 people • Estimated job completion matching rate: 2: 88 / 118 = 74.6% • Number of people who did not apply for a job compared to the number of applicants (number of positions not filled): 118 - 88 = 30 people

[0091] The estimation module 240a may, for example, perform a correction process using the increase in the number of new workers in a predetermined area who are likely to apply. • Estimated number of new applicants (new additions to the number of applicants): +10 people The estimated number of new applicants may be obtained by one of the following methods: Average daily increase in new workers in a given area: up to 10 people The ratio of the number of users × the number of available users in other services linked to the job posting platform within a designated area. • Estimated number of people who completed work on the day: For example, 88 + 10 (estimated number of new applicants) = 98 people • Estimated matching rate of workdays in a designated area: Example 3, 98 / 118 = 83% In the example above, the "estimated number of new applicants" was added to the "estimated number of people who completed work on the day." However, it is also possible to multiply the estimated number of new applicants by the probability of them completing work and then add that.

[0092] According to the example above, the final matching rate for workdays in the designated area is 83% (estimate 3), but it is not limited to this; estimate 1 or estimate 2 may also be used. As estimates 1 through 3 consider more data, the estimation accuracy increases in the order of estimate 1, estimate 2, and estimate 3. Note that the above example of calculating the estimates is merely one example and is not the only one that can be used.

[0093] ≪Specific Example 2≫ Next, we will explain how to estimate the matching rate for a specific category (which may be a specific job posting) using Specific Example 1. The training data includes the training data from Specific Example 1, as well as the types of work and / or job descriptions of past job postings in a given area. The learning model learns from the training data, including the types of work and / or job descriptions of past job postings, to estimate the matching rate for a specific job posting on future workdays. The following example explains how to use a specific job posting as the specific category.

[0094] The learning module 239a uses the aforementioned training data to perform machine learning to estimate the matching rate of a given area and generates a trained model.

[0095] In addition to the statistical data from Specific Example 1, learning module 239a inputs the current date, the date of the workday, and the weather on the workday into the learning model. • Number of openings for a specific job: Example: 5 people

[0096] Assume the following data was output from the learning model. • Estimated job completion matching rate in a designated area: e.g., 85% • Estimated matching rate for the completion of work for specific job postings in a designated area: Example 4, 80% • Estimated number of people who have completed work for a specific job in a designated area: Example: 4 people

[0097] The following is an example of data showing the current matching status (as of November 8, 2024 (Friday)). • Number of people to be recruited for a specific job in a designated area: e.g., 5 people • Number of applicants for a specific job in a designated area: e.g., 3 people

[0098] Next, the estimation module 240a corrects the estimated number of final completed workers based on the relationship between the number of applicants in the past and the number of final completed workers. Figure 4 shows an example of the correlation between the number of applicants three days before application and the number of final completed workers according to the first embodiment. The correlation shown in Figure 4 allows for the calculation of the average value from the data of the number of completed workers each day in the past and the number of applicants three days prior. In specific example 2, there were 3 applicants three days before the work day on November 11th, on November 8th. In the example shown in Figure 4, when there are 3 applicants, the average number of final completed workers is 4.5.

[0099] The estimation module 240a calculates the matching rate for specific job postings in a given area using, for example, the following formula. • Estimated matching rate on the day of work for a specific job posting: Example 5, 4.5 / 5 = 90%

[0100] It should be noted that the above example of estimating the matching rate for a specific job posting is merely one example and is not limited to that example. For example, a learning model could be generated to estimate the matching rate for work days in a specific job posting.

[0101] The above specific examples 1 and 2 illustrate how to calculate the matching rate for workdays in a given area using specific numerical values. In the first embodiment, in addition to the matching rate, it becomes possible to obtain other useful data for the job posting platform, such as the number of people who did not apply for a job on a given workday (the number of people whose positions were not filled) being 30.

[0102] Each of the elements described above may be stored in one or more of the aforementioned storage devices. Each of the modules described above corresponds to a set of instructions for performing the functions described above. The modules or programs (i.e., sets of instructions) described above do not need to be implemented as separate software programs, procedures, or modules, and therefore various subsets of these modules can be combined or reconfigured in various embodiments. In one embodiment, memory 230a may store a subset of the modules and data structures described above. Furthermore, memory 230a may store additional modules and data structures not described above.

[0103] One or more processing units (CPUs) 210 read and execute each module from the memory 230a as needed. For example, one or more processing units (CPUs) 210a may constitute a communication unit by executing the network communication module 232a stored in the memory 230a. Alternatively, one or more processing units (CPUs) 210a may constitute a service control unit, acquisition unit, learning unit, estimation unit, and output unit by executing the service control module 237a, acquisition module 238a, learning module 239a, estimation module 240a, and output module 241a, respectively, which are stored in the memory 230a. Furthermore, the processing of each of the service control module 237a, acquisition module 238a, learning module 239a, estimation module 240a, and output module 241a may be executed by one or more processing units (CPUs) 210a.

[0104] Figure 3 shows a “server,” but it is intended to illustrate the various features that may exist in a set of servers rather than to provide a structural overview of the embodiments described herein. In practice, as will be apparent to those skilled in the art, the items shown separately may be combined, and some items may be configured separately. For example, the items shown separately in Figure 3 may be implemented on a single server, and a single item may be implemented by one or more servers.

[0105] The database 30 may have a configuration similar to the hardware configuration shown in Figure 3. Furthermore, the user information 233a, job posting information 234a, usage history information 235a, and learning model information 236a shown in Figure 3 may be stored in the storage unit of the database 30.

[0106] <Example of data structure> Figure 5 shows an example of user information 233a according to the first embodiment. User information 233a manages information about each member user created by a user (e.g., a worker) using the job recruitment service. The "User ID" includes user identification information (User ID: Identifier) ​​that the server 20a uses to uniquely identify the user. The "User Information" is associated with the User ID.

[0107] "User information" includes personal information of the user, such as "name," "address," "telephone number," "available working hours," and "rating." "Available working hours" includes time information set based on the usage history information 235 described below, and may also include available working hours information set by the worker.

[0108] "Evaluation" includes a value that the employer uses to evaluate the user's work attitude, etc., after the user has worked. "Evaluation" can be an average value or a cumulative value. The user ID may also be included as part of the user information. User information may also include email address, password, work location information, job category information, desired salary information, etc.

[0109] Figure 6 shows an example of job information 234a according to the first embodiment. The "Job ID" includes identification information for the job information. The job information includes various pieces of information associated with the Job ID, such as "Job posting date," "Working hours," "Employer," "Work location," "Salary," "Transportation expenses," and "Job description." In the example shown in Figure 6, "Salary" is used, but "Hourly wage" may also be used. The "Work location" may be specified, for example, by specifying a station, n meters from the station, or n meters from a designated location.

[0110] "Job posting date" includes the date (or period) on which the employer needs the worker. "Working hours" includes the worker's start and end times on the job posting date. "Employer" includes the name of the employer who has posted the job information and is seeking the worker. "Work location" includes at least one of the store name, place name, location name, facility name, etc., where the worker will work. "Salary" includes the salary for the job. "Transportation expenses" includes the salary paid to the worker. "Job description" includes information indicating the content of the work (job category) set by the employer. Job posting 234a may also include other information such as the type of work being offered (information that is specified by a higher-level concept than the job description).

[0111] Figure 7 shows an example of usage history information 235a according to the first embodiment. The "User ID" in the usage history information includes identification information of the user using the job recruitment service. The usage history information includes information such as "Job ID," "Location," "Working Days," "Working Hours," and "Salary" associated with this User ID. In the example shown in Figure 7, "Salary" is used, but "Hourly Wage" may also be used.

[0112] The "Job ID" field includes identification information that identifies the first job that the worker was hired for. The "Location" field includes the worker's workplace. The "Working Days" field includes the days the worker worked. The "Working Hours" field includes the worker's working hours on the job posting date. The "Hourly Wage" field includes the hourly wage for that first job. In addition, the usage history information may include information about the "Employer" and evaluation information about the work. The usage history information may also include the number of times the worker worked within a specified period, the total working hours, etc.

