Recruitment support system, recruitment support method and program

Abstract

To provide a recruitment support system, etc. allowing users to evaluate job postings.SOLUTION: There is provided a recruitment support system. The recruitment support system includes a processor. The processor is configured to execute the following steps: an acquisition step of acquiring information of job postings; and an information generation step of generating expectation value information regarding an occurrence rate or the number of occurrences of actions on the job postings based on content of the job postings and reference information. The reference information includes a correlation between the content of the job postings and the occurrence / non-occurrence or the number of occurrences of actions in the job postings.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a recruitment support system, a recruitment support method, and a program. [Background technology]

[0002] As disclosed in Patent Document 1, a technique is known in which a point is calculated for each item on a job seeker's resume by referring to a prepared evaluation table. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-196433 A Summary of the Invention [Problem to be solved by the invention]

[0004] A job posting, which is a document prepared by an employer when hiring a job seeker, contains different items than a resume prepared by a job seeker and requires evaluation from different perspectives.

[0005] In view of the above circumstances, the present invention provides a recruitment support system etc. capable of evaluating job postings. [Means for solving the problem]

[0006] According to one aspect of the present invention, there is provided a recruitment support system. The recruitment support system includes a processor. The processor is configured to execute the following steps: In the acquisition step, information on a job posting is acquired. In the information generation step, expected value information regarding the occurrence rate or number of occurrences of actions on the job posting is generated based on the content of the job posting and the reference information. The reference information includes a correlation between the content of the job posting and the occurrence or the number of occurrences of actions in the job posting.

[0007] According to this embodiment, it is possible to generate the occurrence rate (expected value) of the action required in the job posting, and therefore the job posting can be objectively evaluated. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a configuration diagram showing a recruitment support system 1. [Diagram 2] 2 is a block diagram showing a hardware configuration of the server device 10. FIG. [Diagram 3] FIG. 2 is a block diagram showing the hardware configuration of the recruiter terminal 20. [Figure 4] FIG. 2 is a block diagram showing functions realized by a server device 10 (control unit 11) and a recruiter terminal 20 (control unit 21). [Diagram 5] FIG. 13 is a diagram showing an example of a job posting JP. [Figure 6] FIG. 11 is an explanatory diagram showing an example of a procedure for acquiring a second vector. [Figure 7] FIG. 2 is a diagram showing an example of an information display screen ID displayed on the recruiter terminal 20. [Figure 8] FIG. 2 is an activity diagram showing the flow of information processing (job posting expected value information generation processing) executed by recruitment support system 1. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. Various characteristic features shown in the following embodiments can be combined with each other.

[0010] Incidentally, the program for realizing the software appearing in this embodiment may be provided as a non-transitory computer-readable recording medium, or may be provided so as to be downloadable from an external server, or may be provided so that the program is started on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0011] In addition, in this embodiment, the term "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In addition, in this embodiment, various information is handled, and this information is represented, for example, by physical values ​​of signal values ​​representing voltage and current, high and low signal values ​​as a binary bit collection consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculation can be performed on the circuit in the broad sense.

[0012] In addition, a circuit in the broad sense is a circuit realized by at least appropriately combining a circuit, circuitry, a processor, a memory, etc. In other words, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0013] 1. Hardware Configuration This section describes the hardware configuration.

[0014] <Recruitment Support System 1> Fig. 1 is a configuration diagram showing a hiring support system 1. The hiring support system 1 comprises a communication line 2, a server device 10, and a plurality of recruiter terminals 20. The server device 10 and the recruiter terminals 20 are configured to be able to communicate with each other via the communication line 2. The connection between the server device 10 and the recruiter terminals 20 may be wired or wireless.

[0015] The hiring support system 1 constitutes part of a recruitment and job search system used by multiple recruiters (a first recruiter U1 and a second recruiter U2). The hiring support system 1 is mainly responsible for managing job postings. In one embodiment, the hiring support system 1 is made up of one or more devices or components. These components are described below.

[0016] <Server device 10> 2 is a block diagram showing a hardware configuration of the server device 10. The server device 10 includes a control unit 11, a storage unit 12, a communication unit 13, and a communication bus 14. The control unit 11, the storage unit 12, and the communication unit 13 are electrically connected to each other inside the server device 10 via the communication bus 14.

[0017] <Control unit 11> The control unit 11 performs processing and control of the overall operation related to the server device 10. The control unit 11 is, for example, a central processing unit (CPU). The control unit 11 realizes various functions related to the server device 10 by reading out a predetermined program stored in the storage unit 12. That is, information processing by software stored in the storage unit 12 can be specifically realized by the control unit 11, which is an example of hardware, and executed as each functional unit included in the control unit 11. These will be described in more detail in the next section. Note that the control unit 11 is not limited to being single, and may be implemented with multiple control units 11 for each function. Also, a combination of these may be used.

[0018] <Storage section 12> The storage unit 12 stores various information defined by the above description. This can be implemented, for example, as a storage device such as a solid state drive (SSD) that stores various programs and the like related to the server device 10 executed by the control unit 11, or as a memory such as a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to the program calculations. The storage unit 12 stores various programs, variables, etc. related to the server device 10 executed by the control unit 11.

[0019] <Communications Division 13> The communication unit 13 is preferably a wired communication means such as USB, IEEE1394, Thunderbolt (registered trademark), wired LAN network communication, etc., but may also include wireless LAN network communication, mobile communication such as 3G / LTE / 5G, BLUETOOTH (registered trademark) communication, etc. as necessary. In other words, it is more preferable to implement it as a collection of multiple communication means. In other words, the server device 10 may communicate various information from the outside via the communication unit 13 and the network.

[0020] The server device 10 may be in an on-premise form or in a cloud form. As the server device 10 in a cloud form, the above-mentioned functions and processes may be provided in the form of, for example, SaaS (Software as a Service) or cloud computing.

[0021] <Employer Terminal 20> Fig. 3 is a block diagram showing the hardware configuration of recruiter terminal 20. As shown in Fig. 3, recruiter terminal 20 comprises a control unit 21, a memory unit 22, a communication unit 23, an input unit 24, an output unit 25, and a communication bus 26. The control unit 21, the memory unit 22, the communication unit 23, the input unit 24, and the output unit 25 are electrically connected inside recruiter terminal 20 via the communication bus 26. Description of the control unit 21, the memory unit 22, and the communication unit 23 will be omitted as they are the same as the description of each unit in server device 10. Note that recruiter terminal 20 may be a terminal operated by a human resources agency that interacts with job seekers on behalf of recruiters.

[0022] <Input section 24> The input unit 24 accepts operation inputs made by the user. The operation inputs are transferred as command signals to the control unit 21 via the communication bus 26. The control unit 21 may execute predetermined control or calculations based on the transferred command signals as necessary. The input unit 24 may be included in the housing of the recruiter terminal 20, or may be externally attached. For example, the input unit 24 may be implemented as a touch panel integrated with the output unit 25. When the input unit 24 is implemented as a touch panel, the user can input tap operations, swipe operations, and the like to the input unit 24. As the input unit 24, a switch button, a mouse, a track pad, a QWERTY keyboard, and the like can be adopted instead of a touch panel.

