Information processing system, information processing apparatus, information processing method, and program

Abstract

To provide a technique for improving matching accuracy by selecting a job seeker suitable to employment conditions of a job-offering company in accordance with a request from the job-offering company.SOLUTION: An information processing apparatus includes a storage unit 13 and a control unit 11, the control unit being configured to: receive scouting e-mail distribution setting information from a terminal device; register the scouting e-mail distribution setting information on the storage unit; search the storage unit to read recruitment information on recruitment selected in the scouting e-mail distribution setting information; generate a prompt for generating a search query for job-seeker extraction which is suitable to the recruitment containing the recruitment information and text data related to the recruitment included in the scouting e-mail distribution setting information; transmit the prompt to a large language model; receive a search query generated based on the prompt on the large language model; search the storage unit using the search query; and specify a job seeker suitable to the recruitment.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a technology for improving matching accuracy by selecting job seekers suitable for job openings at a hiring company in response to a request from the hiring company. [Background technology]

[0002] As labor mobility intensifies, companies seeking to hire have traditionally placed recruitment advertisements in various formats, such as on their own websites and recruitment sites. Job seekers browse these recruitment sites, select jobs that match their desired conditions, and then apply for those jobs. Recently, various technologies have been proposed that select job seekers who meet the needs of a hiring company in response to a request from the hiring company, and then match the two parties together.

[0003] For example, Patent Document 1 discloses a job recruitment matching method having a job entry information sending step of receiving job entry information from a job seeking terminal device via a first interface function providing site that provides terminal devices with the function of sending and receiving information between registered users, and sending job entry related information based on the job entry information to a corresponding job recruitment terminal device corresponding to the job entry information; a job scout information storage step of generating job scout related information based on job scout information and storing it so that it can be sent to a corresponding job seeking terminal device corresponding to the job entry information; and a desire related information sending step of receiving desire information from the job seeking terminal device to which the job scout related information has been sent via the first interface function providing site, and sending the desire information to the corresponding job recruitment terminal device. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-173293 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology disclosed in Patent Document 1 does not disclose how to use a large-scale language model to generate a search query to search for job seekers who meet the needs of a company hiring, how to search for job seekers using the search query, and how to select job seekers who are suitable for the job requirements.

[0006] The present invention has been made in consideration of these problems, and its purpose is to provide a technology that improves the accuracy of matching by selecting job seekers who are suitable for the recruitment conditions of a recruiting company in response to a request from the recruiting company. [Means for solving the problem]

[0007] In order to solve the above problem, an information processing device according to a first aspect of the present invention comprises a memory unit that stores scout email distribution setting information, hiring company information, and job seeker information, and a control unit that receives scout email distribution setting information from a terminal device, registers the scout email distribution setting information in the memory unit, searches the memory unit to read out job information for a job selected in the scout email distribution setting information, generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, incorporating text data related to the job included in the scout email distribution setting information and the job information, sends the prompt to a large-scale language model, receives the search query generated based on the prompt in the large-scale language model, searches the memory unit using the search query, and identifies job seekers suitable for the job.

[0008] An information processing method according to a second aspect of the present invention includes a control unit that receives scout email distribution setting information from a terminal device, registers the scout email distribution setting information in a memory unit, searches the memory unit to read out job information for a job selected in the scout email distribution setting information, generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, incorporating text data related to the job included in the scout email distribution setting information and the job information, sends the prompt to a large-scale language model, receives the search query generated based on the prompt from the large-scale language model, searches the memory unit using the search query, and identifies job seekers suitable for the job.

[0009] An information processing system according to a third aspect of the present invention is an information processing system comprising a terminal device of a hiring company and an information processing device, wherein the terminal device of the hiring company comprises a sending unit that sends scout email distribution setting information to the information processing device, and the information processing device comprises a memory unit that stores the scout email distribution setting information, hiring company information, and job seeker information, and a control unit that receives the scout email distribution setting information from the terminal device of the hiring company, registers the scout email distribution setting information in the memory unit, searches the memory unit to read out job information for a job selected in the scout email distribution setting information, generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, the search query incorporating text data related to the job included in the scout email distribution setting information and the job information, sends the prompt to a large-scale language model, receives the search query generated by the large-scale language model based on the prompt, searches the memory unit using the search query, and identifies job seekers suitable for the job.

