Search support system, search support method and program

Abstract

To provide a search support system and so on capable of efficiently searching human resources.SOLUTION: There is provided a search support system. The search support system includes a processor. The processor is configured to execute the following steps: a receiving step of receiving a search query composed of natural language; a search category presentation step of presenting at least one recommended search category for the search query among categories of registration information of a search target person on the basis of category determination reference information, where the category determination reference information includes a correlation between the search query and the recommended search category; and a search step of searching for the search target person on the basis of relevance between the search query and the information included in a search target category of the registration information, where the search target category is a recommended search category or a category selected by the user among the categories of the registration information.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 a database in which human resource information is registered is referenced and a human resource search is performed according to search conditions. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-327028 Summary of the Invention [Problem to be solved by the invention]

[0004] In talent searches, search accuracy can be improved by setting the search scope (search category) of talent information in accordance with the search conditions (search query). However, selecting a search query and setting the corresponding search scope can be a cumbersome task.

[0005] In view of the above circumstances, the present invention provides a search support system that enables efficient search for human resources. [Means for solving the problem]

[0006] According to one aspect of the present invention, there is provided a search support system. The search support system includes a processor. The processor is configured to execute the following steps: In the receiving step, a search query written in natural language is received; In the search category presenting step, at least one recommended search category for the search query from among categories of the registered information of the search target person is presented based on category determination reference information; The category determination reference information includes a correlation between the search query and the recommended search categories; In the search step, a search is performed for the search target person based on the relevance between the search query and information included in the search target category of the registered information; The search target category is a recommended search category or a category selected by the user from among categories of the registered information.

[0007] According to this aspect, recommended search categories are presented for a search query entered in natural language, allowing the user to refer to the recommended search categories and select a search target category as the search target range, thereby enabling the user to efficiently search for personnel. [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] 2 is a block diagram showing the hardware configuration of a manager terminal 20 and a managed person terminal 30. FIG. [Figure 4] 3 is a block diagram showing the hardware configuration of a job seeker terminal 40. FIG. [Figure 5] 1 is a block diagram showing functions realized by a server device 10 (controller 11), a manager terminal 20 (controller 21), a managed person terminal 30 (controller 31), and a job seeker terminal 40 (controller 41). [Figure 6] 10 is a diagram showing an example of a search screen RD displayed on the administrator terminal 20 or the managed person terminal 30. FIG. [Figure 7]10 is a diagram showing another example of the search screen RD displayed on the administrator terminal 20 or the managed person terminal 30. FIG. [Figure 8] 10 is a diagram showing yet another example of the search screen RD displayed on the administrator terminal 20 or the managed person terminal 30. FIG. [Figure 9] 8 is a diagram showing an example of a change in the search screen RD of FIG. 7. FIG. [Figure 10] 10 is a diagram showing another example of search result information RI displayed in the search result display area RF of the search screen RD of FIG. 6 etc. FIG. [Figure 11] FIG. 10 is a diagram showing yet another example of search result information RI displayed in the search result display area RF of the search screen RD of FIG. 6 etc. [Figure 12] 1 is an activity diagram showing the flow of information processing (personnel search processing) executed by the 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 this 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] In this embodiment, the term "unit" may also include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In addition, various types of information are handled in this embodiment, and this information may be represented by, for example, physical values of signal values representing voltages and currents, 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 may be performed on a circuit in the broad sense.

[0012] In addition, a circuit in the broad sense is a circuit realized by at least appropriately combining a circuit, circuitry, a processor, a memory, etc. 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.

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

[0014] <Search Support System 1> 1 is a configuration diagram showing a search support system 1. The search support system 1 includes a communication line 2, a server device 10, a plurality of manager terminals 20, a plurality of managed person terminals 30, and a plurality of job seeker terminals 40. The server device 10, the manager terminal 20, the managed person terminals 30, and the job seeker terminals 40 are configured to be able to communicate with each other via the communication line 2. The connection between the server device 10, the manager terminal 20, the managed person terminals 30, and the job seeker terminals 40 and the communication line 2 may be wired or wireless.

[0015] The search support system 1 constitutes part of a personnel system used by personnel (multiple people to be managed) subject to personnel management within an organization and multiple managers who have relationships with the people to be managed within the organization, and part of a recruitment and job search system used by 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.

[0016] <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.

[0017] <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.

[0018] <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.

[0019] <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.

[0020] 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.

[0021] <Administrator terminal 20> 3 is a block diagram showing the hardware configuration of the administrator terminal 20 and the managed user terminal 30. As shown in FIG. 3A, the administrator terminal 20 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, an output unit 25, and a communication bus 26. The control unit 21, the storage unit 22, the communication unit 23, the input unit 24, and the output unit 25 are electrically connected via the communication bus 26 inside the administrator terminal 20. The description of the control unit 21, the storage unit 22, and the communication unit 23 is the same as the description of each unit in the server device 10, and therefore will be omitted.

[0022] <Input section 24> The input unit 24 accepts operation inputs made by the user. The operation inputs are transferred as command signals to the control unit 21 via the communication bus 26. The control unit 21 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 administrator terminal 20 or may be externally attached. For example, the input unit 24 may be implemented as a touch panel integrated with the output unit 25. When the input unit 24 is implemented as a touch panel, the user can input tap operations, swipe operations, 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.

[0023] <Output section 25> The output unit 25 displays a screen of a graphical user interface (GUI) that can be operated by a user. The output unit 25 may be included in the housing of the administrator 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 used appropriately depending on the type of administrator terminal 20.

[0024] <Management target terminal 30> 3B, the managed user terminal 30 includes a control unit 31, a storage unit 32, a communication unit 33, an input unit 34, an output unit 35, and a communication bus 36. The control unit 31, the storage unit 32, the communication unit 33, the input unit 34, and the output unit 35 are electrically connected via the communication bus 36 inside the managed user terminal 30. The description of the control unit 31, the storage unit 32, the communication unit 33, the input unit 34, and the output unit 35 will be omitted as they are the same as the description of each unit in the administrator terminal 20.

[0025] <Job Seeker Terminal 40> Fig. 4 is a block diagram showing the hardware configuration of job seeker terminal 40. As shown in Fig. 4, job seeker terminal 40 includes a control unit 41, a memory unit 42, a communication unit 43, an input unit 44, an output unit 45, and a communication bus 46. The control unit 41, the memory unit 42, the communication unit 43, the input unit 44, and the output unit 45 are electrically connected via the communication bus 46 inside job seeker terminal 40. The description of the control unit 41, the memory unit 42, the communication unit 43, the input unit 44, and the output unit 45 is omitted because they are the same as the description of each unit in the administrator 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 (a processor provided in the search support system 1).

[0027] FIG. 5 is a block diagram showing functions realized by the server device 10 (controller 11), the manager terminal 20 (controller 21), the managed person terminal 30 (controller 31), and the job seeker terminal 40 (controller 41).