[0023] <Output section 25> The output unit 25 displays a screen of a graphical user interface (GUI) that can be operated by the user. The output unit 25 may be included in the housing of the recruiter terminal 20, or may be attached externally. Specifically, the output unit 25 may be implemented as a display device such as a CRT display, a liquid crystal display, an organic EL display, or a plasma display. It is preferable that these display devices are implemented by selectively using them according to the type of recruiter terminal 20.

[0024] 2. Functional configuration In this section, the functional configuration of the present embodiment will be described. Information processing by the software stored in the storage unit 12 is specifically realized by the control unit 11, which is an example of hardware, and can be executed as each functional unit included in the control unit 11 (a processor included in the recruitment support system 1).

[0025] FIG. 4 is a block diagram showing functions realized by the server device 10 (control unit 11) and the recruiter terminal 20 (control unit 21).

[0026] As shown in Fig. 4A, the server device 10 (control unit 11) includes a basic display control unit 111, an acquisition unit 112, an information generation unit 113, an information display control unit 114, and an artificial intelligence unit 120. As shown in Fig. 4B, the recruiter terminal 20 (control unit 21) includes a display unit 211 and an operation acquisition unit 212.

[0027] <Basic display control unit 111> The basic display control unit 111 is configured to display various information on the recruiter terminal 20. For example, the basic display control unit 111 displays, on the display unit 211 of the recruiter terminal 20, resumes and curriculum vitae prepared by the job seeker, job advertisements and scouting documents prepared by the recruiter, and the like.

[0028] Recruiters include organizations such as for-profit corporations (e.g., companies), non-profit corporations (e.g., cooperatives, foundations), and public corporations (e.g., local governments). Recruiters also include recruitment agencies that act as agents of organizations to mediate between job seekers and organizations. Recruiters are also called headhunters, agents, etc.

[0029] <Acquisition part 112> The acquisition unit 112 is configured to acquire information on a job posting (hereinafter also referred to as a "target job posting") for which expected value information (score) is to be generated. For example, the acquisition unit 112 displays a list of job postings registered in the job posting database on the recruiter terminal 20, and accepts an input of a selection of the target job posting from the recruiter terminal 20. The acquisition unit 112 may also automatically acquire information on all job postings registered in the job posting database, or any job posting designated by the user, as the target job posting.

[0030] FIG. 5 is a diagram showing an example of a job posting JP. The job posting JP includes a number of items such as the name of the position being advertised, the job content and working conditions (annual salary, job type, industry, work location, etc.), application qualifications, and appealing points as the contents of the job posting. The job posting JP may also include items other than these, such as the title and heading of the job posting. The job posting JP may also include information on the employer (company size (sales, number of employees, etc.), industry, etc.). The job posting database is not limited to the database used by the job posting and job search service provided by the recruiting support system 1, but may be a database in which job postings are registered in an external service, or may be a database of a job posting and job search service other than the recruiting support system 1. The target job posting may be a job posting imported from an external service, or may be a job posting published on the Internet.

[0031] <Information generation unit 113> The information generating unit 113 is configured to generate expected value information regarding the occurrence rate or occurrence number of actions on the target job posting, based on the contents of the target job posting and the reference information. "Actions on the target job posting" include employer actions, which are actions of employers based on the target job posting, and job seeker actions, which are actions of job seekers on the target job posting.

[0032] Employer actions include direct actions initiated by the target job posting (e.g., sending a scouting document based on the target job posting) and indirect actions initiated by another action (e.g., sending a scouting document and making a job offer for a job posting for which a job seeker has responded, or closing a job posting for which a scouting document has been sent). Job seeker actions also include direct actions initiated by the target job posting (e.g., applying for a job posting) and indirect actions initiated by another action (e.g., replying to a scouting document sent by the employer, accepting a job offer from the employer, or closing a job posting for which a job seeker has applied).

[0033] Specifically, the employer actions for which the information generating unit 113 generates expected value information may include at least one of sending a scouting document to a job seeker based on a target job posting, a reply from the job seeker to the scouting document, and a job offer for the job seeker. This makes it possible to obtain the expected value of each action starting from the sending of a scouting document from the employer.

[0034] In addition, the job seeker actions for which the information generating unit 113 generates expected value information may include at least one of an application from a job seeker to a job posting and a job offer made to a job seeker who has applied. This makes it possible to obtain the expected value of each action starting from the job seeker's application.

[0035] Furthermore, the information generating unit 113 may generate, as an action based on the target job posting, information on expectations regarding passing a job interview (first interview, second interview, etc.) and a job offer.

[0036] The reference information includes a correlation between the contents of the job posting and the occurrence or occurrence number of actions in the job posting. The reference information is stored, for example, in the storage unit 12. The reference information is an estimator constructed so as to be able to output expected value information (a predicted occurrence rate or a predicted occurrence number of actions) of the target job posting by inputting the target job posting.

[0037] The reference information may be an expected value calculation model that has been trained to be able to input the target job posting and output expected value information. In this case, the information generating unit 113 inputs the target job posting into the expected value calculation model of the artificial intelligence unit 120, and causes the expected value calculation model to output expected value information. This makes it possible to generate expected value information based on the occurrence or non-occurrence of actual actions in past job postings, or the number of occurrences, by learning and modeling the tendency of job postings that are likely to result in actions such as contract (decision to hire the job seeker) based on a model that has learned information on the reference job posting, which is an actually used job posting, and thereby making it possible to predict the occurrence rate or number of occurrences of actions, thereby improving the accuracy of the predicted occurrence rate or number of occurrences of actions.

[0038] The expected value calculation model is a learning model that predicts whether or not an action will occur in a target job posting, or the number of times it will occur. The expected value calculation model is a learning model that is trained using a reference job posting for learning and data on the occurrence rate or number of occurrences of the corresponding action as training data. A "reference job posting" is a job posting that is registered or has been registered in the job posting database, and that has actually posted a job posting in the past or is currently posting a job posting. In this case, the parameters of the expected value calculation model that have been calculated or tuned by learning correspond to the correlation between the content of the job posting and whether or not an action will occur in the job posting, or the number of times it will occur.

[0039] The expected value calculation model may be a generation AI including a large-scale language model. In this case, the information generation unit 113 inputs the target job posting, inputs a prompt including an instruction to predict and output the occurrence rate or occurrence number of the action of the target job posting to the expected value calculation model, and causes the expected value calculation model to output the occurrence rate or occurrence number of the action. In addition, the information generation unit 113 may input, in addition to the target job posting and the instruction to predict and output the occurrence rate or occurrence number of the action, a prompt in which, for example, one or more samples of the job posting and one or more samples of the occurrence rate or occurrence number of the corresponding action are inserted as input and output samples to the expected value calculation model.

[0040] The information generating unit 113 may generate a probability of a job offer concluded through the scout document as the expected value information. This makes it possible to obtain the probability of a job offer concluded in the target job posting. The probability of a job offer concluded may be expressed as a probability such as a percentage, or may be acquired and expressed as a score. A job offer concluded means that a job seeker accepts a job offer from a recruiter and the employment of the job seeker is decided for the target job posting, and may be called a job offer decision. The probability of a job offer concluded may be rephrased as the ease of conclusion, the expected level of a contract, the possibility of a contract, the ease of decision, the expected level of a decision, or the possibility of decision. In this case, the information generating unit 113 uses a first expected value calculation model that predicts the probability of a job offer concluded in the target job posting as the expected value calculation model. The first expected value calculation model is a learning model that is trained to input the target job posting and output the probability of a job offer concluded (i.e., the occurrence rate of the action of a job offer concluded). The first expectation value calculation model is a learning model that is trained using a reference job posting for learning and the corresponding data on whether or not a job posting has been concluded as training data. In this case, the parameters of the first expectation value calculation model calculated by learning correspond to the correlation between the contents of the job posting and the conclusion rate in the job posting. The data on whether or not a job posting has been concluded is binarized data, for example, with "1" indicating a conclusion and "0" indicating no conclusion. The first expectation value calculation model trained using such data outputs the conclusion probability as a number in the range from 0 to 1.