[0010] A program according to a fourth aspect of the present invention causes a computer having a memory unit that stores scout email distribution setting information, hiring company information, and job seeker information to function as a control unit that accepts scout email distribution setting information from a terminal device, registers the scout email distribution setting information in the memory unit, searches the memory unit to read out job information for a job selected in the scout email distribution setting information, generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, incorporating text data related to the job included in the scout email distribution setting information and the job information, sends the prompt to a large-scale language model, receives the search query generated based on the prompt in the large-scale language model, searches the memory unit using the search query, and identifies job seekers suitable for the job. [Effects of the Invention]

[0011] According to the present invention, it is possible to provide a technology that improves the accuracy of matching by selecting job seekers who suit the recruitment conditions of a recruiting company in response to a request from the recruiting company. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a configuration diagram of an information processing system according to one embodiment of the present invention. [Figure 2] FIG. 2 is a configuration diagram of the first server device of the system. [Figure 3] FIG. 3 is a diagram showing the configuration of a terminal device in the system. [Figure 4] FIG. 4 is a hardware configuration diagram of the first to third server devices of the system. [Figure 5] FIG. 5 is a flowchart showing a series of processing steps related to the scout mail distribution settings by the system. [Figure 6] FIG. 6 is a flowchart showing a series of processing steps involved in generating a search query for searching for job seekers by the system. [Figure 7]Figure 7 shows an example configuration of a prompt requesting the generation of a search query for a job seeker search. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0014] FIG. 1 shows the configuration of an information processing system according to one embodiment of the present invention.

[0015] As shown in the diagram, the information processing system is configured by connecting a first server device 1, a terminal device 2 of a hiring company, a terminal device 3 of a job seeker, a second server device, and a third server device 5 for free communication via a communication network 6 such as the Internet, either wirelessly or via a wired connection. Note that the second server device 4 may be an external device that is not included in the information processing system, and the third server device 5 is an optional component. Furthermore, various types of devices can be used as the terminal devices 2 and 3, such as smartphones, tablet devices, node-type personal computers, and desktop-type personal computers.

[0016] In this configuration, the first server device 1 provides a service of searching for and selecting job seekers who meet the recruitment requirements of a hiring company and matching the two. Conventionally, in searching for and selecting such job seekers, a skilled operator would generate a search query (including, for example, search SQL; the same applies hereinafter) on the first server device 1 and search for and select job seekers who meet the recruitment requirements using the search query. However, one of the features of this embodiment is that such search queries are generated using a large-scale language model and used for the search and selection.

[0017] More specifically, the first server device 1 receives registration information from hiring companies and job seekers in advance and manages this registration information within the system. When a request for sending a scout email is received from a hiring company, the first server device 1 registers scout email delivery setting information and generates a prompt for generating a search query to search for and select a job seeker suitable for the job based on comments (text data) related to the job request included in the scout email delivery setting information and the job data of the hiring company that it manages, and sends the prompt to the second server device 4.

[0018] The second server device 4 receives the prompt, generates a search query based on the prompt using a large-scale language model, and sends it to the first server device 1. It is assumed that the second server device 4 already understands the data structure of hiring companies and job seekers held by the first server device 1. In this way, the first server device 1 receives this search query, searches the registered information of job seekers that it has stored using the search query, selects and registers job seekers who suit the recruitment conditions.

[0019] If the second server device 4 is included as part of the information processing system, i.e., if it is a system owned by the operator, or if a closed relationship has been established with the information processing system in which the operator's data is not used for learning, the second server device 4 can also refer to big data related to matching between hiring companies and job seekers held by the third server device 5 when generating search queries based on a large-scale language model.

[0020] Furthermore, large-scale language models can refer to public data such as labor demographics, job availability ratios, occupational and industry statistics, employment trends, and regional job information when generating search queries.

[0021] According to the above process, job seeker search queries, which previously required time-consuming generation by experienced operators who are familiar with the company's internal systems, can now be generated easily, accurately, and in a short time based on a large-scale language model. This reduces labor costs, significantly improves work efficiency, and is expected to improve matching accuracy.

[0022] FIG. 2 shows the configuration of a first server device as an information processing device in the information processing system, and will be described.

[0023] As shown in the figure, a first server device 1 serving as an information processing device includes a control unit 11 that controls the entire device, and the control unit 11 is connected to a communication unit 12 and a storage unit 13.

[0024] The control unit 11 is composed of a CPU (Central Processing Unit), an MPU (Micro Processing Unit), etc. The communication unit 12 is realized, for example, by a network interface card (NIC), etc., and is connected to a communication network 6 such as the Internet via wired or wireless connection, and is a communication interface for communicating with the hiring company's terminal device 2 and the job seeker's terminal device 3, etc. The storage unit 13 is realized, for example, by a semiconductor memory element such as RAM (Random Access Memory) or flash memory, a hard disk drive (HDD), or an optical disk device, etc., and stores in advance the programs to be executed by the control unit 11. Furthermore, the storage unit 13 has at least a hiring company information storage unit 14, a job seeker information storage unit 15, a scout email distribution setting information storage unit 16, and a scout email storage unit 17.