[0113] Although not shown in the diagram, the learning model information 236a includes a model that estimates the matching rate for a given day in a given area using each data point included in the training data, and the trained parameters of that model. The learning model information 236a may also temporarily include data during or after training.

[0114] <Operation Description> Next, the operation of the information processing system 1 according to the first embodiment will be described. Figure 8 is a flowchart showing an example of the process related to estimating the matching rate of a predetermined area according to the first embodiment. In the example shown in Figure 8, the server 20a is a device that implements a job information provision platform and provides job services, and the flowchart shows the process executed by this server 20a.

[0115] In step S102, the acquisition module 238a acquires location information for a predetermined location. For example, the predetermined location includes the address of each worker, a location where they stay for a predetermined value or a predetermined time or longer, a station used a predetermined number of times or more, the work location of a job posting with a predetermined number of matches or more, or the current location. It is sufficient that the location information to be used is set in advance.

[0116] In step S104, the learning module 239a inputs predetermined statistical data into a learning model that estimates the matching rate of an area (predetermined range) identified based on location information. The learning model used is the previously trained learning model described above. For example, the range of an area may be defined by a map demarcated using a known regional mesh method or Hex. For example, the predetermined statistical data may include at least one of the following: the date (month and day) of each day (working day), the day of the week, whether it is a public holiday or not, and the weather (actual weather, weather forecast data).

[0117] In step S106, the estimation module 240a estimates the matching rate for a predetermined day using the matching rate of an area (predetermined range) output from the learning model. For example, the estimation module 240a may use the matching rate of an area output from the learning model as the final matching rate, or it may input this matching rate as a parameter into a predetermined calculation formula for calculating the matching rate for a predetermined work day. As an example of step S106, the processing in step S162 or step S164 may be performed. In step S162, the estimation module 240a may obtain estimated values ​​for the increase in the number of people to be hired and the increase in the number of applicants up to a future work day, which are estimated (output) by the learning model based on the current statistical data, the number of people to be hired, and the number of applicants. In this case, the estimation module 240a may estimate the matching rate for a future work day using the current estimated number of people to be hired, the number of applicants, the increase in the number of people to be hired, and the increase in the number of applicants. In step S164, the estimation module 240a may train the learning model with learning data that includes category information of past job postings. In this case, the estimation module 240a may estimate the matching rate for each job category.

[0118] Here, Figure 9 shows an example of how the matching rate calculated in the first embodiment is displayed. The example shown in Figure 9 shows an example of a screen where a business operator sets up job postings. In the example shown in Figure 9, the number of applicants relative to the number of positions available in the current job posting is displayed. For example, for a job posting starting at 6 PM on July 25th, the number of applicants is 2 / 5 (number of applicants / number of positions available), indicating that the number of applicants has not yet reached the required number. At this time, the matching rate for a predetermined area including this work location, which was obtained in the first embodiment, may also be displayed. In the example shown in Figure 9, "Estimated matching rate at current hourly wage: 85%" is displayed.

[0119] Returning to Figure 8, in step S108, the estimation module 240a may estimate the matching rate in a predetermined area using the data that can be output from the learning model. For example, the estimation module 240a may calculate Estimate 2 or Estimate 3 as described in Specific Example 1.

[0120] In step S110, the estimation module 240a may estimate the matching rate of a specific job in a given area using data that can be output from a learning model that has learned training data including job categories. For example, the estimation module 240a may calculate the estimated value 4 or 5 described in Specific Example 2.

[0121] Through the above process, it becomes possible to calculate the matching rate for a given area not only by using the matching rate of a specific job posting, but also by using data such as the number of applicants and the number of positions available for each job posting within that area. This enables various processes to improve user convenience as a job posting platform, such as suggesting job posting content based on the area's matching rate to businesses that post jobs within a given area, or changing how workers view job postings.

[0122] Furthermore, in the first embodiment, instead of using pre-made master data such as general prefectures and municipalities, it is possible to identify an area based on the worker's location and adaptively set the area for each worker. In this case, by using an area appropriate to the worker, it becomes possible to show or suggest appropriate job information to the worker using the matching rate of the area appropriate to the worker.

[0123] [Second Embodiment] Traditionally, the use of worker location information has not been given much consideration in job posting platforms, and there was room for improvement in providing services using worker location information. Therefore, in the second embodiment, in order to solve this problem, the job posting platform performs various processes to improve user convenience, such as recommendations and dynamic pricing based on the matching rate between worker location and each job posting.

[0124] <Example of user-side device configuration> The information processing device 10 according to the second embodiment has the same configuration as described in the first embodiment. Any processing specific to the second embodiment will be described as needed.

[0125] <Example of server-side device configuration> Figure 10 is a block diagram showing an example of a server 20b according to the second embodiment. The server 20b is the same as in the first embodiment and includes one or more processing units (CPUs) 210b, one or more network communication interfaces 220b, memory 230b, and one or more communication buses 270b for interconnecting these components. The following mainly describes the processing that differs from the first embodiment.

[0126] The acquisition module 238b acquires location information of the worker's location. For example, when each worker accesses the job posting service or requests to view job postings, the acquisition module 238b acquires information indicating each worker's current location (GPS location information of the worker's terminal). The service control module 237b may allow workers to set a filter for job postings within an area based on their location information. For example, the service control module 237b may filter the search to include jobs within a radius of less than 20 km centered on the worker's location information and / or within a 30-minute commute time (see, for example, Figure 12).

[0127] The learning module 239b has a learning model that estimates the matching rate of each job posting, which is trained using learning data that includes at least the matching rate of each past job posting and the job posting information. For example, the learning module 239b inputs the job posting information, the number of people to be hired, and the number of applicants for each currently open job posting into the learning model that estimates the matching rate of each job posting, which is trained using learning data that includes at least the matching rate of the number of people to be hired relative to the number of people to be hired for each past job posting and the job posting information. As a specific example, the learning module 239b inputs the job posting information for each job posting, the total number of people to be hired for each job posting, and the total number of applicants into the learning model. The learning method in the second embodiment is basically the same as in the first embodiment, although the learning data may differ.

[0128] The estimation module 240b uses data output from the learning model to estimate the matching rate for each currently recruiting job on a specified date. The specified date can be any specific day, including working days. For example, the estimation module 240b may use the matching rate output by the learning model for each job output from the learning model, or it may use a corrected matching rate as in the first embodiment.

[0129] The configuration module 242b sets the jobs to recommend to the worker, using the matching rate of each job on a given day estimated by the estimation module 240b, and the distance between the worker's location (based on the worker's location information) and the work location of each currently recruiting job. For example, the configuration module 242b prioritizes recommending jobs to the worker that are close to the work location.

[0130] In this case, output module 241b may prioritize displaying job information recommended by configuration module 242b to workers. For example, output module 241b may control the display of a job list screen with recommended job information at the top to workers who have made a viewing request (see, for example, Figure 13).

[0131] Through the above process, the matching rate for each job posting can be calculated, and based on that matching rate, job recommendations can be made based on the worker's location. Because workers are recommended jobs that take their location and matching rate into consideration, they are more likely to find a job that matches their needs.

[0132] The learning module 239b may train the learning model with training data that includes job information of jobs that the worker has applied for in the past. As a result, the learning model will output a matching rate for each currently recruiting job that is similar to the job information of jobs that the worker has applied for in the past.

[0133] The job information learned by the learning model includes at least one of the following: type of work, job description, working hours, work location, and salary. For example, learning module 239b can learn the job description, work location, and salary (e.g., above a certain amount) that a worker prefers by training the learning model with application data such as job description, work location, and salary from jobs the worker has applied for in the past.

[0134] In this case, the setting module 242b may set each job to recommend to the worker based on the matching rate of each job on a given day estimated by the estimation module 240b, and trend data based on the worker's past data (job content, work location, salary, etc.).