[0041] The first expectation calculation model may be a generative AI including a large-scale language model. In this case, the information generating unit 113 inputs the target job posting, inputs a prompt including an instruction to predict and output the probability of the job posting of the target job posting into the first expectation calculation model, and causes the first expectation calculation model to output the probability of the job posting. In addition to the instruction to predict and output the probability of the job posting and the target job posting, the information generating unit 113 may input a prompt into the first expectation calculation model, in which, for example, one or more job posting samples and one or more corresponding job posting probability samples are inserted as input and output samples. When the first expectation calculation model is a generative AI, the information generating unit 113 may cause the first expectation calculation model to output text representing the probability of the job posting, such as "easy to conclude" or "difficult to conclude". The text to be output is selected based on the relationship between the threshold value and the probability of the job posting (whether or not the threshold value is exceeded). The threshold value is set based on a statistical value such as the average value of the probability of the job posting.

[0042] The first expectation value calculation model may be trained using a reference job posting in which the number of scout documents transmitted based on the reference job posting is greater than a first threshold value determined in advance, and the presence or absence of a job offer concluded in the reference job posting. That is, the first expectation value calculation model may be machine-trained using a reference job posting in which a certain number of scout documents have been transmitted or more, and data on the presence or absence of a job offer concluded in the reference job posting as training data. The first expectation value calculation model may also be a generation AI including a large-scale language model that has been trained by fine tuning or the like using a reference job posting in which a certain number of scout documents have been transmitted or more, and data on the presence or absence of a job offer concluded in the reference job posting. In many cases, a certain number of scout documents must be transmitted in order to reach a job offer conclusion, and a job posting in which a small number of scout documents have been transmitted (or no scout documents have been transmitted) may have a low probability of a job offer conclusion. Therefore, by excluding a reference job posting in which a small number of scout documents have been transmitted (or no scout documents have been transmitted) from the learning data, the prediction accuracy of the first expectation value calculation model is improved.

[0043] The first threshold value of the number of scout documents sent for the reference job posting used in training the first expectation value calculation model is an arbitrary value, such as 5, 10, 20, etc. In other words, for example, only reference job postings for which 5, 10, or 20 or more scout documents have been sent are used in training the first expectation value calculation model.

[0044] The first expectation value calculation model may also weight the learning of each reference job posting according to the number of scout documents sent. For example, the first expectation value calculation model may be trained using data on reference job postings that are weighted more heavily as the number of scout documents sent based on the reference job posting increases, and the presence or absence of a job offer in the reference job posting.

[0045] Furthermore, the first expected value calculation model may be trained using reference job postings in which the number of scout documents transmitted based on the reference job posting is greater than a first predetermined threshold value, and the presence or absence of a job offer agreement in the reference job posting within a predetermined period from the transmission of the scout document. That is, in the training of the first expected value calculation model, a label of "job offer agreement" is attached to reference job postings in which a certain number of scout documents have been transmitted or more, in which a job offer agreement has been concluded within a certain period from the transmission of the scout document. Also, a label of "no job offer agreement" is attached to reference job postings in which a certain number of scout documents have been transmitted or more, in which no job offer agreement has been concluded, or even if a job offer agreement has been concluded, it has only been concluded after a certain period has passed. Then, the reference job postings with these labels are used as training data. If no conditions are set for the period from the transmission of the scout document to the conclusion of the job offer, the number of job offers that are concluded increases as time passes, and the probability of a job offer agreement may become too high. Therefore, by setting a condition for the period from the sending of the scout document to the conclusion of a job offer in the labeling of the learning data, the number of reference job postings for which a job offer has been concluded can be prevented from continuing to increase, and appropriate learning data can be used. In addition, the conditions for conclusion of a job offer between a reference job posting registered in the database at a relatively old time and a reference job posting registered in the database at a relatively new time can be aligned. As a result, the prediction accuracy of the first expectation value calculation model is improved. The first expectation value calculation model may be trained using reference job postings that have been concluded within a predetermined period from the registration or publication of the job posting.

[0046] The period of time for which a job offer has been concluded for labeling the learning data for the first expectation value calculation model as "job offer concluded" may be any period, such as within 6 months, 12 months, 18 months, etc., from the sending of the scout document. In other words, only reference job postings for which a job offer has been concluded within, for example, 6 months, 12 months, or 18 months from the sending of the scout document are used as data labeled as "job offer concluded."

[0047] When multiple deals have been concluded for one reference job posting, i.e., when multiple job seekers are employed for one reference job posting, the first expectation value calculation model may treat the number of deals concluded per scouting document (i.e., a value normalized by dividing the number of scouting document transmissions by the number of job offers concluded) as the number of deals concluded for the scouting documents for the reference job posting. The number of scouting document transmissions normalized in this way is used in the above-mentioned decision on whether to accept or reject the data for learning and in weighting.

[0048] The conditions for the reference job posting to be used in the learning of the first expectation value calculation model may further include the period of time that has elapsed since the date of registration in the database or the date of disclosure to job seekers, and the period of time from the date of registration or disclosure to the sending of the scouting document. The disclosure date is the date of posting on the job posting / job search service, or the date of disclosure on the Internet. For example, the first expectation value calculation model may use only reference job postings that have been registered between 18 months and 6 months ago from the present time and have had scouting documents sent within 6 months from the date of registration or disclosure as learning data, in addition to meeting the above-mentioned condition of the number of scouting documents sent. The balance between the number of job seekers and the number of job postings may differ depending on the time period, and the ease of closing a deal may differ. However, by limiting the time period when the scouting document was sent, the variation in closing due to the difference in the balance between the number of job seekers and the number of job postings at the time period can be suppressed, and the lead time from the registration of the job posting to closing is also taken into account, improving the prediction accuracy of the first expectation value calculation model.

[0049] The information generating unit 113 may generate the probability of a reply from a job seeker to a scout document as the expected value information. Compared to a contract, a reply from a job seeker to a scout document occurs in a short period from the registration of the job posting, so by acquiring the probability of a reply to the scout document, a score using new information more recent than the contract of a job posting can be obtained. In this case, the information generating unit 113 uses a second expected value calculation model that predicts the probability of a reply from a job seeker for a target job posting as the expected value calculation model. The second expected value calculation model is a learning model that is trained to use the target job posting as input and output the probability of a reply from a job seeker (i.e., the occurrence rate of the action of replying to a scout document). In other words, the second expected value calculation model is a learning model that is trained using the reference job posting and the corresponding data on the presence or absence of a reply to the scout document as training data. In this case, the parameters of the second expected value calculation model calculated by learning correspond to the correlation between the contents of the job posting and the probability of a reply from a job seeker. In addition, the data on the presence or absence of a reply to the scouting document is binarized data, for example, with the presence of a reply being "1" and the absence of a reply being "0." The second expectation value calculation model trained using such data outputs the reply probability as a numerical value in the range from 0 to 1.

