Search support system, search support method and program

Abstract

To provide a search support system, etc. capable of reducing a cost required for search.SOLUTION: There is provided a search support system. The search support system includes at least one processor. The processor is configured to execute the following steps by reading a program: a receiving step of receiving selection of a candidate job opening by an employer from at least one job opening registered in a job opening database; an acquisition step of acquiring job opening information relating to the candidate job opening out of the job opening database; and a presentation step of presenting recommended conditions for searching for candidates suitable for the candidate job opening from search object persons including personnel belonging to organizations or outside organizations that recruit through candidate job openings, on the basis of the job opening information and reference information, where the reference information includes a correlation between the job opening information and the recommended conditions.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] As disclosed in Patent Document 1, a technique is known in which an employer searches for job seekers who meet certain conditions based on job seeker information registered by the job seeker. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-269220 Summary of the Invention [Problem to be solved by the invention]

[0004] In the above-mentioned conventional technology, the employer may have to change the search conditions and repeat the search until he or she finds a job seeker who meets the requirements, which results in high search costs.

[0005] In view of the above circumstances, the present invention provides a search support system and the like that can reduce the cost required for searching. [Means for solving the problem]

[0006] According to one aspect of the present invention, a search support system is provided. The search support system includes at least one processor. The processor is configured to execute the following steps by reading a program: In the reception step, a selection of a candidate job by an employer is accepted from at least one job posting registered in a job posting database. In the acquisition step, job information related to the candidate job posting is acquired from the job posting database. In the presentation step, recommended conditions are presented based on the job posting information and reference information to search for candidates suitable for the candidate job posting from among search subjects, including personnel belonging to an organization that is hiring via the candidate job posting and job seekers outside the organization. The reference information includes a correlation between the job posting information and the recommended conditions.

[0007] According to this aspect, when a recruiter selects a job for which he or she wants to search for candidates, search conditions for job seekers suitable for that job are presented as recommended conditions, thereby reducing the cost required for searches by the recruiter. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a configuration diagram showing a search support system 1. FIG. [Figure 2] 2 is a block diagram showing the hardware configuration of the server device 10. FIG. [Figure 3] FIG. 2 is a block diagram showing the hardware configuration of a recruiting party terminal 20 and a job seeker terminal 30. [Figure 4] 1 is a block diagram showing functions realized by a server device 10 (control unit 11), a recruiting party terminal 20 (control unit 21), and a job seeker terminal 30 (control unit 31). [Figure 5] FIG. 10 is a diagram showing an example of a job seeker search screen SD displayed on the recruiter terminal 20. [Figure 6] FIG. 10 is a diagram showing an example of a job offer selection screen JD displayed on the recruiter terminal 20. [Figure 7] FIG. 10 is a diagram showing an example of a job seeker search screen SD with a candidate job offer selected. [Figure 8] FIG. 10 is a diagram showing an example of a job posting JP registered in a job posting database. [Figure 9] 10 is an example of an interaction table IT included in a record. [Figure 10] FIG. 10 is a diagram showing an example of the distribution of annual incomes of job seekers included in a recommended group. [Figure 11] FIG. 10 is a diagram showing an example of a search condition setting screen CD displayed on the recruiter terminal 20. [Figure 12] FIG. 10 is a diagram showing an example of a job seeker search screen SD in which recommended conditions are set as search conditions. [Figure 13] FIG. 10 is a diagram showing an example of a job seeker search result screen RD displayed on the recruiter terminal 20. [Figure 14] 1 is an activity diagram showing the flow of information processing (job seeker search processing) executed by search support system 1. FIG. DETAILED DESCRIPTION OF THE INVENTION

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

[0010] Incidentally, the program for realizing the software appearing in one embodiment may be provided as a non-transitory computer-readable medium, or may be provided so that it can be downloaded 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] Furthermore, various information processing according to an embodiment may realize input and output corresponding to the input. Here, the form of information referenced in such information processing (hereinafter referred to as reference information) is not limited as long as an output is obtained as a result of the input. The reference information may be, for example, rule-based information such as a database, a lookup table, or a predetermined function (including a decision formula such as a regression formula constructed using a statistical method), a trained model that has previously trained the correlation between input and output, or a large-scale language model that can output a desired result by inputting a prompt.

[0012] In one embodiment, a "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 one 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 set consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculations can be performed on a circuit in the broad sense.

[0013] Furthermore, a circuit in the broad sense is a circuit realized by at least an appropriate combination of a circuit, circuitry, processor, memory, etc. The processor may be a general-purpose processor or a dedicated circuit. That is, 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.

[0014] 1. Hardware Configuration This section explains the hardware configuration.

[0015] <Search Support System 1> Fig. 1 is a configuration diagram showing a search support system 1. The search support system 1 comprises a communication line 2, a server device 10, a plurality of recruiter terminals 20, and a plurality of job seeker terminals 30. The server device 10, the recruiter terminals 20, and the job seeker terminals 30 are configured to be able to communicate with each other via the communication line 2. The connection between the server device 10, the recruiter terminals 20, and the job seeker terminals 30 may be wired or wireless.

[0016] The search support system 1 constitutes, for example, part of a recruitment and job search system used by multiple recruiters (first recruiter U1 and second recruiter U2) and multiple job seekers (first job seeker U3 and second job seeker U4). The search support system 1 mainly performs tasks such as allowing recruiters to search for job seekers and sending scouting documents from recruiters to job seekers. In one embodiment, the search support system 1 is comprised of one or more devices or components. These components will be described below.

[0017] <Server device 10> 2 is a block diagram showing the hardware configuration of server device 10. Server device 10 includes a control unit 11, a storage unit 12, a communication unit 13, and a communication bus 14. Control unit 11, storage unit 12, and communication unit 13 are electrically connected within server device 10 via communication bus 14.

[0018] <Control unit 11> The control unit 11 processes and controls the overall operations 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 predetermined programs stored in the storage unit 12. In other words, information processing by 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. These will be described in more detail in the next section. Note that the control unit 11 is not limited to being single, and multiple control units 11 may be provided for each function. A combination of these may also be used.

[0019] <Storage section 12> The memory unit 12 stores various pieces of information defined above. 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 operations. The memory unit 12 stores various programs, variables, etc. related to the server device 10 executed by the control unit 11.

[0020] <Communications Department 13> The communication unit 13 is preferably a wired communication means such as USB, IEEE1394, Thunderbolt (registered trademark), or wired LAN network communication, but may also include wireless LAN network communication, mobile communication such as 3G / LTE / 5G, or BLUETOOTH (registered trademark) communication as needed. That is, it is more preferable to implement it as a collection of multiple communication means. That is, the server device 10 may communicate various information from the outside via the communication unit 13 and the network.

[0021] The server device 10 may be an on-premise server or a cloud server. The cloud server device 10 may provide the above-described functions and processes in the form of, for example, SaaS (Software as a Service) or cloud computing.

[0022] <Recruiter Terminal 20> Fig. 3 is a block diagram showing the hardware configuration of the recruiting party terminal 20 and the job seeker terminal 30. As shown in Fig. 3A, the recruiting party 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 within the recruiting party terminal 20 via the communication bus 26. The recruiting party terminal 20 is an information processing terminal used in the course of business by each recruiter who is a user belonging to an organization that receives services provided by the server device 10. The description of the control unit 21, the memory unit 22, and the communication unit 23 is omitted here, as they are the same as the description of each unit in the server device 10.

[0023] <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 can 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 attached externally. 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, etc. to the input unit 24. Instead of a touch panel, a switch button, a mouse, a trackpad, a QWERTY keyboard, etc. can be used as the input unit 24.

[0024] <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 depending on the type of recruiter terminal 20.

[0025] <Job Seeker Terminal 30> 3B, the job seeker terminal 30 includes a control unit 31, a memory unit 32, a communication unit 33, an input unit 34, an output unit 35, and a communication bus 36. The control unit 31, the memory unit 32, the communication unit 33, the input unit 34, and the output unit 35 are electrically connected within the job seeker terminal 30 via the communication bus 36. The job seeker terminal 30 is an information processing terminal used by each job seeker who is a user receiving services provided by the server device 10. The description of the control unit 31, the memory unit 32, the communication unit 33, the input unit 34, and the output unit 35 is omitted here as they are the same as the description of each unit in the recruiter terminal 20.

