Processing device, processing program, and processing method

Abstract

[Problem] To improve the degree of matching between an employer and a job seeker and enable a recruitment process that leads to an improvement in worker retention rate. [Solution] At least one processor executes a process for conducting an AI interview using a trained interview model with a job seeker, and extracting a candidate workplace for the job seeker in accordance with the result of the AI interview.

Description

Processing device, processing program, and processing method

[0001] The present disclosure relates to a processing device, a processing program, and a processing method for providing support to employers and employees regarding their recruitment and subsequent employment.

[0002] It has been known for some time that artificial intelligence (AI) is used to match employers and employees and to provide support during and after hiring. For example, Patent Document 1 describes a system for improving the retention rate of specific human resources, comprising: (a) a pre-employment information input means for inputting, when hiring specific human resources, one or more pieces of information from health information obtained during a job interview, information from a question-and-answer session during the job interview, and information on the status of practical training at a work experience session; (b) a hiring decision information providing means for providing hiring decision information for hiring human resources expected to have a high retention rate based on the information input to the pre-employment information input means and using one or more of cooperativeness, business etiquette, and job performance ability as criteria; and (c) a support information providing means for providing information on employment support and technical support to improve the retention rate of members.

[0003] Japanese Patent Application Laid-Open No. 2021-174343

[0004] However, in the system for improving the retention rate of specific human resources described in Patent Document 1, although the results of an interview between the employer and the worker are taken into consideration when deciding whether to hire the candidate, matching is primarily performed based on written or digitalized information prior to the interview, which can make it difficult to fully grasp the job seeker's actual personality and character. Furthermore, with this type of matching, the job seeker's impression of the employer or company can become abstract, and the job seeker may have a different impression from the one they had at the interview. As a result, the job seeker may find it difficult to hire or work in subsequent interviews, and the subsequent interviews themselves may be wasted.

[0005] Furthermore, once actual employment begins after such matching and interviews have been carried out, the employer's impression or evaluation of the employee may change, or the employee's impression or evaluation of the employer may change, resulting in a lower employee retention rate.

[0006] The present disclosure has been made against the above background, and provides a processing device, processing program, and processing method that enable a recruitment process that improves the degree of matching between employers and job seekers and leads to an improvement in the retention rate of employees.

[0007] According to one aspect of the present disclosure, there is provided "a processing device having at least one processor, wherein the at least one processor is configured to execute a process for conducting an AI interview with a job seeker using a trained interview model, and extracting potential employment opportunities for the job seeker based on the results of the AI ​​interview."

[0008] According to one aspect of the present disclosure, there is provided a "processing program for causing a computer having at least one processor to function to conduct an AI interview with a job seeker using a trained interview model, and extract potential employment opportunities for the job seeker based on the results of the AI ​​interview."

[0009] According to one aspect of the present disclosure, there is provided a "processing method executed by at least one processor in a computer having the at least one processor, the processing method including the steps of: conducting an AI interview with a job seeker using a trained interview model; and extracting potential employment opportunities for the job seeker based on the results of the AI ​​interview."

[0010] According to the present disclosure, it is possible to provide a processing device, processing program, and processing method that improve the degree of matching between employers and job seekers and enable a recruitment process that leads to an improvement in the retention rate of employees.

[0011] It should be noted that the above effects are merely illustrative for the sake of convenience and are not limiting. In addition to or instead of the above effects, any effect described in this disclosure or an effect obvious to a person skilled in the art may be achieved.

[0012] FIG. 1 is a diagram schematically illustrating an overview of an employment matching process related to a processing system according to an embodiment of the present disclosure. FIG. 2 is a conceptual diagram schematically illustrating a configuration of a processing system 1 according to an embodiment of the present disclosure. FIG. 3 is a block diagram illustrating an example of a configuration of a terminal device 100 according to an embodiment of the present disclosure. FIG. 4 is a block diagram illustrating an example of a configuration of a server device 200 according to an embodiment of the present disclosure. FIG. 5A is a diagram conceptually illustrating a job seeker management table stored in the server device 200 according to an embodiment of the present disclosure. FIG. 5B is a diagram conceptually illustrating an employer management table stored in the server device 200 according to an embodiment of the present disclosure. FIG. 6A is a diagram illustrating a processing sequence executed between each terminal device 100 and the server device 200 according to an embodiment of the present disclosure. FIG. 6B is a diagram illustrating a processing sequence executed between each terminal device 100 and the server device 200 according to an embodiment of the present disclosure. FIG. 6C is a diagram illustrating a processing sequence executed between each terminal device 100 and the server device 200 according to an embodiment of the present disclosure. FIG. 7A is a diagram illustrating a processing flow executed in the server device 200 according to an embodiment of the present disclosure. FIG. 7B is a diagram showing a processing flow executed in the server device 200 according to an embodiment of the present disclosure.

[0013] Embodiments of the present disclosure will be described with reference to the accompanying drawings, in which the same reference numerals are used to designate common components.

[0014] 1. Overview of Processing System 1 According to the Present Disclosure The processing system according to the present disclosure, for example, conducts an AI interview using artificial intelligence (AI) with a job seeker, taking into account the characteristics of the job seeker, and matches the job seeker with a recruiting employer based on the results. In particular, the processing system is used to conduct an AI interview using a trained interview model developed through machine learning using information (culture, customs) related to the job seeker's hometown (country, region, prefecture, etc.), criminal information, fugitive information, tests previously taken by the job seeker (psychological tests, behavioral tests, aptitude tests, etc.), and information from social networking services (SNS). By conducting such an AI interview, it is possible to match job seekers and employers who are likely to be compatible with each other before the job seeker meets the employer, thereby improving the degree of match between the job seeker and the employer and realizing a hiring process that leads to an improved employee retention rate.

[0015] 1 is a diagram illustrating an outline of a job matching process in a processing system according to an embodiment of the present disclosure. Specifically, FIG. 1 chronologically illustrates an outline of information collection before making a hiring decision on a job seeker, subsequent hiring decisions such as job matching and interviews, and information collection after starting work.