[0050] The second expectation calculation model may be a generative AI including a large-scale language model. In this case, the information generating unit 113 inputs the target job posting, inputs a prompt including an instruction to predict and output the probability of a reply to the scout document of the target job posting, into the second expectation calculation model, and causes the second expectation calculation model to output the probability of a reply to the scout document. In addition, the information generating unit 113 may input a prompt into the second expectation calculation model, in which, for example, one or more samples of job postings and one or more corresponding samples of the probability of a reply to the scout document are inserted as input and output samples, in addition to the instruction to predict and output the probability of a reply to the scout document and the target job posting. When the second expectation calculation model is a generative AI, the information generating unit 113 may cause the second expectation calculation model to output text representing the probability of a reply, such as, for example, "easy to reply" or "hard to reply". The text to be output is selected based on the relationship between the threshold and the probability of a reply (whether or not the threshold is exceeded). The threshold is set based on a statistical value, such as the average value of the probability of a reply, for example.

[0051] The second expectation value calculation model may be trained using a reference job posting in which the number of scout documents sent based on the reference job posting is greater than a second threshold value determined in advance, and the presence or absence of replies to the scout documents in the reference job posting. That is, the second expectation value calculation model may be machine-trained using a reference job posting in which a certain number of scout documents have been sent or more, and data on the presence or absence of replies to the scout documents in the reference job posting as training data. The second expectation value calculation model may also be a generation AI including a large-scale language model that has been trained by fine tuning or the like using a reference job posting in which a certain number of scout documents have been sent or more, and data on the presence or absence of replies to the scout documents in the reference job posting. This allows reference job postings in which a small number of scout documents have been sent (or no scout documents have been sent) to be excluded from the learning data, thereby improving the prediction accuracy of the second expectation value calculation model.

[0052] The second threshold for the number of scout documents sent for the reference job posting used in learning the second expectation calculation model is a number (e.g., 5) smaller than the first threshold in the first expectation calculation model. Similarly to the first expectation calculation model, the second expectation calculation model may weight the learning of each reference job posting according to the number of scout documents sent. For example, the second expectation calculation model may be trained using data on reference job postings that are weighted more heavily the greater the number of scout documents sent based on the reference job posting, and the presence or absence of replies to the scout documents in the reference job posting.

[0053] Furthermore, the second expectation value calculation model may be trained using reference job postings in which the number of scout documents sent based on the reference job posting is greater than a predetermined threshold, and the presence or absence of replies to the scout documents in the reference job postings within a predetermined period from the sending of the scout documents. That is, in the training of the second expectation value calculation model, among the reference job postings in which a certain number of scout documents have been sent or more, those that have received replies within a certain period from the sending of the scout documents are labeled as "with reply", and those that have no reply or have only received replies after a certain period have passed are labeled as "no reply", and are used as training data. In this way, by setting a condition for the period from the sending of the scout documents to the reply in the labeling of the data for training, the number of reference job postings that have received replies is prevented from continuing to increase, thereby improving the prediction accuracy of the second expectation value calculation model.

[0054] The reply period for labeling the learning data for the second expectation calculation model as "reply received" is, for example, within 14 days from the sending of the scout document, which is shorter than the period for concluding a job offer (for example, within 6 months) for labeling the learning data for the first expectation calculation model as "concluded job offer." In other words, only reference job postings that have received a reply within 14 days from the sending of the scout document are used as data labeled as "reply received."

[0055] The conditions for the reference job posting to be used in the training of the second expectation value calculation model may further include the period of time that has elapsed since the date of registration in the database or the date of publication to job seekers, and the period of time from the date of registration or publication to the date of sending the scouting document. For example, in addition to satisfying the above-mentioned number of scouting documents sent, the second expectation value calculation model may use only reference job postings that were registered between 12 months and 14 days ago from the current time and for which a scouting document was sent within one year from the date of registration or publication as training data. This improves the accuracy of the second expectation value calculation model, since the probability can be predicted using a reference job posting that is more recent than the first expectation value calculation model.

[0056] The information generating unit 113 may generate a predicted value of the number of scout document transmissions as the expected value information. When the number of candidates who match the contents of the job posting is small and the number of scout documents that the employer can send is small, it may be difficult to recruit personnel even if the probability of concluding a job posting or the probability of replying to the scout document is high. By generating a predicted value of the number of scout document transmissions, the expected value of the number of scout document transmissions based on the target job posting can be obtained as a score. As a result, it is possible to evaluate the job posting based on the predicted number of scout document transmissions. In this case, the information generating unit 113 uses a third expected value calculation model that predicts the number of scout document transmissions for the target job posting as the expected value calculation model. The third expected value calculation model is a learning model that is trained to use the target job posting as input and output the expected value of the number of scout document transmissions (i.e., the number of occurrences of the action of sending to the scout document). In other words, the second expected value calculation model is a learning model that is trained to use the reference job posting and the data on the number of scout document transmissions corresponding to it as teacher data. In this case, the parameters of the third expectation value calculation model calculated by learning correspond to the correlation between the contents of the job posting and the number of scouting documents sent.

[0057] The third expected value calculation model may be a generative AI including a large-scale language model. In this case, the information generating unit 113 inputs the target job posting, inputs a prompt including an instruction to predict and output the number of scout document transmissions based on the target job posting, into the third expected value calculation model, and causes the third expected value calculation model to output the expected number of scout document transmissions. In addition, the information generating unit 113 may input a prompt into the third expected value calculation model, in addition to the prediction and output instruction for the number of scout document transmissions and the target job posting, as input and output samples, for example, a sample of one or more job postings and a sample of the expected number of scout document transmissions corresponding thereto. When the third expected value calculation model is a generative AI, the information generating unit 113 may cause the third expected value calculation model to output text representing the expected number of scout document transmissions, such as "easy to send" and "difficult to send". The text to be output is selected based on the relationship between the threshold and the expected number of transmissions (whether or not the threshold is exceeded). The threshold is set based on a statistical value such as the average number of transmissions, for example.

[0058] The third expectation value calculation model may be trained using a reference job posting that is within a predetermined period from the date of registration in the database or the date of publication to job seekers, and the number of scout documents sent in the reference job posting. In other words, the third expectation value calculation model may be machine-trained using a reference job posting (including those with zero scout document transmissions) that has not yet passed a certain period from the date of registration or publication, and data on the number of scout documents sent in the reference job posting as training data. The third expectation value calculation model may also be a generation AI including a large-scale language model that has been trained by fine tuning, etc., using a reference job posting that has not yet passed a certain period from the date of registration or publication, and data on the number of scout documents sent in the reference job posting. This makes it possible to suppress variations in the number of scout documents sent due to differences in the balance between the number of job offers and job seekers at the time of registration of the job posting.

[0059] Furthermore, the third expectation calculation model may be trained using a reference job posting and the number of scout document transmissions in the reference job posting within a predetermined period from the registration date or publication date of the reference job posting. That is, in the training of the third expectation calculation model, the number of scout documents transmitted within a certain period from the registration date or publication date of the reference job posting is labeled as the "number of scout document transmissions" and used as training data. Therefore, scout documents transmitted after a certain period from the registration date or publication date are not included in the "number of scout document transmissions." In this way, by setting a condition for the period from the registration date or publication date to the transmission of the scout document in the labeling of the training data, the number of scout document transmissions in each reference job posting is prevented from continuing to increase, thereby improving the prediction accuracy of the third expectation calculation model.