[0026] 2. Functional configuration This section describes the functional configuration of this embodiment. Information processing by 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 (at least one processor included in the search support system 1).

[0027] FIG. 4 is a block diagram showing functions realized by the server device 10 (controller 11), the recruiter terminal 20 (controller 21), and the job seeker terminal 30 (controller 31).

[0028] As shown in Fig. 4A, server device 10 (control unit 11) includes a basic display control unit 111, a reception unit 112, an acquisition unit 113, a presentation unit 114, a search unit 115, a scout document creation unit 116, a list registration unit 117, and an artificial intelligence unit 120. As shown in Fig. 4B, recruiter terminal 20 (control unit 21) includes a display unit 211 and an operation acquisition unit 212. As shown in Fig. 4C, job seeker terminal 30 (control unit 31) includes a display unit 311 and an operation reception unit 312.

[0029] <Basic display control unit 111> The basic display control unit 111 is configured to display various information on the recruiter terminal 20 and the job seeker terminal 30. For example, the basic display control unit 111 displays a job advertisement and scouting document created by the recruiter, registration information created by the job seeker, etc. on the display unit 211 of the recruiter terminal 20 or the display unit 311 of the job seeker terminal 30.

[0030] Recruiters include organizations such as for-profit corporations (such as companies), non-profit corporations (such as cooperatives and foundations), and public corporations (such as local governments). Recruiters also include recruitment agencies that act as agents of organizations to mediate between job seekers and organizations. Recruitment agencies are also called headhunters or agents.

[0031] <Reception Department 112> The reception unit 112 is configured to receive input from the recruiter terminal 20 and the job seeker terminal 30. Specifically, the reception unit 112 receives, from the recruiter terminal 20, a selection of a candidate job offer by the recruiter from at least one (typically multiple) job offers registered in the job offer database.

[0032] Candidate job openings are selected from job openings created or registered by employers among the job openings contained in the job opening database. The job opening database is stored, for example, in the storage unit 12. Furthermore, the job opening database is not limited to a database registered or referenced by the job opening / job search service configured by the search support system 1, but may also be a database in which job openings are registered in a service external to the search support system 1 (for example, a human resources system / service such as an applicant tracking system (ATS)). In other words, the job opening information (typically a job posting) acquired by the acquisition unit 113 described below may be job opening information imported from a service external to the search support system 1, or may be job opening information published on the Internet (i.e., available from the Internet).

[0033] The reception unit 112 may accept an upload of job posting data from the recruiter terminal 20. In this case, the job posting indicated in the uploaded job posting is regarded as a candidate job posting. That is, the reception unit 112 may accept a selection of candidate job postings by uploading. The reception unit 112 may also accept an input of access information indicating the location of the job information on the network (for example, a URL or IP address of a recruitment site, recruitment management system, etc.). In this case, the reception unit 112 regards the job posting indicated in the job information downloaded from the network as a candidate job posting. That is, the reception unit 112 may accept a selection of candidate job postings by input of access information.

[0034] The reception unit 112 may display an input screen for selecting candidate job offers from among the multiple job offers registered in the job offer database on the recruiter terminal 20. This allows the recruiter to easily select candidate job offers.

[0035] Specifically, the reception unit 112 displays a screen for receiving input of job search conditions on the recruiting party terminal 20, receives input of job search conditions from the recruiter, and executes a job search based on the search conditions. After executing the search, the reception unit 112 displays an input screen including the search results on the recruiting party terminal 20, and receives selection of candidate job offers from among the jobs included in the search results.

[0036] 5 is a diagram showing an example of a job seeker search screen SD displayed on the recruiter terminal 20. The job seeker search screen SD includes a job offer selection field JF, a search condition input field CF, and a search execution button B13. A job offer selection button B11 and a search condition reflection button B12 are arranged in the job offer selection field JF.

[0037] The job selection button B11 is an object that is displayed when no candidate job is selected. When an operation input is made to the job selection button B11, the job selection screen JD of FIG. 6 is displayed on the employer terminal 20. The search condition reflection button B12 is a button for creating and reflecting search conditions based on the selected candidate job. When an operation input is made to the search condition reflection button B12, the search condition setting screen CD of FIG. 11, which will be described later, is displayed on the employer terminal 20. The search condition setting screen CD may be displayed in the form of a pop-up window or the like, so as to overlap the job seeker search screen SD. Note that when no candidate job is selected, as in the example of FIG. 5, the search condition reflection button B12 is disabled (for example, grayed out) and does not accept any operation input.

[0038] The search condition input field CF includes an occupation selection field IF, an industry selection field OF, a keyword input field KF, an employed company input field EF, a job title / department input field PF, an age selection field AF, and a gender selection field SF. When an operation input is performed on the search execution button B13, a search for job seekers is executed by the search unit 115 using the search conditions entered in the search condition input field CF.

[0039] 6 is a diagram showing an example of a job selection screen JD displayed on the recruiter terminal 20. The job selection screen JD includes a search condition input field IF, a job search button B21, and a search result display field RF.

[0040] The search condition input field IF accepts input of search conditions for candidate job openings from the employer terminal 20. Examples of search conditions for candidate job openings include the job title (position name) and identification number (administration ID). When search conditions have been entered in the search condition input field IF and an operation input is performed on the job search button B21, the reception unit 112 searches for job openings that meet the search conditions from among the job openings registered in the job openings database by the employer. The job opening search results (searched job openings) by the reception unit 112 are displayed in the search result display field RF, for example, in list form. Note that each job opening included in the search results is accompanied by a details confirmation button B22 for displaying the job opening details (job posting).

[0041] The employer selects candidate jobs for which recommended conditions are to be presented from among the jobs displayed in the search result display field RF. For example, the employer makes a selection input in the display area of ​​the job(s) to be used as candidate jobs in the search result display field RF.

[0042] Fig. 7 is a diagram showing an example of the job seeker search screen SD with a candidate job offer selected. In Fig. 7, the job offer selection field JF displays the candidate job offer OJ (job offer name of the candidate job offer) selected by the employer instead of the job offer selection button B11. In addition, the search condition reflection button B12 is enabled and ready to accept input.

[0043] <Acquisition part 113> The acquisition unit 113 is configured to acquire, from the job database, job information related to the candidate job offer received by the reception unit 112. The job information includes not only the job advertisement of the candidate job offer but also behavioral history related to the behavior of the employer or job seeker related to the candidate job offer.

[0044] 8 is a diagram showing an example of a job posting JP registered in a job posting database. The job posting JP includes multiple items such as the name of the position being recruited, job content and working conditions, application qualifications, and appealing points. Note that the job posting JP may also include other items.

[0045] The behavioral history of the candidate job offer includes at least one of a first interaction that the recruiter has with the job seeker based on the candidate job offer and a second interaction that the job seeker has with the candidate job offer.

[0046] Examples of first interactions include an employer sending a scouting document to a job seeker, an employer flagging job seeker information (registered job seeker information) related to a candidate job (e.g., by adding it to favorites, bookmarks, or target lists), passing a job screening (e.g., document screening or interview screening), and concluding a job offer. Job seeker information includes the job seeker's resume, curriculum vitae, and other profile information. A "resume" is a document that primarily contains the job seeker's profile, current situation, educational background, work history, desired working conditions, etc., while a "curriculum vitae," also known as a résumé, is a document in which the job seeker conveys to the employer their work history, experience, skills, qualifications, etc., related to their previous work. Job seeker information may also include the job seeker's desired industry and job type.

[0047] Examples of secondary interactions include a job seeker replying to a scouting document received from an employer, a job seeker viewing job information (such as a job posting), a job seeker flagging a job (such as adding it to a favorite, bookmark, or target list), or applying for a job.

[0048] The job information acquired by the acquisition unit 113 may include at least the job advertisements of the candidate job offers. This improves the quality of the recommended conditions presented by the presentation unit 114. Furthermore, the acquisition unit 113 may acquire both the job advertisements of the candidate job offers and the behavioral history of the candidate job offers as job information.