[0025] More specifically, the hiring company information storage unit 14 stores various data constituting hiring company information, such as the hiring company's name, location, industry, company size, and other attribute information, linked to the hiring company ID, as well as the hiring ID. Furthermore, linked to this hiring ID, hiring data such as job content, job type, educational background, work location, salary, holidays and vacations, treatment and benefits, training system, images, and hiring process are stored as part of the hiring company information.

[0026] The job seeker information storage unit 15 stores various data that constitute job seeker information, such as the job seeker's name, age, sex, address, highest level of education, number of job changes, industry experience, most recent annual salary, desired job type, desired annual salary, and desired place of work, linked to the job seeker ID.

[0027] The scout mail delivery setting information storage unit 16 stores various data that constitutes scout mail delivery setting information, such as activation of delivery, selection of target job information (job ID), and other text data related to job-related requests, linked to a delivery ID. The scout mail storage unit 17 stores data such as the job ID and the generated mail body, linked to a mail ID.

[0028] By reading and executing the programs stored in the storage unit 13, the control unit 11 functions as a reception unit 11a, a job execution unit 11b, a search unit 11c, a prompt generation unit 11d, a transmission unit 11e, and a main control unit 11f.

[0029] The reception unit 11a receives, for example, a scout email distribution request from the terminal device 2 of a recruiting company. This request includes scout email distribution setting information. The scout email distribution setting information includes data such as enabling distribution, selecting the desired job offer (job offer ID), and a request (text data) related to the job offer. The job execution unit 11b controls overall processing such as searches by the search unit 11c (described later) and prompt generation by the prompt generation unit 11d (described later).

[0030] When selecting job seekers to whom the scout email is to be sent, the search unit 11c refers to the hiring company information storage unit 14 to search for and extract job data for the job openings selected at the time the scout email delivery request was made by the company that made the scout email delivery request, and when generating the scout email, the search unit 11c refers to the hiring company information storage unit 14 to extract employer data, and refers to the job seeker information storage unit 15 to search for and extract job seeker data.

[0031] When extracting job seekers to whom distribution is to be made, the prompt generation unit 11d generates a prompt that requests the generation of a search query for job seekers suitable for the job, incorporating the recruitment data of target companies searched and extracted by the search unit 11c. When generating a scout email, the prompt generation unit 11d generates a prompt that requests the generation of a scout email, incorporating the recruitment data and job seeker data searched and extracted by the search unit 11c. The prompt generation unit 11d then sends these prompts to the second server device 4, requesting the generation of the search query and scout email.

[0032] The second server device 4 is equipped with a large-scale language model (LLM) and uses the LLM to generate various types of prompts. Here, a LLM is a generation AI that learns from large amounts of text data to understand sentences as if they were human and generate various types of prompts in response to requests. For example, ChatGPT is based on a large-scale language model called GPT-3.5. When generating various types of prompts using the LLM, for example, if the second server device 4 is incorporated as part of the operator's system or if a contractual relationship has been established under which information such as prompts is not used for learning, the second server device 4 may refer to big data related to job searches held by the third server device 5. Here, big data includes various data related to job searches, such as job data, company data, job seeker data, application data, and behavioral history.

[0033] Furthermore, when generating search queries, large-scale language models can refer to publicly available data such as labor demographic information, job availability ratios, occupational and industry statistics, employment trends, and regional job information.

[0034] The sending unit 11e sends the scout email generated using the large-scale language model to the job seeker as a distribution destination. In addition, the main control unit 11f is in charge of overall control, such as generating display data such as HTML for screen display.

[0035] The configuration of the terminal device in the information processing system is shown and explained in Figure 3. Note that since the terminal device 2 of the hiring company and the terminal device 3 of the job seeker have the same configuration, the details will be explained here using the terminal device 2 of the hiring company as an example.

[0036] As shown in the figure, the terminal device 2 (or 3) has a control unit 21, a communication unit 22, an operation input unit 23, a display unit 24, and a storage unit 25. The units 21 to 25 are connected to each other so as to be able to communicate with each other via a system bus. The communication unit 22 is realized by, for example, an NIC or the like, and is a communication interface that is connected to a communication network 6 such as the Internet by wire or wirelessly and communicates with the first server device 1 and the like.