[0060] In the learning data for the third expectation value calculation model, the transmission period of the scouting documents to be counted as the "number of scouting documents sent" is, for example, within six months from the registration date or publication date of the reference job posting. In other words, only scouting documents sent within six months from the registration date or publication date of the reference job posting are counted as the "number of scouting documents sent."

[0061] The information generating unit 113 may generate, as the expected value information, first expected value information which is the probability of concluding a job offer, second expected value information which is the probability of receiving a reply from a job seeker, and third expected value information which is a predicted value of the number of scout documents to be sent, and may calculate a score for the job posting based on the first expected value information, the second expected value information, and the third expected value information. This allows the target job posting to be given an overall score which is a combination of the three expected value information.

[0062] For example, the information generating unit 113 may calculate the average value of the weighted values ​​of the first expectation information, the second expectation information, and the third expectation information as the overall score of the job posting. The weighting of the first expectation information may be smaller than the weighting of the second expectation information and the third expectation information. This allows the calculation of an overall score that places emphasis on the second expectation information (scout reply rate) and the third expectation information (number of scout transmissions), which are more recent than the first expectation information (job offer success rate) and also have a correlation with job offer success. As a result, a highly reliable score can be given to the target job posting. The information generating unit 113 may also calculate the overall score from the first expectation information, the second expectation information, and the third expectation information by other methods, without relying on weighting.

[0063] The information generating unit 113 may normalize the overall score to a value between 0 and 1. Furthermore, the information generating unit 113 may rank the target job listings with labels such as "S", "A", "B", "none", etc. according to the distribution of the normalized overall scores. For example, "S" is assigned to target job listings having an overall score in the top xx% or more of the distribution of overall scores.

[0064] The distribution of overall scores used for ranking is determined based on the evaluation criteria of the target job posting set by the employer. For example, if it is desired to evaluate job postings handled by individuals or departments, the target job postings are ranked using the distribution of overall scores for job postings handled by individuals or departments. Also, for example, if it is desired to evaluate job postings handled by an organization, the target job postings are ranked using the distribution of overall scores for job postings within the same organization. Furthermore, for example, if it is desired to check the evaluation of job postings issued by an organization in the world as a whole or within a given industry, the target job postings are ranked using the distribution of overall scores for all job postings, including those of other organizations.

[0065] The information generation unit 113 may classify each of the first expectation information, the second expectation information, and the third expectation information into a level, for example, "high" or "low," and then generate a combined score that combines the level of the first expectation information, the level of the second expectation information, and the level of the third expectation information.

[0066] Furthermore, the information generating unit 113 may input the target job posting to the expectation value calculation model, and cause the expectation value calculation model to output an overall score, which is expected value information (for example, an average value of values ​​weighted for each of the first expectation value information, the second expectation value information, and the third expectation value information). In this case, the information generating unit 113 uses an integrated expectation value calculation model as the expectation value calculation model. The integrated expectation value calculation model is a learning model that is trained to input the target job posting and output the overall score. In other words, the integrated expectation value calculation model is a learning model that is trained using the reference job posting and the corresponding overall score data as teacher data. The integrated expectation value calculation model may be a generation AI that includes a large-scale language model. In this case, the information generating unit 113 inputs the target job posting, inputs a prompt including an instruction to calculate and output the overall score of the target job posting to the integrated expectation value calculation model, and causes the integrated expectation value calculation model to output the overall score. Furthermore, the information generating unit 113 may input to the integrated expectation value calculation model, as input and output samples, a prompt in which, for example, one or more sample job advertisements and one or more corresponding samples of the overall score are inserted, in addition to the overall score calculation / output instruction and the target job advertisement. If the integrated expectation value calculation model is a generation AI, the information generating unit 113 may cause the integrated expectation value calculation model to output text indicating the probability of closing, such as "likely to close" or "hard to close".

[0067] The information generating unit 113 may generate a predicted value of the number of applications from job seekers as the expected value information. This makes it possible to acquire the expected value of the number of applications from job seekers based on the target job advertisement as a score. In this case, the information generating unit 113 uses a fourth expected value calculation model as the expected value calculation model. The fourth expected value calculation model is a learning model trained to take the target job advertisement as input and to output the expected value of the number of applications from job seekers (i.e., the number of occurrences of the action of applying from a job seeker). In other words, the fourth expected value calculation model is a learning model trained using the reference job advertisement and the corresponding data on the number of applications from job seekers as training data.

[0068] The fourth expectation calculation model may be a generative AI including a large-scale language model. In this case, the information generating unit 113 inputs the target job posting, inputs a prompt including an instruction to predict and output the number of applications from job seekers based on the target job posting, to the fourth expectation calculation model, and causes the fourth expectation calculation model to output the expected value of the number of applications from job seekers. In addition, the information generating unit 113 may input, in addition to the instruction to predict and output the number of applications from job seekers and the target job posting, a prompt into which, for example, one or more samples of job postings and samples of the expected value of the number of applications from one or more job seekers corresponding thereto are inserted as input and output samples to the fourth expectation calculation model. When the fourth expectation calculation model is a generative AI, the information generating unit 113 may cause the fourth expectation calculation model to output text expressing the expected value of the number of applications, such as, for example, "easy to apply" and "hard to apply". The text to be output is selected based on the relationship between the threshold value and the expected value of the number of applications (whether or not the threshold value is exceeded). The threshold value is set based on, for example, a statistical value such as the average value of the number of applications.

[0069] The information generating unit 113 may generate the probability of a job offer being concluded through a voluntary application from a job seeker as the expected value information. This makes it possible to obtain the probability of a job offer being concluded in the target job posting as a score. In this case, the information generating unit 113 uses the fifth expected value calculation model as the expected value calculation model. The fifth expected value calculation model is a learning model that is trained to take the target job posting as input and to output the probability of a job offer being concluded (i.e., the occurrence rate of the action of a job offer being concluded). In other words, the fifth expected value calculation model is a learning model that is trained using the reference job posting and the corresponding data on whether or not a job offer has been concluded as training data.

[0070] The fifth expectation calculation model may be a generative AI including a large-scale language model. In this case, the information generating unit 113 inputs the target job posting, inputs a prompt including an instruction to predict and output the probability of the job posting of the target job posting into the fifth expectation calculation model, and causes the fifth expectation calculation model to output the probability of the job posting. In addition to the instruction to predict and output the probability of the job posting and the target job posting, the information generating unit 113 may input, as input and output samples, a prompt into the fifth expectation calculation model, for example, in which one or more job posting samples and one or more corresponding job posting probability samples are inserted. When the fifth expectation calculation model is a generative AI, similar to the first expectation calculation model, the information generating unit 113 may cause the fifth expectation calculation model to output text representing the probability of the job posting, such as "easy to conclude" or "difficult to conclude".