[0028] As shown in FIG. 5A, the server device 10 (control unit 11) includes a basic display control unit 111, a reception unit 112, a search category presentation unit 113, a search unit 114, and an artificial intelligence unit 120. As shown in FIG. 5B, the administrator terminal 20 (control unit 21) includes a display unit 211 and an operation reception unit 212. As shown in FIG. 5C, the managed person terminal 30 (control unit 31) includes a display unit 311 and an operation reception unit 312. As shown in FIG. 5D, the job seeker terminal 40 (control unit 41) includes a display unit 411 and an operation reception unit 412.

[0029] <Basic display control unit 111> The basic display control unit 111 is configured to display various information on the manager terminal 20, the managed person terminal 30, or the job seeker terminal 40. For example, the basic display control unit 111 displays registration information of human resources registered in the database on the display unit 211 of the manager terminal 20, the display unit 311 of the managed person terminal 30, or the display unit 411 of the job seeker terminal 40, in response to a request from each user (managers U1, U2, managed persons U3, U4, or job seekers U5, U6).

[0030] A "person subject to management" is someone who works in an organization, such as an employee of the organization (full-time employee, temporary employee, fixed-term employee, etc.). Organizations include for-profit corporations (such as companies), non-profit corporations (such as cooperatives and incorporated foundations), and public corporations (such as local governments). A "manager" includes a person on the reporting line of a person subject to management (such as a superior, line manager, etc.), a human resources manager, etc. Note that a "person subject to management" in the search support system 1 can also be the "manager" of another person subject to management.

[0031] The databases handled by the search support system 1 include an organization database and a job seeker database. The organization database is accessible only to those involved with the organization and includes an employee database in which data is not made public or provided to anyone outside the organization. The organization database registers administrators and people to be managed.

[0032] Information about the managed persons registered in the organization database (hereinafter, "managed person registration information") includes, for example, the name, gender, age, employee code, contact information, affiliation, employment status, work location, job duties, skills, qualifications, educational background, work history, job types, job titles, organizational goals set by the managed person, managerial evaluations of those goals, information about the managed person before joining the organization (hiring information), the managed person's condition, engagement (e.g., eNPS), aptitude or personality, interview history of the managed person (interviewee, comments, etc.), and attendance information of the managed person (overtime hours, number of annual leave taken, etc.). Hiring information includes the position offered at the time of hiring, annual salary, hiring interviewer, hiring evaluation, aptitude test results, reference check information, etc. Information other than hiring information is information after joining the organization (post-hiring information).

[0033] The job seeker database is different from the organizational database within the organization and is a database that is open to employers who are users of the recruitment service, and is typically provided outside the organization. The job seeker database contains job seekers outside the organization.

[0034] Information on job seekers registered in the job seeker database (hereinafter referred to as "job seeker registration 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 informs the employer of their work history, experience, skills, qualifications, etc., related to their past work.

[0035] <Reception Department 112> The reception unit 112 is configured to receive input from the manager terminal 20, the managed person terminal 30, or the job seeker terminal 40. Specifically, the reception unit 112 receives a search query written in natural language from the manager terminal 20. The search query includes at least one word or at least one sentence. The search query may include proper nouns such as the name of a human resource (first and last name), the name of an organization, the name of a place, or the name of a product. The reception unit 112 may also receive a search query from the managed person terminal 30 of an employee (managed person) who is not a manager.

[0036] 6 is a diagram showing an example of a search screen RD displayed on the administrator terminal 20 or the managed user terminal 30. The search screen RD includes a search query input field IF, a search execution button B11, a check box CB, a search mode display area MF, and a search result display area RF.

[0037] The search query input field IF is a field into which a search query is input by the administrator terminal 20. When the search execution button B11 is operated with a word or sentence input into the search query input field IF, the reception unit 112 receives the word or sentence input into the search query input field IF as a search query.

[0038] The search mode display area MF is an area where a search mode and a search keyword, which will be described later, are displayed. The search result display area RF is an area where the search results by the search unit 114, which will be described later, are displayed.

[0039] <Search category presentation unit 113> The search category presentation unit 113 is configured to present, based on the category determination reference information, at least one recommended search category for the search query received by the reception unit 112, from among the categories of the registration information of the person to be searched. The "person to be searched" is a person to be managed and / or a job seeker, and the "registration information of the person to be searched" is registration information of a person to be managed and / or registration information of a job seeker.

[0040] The "category of registered information" is the search target range in the registered information that is referenced in a search by the search unit 114. Examples of categories of registered information include "educational background," "work history," "qualifications," "job type," "position," and "skills." The "educational background" category includes information such as the names of universities, graduate schools, vocational schools, high schools, faculties, departments, etc. that a person is enrolled in or has graduated from (completed). The "work history" category includes information such as the names of organizations that a person has enrolled in. The "qualifications" category includes information such as the names of qualifications held. The "job type" category includes information such as the names of job types that a person has held. The "position" category includes information such as the names of job positions that a person has held. The "skills" category includes information such as the names of skills a person possesses.

[0041] The categories of registered information may also include a "general" category, which searches all registered information. The "general" category may include information (information included in each category) included in categories other than "general" (in the example of FIG. 6, the categories of "educational background," "work history," "qualifications," "job type," "job title," and "skills"), as well as information not included in any of the categories other than "general" (i.e., information not classified into any of the categories of "educational background," "work history," "qualifications," "job type," "job title," and "skills"). The categories of registered information may be set in common for all registered information (managed person registered information and job seeker registered information), or may differ for each database (organization database or job seeker database) to be searched. In other words, there may be categories set only for managed person registered information and categories set only for job seeker registered information.

[0042] For example, as shown in FIG. 6 , when the sentence “person with experience in personnel system design” is input as a search query, the search category presentation unit 113 sets the “skills” category as a recommended search category using the category determination reference information. In the search mode display area MF on the search screen RD of FIG. 6 , multiple categories included in the registered information are displayed in parallel as category tags TG1. Among the multiple category tags TG1, the category tag TG1 of a category selected as a search target category (described later) is highlighted (displayed with a bold frame and shading in the example of FIG. 6 ) to indicate that it has been selected or presented relative to the category tags TG1 of other categories (non-selected categories). The search category presentation unit 113 may also present the category tag TG1 of an unregistered category not included in the registered information as a search mode option. When the user selects the category tag TG1 of an unregistered category as a search target category, the search unit 114 sets the category closest to the unregistered category among the categories included in the registered information as a search target category.

