Matching device, method for matching, and matching program

Abstract

To provide a matching device which generates a job description including information other than superficial information, and performs matching between a recruiter and a job seeker.SOLUTION: The matching device includes: a first input unit for receiving an input of a first picture including information related to the responsibilities of a human resource that a recruiter is looking for; a first generation unit for analyzing the first picture and generating recruiting information including the atmosphere of the office of the recruiter; a second input unit for receiving input of a second picture including information related to a job seeker; a second generation unit for analyzing the second picture and generating job seeker information including the nature of the job seeker; a calculation unit for calculating the matching degree of recruiter information and job seeker information; and an output unit for outputting the matching degree.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a matching device, a matching method, and a matching program. [Background technology]

[0002] In recent years, with the wide variety of job-seeking websites available, job seekers are expected to effectively create resumes that better present themselves to employers. At the same time, employers are also expected to create accurate job information for the personnel they are looking for. Such job information is sometimes called a job description, as it is a document that specifies the job duties.

[0003] Therefore, Patent Document 1 discloses a technique for generating a job description. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] U.S. Patent No. 11,373,146 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the job description generation method shown in the above Patent Document 1 can generate a job description that indicates the necessary techniques and skills, but does not include other information, and more accurate job descriptions are desired. Also, the technology described in the above Patent Document 1 has the problem that it cannot create a resume for a job seeker.

[0006] Therefore, an object of the present invention is to provide a matching device, a matching method, and a matching program that can generate job seeker information for job seekers and job information for employers, and can introduce those who meet the conditions to each other. [Means for solving the problem]

[0007] A matching device according to one embodiment of the present invention includes a first input unit that accepts input of a first video including information related to the job of the personnel desired by an employer, a first generation unit that analyzes the first video to generate job information including the atmosphere of the employer's workplace, a second input unit that accepts input of a second video including information related to a job seeker, a second generation unit that analyzes the second video to generate job seeker information including the temperament of the job seeker, a calculation unit that calculates the degree of matching between the job information and the job seeker information, and an output unit that outputs the degree of matching.

[0008] In the above matching device, the first generation unit may generate job information by inputting the first video received by the first input unit into a first trained model that has learned the relationship between a video including information related to the job duties of the personnel desired by the employer and job information to be generated based on the video.

[0009] In the above matching device, the second generation unit may generate job seeker information by inputting the second video received by the second input unit into a second trained model that has learned the relationship between the video including information related to the job seeker and the job seeker information to be generated based on the video.

[0010] In the above matching device, the second trained model may further be a model that has learned the relationship between technical documents and job seeker information, and the second input unit may further accept input of documents that the job seeker has previously filled out.

[0011] Furthermore, a matching method according to one embodiment of the present invention includes a first input step in which a computer receives input of a first video including information related to the job of the personnel desired by the employer, a first generation step in which the computer analyzes the first video to generate job information including the atmosphere of the employer's workplace, a second input step in which the computer receives input of a second video including information related to the job seeker, a second generation step in which the computer analyzes the second video to generate job seeker information including the temperament of the job seeker, a calculation step in which the computer calculates the degree of matching between the job information and the job seeker information, and an output step in which the computer outputs the degree of matching.

[0012] In addition, a matching program according to one embodiment of the present invention provides a computer with a first input function for accepting input of a first video of an interview with an employer, a first generation function for analyzing the first video to generate job information including the atmosphere of the employer's workplace, a second input function for accepting input of a second video of an interview with a job seeker, a second generation function for analyzing the second video to generate job seeker information including the temperament of the job seeker, a calculation function for calculating the degree of matching between the job information and the job seeker information, and an output function for outputting the degree of matching. Effect of the Invention

[0013] According to the matching device, matching method, and matching program of the present disclosure, job descriptions for employers and job descriptions for job seekers can be automatically generated and matched, making it possible to introduce suitable jobs to job seekers and suitable human resources to employers. [Brief description of the drawings]

[0014] [Figure 1] FIG. 1 is a system diagram showing an example of the configuration of a matching system. [Diagram 2] FIG. 2 is a block diagram showing a configuration example of a matching device. [Diagram 3] 13 is a flowchart showing an example of a process for generating a recruiter description. [Figure 4]13 is a flowchart showing an example of a process for generating a recruiter description. [Diagram 5] 13 is a flowchart illustrating an example of a matching process. [Figure 6] FIG. 2 is a diagram showing an example of job information. [Figure 7] FIG. 2 is a diagram showing an example of job seeker information. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] A matching device, a matching method, and a matching program according to the present invention will be described in detail below with reference to the drawings.