[0135] Through the above process, by using workers' past application data as training data, it becomes possible to recommend job postings according to the workers' application trends.

[0136] In addition to any of the above-described processes, the learning module 239b may also include inputting job information, the number of positions available, and the number of applicants for each currently open job within an area (a predetermined range) identified based on the worker's location information into a learning model trained using learning data of past job postings within that area. The learning method in the predetermined area is the same as in the first embodiment.

[0137] For example, the learning module 239b can reduce the amount of learning data and perform learning efficiently by identifying each job posting within a predetermined area based on the worker's location.

[0138] By performing the above steps, the server's processing resources can be used efficiently, which can also contribute to faster processing.

[0139] Proposal module 243b proposes that businesses that create a first job posting should be given an incentive to workers if the matching rate on a given day is below a predetermined value. For example, proposal module 243b proposes that businesses that create first job postings should be given a special incentive to workers if the matching rate on a given day is 50% or less, or if the recommendation ranking is 5th or lower. The source of the incentive may be contributed by the job posting business or the operator of the job information provision platform.

[0140] Furthermore, incentives may be adjusted according to the number of workers (potential workers) who are located near the work location (residents, frequent users, or currently present). For example, if there are many potential workers, matching may occur without any effort, but if there are few potential workers, incentives should be actively offered to encourage applications from workers.

[0141] Through the above process, we can make suggestions to businesses to improve the matching rate. As a result, there is a possibility that the number of applications from workers will increase, which can contribute to alleviating the labor shortage for businesses.

[0142] In addition to any of the above processes, proposal module 243b may also include proposing an increase in the salary information of the first job. For example, proposal module 243b may propose an increase of a predetermined amount or an increase of a predetermined percentage to the salary of the first job (see, for example, Figure 14). Note that incentives may also include benefits for workers other than salary increases, such as increased transportation expenses, provision of predetermined vouchers, or increased points.

[0143] Through the above process, it becomes possible to offer specific suggestions as incentives that can contribute to improving the matching rate, such as enabling dynamic pricing.

[0144] In addition to any of the above-described processes, proposal module 243b may also include proposing an increase in the salary information of the first job posting based on the salary information of job postings similar to the first job posting. For example, proposal module 243b may propose a predetermined percentage increase, a predetermined amount increase, etc., based on the salaries of other job postings that have similar job duties to the first job posting and / or have a matching rate of a predetermined value or higher. As a specific example, if the salary of a job posting similar to the first job posting and with a matching rate of 90% or higher is higher, proposal module 243b may propose to the employer of the first job posting an increase in salary to bring it up to the same level as the salary of that job posting.

[0145] Through the above process, it becomes possible to improve the justification for increasing incentives based on objective data.

[0146] In addition to any of the processes described above, learning module 239b may also include inputting the job information of the first job posting, including the increased salary information, into the learning model. For example, if a salary increase is proposed by proposal module 243b and approved by the employer, learning module 239b retrains the learning model using the job information of the first job posting, including the increased salary information.

[0147] In this case, the estimation module 240b may also include obtaining the matching rate of the first job based on the increased salary information from the learning model, and using this matching rate to estimate the increase in the matching rate for the first job on a given day. For example, the estimation module 240b may estimate the increase in the matching rate for the working day of the first job from the matching rates before and after the salary increase, which are output from the learning model. As a specific example, if the matching rate before the salary increase is 85% and after the salary increase it becomes 92%, the increase in the matching rate is 7%. Note that estimating the increase in the matching rate is synonymous with estimating the matching rate after the increase.

[0148] The output module 241b may output the increase in the matching rate obtained by the above-described process to a predetermined device. For example, the output module 241b may output data on the increase in the matching rate due to the provision of incentives to the information processing device 10A of the business operator. When the output module 241b makes a proposal on the screen showing the application status of the business operator's job postings, it may display the screen shown in Figure 15 described later, and when the output module 241b makes a proposal on the settings screen for setting up job postings, it may display the screen shown in Figure 16 described later.

[0149] Through the above process, it is possible to determine the change in the matching rate when incentives (especially salary) are provided. Furthermore, by presenting the increase in the matching rate to the business operator as a numerical value, the business operator can be made to understand the effect of providing incentives. In addition, estimation module 240b may estimate the amount of increase in the matching rate using the matching rate of the first job posting based on the planned salary increase, from a pre-trained learning model, without obtaining approval from the business operator. In this case, output module 241b can notify the business operator how much the matching rate will increase with this salary before approval is granted.

[0150] 《Specific Example 3》 Next, we will explain how to set up the recommended job postings in the second embodiment using a specific example. For example, the training data may include job posting information for each job. The training model can be generated, for example, by training the training model for estimating the matching rate of a specific job posting, as explained in Specific Example 2, with the job posting information for each job.

[0151] Learning module 239b calculates, for example, the work completion matching rate for the workday of November 11, based on the data as of November 8. For example, learning module 239b identifies job openings within an area based on worker location information and generates a job listing that includes the identified job openings. For each job opening in the job listing, learning module 239b takes the current number of applicants / number of openings (as of November 11) and the distance from the location of a specific worker to the work location as input data. A specific worker is a worker whose location information is obtained, and includes workers who request to view job openings or access job services. The following is an example of input data to be entered into the learning model. Input data Current number of applicants / number of openings for each job posting (A, B, C), distance (distance between the specific worker and the job posting's work location) Job posting A: 2 / 3, distance 1km ·Recruitment B: 3 / 4, distance 2km ·Recruitment C:0 / 2, distance 3km

[0152] Assume the following data (job completion matching rate) is output from the learning model. Note that the output data may also include the job completion matching rate for a specific job, as in Example 2. Note that the job completion matching rate in Example 3 may be the application rate, which uses the number of applicants for each workday without considering cancellations. Job completion matching rate Job posting A: 95% ·Recruitment B:90% ·Recruitment C: 60% In the calculation formula described later, parameter Z represents this work completion matching rate.

[0153] The configuration module 242b assigns a first weight to the work location of the job that a specific worker applies for, based on the worker's past application data. The first weight is a weight related to the distance between the worker's location and the work location of the applied job, using the worker's past application data. The first weight may be a pre-set value or a value obtained by a learning model. The first weight may also be calculated at predetermined distance intervals. An example of the first weight value is shown below. First weight ·Distance 0~3km:2 ·Distance 3~5km:1 ·Distance 5~20km: 0.5 • Distance 20~km: 0.1

[0154] Next, the configuration module 242b calculates a second weight for the job completion matching rate (or application rate) using the following formula. The second weight is the weight for the matching rate for a specific job posting. Second weight: 1.5-Z Job posting A: 0.55 (=1.5 - 0.95) ·Recruitment B:0.6(=1.5-0.9) ·Recruitment C:0.9(=1.5-0.6)

[0155] Next, the configuration module 242b obtains a third weight value related to the affinity between a specific worker and each job posting, which is output from the learning model. The third weight is the weight related to the affinity between each job posting A-C and the specific worker, using the worker's past application data. Third weight Job posting A: 1.0 ·Recruitment B:0.8 ·Recruitment C:1.6

[0156] The configuration module 242b calculates the recommendation priority for each job posting by using the following calculation, which involves the first to third weights. The recommendation priority (total weight) for each job posting = 1st weight + 2nd weight + 3rd weight Job posting A: 3.55 (=2 + 0.55 + 1.0) ·Recruitment B:3.4(=2+0.6+0.8) ·Recruitment C:3.6(=1+0.9+1.6)

[0157] The configuration module 242b sets the recommendation priority for each job posting A to C in the following order. 1.Recruitment C(3.6) 2. Job posting A (3.55) 3.Recruitment B(3.4)

[0158] For example, output module 241b may publish job information to workers in the order of priority set by configuration module 242b. According to the above example, output module 241b configures the job list screen so that job C, job A, and job B are displayed on the screen in that order.