[0043] When the search execution button B11 is operated on the search screen RD, the search category presentation unit 113 displays the category tag TG1 of the recommended search category as the initially selected category tag TG1 in the search mode display area MF, thereby presenting the recommended search category to the user. Note that, although one recommended search category is presented in the example of FIG. 6, the search category presentation unit 113 may present multiple recommended search categories. In this case, multiple category tags TG1 are initially selected. Note that, when the "general" category is initially selected as the recommended search category, the category tag TG1 (search mode) does not necessarily have to be displayed.

[0044] The category determination reference information is information that includes a correlation between a search query and a recommended search category. The category determination reference information is an estimator that is constructed so as to be able to output at least one recommended search category using a search query received by the receiving unit 112 as input.

[0045] The category determination reference information may be a category determination model that has been trained to receive a search query as input and output recommended search categories. In this case, the search category presentation unit 113 inputs the search query to the category determination model and causes the category determination model to output recommended search categories. This allows accurate recommended search categories to be determined from the search query.

[0046] The category determination model is a learning model trained to receive a search query (natural language) as input and output a recommended search category. That is, the category determination model is a learning model trained using a search query and its corresponding recommended search categories as training data. The category determination model may also be a generative AI including a large-scale language model. In this case, the search category presentation unit 113 receives a search query as input, inputs a prompt including an instruction to output a recommended search category based on the search query to the category determination model, and causes the category determination model to output the recommended search category. The search category presentation unit 113 may also input a prompt into the category determination model, in which, for example, one or more search query samples and one or more corresponding recommended search category samples are inserted, as input and output samples, in addition to the instruction to output the recommended search category and the search query.

[0047] Furthermore, the search category presentation unit 113 may input the search query, the category names included in the registration information, and an instruction to output recommended search categories for the search query into a category determination model, which is a large-scale language model. A "category name" is, for example, a name (a label such as "skill") assigned to each of multiple categories included in the registration information. This allows the category determination model to determine recommended search categories without using samples of the content of each category (information included in each category), thereby reducing the amount of information input to the recommended search categories and reducing the processing load, while improving the accuracy of determining recommended search categories.

[0048] The search category presentation unit 113 may convert the search query received by the reception unit 112 into a revised search query suitable for searching registered information based on the reference information for query conversion, and may also present recommended search categories for the revised search query. The reference information for query conversion is information including a correlation between the search query and the revised search query. Like the original search query, the "revised search query" is composed of natural language and includes at least one word or at least one sentence. The revised search query may also include proper nouns such as the name of a human resource (full name), organization name, place name, or product name. This allows the search query entered by the user to be optimized, and then recommended search categories can be presented and a search for managed individuals and / or job seekers can be performed, thereby improving the accuracy of human resource searches.

[0049] Specifically, when the search query includes a talent name, the search category presentation unit 113 may use at least part of the registered information of the talent name as the modified search query. In this case, the query conversion reference information includes information for referencing the registered information of the talent name included in the search query. As a result, recommended search categories are presented and managed persons and / or job seekers are searched for based on the registered information of the talent name entered by the user, so that the user can efficiently search for the talent they are looking for while ensuring search accuracy.

[0050] FIG. 7 is a diagram showing another example of the search screen RD displayed on the administrator terminal 20 or the managed person terminal 30. FIG. 7 shows an example of the display of the search mode display area MF and the search result display area RF when a search query including a human resource name is entered in the search query input field IF. When the sentence "Is there anyone like Mr. / Ms....?" is entered as the search query, as in this example, the search category presentation unit 113 uses the query conversion reference information to replace the search query with the registered information of the managed person or job seeker corresponding to the human resource name included in the search query. In other words, the search category presentation unit 113 regards the entire information included in the registered information as the corrected search query. In this case, the query conversion reference information includes, for example, access information to an organization database or a job seeker database as information for referencing the registered information.

[0051] After generating a revised search query based on the talent name, the search category presentation unit 113 sets the "general" category as a recommended search category using the category determination reference information. That is, when the search query includes a talent name, the search category presentation unit 113 sets a category including all of the registered information as a recommended search category. However, even when the search query includes a talent name, if the search query includes words or sentences other than the talent name (hereinafter referred to as "non-talent name elements"), the search category presentation unit 113 may set a category other than "general" as a recommended search category based on the non-talent name elements and the category determination registration information. In this case, the revised search query is composed of information included in the recommended search categories (i.e., part of the registered information) from the registered information of the managed person or job seeker corresponding to the talent name, and the non-talent name elements included in the input search query.

[0052] The search category presentation unit 113 determines whether or not a search query contains a personnel name by using, for example, a learning model trained to be able to extract personnel names from natural language (search query), a table listing personnel names, etc. The search category presentation unit 113 may also detect a specific symbol ("@" in the example of FIG. 7) contained in the search query and determine that the word following the symbol is a personnel name.

[0053] The search category presentation unit 113 may input the search query into a query conversion model and cause the query conversion model to output a revised search query. The query conversion model is a learning model that has been trained to be able to input a search query and output a revised search query, and is included in the query conversion reference information. This allows the search query entered by the user to be revised to content suitable for the search, and then recommended search categories can be presented and a search for managed persons and / or job seekers can be performed, thereby increasing the degree of freedom in entering a search query by the user. As a result, the user can search for personnel by intuitively entering a search query.

[0054] The query conversion model is a learning model trained to receive a search query as input and output a revised search query. That is, the query conversion model is a learning model trained using a search query and a corresponding revised search query as training data. The query conversion model may also be a generative AI including a large-scale language model. In this case, the search category presentation unit 113 receives the search query as input, inputs a prompt including an instruction to output a revised search query based on the search query to the query conversion model, and causes the query conversion model to output the revised search query. The search category presentation unit 113 may also input a prompt into the query conversion model, in which, for example, one or more search query samples and one or more corresponding revised search query samples are inserted, as input and output samples, in addition to the instruction to generate and output the revised search query and the search query.

[0055] FIG. 8 is a diagram showing yet another example of the search screen RD displayed on the administrator terminal 20 or the managed person terminal 30. FIG. 8 shows an example of the display of the search mode display area MF and the search result display area RF when a search query entered in the search query input field IF is converted into a revised search query. As in this example, a search query such as "someone who is good at cooking" that is unlikely to be included in the managed person registration information or job seeker registration information (the expected value of matching content being present in the registration information is low) is converted into a revised search query such as "cooking" by the search category presentation unit 113. Note that, although "general" is presented as a recommended search category in the example of FIG. 8, "skills," for example, may also be presented as a recommended search category.

[0056] After generating the revised search query, the search category presentation unit 113 sets the "general" category as a recommended search category using the revised search query ("cooking") and the category determination reference information. That is, the search category presentation unit 113 presents at least one recommended search category for the revised search query from among the categories of the registered information of the search target person, based on the category determination reference information.