[0016] First, a server device of a business that matches a first user (job seeker) with a second user (employer) acquires job seeker information from the job seeker, company information from the employer, and web information from the web before making a hiring decision for the job seeker. The server device also administers a test to the job seeker and receives information related to the test results.

[0017] The job seeker information obtained from the job seeker includes basic personal information of the job seeker, such as the job seeker's country of origin, region of origin, age, gender, etc. The job seeker information also includes personal characteristic information such as the job seeker's skills, experience, or information listed on their resume. Here, the job seeker may be either Japanese or foreign.

[0018] In contrast, company information obtained from employers includes basic employer information related to the employer's name, location, business area, business scale, turnover rate, age ratio, ratio of foreign nationals, ratio of men to women, ratio of disabled people, and job information (job type, work location, job content, etc.). Here, we assume that the employer is a general company (enterprise), but the employer may also be a sole proprietor, union, organization, or other legal entity.

[0019] Furthermore, web information obtained from the web includes information on national or local culture, criminal information, fugitive information, and posts or comments by job seekers on SNS. In particular, information related to the SNS may be obtained by crawling.

[0020] The server device uses the acquired information to perform machine learning for a trained interview model for AI interviews and a trained matching model for matching job seekers and employers, as well as additional machine learning. This makes it possible to conduct a preliminary AI interview (first interview) with the job seeker using the trained interview model, with optimal questions selected based on various information. Furthermore, based on the results of the AI ​​interview and other information, the trained matching model can match job seekers and employers with high accuracy, achieving an optimal match for both parties. Note that the trained interview model for AI interviews and the trained matching model for matching job seekers and employers are not limited to being separate models, and may be a single model capable of AI interviews and matching.

[0021] Then, when the hiring decision stage begins, the job seeker operates their own terminal device to access the server device and undergoes a first interview (AI interview) using a trained interview model for AI interviews. During the AI ​​interview, the server device converts responses made during the interview into text and stores the text information in memory. The server device also performs processing related to simultaneous language interpretation for foreign job seekers. This processing may be performed by the trained interview model for AI interviews or by another translation program, etc.

[0022] Furthermore, during the AI ​​interview, the server device may receive status information related to the job seeker's question and answer session and determine whether the job seeker's words and actions contain any falsehoods. For example, the server device may receive photographic information of the job seeker as status information and analyze the job seeker's eye movements and body movements from the photographic information. Specifically, if the job seeker's eyes wander or their limbs move restlessly, it may be determined that the job seeker is lying. Note that this analysis may be performed by a trained interview model for AI interviews, or by a program dedicated to this analysis or another trained model.

[0023] Next, the server device uses a trained matching model for matching job seekers and employers to extract employers that would be suitable for the job seeker who has undergone the AI ​​interview. This extraction may use not only the AI ​​interview result information, but also job seeker information obtained before the AI ​​interview and information related to test results. The server device then sends job seeker information to the extracted employers and employer information to the job seeker, inquiring whether or not to conduct an interview (second interview). When the server device receives notification from both the job seeker and employer that they wish to conduct an interview, it coordinates the interview. Furthermore, the server device acquires the results of the interview (information related to whether or not the job seeker has been hired) and stores them in memory, and performs additional machine learning on the trained interview model for AI interviews and the trained matching model for matching job seekers and employers.

[0024] After that, after a certain period of time has passed since the job seeker's employment was confirmed and the job seeker began working for the employer, the server device acquires working environment information from the worker and worker evaluation information from the employer. The server device then stores the acquired information in memory and performs additional machine learning on a trained interview model for AI interviews and a trained matching model for matching job seekers with employers. Here, the working environment information may include, for example, the worker's evaluation of the employer, evaluation of the work content, evaluation of the working conditions, and results of engagement scoring. Meanwhile, the worker evaluation information may include promotion / demotion, work attitude, attendance hours, resignation, personnel evaluation, etc.

[0025] In this way, the server device performs additional machine learning on the trained interview model and the trained matching model using various judgment results related to hiring decisions, making it possible to update the models in accordance with information related to the hiring process and information related to post-hiring evaluations, etc. Therefore, by accumulating hiring experience, it is possible to further improve the accuracy of AI interviews and matching between job seekers and employers.

[0026] In this disclosure, even if a term such as "first" or "second" is used, it does not mean that the term is limited to only the two elements to which it is attached. Naturally, a "third," a "fourth," or more elements may also be included.

[0027] 2. Configuration of Processing System 1 Fig. 2 is a conceptual diagram illustrating a configuration of a processing system 1 according to an embodiment of the present disclosure. According to Fig. 2, a first user terminal device 100-1 operated by a first user who is a job seeker looking for employment, a second user terminal device 100-2 operated by a second user who is an employer looking for workers, and a server device 200 that conducts AI interviews and matching are communicatively connected to each other via a network. The network may be configured wirelessly, wired, or a combination thereof.

[0028] 2, only one first user terminal device 100-1 and one second user terminal device 100-2 are shown, but it is possible to include two or more of each device. Also, although the server device 200 is shown as a single device, it is also possible to distribute the components and processes of the server device 200 to multiple server devices or cloud server devices.

[0029] 2, the first user terminal device 100-1 and the second user terminal device 100-2 are described separately, but they may be collectively referred to as the terminal device 100. However, even in such a case, the terminal devices are simply referred to collectively, and do not mean that the first user terminal device 100-1 and the second user terminal device 100-2 have the same processing or configuration.

[0030] 3. Configuration of the terminal device 100 Fig. 3 is a block diagram showing an example of the configuration of the terminal device 100 according to an embodiment of the present disclosure. The terminal device 100 does not need to include all of the components shown in Fig. 3, and may have a configuration in which some components are omitted, or may include other components.

[0031] The terminal device 100 is typically a terminal device capable of wireless communication, such as a smartphone, but is not limited to such devices. For example, the terminal device may be a feature phone, a personal digital assistant (PDA), a laptop computer, a desktop computer, a portable game console, a stationary game console, or any other device capable of executing the program of the present disclosure. As described above, the terminal device 100 includes a first user terminal device 100-1 and a second user terminal device 100-2. However, these terminal devices do not necessarily have to be the same type or identical, and may be different types of terminal devices. As described above, the first user terminal device 100-1 and the second user terminal device 100-2 in the processing system 1 may each include multiple terminal devices, but the terminal devices do not necessarily have to be the same type and may be different types of terminal devices.