[0159] The job recommendation method in Specific Example 3 is merely an example and is not limited to the above example. For example, it is sufficient to use at least one of the first to third weights, and the calculation method is not particularly important as long as jobs that are close to and highly compatible with a specific worker receive higher priority. Regarding the display of recommended jobs, it is sufficient that the recommended jobs are highlighted and displayed to distinguish them from other jobs.

[0160] <Example of data structure> The various data in the second embodiment are the same as the various data corresponding to the first embodiment. In the second embodiment, the data output from the learning model and the final output data are different, but it is sufficient that they be stored in memory 230b.

[0161] <Operation Description> Next, the operation of the information processing system 1 according to the second embodiment will be described. Figure 11 is a flowchart showing an example of processing related to job recommendation according to the second embodiment. In the example shown in Figure 11, the server 20b is a device that implements the job information provision platform and provides the job service according to the second embodiment, and the flowchart shows the processing performed by this server 20b.

[0162] In step S202, the acquisition module 238b acquires location information indicating the worker's location. For example, the location information may be GPS location information installed in the information processing device 10B used by the worker. For example, the service control module 237b may narrow the search to a radius of less than 20km centered on the worker's location information and / or a commute time of 30 minutes or less (see, for example, Figure 12).

[0163] Figure 12 shows an example of a job filtering screen for workers according to the second embodiment. The screen shown in Figure 12 is used by workers to filter the displayed job postings when they request to view job postings. In the example shown in Figure 12, the filtering items include start time, working hours, compensation, whether or not transportation expenses are covered, distance, and commute time.

[0164] For "Start Time," you can select the time when the job starts (e.g., 12:00-15:00). For "Working Hours," you can select at least one option from, for example, less than 3 hours, 3-4 hours, 4-5 hours, 5-6 hours, etc. For "Benefits," you can select at least one option from, for example, no experience necessary, casual dress code, hair color OK, nail polish OK, etc. For "Transportation Allowance," you can choose to display only jobs that include transportation allowance, or to display jobs regardless of whether transportation allowance is included.

[0165] "Distance" and "Commute Time" are examples of filtering options based on the worker's location information. "Distance" allows users to select, for example, a radius of a certain number of kilometers from the worker's current location to display job postings. "Commute Time" allows users to select a minimum commute time from the worker's current location to the workplace. The worker's current location can be a user-specified location (such as their home address or frequently used train station). Route searches can be performed using the API of a publicly known web service, allowing users to set their starting point and destination to obtain the optimal transportation route (train, bus, etc.).

[0166] By setting distance and commute time, workers can search for job postings based on their location, using the matching rate described below.

[0167] Returning to Figure 11, in step S204, the learning module 239b inputs the job information, number of positions available, and number of applicants for each currently open job into a learning model that estimates the matching rate of each job, which has been trained using training data that includes at least the matching rate of each past job and the job information. For example, the learning module 239b inputs the job information for each job, the total number of positions available for each job, and the total number of applicants into the learning model.

[0168] In step S206, the estimation module 240b uses the data output from the learning model to estimate the matching rate for each currently recruiting job on a specified date. The specified date is, for example, a working day. For example, the estimation module 240b may use the matching rate output by the learning model for each job output from the learning model, or it may use a corrected matching rate as in the first embodiment.

[0169] In step S208, the configuration module 242b recommends jobs to the worker, using the matching rate of each job on a predetermined day estimated by the estimation module 240b, and the distance between the worker's location and the work location of each currently recruiting job, based on the worker's location information. For example, the configuration module 242b prioritizes recommending jobs to the worker that have a low matching rate and are close to the worker's location. As a method of recommendation, for example, the output module 241b may publish job information to the worker in the priority order set by the configuration module 242b (see, for example, Figure 13).

[0170] Figure 13 shows an example of a job listing screen including recommended jobs according to the second embodiment. In the example shown in Figure 13, the job listings for a specified day (e.g., June 5th) are shown from the daily job listings, and the worker has selected "closest distance / shortest commute time" as the display priority. When "closest distance / shortest commute time" is selected, the two "recommended" jobs at the top are jobs recommended by the setting module 242b.

[0171] As a concrete example, if a worker selects "closest distance / shortest commute time," the acquisition module 238b acquires the worker's location information (the timing of acquiring the worker's location information is irrelevant). The learning module 239b outputs the matching rate for each job posting, and the setting module 242b sets a higher priority for jobs that are closer to the worker's location, for example. As a result, a screen like the one shown in Figure 13 is displayed. Suppose the worker selects job posting W10. At this point, the screen transitions to the screen shown in Figure 14.

[0172] Figure 14 shows an example of specific job information accessed from the job listings according to the second embodiment. In the example shown in Figure 14, job information for job W10 is displayed, and it can be seen that the salary for the job has been increased ("10% higher than usual due to weather conditions") because the weather data for the working day is bad (rain, etc.) based on statistical data. The salary has been increased by 10% from 8,000 yen due to the weather forecast (dynamic pricing).

[0173] Figure 15 shows an example of a proposal screen for businesses according to the second embodiment. In the screen shown in Figure 15, the proposal module 243b makes a proposal to provide incentives to job postings with low matching rates. For example, the proposal module 243b proposes to two job postings, one starting at 6 PM and the other at 5 PM on July 25th, "Why not raise the hourly wage to improve the matching rate?" These two job postings have ample time before the start date, the number of applicants / number of positions is "2 / 5", and the current matching rate is not high, so the proposal module 243b makes a proposal to provide incentives.

[0174] Proposal module 243b may also determine the amount of the hourly wage increase and the increase in the matching rate resulting from the wage increase, based on other job postings. In the example shown in Figure 15, proposal module 243b suggests to the employer that increasing the hourly wage by 100 yen would result in a 100% matching rate. This allows the employer to understand how many positions can be filled by providing incentives while monitoring the application status.

[0175] Figure 16 shows an example of a business operator's settings screen according to the second embodiment. In the screen shown in Figure 16, the amount and matching rate are displayed in association with the hourly wage setting item on the job information settings screen. For example, the suggestion module 243b calculates the hourly wage and matching rate from past data based on the matching rate on working days, the job description of the job, weather data, etc. In the example shown in Figure 16, the suggestion information for the hourly wage is displayed as "1,250 yen or more with a matching rate of 90%". This allows the business operator to set up the job information based on suggestions from the job provision platform. Also, at the stage of the settings screen shown in Figure 16, worker location information does not necessarily have to be used to estimate the matching rate.

[0176] Through the above processing, according to the second embodiment, various processes such as recommendations and dynamic pricing become possible as a job information provision platform, based on the worker's location and the matching rate of each job posting, in order to improve user convenience.

[0177] [Third Embodiment] Traditionally, on job posting platforms, workers had to independently research transportation information to their workplaces, which involved the time and effort of checking transportation costs and travel time. Furthermore, there are currently no services that allow users to browse job postings while considering transportation information. Therefore, in the third embodiment, in order to solve this problem, the job posting platform will perform various processes to improve user convenience, such as recommendations and dynamic pricing based on transportation information from the worker's location.

[0178] <Example of user-side device configuration> The information processing device 10 according to the third embodiment has the same configuration as described in the first and second embodiments. Any processing specific to the third embodiment will be described as needed.

[0179] <Example of server-side device configuration> Figure 17 is a block diagram showing an example of a server 20c according to the third embodiment. The server 20c is the same as in the first and second embodiments and includes one or more processing units (CPUs) 210c, one or more network communication interfaces 220c, memory 230c, and one or more communication buses 270c for interconnecting these components. The following mainly describes the processing that differs from the first and second embodiments.

[0180] The acquisition module 238c acquires location information indicating the location of a worker. The worker's location information is the same as the location information described in the second embodiment. For example, a worker whose location information has been acquired is designated as a specific worker.

