Information processing systems, information processing methods, and programs

Abstract

This invention provides an information processing system, method, and program for efficiently searching for personnel. [Solution] The method accepts a search query input composed of natural language, extracts search keywords from the search query, and classifies the search keywords into search categories based on classification reference information. Here, the search category is the search scope in the registered information of the search target, and the classification reference information includes the correlation between the search keywords and the search category. The method also accepts the selection of a specific category from among the search categories, searches for the search target by comparing the information contained in the search category of the registered information with the search keywords contained in that search category, displays guidance information for the search target found, displays guidance information that includes the relationship between the search keywords contained in the specific category and the registered information, or performs a search with a broader information search scope than that applied to the search keywords contained in search categories other than the specific category.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a technique for performing a human resource search according to search conditions by referring to a database in which human resource information is registered.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] There is a need for a technique that can efficiently search for human resources.

[0005] In view of the above circumstances, the present invention aims to provide an information processing system and the like that can efficiently search for human resources.

Means for Solving the Problems

[0006] According to one aspect of the present invention, an information processing system is provided, comprising at least one processor, the processor configured to perform the following steps by reading a program, the query receiving step receiving input of a search query composed of natural language, the classification step extracting at least one search keyword from the search query and classifying the search keyword into a plurality of search categories based on classification reference information, where the search category is the search target range in the registered information of the search target, and the classification reference information includes the correlation between the search keyword and the search category, the search condition receiving step displaying the search keyword for each search category and accepting the selection of a specific category from the plurality of search categories, the search step performing a search for the search target by comparing the information included in the search category of the registered information with the search keyword included in the search category and displaying guidance information for the searched search target, the search step displays guidance information including correlation information showing the relationship between the search keyword included in the specific category and the registered information, or performs a search for the search target by expanding the information search range applied to the search keyword included in the specific category to the information search range applied to the search keyword included in the search category other than the specific category.

[0007] In this configuration, search keywords extracted from search queries entered as natural language are presented for each search category, and search results customized for a specific category are displayed. Therefore, by selecting the search category that the user wants to prioritize, they can search for personnel with a high degree of matching within that search category. This allows users to search for personnel efficiently. [Brief explanation of the drawing]

[0008] [Figure 1] This is a diagram showing the configuration of Information Processing System 1. [Figure 2] This is a block diagram showing the hardware configuration of server device 10. [Figure 3] This block diagram shows the hardware configuration of the administrator terminal 20 and the managed user terminal 30. [Figure 4] This block diagram shows the functions realized by the server device 10 (control unit 11), the administrator terminal 20 (control unit 21), and the managed user terminal 30 (control unit 31). [Figure 5] This figure shows an example of the search screen RD displayed on the administrator terminal 20. [Figure 6] This figure shows an example of the search condition setting area SA being expanded in the search screen RD of Figure 5. [Figure 7] This figure shows an example of the condition input screen CD displayed on the administrator terminal 20. [Figure 8] This figure shows an example of a condition input field CF in the filtering condition setting area NA. [Figure 9] This figure shows an example of the search screen RD when a different search query (a search query related to experienced organizations) is entered than in Figure 5. [Figure 10] This figure shows an example of the SD screen displaying detailed information of the search target, as shown on the administrator terminal 20. [Figure 11] This figure shows another example of the detailed information display screen SD. [Figure 12] This figure shows an example of the bookmark registration screen BD displayed on the administrator terminal 20. [Figure 13] This is an activity diagram showing an example of the flow of information processing (search process for managed persons) performed by Information Processing System 1. [Modes for carrying out the invention]

[0009] Embodiments of the present invention will be described below with reference to the drawings. The various features shown in the embodiments below can be combined with each other.

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

[0011] Furthermore, in various information processing according to one embodiment, an input and an output corresponding to the input can be realized. Here, as long as an output is obtained as a result of the input, the form of the information referenced in such information processing (hereinafter referred to as "reference information") is not limited. 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 equation constructed by a statistical method), or a pre-trained model that has learned the correlation between input and output in advance, or a large-scale language model that can output a desired result by inputting a prompt.

[0012] Furthermore, in one embodiment, "part" may include, for example, hardware resources implemented by a circuit in a broad sense, and the information processing of software that can be specifically realized by these hardware resources. Also, in one embodiment, various types of information are handled, and this information can be represented, for example, by the physical values ​​of signal values ​​representing voltage and current, the high or low values ​​of signal values ​​as a set of binary bits composed of 0s or 1s, or by quantum superposition (so-called qubits), and communication and calculations can be performed on a circuit in a broad sense.

[0013] Furthermore, a circuit in a broad sense is a circuit realized by appropriately combining at least a circuit, circuitry, a processor, a memory, etc. The processor may be a general-purpose processor or a dedicated circuit. That is, it includes an application-specific integrated circuit (ASIC), programmable logic devices (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 In this section, the hardware configuration will be described.

[0015] <Information Processing System 1> FIG. 1 is a configuration diagram showing an information processing system 1. The information processing system 1 includes a communication line 2, a server device 10, a plurality of administrator terminals 20, and a plurality of managed person terminals 30. The server device 10, the administrator terminals 20, and the managed person terminals 30 are configured to be able to communicate with each other through the communication line 2. The connections of the server device 10, the administrator terminals 20, and the managed person terminals 30 may be wired or wireless.

[0016] The information processing system 1 constitutes at least a part of a personnel system used by, for example, a plurality of administrators (the first administrator U1 and the second administrator U2) and a plurality of managed persons (the first managed person U3 and the second managed person U4). The information processing system 1 mainly performs information management of managed persons by administrators. In one embodiment, the information processing system 1 consists of one or more devices or components. Hereinafter, these components will be described.

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

[0018] <Control Unit 11> The control unit 11 performs processing and control of the overall operation related to the server device 10. The control unit 11 is, for example, a Central Processing Unit (CPU). The control unit 11 realizes various functions related to the server device 10 by reading predetermined programs stored in the memory unit 12. That is, information processing by software stored in the memory unit 12 is concretely 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 a single unit, and the server device 10 may have multiple control units 11 for each function. The server device 10 may also be composed of a combination of these.

[0019] <Storage section 12> The storage unit 12 stores various types of information as defined above. This can be done, for example, as a storage device such as a solid-state drive (SSD) that stores various programs related to the server device 10 executed by the control unit 11, or as memory such as random access memory (RAM) that stores temporarily necessary information (arguments, arrays, etc.) related to program calculations. The storage 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 preferably uses wired communication methods such as USB, IEEE1394, Thunderbolt®, and wired LAN network communication, but may also include wireless LAN network communication, mobile communication such as LTE / 5G, and Bluetooth® communication as needed. In other words, it is more preferable to implement it as a collection of these multiple communication methods. 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 on-premises or in a cloud environment. A cloud-based server device 10 may provide the above-mentioned functions and processing in the form of, for example, SaaS (Software as a Service) or cloud computing.

[0022] <Administrator terminal 20> Figure 3 is a block diagram showing the hardware configuration of the administrator terminal 20 and the managed person terminal 30. The administrator terminal 20 is an information processing terminal used by the administrator. An "administrator" is a person who works for an organization and manages managed persons who work for the same organization. Administrators include, for example, people on the reporting line (e.g., supervisors, line managers, etc.) and personnel managers. In information processing system 1, a "managed person" can also be the "administrator" of other managed persons.

[0023] Furthermore, "organizations" include for-profit corporations (e.g., companies), non-profit corporations (e.g., cooperatives, foundations, etc.), and public corporations (e.g., local governments, etc.).

[0024] As shown in Figure 3A, the administrator terminal 20 comprises 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, storage unit 22, communication unit 23, input unit 24, and output unit 25 are electrically connected within the administrator terminal 20 via the communication bus 26. The descriptions of the control unit 21, storage unit 22, and communication unit 23 are the same as those of the components in the server device 10 and are therefore omitted.