[0057] <Search section 114> The search unit 114 is configured to search for any managed person and / or job seeker in the organization database and / or job seeker database. Specifically, the search unit 114 searches for the search target person based on the relevance between the search query and information included in the search target category of the registered information. The search target category is either a recommended search category or a category selected by the user from the categories of the registered information. Furthermore, if the search category presentation unit 113 has generated a revised search query, the search unit 114 searches for the search target person based on the relevance between the revised search query and information included in the search target category.

[0058] The search unit 114 may execute, as an initial search, a search for the recommended search categories presented (determined) by the search category presentation unit 113, without receiving input from the user. That is, the search unit 114 may execute a search for the search target person based on the relevance between the search query or the revised search query and information included in the recommended search categories of the registered information, simultaneously with the presentation of the recommended search categories by the search category presentation unit 113, and display the search results of the search.

[0059] When the search unit 114 executes an initial search for recommended search categories, the search unit 114 receives an input of at least one search target category from the user after executing the initial search. Upon receiving the user's selection of a search target category, the search unit 114 executes a search for the search target category.

[0060] Furthermore, the search unit 114 does not necessarily have to perform an initial search for the recommended search categories. If an initial search is not performed, the search unit 114 receives a search target category input from the user after the search category presentation unit 113 presents the recommended search categories, and then performs a search for the search target category. That is, after the search category presentation unit 113 presents the recommended search categories, the search unit 114 may perform a search for the search target person based on the relevance between the search query or revised search query and information included in the selected search target category among the registered information, and display the search results for the search. In this case, if the user selects the recommended search category as the search target category, a search is performed for the recommended search category. On the other hand, if the user selects a category other than the recommended search category as the search target category, a search is not performed for the recommended search category.

[0061] When the search category is changed by the user's selection, the search unit 114 immediately displays the search results for the changed search target category. That is, the search unit 114 accepts the user's input of the search target category and updates the search results in real time. Note that the search unit 114 may execute a search in the search target category after changing the search target category, or may execute a search in all categories in the background when accepting a search query, and switch the search results in accordance with the change in the search target category.

[0062] Fig. 9 is a diagram showing an example of a change in the search screen RD of Fig. 7. When the user changes the search mode (search target category) from "general" to "work history" on the search screen RD of Fig. 7, the search mode display area MF and the search result display area RF are instantly updated. Specifically, the search keywords RK (details will be described later) displayed in the search mode display area MF and the search matched persons displayed in the search result display area RF change dynamically.

[0063] The search unit 114 searches the organization database as a search target and searches for managed individuals as search targets, thereby making it possible to search for and scout personnel suitable for positions that need to be filled within the organization.

[0064] The search unit 114 also searches the job seeker database as a search target, searching for job seekers as search targets, thereby enabling the search unit 114 to search for and scout personnel suitable for positions that need to be filled within the organization from outside the organization.

[0065] The search unit 114 may perform a search based on the search query or the modified search query in both the organization database and the external job seeker database. That is, the search unit 114 may simultaneously search for the target administrator and job seekers based on the search query or the modified search query. In this case, the search results include both the target administrator and job seekers. The search unit 114 may simultaneously present to the user search matches extracted from the managed individuals (i.e., in-organization candidates) and search matches extracted from the job seekers (i.e., outside-organization candidates). In this case, the search unit 114 may display labels to distinguish between the in-organization candidates and the outside-organization candidates, or may change the display style (color, size, font, etc.) between the in-organization candidates and the outside-organization candidates. For example, the search unit 114 may display the number of search matches for the outside-organization candidates on the administrator terminal 20 or the managed-person terminal 30 along with the search results for the in-organization candidates. Furthermore, the search unit 114 may display the number of search results for in-organization candidates on the administrator terminal 20 or the managed person terminal 30 along with the search results for candidates outside the organization, or may display only the number of search results for in-organization candidates and candidates outside the organization on the administrator terminal 20 or the managed person terminal 30.

[0066] The search unit 114 may search either the organization database or the job seeker database, whichever database the user selects. For example, the search unit 114 causes the administrator terminal 20 to display a selection screen including an object (e.g., a button) for searching the organization database (for searching for personnel within the organization) and an object for searching the job seeker database (for searching for personnel outside the organization). The search unit 114 receives a search instruction for the organization database or a search instruction for the job seeker database through an input operation on either object.

[0067] The search unit 114 may be configured to search only one of the organization database and the job seeker database. In other words, the search support system 1 may be a system that searches only the organization database, or a system that searches only the job seeker database.

[0068] The search unit 114 determines search subjects who match the input search query as search hits included in the search results, for example, based on the similarity between the search query or revised search query and the information included in the search target category of each search subject (managed subjects and job seekers). Specifically, the search unit 114 calculates a matching index that indicates the degree of match between the search query or revised search query and the information included in the search target category, and displays the search subjects extracted based on the size of the matching index as search hits. This makes it possible to extract managed subjects or job seekers who match the search query using an objective index.

[0069] The matching index is calculated as the similarity (e.g., cosine similarity) between the vector data of the search query or revised search query and the vector data of the information included in the search target category among the managed person registration information or job seeker registration information. The vector data is a vector value obtained by quantifying the search query or revised search query and the managed person registration information or job seeker registration information using known methods such as natural language processing using morphological analysis or encoding of categorical variables.

[0070] For example, if "skills" is the search target category, the search unit 114 compares the vector data of the search query or the revised search query with the vector data of information included in the "skills" category in the registration information of the person to be searched, and calculates a matching index. Also, for example, if "general" is the search target category and the search query includes a human resource name, the search unit 114 compares the overall vector data of the managed person registration information or job seeker registration information of the human resource name as the revised search query with the vector data of the overall registration information of the person to be searched, and calculates a matching index. Furthermore, if multiple search target categories (e.g., "qualifications" and "skills") are selected, the search unit 114 compares the vector data of the search query or the revised search query with the vector data of information that combines information included in each of the multiple search target categories in the registration information of the person to be searched (e.g., information included in the "qualifications" category and information included in the "skills" category).

[0071] Furthermore, the matching index may be calculated based on, for example, tags assigned to the managed persons or job seekers and weighted scores of the tags. For example, the search unit 114 extracts tag words from the search query or revised search query, and assigns the weighted scores set for the tags as a matching index to search subjects (managed persons and job seekers) who have been tagged with tags corresponding to the tag words. For search subjects who have been tagged with multiple tags corresponding to multiple tags included in the search query or revised search query, the sum of the weighted scores of the multiple tags is used as the matching index. The weighted scores of the tags are set in advance for each tag by a settings manager or the like of the search support system 1.

[0072] Furthermore, the search unit 114 may calculate a matching index for the search subject based on the number of tags that match tag words extracted from the search query or the revised search query. For example, if the number of tag words extracted from the search query or the revised search query is five, the matching index for the search subject to which tags that match four of these tags are attached is set to 80%.