[0181] The configuration module 242c recommends jobs to workers based on location information acquired by the acquisition module 238c, and on transportation information to the work locations of each job within a predetermined range identified based on that location information. For example, the configuration module 242c sets the jobs to recommend to workers based on the transportation information from this location to the work location of each job within the area identified based on the location information acquired by the acquisition module 238c. For example, the configuration module 242c sets the jobs to recommend to a specific worker based on the transportation route from the specific worker's location to the work location for each job with a work location within a predetermined area (predetermined range), taking into account factors such as ease of commuting, travel time, and effort required for travel.

[0182] As a specific example, the configuration module 242c uses the location information of a specific worker who requested to view job postings to identify job postings within a 20km radius, and obtains a transportation route from the specific worker's location to the work location of each identified job posting. The transportation route may be obtained from a publicly known transportation information service that can obtain transportation route information by setting a departure point and destination. The obtained transportation route may have multiple options, each showing a different route from the departure point to the destination.

[0183] For example, the configuration module 242c can obtain a transportation route by using the API published by the transportation information service to set the worker's location as the departure point and the work location as the destination point via the API. The transportation route information may include transportation costs, number of transfers, walking distance, and travel time.

[0184] In addition to any of the above processes, the configuration module 242c may also select and set job postings to recommend to specific workers based on predetermined recommendation conditions for the acquired transportation route information. The predetermined conditions include at least one of the following: travel time is less than or equal to a predetermined time, the number of transfers is less than or equal to a predetermined number, and transportation costs are less than or equal to a predetermined amount. The more burdensome the transportation route is for the specific worker (for example, fewer transfers, shorter travel time, lower transportation costs, etc.), the higher the recommendation priority. On the other hand, the configuration module 242c may also be set to give a higher recommendation priority to job postings with unfavorable conditions indicated by the transportation route, in order to facilitate matching.

[0185] The output module 241c publishes the job information for each job configured by the configuration module 242c to specific workers. For example, the output module 241c may configure the job screen so that jobs with higher recommendation priority are displayed at the top of the job list.

[0186] Through the above processes, it becomes possible to implement various features that improve user convenience as a job information provision platform, such as recommendations based on traffic information from the worker's location and dynamic pricing.

[0187] In addition to any of the above-described processes, the learning module 239c inputs the number of people to be hired and the number of applicants for each job currently being advertised in the area into a learning model that estimates the matching rate of the area, which is trained using learning data that includes at least the matching rate of the number of people who completed work relative to the number of people being hired for each past job posting within the area (predetermined range). For example, the learning model is a model that estimates the number of people who completed work / number of people being hired for each job posting on future work days, based on the current application status (number of applicants / number of people being hired) for each job posting, in addition to traffic information. The learning model in the third embodiment may be the learning model in the second embodiment. For example, the recommended job postings are sorted by weighting using the location of a specific worker.

[0188] The configuration module 242c may also include setting each job to recommend to workers based on the matching rate of each currently available job output from the learning model. For example, the configuration module 242c may set jobs with a matching rate lower than a predetermined value as jobs to recommend to workers in order to increase the matching rate as much as possible.

[0189] Through the above process, it becomes possible to set job recommendations for workers based not only on transportation information but also on the matching rate of each job posting.

[0190] In addition to any of the processes described above, the configuration module 242c may also include obtaining transportation information for each job posting identified based on predetermined conditions. The predetermined conditions include, but are not limited to, the following: Condition 1: The matching rate is less than or equal to a predetermined value (the matching rate can be the one calculated in the first embodiment). Condition 2: Within a specified distance from the worker's location Condition 3: Conditions 1 and 2

[0191] By performing the above process, the amount of data collected for job postings that require traffic information can be reduced, thereby reducing the amount of processing power required by the computer.

[0192] In addition to any of the above-described processes, the proposed module 243c may also perform the action of providing an incentive to each job recommended by the configuration module 242c. The incentive can be, for example, a perk for the user, and any perk that contributes to applications is acceptable. The source of the incentive may be the employer of the job or the operator of the job information provision platform.

[0193] Through the above process, for example, it becomes possible to increase the likelihood of applications for job postings that are located in inconvenient locations or have low matching rates by providing incentives to workers.

[0194] Proposal module 243c may include, as an incentive, an increase in the transportation allowance set for each job. For example, proposal module 243c may increase the transportation allowance for jobs located in inconvenient locations to cover the worker's transportation costs. Specifically, proposal module 243c may increase the transportation allowance so that it does not fall below the transportation allowance included in the transportation information.

[0195] According to the above process, while job postings for workplaces in inconvenient locations tend to have higher transportation costs than usual, increasing the transportation allowance for workers can prevent them from feeling financially disadvantaged.

[0196] In addition to any of the above processes, proposed module 243c may also include setting an increased transportation allowance that is below the upper limit of the transportation allowance set based on salary. For example, proposed module 243c may set an upper limit to prevent transportation expenses from becoming excessively high.

[0197] Based on the above process, while job postings for workplaces in inconvenient locations tend to have higher transportation costs than usual, even if transportation costs for workers are increased, setting a limit can prevent imposing an excessive burden on employers.

[0198] In addition to any of the above processes, proposal module 243c may also propose that the business that created the first job posting provide an incentive to the business that created the first job posting if the matching rate is below a predetermined value. For example, proposal module 243c may propose to the business of the first job posting that it provide a reward as an incentive to workers for the first job posting if the matching rate for the workday is 50% or less, or if the recommendation ranking is 5th or lower. Providing incentives to job postings with low matching rates is done, for example, to encourage as many workers as possible to be matched and work. Job postings with high matching rates are likely to attract the required number of applicants without any special action, so it is advisable to take measures for job postings with low matching rates.

[0199] Through the above process, we can propose ways to improve the matching rate for businesses whose workplaces are located in areas with inconvenient transportation. As a result, there is a possibility that the number of applications from workers will increase, which can contribute to alleviating the labor shortages of businesses.

[0200] In addition to any of the above-described processes, proposal module 243c may also include notifying the business operator that the number of applicants may increase as a result of the provision of incentives. For example, proposal module 243c may determine the possibility of an increase in the number of applicants using the increase in the matching rate described in the second embodiment. Specifically, if the increase in the matching rate is positive, it may be determined that there is a possibility of an increase, and if the increase in the matching rate is negative, it may be determined that there is no possibility of an increase.

[0201] Through the above process, it is possible to determine the change in the matching rate when incentives are provided. Furthermore, by presenting the increase in the matching rate to the businesses as a numerical value, it becomes possible to make the businesses realize the effectiveness of providing incentives.

[0202] In addition to any of the above processing, output module 241c may also include prioritizing the publication of job postings with incentives. For example, output module 241c may configure the screen so that job postings with incentives are displayed at the top of the job list.

[0203] Through the above process, it becomes possible to display job postings with incentives in a location that is more likely to be noticed by workers, thereby increasing the likelihood of applications.

[0204] ≪Specific Example 4≫ Next, we will explain the method for setting recommended jobs in the third embodiment using a specific example. For example, in the third embodiment, the recommended jobs for each worker are adaptively changed using transportation information to the work location. For example, the matching probability is improved by increasing the display priority of jobs with poor transportation access or by offering incentives.

[0205] Figure 18 shows the current status of each job posting according to the third embodiment. In the example shown in Figure 18, job postings D to F that satisfy predetermined conditions (e.g., condition 3) for obtaining traffic information have been identified. For example, the following conditions are used as the predetermined conditions. Condition 1: Job completion matching rate ≤ 80% Condition 2: Within 20km of the worker's location (First weight > 0.1 in the second embodiment) Condition 3: Conditions 1 and 2

[0206] In the example shown in Figure 18, "Current number of applicants" represents the number of applicants divided by the number of positions available. "Weight" is the priority (total weight) calculated in the second embodiment, but weights calculated by other methods may also be used. The setting module 242c acquires traffic information for job postings D to F shown in Figure 18. This reduces the processing load because the job postings are filtered by condition 3.