[0025] <Input section 24> The input unit 24 receives operation inputs made by the user. The operation inputs are transmitted as command signals to the control unit 21 via the communication bus 26. The control unit 21 can perform predetermined controls or calculations based on the transmitted command signals as needed. The input unit 24 may be included in the casing of the administrator terminal 20 or it may be an external component. 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, the input unit 24 can be a switch button, mouse, trackpad, QWERTY keyboard, etc.

[0026] <Output section 25> The output unit 25 displays a graphical user interface (GUI) screen that can be operated by the user. The output unit 25 may be included in the casing of the administrator terminal 20 or it may be an external device. Specifically, the output unit 25 can be implemented as a display device such as a CRT display, liquid crystal display, organic EL display, or plasma display. It is preferable that these display devices be used depending on the type of administrator terminal 20.

[0027] <Managed user terminal 30> The managed user terminal 30 is an information processing terminal used by the managed user. The "managed user" is a person who works for the organization, for example, an employee of the organization (regular employee, temporary employee, contract employee, etc.).

[0028] As shown in Figure 3B, the managed user terminal 30 comprises 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, storage unit 32, communication unit 33, input unit 34, and output unit 35 are electrically connected within the managed user terminal 30 via the communication bus 36. The descriptions of the control unit 31, storage unit 32, communication unit 33, input unit 34, and output unit 35 are the same as the descriptions of each part in the administrator terminal 20 and are therefore omitted.

[0029] 2. Functional Configuration This section describes the functional configuration of this embodiment. Information processing by software stored in the memory 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 provided by the information processing system 1).

[0030] Figure 4 is a block diagram showing the functions realized by the server device 10 (control unit 11), the administrator terminal 20 (control unit 21), and the managed user terminal 30 (control unit 31).

[0031] As shown in Figure 4A, the server device 10 (control unit 11) comprises a basic display control unit 111, a query reception unit 112, a classification unit 113, a search condition reception unit 114, a search unit 115, a registration reception unit 116, and an artificial intelligence unit 120. As shown in Figure 4B, the administrator terminal 20 (control unit 21) comprises a display unit 211 and an operation acquisition unit 212. As shown in Figure 4C, the managed user terminal 30 (control unit 31) comprises a display unit 311 and an operation acquisition unit 312.

[0032] <Basic display control unit 111> The basic display control unit 111 is configured to display various information on the administrator terminal 20 or the managed user terminal 30. For example, in response to requests from each user (administrators U1, U2 or managed users U3, U4), the basic display control unit 111 displays the registration information of managed users registered in the database on the display unit 211 of the administrator terminal 20 or the display unit 311 of the managed user terminal 30.

[0033] The registration information of managed individuals registered in the database includes, for example, the individual's name, gender, age, employee code, contact information, department, job title, employment type, work location, duties, skills, qualifications, educational background, work history, previous job titles, previous positions, goals set by the individual within the organization, managerial evaluations of those goals, information about the individual before joining the organization (recruitment information), the individual's condition, engagement (e.g., eNPS), suitability or personality, interview history (interviewer, content of statements, etc.), attendance information (overtime hours, number of annual holidays taken, etc.), and summary documents describing their career. Recruitment information includes the offered position, annual salary, interviewer, evaluation at the time of recruitment, aptitude test results, and reference check information. Information other than recruitment information is information obtained after joining the organization (post-recruitment information).

[0034] <Query Reception Unit 112> The query reception unit 112 is configured to receive search queries written in natural language from the administrator terminal 20 for searching for search targets (e.g., managed persons). A search query includes at least one word or at least one sentence. A search query may include proper nouns such as personnel names (names), organization names, department names, place names, skill names, job titles, and product names. The query reception unit 112 may also receive search queries from managed person terminals 30 of employees who are not administrators (managed persons). Search targets may include both administrators and managed persons, and typically, search targets may be employees belonging to an organization.

[0035] <Classification section 113> The classification unit 113 is configured to extract at least one search keyword from the search query received by the query reception unit 112, and to classify the search keyword into multiple search categories based on classification reference information.

[0036] The search category is the scope of the search in the registered information of the person being searched. In other words, the search category is the category from the registered information of the person being searched that the search unit 115 (described later) targets for the search keyword (compared with the search keyword). The search category may also be interpreted as the axis of the search.

[0037] Examples of categories for registered information include "Skills and Experience," "Qualifications," "Job Title," "Position," "Work History," "Educational Background," and combinations thereof. The "Skills and Experience" category includes information such as the skills possessed and the names of positions held. The "Qualifications" category includes information such as the names of qualifications held. The "Job Title" category includes information such as the names of job titles held. The "Position" category includes information such as the names of positions held. The "Work History" category includes information such as the names of organizations previously employed. The "Educational Background" category includes information such as the names of universities, graduate schools, vocational schools, high schools, faculties, and departments attended or graduated from.

[0038] The classification unit 113 extracts words included in the search query as search keywords, for example, by morphological analysis. The classification unit 113 may extract the words included in the search query as search keywords as they are, or it may extract from pre-registered master words the closest to the words included in the search query, a higher-level conceptualization of the words, a lower-level conceptualization of the words, etc., as search keywords.

[0039] Master words may include pre-registered keywords such as skill names, position names, qualification names, job titles, positions, products, and organization names, or they may include keywords extracted from words included in the registered information of the person being managed. Master words may also be tag words included in the registered information of the person being managed.

[0040] A tag word is a keyword represented by a tag assigned to registration information. Tags are labels that indicate the attributes of the managed person, and are assigned according to words that represent skills, qualifications, educational background, work history, job experience, job titles, etc., included in the registration information. A tag word may be a word included in the registration information, or it may be a word or sentence included in the registration information converted based on a table or tag list, etc. The relationship between the word or sentence included in the registration information and the tag word is set in advance in relation definition information such as a table or tag list. Tags are assigned to the registration information of managed persons by the administrator for purposes such as searching for and classifying managed persons. Tags may also be assigned automatically based on a tag list when the registration information is registered.

[0041] The classification unit 113 may replace (split) the words included in the search query into multiple keywords and extract these as search keywords. For example, the word "MGR or higher" for a job title may be replaced with "MGR," "Department Head," "Section Chief," "General Manager," etc., and these may be used as search keywords. The word splitting is performed, for example, by referring to a rank list prepared for each category (e.g., job title, job grade, qualifications, etc.).

[0042] The classification unit 113 may also input the search query into a keyword extraction model and have the keyword extraction model output the search keywords. The keyword extraction model is, for example, a learning model included in the artificial intelligence unit 120 that has been trained to take a search query as input and output search keywords. The keyword extraction model is, for example, a learning model that has been trained using data of search queries and their corresponding search keywords as training data.

[0043] The keyword extraction model may be a generative AI that includes a large-scale language model. In this case, the classification unit 113 takes a search query as input, inputs a prompt to the keyword extraction model that includes an instruction to extract search keywords from the search query, and causes the keyword extraction model to output the search keywords. The classification unit 113 may also generate a prompt that gives an instruction to the keyword extraction model to extract search keywords from the search query, and input this prompt to the keyword extraction model. In addition to the search keyword extraction and output instructions and the search query, the classification unit 113 may also input a prompt to the keyword extraction model that includes, for example, one or more sample search queries and one or more sample search keywords corresponding to them, as examples, samples, or training data of input and output pairs.

[0044] The classification reference information includes information that shows the correlation between search keywords and search categories. The classification reference information is stored, for example, in the memory unit 12. The classification reference information is, for example, an estimator constructed to take a search keyword as input and output the search category corresponding to that search keyword.

[0045] The classification reference information may include, for example, tables, functions, simple algorithms, etc., that show the correlation between search keywords and search categories. The correlations included in the classification reference information can be constructed, for example, by extracting and aggregating keywords included in the actual registration information for each category of the registration information.