[0032] 3, the terminal device 100 includes an output interface 111, a processor 112, a memory 113 including RAM, ROM, or non-volatile memory (or HDD in some cases), a communication interface 114 including a communication processing circuit and an antenna, a camera 115, and an input interface 116 including a touch sensor and hard keys. These components are electrically connected to each other via control lines and data lines.

[0033] The output interface 111 functions as an output unit that outputs images captured by the camera 115 and various displays output by executing the program according to the present disclosure to devices such as a display or a printer in response to instructions from the processor 112. Note that such a display may be, for example, a liquid crystal display, an organic EL display, or electronic paper.

[0034] The processor 112 is configured with a CPU (microcomputer) and functions as a control unit that controls other connected components based on various programs stored in the memory 113. Specifically, the processor 112 reads and executes programs for executing applications and programs for executing an OS according to the present disclosure from the memory 113. In the present disclosure, the processor 112 particularly executes the processes described in the processing sequences of FIGS. 6A and 6C (details of the processes are described in FIGS. 6A and 6C). The processor 112 may be configured with a single CPU, or may be configured with a combination of multiple CPUs and GPUs.

[0035] The memory 113 is composed of a ROM, a RAM, a non-volatile memory, a HDD, etc., and functions as a storage unit. The ROM stores instructions and commands for executing the application and OS according to the present disclosure as a program. The RAM is used to write and read data while the program stored in the ROM is being processed by the processor 112. The non-volatile memory is memory into which data is written and read as the program is executed, and the data written therein is retained even after the program execution has ended. In the present disclosure, the memory 113 particularly stores programs for executing the processes described in the processing sequences of Figures 6A and 6C (details of the processes will be described in Figures 6A and 6C).

[0036] The communication interface 114 functions as a communication unit that transmits and receives information to and from the server device 200 and other terminal devices installed remotely via a communication processing circuit and an antenna. The communication processing circuit processes programs and various information used in the processing system 1 to transmit and receive information to and from the server device 200 and other terminal devices according to the progress of processing.

[0037] The communication processing circuit processes based on a wideband wireless communication system such as the LTE system, but can also process based on a system related to narrowband wireless communication or a system related to contactless wireless communication such as a wireless LAN such as IEEE802.11 or Bluetooth (registered trademark).In addition, wired communication can be used instead of or in addition to wireless communication.

[0038] The camera 115 is, for example, a general web camera or video camera, and constantly captures images of the job seeker or employer. In particular, the camera 115 is used during the first interview, which is an AI interview, or when the second interview is conducted online.

[0039] The input interface 116 is composed of a touch sensor, hard keys, etc., and functions as an input unit that accepts instruction inputs related to the execution of the program according to the present disclosure, operation inputs for registering various information, etc. The touch sensor is arranged to cover the output interface 111, and transmits position coordinate information corresponding to image data output from the output interface 111 to the display to the processor 112. Known touch sensor methods can be used, such as a resistive film method, a capacitive coupling method, and an ultrasonic surface acoustic wave method. In the present disclosure, the touch sensor detects swipe operations and tap operations on each icon, etc. displayed on the output interface 111 by an indicator. Note that, although the present disclosure uses the input interface 116 provided in the terminal device 100, it is also possible to use an input interface 116, such as a mouse, that is connected wirelessly or via a wire to a main body including the processor 112, etc.

[0040] 4. Configuration of Server Device 200 FIG. 4 is a block diagram showing an example of the configuration of the server device 200 according to an embodiment of the present disclosure. The server device 200 does not need to include all of the components shown in FIG. 4 , and it is possible to omit some of the components, or to add other components. Furthermore, the server device 200 does not need to include the components shown in FIG. 4 in a single housing, and it is also possible to distribute the components and processes of the server device 200 across multiple server devices or cloud server devices.

[0041] 4, the server device 200 includes a memory 211 including RAM, ROM, non-volatile memory, and an HDD, a processor 212 including a CPU, and a communication interface 213. These components are electrically connected to each other via control lines and data lines.

[0042] The memory 211 is composed of ROM, RAM, non-volatile memory, HDD, etc., and functions as a storage unit. The ROM stores instructions and commands for executing the application and OS according to the present disclosure as programs. Such programs are loaded and executed by the processor 212. The RAM is used to write and read data while the program stored in the ROM is being processed by the processor 212. The non-volatile memory is memory into which data is written and read as the program is executed, and data written therein is retained even after execution of the program has ended. In the present disclosure, the memory 211 particularly stores programs for "processing for conducting an AI interview with a job seeker using a trained interview model" and "processing for extracting employment candidates for the job seeker based on the results of the AI ​​interview." The memory 211 also stores various information listed in the job seeker management table ( FIG. 5A ) and the employer management table ( FIG. 5B ).

[0043] The processor 212 is composed of a CPU (microcomputer) and functions as a control unit for controlling other connected components based on various programs stored in the memory 211. Specifically, the processor 212 reads out from the memory 211 programs for executing applications according to the present disclosure and programs for executing an OS, and executes these programs. In the present disclosure, the processor 212 particularly executes "processing for conducting an AI interview with a job seeker using a trained interview model," "processing for extracting potential employment opportunities for the job seeker based on the results of the AI ​​interview," and the like. The processor 212 may be composed of a single CPU, or may be composed of a combination of multiple CPUs and GPUs.

[0044] The communication interface 213 functions as a communication unit that transmits and receives information to and from other remotely installed server devices and terminal devices 100 via a communication processing circuit and an antenna. The communication processing circuit processes programs and various information used in the processing system 1 to transmit and receive information to and from other server devices and terminal devices 100 according to the progress of processing.

[0045] The communication processing circuit processes based on a wideband wireless communication system such as the LTE system, but can also process based on a system related to narrowband wireless communication or a system related to contactless wireless communication such as a wireless LAN such as IEEE802.11 or Bluetooth (registered trademark).In addition, wired communication can be used instead of or in addition to wireless communication.