[0207] Figure 19 shows an example of transportation information for each job posting according to the third embodiment. In the example shown in Figure 19, the transportation route as transportation information includes transportation costs, number of transfers, walking distance, and travel time. If there are multiple transportation routes, it is preferable to obtain the highest-level transportation route suggested as optimal by the transportation information service. In the examples shown in Figures 18 and 19, it is assumed that the setting module 242c is set as a job posting that recommends job postings D to F.

[0208] Next, proposed module 243c calculates the incentive to be given to each job posting using the following procedure. (1) Incentive limit = MIN (salary x 15%, 1000 yen) Job posting D: MIN(450,1000) = 450 yen Job offer E:MIN(1100,1000)=1000 yen Job posting F: MIN(750,1000) = 750 yen

[0209] (2) Transportation expense incentive = MIN (incentive limit, transportation cost of the route × incentive coefficient - transportation cost of the job set by the employer) Incentive coefficient for transportation expenses: Dynamically adjustable. Let's assume it's 1.3. Job offer D = MIN(450,150 × 1.3 - 200) = -5 yen *Sufficient transportation expenses are already included in the job posting. Job offer E = MIN(1000,500 × 1.3 - 200) = 450 yen Job offer F = MIN(750,800 × 1.3 - 300) = 740 yen For example, as an incentive for transportation expenses, it would be good to set it at 0 yen for job D (sufficient transportation expenses already set), 450 yen for job E, and 740 yen for job F. Note that the employer may adjust the transportation expense incentive for each job from the recommended amounts mentioned above as appropriate.

[0210] (3) Weighting of the sorting order in the job listings The weight setting determines the order in which recommendations are displayed in the job listings. Original weight (total weight) + |Transportation expense incentive| / 200 ·Recruitment D:3.6+5 / 200=3.625 Job posting E: 2.8 + 450 / 200 = 5.05 Job offer F: 2.1 + 800 / 200 = 6.1

[0211] Output module 241c configures the screen so that job postings F, E, and D are displayed in the order of the weights determined above, and controls the display of these postings on the worker's job list screen. Job postings eligible for incentives are, for example, those with a matching rate of 75% or less. This is because job postings with a high matching rate are likely to attract applicants without any additional measures.

[0212] For example, proposed module 243c would recommend a travel expense incentive next to a job posting when it is created, or when the estimated matching rate after X days has passed since posting is Y. For example, "If transportation costs are increased by 400 yen, the estimated matching rate is likely to increase from 40% to 90%!"

[0213] If the employer approves the proposal to increase transportation expenses, output module 241c may add weight to increase the visibility of this job posting. The following are examples of measures to increase visibility. • Measure 1: For all job postings that offer incentives (e.g., increased transportation allowance), add the weight of this job posting and add "High transportation allowance!" to the thumbnail image. Example: Weight Plus = (Transportation Incentive / 400) Upper Limit + 1.0 • Strategy 2: Place this job posting in the recommended workers section within a radius of Xkm. • Measure 3: Increase the frequency of CRM (Customer Relationship Management) communication (push notifications, emails, etc.) to workers within Xkm of this job posting.

[0214] The job recommendation method in Specific Example 4 is merely an example and is not limited to the above example. For example, the formulas for calculating transportation expense incentives and the formulas for calculating the weights of the sorting order are just examples. It is sufficient to be able to propose appropriate transportation expenses for jobs with poor transportation access, and job postings from businesses that approve the proposal should be displayed preferentially in the job list.

[0215] <Example of data structure> The various data in the third embodiment are the same as the various data corresponding to the first and second embodiments. In the third embodiment, the data output from the learning model and the final output data are different, but it is sufficient that they are stored in memory 230c.

[0216] <Operation Description> Next, the operation of the information processing system 1 according to the third embodiment will be described. Figure 20 is a flowchart showing an example of the processing related to job recommendation according to the third embodiment. In the example shown in Figure 20, the server 20c is a device that implements the job information provision platform and provides the job service according to the third embodiment, and the flowchart shows the processing performed by this server 20c.

[0217] In step S302, the acquisition module 238c acquires location information indicating the worker's position. For example, the location information may be GPS location information from the information processing device 10B used by the worker.

[0218] In step S304, the setting module 242c recommends to the worker each job posting within the area identified based on the location information acquired by the acquisition module 238c, based on the transportation information from this location to the workplace of each job posting. For example, the setting module 242c sets up each job posting to recommend to a specific worker based on the transportation route from the specific worker's location to the workplace, taking into account factors such as ease of commuting, travel time, and effort of travel, for each job posting with a workplace within a predetermined area, based on the location information of the specific worker. Note that as an example of step S304, the processing of steps S342 and S344 may also be performed.

[0219] In step S342, the learning module 239c may input the number of people to be hired and the number of applicants for each job currently being advertised in the area into a learning model that estimates the matching rate of the area, which is trained using learning data that includes at least the matching rate of the number of people who completed work relative to the number of people to be hired for each past job in the area. For example, the learning model is a model that estimates the number of people who completed work / number of people to be hired on future work days for each job, based on the current application status (number of applicants / number of people to be hired) for each job. The learning model in the third embodiment may be the learning model in the second embodiment.

[0220] In step S344, the configuration module 242c may also include setting each job to recommend to the worker based on the matching rate of each currently available job output from the learning model. For example, the configuration module 242c may set jobs with a matching rate lower than a predetermined value as jobs to recommend to the worker in order to increase the matching rate as much as possible.

[0221] In step S306, the output module 241c publishes the job information for each job set by the configuration module 242c to the specified workers. For example, the output module 241c may configure the job screen so that jobs with higher recommendation priority are displayed at the top of the job list.

[0222] In step S308, the suggestion module 243c may perform the action of providing an incentive to each job recommended by the configuration module 242c. The incentive may be, for example, a perk for the user, and any perk that contributes to the application process may be used.

[0223] In step S310, if the proposal module 243c obtains approval for the proposal from the business operator, it provides an incentive to the job posting. If an incentive is provided, the output module 241c implements measures 1 to 3 described above, for example. Alternatively, the proposal module 243c may provide an incentive to the job posting even without approval from the business operator.

[0224] Figure 21 shows an example of a push notification according to the third embodiment. In the example shown in Figure 21, the output module 241c sends a push notification to each worker to proactively inform them of job postings with increased transportation allowances (measure 3). The push notification displays "Job postings with increased transportation allowances! Job postings with increased transportation allowances for a limited time have been posted!", and the information processing device 10 used by the worker waits for an operation from the worker. When a worker activates the push notification, the job application app is launched, and the screen shown in Figure 22 may be displayed directly or indirectly.

[0225] Figure 22 shows an example of a job listing screen according to the third embodiment. In the screen shown in Figure 22, a job listing with "Increased Transportation Allowance!" written at the top of the screen is displayed (Measure 2). This makes it more likely that workers will be interested in jobs with increased transportation allowance and will refer to these jobs. Assume that job listing W20 is operated by a worker in the screen shown in Figure 22. At this time, the detailed screen of job listing W20 shown in Figure 23 is displayed on the screen. Note that if the job listing shown in Figure 22 is recommended even without "Increased Transportation Allowance," it may also be the screen displayed in step S306 without "Increased Transportation Allowance" (for example, if it is changed to "Recommended!").

[0226] Figure 23 shows an example of a job details screen according to the third embodiment. In the example shown in Figure 23, it is easy to see that the transportation allowance for this job has increased by 740 yen. Furthermore, because it is limited to workers within a specific distance, if the worker's location is within a predetermined distance from the work location of this job, they can receive the push notification shown in Figure 21 and check the job details screen shown in Figure 23.