[0037] The operation input unit 23 is realized by, for example, a mouse, a keyboard, etc., and accepts various operation inputs from the user. The display unit 24 is realized by, for example, a liquid crystal display, etc., and displays various types of information. The operation input unit 23 and the display unit 24 may be integrated into a touch panel. The storage unit 25 is realized by, for example, a semiconductor memory element such as RAM or flash memory, a HDD, or an optical disk device, etc., and stores programs executed by the control unit 21. The control unit 11 may be configured by an integrated circuit such as an ASIC or FPGA.

[0038] In such a configuration, the control unit 21 is realized, for example, by a CPU or MPU, and functions as a transmission unit 21a, a reception unit 21b, a display control unit 21c, and a main control unit 21d by executing a program stored in the memory unit 25.

[0039] The sending unit 21a sends a scout email distribution request to the first server device 1. The receiving unit 21b receives data and content in HTML format or the like sent from the server device 1. If the terminal device is a job seeker's terminal device 3, the scout email is received. The display control unit 21c displays the content received by the receiving unit 21b on the display unit 24 so that it can be viewed. In addition, the main control unit 21d controls the entire terminal device.

[0040] Next, the hardware configuration of the server device will be described with reference to Fig. 4. Since the first server device 1, the second server device 4, and the third server device 5 have substantially the same configuration, the configuration of the first server device 1 will be exemplified and described here.

[0041] As shown in the figure, the first server device 1 includes a CPU 101, a RAM 102, a ROM 103, a HDD 104, a communication I / F 105, and an external I / F 106, which are communicatively connected via a system bus.

[0042] In this configuration, the CPU 101 controls the entire first server device 1. The RAM 102 is a volatile memory that provides a work area when a program is executed. The ROM 103 is a non-volatile memory that stores programs and various data. The HDD 104 is a non-volatile storage device that holds programs and various data.

[0043] The CPU 101 performs various functions as the control unit 11 that executes the above-mentioned prompt generation, etc., by reading and executing programs from the HDD 104, etc. In addition, the communication I / F 105 is an interface that realizes communication with the terminal devices 2, 3, etc., and an external device such as a monitor is connected to the external I / F 106.

[0044] A series of processes relating to the registration of scout mail distribution setting information will be described below with reference to the flowchart of FIG.

[0045] The sending unit 21a of the hiring company's terminal device 2 sends a scout email distribution request to the first server device (S1). Then, in the first server device 1, the receiving unit 11a receives the scout email distribution request from the hiring company's terminal device 2 (S2). The scout email distribution request includes scout email distribution setting information. The scout email distribution setting information includes various data such as enabling distribution, selection of the desired job offer (job offer ID), and a request related to the job offer (text data).

[0046] Next, in the first server device 1, the job executing unit 11b registers this scout mail distribution setting information in the scout mail distribution setting information storage unit 16 (S3). In this way, a series of processes relating to the registration of scout mail distribution setting information is completed.

[0047] Next, a series of processes related to generating search queries for job seekers to be distributed will be described with reference to the flowchart in Fig. 6. Here, Fig. 7 will also be referenced as appropriate.

[0048] When processing starts, the job execution unit 11b reads out the scout mail distribution setting information from the scout mail distribution setting information storage unit 16 (S11). Here, the scout mail distribution setting information includes various data such as enabling distribution, selecting the desired job offer (job offer ID), and a request (text data) related to the job offer.

[0049] Next, in order to select job seekers to whom the scout email distribution is to be sent, the search unit 11c refers to the hiring company information storage unit 14 to search for and extract job data for the jobs selected at the time of the scout email distribution request by the company that made the scout email distribution request (S12).

[0050] Next, the prompt generation unit 11d generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, incorporating the job data of the target companies searched and extracted by the search unit 11c and the request (text data) related to the job included in the scout email distribution information, in order to extract job seekers to whom the job should be distributed, and sends the prompt to the second server device 4 to request the generation of a search query (S13).

[0051] Here, the generation of a prompt requesting the generation of a search query is as shown in Fig. 7. As shown in the figure, the job data searched and extracted by the search unit 11c (job content, job type, educational background, work location, salary, holidays and vacations, treatment and benefits, educational system, images, recruitment process, etc.) and the request (text data) related to the job included in the scout email distribution setting information are inserted into the template "Please generate a search query for searching for job seekers that match the job data and job request."