[0071] The fifth expectation value calculation model may be trained using reference job postings with a number of applications from job seekers greater than a third threshold value determined in advance and the presence or absence of a job offer in the reference job posting. In other words, the fifth expectation value calculation model may be machine-trained using reference job postings with a certain number of applications and data on the presence or absence of a job offer in the reference job posting as training data. The fifth expectation value calculation model may also be a generation AI including a large-scale language model trained by fine tuning or the like using reference job postings with a certain number of applications and data on the presence or absence of a job offer in the reference job posting. This eliminates reference job postings with a small number of applications (or no applications) from the training data, improving the prediction accuracy of the fifth expectation value calculation model.

[0072] The third threshold value of the number of applications for the reference job posting used in training the fifth expectation value calculation model is, for example, 100. In other words, only reference job postings with 100 or more applications are used in training the fifth expectation value calculation model.

[0073] Specifically, the feature quantities of a job posting are used as input to the expected value calculation model. The feature quantities of a job posting are data derived from the contents of the job posting, regardless of the format of the job posting. As the feature quantities, for example, feature vectors obtained by converting the contents of a target job posting into vectors expressing its features are used. Here, multiple feature vectors may be obtained from one target job posting, and the multiple feature vectors may be used as feature quantities to be input to the expected value calculation model. Note that the feature quantities of a job posting may be extracted by a method other than vectorization.

[0074] Specifically, the information generating unit 113 may input a feature amount including a first vector obtained by vectorizing the sentences included in the job advertisement to the expected value calculation model, and cause the expected value calculation model to output expected value information. This eliminates the influence of the job advertisement format (item types, order, etc.) and spelling variations, and allows the expected value information based on text information such as sentences included in the job advertisement to be output. This improves the accuracy of the expected value information.

[0075] The information generating unit 113 may input a feature amount including a second vector obtained by vectorizing the attribute information included in the job posting to the expected value calculation model, and cause the expected value calculation model to output expected value information. This improves the accuracy of the expected value information based on the category information (annual salary, job type, industry, work location, etc.) included in the job posting.

[0076] Furthermore, the information generating section 113 may input a feature amount including both the first vector and the second vector to an expected value calculation model, and cause the expected value calculation model to output expected value information.

[0077] The expected value calculation model is trained to receive as input features including at least one (preferably both) of a first vector obtained by vectorizing the text included in the job advertisement and a second vector obtained by vectorizing the attribute information included in the job advertisement, and to output expected value information.

[0078] The information generating unit 113 converts the sentences included in the job posting into a first vector, for example, in the following procedure. First, the information generating unit 113 performs morphological analysis on the text data included in the job posting data, and divides the sentences into words. Furthermore, the divided words are filtered to remove stop words (functional words such as particles and auxiliary verbs) and extract only nouns. Stop words are determined, for example, based on dictionary definitions, frequency of occurrence, and the like. Next, the information generating unit 113 performs word normalization (absorption of spelling variations) on the extracted nouns. Word normalization includes procedures such as standardizing character sets, replacing numbers, and standardizing words using a dictionary. The standardization of character sets includes standardizing uppercase letters to lowercase letters, standardizing half-width kana to full-width kana, and the like. The replacement of numbers is the replacement of numbers included in words with representative symbols representing numbers (for example, "0").

[0079] Thereafter, the information generating unit 113 acquires a first vector by using the text of the job advertisement and the words that have been subjected to the above-mentioned processing, through any vectorization method such as the TF-IDF method.

[0080] The TF-IDF method is a method of quantifying importance using the frequency of occurrence of words in a sentence, and the information generating unit 113 calculates the number of occurrences of each word included in the sentence of the job posting in the job posting. The calculation of the number of occurrences is performed, for example, by a sentence analysis model such as Bag of Words. Next, the information generating unit 113 converts the number of occurrences of words in one job posting into the term frequency (tf) of the word, and further calculates the inverse document frequency (idf), which is the rarity of the frequency of occurrence of each word, in the entire job posting from which the data was obtained, based on the term frequency (tf) of the word in each job posting. Finally, the information generating unit 113 calculates the importance (tf-idf) of each word in each job posting from the term frequency (tf) and the inverse document frequency (idf), and generates a first vector whose components are the importance of each word. The dimension of the first vector is the number of words whose importance is calculated.

[0081] In addition to the above-mentioned TF-IDF method, the LSI method or the LDA method may be used to quantify each word. Furthermore, the information generating unit 113 may generate a first vector from the text of the job posting using a model that performs vectorization based on distributed representation of words, such as Word2vec or BERT. The TF-IDF method has the advantage of being strong in keyword processing and light in processing. Since keywords that appear in job postings have characteristics, it is preferable to use the TF-IDF method. In addition, the TF-IDF method, for example, reduces the weight of words (such as "company" and "department") that appear in many job postings regardless of whether the job postings have been concluded, and therefore the feature amount of the job posting can be suitably determined.

[0082] The information generating unit 113 converts the attribute information included in the job posting into a second vector, for example, by the following procedure. The information generating unit 113 converts the attribute information, which is numerical data or categorical (qualitative) data, into a vector, for example, by a method such as One-Hot encoding. In One-Hot encoding, the information generating unit 113 converts the attribute information, which is numerical data or categorical (qualitative) data, into a One-Hot vector in which each component is 0 or 1. The information generating unit 113 generates the second vector by combining the One-Hot vectors of each piece of attribute information. The "attribute information" is information that is input by selecting from predetermined categories and numerical values, for example, gender, age, annual income, industry, etc.

[0083] FIG. 6 is an explanatory diagram showing an example of a procedure for acquiring a second vector. For example, as shown in FIG. 6A, the information generating unit 113 converts the industries of each job posting shown in the table on the left into the One-Hot vectors shown in the table on the right. In the One-Hot vector in FIG. 6A, if the industry corresponds to the industry assigned to each component, the component is "1", and if not, the component is "0". If the industry of the job posting corresponds to multiple industries of the components of the One-Hot vector, the multiple components are "1". Also, as shown in FIG. 6B, for example, the information generating unit 113 converts the annual income of each job posting shown in the table on the left into the One-Hot vectors shown in the table on the right. In the One-Hot vector in FIG. 6B, the annual income is divided into multiple annual income bands in a certain interval, and if it corresponds to the annual income band assigned to each component, the component is "1", and if it does not correspond, the component is "0". If the annual salary listed in the job posting falls within multiple annual salary bands of the components of the One-Hot vector, multiple components will be set to "1".

[0084] The information generating unit 113 may vectorize the job posting using a vectorization model, which is a generative AI including a large-scale language model. In this case, the information generating unit 113 inputs the job posting, inputs a prompt including an instruction to vectorize and output the job posting to the vectorization model, and causes the vectorization model to output the first vector and / or the second vector. Furthermore, in addition to the job posting and the instruction to vectorize and output the job posting, the information generating unit 113 may input, to the vectorization model, a prompt in which, for example, one or more job posting samples and one or more corresponding first vector and / or second vector samples are inserted.

[0085] <Information display control unit 114> The information display control unit 114 is configured to display the expected value information generated by the information generation unit 113 on the recruiter terminal 20 or the like. The information display control unit 114 presents, for example, job postings with high expected values ​​(scores) on the recruiter terminal 20. FIG. 7 is a diagram showing an example of an information display screen ID displayed on the recruiter terminal 20. On the information display screen ID, the IDs (registration numbers) of job postings with high expected decision values ​​(predicted job closing rates), available positions, and details of special offers for these job postings are displayed. Examples of special offers include campaigns in which the fee for sending scout documents is free or discounted up to a certain number of times. This can promote the sending of scout documents for job postings with high closing probabilities, and as a result, can increase the number of deals that users close.