[0227] Figure 24 shows an example of a business operator's settings screen according to the third embodiment. In the screen shown in Figure 24, the amount and matching rate are displayed in association with the transportation expense setting item on the job information settings screen. For example, the suggestion module 243c calculates the amount of the increase in transportation expenses to suggest to the business operator, referring to the transportation expenses included in the transportation information. In the example shown in Figure 24, the suggested information for transportation expenses is displayed as "+400 yen for an 85% matching rate". Also, the screen shown in Figure 24 displays "If the recommendation content is reflected, this job will be displayed preferentially on the job screen", so the business operator knows that by approving this suggestion, the job will be displayed preferentially. This allows the business operator to set up job information based on suggestions from the job provision platform.

[0228] Through the above processing, according to the third embodiment, various processes that improve user convenience become possible as a job information provision platform, such as recommendations based on traffic information from the worker's location and dynamic pricing.

[0229] Furthermore, assuming that each of the processes described in the first to third embodiments can be executed by each of the configurations in the third embodiment, then each of the processes shown in Figure 25 can be executed. Figure 25 is a flowchart showing an example of each process in the third embodiment.

[0230] In steps S402 and S404, the worker device (information processing device 10B, etc.) uses a location estimation service (e.g., GPS) to acquire its own location information and outputs it to the server 20c. Note that the use of a location estimation service is not mandatory; location information such as a location on a map specified by the worker may also be output.

[0231] In step S406, the setting module 242c of the server 20c identifies job offers that the worker is eligible to apply for based on the worker's location. For example, job offers within a predetermined distance from the worker's location may be considered eligible for application.

[0232] In step S408, the output module 241c of the server 20c publishes the list of job offers to the worker. At this time, in response to a viewing request from the worker, screen information of the list of job offers including job offers that the worker is eligible to apply for may be output to the worker's device.

[0233] In step S410, the setting module 242c of the server 20c may sort the job offers according to the worker's location. The sorting method can use the method of Specific Example 3 or 4.

[0234] In step S412, the setting module 242c of the server 20c accesses the traffic information providing service via an API in order to set the worker's location and the work location of the job offer and obtain traffic information.

[0235] In step S414, the setting module 242c of the server 20c obtains, as traffic information, options for traffic routes from the traffic information providing service, and obtains, as information for each route, travel time, number of transfers, transportation costs, etc.

[0236] In step S416, the learning module 239c of the server 20c learns the job offer information of the job offers that the worker has applied for in the past, thereby grasping the worker's application tendency.

[0237] In step S418, the learning module 239c of the server 20c estimates how much the user tolerates each means of transportation such as bus, train, walking, etc. for the worker. The learning module 239c can estimate the user's tolerance status by grasping the worker's application tendency.

[0238] In step S420, the setting module 242c of the server 20c sets the priority to be recommended to the worker based on each job offer sorted in step S410, the traffic information for each job offer acquired in step S414, and the acceptance degree of the worker's means of transportation acquired in step S418, and generates a job offer list based on the priority.

[0239] In steps S422 and S424, the estimation module 240c of the server 20c estimates the matching rate for each job offer. At this time, the matching rate may be estimated for each area. Also, the matching rate is obtained using past data, the current number of applicants, the number of job offers, and the like. For example, the matching rate can be calculated using any of the estimation methods of the matching rate disclosed in the present disclosure.

[0240] In step S426, the proposal module 243c of the server 20c determines the incentive to be proposed for each job offer. The incentive is a privilege given to the worker and includes at least one of salary, transportation expenses, points, service vouchers, and the like.

[0241] In step S428, the proposal module 243c of the server 20c notifies or displays on the screen the incentive to be given to the worker by applying for the job offer to which the incentive is given. Through the above processing, the integrated processing of the first to third embodiments can be executed.

[0242] Note that the disclosed technology is not limited to the above-described embodiments, and can be implemented in various other forms without departing from the gist of the disclosed technology. Therefore, the above embodiments are merely illustrative in every respect and are not to be construed in a limiting sense. For example, the above-described processing steps can be arbitrarily changed in order or executed in parallel as long as there is no contradiction in the processing content.

[0243] The programs of embodiments of this disclosure may be provided stored on a computer-readable storage medium. The storage medium is a “tangible medium that is not temporary” on which the program can be stored. The program includes, but is not limited to, software programs and computer programs.