[0046] 5A is a diagram conceptually illustrating a job seeker management table stored in the server device 200 according to an embodiment of the present disclosure. Of the information stored in the job seeker management table, the job seeker ID and personal basic information are generated each time a new registration request is received from the first user terminal device 100-1, and the other information is updated as the processing progresses.

[0047] The job seeker management table is additionally generated row by row for each newly generated job seeker ID. In FIG. 5A , "basic personal information," "personal characteristic information," "AI interview information," "work performance information," and "work preference information" are stored in association with a "job seeker ID." A "job seeker ID" is a unique identifier assigned to each job seeker who is a first user seeking employment. The "job seeker ID" may be, for example, a serial number, or may be composed of arbitrary letters and numbers set by the job seeker, or may be an email address or a part thereof. Therefore, the "job seeker ID" may be freely set as long as it can identify the job seeker.

[0048] "Basic personal information" includes information of a job seeker that is not dependent on abilities, personality, etc., among the job seeker's personal information. For example, "basic personal information" may include name, gender, age, birthplace, nationality, current address, past addresses, work history, etc. In contrast, "personal characteristic information" includes information of a job seeker that is dependent on abilities, personality, etc., among the job seeker's personal information. For example, "personal characteristic information" may include the results of general psychological tests, behavioral tests, technical tests, language tests, or aptitude tests. These tests are not limited to those conducted by the server device 200, but may also be tests that the job seeker takes independently. Furthermore, these tests may be designed so that the job seeker can take them unconsciously by playing a general game. In particular, a game may be played in the metaverse, a three-dimensional virtual space constructed on the Internet, and the job seeker's behavioral test may be conducted by the game in the metaverse.

[0049] "AI interview information" is information related to the results of an AI interview that the server device 200 provides to a job seeker before the matching process. For example, the "AI interview information" may include text information on the content of the question and answer session between the job seeker and the AI ​​during the AI ​​interview. In other words, data such as the job seeker's answers to the AI's questions is stored. The "AI interview information" may also include the job seeker's movements and gestures during the AI ​​interview and the analysis results thereof. The movements and gestures may include eye and mouth movements, as well as other body movements. The analysis results may also include whether the job seeker is calm, whether there is a possibility of suspicious behavior, etc. The "AI interview information" may also include an analysis result of whether the question and answer session contains lies based on the analysis results.

[0050] "Work performance information" is information related to the place of employment when a job seeker receives an employment offer, and information related to the employer's evaluation of the worker after the worker has started working. Therefore, if the place of employment has not been determined, no information will be stored in this information field. Examples of information related to the employer's evaluation of the worker may include information related to work attitude, attendance hours, resignation, personnel evaluation, and promotion / demotion.

[0051] "Work preference information" is information about the job seeker's desired working conditions, place of employment, working environment, working period, etc. This information is stored when the job seeker is registered (i.e., when the job seeker ID is generated), but may be updated each time desired conditions are received from the job seeker.

[0052] 5B is a diagram conceptually illustrating an employer management table stored in the server device 200 according to an embodiment of the present disclosure. Of the information stored in the employer management table, the employer ID and employer basic information are generated each time a new registration request is received from the second user terminal device 100-2, and the other information is updated as the processing progresses.

[0053] The employer management table is additionally generated row by row for each newly generated employer ID. In FIG. 5B , "employer basic information," "working conditions information," "recruitment history information," and "employee evaluation information" are stored in association with an "employer ID." An "employer ID" is a unique identifier assigned to each employer, who is a second user searching for workers. The "employer ID" may be, for example, a serial number, or may be composed of arbitrary letters and numbers set by the employer, or may be an email address or a part thereof. Therefore, the "employer ID" may be freely set as long as it can identify the employer.

[0054] "Basic employer information" includes publicly available information about the employer company, etc., as well as non-public information obtained from the employer. For example, "basic employer information" may include the company's name, location, base, business area, number of employees, turnover rate, age ratio, ratio of disabled employees, ratio of foreign employees, etc. Furthermore, "basic employer information" may include a score (company score) calculated based on this information and evaluation information from employees.

[0055] "Working conditions information" is information related to various conditions when a job seeker is hired. For example, "working conditions information" may include information such as salary and other benefits, work location, working hours, and holidays. "Recruitment history information" is information related to the employer's past hiring history. For example, "recruitment history information" may include the number of people hired in the past and the job seeker IDs of those who have been hired in the past.

[0056] "Employee evaluation information" is information in which a worker evaluates an employer after a specified period of time has passed since the job seeker began working as an employee, taking into account the worker's work up to that point. Therefore, if no evaluation has been made by the worker, no information will be stored in this information column. Examples of information related to an employee's evaluation of an employer may include the results of engagement scoring, the worker's subjective impression, the work environment, differences from the initial working conditions, etc.

[0057] 6. Processing sequence executed by the processing system 1 (A) Processing up to the implementation of the AI ​​interview Fig. 6A is a diagram showing a processing sequence executed between each terminal device 100 and the server device 200 according to an embodiment of the present disclosure. Specifically, Fig. 6A is a diagram showing a processing sequence executed between each terminal device 100 and the server device 200 from the login process by the first user terminal device 100-1 until the result of the AI ​​interview is registered in the server device 200.

[0058] 6A, first, the processor 112 of the first user terminal device 100-1 receives operational input from the job seeker via the input interface 116 and generates login information (S11). The login information includes the job seeker's ID and password. More specifically, the job seeker's operational input refers to the job seeker operating the first user terminal device 100-1 to launch an employment support service app, and then entering the user's ID and password via the input interface 116 on the app's startup screen. The processor 112 then transmits the generated login information via the communication interface 114 to the server device 200, along with a login request (T11) indicating the job seeker's desire to begin employment support.

[0059] When the login information and login request generated by the first user terminal device 100-1 are received by the server device 200 via the communication interface 213, the processor 212 of the server device 200 performs authentication processing for login based on the login information (S12). Specifically, the processor 212 of the server device 200 compares the received login information with the job seeker information stored in the memory 211 to confirm whether the person making the login request is a registered job seeker. If the processor 212 of the server device 200 confirms registration, it approves the job seeker's login as a result of the authentication processing (T12). On the other hand, if the processor 212 of the server device 200 does not confirm registration, it requests the first user terminal device 100-1 to input the job seeker ID and password again.