[0086] The information display control unit 114 may display the score or rank of the job posting generated by the information generation unit 113 on the employer terminal 20 or the job seeker's terminal together with the job posting. The information display control unit 114 may also sort the job postings in order of score or rank and display them as a list, or may present suggestions for modifying the content of job postings with low scores.

[0087] Furthermore, the information display control unit 114 may display the score or rank of the job posting on the terminal of a prospective user (employer or job seeker) who is not registered with the job recruitment / job search service provided by the recruitment support system 1. This can encourage the prospective user to register with the job recruitment / job search service.

[0088] <Artificial Intelligence Department 120> The artificial intelligence unit 120 is configured to receive input from each functional unit and return the instructed output. The artificial intelligence used by the server device 10 in each functional unit may be a common one, or may be prepared individually for each functional unit.

[0089] The artificial intelligence unit 120 is an AI (Artificial Intelligence) equipped with a transformer including GPT (Generative Pretrained Transformer, including GPT-1, GPT-2, and GPT-3), BERT (Bidirectional Encoder Representations from Transformers), BART (Bidirectional and Auto-regressive Transformer), etc., and a language model such as a recurrent neural network (RNN), and may include a generative AI.

[0090] The language model is an example of a learning model based on a machine learning algorithm. Specific examples of machine learning algorithms include nearest neighbor methods, naive Bayes methods, decision trees, support vector machines, and deep learning using neural networks. The artificial intelligence unit 120 can apply the above algorithms as appropriate.

[0091] The artificial intelligence unit 120 may have a trained model constructed by a learning method such as supervised learning, unsupervised learning, or self-supervised learning. In supervised learning, machine learning is performed using training data (training data). The training data is composed of a pair of input data for learning and output data (correct answer data). In addition, the language model may not only be one trained for a specific task, but also a general-purpose model that can be used for a wide range of tasks.

[0092] The artificial intelligence unit 120 may be a general-purpose natural language processing learning model such as a large-scale language model (LLM) that has learned a huge amount of data as an artificial intelligence. Such a general-purpose learning model includes a language model that can handle various tasks without fine tuning by one-shot learning, few-shot learning, etc. Also, the general-purpose learning model may be configured to be capable of handling various tasks by zero-shot learning. The artificial intelligence used in each functional unit of the control unit 11 may be a separate learning model, or may be a common general-purpose learning model.

[0093] The learning model included in the artificial intelligence unit 120 can perform additional learning. For example, the artificial intelligence unit 120 may perform additional learning and fine-tuning each time a new reference job posting is registered and an action is taken on the reference job posting, using these as new training data. This improves the accuracy of the expected value information output from the learning model.

[0094] <Display> The display unit 211 of the recruiter terminal 20 displays a screen indicated by the screen data transmitted from the server device 10.

[0095] <Operation acquisition part> The operation acquisition unit 212 of the recruiter terminal 20 accepts operations by a user (recruiter) who uses the recruiter terminal 20.

[0096] 3. Information processing method This section describes an information processing method of the server device 10. This information processing method is executed by a computer, with each unit of the server device 10 acting as each step.

[0097] Specifically, this information processing method includes an acquisition step, an information generation step, and an information display step. In the acquisition step, information on a target job posting is acquired. In the information generation step, expected value information regarding the occurrence rate or occurrence number of actions on the target job posting is generated based on the contents of the target job posting and reference information. In the information display step, the expected value information generated in the information generation step is displayed.

[0098] 8 is an activity diagram showing the flow of information processing (job advertisement expected value information generation processing) executed by the recruitment support system 1. The information processing will be described below along with each activity in this activity diagram.

[0099] The process of generating expected value information for a job posting begins with the selection of a target job posting by the recruiter. The recruiter selects the target job posting for which expected value information is to be generated from a list presented by the server device 10 on the recruiter terminal 20 (activity A101). Note that if the server device 10 automatically selects the target job posting, the recruiter does not need to select the target job posting (A101).

[0100] The server device 10 acquires information on the target job posting from the job posting database (acquisition step, activity A102). Next, the server device 10 generates expected value information using the contents of the target job posting and reference information (e.g., an expected value calculation model) (information generation step, activity A103).

[0101] The server device 10 outputs the generated expected value information to the recruiting party terminal 20 (information display step, activity A104). As a result, the expected value information of the target job posting is displayed on the recruiting party terminal 20 (activity A105).

[0102] 4. Effect The operation of this embodiment can be summarized as follows. That is, the occurrence rate (expected value) of the action required for the job posting can be generated by using the correlation between the content of the job posting and whether or not an action is taken on the job posting. Therefore, the job posting can be objectively evaluated. Furthermore, it is possible to prioritize jobs for which an action such as sending a scout message is required, using objective data based on actual data from around the world that is free of subjectivity and personal influence.

[0103] Although the embodiment of the present invention has been described above, the present invention is not limited to this, and can be modified as appropriate without departing from the technical concept of the invention.

[0104] 5.Other In the above embodiment, the server device 10 performs various storage and control, but multiple external devices may be used instead of the server device 10. That is, various information and programs may be distributed and stored in multiple external devices using block chain technology or the like. In particular, the artificial intelligence unit 120 may be an external configuration of the server device 10. In that case, the artificial intelligence unit 120, which is an external configuration, is configured to receive input from each functional unit of the server device 10 and return the instructed output to the server device 10.

[0105] The reference information used by the information generating unit 113 does not necessarily have to be a learned model. For example, the reference information may be a table, a function, a simple algorithm, or the like, describing the correlation between a feature extracted from the contents of the target job posting and the occurrence or non-occurrence of an action or the number of occurrences (expected value information). Such reference information is constructed, for example, by a statistical method based on the data of the reference job posting. In other words, the correlation calculated from the reference job posting and the occurrence or non-occurrence or the number of occurrences of an actual action for the reference job posting (expected value information) may be used as reference information to predict the expected value information of the target job posting.

[0106] The aspect of the present embodiment is not limited to the hiring support system 1, and may be an information processing method or a program. The hiring support method includes each step executed by the hiring support system 1. The program causes a computer to execute each step of the hiring support system 1.

[0107] It may be provided in any of the following ways:

[0108] (1) A recruiting support system comprising a processor configured to execute the following steps: an acquisition step acquires job posting information; and an information generation step generates expected value information regarding the occurrence rate or number of actions on the job posting based on the content of the job posting and reference information, wherein the reference information includes a correlation between the content of the job posting and the occurrence or number of occurrences of the action in the job posting.

[0109] (2) In the recruitment support system described in (1) above, the actions include at least one of sending a scouting document to a job seeker based on the job posting, a reply from the job seeker to the scouting document, and concluding a job offer for the job seeker.

[0110] (3) In the recruitment support system described in (2) above, in the information generation step, a probability of the job offer being concluded is generated as the expected value information.

[0111] (4) The recruitment support system according to (2) or (3) above, wherein in the information generating step, a probability of a reply from the job seeker is generated as the expected value information.

[0112] (5) A recruiting support system according to any one of (2) to (4) above, wherein in the information generation step, a predicted value of the number of times the scout document will be sent is generated as the expected value information.