[0073] Extraction of tag words from the search query or revised search query is performed, for example, by referring to a tag list that describes multiple tag words corresponding to tags assigned to the managed person registration information and job seeker registration information. Furthermore, when the revised search query includes at least a part of the managed person registration information or job seeker registration information (i.e., when a human resource name is entered as the search query), tag words are extracted from tags that have been assigned in advance to the managed person registration information or job seeker registration information.

[0074] Furthermore, the search unit 114 may input the registration information of the search target (managed person registration information or job seeker registration information) and the search query or the revised search query into the matching index calculation model, and cause the matching index calculation model to output a matching index. The matching index calculation model is a learning model that receives the registration information and the search query as input and is trained to output a matching index. In other words, the matching index calculation model is a learning model that is trained using the registration information, the search query, and the corresponding matching index data as training data. The matching index calculation model may also be a generative AI that includes a large-scale language model. In this case, the search unit 114 receives the registration information and the search query as input, and inputs a prompt including an instruction to create a matching index based on the registration information and the search query into the matching index calculation model, and causes the matching index calculation model to output the matching index. In addition, the search unit 114 may input a prompt into the matching index calculation model, in which, in addition to the matching index calculation and output instructions, the registered information, and the search query, as input and output samples, for example, one or more samples of registered information and search queries and one or more corresponding samples of matching indexes are inserted.

[0075] The search unit 114, for example, displays search subjects whose matching index for the search target category exceeds a predetermined threshold (for example, 50%) as search candidates on the administrator terminal 20 or the managed person terminal 30. Furthermore, the search unit 114 preferentially displays information on search candidates (for example, displays them at the top of the list) in descending order of the matching index for the search target category.

[0076] The search result information RI for each matched person is displayed in a list in the search result display area RF of the search screen RD shown in Figures 6, 7, and 8. The search result information RI includes basic information BI, explanatory text ET, matching information MI, and attribute information AT.

[0077] The search result information RI is a card-like object having a frame that separates another area (external area) from the information display area (internal area). The information display area of the search result information RI contains basic information BI, explanatory text ET, matching information MI, and attribute information AT. Information other than these may also be placed in the information display area.

[0078] The basic information BI includes the facial image, name, ID (employee number, etc.), affiliation, job title, etc. of the search matched person. The explanatory text ET includes text created based on the information contained in the search target category. Details of the explanatory text ET will be described later. The matching information MI includes a matching index MP of the search target category and basis keywords BK. The matching index MP is composed of, for example, a combination of the search target category name and a numerical value. The display format of the matching index MP is not limited to numbers, and may be a graphic such as a gauge, bar, or graph that indicates the size of the matching index. Details of the basis keywords BK will be described later.

[0079] In the attribute information AT, the attributes of the search matched person are displayed as attribute tags TG2. The attribute tags TG2 are tags included in the registration information of the search matched person (previously assigned to the search matched person), or tags assigned by the search unit 114 based on the registration information of the search matched person. The attribute tags TG2 are assigned according to words representing skills, qualifications, educational background, work history, experienced job types, experienced job titles, etc. included in the registration information. The attribute tags TG2 may be tags obtained by converting words included in the registration information into tags, or may be tags obtained by converting words or sentences included in the registration information into corresponding tag words. The relationship between words or sentences included in the registration information and tag words is set in advance in relationship definition information such as a table or tag list. In the attribute information AT, the attribute tags TG2 are displayed grouped by category of the registration information (categories selectable as search target categories), for example. Furthermore, the attribute tags TG2 of multiple categories may be displayed grouped. In the example of FIG. 6, the attribute tags TG2 of the "educational background" category and the attribute tags TG2 of the "work history" category are grouped.

[0080] Of the attribute tags TG2, the attribute tag TG2 related to the "skill" category can be assigned based on skill words registered in a pre-prepared skill list. This enables a search for talent based on standardized skills, improving the efficiency of talent searches. It also makes it easier to seamlessly search for talent using skills, regardless of whether they are inside or outside the organization. Multiple skill words are registered in the skill list. Skill words are generated by referring to, for example, an occupation dictionary, skill dictionary, or in-house terminology dictionary used in the job seeker database.

[0081] In addition, the search unit 114 displays detailed information (information included in the registration information) of each search result matched person on the administrator terminal 20 or the managed person terminal 30 by inputting an input operation into the information display area of the search result information RI of each search result matched person (specifically, text or icon (e.g., text such as name, department, facial image, etc.)).

[0082] Furthermore, the user can create and send a scouting document to any search match (person to be managed or job seeker) by selecting the search result information RI of that search match displayed in the search result display area RF.

[0083] The search unit 114 may adjust the information to be displayed as search results depending on the authority of the searcher (administrator or managed person). That is, the search unit 114 may limit the information of the searched person to be displayed on the administrator terminal 20 or the managed person terminal 30 depending on the satisfaction level of the searcher's viewing conditions. The viewing conditions are set, for example, depending on a combination of the type of information that can be displayed (restriction level) and the type of searcher (job title, position, or whether or not the searcher has individual authority).

[0084] The search unit 114 may further calculate a matching index between the search query or the revised search query and information included in a category other than the search target category. This allows search results based on the matching index to be displayed immediately when the user changes the search target category. Furthermore, the search unit 114 may display a matching index for each category as the search result. This allows the user to refer to the matching index for each category and switch the search target category.

[0085] FIG. 10 is a diagram showing another example of search result information RI displayed in the search result display area RF of the search screen RD of FIG. 6 or the like. The search result information RI of FIG. 10 further includes a plurality of additional matching indicators AMP and additional matching information AMI. The additional matching indicators AMP are matching indicators of categories other than the search target category. The additional matching indicators AMP are, for example, composed of a combination of a category name and a numerical value (or a graphic). In the example of FIG. 10, the additional matching indicator AMP of the "general" category is arranged above the matching indicator MP of the search target category (the "skills" category). Furthermore, the additional matching indicators AMP of categories other than "general" and "skills" are arranged below the matching information MI as additional matching information AMI. Note that the arrangement of the matching indicators MP and additional matching indicators AMP in FIG. 10 is just an example. For example, the matching indicator MP of the search target category may be arranged above other information, or among the additional matching indicators AMP, the additional matching indicators AMP of categories other than "general" may be arranged above the matching indicator MP.

[0086] The search unit 114 may extract keywords related to the search target category from the search query as search keywords and display the search keywords. Furthermore, when a revised search query is generated in the search category presentation unit 113, the search unit 114 may extract keywords related to the search target category from the revised search query as search keywords and display the search keywords. This allows the user to check what keywords are being used in the search, making it easier to revise the search query, etc.