[0060] When the result of the authentication process (approval result) by the server device 200 is received by the first user terminal device 100-1 via the communication interface 114, the employment support service is initiated in the first user terminal device 100-1. When the service is initiated, the processor 112 of the first user terminal device 100-1 accepts the job seeker's operational input via the input interface 116 and generates personal basic information and employment preference information (S13). The personal basic information includes the job seeker's name, gender, age, birthplace, nationality, current address, past addresses, work history, etc. The employment preference information includes information such as the job seeker's desired working conditions, place of employment, working environment, and employment timing. The processor 112 of the first user terminal device 100-1 then transmits a registration request to the server device 200 via the communication interface 114, along with the generated personal basic information and employment preference information (T13).

[0061] Next, upon receiving the registration request, the basic personal information, and the employment preference information, the processor 212 of the server device 200 starts an external information acquisition process (S14). Specifically, the processor 212 of the server device 200 identifies the job seeker based on the received basic personal information and acquires information related to the job seeker from outside the server device 200. For example, the processor 212 of the server device 200 may perform tracking on social media or the web to acquire information such as posts by the job seeker on social media, information on related parties such as friends connected on social media (i.e., information on whether or not they have connections with criminals), and posts on various Internet sites. The processor 212 of the server device 200 may also inquire with an agency in the country in which the job seeker resides to confirm that the job seeker is not a criminal or a fugitive, or may acquire information on criminals or fugitives in the relevant country from an external database. Furthermore, the processor 212 of the server device 200 may obtain information on criminals or fugitives from relevant parties as analog information such as on paper, digitize the analog information, and store it in the memory 211 .

[0062] Next, the processor 212 of the server device 200 transmits a request to participate in the game to the first user terminal device 100-1 via the communication interface 213 (T14). Specifically, the processor 212 of the server device 200 requests the job seeker to participate in the game as an example of a pre-test (psychological test, aptitude test, behavioral test) for the job seeker before the AI ​​interview. The reason for conducting a game as an example of a test is to enable the job seeker to take the test in a natural state without being conscious of the test, and to obtain information about the job seeker's latent characteristics.

[0063] When the first user terminal device 100-1 receives the game participation request sent from the server device 200 via the communication interface 114, game processing is performed in the first user terminal device 100-1 and the server device 200 as a test for the job seeker (S15). Specifically, the processor 212 of the server device 200 reads the game program from the memory 211 and provides an environment in which the first user terminal device 100-1 can play the game. The processor 112 of the first user terminal device 100-1 then accepts operational inputs from the job seeker via the input interface 116 and progresses the game. In other words, the game is provided to the first user terminal device 100-1 from the server device 200 via a communication line such as the Internet, and progresses sequentially through the job seeker's operations. Note that testing of job seekers is not limited to games; general psychological tests, behavioral tests, aptitude tests, language tests, and technical tests may be conducted instead of games.

[0064] When the game ends, the processor 212 of the server device 200 stores the received personal basic information, employment preference information, and game results (i.e., test results) in the memory 211 (S16). More specifically, the processor 212 stores each piece of information in the corresponding location in the job seeker management table shown in FIG. 5A.

[0065] Next, the processor 212 of the server device 200 sends a request to conduct an AI interview to the first user terminal device 100-1 via the communication interface 213 (T15). Specifically, the processor 212 of the server device 200 requests the job seeker to conduct an AI interview as a first interview in order to understand the characteristics of the job seeker before matching the job seeker with an employer. By conducting a first interview using an AI interview, it becomes possible to understand the compatibility between the job seeker and the employer before they meet in person, and to perform optimal matching so that the direct interview between the job seeker and the employer is not wasted.

[0066] When the first user terminal device 100-1 receives the request to conduct an AI interview from the server device 200 via the communication interface 114, an AI interview is conducted as a first interview with the job seeker in the first user terminal device 100-1 and the server device 200 (S17). Specifically, the processor 212 of the server device 200 reads the trained interview model from the memory 211 and provides an environment in which the AI ​​interview can proceed in the first user terminal device 100-1. The processor 112 of the first user terminal device 100-1 then accepts operation inputs and voice from the job seeker via the input interface 116 and transmits the voice, etc. to the server device 200. The trained interview model of the server device 200 executes processing to carry out the next question and answer session, etc., in response to the received voice, etc. These processes are repeated, and the AI ​​interview progresses. The processor 212 of the server device 200 or the trained interview model may then provide simultaneous interpretation or convert the conversation into text for foreigners, etc., and transmit the interpretation information or the text information to the first user terminal device 100-1. The processor 212 of the server device 200 or the trained interview model may also record and analyze the movements and gestures of the job seeker during the AI ​​interview. That is, the processor 212 of the server device 200 or the trained interview model may analyze the job seeker's calmness and the possibility of suspicious behavior based on the job seeker's eye and mouth movements or other body movements, and analyze whether the question and answer session during the interview contains any lies.

[0067] When the AI ​​interview is completed, the processor 212 of the server device 200 registers the results of the AI ​​interview (S18). Specifically, the processor 212 of the server device 200 stores the information obtained by the AI ​​interview in the AI ​​interview information column of the job applicant management table shown in Fig. 5A. Furthermore, if personal characteristic information or employment preference information of the job applicant is obtained in the AI ​​interview, the processor 212 of the server device 200 may add this information to the personal characteristic information column or the employment preference information column of the job applicant management table shown in Fig. 5A.

[0068] 6B is a diagram showing a processing sequence executed between each terminal device 100 and the server device 200 according to an embodiment of the present disclosure. Specifically, it is a diagram showing a processing sequence executed between each terminal device 100 and the server device 200 from matching between a job seeker and an employer to performing additional machine learning of the trained interview model and the trained matching model based on pass / fail information after the second interview is conducted.