[0049] <Presentation part 114> The presentation unit 114 is configured to present recommended conditions for searching for candidates suitable for the candidate job offer from among search subjects including job seekers outside the organization that is hiring through the candidate job offer, based on the job offer information and reference information acquired by the acquisition unit 113.

[0050] Job seekers who are search targets have their registration information registered in a job seeker database stored in, for example, the storage unit 12. The recommendation conditions are search conditions for searching for job seekers who are suitable for candidate job openings (job seekers who are candidates) from among the job seekers registered in the job seeker database.

[0051] The reference information is information including a correlation between job information and recommended conditions, and is stored, for example, in the storage unit 12. The reference information is, for example, an estimator constructed so that job information is input and recommended conditions can be output. The reference information may include, for example, a table, a function, a simple algorithm, etc., indicating the correlation between the feature (vector data) extracted from the job information and the recommended conditions. The correlation included in the reference information can be constructed, for example, by statistically analyzing data (search log) that records the job information and search results linked to the job information. Note that the vector data is a vector value quantified from text data using known methods such as natural language processing using morphological analysis, encoding of categorical variables, etc.

[0052] The recommended conditions include various conditions that can be set as search conditions used in searching for job seekers. The recommended conditions include, for example, recommended keywords (recommended queries) for searching for job seekers based on keywords included in job seeker information, recommended attributes for searching for job seekers based on job seeker attributes, etc. The job seeker attributes include, for example, current or past occupations, current or past industries, current or past affiliated organizations, possessed skills, possessed qualifications, annual income, age, etc.

[0053] The recommended conditions may include recommended keywords. That is, the presentation unit 114 may present recommended conditions including recommended keywords. This allows the recruiter to be presented with keywords that verbalize the characteristics of candidate jobs, making it easier for the recruiter to set search keywords themselves.

[0054] The reference information may be a recommended conditions output model of the artificial intelligence unit 120 that has been trained to be able to input job information (job postings and / or behavioral history) and output recommended conditions. In this case, the presentation unit 114 inputs the job information to the recommended conditions output model and causes the recommended conditions output model to output recommended conditions.

[0055] The recommended conditions output model is a learning model trained using a combination of job information for learning and recommended conditions linked to the job information as training data. The recommended conditions used in the training data may be search conditions that were actually used and recorded in the search log. Furthermore, among the search conditions recorded in the search log, only search conditions that hit job seekers who took actions (led to actions) such as sending a scouting document, replying to a scouting document, a job seeker flagging a job offer, a job seeker flagging a job offer, viewing a job offer, applying for a job offer, passing a screening (document screening or interview screening), or concluding a job offer may be used as training data. In the recommended conditions output model, parameters calculated, tuned, etc. through training constitute the correlation of the reference information.

[0056] The recommended conditions output model may also be a generative AI including a large-scale language model. In this case, the presentation unit 114 receives job information as input, inputs a prompt including an instruction to output recommended conditions to the recommended conditions output model, and causes the recommended conditions output model to output the recommended conditions. The presentation unit 114 may generate a prompt that instructs the recommended conditions output model to create recommended conditions from the job information, and input the prompt to the recommended conditions output model. The presentation unit 114 may also input, in addition to the instruction to create and output recommended conditions and the job information, a prompt that inserts, as input and output samples, for example, one or more job information samples and one or more corresponding recommended conditions samples to the recommended conditions output model. The recommended conditions output model creates recommended conditions from the job information according to the input prompt.

[0057] The presentation unit 114 may extract a group (recommended group) of job seekers (search subjects) recommended for the candidate job offer from the job seeker database based on the job information acquired by the acquisition unit 113 and the group extraction information included in the reference information, and further extract recommended conditions based on the recommended group and the condition extraction information included in the reference information. This allows the recommended conditions to be extracted based on information about actual job seekers, thereby improving the quality of the presented recommended conditions.

[0058] <Recommended group selection> The group extraction information is information including a first correlation between job information and job seekers (search subjects) included in the recommended group, and is stored, for example, in the storage unit 12. The group extraction information is, for example, an estimator constructed so as to be able to input job information and output a recommended group.

[0059] For example, the presentation unit 114 first determines the attributes of the job posting of the job information (i.e., the job posting of the candidate job) from the items included in the job posting, and extracts from the records the first instructions or second interactions that have previously involved similar job postings with attributes close to those of the job posting of the candidate job (the same or similar attributes). The records are aggregate data of the behavioral history of many job postings included in the job posting database, and are stored in the storage unit 12, for example, as part of the group extraction information.

[0060] Next, the presentation unit 114 extracts job seekers who are related to the extracted first instruction or second interaction as recommended job seekers. A group consisting of the extracted recommended job seekers is set as a recommended group to be recommended for the candidate job offer.

[0061] Attributes of a job posting include, for example, job type, industry, organization, department, job title, required skills, required qualifications, expected annual salary, and age range for job posting. The closeness of attributes (range of similarity) may be defined in the group extraction information for each keyword indicating an attribute, or may be defined as a threshold for the difference in feature values ​​(vector distance) between keywords representing two attributes. For example, job types that are similar to each other may be defined in the form of a table, or a threshold for the difference in feature values ​​at which two job types are determined to be similar to each other (for example, the lower limit of cosine similarity) may be defined.

[0062] The similarity of attributes between job postings may be determined using a similarity determination model. The similarity determination model is a learning model that is trained to be able to calculate the similarity between two job postings. In addition, the presentation unit 114 may measure the distance between vector data obtained by vectorizing data included in the job posting of a candidate job and vector data obtained by vectorizing data included in the job postings of other jobs, and extract job postings with a close distance as similar job postings.

[0063] "Job seekers related to the first interaction" are, for example, job seekers to whom a scouting document has been sent, job seekers whose job seeker information has been flagged by an employer, job seekers who have passed the screening process for a job offer, job seekers for whom a job offer has been concluded, etc. "Job seekers related to the second interaction" are, for example, job seekers who have replied to a scouting document, and job seekers who have viewed, flagged, applied, etc. for a job posting.

[0064] The record includes an interaction rate for each job seeker for each attribute of the job posting, calculated from the history of first interactions performed by multiple recruiters and second interactions performed by multiple job seekers. The interaction rate indicates the likelihood of a reaction (e.g., a reply to a scouting message) from the job seeker when an action (e.g., sending a scouting message) is taken against the job seeker. In other words, the presentation unit 114 extracts job seekers with a high probability of interaction based on data (records) indicating the type of interaction that has occurred between similar job seekers and candidate job seekers for which a job seeker is to be searched. For example, if the interaction is "sending a scouting message," the presentation unit 114 extracts job seekers with a high probability of receiving a scouting message for the candidate job seeker for which a job seeker is to be searched, based on a combination of data on job seekers that have sent scouting messages in the past and the job seekers to whom those job seekers have sent scouting messages.

[0065] FIG. 9 is an example of an interaction table IT included in a record. The interaction table IT in FIG. 9A includes aggregate values ​​based on the interaction history for combinations of job seekers C1 to C4 and job postings J1 to J5. The numbers in the table are, for example, the total value of points (described below) resulting from interactions that have occurred. For example, a "0" in the table indicates that no interaction has occurred, and a "1" indicates that an interaction (for example, a reply to a scouting document) has occurred.

[0066] The numerical values ​​contained in the interaction table IT of a record may be numerical values ​​related to only the first interaction, or may be numerical values ​​related to only the second interaction. For example, the numerical values ​​contained in the interaction table IT may indicate, with a value of 1 or 0, whether or not the employer sent a scout to the job seeker, or may indicate, with a value of 1 or 0, whether or not the job seeker replied to the scout from the employer.

[0067] In the example of the interaction table IT in Figure 9A, interactions (e.g., replies to the scouting document) have occurred with job posting J1 from job seekers C1, C3, and C4, and future interactions can be expected, whereas no interaction has occurred from job seeker C2, and future interactions cannot be expected.

[0068] The presentation unit 114 identifies job seekers with whom an interaction may occur (for example, job seekers who are likely to receive a scouting document) based on the interaction table IT and the attributes of the job postings of candidate job openings. Specifically, the presentation unit 114 first calculates the similarity between the attributes of the job postings of candidate job openings and the attributes of the job postings included in the record. In more detail, the presentation unit 114 vectorizes the attributes of the job postings and calculates the similarity between the job postings of candidate job openings and each of the multiple job postings included in the record using cosine similarity or the like for the vectors.