[0087] The search unit 114 extracts, as search keywords, words that are identical or similar to words included in the search query or revised search query from among words included in the search target category. The similarity of words is determined, for example, by comparing features such as vector data, referring to a database in which synonyms, similar words, and spelling variations are registered, or by determining using a learning model (large-scale language model). The search unit 114 may also extract, as search keywords, tag words that are identical or similar to words included in the search query or revised search query from among tag words of tags that are pre-assigned to the registered information of the search target person or tags that are assigned by the search unit 114 (for example, attribute tag TG2 displayed in attribute information AT in FIG. 10, etc.).

[0088] After extracting search keywords from the search query or the revised search query, the search unit 114 may calculate a matching index for each category as the degree of match between the search keywords and information in each category in the registered information. The search unit 114 may also calculate a matching index for each category as the degree of match between the search query or the revised search query and information in each category in the registered information, and extract search keywords independently of the calculation of the matching index. If the search query includes a human resource name, the search unit 114 may use the human resource name as a search keyword.

[0089] As shown in FIG. 6, the search keyword RK is displayed as a tag-like object in the search mode display area MF of the search screen RD. Also, as shown in FIG. 7, when a search query includes a talent name, the talent name extracted from the search query is displayed as the search keyword RK together with, for example, a facial image FI of the talent name. The facial image FI is obtained, for example, from the registration information of a managed person or job seeker corresponding to the talent name. Also, as shown in FIG. 9, when a search mode (search target category) other than "general" is selected, in addition to the talent name similar to that shown in FIG. 7, words extracted from the revised search query (the registration information of a managed person or job seeker corresponding to the talent name) are further displayed as the search keyword RK. Note that the search unit 114 may display the search keyword RK when a category other than "general" is selected as the search target category, and may not display the search keyword RK when the "general" category is selected as the search target category. This prevents a user from mistaking the search keyword RK for a specific category when searching in the "general" category.

[0090] The search unit 114 may display, in the search results, explanatory text created based on descriptions related to the search query from among information included in the search target category. This allows the user to easily understand the relevance (matching parts) between the search query and the search matched person, as well as important points in the registration information of the search matched person.

[0091] The explanatory text includes a text newly created using information included in the search target category of the registered information. The explanatory text is a summary of the text included in the registered information, a text reconstructed from the text of the registered information using keywords included in the registered information, or a text newly constructed using keywords included in the registered information.

[0092] The creation of a summary of the text included in the registered information is performed by referencing summarization reference information that includes correlations between the registered information and explanatory text. For example, the summarization reference information may be a summary creation model, which is a learning model that takes the registered information as input and trains it to output explanatory text that summarizes the registered information.

[0093] The keywords used to create the explanatory text are words extracted from information included in the search target category of the registered information that are similar to or match the words included in the search query. The explanatory text may include multiple keywords extracted from the registered information. To extract the keywords, for example, a keyword extraction model may be used, which is a learning model that receives the registered information and the search query as input and is trained (by machine learning or fine tuning) to output keywords for creating the explanatory text.

[0094] As shown in FIG. 6 and other figures, the explanatory text created by the search unit 114 is displayed as an explanatory text ET in the search result information RI of the search match. The search unit 114 may display the explanatory text ET in the search result information RI of all search matched persons, or may display the explanatory text ET only in the search result information RI of search matched persons who meet certain conditions for adding the explanatory text. Examples of the conditions for adding the explanatory text include the size of the matching index (e.g., 80% or more, within the top 10, etc.). The conditions for adding the explanatory text may be changed depending on the size of the search target population (e.g., the number of managed individuals registered in the organization database, the number of job seekers registered in the job seeker database, etc.). For example, if the size of the population is equal to or less than a threshold, the explanatory text ET may be displayed for search matched persons whose matching index is within the top 10. If the size of the population exceeds the threshold, the explanatory text ET may be displayed for search matched persons whose matching index is within the top 30 to 50. This reduces the amount of information in the search result information RI of search matched persons with low matching indexes, thereby improving the efficiency with which users can check search results. Furthermore, the processing load on the control unit 11 for creating the explanatory text can be reduced.

[0095] The search unit 114 may create and display an explanatory text upon receiving an instruction from a user to create and display the explanatory text. For example, the reception unit 112 may receive an instruction to create the explanatory text along with a search query. The instruction to create the explanatory text is received, for example, by inputting an operation on the search execution button B11 while a check box CB shown in FIG. 6 or the like is checked. The check box CB is, for example, located near the search execution button B11 (on the right side in the example of FIG. 6).

[0096] When a revised search query is generated in the search category presentation unit 113, the search unit 114 may create an explanatory sentence based on a description related to the revised search query, instead of the search query.

[0097] The search unit 114 may display a job summary (career summary) as the explanatory text ET. A "job summary" is a text describing the work history of a management target or job seeker. The job summary, for example, chronologically describes the department, role or job type, job content, results or achievements, etc. The job summary may also include time-series data indicating changes over time in the attributes of the management target, such as personality, characteristics, and condition. The job summary may be created in advance based on information included in the registration information, or may be created by the search unit 114 when a search is performed. The search unit 114 may display both the explanatory text and the job summary as the explanatory text ET, or may display only one of the explanatory text and the job summary. The job summary may also be displayed in a separate area from the explanatory text.

[0098] FIG. 11 is a diagram showing yet another example of search result information RI displayed in the search result display area RF of the search screen RD of FIG. 6, etc. The search result information RI of FIG. 11 further includes a job summary display area SA. A job summary (career summary) is displayed in the job summary display area SA. The job summary display area SA is located, for example, above the matching information MI. In the example of FIG. 11, the matching index MP of the search target category is located outside the matching information MI (specifically, to the right of the basic information BI). Also, in the example of FIG. 11, an explanatory text as a commentary ET is located inside the matching information MI (specifically, below the basis keyword BK). Note that the search result information RI of FIG. 11 may also include additional matching index AMP, additional matching information AMI, attribute information AT, etc., included in the search result information RI of FIG. 10.

[0099] The search unit 114 may extract basis keywords that match or are similar to the search query from the information included in the search target category and display the basis keywords in the search result, thereby allowing the user to instantly grasp points that match the search query among the registered information of the searched person.

[0100] The search unit 114 extracts, for example, from among words included in the search target category, words that are the same as or similar to words included in the search query as basis keywords. Furthermore, the search unit 114 may extract, as basis keywords, tag words of tags that are pre-assigned to the registered information of the search target person or tags assigned by the search unit 114 (for example, attribute tag TG2 displayed in attribute information AT in FIG. 6, etc.) that correspond to words included in the search query. Furthermore, the search unit 114 may extract, as basis keywords, words that are the same as or similar to the search keywords from the registered information.

[0101] The search unit 114 may highlight the basis keywords according to the degree of relevance with the search query. This makes it easier for the user to grasp points that are particularly highly relevant to the registered information of the search matched person. "Relevance" is the degree of similarity with words included in the search query. For example, the search unit 114 highlights basis keywords that are the same as words or search keywords included in the search query. Examples of highlighting include coloring, bolding, enlarging the size, and individual display (displaying in a position or area separate from other basis keywords).