[0046] The classification reference information may include a keyword classification model that has been trained to take search keywords as input and output search categories. In this case, the classification unit 113 inputs the search keywords into the keyword classification model and causes the keyword classification model to output search categories. The keyword classification model is, for example, a learning model included in the artificial intelligence unit 120 that has been trained to take search keywords as input and output search categories. The keyword classification model is, for example, a learning model that has been trained using data of search keywords and their corresponding search categories as training data.

[0047] The keyword classification model may be a generative AI that includes a large-scale language model. In this case, the classification unit 113 takes the search keyword as input, inputs a prompt to the keyword classification model that includes an instruction to determine the search category corresponding to the search keyword, and outputs the search category to the keyword classification model. The classification unit 113 may also generate a prompt that gives the keyword classification model an instruction to determine the search category corresponding to the search keyword, and input this prompt to the keyword classification model. In addition to the search category determination / output instruction and the search keyword, the classification unit 113 may also input a prompt to the keyword classification model that includes, for example, one or more sample search keywords and one or more sample search categories corresponding to them, as examples, samples, or training data of input and output pairs. In the keyword classification model, parameters calculated and tuned through learning constitute the correlation of the classification reference information.

[0048] The classification unit 113 may classify search keywords that it determines cannot be classified into a default category. For example, a skill or experience search category may be used as the default category. The "default category" is a search category that is pre-set to classify search keywords that cannot be classified into any of the search categories. The default category may also be a search category that is initially selected as a specific category in the search condition receiving unit 114, which will be described later. Furthermore, the classification unit 113 may insert an instruction in the prompt to be input to the keyword classification model to classify search keywords into the default category if they cannot be classified into any of the search categories, and may also specify a specific search category as the default category.

[0049] The classification unit 113 may display a list of extracted search keywords in the vicinity of the input receiving object where the query receiving unit 112 receives the input of the search query on the administrator terminal 20. This makes it easier for the administrator to see what search keywords are being used to perform the search (i.e., what search keywords have been extracted from the search query). "Nearby of the input receiving object" is, for example, directly below the input receiving object (the input field for the search query).

[0050] The classification unit 113 may display the search keyword along with its search category near the input receiving object. If search keywords corresponding to different search categories are extracted, the classification unit 113 may combine the search category and the corresponding search keyword for each search category and display them near the input receiving object.

[0051] Figure 5 shows an example of a search screen RD displayed on the administrator terminal 20. The search screen RD includes a query input area QA and a results display area RA.

[0052] The query input area QA displays the search query input field IF (an example of an input receiving object), the search execution button B11, the search condition expansion object DO, the search category SC, and the search keyword SK.

[0053] The search query input field IF accepts search queries from the administrator terminal 20. When a word or phrase is entered in the search query input field IF and the search execution button B11 is pressed, the search is performed by the search unit 115, which will be described later.

[0054] The search condition expansion object DO accepts instructions to expand the search condition setting area SA (Figure 6), which will be described later. Figure 5 shows the state in which the search condition setting area SA is not expanded (hidden).

[0055] The search category SC and search keyword SK are displayed in a single line directly below the search query input field IF. The search category SC is the name of the search category to which the search keyword SK extracted from the search query entered in the search query input field IF belongs. If search keyword SK belonging to multiple search categories SC is extracted, multiple search category groups combining one search category SC and the search keyword SK belonging to that search category SC will be displayed in the list.

[0056] The extraction of search keywords SK and classification into search categories SC are triggered when text is entered into the search query input field IF (acceptance of at least a portion of the search query). Therefore, the search category SC and search keywords SK are displayed below the search query input field IF in parallel with the administrator's input of the search query. When the search query is edited (added, changed, deleted, etc.) in the search query input field IF, the search category SC and search keywords SK are also updated immediately.

[0057] <Search Criteria Reception Unit 114> The search condition receiving unit 114 is configured to receive input of search conditions or display conditions for search results from the administrator terminal 20 via the search unit 115.

[0058] Specifically, the search condition receiving unit 114 displays the search keywords extracted by the classification unit 113 on the administrator terminal 20 for each search category classified by the classification unit 113, and also accepts the selection of a specific category from among multiple search categories from the administrator terminal 20.

[0059] A "specific category" is a search category in which search keywords belonging to that category are used with greater weight (preferred use) than search keywords belonging to other search categories in the search unit 115. The search condition receiving unit 114 may accept the selection of multiple specific categories.

[0060] The search condition receiving unit 114 displays, for example, multiple search categories and the search keywords included in each search category on the administrator terminal 20, and accepts the selection of one specific category from the administrator terminal 20. The search condition receiving unit 114 may display only the search categories to which the user belongs (at least one search keyword is classified) from among the multiple pre-prepared search categories, or it may display all search categories, including those to which the user belongs (no search keywords are classified). The search condition receiving unit 114 may also choose not to accept the selection of a specific category for search categories to which the user belongs.

[0061] The search condition receiving unit 114 may accept the selection of a specific category from the administrator terminal 20, with one of several search categories initially selected as the specific category. The initially selected search category is determined, for example, from among the search categories to which at least one search keyword belongs. The search condition receiving unit 114 may also determine the initially selected search category based on a predetermined priority order of search categories. For example, if the "Skills" search category has the highest priority, and the classification unit 113 has extracted search keywords belonging to the "Skills" search category, then the "Skills" search category will be initially selected as the specific category. In this case, if no search keywords belonging to the "Skills" search category have been extracted, then the search category with the next highest priority after "Skills" becomes a candidate for initial selection.

[0062] The search condition receiving unit 114 may accept requests from the administrator terminal 20 to delete or add search keywords for each search category. This allows the administrator to directly edit search keywords, improving the search efficiency of the managed users compared to having to re-enter the search query.

[0063] Search keywords can be added, for example, by entering text or selecting from keyword suggestions registered for each search category.

[0064] The search condition reception unit 114 may accept input of filtering keywords from the administrator terminal 20. The filtering keywords are used to narrow down the search target when the search unit 115 performs a search.

[0065] The search condition receiving unit 114 may accept text entered on the administrator terminal 20 as filtering keywords. Alternatively, the search condition receiving unit 114 may present candidate filtering keywords to the administrator terminal 20 and accept the selection of candidates on the administrator terminal 20 as input for filtering keywords.

[0066] In particular, the registration acceptance unit 116, described later, presents administrators with candidate filtering keywords, enabling them to search using filtering keywords prepared for each administrator (or organization). This improves the search efficiency of the managed users.

[0067] Figure 6 shows an example of the search condition setting area SA being expanded in the search screen RD of Figure 5. The search condition setting area SA is displayed below the query input area QA and above the result display area RA (see Figure 5) when an input operation is performed on the search condition expansion object DO in the query input area QA.

[0068] The search condition setting area SA includes the category setting area CA, the refinement condition setting area NA, and the search execution button B12. The category setting area CA is the area where users can select a specific category and edit search keywords for each search category. For each of the multiple search categories, the category setting area CA displays a keyword display field KF and a specific category selection object SO.

[0069] In the example shown in Figure 6, the category setting area CA displays six search categories: "Skills and Experience," "Qualifications Held," "Experienced Job Titles," "Experienced Positions," "Company Name," and "Educational Background / School Name." The keyword display field KF for each search category displays the search keywords SK extracted by the classification unit 113 or the search keywords SK added by the administrator.

[0070] Each search keyword SK in the keyword display field KF is assigned a deletion object EO. ​​When an input operation is performed on the deletion object EO, the search keyword SK is deleted from the keyword display field KF. Also, for example, selecting an internal area of ​​the keyword display field KF will display a keyword addition input field that accepts keyword input. Keywords entered in the keyword addition input field are added as search keywords SK. Note that even if the search keyword SK in the keyword display field KF is edited (deleted or added), the search keyword SK in the query input area QA will not be changed. Furthermore, in the search performed by the search unit 115 described later, the search is performed using the search keywords SK displayed in the keyword display field KF.