[0069] After calculating the similarity, the presentation unit 114 weights the interaction-based numerical values ​​(interaction aggregate values) corresponding to the multiple job postings included in the record according to the similarity with the job posting of the candidate job. The presentation unit 114 identifies job seekers with a large total of the weighted interaction aggregate values ​​(interaction expectation value) as job seekers who may interact with the job posting of the candidate job. The presentation unit 114 extracts, for example, the top 100 people with the highest interaction expectation value as a recommended group. Note that the number of job seekers included in the recommended group is not particularly limited.

[0070] In the example of Figure 9A, when searching for candidates for a new job posting J6, the presentation unit 114 compares the job postings to calculate the similarity between the new job posting J6 and job postings J1 to J5 stored in the records. For example, if the similarities between job postings J1 to J5 and job posting J6 are 0.5, 0.3, 0.5, 0.8, and 0.2, respectively, as shown in Figure 9B, the interaction expectations of job seekers C1 to C4 are 1.2, 1.6, 1.0, and 2.3, respectively, as shown in Figure 9C. Therefore, among job seekers C1 to C4, job seeker C4 is the job seeker with whom interaction is most likely to occur for job posting J6.

[0071] The presenting unit 114 may extract a recommended group using information on job postings of multiple candidate job offers. In this case, the presenting unit 114 may use a vector obtained by combining the vectors of the job postings of multiple candidate job offers as the vector of the attributes of the job posting.

[0072] The presentation unit 114 may extract a recommended group based on points set according to the types of the first interaction and the second interaction. This allows weighting of interactions on both the recruiter's side and the job seeker's side, making it easier to extract job seekers suitable for recommendation.

[0073] For example, the presentation unit 114 assigns 3 points to a job seeker who has received a scouting message, 5 points to a job seeker who has replied to the scouting message, and 1 point to a job seeker who has viewed the job information, and extracts job seekers whose total number of points assigned to them is equal to or greater than a certain value or whose ranking based on the total number of points is equal to or greater than a certain value as a recommended group. In other words, the presentation unit 114 assigns higher points in the order of replying to a scouting message, sending a scouting message, and viewing information.

[0074] The group extraction information may be a group extraction model of the artificial intelligence unit 120 that has been trained to be able to input job information for candidate jobs and output a recommended group. In this case, the presentation unit 114 inputs the job information for candidate jobs into the group extraction model and causes the group extraction model to output a recommended group. This makes it possible to present recommended conditions based on extraction examples of recommended groups from a large number of job postings.

[0075] The group extraction model is a learning model trained using a combination of job information for training and job seekers linked to the job information as training data. Examples of "job seekers linked to job information" include job seekers to whom scouting documents have been sent based on the job indicated in the job information, job seekers for whom the employer has performed an operation such as flagging the job indicated in the job information in their job seeker information, job seekers who have passed the screening for the job indicated in the job information, job seekers for whom the job indicated in the job information has been concluded, job seekers who have replied to scouting documents based on the job indicated in the job information, and job seekers who have performed operations such as viewing, flagging, or applying for the job indicated in the job information. In the group extraction model, parameters calculated, tuned, etc. through learning constitute the first correlation of the group extraction information. The group extraction model can extract job seekers who are likely to interact with candidate job offers (e.g., job seekers who are likely to send scouting documents).

[0076] In particular, the group extraction model may be trained so that it can input job postings of candidate jobs and output recommended groups by referencing records. In this case, the presentation unit 114 inputs the job postings to the group extraction model and causes the group extraction model to output recommended groups. This allows the recommended groups to be appropriately extracted by utilizing a summary of past first instructions or second interactions.

[0077] The group extraction model may also be a generative AI including a large-scale language model. In this case, the presentation unit 114 receives job information as input, inputs a prompt including an instruction to reference the records and output a recommended group to the group extraction model, and causes the group extraction model to output the recommended group. The presentation unit 114 may generate a prompt that instructs the group extraction model to create a recommended group from the job information, and input the prompt to the group extraction model. The presentation unit 114 may also input a prompt to the group extraction model that inserts, as input and output samples, for example, one or more job information samples and one or more corresponding recommended group samples, in addition to the instruction to extract and output the recommended group and the job information. The group extraction model creates a recommended group from the job information in accordance with the input prompt.

[0078] <Extraction of recommended conditions from the recommended population> The condition extraction information used to extract recommended conditions from the recommended population is information including a second correlation between the recommended population and the recommended conditions, and is stored, for example, in the storage unit 12. The condition extraction information is, for example, an estimator constructed so as to be able to output recommended conditions by inputting information on job seekers included in the recommended population.

[0079] The presentation unit 114 may refer to the condition extraction information, compare keywords included in the first search subject information related to job seekers (search subjects) included in the recommended group with keywords included in the second search subject information related to job seekers (search subjects) outside the recommended group, and present the selected keywords as recommended conditions. This makes it possible to present search conditions that make it more likely that job seekers included in the recommended group will be included in the search results, and less likely that job seekers not belonging to the recommended group will be included in the search results.

[0080] Here, "job seekers outside the recommended group" refers to, for example, all job seekers other than those included in the recommended group among all job seekers searched for by the user (employer). Also, "job seekers outside the recommended group" may be job seekers not included in the recommended group who have experience in the same or similar occupations or industries as job seekers included in the recommended group.

[0081] A specific procedure for selecting recommended keywords to be presented as recommendation conditions is, for example, as follows: First, the presentation unit 114 extracts, by natural language processing, primary keywords contained in the job seeker information included in the recommended group. Similarly, the presentation unit 114 extracts, by natural language processing, secondary keywords contained in the job seeker information not belonging to the recommended group. Next, the presentation unit 114 extracts and presents, from the extracted primary keywords, keywords that have not been extracted as secondary keywords as recommended keywords.

[0082] The presentation unit 114 may assign scores to the keywords selected by the above comparison according to the frequency of use in the job seeker information of job seekers included in the recommended group, and may preferentially present keywords with higher scores as recommended keywords. This can improve the accuracy of searches for job seekers included in the recommended group. Specifically, the presentation unit 114 presents, as recommended keywords, keywords with scores above a certain level, or keywords with a rank above a certain level when sorted in descending order of scores. The presentation unit 114 may also present all selected keywords in descending order of score. The score threshold or rank for presenting as recommended keywords is defined in the condition extraction information. Furthermore, the presentation unit 114 may present the plurality of recommended keywords that have been presented to the employer in descending order of score.

[0083] The presentation unit 114 may assign a higher score to the selected keyword the more frequently it is used in the registration information of job seekers included in the recommended group. This makes it easier to search for job seekers included in the recommended group and job seekers who have attributes similar to these job seekers.

[0084] Furthermore, the presentation unit 114 may assign a higher score to the selected keyword the lower the frequency of use of the keyword in the registration information of job seekers outside the recommended group. This makes it more difficult to search for job seekers who do not belong to the recommended group. Note that the presentation unit 114 does not necessarily need to refer to the registration information of job seekers outside the recommended group when selecting recommended keywords. For example, the presentation unit 114 may extract recommended keywords by performing natural language analysis on the job seeker information (resumes, etc.) of job seekers included in the recommended group, score the extracted recommended keywords based on the frequency of appearance in the job seeker information, and present the recommended keywords on the employer terminal 20 in descending order of score.

[0085] Furthermore, the presentation unit 114 may present, as the recommended conditions, parameter ranges indicating the attributes of job seekers (search subjects) included in the recommended group based on the distribution of job seekers (search subjects) included in the recommended group. This makes it possible to present, as the recommended conditions, conditions that specify a volume zone in which the number of job seekers belonging to a specific range is large based on the distribution of job seekers included in the recommended group. The threshold value for the number of job seekers used to set the parameter range (i.e., the threshold value for the parameter range to be set as the recommended conditions) is defined in the condition extraction information.