[0016] <Embodiment Mode> <Summary> FIG. 1 is a system diagram showing an example of the configuration of a matching system.

[0017] As shown in FIG. 1, the matching system includes a matching device 100 and communicates with a recruiter terminal 200 and a job seeker terminal 300 via a network 400 .

[0018] The matching device 100 generates recruitment information related to the job of the human resources desired by the recruiter and job seeker information of the job seeker, and performs a matching process between the recruiter and the job seeker based on the generated recruitment information and job seeker information. The recruiter is a person or organization that employs, and the job seeker is a person who seeks a job and is employed. The matching device 100 acquires an interview video 20 conducted with the recruiter, and generates recruitment information about the recruiter. The matching device 100 also acquires an interview video 30 conducted with the job seeker, and generates job seeker information about the job seeker. The matching process may be a process of introducing one person to the other. More specifically, the matching process may be a process of introducing a job seeker to the recruiter, or a process of introducing a recruiter to the job seeker, and may be a process of introducing one person to the other when the conditions of the recruiter match the conditions of the job seeker, or when the conditions of the job seeker match the conditions of the recruiter, and preferably, the introduction may be performed when the conditions of both the recruiter and the job seeker match to a predetermined degree or more.

[0019] Job information is information that indicates the content and conditions of a job, and is information about the job of personnel desired by employers. Job information may include information such as the content of the job the job seeker will be doing, the skills and work history (experience) that the employer is looking for in the job seeker, the salary, vacation, and other conditions that the employer can offer to the job seeker, the qualifications required if the job seeker ends up working at the employer's workplace, and the atmosphere of the workplace. The qualifications of the job seeker included in the job information may also include information such as the personality and communication skills of the job seeker. The job seeker information may include not only the skills and qualifications possessed by the job seeker, and the salary, vacation, and other conditions the job seeker requests from the employer, but also information on the job seeker's personality, character, temperament, and the work environment the job seeker requests from the employer. The job seeker information may be a resume or curriculum vitae.

[0020] Traditional job information only had superficial information (for example, a rough description of the job duties, such as clerical or secretarial), and job seekers would often find that the job was different from what they had imagined, the atmosphere in the workplace was different from what they had expected, or the way of working did not suit their own way of working. The reverse was also true, and it was possible for the employer to find that the way of working was different from what they expected from the job seeker, or that the job seeker had not acquired the skills they were looking for. For this reason, it is preferable for job information and job seeker information to be as detailed as possible, but on the other hand, there is the problem that creating such information is cumbersome for both parties, and it may not be appropriate as an expression. There is also the problem that information cannot be expressed that the employer or job seeker is not aware of. In other words, there is a problem that the visualization of the type of person the employer is looking for, and the visualization of the type of person the job seeker is looking for and the type of job they are looking for are insufficient or not possible. In particular, Japanese people are said to be poor at expressing themselves, and there are many cases where there are many things that are not included in resumes. The matching device 100 according to this embodiment automatically creates job information for job recruiters and job seeker information for job seekers, and performs matching between job recruiters and job seekers.

[0021] The matching device 100 transmits job seeker information 11 indicating job seekers who meet a predetermined level of the conditions required by the recruiter to the recruiter terminal 200 via the network 400. The matching device 100 also transmits job seeker information 12 indicating recruiters who meet a predetermined level of the conditions required by the job seeker to the job seeker terminal 300 via the network 400. In this way, the matching device 100 can introduce job seekers who meet the conditions desired by the recruiter, or recruiters who meet the conditions desired by the job seeker.

[0022] The network 400 includes a wireless network and a wired network. Specifically, the network 400 may be, for example, a wireless LAN (WLAN), a wide area network (WAN), integrated service digital networks (ISDNs), wireless LANs, code division multiple access (CDMA), long term evolution (LTE), LTE-Advanced, 4th generation communication (4G), 5th generation communication (5G), 6th generation communication (6G) or later mobile communication system. The network 400 is not limited to these examples, and may be, for example, a public switched telephone network (PSTN), Bluetooth (Bluetooth (registered trademark)), an optical line, an asymmetric digital subscriber line (ADSL) line, a satellite communication network, or the like. The network 400 may also be a combination of these. Although one recruiter terminal 200 and one job seeker terminal 300 are shown in FIG. 1, it goes without saying that there may be a plurality of them. Furthermore, the matching device 100 may be realized by parallel computing using a plurality of computers. The details will be explained below.