[0113] (6) In the recruitment support system described in (2) above, in the information generation step, first expected value information which is the probability of the job offer being concluded, second expected value information which is the probability of a reply from the job seeker, and third expected value information which is a predicted number of the scout document to be sent are generated as the expected value information, and a score of the job posting is calculated based on the first expected value information, the second expected value information, and the third expected value information.

[0114] (7) In the recruitment support system described in (6) above, in the information generation step, the score is calculated as an average of weighted values ​​for the first expectation information, the second expectation information, and the third expectation information, and the weighting of the first expectation information is smaller than the weighting of the second expectation information and the third expectation information.

[0115] (8) A recruitment support system according to any one of (1) to (7) above, wherein the action includes at least one of an application from a job seeker in response to the job posting and a job offer being concluded for the job seeker who has applied.

[0116] (9) In the recruitment support system described in any one of (1) to (8) above, the reference information is an expectation value calculation model that has been trained to be able to input the job posting and output the expected value information, and in the information generation step, the job posting is input to the expectation value calculation model and the expectation value calculation model is caused to output the expected value information.

[0117] (10) In the recruitment support system described in (9) above, the actions include at least one of a reply from the job seeker to a scout document sent to the job seeker based on the job posting, and a job offer concluded for the job seeker, and the expected value calculation model is trained using reference job postings among the reference job postings in which the number of scout documents sent based on the reference job posting is greater than a predetermined threshold, and the presence or absence of the reply or the job offer concluded in the reference job posting.

[0118] (11) In the recruitment support system described in (10) above, the action includes the conclusion of the job offer, and the expected value calculation model is learned using reference job listings among the reference job listings in which the number of scout documents sent based on the reference job listings is greater than a predetermined threshold, and whether or not the job offer has been concluded in the reference job listing within a predetermined period from the sending of the scout document.

[0119] (12) A recruitment support system according to any one of (9) to (11) above, wherein the action includes sending a scouting document to a job seeker based on the job posting, and the expected value calculation model is trained using reference job postings that are within a predetermined period from the date of registration in a database or the date of disclosure to job seekers, and the number of times the scouting document has been sent for the reference job posting.

[0120] (13) In the recruitment support system described in any one of (9) to (12) above, the expectation value calculation model is trained to input features including vectors of sentences included in the job advertisement and output the expectation value information, and in the information generation step, features including vectors of sentences included in the job advertisement are input to the expectation value calculation model and the expectation value calculation model is caused to output the expectation value information.

[0121] (14) In the recruitment support system described in any one of (9) to (13) above, the expectation value calculation model is trained to input a feature including a vector of attribute information included in the job advertisement and output the expectation value information, and in the information generation step, a feature including a vector of attribute information included in the job advertisement is input to the expectation value calculation model and the expectation value calculation model is caused to output the expectation value information.

[0122] (15) A recruitment support method comprising steps executed by a recruitment support system described in any one of (1) to (14) above.

[0123] (16) A program that causes a computer to execute each step of the recruitment support system described in any one of (1) to (14) above. Of course, this is not the case.

[0124] Finally, although various embodiments according to the present disclosure have been described, these are presented as examples and are not intended to limit the scope of the invention. The novel embodiment can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0125] 1: Recruitment support system 2: Communication lines 10: Server device 11: Control section 12: Storage section 13: Communications Department 14: Communication bus 20: Recruiter terminal 21: Control section 22: Storage section 23: Communications Department 24: Input section 25: Output section 26: Communication bus 111: Basic display control unit 112: Acquisition Department 113: Information generation section 114: Information display control unit 120: Artificial Intelligence Department 211:Display section 212: Operation acquisition section

Claims

1. A recruitment support system, A processor is provided. The processor is configured to perform the steps of: In the acquisition step, the job posting information is acquired, In an information generation step, expected value information regarding the occurrence rate or number of actions on the job posting is generated based on the content of the job posting and reference information, wherein the reference information includes a correlation between the content of the job posting and the occurrence or number of occurrences of the action in the job posting.A recruitment support system.

2. The recruitment support system according to claim 1, The action includes at least one of sending a scouting document to the job seeker based on the job posting, a reply from the job seeker to the scouting document, and concluding a job offer for the job seeker.

3. The recruitment support system according to claim 2, In the information generating step, a probability of the job offer being concluded is generated as the expected value information.

4. The recruitment support system according to claim 2, In the information generating step, a probability of a reply from the job seeker is generated as the expected value information.

5. The recruitment support system according to claim 2, In the information generating step, a predicted value of the number of transmissions of the scouting document is generated as the expected value information.

6. The recruitment support system according to claim 2, In the information generating step, As the expected value information, first expected value information is a probability of the job offer being concluded, second expected value information is a probability of a reply from the job seeker, and third expected value information is a predicted value of the number of transmissions of the scouting document; and A recruitment support system that calculates a score for the job posting based on the first expectation information, the second expectation information, and the third expectation information.

7. 7. The recruitment support system according to claim 6, In the information generating step, Calculating the score by averaging values ​​obtained by weighting the first expectation information, the second expectation information, and the third expectation information; A recruitment support system, wherein the weighting of the first expectation information is smaller than the weightings of the second expectation information and the third expectation information.

8. The recruitment support system according to claim 1, The action includes at least one of an application from a job seeker in response to the job posting and a job offer being concluded for the job seeker who applied.

9. The recruitment support system according to claim 1, The reference information is an expectation value calculation model that has been trained to be able to use the job posting as an input and the expectation value information as an output, In the information generation step, the job posting is input to the expectation value calculation model, and the expectation value calculation model is caused to output the expectation value information.

10. The recruitment support system according to claim 9, The action includes at least one of a reply from the job seeker to a scout document for the job seeker based on the job posting, and a job offer for the job seeker; A recruitment support system in which the expected value calculation model is trained using reference job listings where the number of scout documents sent based on the reference job listing is greater than a predetermined threshold, and the presence or absence of the reply or the conclusion of the job offer in the reference job listing.

11. The recruitment support system according to claim 10, The action includes closing the job offer; A recruitment support system in which the expected value calculation model is trained using reference job listings among the reference job listings, where the number of scout documents sent based on the reference job listings is greater than a predetermined threshold, and whether or not the job offer has been concluded in the reference job listing within a predetermined period from the sending of the scout document.

12. The recruitment support system according to claim 9, The action includes sending a scouting document to a job seeker based on the job posting; A recruitment support system in which the expected value calculation model is trained using reference job postings that are within a predetermined period of time from the date of registration in a database or the date of disclosure to job seekers, and the number of times the scout document was sent for those reference job postings.

13. The recruitment support system according to claim 9, The expected value calculation model is trained to receive a feature amount including a vector of a sentence included in the job posting as an input and to output the expected value information, In the information generation step, a feature amount including a vector of the sentence included in the job posting is input to the expectation value calculation model, and the expectation value calculation model is caused to output the expectation value information, in the recruitment support system.

14. The recruitment support system according to claim 9, The expected value calculation model is trained to receive a feature amount including a vector of attribute information included in the job posting as an input and to output the expected value information, In the information generation step, a feature amount including a vector of attribute information included in the job posting is input to the expectation value calculation model, and the expectation value calculation model is caused to output the expectation value information.

15. A recruitment support method, comprising: A recruitment support method comprising the steps executed by the recruitment support system according to any one of claims 1 to 14.

16. A program, A program causing a computer to execute each step of the recruitment support system according to any one of claims 1 to 14.

Images