[0069] First, according to FIG. 6B , the processor 212 of the server device 200 performs a matching process to select potential employers (employers) for the job seeker who has undergone an AI interview (S21). Specifically, the processor 212 of the server device 200 reads the trained matching model stored in the memory 211, inputs various information about the job seeker into the trained matching model, and obtains an output of optimal potential employers. In this case, the number of potential employers is not limited to one, and multiple candidates may be output. The following describes the case where one potential employer is output.

[0070] Next, once the candidate employment destinations are obtained, the processor 212 of the server device 200 transmits the output candidate employment destination information related to the candidate employment destination to the job seeker's first user terminal device 100-1 via the communication interface 213 (T21). The candidate employment destination information includes a reply request regarding whether or not to conduct an interview with the candidate employment destination. The job seeker checks the candidate employment destination information output from the first user terminal device 100-1 and decides whether or not to conduct a face-to-face interview with the candidate employment destination. In response, the processor 212 of the server device 200 transmits the job seeker's information as worker candidate information via the communication interface 213 to the second user terminal device 100-2 of the output candidate employment destination (T22). The candidate employment destination information includes the job seeker's basic personal information, personal characteristic information, AI interview information, and a reply request regarding whether or not to conduct an interview with the job seeker who is a candidate for employment. Therefore, the employer checks the various pieces of information about the job seeker output from the second user terminal device 100-2 and decides whether or not to conduct a face-to-face interview with the job seeker.

[0071] Thereafter, if the job seeker requests an interview with a potential employer, the processor 112 of the first user terminal device 100-1 accepts the job seeker's operational input via the input interface 116 and sends a request for a second interview with the potential employer to the server device 200 (T23). On the other hand, if the employer requests an interview with the job seeker, the processor 112 of the second user terminal device 100-2 accepts the employer's operational input via the input interface 116 and sends a request for a second interview with the job seeker to the server device 200 (T24). In this way, if the job seeker and the employer's requests to conduct an interview match, the processor 212 of the server device 200 performs a second interview arrangement process (S22). Specifically, the processor 212 of the server device 200 arranges the date, time, and location of the second interview, and provides support to the job seeker and employer until the second interview is conducted.

[0072] After that, a second interview is conducted between the job seeker and the employer, and when the employer decides whether or not to hire the job seeker, the processor 112 of the second user terminal device 100-2 accepts the employer's operational input via the input interface 116, generates pass / fail information, and transmits the pass / fail information to the first user terminal device 100-1 and the server device 200 via the communication interface 114 (T25, T26).

[0073] Next, when the server device 200 receives the pass / fail information transmitted from the second user terminal device 100-2, the processor 212 of the server device 200 registers the pass / fail information in the memory 211 (S23). Specifically, the processor 212 of the server device 200 registers the received pass / fail information in the employment performance information of the job seeker management table shown in Fig. 5A. Note that the pass / fail information is output via the output interface 111 of the first user terminal device 100-1, allowing the job seeker to know whether or not they will be hired.

[0074] Next, the processor 212 of the server device 200 performs additional machine learning on the trained interview model and the trained matching model based on various information related to the job seeker and the employer obtained before the pass / fail decision for the job seeker is made (S24). Specifically, the processor 212 of the server device 200 reads the information added to the job seeker management table and the employer management table, and performs additional machine learning based on the read information. This makes it possible to update each model based on information related to the most recent game test, information related to the AI ​​interview, and information related to the second interview (pass / fail information), thereby further improving the accuracy of the AI ​​interview and matching process from the next time onwards.

[0075] (C) Processing Related to Evaluation After Starting Employment Fig. 6C is a diagram showing a processing sequence executed between each terminal device 100 and the server device 200 according to an embodiment of the present disclosure. Specifically, Fig. 6C is a diagram showing a processing sequence executed between each terminal device 100 and the server device 200 after a predetermined period has elapsed since the job seeker started working, until an evaluation of the employer and additional machine learning are performed based on information acquired from the former job seeker who has become an employee and information acquired from the employer.

[0076] 6C , first, the processor 112 of the first user terminal device 100-1 receives operational input from the former job seeker who has become employed via the input interface 116 and generates working environment information (S31). Here, the working environment information may include, for example, the worker's evaluation of the employer, the work content, the working conditions, and the results of engagement scoring. Then, the processor 112 of the first user terminal device 100-1 transmits the generated working environment information to the server device 200 via the communication interface 114 (T31).

[0077] Next, the processor 112 of the second user terminal device 100-2 accepts the employer's operational input via the input interface 116 and generates worker information (S32). Here, the worker information may include, for example, worker evaluation information, such as promotion / demotion, work attitude, attendance time, retirement, and personnel evaluation. The processor 112 of the second user terminal device 100-2 then transmits the generated worker information to the server device 200 via the communication interface 114 (T32).

[0078] Next, the processor 212 of the server device 200 stores the received working environment information and worker evaluation information in memory 211 and evaluates the employer based on the working environment information and information related to other employers (S33). Specifically, the processor 212 of the server device 200 stores various information included in the received working environment information in the "Employer Evaluation Information" field of the employer management table in FIG. 5B, etc. The processor 212 of the server device 200 also stores various information included in the received worker evaluation information in the "Work Performance Information" field of the job seeker management table in FIG. 5A, etc. The processor 212 of the server device 200 then reads and executes a scoring program from memory 211 to perform a scoring of the employer's evaluation. The program uses not only previously stored employer information but also the most recently obtained worker evaluation information (working environment information) to perform an up-to-date scoring that is tailored to the actual employer. The processor 212 of the server device 200 may store the scoring results in the employer management table of FIG. 5B, or may transmit the results to the second user terminal device 100-2.

[0079] Next, the processor 212 of the server device 200 performs additional machine learning on the trained interview model and the trained matching model based on the newly added working environment information, worker evaluation information, and employer scoring information (S34). Specifically, the processor 212 of the server device 200 reads the information added to the job seeker management table and the employer management table and performs additional machine learning based on the read information. This allows each model to be updated based on information on worker and employer evaluations after a certain period of time has passed since the start of employment, thereby further improving the accuracy of subsequent AI interview and matching processes. Furthermore, each model is updated appropriately using information on other workers and other employers, so even if a job seeker or their company has a poor hiring and employment track record, they can receive services related to AI interview and matching processes using more accurate models.