[0244] The following are additional notes regarding the disclosed technology described above. [Note 1] Information processing device, To obtain location information of a predetermined location, Inputting predetermined statistical data into a learning model that estimates the matching rate within a predetermined range, which is trained using learning data that includes at least the matching rate and statistical data of each past job posting within a predetermined range identified based on the location information. Using the data output from the aforementioned learning model, estimate the matching rate of job postings within the predetermined range for a given day. An information processing method that performs [this action]. [Note 2] To obtain the above means, The information processing method described in Appendix 1, which includes obtaining location information indicating the location of the worker from a processing device used by the worker. [Note 3] The information processing method according to Appendix 1 or 2, which involves training a learning model with the learning data, which includes statistical data, including at least one piece of data on the date, day of the week, public holiday, and weather of past working days. [Note 4] The information processing method according to any one of the appendices 1 to 3, which involves training the learning model with the learning data, which includes at least the number of cancellations on working days. [Note 5] The above estimation means that Based on current statistical data, the number of positions available, and the number of applicants, obtain estimated values ​​for the increase in the number of positions available and the increase in the number of applicants up to the future work date, as estimated by the aforementioned learning model. An information processing method according to any one of Appendix 1 to 4, which includes estimating the matching rate for future work days using the current number of recruits, the number of applicants, the estimated increase in the number of recruits, and the estimated increase in the number of applicants. [Note 6] The above estimation means that The information processing method described in Appendix 5, which includes estimating the matching rate for the aforementioned future workdays based on the number of cancellations for those future workdays. [Note 7] The above estimation means that To estimate the increase in the number of available workers from the present to the aforementioned future work date, The information processing method described in Appendix 5, which includes estimating the matching rate of future workdays based on the increase in the number of workers. [Note 8] The learning model is trained using the aforementioned learning data, which includes category information of past job postings. The above estimation means that An information processing method described in any one of Appendix 1 to 7, which includes estimating the matching rate for each job category. [Note 9] The aforementioned training is, The information processing method described in Appendix 8, which includes training the learning model with the category information, including the type and / or content of the job postings. [Note 10] In an information processing device, To obtain location information of a predetermined location, Based on the location information identified, a learning model for estimating the matching rate within a predetermined range is trained using learning data that includes at least the matching rate of the number of completed jobs relative to the number of available positions and statistical data on the number of days worked. Using the data output from the aforementioned learning model, estimate the matching rate of job postings within the predetermined range for a given day. A program that executes the command. [Note 11] An information processing apparatus including one or more processors, wherein the one or more processors acquire position information of a predetermined position, input predetermined statistical data into a learning model that estimates a matching rate of a predetermined range, which is learned using learning data including at least a matching rate of the number of completed workdays to the number of recruited people and statistical data of workdays for each past job offer within the predetermined range specified based on the position information, estimate a matching rate of a predetermined day of a job offer in the predetermined range using data output from the learning model, and execute the above operations. [Appendix 21] The information processing apparatus acquires position information of workers, inputs job offer information, the number of recruited people, and the number of applicants for each currently recruiting job offer into a learning model that estimates a matching rate of each job offer, which is learned using learning data including at least a matching rate of each past job offer and job offer information, estimates a matching rate of a predetermined day of each currently recruiting job offer using data output from the learning model, recommends each job offer set using the estimated matching rate of each job offer on the predetermined day and the distance between the position of the worker based on the position information and the work location of each currently recruiting job offer to the worker, and executes the above operations. [Appendix 22] The learning model learns the learning data including job offer information of jobs to which the worker has applied in the past, and outputs a matching rate of each currently recruiting job offer that is similar to the job offer information of jobs to which the worker has applied in the past from the learning model. The information processing method according to claim 1. [Appendix 23] The job offer information learned by the learning model includes at least one of job type, job content, working hours, work location, and salary. The information processing method according to Appendix 21 or 22. [Appendix 24] The above input means, The information processing method described in any one of Appendix 21 to 23, which includes inputting job information, the number of people to be hired, and the number of applicants for each job currently being advertised within the predetermined range into a learning model that has been trained using the learning data of each past job within a predetermined range identified based on the location information. [Note 25] The information processing method described in any one of Appendix 21 to 24, which involves proposing to the business operator that created the first job posting that an incentive be given to workers for the first job posting whose matching rate on the aforementioned specified day is below a specified value. [Note 26] The above proposal is to The information processing method described in Appendix 25, which includes proposing to increase the salary information of the first job posting. [Note 27] The above proposal is to The information processing method described in Appendix 26, which includes proposing to increase the salary information of the first job posting based on the salary information of job postings similar to the first job posting. [Note 28] The above input means, This includes inputting the job information of the first job, including the increased salary information, into the learning model. The above estimation means that The information processing method according to Appendix 26 or 27, which includes estimating the amount of increase in the matching rate for the first job on the specified date using the matching rate of the first job based on the increased salary information. [Note 29] In an information processing device, Obtaining the worker's location information, The learning model, which estimates the matching rate for each job posting, is trained using training data that includes at least the matching rates and job posting information for each past job posting. The job posting information, number of positions available, and number of applicants for each currently open job posting are then input into this model. Using the data output from the aforementioned learning model, estimate the matching rate for each of the currently recruiting job openings on a specified date. The job postings are recommended to the worker based on the estimated matching rate of each job posting on the specified date, and the distance between the worker's location based on the location information and the work location of each currently recruiting job posting. A program that executes something. [Note 30] An information processing device including one or more processors, The one or more processors described above are: Obtaining the worker's location information, The learning model, which estimates the matching rate for each job posting, is trained using training data that includes at least the matching rates and job posting information for each past job posting. The job posting information, number of positions available, and number of applicants for each currently open job posting are then input into this model. Using the data output from the aforementioned learning model, estimate the matching rate for each of the currently recruiting job openings on a specified date. The job postings are recommended to the worker based on the estimated matching rate of each job posting on the specified date, and the distance between the worker's location based on the location information and the work location of each currently recruiting job posting. An information processing device that performs this task. [Note 41] Information processing device, Obtaining the worker's location information, Based on the location information, the system recommends to the worker each job posting that is set based on the transportation information to the work location of each job posting within a predetermined range identified based on the location information. An information processing method that performs the following. [Note 42] The following steps are performed: input the number of positions available and the number of applicants for each job currently being advertised within the predetermined range into a learning model that estimates the matching rate within the predetermined range, which has been trained using learning data that includes at least the matching rate of each past job posting within the predetermined range. The above setting means, The information processing method according to Appendix 41, further comprising setting each job to recommend to the worker based on the matching rate of each currently recruiting job output from the learning model. [Note 43] The above setting means, An information processing method as described in Appendix 41 or 42, which includes obtaining the aforementioned transportation information for each job opening identified based on predetermined conditions. [Note 44] An information processing method described in any of the locations in Appendix 41 to 43, which performs the action of providing an incentive to each of the recommended job postings. [Note 45] The aforementioned granting means The information processing method described in Appendix 44, which includes increasing the transportation allowance set for each of the aforementioned job postings. [Note 46] The aforementioned granting means The information processing method described in Appendix 45, which includes setting an increased transportation allowance that is less than or equal to the upper limit of the transportation allowance set based on salary. [Note 47] The information processing method described in Appendix 42, which involves proposing to provide an incentive to the business operator that created the first job posting for which the matching rate is below a predetermined value. [Note 48] The above proposal is to The information processing method described in Appendix 47, which includes notifying the business operator that the number of applicants may increase as a result of the provision of the aforementioned incentive. [Note 49] The information processing method described in Appendix 47 or 48, which performs the action of prioritizing the publication of job postings to which the aforementioned incentives have been granted. [Note 50] In an information processing device, Obtaining the worker's location information, Based on the location information, the system recommends to the worker each job posting that is set based on the transportation information to the work location of each job posting within a predetermined range identified based on the location information. A program that executes the command. [Note 51] An information processing device including one or more processors, The one or more processors described above Obtaining the worker's location information, Based on the location information, the system recommends to the worker each job posting that is set based on the transportation information to the work location of each job posting within a predetermined range identified based on the location information. An information processing device that performs the following actions. [Explanation of Symbols]

[0245] 1. Information Processing System 10, 10A, 10B, 10C Information Processing Device 20. Information Processing Equipment (Server) 110, 210 Processing Units (CPUs) 120, 220 Network Communication Interfaces 130, 230 memory 131, 231 Operating Systems 132, 232 Network Communication Modules 133 App Data 134 Service Processing Module 135 Acquisition Module 136 Output Modules 137 Processing Modules 138 Display Control Module 150 User Interfaces 170, 270 Communications Bus 233 User Information 234 Job postings 235 Usage History Information 236 Learning Model Information 237 Service Control Module 238 Acquisition Modules 239 Learning Modules 240 Estimated Modules 241 Output Module 242 Configuration Module 243 Proposed Modules

Claims

1. Information processing device, Obtaining the worker's location information, The learning model, which estimates the matching rate for each job posting, is trained using training data that includes at least the matching rates and job posting information for each past job posting. The job posting information, number of positions available, and number of applicants for each currently open job posting are then input into this model. Using the data output from the aforementioned learning model, estimate the matching rate for each of the currently recruiting job openings on a specified date. The job postings are recommended to the worker based on the estimated matching rate of each job posting on the specified date, and the distance between the worker's location based on the location information and the work location of each currently recruiting job posting. An information processing method that performs [this action].

2. The information processing method according to claim 1, wherein the learning model learns the learning data, which includes job information of jobs that the worker has applied for in the past, and the learning model outputs a matching rate for each currently recruiting job that is similar to the job information of jobs that the worker has applied for in the past.

3. The information processing method according to claim 1 or 2, wherein the job information learned by the learning model includes at least one of the type of work, the content of the work, the working hours, the work location, and the salary.

4. The above input means, The information processing method according to claim 1, further comprising inputting job information, number of applicants, and number of applicants for each currently recruiting job within a predetermined range into a learning model trained using the learning data of each past job within a predetermined range identified based on the location information.

5. The information processing method according to claim 1, wherein, for the first job posting whose matching rate on the predetermined date is below a predetermined value, the method proposes to the business operator that created the first job posting that an incentive be given to the worker.

6. The above proposal is to The information processing method according to claim 5, comprising proposing to increase the salary information of the first job posting.

7. The above proposal is to The information processing method according to claim 6, comprising proposing to increase the salary information of the first job posting based on the salary information of job postings similar to the first job posting.

8. The above input means, This includes inputting the job information of the first job, including the increased salary information, into the learning model. The above estimation means that The information processing method according to claim 6 or 7, further comprising estimating the amount of increase in the matching rate for the first job on a predetermined date using the matching rate of the first job based on the increased salary information.

9. In an information processing device, Obtaining the worker's location information, The learning model, which estimates the matching rate for each job posting, is trained using training data that includes at least the matching rates and job posting information for each past job posting. The job posting information, number of positions available, and number of applicants for each currently open job posting are then input into this model. Using the data output from the aforementioned learning model, estimate the matching rate for each of the currently recruiting job openings on a specified date. The job postings are recommended to the worker based on the estimated matching rate of each job posting on the specified date, and the distance between the worker's location based on the location information and the work location of each currently recruiting job posting. A program that executes something.

10. An information processing device including one or more processors, The one or more processors described above are: Obtaining the worker's location information, The learning model, which estimates the matching rate for each job posting, is trained using training data that includes at least the matching rates and job posting information for each past job posting. The job posting information, number of positions available, and number of applicants for each currently open job posting are then input into this model. Using the data output from the aforementioned learning model, estimate the matching rate for each of the currently recruiting job openings on a specified date. The job postings are recommended to the worker based on the estimated matching rate of each job posting on the specified date, and the distance between the worker's location based on the location information and the work location of each currently recruiting job posting. An information processing device that performs this task.

Images