[0071] A specific category selection object (SO) accepts the selection of a specific category. A specific category selection object (SO) is provided for each search category, and the administrator terminal 20 accepts exclusive input, where only one specific category selection object (SO) for any given search category is selected. Selected specific category selection objects (SO) are displayed in a different form than other unselected specific category selection objects (SO). In the example in Figure 6, the selected specific category selection object (the "Skills and Experience" specific category selection object (SO) has an icon common to all specific category selection objects (SO), as well as information indicating that it is selected.

[0072] The filtering criteria setting area NA is the area where filtering criteria are set when searching for managed persons. The filtering criteria setting area NA displays the condition input field CF where filtering criteria are entered, and the call object CO which accepts the call to the condition input screen.

[0073] In the example in Figure 6, the corresponding condition input fields CF are displayed for the categories "Grade," "Work Location," and "Other Keywords." The condition input fields CF for "Grade" and "Work Location" are dropdown lists from which conditions are selected from a pre-configured list.

[0074] Furthermore, in the example in Figure 6, the corresponding call object COs are displayed for the "Department" and "Company-Defined Skills" categories. Figure 7 shows an example of the condition input screen CD displayed on the administrator terminal 20. The condition input screen CD is displayed overlaid on the search screen RD (specifically, the search condition setting area SA) when an input operation is performed on the call object CO for "Company-Defined Skills" in Figure 6, for example. Note that "Company-Defined Skills" are items that are entered by selecting a specific option from multiple options that have been pre-set (registered) for each organization, and may also be "Department-Defined Skills" that have been pre-set for each department.

[0075] The condition input screen CD includes the first selection field SI1, the second selection field SI2, the skill addition object AO (see Figure 7B), the cancel button B21, and the select button B22.

[0076] As shown in Figure 7A, the first selection field SI1 displays a list of registered skill types (selection candidates) in a dropdown menu, and accepts the selection of a skill type from the administrator terminal 20. As shown in Figure 7B, the second selection field SI2 displays a list of registered skill levels (ranks) selected in the first selection field SI1 in a dropdown menu, and accepts the selection of a skill level from the administrator terminal 20. Note that multiple skill levels may be selected.

[0077] The Skill Addition Object AO accepts the addition of the first selection field SI1 and the second selection field SI2. When an input operation is performed on the Skill Addition Object AO, new first selection field SI1 and second selection field SI2 are added.

[0078] If an input operation is performed on the Cancel button B21, the input contents in the first selection field SI1 and the second selection field SI2 are discarded, and the condition input screen CD is closed. If an input operation is performed on the Select button B22, the input contents in the first selection field SI1 and the second selection field SI2 are accepted as filtering conditions for skills, and the condition input screen CD is closed. The accepted filtering conditions are displayed in the display area of ​​the corresponding category in the filtering condition setting area NA (for example, the "Company-defined skills" field).

[0079] Furthermore, a dropdown list-style condition input field (CF) may be displayed for "Department" and "Company-defined skills." Figure 8 shows an example of a condition input field (CF) in the filtering condition setting area (NA). In the example in Figure 8, a list of registered skills is displayed in the condition input field (CF) for the "Company-defined skills" category.

[0080] When an input operation is performed on the search execution button B12 in the search condition setting area SA shown in Figure 6, a search is executed by the search unit 115, which will be described later. In other words, the search execution button B12 has the same function as the search execution button B11 in the query input area QA.

[0081] <Search section 115> The search unit 115 is configured to search for registration information of the search target (managed person) registered in the database.

[0082] Specifically, the search unit 115 performs a search for the target person by comparing the information included in the search category of the registered information with the search keywords included in the search category, and displays the guidance information of the searched target person on the administrator terminal 20.

[0083] For example, if a search keyword (e.g., "SaaS") included in the "Skills" search category is extracted by the classification unit 113 (or entered by the administrator), the search unit 115 searches for the target user by determining whether the keyword corresponding to the search keyword is included in the "Skills" category (field) of the target user's registration information.

[0084] If a single search category contains multiple search keywords, the search unit 115 performs a search using these keywords as an OR condition, for example. However, the search unit 115 may also perform a search using multiple search keywords as an AND condition, or it may receive a request from the administrator terminal 20 to specify whether to use an OR condition or an AND condition.

[0085] If there are multiple search categories that contain the search keyword, the search unit 115 will, for example, perform a search using the group of search keyword contained in these search categories as an AND condition. However, the search unit 115 may also perform a search using the group of search keyword for each of the multiple search categories as an OR condition, or it may accept from the administrator terminal 20 whether to use an OR condition or an AND condition.

[0086] Furthermore, the search unit 115 performs a first specific process that displays guidance information including relevance information showing the relationship between search keywords included in a specific category and registered information, or a second specific process that performs a search for the target person by expanding the information search range applied to search keywords included in a specific category to a wider range applied to search keywords included in search categories other than the specific category.

[0087] The search unit 115 may perform only the first specific processing, only the second specific processing, or both the first and second specific processing. The "information search range" refers to the range of information that is determined to be extracted (matches the search keyword) by comparison with the search keyword.

[0088] The search unit 115 extracts search targets whose information in the search category of the registered information contains keywords that match or are similar to the search keyword. When the search unit 115 performs the second identification process, it may perform a search for search targets in the specified category using keywords that match or are similar to the search keyword as the search keyword information search scope, and in search categories other than the specified category, it may perform a search for search targets using only keywords that match the search keyword as the search keyword information search scope. This ensures that the search keywords in the specified category selected by the administrator are reflected preferentially in the search results, making it easier for the administrator to find the search targets (managed persons) they intend.

[0089] Similarity between keywords is determined by methods such as comparison using feature vectors, referencing databases containing synonyms, related words, and variations in spelling, and using learning models (large-scale language models).

[0090] The search unit 115 may search for search targets by determining similarity through a comparison of the features of the search keyword with the features of the keywords included in the registered information. This enables flexible searching of search targets, where the extraction target is not limited to searching by the search keyword itself. It also makes it easier to set the threshold for similarity determination of search keywords. For example, when the search unit 115 performs a second identification process (expanding the information search range for a specific category), the threshold for similarity determination (lower limit of similarity) for extracting search targets for search keywords in a specific category may be set lower than the threshold for similarity determination for search keywords in other search categories.

[0091] When an organization name, such as a company name, is used as a search keyword, the search unit 115 may determine the similarity between organization names based on the organization's attributes (e.g., industry, sales, number of employees, etc.). For example, the search unit 115 may perform the similarity determination using the following procedure. First, the search unit 115 pre-registers reference data in the database that defines the attributes of each organization (organizational attributes). Here, organizational attributes are information about an organization, and organizational information may include, for example, sales, profit, industry, capital, location, listing status, number of employees, year of establishment, etc. Next, the search unit 115 refers to the reference data and identifies the corresponding organizational attributes from the organization names contained in the registered information of the search target. Furthermore, the search unit 115 determines the similarity between the two by comparing the vector data obtained by vectorizing the organizational attributes of the search keyword and the organizational attributes of the registered information, respectively.

[0092] Similarity is defined, for example, by the distance between a first feature obtained by vectorizing search keywords and a second feature obtained by vectorizing keywords included in the registration information. Cosine similarity is used as a measure of similarity, and the closer the cosine similarity is to 1, the greater the similarity.

[0093] Vectorization is performed, for example, by quantification using known methods such as natural language processing using morphological analysis or encoding. Furthermore, the first and second features may be obtained by referring to a table in which features are defined for each keyword.

[0094] The search unit 115 may search for search targets by determining similarity through a comparison of the features of the search keyword with the features of the tag words included in the registered information. This makes it possible to search for search targets based on tags that have been previously assigned to the registered information. In this case, the similarity determination is the same as when comparing the search keyword with keywords other than tag words, as described above.