[0080] Next, the processor 212 of the server device 200 generates management information for a specific employer using various information in the job seeker management table of FIG. 5A and the employer management table of FIG. 5B (S35). Here, the management information refers to information that can be utilized for organizational management based on success stories of each employer, which are based on information related to recruitment and actual employment stored in the memory 211 of the server device 200. In other words, information that can be applied to the recruitment and employment of other job seekers is generated based on cases in which each employer successfully hired and employed a job seeker and successfully trained each job seeker. Therefore, even if a company has few success stories, it can receive management information generated based on the success stories of multiple other employers. In other words, each employer can obtain information that can be used to utilize the success stories of other companies in its own management. Specifically, the processor 212 of the server device 200 may generate the management information by reading and executing a program for generating management information from the memory 211, or may generate the management information using another trained model.

[0081] The processor 212 of the server device 200 then transmits the management information to the second user terminal device 100-2 via the communication interface 213 (T33). The employer can use the management information output by the second user terminal device 100-2 to manage the organization. The processor 212 of the server device 200 may also transmit the engagement scoring information described above, which allows the employer to utilize the management information and engagement scoring information to further strengthen the organization of the company.

[0082] Thereafter, the processor 212 of the server device 200 uses the trained matching model that has undergone additional machine learning to perform a matching process with other job seekers and other employers, and performs an output process of the results (S36). Specifically, when a new job seeker makes an inquiry, the processor 212 of the server device 200 may use the trained matching model to extract matching employers from employers around the world and introduce the employer to the job seeker using a direct message or the like (output process). Furthermore, when an employer makes an inquiry about a job opening, the processor 212 of the server device 200 may use the trained matching model to extract matching job seekers from job seekers around the world and introduce the employer to the job seeker using a direct message or the like (output process).

[0083] 7. Processing Flow Executed in the Server Device 200 The processing flow executed in the server device 200 in generating and using the trained interview model and trained matching model will now be described in detail.

[0084] (A) Generation and Use of Trained Estimation Model Fig. 7A is a diagram showing a processing flow executed in the server device 200 according to an embodiment of the present disclosure. Specifically, Fig. 7A is a diagram showing a processing flow relating to the generation to use of a trained interview model according to an embodiment of the present disclosure. Note that this processing flow may be executed by the processor 212 of the server device 200 or by a processor of another processing device.

[0085] 7A, the processor 212 reads, as learning interview data, various pieces of information from the job seeker management table of FIG. 5A and various pieces of information from the employer management table of FIG. 5B (S111). For example, the processor 212 may extract personal basic information, personal characteristic information, employer basic information, and other related information. Furthermore, the processor 212 also reads, as learning data, information related to employment history for which a hiring decision has been made (S112). For example, the processor 212 may extract work history information from the job seeker management table of FIG. 5A and employment history information from the employer management table of FIG. 5B.

[0086] Once the various pieces of information described above are obtained, the processor 212 executes a step of performing machine learning of a trained interview model using the information (S113). As an example, the machine learning is performed by providing these sets of information to a neural network that combines neurons, and repeating learning while adjusting the parameters of each neuron so that the output from the neural network is the same as the correct information. The processor 212 then executes a step of acquiring the trained interview model and storing it in the memory 211 (S114). The acquired trained interview model may be stored in another device connected to the server device 200 via a wired or wireless network.

[0087] On the other hand, when the processor 212 conducts an AI interview using the trained interview model, the processor 212 reads from the memory 211 the basic personal information, personal characteristic information, employment preference information, etc. of the job seeker who will be conducting the AI ​​interview as initial input data (S115). The processor 212 then inputs the input data into the trained estimation model and conducts the AI ​​interview using the trained interview model (S116). Thereafter, the processor 212 stores the AI ​​interview results in the job seeker management table of FIG. 5A (S117).

[0088] Thereafter, the processor 212 performs additional machine learning (S113) based on the AI ​​interview results (S117). The processor 212 also performs additional machine learning (S113) based on working environment information and evaluation information (S118) obtained after a predetermined period has elapsed since the job seeker began working as an employee. That is, the processor 212 repeats additional learning (iterative learning) each time the above-mentioned learning interview data, AI interview results, working environment information, and evaluation information are obtained, thereby improving the output accuracy of the trained interview model.

[0089] (B) Generation and Use of Trained Matching Model Fig. 7B is a diagram showing a processing flow executed in the server device 200 according to an embodiment of the present disclosure. Specifically, Fig. 7B is a diagram showing a processing flow relating to the generation to use of a trained matching model according to an embodiment of the present disclosure. Note that this processing flow may be executed by the processor 212 of the server device 200 or by a processor of another processing device.

[0090] 7B, the processor 212 reads, for example, job seeker information from the job seeker management table of FIG. 5A and employer information from the employer management table of FIG. 5B as learning interview data (S211). For example, the processor 212 may extract personal basic information, personal characteristic information, employer basic information, and other related information. Furthermore, the processor 212 may also read, as learning data, information related to employment history for which a hiring decision has been made (S212). For example, the processor 212 may extract work history information from the job seeker management table of FIG. 5A and hiring history information from the employer management table of FIG. 5B.

[0091] Once the various pieces of information described above are obtained, the processor 212 executes a step of performing machine learning of a trained matching model using the information (S213). For example, the machine learning is performed by providing a set of information to a neural network made up of a combination of neurons, and repeating learning while adjusting the parameters of each neuron so that the output from the neural network is the same as the correct information. The processor 212 then executes a step of acquiring the trained matching model and storing it in the memory 211 (S214). The acquired trained matching model may be stored in another device connected to the server device 200 via a wired or wireless network.

[0092] On the other hand, when performing matching processing using a trained matching model, the processor 212 reads job seeker information or employer information as input data from the memory 211 (S215). The processor 212 then inputs the input data into the trained matching model and executes a matching determination (S216). Thereafter, the processor 212 stores the matching results in the job seeker management table and employer management table of FIG. 5A (S217).