[0102] As shown in FIG. 6 and other figures, the basis keywords BK are displayed as tag-like objects as part of the matching information MI on the search screen RD. The basis keywords BK are arranged, for example, alongside the matching indicators MP. Furthermore, as shown in FIGS. 10 and 11, the basis keywords BK that are highly relevant to the search query ("Angular" and "Kotlin" in FIG. 10 and "Data Analysis" in FIG. 11) are highlighted (displayed with a bold frame and hatching in the examples of FIGS. 10 and 11).

[0103] Furthermore, the number of basis keywords BK and the number of basis keywords BK that are highly relevant to the search query (the same as the search keyword) among the basis keywords BK may be displayed. For example, in the example of the search result information RI in FIG. 11 , the matching information MI (e.g., above the basis keywords BK) may display the degree of match of the basis keywords BK together with the name of the search target category, such as "skill match 1 / 4." Note that the search unit 114 may display the basis keywords BK when a category other than "general" is selected as the search target category, and may not display the basis keywords BK when the "general" category is selected as the search target category. This prevents a user from having the illusion that a search is being performed in a specific category using the basis keywords BK when performing a search in the "general" category.

[0104] The search unit 114 may determine a matching index for the search subject based on the similarity between the basis keyword and the basis keyword. For example, the search unit 114 may set the matching index for the search subject that includes a basis keyword that matches the search keyword to 100%. Furthermore, when multiple search keywords are extracted from the search query, the search unit 114 may set the matching index for the search subject that includes multiple basis keywords that match all of the multiple search keywords to 100%.

[0105] The search unit 114 may accept a selection input for a basis keyword by the user and add the selected basis keyword to the search query. For example, the user adds the basis keyword as a search keyword by clicking the basis keyword, dragging and dropping the basis keyword into the search mode display area MF, etc. Furthermore, the search unit 114 performs a narrowed search of search targets based on the added search keyword.

[0106] When a revised search query is generated in the search category presentation unit 113, the search unit 114 may display basis keywords that match or are similar to the revised search query, instead of the search query.

[0107] <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 for each functional unit may be a common one, or may be prepared individually for each functional unit.

[0108] The artificial intelligence unit 120 is an AI (Artificial Intelligence) equipped with language 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 Recurrent Neural Networks (RNNs), and includes generative AI.

[0109] 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.

[0110] The artificial intelligence unit 120 has 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.

[0111] The artificial intelligence unit 120 includes, as its artificial intelligence, a general-purpose natural language processing learning model such as a large-scale language model (LLM) that has learned a huge amount of data. Such a general-purpose learning model includes a language model that can handle various tasks without fine tuning using one-shot learning, few-shot learning, etc. Furthermore, the general-purpose learning model can also handle various tasks using zero-shot learning. The artificial intelligence used in each functional unit of the control unit 11 may be a separate learning model, or may be a common general-purpose learning model.

[0112] The learning model included in the artificial intelligence unit 120 is capable of additional learning. For example, the artificial intelligence unit 120 learns whether the presented recommended search categories were used as search target categories (whether categories other than the recommended search categories were used as search target categories). In other words, the artificial intelligence unit 120 performs additional learning and fine-tuning using a training dataset in which the selection results of recommended search categories are labeled as feedback for the recommended search categories created by the learning model. This optimizes the position information output from the learning model.

[0113] <Display> The display unit 211 of the manager terminal 20, the display unit 311 of the managed person terminal 30, and the display unit 411 of the job seeker terminal 40 each display a screen indicated by the screen data transmitted from the server device 10.

[0114] <Operation reception section> The operation acceptance unit 212 of the manager terminal 20 accepts operations by a user (manager) who uses the manager terminal 20. The operation acceptance unit 312 of the managed person terminal 30 accepts operations by a user (managed person) who uses the managed person terminal 30. The operation acceptance unit 412 of the job seeker terminal 40 accepts operations by a user (job seeker) who uses the job seeker terminal 40.

[0115] 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.

[0116] This information processing method includes a receiving step, a search category presenting step, and a search step. In the receiving step, a search query written in natural language is received. In the search category presenting step, at least one recommended search category for the search query from among categories of the registered information of the search target person is presented based on category determination reference information. The category determination reference information includes a correlation between the search query and the recommended search categories. In the search step, a search is performed for the search target person based on the relevance between the search query and information included in the search target category of the registered information. The search target category is a recommended search category or a category selected by the user from among categories of the registered information.

[0117] 12 is an activity diagram showing the flow of information processing (personnel search processing) executed by the search support system 1. Below, the information processing will be explained along with each activity in this activity diagram.

[0118] The talent search process begins when an administrator inputs a search query from the administrator terminal 20 (activity A101). The server device 10 accepts the search query sent from the administrator terminal 20 (activity A102). Next, the server device 10 determines recommended search categories for the search query (activity A103).

[0119] After determining the recommended search categories, the server device 10 executes a search (calculation of a matching index) using a search query for each category included in the registration information of the search target person (activity A104). Note that the server device 10 may execute the determination of recommended search categories in activity A013 in parallel with or after the search for each category. After executing the search for each category, the server device 10 outputs the search results for the recommended search categories together with the recommended search categories to the administrator terminal 20 (activity A105). As a result, the recommended search categories and the search results for the recommended search categories are displayed as initial results on the administrator terminal 20 (activity A106).

[0120] If necessary, the administrator inputs a change to the search target category on the administrator terminal 20 (activity A107). Upon receiving the instruction to change the search target category, the server device 10 outputs the search results for the changed search target category to the administrator terminal 20 (activity A108). As a result, the search results for the changed search target category are displayed on the administrator terminal 20 (activity A109).

[0121] 4. Effect The operation of this embodiment can be summarized as follows: Since recommended search categories are presented for a search query entered in natural language, the user can refer to the recommended search categories and select a search target category as the search target range. This allows the user to efficiently search for talent.

[0122] 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.

[0123] 5.Other In the above embodiment, the server device 10 performs various storage and control operations. 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 configured external to the server device 10. In this case, the artificial intelligence unit 120, which is an external configuration, is configured to receive input from each functional unit of the server device 10 and return instructed output to the server device 10.

[0124] The category determination reference information used by the search category presentation unit 113 does not necessarily have to be a trained model. For example, the category determination reference information may be a table, function, simple algorithm, or the like that describes the correlation between features (vector data) extracted from a search query and each category in the registered information. This category determination reference information is constructed, for example, by extracting and recording words that are frequently used as search queries from the content described in each category of the registered information.

[0125] The aspect of this embodiment is not limited to the search support system 1, but may be a search support method or a program. The search support method includes steps executed by the search support system 1. The program causes a computer to execute the steps of the search support system 1.