[0095] If the search condition receiving unit 114 has received refinement keywords, the search unit 115 may search for the target user by performing a first similarity judgment by comparing the feature quantities of the search keywords with the feature quantities of the keywords included in the registered information, and a second similarity judgment by comparing the refinement keywords with the keywords included in the registered information. This makes it possible to add direct refinement using refinement keywords to the highly flexible search based on feature quantities, thereby improving the search efficiency of the administrator.

[0096] The search unit 115, for example, determines whether the registered information containing keywords similar to the search keyword and that are included in the search category of the search keyword (first similarity determination) also contains the same or similar keywords as the refinement keyword (second similarity determination). For example, the search unit 115 may perform a search of registered information by vector matching of the search keyword and a search of registered information by keyword matching of the refinement keyword. The search unit 115 extracts the search targets of the registered information containing these keywords as the searched (hit) search targets. Furthermore, when the search unit 115 performs the first similarity determination and the second similarity determination, a second specific processing (expansion of the information search range for a specific category) may be performed in the first similarity determination. In addition, for keywords related to proper nouns such as organization names and school names among the search keywords, the search of registered information may be performed by keyword matching.

[0097] The search unit 115 displays the guidance information of the searched (extracted) individuals on the administrator terminal 20, for example, in a list format. The "guidance information" includes some of the registered information of the searched individuals (for example, name, employee number, department, job title, work location, etc.).

[0098] When the search unit 115 performs the first specific processing (display of relevance with a specific category), the guidance information for the search target includes relevance information that shows the relationship between the search keywords included in the specific category (hereinafter, "specific keywords") and the registered information. The "relevance information" includes, for example, the strength of the relationship (degree of relevance) between the specific keywords and the keywords included in the registered information, and the location (extraction location) where the specific keywords are described in the registered information.

[0099] The search unit 115 may display guidance information, including a matching score as relevant information, on the administrator terminal 20. The matching score represents the degree of agreement between the search keywords (specific keywords) included in a particular category and the registered information. This allows the administrator to quantitatively understand the suitability of the search target in a specific category designated by the administrator.

[0100] The search unit 115 may calculate a matching score as the similarity between the feature quantities of a specific keyword and the feature quantities of keywords included in the registered information. This allows for the output of a matching score as an objective indicator.

[0101] Features are, for example, vectorized keywords (vector data). The matching score is calculated, for example, as the similarity (e.g., cosine similarity) between the vector data of a specific keyword and the vector data of keywords included in the registration information. For example, if "skills" is a specific category, the search unit 115 compares the vector data of each search keyword included in the "skills" search category with the vector data of keywords included in the "skills" category in the registration information of the search target, and calculates the matching score.

[0102] If multiple specific keywords exist (i.e., if a specific category contains multiple search keywords), the search unit 115 may use, for example, the average, maximum, or minimum value of the matching scores calculated for each specific keyword as the matching score. Alternatively, the search unit 115 may calculate the matching score as the similarity between the feature quantities of the specific keywords and the feature quantities of the tag words included in the registration information.

[0103] Furthermore, the search unit 115 may calculate a matching score based on the cosine similarity between the average vector data obtained by averaging the vector data of all tag words included in the registration information of each target person, such as an employee, and the vector data of the search keyword (if multiple search keywords are included, the average of the vector data of multiple search keywords). The search unit 115 converts the cosine similarity into a matching score using, for example, a predetermined rule (e.g., a relational expression, a function, etc.).

[0104] Furthermore, the search unit 115 may extract tag words from the individual tag words included in the registration information of each search target, such as an employee, whose vector data similarity to the search keyword is above a predetermined value, and calculate a matching score based on the average vector data obtained by averaging these similar vector data for each search target, and the cosine similarity between this average vector data and the vector data of the search keyword. This makes it possible to suppress the decrease in the matching score due to tag words that are not related to the search keyword.

[0105] The search unit 115 may input specific keywords and the registered information of the search target into a matching score calculation model and have the matching score calculation model output a matching score. The matching score calculation model is a learning model included in the artificial intelligence unit 120 that has been trained to take specific keywords and registered information as input and output a matching score. For example, the matching score calculation model is a learning model that has been trained using pairs of specific keywords and registered information and the corresponding matching score data as training data.

[0106] The matching score calculation model may be a generative AI including a large-scale language model. In this case, the search unit 115 takes specific keywords and registered information as input, inputs a prompt to the matching score calculation model that includes an instruction to calculate a matching score from the specific keywords and registered information, and causes the matching score calculation model to output the matching score. The search unit 115 may also generate a prompt that gives the matching score calculation model an instruction to calculate a matching score from the specific keywords and registered information, and input this prompt to the matching score calculation model. In addition to the matching score calculation and output instruction and the specific keywords and registered information, the search unit 115 may also input a prompt to the matching score calculation model that includes, for example, one or more samples of specific keywords and registered information and one or more samples of corresponding matching scores as examples, samples, or training data of input and output pairs.

[0107] The matching score may include the degree of agreement between search keywords other than the specific keyword (search keywords included in search categories other than the specific category) and the registered information. In other words, the matching score may be an indicator that prioritizes showing the degree of agreement between the specific keyword and the registered information. In this case, the search unit 115 may calculate the matching score for each search target by summing the matching score (subscore) of the specific keyword and the subscores of search keywords other than the specific keyword, each weighted (for example, by multiplying by individual weighting coefficients). The weighting of the specific keyword is greater than the weighting of other search keywords.

[0108] The guidance information may include specific tag words that belong to a particular category among the tag words assigned to the registration information. In this case, the search unit 115 may highlight the specific tag words according to the degree of match with the search keyword. This allows the administrator to verify the appropriateness of the search target based on the tags assigned to the registration information.

[0109] "Specific tag words included in a specific category" refers to, for example, if the specific category is "skills," the tag words that represent tags assigned to the "skills" category (or that have the attribute of "skills") in the registration information.

[0110] The "degree of match with search keywords" is calculated, for example, by comparing features such as vector data, or by using a learning model (large-scale language model), similar to the matching score mentioned above.

[0111] Examples of "highlighting" include adding a box, coloring, bolding, increasing the size, and displaying it individually (displaying it in a different position or area from other tag words). The search unit 115 may, for example, highlight specific tag words whose degree of matching with the search keyword is above a predetermined threshold. The search unit 115 may also change the degree of emphasis on specific tag words according to the degree of matching with the search keyword. For example, a specific tag word that exactly matches the search keyword may be highlighted more than a specific tag word that is similar to the search keyword.

[0112] The guidance information may include an explanatory text that summarizes the registration information, created based on the registration information and explanatory reference information. In this case, the search unit 115 may highlight keywords included in the explanatory text according to the degree of match with the search query received by the query reception unit 112. This allows the administrator to efficiently grasp the information of the searched person and to confirm the basis of the search (the keywords that were hit).

[0113] The explanatory text includes, for example, a job summary (career summary) of the person being searched. A "job summary" is a document describing the job history of the person being searched. For example, a job summary may include information such as department, role or job title, job duties, results or achievements, etc. The job summary may also include descriptions of attributes such as the personality, characteristics, and condition of the person being managed.

[0114] The explanatory reference information is information that includes the correlation between the registered information and the explanatory text. The explanatory reference information is stored, for example, in the memory unit 12. The explanatory reference information is, for example, a text generator that is constructed to take registered information as input and output an explanatory text corresponding to that registered information.

[0115] The explanatory reference information may include, for example, a simple algorithm for extracting explanatory text from the registration information (summarizing the registration information). The correlations included in the explanatory reference information can be constructed, for example, by extracting and aggregating keywords used in the explanatory text from the actual registration information.

[0116] The explanatory reference information may include an explanatory text generation model that has been trained to take registered information as input and output explanatory text. In this case, the search unit 115 inputs the registered information into the explanatory text generation model and causes the explanatory text generation model to output explanatory text. The explanatory text generation model is a learning model included in the artificial intelligence unit 120 that has been trained to take registered information as input and output explanatory text. For example, the explanatory text generation model is a learning model that has been trained using registered information and data of corresponding explanatory texts as training data.