[0093] Thereafter, processor 212 performs additional machine learning (S213) based on the matching result (S217). Processor 212 also performs additional machine learning (S213) based on working environment information and evaluation information (S218) obtained after a predetermined period has elapsed since the job seeker began working as a worker. That is, processor 212 repeats additional learning (iterative learning) each time the above-mentioned learning interview data, matching result, working environment information, and evaluation information are obtained, thereby improving the output accuracy of the trained matching model.

[0094] 8. Modifications In the above-described embodiment, the matching process is performed after the AI ​​interview, but this is not limited to this. That is, the processor 212 of the server device 200 may first perform the matching process after performing the acquired information registration process after the game process. Then, the AI ​​interview with the job seeker may be performed in association with the matching destination.

[0095] The processes and procedures described herein can be realized not only by those explicitly described in this disclosure, but also by software, hardware, or a combination thereof. Specifically, the processes and procedures described herein can be realized by implementing logic corresponding to the processes in a medium such as an integrated circuit, volatile memory, non-volatile memory, magnetic disk, or optical storage. Furthermore, the processes and procedures described herein can be implemented as computer programs and executed by various computers, including terminal devices and server devices.

[0096] Although processes and procedures described herein are described as being performed by a single device, software, component, or module, such processes or procedures may be performed by multiple devices, multiple software, multiple components, and / or multiple modules. Furthermore, although various information described herein is described as being stored in a single memory or storage unit, such information may be stored in multiple memories within a single device or multiple memories distributed across multiple devices. Furthermore, software and hardware elements described herein may be realized by integrating them into fewer components or by decomposing them into more components.

[0097] 9. Summary of an embodiment of the present disclosure A processing device according to an embodiment of the present disclosure is a processing device including at least one processor, wherein the at least one processor is configured to execute a process for conducting an AI interview with a job seeker using a trained interview model, and extracting employment candidates for the job seeker based on a result of the AI ​​interview.

[0098] In the above configuration, the at least one processor may conduct the AI ​​interview taking into account at least one of the job seeker's basic personal information and personal characteristic information. In this case, the job seeker's personal characteristic information may include the results of a test previously administered to the job seeker. Furthermore, in this case, the job seeker's basic personal information may include information related to the job seeker's hometown.

[0099] In the above configuration, the at least one processor may analyze the attitude and behavior of the job seeker during the AI ​​interview and analyze whether there is any lie in the answers of the job seeker.

[0100] In the above configuration, the at least one processor may convert the exchanges in the AI ​​interview into text and store the text in a memory, and in this case, the trained interview model may be subjected to additional machine learning based on the converted exchanges.

[0101] In the above configuration, the trained interview model may be machine-learned using personal characteristic information of individuals who are deemed unsuitable as workers as training interview data.

[0102] In the above configuration, the at least one processor may extract the potential employment destinations using a trained matching model trained by machine learning based on job seeker information, employer information, and hiring history. In this case, the at least one processor may acquire post-employment work environment information from the job seeker whose employer was determined through the AI ​​interview, and perform additional machine learning of the trained interview model and the trained matching model based on the work environment information. In addition, in the above case, the at least one processor may acquire post-employment evaluation information of the job seeker from the employer whose employer was determined through the AI ​​interview, and perform additional machine learning of the trained interview model and the trained matching model based on the evaluation information.

[0103] A processing program according to an embodiment of the present disclosure causes a computer having at least one processor to function in such a way that the at least one processor conducts an AI interview with a job seeker using a trained interview model, and extracts potential employment opportunities for the job seeker based on the results of the AI ​​interview.

[0104] A processing method according to an embodiment of the present disclosure is a processing method executed by at least one processor in a computer having the at least one processor, and includes the steps of conducting an AI interview with a job seeker using a trained interview model, and extracting potential employment opportunities for the job seeker based on the results of the AI ​​interview.

[0105] 100 Terminal device 100-1 First user terminal device 100-2 Second user terminal device 200 Server device

Claims

1. A processing device having at least one processor, the at least one processor being configured to execute a process for: conducting an AI interview with a job seeker using a trained interview model; and extracting potential employment opportunities for the job seeker based on the results of the AI ​​interview.

2. The processing device according to claim 1, wherein the at least one processor conducts the AI ​​interview by taking into account at least one of basic personal information or personal characteristic information of the job seeker.

3. The processing device according to claim 2, wherein the personal characteristic information of the job seeker includes the results of a test previously administered to the job seeker.

4. The processing device according to claim 3, wherein the basic personal information of the job seeker includes information related to the job seeker's hometown.

5. The processing device according to claim 1, wherein the at least one processor analyzes the attitude and behavior of the job seeker during the AI ​​interview and analyzes whether there is any lie in the job seeker's answers.

6. The processing device according to claim 1, wherein the at least one processor converts the exchange in the AI ​​interview into text and stores it in memory.

7. The processing device according to claim 6, wherein the trained interview model has undergone additional machine learning based on the transcribed interactions.

8. The processing device according to claim 1, wherein the trained interview model is trained by machine learning using personal characteristic information of individuals who are judged to be unsuitable as workers as training interview data.

9. The processing device according to claim 1, wherein the at least one processor extracts the potential employers using a trained matching model that is machine-learned based on job seeker information, employer information, and hiring history.

10. The processing device according to claim 9, wherein the at least one processor acquires post-employment work environment information from the job seeker whose employment has been decided through the AI ​​interview, and performs additional machine learning of a trained interview model and a trained matching model based on the work environment information.

11. The processing device described in claim 9, wherein the at least one processor obtains evaluation information of the job seeker after employment from the employer determined through the AI ​​interview, and performs additional machine learning of a trained interview model and a trained matching model based on the evaluation information.

12. A processing program in a computer having at least one processor that causes the at least one processor to function as follows: conduct an AI interview with a job seeker using a trained interview model; and extract potential employment opportunities for the job seeker based on the results of the AI ​​interview.

13. A processing method executed by at least one processor in a computer having the at least one processor, comprising: a step of conducting an AI interview with a job seeker using a trained interview model; and a step of extracting potential employment opportunities for the job seeker based on the results of the AI ​​interview.

Images