[0052] The second server device 4 is equipped with a large-scale language model. Upon receiving a prompt, the second server device 4 generates a search query in response to the prompt using the large-scale language model and transmits the generated query to the first server device 1 (S14, S15). Here, a "large-scale language model" refers to a generation AI that learns from large amounts of text data to understand sentences as if they were human and generate various queries in response to requests. For example, ChatGPT is based on a large-scale language model called GPT-3.5. When generating a search query using a large-scale language model, for example, if the second server device 4 is incorporated as part of an operator's system or if a contractual relationship has been established under which information such as prompts is not used for learning, the second server device 4 may refer to big data related to job searches held by the third server device 5. This big data includes various data related to job searches, such as job data, company data, job seeker data, application data, and behavioral history.

[0053] In the first server device 4, the reception unit 11a receives the search query (S16), and under the control of the job execution unit 11b, the search unit 11c executes a search in the job seeker information storage unit 15 based on the search query (S17). Then, the extracted job seekers who are suitable for the job offer are registered in the scout email distribution setting information storage unit 16 (S18), thus completing the series of processes.

[0054] As described above, according to one embodiment of the present invention, in response to a scout email delivery request from a hiring company, a search query for extracting job seekers suitable for receiving a scout email is generated using a large-scale language model, and the search query makes it possible to easily and efficiently select job seekers suitable for the job offer, thereby improving work efficiency and also improving the accuracy of matching between hiring companies and job seekers.

[0055] Although one embodiment of the present invention has been described above, the present invention is not limited to this embodiment, and various improvements and modifications can be made without departing from the spirit and scope of the present invention.

[0056] For example, in the above-described embodiment, an example was shown in which a prompt for generating a search query for searching for job seekers was generated based on a job-related request (text data) and job information, but this is not limited to this, and other elements may also be incorporated into the prompt. [Explanation of symbols]

[0057] 1...first server device, 2...terminal device of hiring company, 3...terminal device of job seeker, 4...second server device, 5...third server device, 6...communication network, 11...control unit, 11a...reception unit, 11b...job execution unit, 11c...search unit, 11d...prompt generation unit, 11e...sending unit, 11f...main control unit, 12...communication unit, 13...memory unit, 14...hiring company information memory unit, 15...job seeker information memory unit, 16...scout email setting information memory unit, 17...scout email memory unit, 21...control unit, 21a...sending unit, 21b...receiving unit, 21c...display control unit, 21d...main control unit, 22...communication unit, 23...operation input unit, 24...display unit, 25...memory unit.

Claims

1. a storage unit for storing scout email distribution setting information, recruiting company information, and job seeker information; a control unit that receives scout mail distribution setting information from a terminal device, registers the scout mail distribution setting information in the storage unit, searches the storage unit to read out job information for a job selected in the scout mail distribution setting information, generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, incorporating text data related to the job included in the scout mail distribution setting information and the job information, transmits the prompt to a large-scale language model, receives the search query generated based on the prompt in the large-scale language model, searches the storage unit using the search query, and identifies job seekers suitable for the job. Information processing device.

2. A control unit receives scout mail distribution setting information from a terminal device, registers the scout mail distribution setting information in a storage unit, searches the storage unit to read out job information for a job selected in the scout mail distribution setting information, generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, incorporating text data related to the job included in the scout mail distribution setting information and the job information, transmits the prompt to a large-scale language model, receives the search query generated based on the prompt in the large-scale language model, searches the storage unit using the search query, and identifies job seekers suitable for the job. Information processing methods.

3. In an information processing system comprising a terminal device of a recruiting company and an information processing device, The terminal device of the recruiting company a transmission unit that transmits scout mail distribution setting information to the information processing device; The information processing device includes: a storage unit that stores the scout email distribution setting information, recruiting company information, and job seeker information; a control unit that receives scout email distribution setting information from a terminal device of the hiring company, registers the scout email distribution setting information in the storage unit, searches the storage unit to read out job information for a job selected in the scout email distribution setting information, generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, incorporating text data related to the job included in the scout email distribution setting information and the job information, transmits the prompt to a large-scale language model, receives the search query generated based on the prompt in the large-scale language model, searches the storage unit using the search query, and identifies job seekers suitable for the job. Information processing system.

4. A computer having a storage unit for storing scout email distribution setting information, recruiting company information, and job seeker information, The system functions as a control unit that accepts scout mail distribution setting information from a terminal device, registers the scout mail distribution setting information in the storage unit, searches the storage unit to read out job information for a job selected in the scout mail distribution setting information, generates a prompt requesting the generation of a search query for extracting job seekers suitable for the job, incorporating text data related to the job included in the scout mail distribution setting information and the job information, transmits the prompt to a large-scale language model, receives the search query generated based on the prompt in the large-scale language model, searches the storage unit using the search query, and identifies job seekers suitable for the job. program.

Images