[0126] It may be provided in the following manner.

[0127] (1) A search support system comprising a processor configured to execute the following steps: a receiving step receives a search query written in natural language; a search category presentation step presents at least one recommended search category for the search query from among categories of the registered information of the person to be searched based on reference information for category determination, wherein the reference information for category determination includes a correlation between the search query and the recommended search category; and a search step performs a search for the person to be searched based on the relevance between the search query and information included in the search target category of the registered information, wherein the search target category is the recommended search category or a category selected by the user from among the categories of the registered information.

[0128] (2) In the search support system described in (1) above, the reference information for category determination is a category determination model that has been trained to be able to input the search query and output the recommended search category, and in the search category presentation step, the search query is input to the category determination model and the category determination model is caused to output the recommended search category.

[0129] (3) In the search support system described in (2) above, the category determination model is a large-scale language model, and in the search category presentation step, the search query, a category name included in the registration information, and an instruction to output the recommended search category for the search query are input to the category determination model.

[0130] (4) In the search support system described in any one of (1) to (3) above, in the search step, a matching index indicating the degree of match between the search query and the information included in the search target category is calculated, and the search target person extracted based on the magnitude of the matching index is displayed as a search result.

[0131] (5) In the search support system described in (4) above, in the search step, the matching index between the search query and information included in a category other than the search target category is further calculated, and the matching index for each category is displayed as the search result.

[0132] (6) In the search support system described in (4) or (5) above, in the search step, a sentence created based on a description related to the search query from among the information included in the search target category is displayed in the search results.

[0133] (7) In the search support system described in any one of (4) to (6) above, in the search step, basis keywords that match or are similar to the search query are extracted from the information included in the search target category, and the basis keywords are displayed in the search results.

[0134] (8) In the search support system described in (7) above, in the searching step, the basis keywords are highlighted according to the degree of relevance to the search query.

[0135] (9) In the search support system described in any one of (1) to (8) above, in the search step, keywords related to the search target category are extracted as search keywords from the search query, and the search keywords are displayed.

[0136] (10) In the search support system described in any one of (1) to (9) above, in the search category presentation step, the search query received in the reception step is converted into a revised search query suitable for searching the registered information based on reference information for query conversion, and the recommended search category for the revised search query is presented, wherein the reference information for query conversion includes a correlation between the search query and the revised search query, and in the search step, a search for the search subject is performed based on the relevance between the revised search query and information included in the search target category.

[0137] (11) In the search support system described in (10) above, in the search category presentation step, if the search query includes a personnel name, at least a part of the registered information of the personnel name is used as the modified search query, and the reference information for query conversion includes information for referencing the registered information of the personnel name included in the search query.

[0138] (12) In the search support system described in (10) or (11) above, the reference information for query conversion includes a query conversion model that has been trained to be able to input the search query and output the revised search query, and in the search category presentation step, the search query is input to the query conversion model and the query conversion model is caused to output the revised search query.

[0139] (13) In the search support system described in any one of (10) to (12) above, in the search step, keywords related to the search target category are extracted as search keywords from the revised search query, and the search keywords are displayed.

[0140] (14) In the search support system described in any one of (1) to (13) above, in the search step, an organization database in which managed persons working in the organization are registered is searched for, and the managed persons are searched for as the search target.

[0141] (15) In the search support system described in any one of (1) to (14) above, in the search step, a job seeker database in which job seekers outside the organization are registered is searched for, and the job seeker is searched for as the search target.

[0142] (16) A search support method comprising the steps executed by the search support system according to any one of (1) to (15) above.

[0143] (17) A program that causes a computer to execute each step of the search support system described in any one of (1) to (15) above. Of course, this is not the case.

[0144] 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]

[0145] 1: Search support system 2: Communication line 10: Server device 11: Control section 12: Storage section 13: Communications Department 14: Communication bus 20: Administrator terminal 21: Control unit 22: Storage section 23: Communications Department 24: Input section 25: Output section 26: Communication bus 30: Managed device 31: Control unit 32: Storage section 33: Communications Department 34: Input section 35: Output section 36: Communication bus 40: Job seeker terminal 41: Control unit 42: Storage section 43: Communications Department 44: Input section 45: Output section 46: Communication bus 111: Basic display control section 112: Reception 113: Search category presentation section 114: Search section 120: Artificial Intelligence Department 211:Display section 212: Operation reception unit 311: Display section 312: Operation reception section 411: Display section 412: Operation reception section

Claims

1. It is a search support system, Equipped with a processor, The aforementioned processor is configured to perform the following steps: The registration step accepts search queries that include the name of the person in question. In a specific step, based on the aforementioned person's name, the registration information of the person to be managed who works for the organization is identified. Here, the registration information includes at least one of the following: age, affiliation, job duties, skills, qualifications, occupation, position, work history, and evaluation. In the search step, based on the registration information, the system searches for candidates within the organization as search targets from the organizational database where the managed person is registered. In the calculation step, a matching index is calculated that indicates the degree of agreement between the registration information and the registration information of the search target. Here, the matching index is calculated for each category of the registration information, and each category includes at least one of the following: work experience, job type, skills, and qualifications. In the display step, the search support system displays attribute information indicating the attributes of the search target and the matching indicators for each category as search results.

2. In the search support system described in claim 1, The search support system, in the display step, displays the search targets extracted based on the magnitude of the matching indicator as the search results.

3. In the search support system described in claim 2, In the display step, the search support system displays in the search results a text created based on the information related to the search query, which is included in the registration information of the person being searched.

4. In the search support system described in claim 2, A search support system that, in the search step, extracts supporting keywords that match or are similar to the search query from the information contained in the registration information of the person being searched, and in the display step, displays the supporting keywords in the search results.

5. In the search support system described in claim 4, The search support system, in the display step, highlights the relevant keywords according to their relevance to the search query.

6. In the search support system described in claim 1, The search step further includes searching a job seeker database containing registered job seekers from outside the organization, and then searching for those job seekers as search targets in the search support system.

7. In the search support system described in Claim 1, The aforementioned reception step further accepts input operations for an object used to search for personnel within the organization. The search step involves a search support system that performs a search for the target person in response to the input operation.

8. In the search support system described in Claim 1, In the calculation step described above, the matching index for the overall category covering all of the registered information is further calculated, A search support system that, in the display step, displays the matching indicator for the overall category as the search result.

9. In the search support system described in Claim 1, In the display step, the search results are displayed in association with a descriptive text summarizing the registration information of the search target and matching information including the matching indicator. Here, the explanatory text refers to a search support system that is output from a learning model by inputting the registration information of the person to be searched into the learning model.

10. A search support method, A search support method comprising each step performed by the search support system according to any one of claims 1 to 9.

11. It is a program, A program that causes a computer to perform each step of the search support system described in any one of claims 1 to 9.

Images