[0086] An example of a "parameter indicating the attributes of a job seeker" is annual income. FIG. 10 is a diagram showing an example of the distribution of annual incomes of job seekers included in a recommended population. The annual incomes of job seekers are extracted from the current annual incomes of job seekers described in the job seeker information (resumes or curriculum vitae). In the distribution example DT of FIG. 10A, the presentation unit 114 determines that the range of annual incomes from 6 million to 12 million yen is a volume zone (a parameter indicating the attributes of job seekers) in which a large number of job seekers fall, and presents this range as a recommended condition. Note that the number of job seekers, the ratio of the number of job seekers to the total number of job seekers, or the like may be set in advance as a threshold for determining the volume zone. The presentation unit 114 may determine the volume zone based on this threshold and present the determined volume zone as a recommended condition.

[0087] The distribution of job seekers may also be determined using two or more items as axes. For example, as shown in FIG. 10B , the presentation unit 114 may use a heat map HM showing the distribution of annual income and occupational type of job seekers included in the recommended group to present, as recommended conditions, occupations with a large number of job seekers within an annual income range set as the recommended conditions. The heat map HM in FIG. 10B shows the distribution of job seekers included in the recommended group, with annual income information and occupation type information included in the job seeker information as axes. For example, the presentation unit 114 presents, as search conditions, an annual income range of "6 million to 12 million yen" and "Occupation A," "Occupation C," and "Occupation D," which have a large number of job seekers within this annual income range. Furthermore, the presentation unit 114 may present, instead of occupations, annual income ranges and industries as recommended conditions based on the distribution of annual income and industries. Furthermore, recommended conditions may be presented based on the distribution of job seekers with respect to items included in the job seeker information other than annual income, occupation, and industry. In this way, the items used to determine the distribution of job seekers are not limited to items expressed as numerical values.

[0088] The condition extraction information may be a condition extraction model of the artificial intelligence unit 120 that has been trained to be able to input a recommended population and output recommended conditions. In this case, the presentation unit 114 inputs information about job seekers included in the recommended population into the condition extraction model and causes the condition extraction model to output recommended conditions. This makes it possible to present recommended conditions based on examples of recommended conditions extracted from the recommended population.

[0089] The condition extraction model is a learning model that is trained using a combination of a recommended population for learning and the recommended conditions linked to the recommended population as training data. In the condition extraction model, parameters calculated, tuned, etc. through learning constitute the second correlation of the condition extraction information.

[0090] The condition extraction model may also be a generative AI including a large-scale language model. In this case, the presentation unit 114 receives information on multiple job seekers included in the recommended population as input, inputs a prompt including an instruction to output recommended conditions to the condition extraction model, and causes the condition extraction model to output the recommended conditions. The presentation unit 114 may generate a prompt that instructs the condition extraction model to create recommended conditions from the recommended population, and input the prompt to the condition extraction model. In addition to the recommended population and the instruction to extract and output recommended conditions, the presentation unit 114 may also input a prompt to the condition extraction model that includes, as input and output samples, one or more samples of the recommended population and one or more corresponding samples of the recommended conditions. The condition extraction model creates recommended conditions from the recommended population according to the input prompt.

[0091] The presentation unit 114 may display statistical data of job seekers included in the recommended group in a manner that does not identify individuals on the recruiter terminal 20. Examples of statistical data include annual income (average, maximum, or minimum), the number of people by industry or company from which they come, etc.

[0092] Before presenting the created search criteria to the user (employer), the presentation unit 114 may modify the recommended criteria using search results obtained by searching for job seekers using the recommended criteria, and present the modified recommended criteria to the employer. This improves the accuracy of searching for job seekers included in the recommended group, thereby facilitating a reduction in search costs incurred by the employer.

[0093] Specifically, the presentation unit 114 causes the search unit 115 to perform a search using the recommended keywords, parameter ranges, etc. included in the recommendation conditions created based on the recommended group. The presentation unit 114 changes, adds, or deletes the recommended keywords and parameter ranges included in the recommendation conditions in accordance with the proportion of job seekers included in the recommended group in the search results obtained by the search unit 115, etc.

[0094] More specifically, the presentation unit 114 may modify the recommended conditions based on the number of job seekers included in the recommended group among the job seekers included in the search results, and present the modified recommended conditions to the employer. This makes it possible to present recommended conditions that ensure that a certain number or more of recommended job seekers included in the recommended group are detected. For example, the presentation unit 114 repeats modifying the search results and executing the search until the search results include a predetermined threshold or more of recommended job seekers (or until the ratio of recommended job seekers included in the search results to the number of all recommended job seekers is equal to or greater than the threshold). Furthermore, the presentation unit 114 may determine whether or not the recommended conditions need to be modified based on the number of recommended job seekers included in a certain range of the top rankings in the display order of the search results. After completing the modification of the recommended conditions, the presentation unit 114 presents the modified recommended conditions to the employer terminal 20.

[0095] In the flow for correcting the recommended conditions, the presentation unit 114 refers to the job seeker information (resume and curriculum vitae) of the recommended job seeker and corrects the recommended conditions. For example, the presentation unit 114 makes adjustments such as correcting (lowering) the criteria (threshold) for the scores of the recommended keywords to be adopted when selecting recommended keywords based on scores according to the frequency of use in the job seeker information.

[0096] The presentation unit 114 may input the recommended conditions and the search results of the job seeker under the recommended conditions to a condition modification model of the artificial intelligence unit 120, and cause the condition modification model to output modified recommended conditions. The condition modification model is a learning model that is trained so that it can input the recommended conditions and the search results and output modified recommended conditions, and is trained using combinations of the recommended conditions and search results for learning and the modified recommended conditions linked to the recommended conditions and the search results as training data.

[0097] The condition modification model may also be a generative AI including a large-scale language model. In this case, the presentation unit 114 receives the recommended conditions and their search results as input, inputs a prompt including an instruction to output the modified recommended conditions (to re-output the recommended conditions) to the condition modification model, and causes the condition modification model to output the recommended conditions. The presentation unit 114 may generate a prompt that instructs the condition modification model to create modified recommended conditions from the recommended conditions and the search results, and input the prompt to the condition modification model. The presentation unit 114 may also input, to the condition modification model, a prompt that inserts, as input and output samples, for example, one or more sample recommended conditions and search result samples and one or more corresponding samples of modified recommended conditions, in addition to the instruction to modify and output the recommended conditions and the recommended conditions and the search results. The condition modification model creates modified recommended conditions from the recommended conditions and the search results according to the input prompt.

[0098] The presentation unit 114 may present recommended conditions including a first condition including candidates for job type or industry and a second condition including at least one search keyword. Furthermore, the presentation unit 114 may re-present the second condition based on a selection of a candidate included in the first condition by the user (recruiter). This allows the search keyword to be re-presented based on the job type or industry selected by the user (recruiter). As a result, the presentation accuracy of recommended conditions that enable a search for job seekers suitable for a job is improved.

[0099] The presentation unit 114 may, for example, cause a recommended condition output model that outputs recommended conditions in response to input of job information to output the first condition and the second condition, or may use information for condition extraction (including the condition extraction model) to extract the first condition and the second condition from the recommended group using the above-mentioned procedure.

[0100] The presentation unit 114, for example, references the re-presentation information, sets recommended keywords corresponding to at least one job type or industry selected from the candidates included in the first condition, and re-presents the recommended keywords as the second condition to the recruiter terminal 20. The "recommended keywords corresponding to job types or industries" may be, for example, keywords associated with each job type or industry in a table included in the re-presentation information, or may be keywords having features similar to the features (e.g., vector data) of the keywords representing the job type or industry (e.g., cosine similarity equal to or greater than a threshold included in the re-presentation information).

[0101] The re-presentation information may be a re-presentation model of the artificial intelligence unit 120 that has been trained so as to be able to input a job type or industry and output recommended keywords (second conditions). In this case, the presentation unit 114 inputs the job type or industry selected by the employer into the re-presentation model and causes the re-presentation model to output recommended keywords for re-presentation.