[0023] <Configuration> Fig. 2 is a block diagram showing an example of the configuration of the matching device 100. As described above, the matching device 100 is a computer system that generates job information of a recruiter and job seeker information of a job seeker, and matches the recruiter and the job seeker. As shown in Fig. 2, the matching device 100 includes a communication unit 110, an input unit 120, an output unit 130, a storage unit 140, and a control unit 150.

[0024] The communication unit 110 is a communication interface capable of communicating with an external device. The communication unit 110 communicates with the recruiter terminal 200 and the job seeker terminal 300 via the network 400. The communication unit 110 receives the recruiter source information transmitted from the recruiter terminal 200 and transmits it to the control unit 150. The recruiter source information includes information about the job, and may include an interview video 20 as a first video including information about the job of the human resource desired by the recruiter. Note that the first video is not limited to the interview video 20, and may be any video that includes information about the job, such as an introductory video introducing the workplace of the recruiter or a video of an event held in relation to the job of the recruiter. The communication unit 110 also receives the job seeker source information transmitted from the job seeker terminal 300 and transmits it to the control unit 150. The job seeker source information includes information about the job search, and may include an interview video 30 as a second video including information about the job seeker, and may include technical documents described later. The second video is not limited to interview video 30, and may be any video containing information about the job seeker, such as the job seeker's skills, temperament, and personality, and may be, for example, a video of the job seeker introducing himself / herself, a video of the job seeker making some kind of presentation, or a video posted by the job seeker on a social networking site, etc. Based on an instruction from the control unit 150, the communication unit 110 transmits job information 12 transmitted from the control unit 150 to the job seeker terminal 300. Based on an instruction from the control unit 150, the communication unit 110 transmits job seeker information 11 transmitted from the control unit 150 to the recruiter terminal 200.

[0025] The input unit 120 is an input interface having a function of accepting input from an operator of the matching device 100 and transmitting it to the control unit 150. The input unit 120 may be realized by, for example, input devices such as a keyboard or a mouse, but is not limited to these. The input unit 120 may, for example, accept a connection of an external memory such as a flash memory, read out the recruiter source information or job seeker source information stored in the external memory, and transmit it to the control unit 150. The input unit 120 may also accept input of information related to the recruiter or job seeker via a keyboard or the like, and transmit the accepted input contents to the control unit 150.

[0026] The output unit 130 outputs specified information in accordance with instructions from the control unit 150. The output unit 130 may be realized by a monitor, a speaker, etc., or may be realized by transmitting information to an external device via the communication unit 110. The output unit 130 may output, for example, job information, job seeker information, and matching information generated by the control unit 150.

[0027] The storage unit 140 is a storage medium having a function of storing various programs and various data required for the operation of the matching device 100. The storage unit 140 can be realized, for example, by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, etc., but is not limited to these. The storage unit 140 may also be realized by cloud storage on a network. The storage unit 140 stores a first trained model 141 and a second trained model 142. The storage unit 140 may also store a program for the matching device 100 to generate job information, a program for generating job seeker information, a program for matching between the generated job information and job seeker information, etc.