[0117] The explanatory text generation model may be a generative AI including a large-scale language model. In this case, the search unit 115 takes registered information as input, inputs a prompt to the explanatory text generation model that includes an instruction to create an explanatory text corresponding to the registered information, and causes the explanatory text generation model to output the explanatory text. The search unit 115 may also generate a prompt that gives the explanatory text generation model an instruction to create an explanatory text corresponding to the registered information, and input this prompt to the explanatory text generation model. In addition to the explanatory text creation / output instruction and registered information, the search unit 115 may also input a prompt to the keyword classification model that includes, for example, one or more samples of registered information and one or more samples of corresponding explanatory texts as examples, samples, or training data of input and output pairs. In the explanatory text generation model, parameters calculated and tuned through learning constitute the correlation of the explanatory reference information.

[0118] The search unit 115 may create an explanatory text that includes at least a specific keyword (a search keyword for a specific category) or a keyword similar to the specific keyword. For example, the search unit 115 may provide the explanatory text creation model, which is a large-scale language model, with a prompt that includes instructions to input the specific keyword in addition to the registration information, and to create an explanatory text that includes at least the specific keyword or a keyword similar to it.

[0119] The "degree of match with the search query" refers to the degree of match with the keywords included in the search query. For example, similar to the matching score mentioned above, it is calculated by comparing features such as vector data, or by using a learning model (large-scale language model).

[0120] Examples of "highlighting" in explanatory text include adding boxes, coloring, bolding, increasing the size, and displaying them individually (in a different location or area from the explanatory text). The search unit 115 may, for example, highlight keywords whose degree of matching with the keywords included in the search query is above a predetermined threshold. The search unit 115 may also change the degree of emphasis on keywords according to the degree of matching with the keywords included in the search query.

[0121] The search unit 115 may display a list of searched individuals on the administrator terminal 20, and may also display detailed information (more detailed than the guidance information) of the selected individual on the administrator terminal 20. The detailed information may be, for example, all or part of the registration information.

[0122] The search unit 115 may accept registration of the searched subject to the talent pool from the administrator terminal 20. The talent pool may also be called a bookmark list, favorites list, target list, etc., and is a collection (list) of managed persons (search subjects) such as employees registered in the database, in which the administrator has selected or extracted managed persons. The collection is created by the administrator, etc. Each collection is assigned a name, and in each collection, the name and its constituent members (registered managed persons) are stored in association. The constituent members of each collection may be, for example, candidates for transfer, candidates for promotion, candidates for a specific project, etc.

[0123] As shown in Figure 5, the search results from the search unit 115 are displayed, for example, in the results display area RA of the search screen RD. The results display area RA displays the condition input area TA, the overall checkbox CB1, the guidance information GI for the multiple search targets found, and the batch registration button B13.

[0124] The condition input area TA accepts input for filtering conditions of the guidance information GI displayed in the result display area RA. The condition input area TA may accept input for the same conditions as the filtering condition setting area NA in Figure 6. The overall checkbox CB1 is an object that accepts the bulk selection of the displayed guidance information GI (search target). When the overall checkbox CB1 is selected, all individual checkboxes CB2 of each guidance information GI described later are selected. In the example in Figure 5, the total number of search target individuals that have been searched is displayed next to the overall checkbox CB1. The bulk registration button B13 is an object that accepts the registration of selected search target individuals (individual checkboxes CB2 checked) to the talent pool. In the example in Figure 5, the number of selected search target individuals is displayed next to the bulk registration button B13.

[0125] Multiple information GIs are displayed in a list in a vertical direction. The information GIs are displayed, for example, in order of matching score MS. Each information GI displays the basic information of the search target (e.g., face image, name, employee number, department, position, work location, etc.). In addition, each information GI displays an individual checkbox CB2, an individual registration button B14, a detailed display button B15, matching information MI, and an explanatory text ET.

[0126] The individual checkbox CB2 is an object that accepts individual selections of the search target. The individual registration button B14 is an object that accepts individual registration of the search target to the talent pool. The detail display button B15 is an object that accepts instructions to display the search target's detailed information.

[0127] Matching Information (MI) includes matching score (MS), tag words (TG), etc. In the example in Figure 5, the matching score (MS) for a specific category ("Skills & Experience") and the tag words (TG) assigned to the registered information of the search target in that specific category are displayed as Matching Information (MI). Among the tag words (TG), tag words (TG) that match or are similar to the search keyword ("Product Manager") are highlighted compared to other tag words (TG).

[0128] The explanatory text (ET) includes a summary of the registration information. Keywords in the explanatory text (ET) that match or are similar to keywords in the search query (e.g., "product manager") are highlighted within the explanatory text (ET).

[0129] Figure 9 shows an example of the search screen RD when a different search query (a search query related to experienced organizations) is entered than in Figure 5. In this example, "□□ Company" included in the search query is extracted as the search keyword SK and displayed in the keyword display field KF of the search category "Name of current company". In addition, the specific category is initially set as "Name of current company". Furthermore, in the results display area RA, matching information MI is displayed in each guidance information GI with "Name of current company" as the specific category.

[0130] Figure 10 shows an example of the detailed information display screen SD of the search target displayed on the administrator terminal 20. The detailed information display screen SD is displayed overlaid on the search screen RD when, for example, an input operation is performed on the detailed information display button B15 of the guidance information GI displayed on the search screen RD.

[0131] The detailed information display screen SD includes the basic information display area BA, the information switching tab TB, and the information display area IA. The basic information display area BA displays the basic information of the search target and the individual registration button B14. The information switching tab TB is an object that accepts switching of the information displayed in the information display area IA. In Figure 10, the state in which "Career Summary" is selected in the information switching tab TB is shown, and the information display area IA displays the job summary and the category-specific tag words TG.

[0132] Figure 11 shows another example of the detailed information display screen SD. In Figure 11, the "Internal Career History" is selected in the information switching tab TB, and the information display area IA displays an overview of the work history, the job duties for each work history, and tag words TG related to acquired skills or experience, in chronological order.

[0133] Figure 12 shows an example of the bookmark registration screen BD displayed on the administrator terminal 20. The bookmark registration screen BD is displayed, for example, when an input operation is performed on the batch registration button B13 displayed on the search screen RD, or on the individual registration button B14 on the guidance information GI or detailed information display screen SD.

[0134] The bookmark registration screen BD includes a list input field LF. The list input field LF accepts the selection of a talent pool to which the selected search target will be registered. The talent pool to be registered may be selected from a pull-down menu or by text search. If an input operation is performed on the cancel button B31, the contents entered in the list input field LF will be discarded. If an input operation is performed on the register button B32, the selected search target will be registered in the talent pool entered in the list input field LF.

[0135] <Registration Reception Department 116> The registration reception unit 116 is configured to receive registrations of candidate filtering keywords from the administrator terminal 20, which are then received by the search condition reception unit 114. This allows administrators to pre-prepare arbitrary filtering keywords, for example, according to labels (skill tags, etc.) used within the organization, thereby improving the search efficiency for search targets. The candidates registered by the registration reception unit 116 are called up when the search condition reception unit 114 receives input for filtering keywords.

[0136] <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 each functional unit of the server device 10 may be common to all units, or it may be prepared individually for each functional unit.

[0137] 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, GPT-3, and GPT-4), BERT (Bidirectional Encoder Representations from Transformers), BART (Bidirectional and Auto-regressive Transformer), and language models such as recurrent neural networks (RNNs), and may include generative AI including large-scale language models. Large-scale language models are a type of generative AI and include models provided by services such as OpenAI's GPT, Google's Gemini, and Microsoft's Azure AI Studio. In addition, the artificial intelligence unit 120 can include any machine learning model, deep learning model, artificial intelligence model, etc.

[0138] The language model is an example of a learning model using a machine learning algorithm. Specific machine learning algorithms include nearest neighbors, naive Bayes, decision trees, support vector machines, and deep learning using neural networks. The artificial intelligence unit 120 can apply the above algorithms as appropriate.