[0102] The re-presentation model is a learning model trained using a combination of a training occupation or industry and recommended keywords associated with the occupation or industry as training data. The re-presentation model may also be a generative AI including a large-scale language model. In this case, the presentation unit 114 inputs the occupation or industry and inputs a prompt including an instruction to output recommended keywords to the re-presentation model, causing the re-presentation model to output the recommended keywords. The presentation unit 114 may generate a prompt that instructs the re-presentation model to create recommended keywords from the occupation or industry, and input the prompt to the re-presentation model. The presentation unit 114 may also input a prompt to the re-presentation model that includes, for example, one or more occupation or industry samples and one or more corresponding recommended keyword samples as input and output samples, in addition to the occupation or industry and the instruction to create and output recommended keywords. The re-presentation model creates recommended keywords from the occupation or industry according to the input prompt.

[0103] 11 is a diagram showing an example of a search condition setting screen CD displayed on the recruiter terminal 20. The search condition setting screen CD includes an occupation selection area OA, an industry selection area IA, a keyword input area KA, a cancel button B31, and a confirm button B32.

[0104] The job type selection area OA displays job type candidates (part of the first conditions) included in the recommended conditions created by the presentation unit 114. The user, who is the employer, can select candidates to use as search conditions (job type narrowing conditions) by performing an operation input (checking the checkbox) on the checkbox attached to each candidate. The employer can also select all of the presented job type candidates by checking the checkbox attached to the title ("recommended job type") of the job type selection area OA. It is also possible not to select all job types (not to narrow down by job type).

[0105] The industry selection area IA displays industry candidates (part of the first conditions) included in the recommended conditions created by the presentation unit 114. As with the job selection area OA, the recruiter can select candidates to use as search conditions (industry narrowing conditions) by checking the checkbox attached to each candidate. The recruiter can also select all of the presented industry candidates by checking the checkbox attached to the title ("Recommended Industry") of the industry selection area IA. It is also possible not to select all industries (not to narrow down by industry).

[0106] The keyword input area KA displays recommended keywords (second conditions) included in the recommended conditions created by the presentation unit 114. The recommended keywords are initially placed in either the AND condition box AC or the OR condition box OC (the state immediately after the search condition setting screen CD is displayed). In the example of FIG. 11, the recommended keywords are initially placed in the OR condition box OC, but the recommended keywords may also be initially placed in the AND condition box AC. Also, recommended keywords may be initially placed in each of the AND condition box AC and the OR condition box OC. Furthermore, the recommended keywords initially placed in the AND condition box AC may be different from the recommended keywords initially placed in the OR condition box OC. In other words, the presentation unit 114 may present a group of recommended keywords that can be combined with an AND condition and a group of recommended keywords that can be combined with an OR condition.

[0107] The employer sets (enables) the recommended keywords as search criteria candidates by checking the checkbox next to the title ("Recommended Keywords") of the keyword input area KA. In the initial state, the checkbox next to the title is checked, and the recommended keywords are enabled.

[0108] The presentation unit 114 may accept input to the AND condition frame AC and the OR condition frame OC (editing, adding, or deleting recommended keywords using the input unit 24 of the recruiter terminal 20). The presentation unit 114 may also accept movement of recommended keywords from the AND condition frame AC to the OR condition frame OC, or movement of recommended keywords from the OR condition frame OC to the AND condition frame AC, by a drag operation or the like.

[0109] When a job type or industry is selected in the job type selection area OA or industry selection area IA, the presentation unit 114 re-presents the recommended keywords (second conditions) corresponding to the selected job type or industry in the keyword input area KA. That is, the presentation unit 114 updates the keyword input area KA and replaces the recommended keywords in accordance with the selection of the job type or industry. The re-presented recommended keywords may be displayed in the AND condition box AC or in the OR condition box OC.

[0110] When an operation input is performed on the Cancel button B31, all check boxes for job type, industry, and recommended keywords (candidate settings as search conditions) are cleared. When an operation input is performed on the Confirm button B32, the recommended conditions whose check boxes are checked (candidate settings as search conditions) are reflected in the search condition input field CF on the job seeker search screen SD.

[0111] FIG. 12 is a diagram showing an example of the job seeker search screen SD in which the recommended conditions have been set as search conditions. The job seeker search screen SD in FIG. 12 is obtained by inputting the search conditions (job type, industry, and search keyword) set on the search condition setting screen CD into the search condition input field CF of the job seeker search screen SD in FIG. 7, and is displayed on the recruiter terminal 20, for example, in response to an operation input to the decision button B32 on the job seeker search screen SD in FIG. 11. Note that each job type and industry condition is provided with a years of experience setting object YE and a delete button DB. The years of experience setting object YE is an object that allows the recruiter to select, for example, the number of years of experience to be used as a search condition for each job type or industry from a pull-down list. The years of experience setting object YE allows the recruiter to individually specify the number of years of experience for each set job type or industry, thereby enabling flexible setting of search conditions. The delete button DB is an object for deleting a job type or industry from the search conditions.

[0112] The search unit 115 that executes the search accepts edits of the search conditions from the user (employer) on the job seeker search screen SD on which the recommended conditions have been inserted. Therefore, the employer may execute a search using the search conditions inserted on the job seeker search screen SD (i.e., presented by the presentation unit 114) as they are, or may execute a search after editing the search conditions.

[0113] <Search Section 115> The search unit 115 is configured to execute a job seeker search in the job seeker database based on the search criteria entered by the employer. Specifically, the search unit 115 searches for candidates (job seekers in this embodiment) using the search criteria set by the employer based on the recommended criteria, and links at least one of the searched candidates to candidate jobs. This allows the employer to manage or execute actions, etc., regarding candidates (job seekers) obtained in the search results by linking them to the job information of the candidate jobs selected by the employer. Specifically, the job postings of the candidate jobs selected by the employer, the presented recommended criteria, the search criteria used in the search, and a list of hit job seekers are stored in association with each other.

[0114] The search results by the search unit 115 are linked to information on candidate job openings selected by the employer and stored at least temporarily in the storage unit 12, etc. When an action is executed based on the search results (viewing registered information, creating a scout document, adding to a candidate list, etc.), the information on the candidate job openings linked to the search results is referenced as appropriate.

[0115] 13 is a diagram showing an example of a job seeker search result screen RD displayed on the recruiter terminal 20. The job seeker search result screen RD includes an action input area AA and a job seeker information display area JA.

[0116] The action input area AA is an area for accepting actions for multiple job seekers included in the search results all at once. The action input area AA is arranged with a job seeker selection object SO, a send scout button B41, an add to target list button B42, and an output information button B43. The job seeker selection object SO accepts an operation input for selecting all job seekers included in the search results (100 job seekers in the example of FIG. 13) at once as targets for the action. When a bulk selection is made, the check boxes attached to the job seeker selection object SO are checked. Furthermore, in the job seeker information display area JA, the check boxes of the job seekers who have been selected at once are also checked.

[0117] The send scout button B41 is a button for creating scout documents all at once for the selected job seekers (job seekers who are the target of the action). When an operation input is made to the send scout button B41, the scout document creation unit 116 creates the scout document.

[0118] The Add to Target List button B42 is a button for registering selected job seekers (job seekers who are the target of the action) in a target list all at once. When an operation input is made to the Add to Target List button B42, the list registration unit 117 adds the job seekers to the candidate list.

[0119] The information output button B43 is a button for outputting information about the selected job seeker (job seeker who is the target of the action) on a separate screen or all at once as data (e.g., PDF data). When an operation input is performed on the information output button B43, the information about the selected job seeker is output to the recruiter terminal 20, etc.

[0120] <Scout Document Creation Department 116> The scouting document creation unit 116 is configured to create and send scouting documents to job seekers selected by the recruiter. The scouting documents are sent to job seekers who are registered in the search support system 1 and are candidates for the job.

[0121] Specifically, the scout document creation unit 116 creates a scout document with job postings of candidate job postings attached in advance for job seekers selected by the employer from among the job seekers included in the search results by the search unit 115. This saves the employer the trouble of attaching job postings to the scout document (selecting the job postings to attach). Therefore, by selecting candidate job postings before searching for job seekers, the employer can efficiently perform the process from searching for job seekers to sending the scout document.

[0122] The scout document creation unit 116 stores the linking information stored by the search unit 115 (job postings for candidate job postings selected by the recruiter, the presented recommended conditions, the search conditions used in the search, and a list of hit job seekers), further linking it with the job seeker selected as the recipient of the scout document. When the scout document creation unit 116 receives an instruction to create a scout document for the selected job seeker (for example, the recruiter executes an operation input on the send scout button B41), it retrieves the job postings stored in association with the job seeker and attaches them in advance to the scout document to be created.