[0139] 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 consists of pairs of input data and output data (correct answer data) for training. Furthermore, the language model may not only be one trained for a specific task, but also a general-purpose model that can be used universally for a wide range of tasks.

[0140] The artificial intelligence unit 120 may be a general-purpose natural language processing learning model, such as a Large Language Model (LLM), which has learned from a vast amount of data. An LLM is a learning model that has been pre-trained on a large amount of data consisting of text data, etc. (for example, (i) web content on the internet, or (ii) data stored in a predetermined database), and can perform various language processing tasks by being given a task. It can perform a wide range of natural language processing tasks, such as understanding sentence patterns and context, responding to questions, and generating sentences, according to the given prompts. Such a general-purpose learning model includes language models that can handle various tasks without fine-tuning using One-shot Learning or Few-shot Learning. Furthermore, the general-purpose learning model may also be configured to 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 it may be a common general-purpose learning model.

[0141] The learning models included in the artificial intelligence unit 120 (such as keyword extraction models and keyword classification models used in each functional unit) can undergo additional learning through transfer learning or fine-tuning. For example, the artificial intelligence unit 120 may perform additional learning and fine-tuning each time new data is registered, using it as new training data. This improves the accuracy of the information output from the learning models.

[0142] The learning model included in the artificial intelligence unit 120 may be a learning model (distilled model) obtained by knowledge distillation using the original learning model. In knowledge distillation, a pre-trained model, such as a large-scale language model, is used as the teacher model, and the parameters of the student model are adjusted so that the output loss of the student model (distilled model) relative to the output (Soft Target Loss) of the teacher model is small. The student model is then trained, and this student model becomes the distilled model. Alternatively, the student model may be trained so that the output loss of the student model relative to the correct labels (Hard Target) of the teacher data (combinations of input and output data of the learning model) is small. Compared to the original learning model (teacher model), the distilled model has similar performance to the original learning model, but with fewer parameters and a lower processing load. Therefore, using a distilled model can reduce the cost of the information processing system 1.

[0143] For example, the learning model used in each functional unit may be a distilled model that has been trained using combinations of input and output data from a large-scale language model as training data. Alternatively, when the information processing system 1 is introduced, a large-scale language model may be used as the learning model in each functional unit, and once training data from the large-scale language model has been accumulated, the distilled model obtained by knowledge distillation using that training data may be used as the learning model in each functional unit.

[0144] <Display> The display unit 211 of the administrator terminal 20 and the display unit 311 of the managed user terminal 30 each display the screen indicated by the screen data transmitted from the server device 10.

[0145] <Operation acquisition section> The operation acquisition unit 212 of the administrator terminal 20 accepts operations from the administrator using the administrator terminal 20. The operation acquisition unit 312 of the managed user terminal 30 accepts operations from the managed user using the managed user terminal 30.

[0146] 3. Information Processing Methods This section describes the information processing method of the server device 10. In this information processing method, each part of the server device 10 is executed by a computer as a step.

[0147] This information processing comprises a query reception step, a classification step, a search condition reception step, and a search step. The query reception step accepts input of a search query composed of natural language. The classification step extracts at least one search keyword from the search query and classifies the search keyword into multiple search categories based on classification reference information. The search condition reception step displays the search keyword for each search category and accepts the selection of a specific category from among the multiple search categories. The search step performs a search for the target person by comparing the information contained in the search category of the registered information with the search keyword contained in that search category, and displays the guidance information for the searched target person.

[0148] Figure 13 is an activity diagram showing an example of the flow of information processing (searching for managed persons) performed by the information processing system 1. The information processing will be explained below in accordance with each activity in this activity diagram.

[0149] The process of searching for managed individuals begins with the administrator entering a search query to find the managed individuals. The administrator enters the search query on the administrator terminal 20 (Activity A101). The server device 10 extracts search keywords from the entered search query (Activity A102). Subsequently, the server device 10 classifies the search keywords into search categories and outputs the results to the administrator terminal 20 (Activity A103). As a result, the search keywords are displayed on the administrator terminal 20 according to their search categories (Activity A104).

[0150] After the search keywords are displayed, the administrator selects a specific category on the administrator terminal 20 (Activity A105). After selecting the specific category, the administrator enters a command to execute the search on the administrator terminal 20 (Activity A106).

[0151] The server device 10 receives a search command from the administrator terminal 20 and performs a search for managed persons based on the search keyword and specific category (Activity A107). Subsequently, the server device 10 outputs guidance information for the found managed persons to the administrator terminal 20 (Activity A108). As a result, the guidance information for the managed persons is displayed on the administrator terminal 20 (Activity A109).

[0152] 4. Effect The operation of this embodiment can be summarized as follows: Search keywords extracted from search queries entered as natural language are presented for each search category, and search results customized for a specific category are displayed. Therefore, by selecting a search category that the user wants to prioritize, they can search for personnel with a high degree of matching in that search category. This allows users to search for personnel efficiently.

[0153] Although embodiments of the present invention have been described above, the present invention is not limited thereto and can be modified as appropriate without departing from the technical spirit of the invention.

[0154] 5. Others In the above embodiment, the server device 10 performed various storage and control functions, but instead of the server device 10, multiple external devices may be used. That is, various information and programs may be stored in a distributed manner across multiple external devices using blockchain technology or the like. In particular, the artificial intelligence unit 120 may be an external configuration of the server device 10. In that case, the external artificial intelligence unit 120 may be provided by, for example, an artificial intelligence service server, and is configured to receive input 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 an LLM. The artificial intelligence service server receives prompt input in the form of text, images, audio, etc., and generates and responds with answers to the prompts.

[0155] The search unit 115 may search for job seekers. In other words, the search unit 115 may use search keywords to search for registered information of job seekers. "Job seekers" include, for example, currently employed individuals (those seeking a career change) and prospective graduates (job seekers). The registered information of job seekers includes their resumes, work histories, and other profile information. A "resume" is a document that mainly describes the job seeker's profile, current situation, educational background, work history, and desired working conditions, while a "work history," also called a resume, is a document in which a job seeker communicates their past work experience, skills, qualifications, etc., to potential employers. The registered information of job seekers may also include conditions such as the industry and job type that the job seeker desires.

[0156] The embodiments of this model are not limited to the information processing system 1, but may also be an information processing method or a program. The information processing method comprises each step executed by the information processing system 1. The program causes a computer to execute each step of the information processing system 1.

[0157] The product may be provided in any of the following embodiments.

[0158] (1) An information processing system comprising at least one processor, the processor configured to perform the following steps by reading a program, the query receiving step receiving input of a search query composed of natural language, the classification step extracting at least one search keyword from the search query and classifying the search keyword into a plurality of search categories based on classification reference information, where the search category is the search target range in the registered information of the search target, the classification reference information includes the correlation between the search keyword and the search category, and the search condition receiving step displaying the search keyword for each of the search categories, and a plurality An information processing system that accepts the selection of a specific category from the aforementioned search categories, and in the search step, performs a search for the target person by comparing the information included in the search category of the registered information with the search keywords included in the search category, and displays guidance information for the searched target person, and in the search step, displays the guidance information including relation information showing the relationship between the search keywords included in the specific category and the registered information, or performs a search for the target person by expanding the information search range applied to the search keywords included in the specific category to the information search range applied to the search keywords included in the search categories other than the specific category.

[0159] (2) An information processing system as described in (1) above, wherein the search step displays the guidance information including the matching score as related information, wherein the matching score is the degree of agreement between the search keyword included in the specific category and the registered information.

[0160] (3) An information processing system as described in (2) above, wherein in the search step, the matching score is calculated as the similarity between the feature quantity of the search keyword and the feature quantity of the keyword included in the registered information.

[0161] (4) An information processing system according to any one of (1) to (3) above, wherein in the search step, the information processing system searches for the target person by making a similarity determination by comparing the feature quantity of the search keyword with the feature quantity of the keyword included in the registered information.

[0162] (5) In the information processing system described in (4) above, the search step searches for the target person by similarity determination by comparing the feature quantity of the search keyword with the feature quantity of the tag word included in the registered information, wherein the tag word is a keyword represented by a tag attached to the registered information.

[0163] (6) An information processing system as described in (4) or (5) above, wherein in the search condition acceptance step, the system accepts input of a narrowing keyword, and in the search step, it searches for the target person by performing a first similarity judgment by comparing the feature quantity of the search keyword with the feature quantity of the keyword included in the registered information, and a second similarity judgment by comparing the narrowing keyword with the keyword included in the registered information.

[0164] (7) An information processing system as described in (6) above, wherein the processor is configured to further perform the following steps: in the registration acceptance step, it accepts the registration of candidate refinement keywords; and in the search condition acceptance step, it presents the candidates and accepts the selection of the candidates as input of refinement keywords.

[0165] (8) An information processing system according to any one of (1) to (7) above, wherein in the search step, in the specific category, the search for the search subject is performed with keywords that match or are similar to the search keyword as the information search range for the search keyword, and in the search category other than the specific category, the search for the search subject is performed with only keywords that match the search keyword as the information search range for the search keyword.

[0166] (9) An information processing system according to any one of (1) to (8) above, wherein in the classification step, the information processing system displays a list of the search keywords near the object that accepts the input of the search query.

[0167] (10) An information processing system according to any one of (1) to (9) above, wherein in the search condition acceptance step, the system accepts the deletion or addition of the search keyword for each search category.

[0168] (11) An information processing system according to any one of (1) to (10) above, wherein the guidance information includes a specific tag word that is included in the specific category among the tag words assigned to the registration information, where the tag word is a keyword represented by the tag assigned to the registration information, and in the search step, the information processing system highlights the specific tag word according to the degree of match with the search keyword.

[0169] (12) An information processing system according to any one of (1) to (11) above, wherein the guidance information includes an explanatory text that summarizes the registration information, which is created based on the registration information and the explanatory reference information, wherein the explanatory reference information includes the correlation between the registration information and the explanatory text, and in the search step, the system highlights keywords included in the explanatory text according to the degree of match with the search query.

[0170] (13) An information processing method comprising each step performed by the information processing system described in any one of (1) to (12) above.

[0171] (14) A program that causes a computer to perform each step of the information processing system described in any one of (1) to (12) above. Of course, this is not always the case.

[0172] Finally, while various embodiments relating to this disclosure have been described, these are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be implemented in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of symbols]

[0173] 1: Information Processing System 2: Communication lines 10: Server device 11: Control Unit 12: Storage section 13: Communications Department 14: Communications bus 20: Administrator terminal 21: Control Unit 22: Storage section 23: Communications Department 24: Input section 25: Output section 26: Communications bus 30: Managed terminals 31: Control Unit 32: Storage section 33: Communications Department 34: Input section 35: Output section 36: Communications bus 111: Basic Display Control Unit 112: Query Reception Department 113: Classification section 114: Search Criteria Reception Department 115: Search section 116: Registration Department 120: Artificial Intelligence Department 211:Display section 212: Operation acquisition section 311: Display section 312: Operation acquisition section AI: Generation AO: Skill Addition Object B11: Search execution button B12: Search execution button B13: Batch registration button B14: Individual registration button B15: Show Details button B21: Cancel button B22: Select button B31: Cancel button B32: Register button BA: Basic information display area BD: Bookmark registration screen CA: Category setting area CB1: Overall checkbox CB2: Individual checkboxes CD: Condition Input Screen CF: Condition input field CO: Calling object DO: Search criteria expansion object EO: Delete object ET: Explanation GI: Information IA: Information display area IF: Search query input field KF: Keyword display field LF: List input field MI: Matching Information MS: Matching Score NA: Filtering condition setting area QA: Query input area RA:Result display area RD: Search screen SA: Search condition setting area SC: Search Category SD: Detailed information display screen SI1: First selection field SI2: Second selection field SK: Search Keywords SO: Select a specific category object TA: Condition input area TB: Information Switching Tab TG: Tag words

Claims

1. An information processing system, Equipped with at least one processor, The aforementioned processor is configured to perform the following steps by reading a program: In the query acceptance step, the input of a search query composed of natural language is accepted. In the classification step, at least one search keyword is extracted from the search query, and the search keyword is classified into multiple search categories based on classification reference information. Here, the search category is the search scope in the registered information of the search target, and the classification reference information includes the correlation between the search keyword and the search category. In the search criteria acceptance step, the search keywords are displayed for each search category, and the selection of a specific category from among multiple search categories is accepted. In the search step, the search for the target person is performed by comparing the information included in the search category of the registration information with the search keywords included in the search category, and the guidance information of the searched target person is displayed. An information processing system that, in the search step, displays the guidance information including relevance information showing the relationship between the search keywords included in the specific category and the registered information, or performs a search for the target person by broadening the information search range applied to the search keywords included in the specific category to a wider range applied to the search keywords included in the search categories other than the specific category.

2. In the information processing system described in claim 1, In the search step, the guidance information including the matching score as the relevance information is displayed. Herein, the matching score is the degree of agreement between the search keyword included in the specific category and the registered information in the information processing system.

3. In the information processing system described in claim 2, An information processing system that, in the search step, calculates the matching score as the degree of similarity between the feature quantities of the search keyword and the feature quantities of the keyword included in the registered information.

4. In the information processing system described in claim 1, The information processing system searches for the target person in the search step by determining similarity by comparing the feature quantities of the search keyword with the feature quantities of the keyword included in the registered information.

5. In the information processing system described in claim 4, In the search step, the search target is found by determining similarity by comparing the features of the search keyword with the features of the tag words included in the registration information. Here, the tag word is the keyword represented by the tag assigned to the registration information, and the information processing system.

6. In the information processing system described in claim 4, In the aforementioned search criteria acceptance step, the input of refinement keywords is accepted. The information processing system searches for the target person in the search step by performing a first similarity determination by comparing the feature quantity of the search keyword with the feature quantity of the keyword included in the registered information, and a second similarity determination by comparing the refinement keyword with the keyword included in the registered information.

7. In the information processing system described in claim 6, The aforementioned processor is configured to perform the following steps: In the registration acceptance step, we accept registration of candidate keywords for the aforementioned filtering process. An information processing system that, in the search condition acceptance step, presents the candidates and accepts the selection of the candidates as input for the narrowing keywords.

8. In the information processing system described in claim 1, An information processing system that, in the search step, performs a search for the target person in a specific category, using keywords that match or are similar to the search keyword as the information search scope for the search keyword, and in search categories other than the specific category, performs a search for the target person using only keywords that match the search keyword as the information search scope for the search keyword.

9. In the information processing system described in claim 1, In the classification step, the information processing system displays a list of the search keywords near the object that accepts the input of the search query.

10. In the information processing system described in claim 1, The information processing system in the search condition acceptance step accepts the deletion or addition of search keywords for each search category.

11. In the information processing system described in claim 1, The aforementioned guidance information includes, among the tag words assigned to the registration information, specific tag words included in the specific category, Here, the aforementioned tag word is a keyword represented by the tag assigned to the registration information. An information processing system that, in the search step, highlights the specific tag word according to the degree of match with the search keyword.

12. In the information processing system described in claim 1, The aforementioned guidance information includes an explanatory text that summarizes the aforementioned registration information, which is created based on the aforementioned registration information and explanatory reference information. Here, the explanatory reference information includes the correlation between the registration information and the explanatory text. An information processing system that, in the search step, highlights keywords contained in the explanatory text according to the degree of match with the search query.

13. Information processing method, An information processing method comprising each step performed by the information processing system according to any one of claims 1 to 12.

14. It is a program, A program for causing a computer to perform each step of the information processing system described in any one of claims 1 to 12.

Images