[0123] The scout document creation unit 116 may create a scout document by accepting input of characters, etc. from the recruiter terminal 20, or may input a job posting for a candidate job into a scout document creation model in the artificial intelligence unit 120 and have the scout document creation model output a scout document. The scout document creation unit 116 may prepare a scout text template or scout document creation model for each job type or industry.

[0124] The scout document creation model is a learning model that is trained to use job postings as input and scout documents as output. In other words, the scout document creation model is a learning model that is trained using a combination of a training job posting and the scout document data corresponding to that job posting as training data.

[0125] The scout document creation model may also be a generative AI that includes a large-scale language model. In this case, the scout document creation unit 116 inputs a job posting, inputs a prompt including an instruction to output a scout document to the scout document creation model, and causes the scout document creation model to output the scout document. The scout document creation unit 116 may generate a prompt that instructs the scout document creation model to create a scout document from the job posting, and input the prompt to the scout document creation model. The scout document creation unit 116 may also input a prompt that inserts, for example, one or more sample job postings and one or more corresponding scout document samples to the scout document creation model in addition to the scout document creation and output instruction and the job posting. The scout document creation model creates a scout document from the job posting according to the input prompt. The scout document creation unit 116 may further input information about the job seeker to whom the scout document is to be sent into the scout document creation model, which is a large-scale language model, and cause the scout document creation model to create a scout document based on the job seeker information.

[0126] The scout document creation unit 116 may attach a job posting for a candidate job to a scout document created manually by a recruiter or an existing scout document, or may attach a job posting for a candidate job to a scout document output (i.e., automatically generated) by the scout document creation model described above. When attaching a job posting for a candidate job to a scout document created manually by a recruiter or the like, the scout document creation unit 116 acquires a scout document created by the recruiter or the like inputting text using the input unit 24 of the recruiter terminal 20, and attaches the job posting acquired by the acquisition unit 113 to the acquired scout document. When attaching a job posting to an existing scout document, the scout document creation unit 116 acquires an existing scout document by means of uploading a file from the recruiter terminal 20, or the like, and attaches the job posting acquired by the acquisition unit 113 to the acquired existing scout document.

[0127] The scout document creation model may also be a learning model that has been trained to be able to output a scout document with a job posting for a candidate job posting attached. That is, the scout document creation unit 116 may input a job posting into the scout document creation model and cause the scout document creation model to output a scout document with the job posting attached. Alternatively, the scout document creation unit 116 may input an existing scout document and a job posting into the scout document creation model and cause the scout document creation model to output a scout document with the job posting attached.

[0128] <List Registration Section 117> The list registration unit 117 is configured to add job seekers selected by the employer to a candidate list for each job offer. Specifically, the list registration unit 117 adds candidates selected by the employer from among the candidates (job seekers in this embodiment) included in the search results by the search unit 115 to the candidate list for the candidate job offer. This saves the employer the trouble of selecting the target job offer (candidate list) when registering a candidate in the job seeker candidate list. Therefore, by selecting candidate job offers before searching for job seekers, the employer can efficiently perform the process from searching for job seekers to adding them to the candidate list.

[0129] Job seekers included in the candidate list are those selected by the employer to be assigned a candidate label (bookmark) (i.e., instructed to assign a candidate label). Job seekers who have been assigned a candidate label are registered in a candidate list prepared for each job posting. A candidate list is called, for example, a favorite list, bookmark list, or target list. By referring to the candidate list, registered employers can select job seekers (candidates) who have been assigned a candidate label to take actions such as viewing registered information, creating and sending scouting documents, etc.

[0130] <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.

[0131] The artificial intelligence unit 120 is an AI (Artificial Intelligence) equipped with learning models such as Transformers including GPT (Generative Pretrained Transformer, including GPT-1, GPT-2, and GPT-3), BERT (Bidirectional Encoder Representations from Transformers), and BART (Bidirectional and Auto-regressive Transformer), and language models such as Recurrent Neural Networks (RNNs), and may include generative AI.

[0132] 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.

[0133] 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 consists of pairs of input data for learning and output data (correct answer data). Furthermore, the language model may not only be trained for a specific task, but also be a general-purpose model that can be used for a wide range of tasks.

[0134] The artificial intelligence unit 120 may be a general-purpose natural language processing learning model, such as a large-scale language model (LLM), trained on a huge amount of data. LLMs are learning models that have previously trained on a large amount of data, such as text data (e.g., (i) web content on the Internet, or (ii) data stored in a specified database). They can execute various language processing tasks when given tasks, and can perform a wide range of natural language processing tasks, such as grasping sentence patterns and contexts, answering questions, and generating sentences, according to given prompts. Such general-purpose learning models include language models that can handle various tasks without fine-tuning, such as through one-shot learning or few-shot learning. Furthermore, general-purpose learning models may also be configured to handle various tasks through zero-shot learning. The artificial intelligence used in each functional unit of the control unit 11 may be a separate learning model, or a common general-purpose learning model.

[0135] The learning models included in the artificial intelligence unit 120 (learning models used in each functional unit, such as a group extraction model) can undergo additional learning as transfer learning or fine tuning. For example, each time new job seeker registration information, job posting registration, etc. is generated, the artificial intelligence unit 120 may perform additional learning and fine tuning using this as new training data. This improves the accuracy of the information output from the learning model.

[0136] The learning model included in the artificial intelligence unit 120 may be a learning model (distilled model) obtained by knowledge distillation using an original learning model. In knowledge distillation, a trained model such as a large-scale language model is used as a teacher model, and the parameters of the student model (distilled model) are adjusted to reduce the output loss (Soft Target Loss) of the student model (distilled model) relative to the output (Soft Target) of the teacher model, thereby learning the student model, which becomes the distilled model. Alternatively, the student model may be learned to reduce the output loss (Hard Target Loss) of the student model relative to the correct label (Hard Target) of the teacher data (combination of input data and output data of the learning model). Compared to the original learning model (teacher model), the distilled model has a smaller number of parameters and a smaller processing load while maintaining performance similar to the learning model. Therefore, using a distilled model can reduce the cost of the search support system 1.

[0137] For example, the group extraction model etc. may be a distilled model trained using a combination of input data and output data in a large-scale language model as training data. Furthermore, when the search support system 1 is introduced, a large-scale language model may be used as the group extraction model etc., and once training data from the large-scale language model has been accumulated, a distilled model obtained by knowledge distillation using the training data may be used as the group extraction model etc.

[0138] <Display> The display unit 211 of the recruiting party terminal 20 and the display unit 311 of the job seeker terminal 30 each display a screen indicated by the screen data transmitted from the server device 10.

[0139] <Operation acquisition section> The operation acquisition unit 212 of the recruiting party terminal 20 accepts operations by a user (recruiter) who uses the recruiting party terminal 20. The operation acceptance unit 312 of the job seeker terminal 30 accepts operations by a user (job seeker) who uses the job seeker terminal 30.

[0140] 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.

[0141] This information processing comprises a reception step, an acquisition step, a presentation step, and a search step. In the reception step, the employer's selection of candidate jobs is accepted from among multiple job openings registered in the job database. In the acquisition step, job information related to the candidate jobs is acquired from the job database. In the presentation step, recommended conditions for searching for job seekers suitable for the candidate jobs are presented based on the job information and reference information. In the search step, job seekers are searched for using search conditions set by the employer based on the recommended conditions, and at least one of the job seekers found is linked to the candidate jobs.

[0142] 14 is an activity diagram showing the flow of information processing (job seeker search processing) executed by the search support system 1. Below, the information processing will be explained along with each activity in this activity diagram.

[0143] The job seeker search process begins with the user (recruiter) selecting candidate job offers. The recruiter selects candidate job offers from job offers registered in the job offer database using the recruiter terminal 20 (activity A101). The server device 10 accepts the selection of candidate job offers from the recruiter terminal 20 and obtains job information related to the candidate job offers from the job offer database (activity A102).

[0144] After acquiring the job information, the server device 10 creates recommended conditions based on the job information and the reference information (activity A103). Next, the server device 10 outputs the recommended conditions to the recruiting party terminal 20 (activity A104). As a result, the recommended conditions are displayed (presented) on the recruiting party terminal 20 (activity A105).

[0145] The recruiter edits the presented recommended conditions as appropriate on the recruiter terminal 20 and sets the search conditions (activity A106). After setting the search conditions, the recruiter inputs an instruction to execute a search on the recruiter terminal 20 (activity A107). Upon receiving the instruction to execute a search, the server device 10 executes a search for job seekers using the set search conditions (activity A108). Next, the server device 10 outputs the search results to the recruiter terminal 20 (activity A109). As a result, the search results for job seekers are displayed on the recruiter terminal 20 (activity A110).

[0146] 4. Effect The operation of this embodiment can be summarized as follows: When a recruiter selects a job for which they wish to search candidates, search conditions for job seekers suitable for that job are presented as recommended conditions. This reduces the cost required for searches by recruiters.

[0147] 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 within the scope of the technical idea of ​​the invention.

[0148] 5.Other In the above embodiment, the server device 10 performs various storage and control functions. However, 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 blockchain technology or the like. In particular, the artificial intelligence unit 120 may be an external component of the server device 10. In this case, the external artificial intelligence unit 120 may be provided, for example, by an artificial intelligence service server and configured to receive inputs from each functional unit of the server device 10, receive requests to execute artificial intelligence services, and return the instructed output as a processing result to the server device 10. The artificial intelligence service server may be a server that provides services using a language model as a learning model, or a server that executes language processing tasks using a language model. The artificial intelligence service server may be constructed using LLM. The artificial intelligence service server receives inputs of prompts such as text, images, and voice, and generates and responds to the prompts.

[0149] The control unit 11 may set, as the search target, human resources belonging to the organization that is hiring through the candidate job offer. In other words, the presentation unit 114 may present recommended conditions for searching for in-organization human resources suitable for the candidate job offer. The in-organization human resources who are the search target have human resource information registered in a human resources database stored in the memory unit 12, for example. Furthermore, the search unit 115 may execute an in-organization human resources search in the human resources database based on the search conditions entered by the recruiter, and display the search results for in-organization human resources on the recruiter terminal 20. In this way, when the search target and candidate are in-organization human resources, the processing of each unit can be realized by appropriately replacing "job seeker" in the above embodiment with "in-organization human resources."

[0150] The control unit 11 does not necessarily have to link the candidate job vacancy to the searched candidate (that is, the search result) in the search unit 115.

[0151] The aspect of this embodiment is not limited to the job search support system 1, but may also be an information processing method or a program. The job search support method includes steps executed by the job search support system 1. The program causes a computer to execute the steps of the job search support system 1.

[0152] It may be provided in the following manner.

[0153] (1) A search support system comprising at least one processor, the processor being configured to execute the following steps by reading a program: a reception step of receiving a selection of candidate job openings by an employer from at least one job opening registered in a job opening database; an acquisition step of acquiring job opening information related to the candidate job opening from the job opening database; and a presentation step of presenting recommended conditions for searching for suitable candidates for the candidate job opening based on the job opening information and reference information from among search subjects, including personnel belonging to an organization that is recruiting through the candidate job opening or job seekers outside the organization, wherein the reference information includes a correlation between the job opening information and the recommended conditions.

[0154] (2) In the search support system described in (1) above, the presenting step presents the recommended conditions for searching for job seekers suitable for the candidate job openings.

[0155] (3) In the search support system described in (1) or (2) above, the job information includes at least a job posting.

[0156] (4) The search support system according to any one of (1) to (3) above, wherein the recommended conditions include recommended keywords.

[0157] (5) In the search support system described in any one of (1) to (4) above, in the presentation step, a group of search subjects recommended for the candidate job offer is extracted based on the job information and group extraction information included in the reference information, and the recommended conditions are further extracted based on the group and condition extraction information included in the reference information, wherein the group extraction information includes a correlation between the job information and the search subjects included in the group, and the condition extraction information includes a correlation between the group and the recommended conditions.

[0158] (6) In the search support system described in (5) above, the group extraction information is a group extraction model that has been trained to be able to input the job information and output the group, and in the presentation step, the job information is input to the group extraction model and the group extraction model is made to output the group.

[0159] (7) In the search support system described in (5) or (6) above, in the presentation step, keywords selected by comparing keywords included in first search subject information related to search subjects included in the group with keywords included in second search subject information related to search subjects outside the group are presented as the recommended conditions.

[0160] (8) In the search support system described in (5) or (6) above, in the presentation step, a range of parameters indicating the attributes of the search subjects included in the group is presented as the recommended conditions based on the distribution of the search subjects included in the group.

[0161] (9) In the search support system described in any one of (1) to (9) above, in the reception step, an input screen is displayed for selecting the candidate job from among a plurality of job vacancies registered in the job vacancy database.

[0162] (10) In the search support system described in any one of (1) to (9) above, the recommended conditions include a first condition including candidates for job type or industry and a second condition including at least one search keyword, and in the presentation step, the second condition is re-presented based on the employer's selection of a candidate included in the first condition.

[0163] (11) In the search support system described in any one of (1) to (10) above, the processor is configured to further execute the following steps: in the search step, searching for candidates using search conditions set by the employer based on the recommended conditions, and linking the candidate job openings to at least one of the searched candidates.

[0164] (12) In the search support system described in (11) above, the processor is further configured to execute the following steps: in the scout document creation step, a scout document is created with a job posting for the candidate job already attached for a job seeker selected by the employer from among the job seekers included in the search results in the search step.

[0165] (13) In the search support system described in (11) or (12) above, the processor is configured to further execute the following steps: in the list registration step, candidates selected by the employer from among the candidates included in the search results in the search step are added to a candidate list for the candidate job opening.

[0166] (14) A search support method comprising the steps executed by the search support system according to any one of (1) to (13) above.

[0167] (15) A program for causing a computer to execute each step of the search support system described in any one of (1) to (13) above. Of course, this is not the case.

[0168] Finally, while various embodiments of the present disclosure have been described, they are presented as examples and are not intended to limit the scope of the invention. The novel embodiments may be embodied in various other forms, and various omissions, substitutions, and modifications may be made without departing from the spirit of the invention. Such embodiments and modifications are intended to be included within the scope and spirit of the invention, as well as within the scope of the inventions and their equivalents as defined in the claims. [Explanation of symbols]

[0169] 1: Search support system 2: Communication line 10: Server device 11: Control section 12: Storage section 13: Communications Department 14: Communication bus 20: Recruiter terminal 21: Control unit 22: Storage section 23: Communications Department 24: Input section 25: Output section 26: Communication bus 30: Job seeker terminal 31: Control unit 32: Storage section 33: Communications Department 34: Input section 35: Output section 36: Communication bus 111: Basic display control section 112: Reception 113: Acquisition Department 114: Presentation part 115: Search section 116: Scout Document Creation Department 117: List registration section 120: Artificial Intelligence Department 211:Display section 212: Operation acquisition section 311: Display section 312: Operation reception section

Claims

1. A search support system, at least one processor; The processor is configured to execute the following steps by reading the program: In the acquisition step, job information about candidate jobs is acquired, In the presentation step, the job information is input into a large-scale language model and recommended conditions output by the large-scale language model are presented, wherein the recommended conditions are conditions for searching for candidates suitable for the candidate job from among search subjects including personnel belonging to an organization that is recruiting through the candidate job and job seekers outside the organization.

2. In the search support system according to claim 1, The search support system, wherein the recommendation conditions include conditions regarding at least one of job type and annual salary.

3. In the search support system according to claim 1, In the reception step, the job information is uploaded from the employer's terminal, In the acquiring step, the search support system acquires the uploaded job information.

4. 2. The search support system according to claim 1, the recommendation conditions include a plurality of recommendation keywords; In the presenting step, the search support system presents the recommended keywords in descending order of the scores assigned to each of the recommended keywords.

5. A search support method, comprising: A search support method comprising the steps executed by the search support system according to any one of claims 1 to 4.

6. A program, A program for causing a computer to execute each step of the search support system according to any one of claims 1 to 4.

Images