[0028] The first trained model 141 is a learning model that has learned the relationship between the first video including information about the job of the human resource desired by the recruiter and the job information created based on the first video. Here, the information about the job may include information indicating the specific content of the job that the actual job seeker will perform, information indicating the atmosphere of the workplace, information about the job policy of the department to which the job seeker will be assigned, and the like. As a more specific example, the first trained model 141 may be a learning model that has learned the relationship between the interview video 20 showing an interview conducted with the recruiter about the desired human resource and the job information created based on the interview video 20. That is, it is a learning model that has learned, as teacher data, a video of an interview to inquire about the desired human resource (approaching a job seeker) and information annotated with the video that should be expressed as job information (information on the recruiter's skills, personality, various conditions such as salary and vacation that can be provided, and the work environment that can be provided). Various existing algorithms can be used for training the first trained model 141, and examples include, but are not limited to, support vector machines and neural networks. Therefore, the first trained model 141 is a model that estimates and generates job information of a recruiter by inputting a first video (e.g., an interview video of the recruiter regarding the desired human resource) containing information about the job of the human resource desired by the recruiter. Note that the recruiter is a person or organization that is recruiting human resources. The first trained model 141 may also be a model that has learned the relationship between the character string obtained by speech recognition of the first video (e.g., an interview video 20 of the recruiter) containing information about the job of the human resource desired by the recruiter and the job information. The first trained model 141 may also be a model that has learned information that associates the voice contained in the first video with the items (character strings) that should be included in the job information that can be derived from the voice, as training data.As an example, when the interview video 20 contains a voice saying "I want someone who can support everyone," information such as "a cooperative person" or "a behind-the-scenes helper" may be associated and used as teacher data. As another example, when the interview video 20 contains a voice saying "I want someone who can lead the team," information such as "a person with leadership skills" or "a person who demonstrates captaincy" may be associated and used as teacher data. As another example, when the interview video 20 contains a voice saying "It's a problem if someone is not punctual," information such as "a person who is punctual" may be associated and used as teacher data. Here, the character string associated as teacher data may be determined manually.

[0029] The second trained model 142 is a learning model that has learned the relationship between the second video including information about the job seeker and the job seeker information created based on the second video. Here, the second video including information about the job seeker may include information that allows the job seeker to demonstrate his or her confidence to the employer, and may include information such as the job seeker's skills, abilities, experience, personality, and character. As a more specific example, the second trained model 142 may be a learning model that has learned the relationship between the interview video 30 of the job seeker and the job seeker information created based on the interview video 30. That is, the second trained model 142 is a learning model that has learned the relationship between the interview video 30 of the job seeker and the job seeker information created based on the interview video 30 as teacher data, which is a video of an interview conducted with the job seeker and information annotated with the video as information to be expressed as job seeker information (such as the job seeker's skills, personality, salary required, and what is required for a job or workplace). Various existing algorithms can be used for training the second trained model 142, and examples of the second trained model 142 include, but are not limited to, support vector machines and neural networks. Therefore, the second trained model 142 is a model that receives the second video (for example, a video of an interview with a job seeker regarding the job he or she is seeking) including information about the job seeker, and estimates and generates job seeker information indicating the information about the job seeker. The job seeker is a person who is looking for a job and is hoping to find employment. The second trained model 142 may be a model trained using information that associates the voice included in the second video with the items (character strings) that should be included in the job information that can be derived from the voice, as training data. As an example, when the interview video 30 includes a voice saying "I can't refuse requests," information such as "a cooperative person," "kind," or "timid" may be associated and used as training data. As another example, when the interview video 30 includes a voice saying "I couldn't do it with the method of ~, so I decided to do ○○ and it worked," information such as "I can change my way of thinking" or "I have the ability to apply" may be associated and used as training data.As another example, when the interview video 30 includes a voice saying "I often have trouble waking up in the morning," information such as "low blood pressure" or "there is a possibility that I will be late" may be associated with the voice and used as training data. Here, the character strings associated as training data may be determined manually.

[0030] The second trained model 142 may further be a learning model that learns the relationship between technical documents, etc., created in the past by the job seeker and job seeker information. In other words, it may be a model that estimates and generates job seeker information from various terms included in the technical documents. The technical documents may be any documents created in the past by the recruiter, and are preferably documents that contain information that can identify the skills of the recruiter or the technical field in which the recruiter excels. For example, the technical documents may be papers, patent specifications, reports, etc. The technical documents may also be resume manuscripts, presentation materials, etc., created in past jobs, etc.

[0031] In addition, the second trained model 142 may be a model that has learned the relationship between a character string obtained by speech recognition of a second video (e.g., an interview video of a job seeker) containing information about the job seeker and the job seeker information.

[0032] The control unit 150 is a processor having a function of controlling each unit of the matching device 100. The control unit 150 functions as the matching device 100 by executing a program stored in the storage unit 140. The control unit 150 includes a first input unit 151, a first generation unit 152, a second input unit 153, a second generation unit 154, a calculation unit 155, and an output unit 136.

[0033] The first input unit 151 accepts input of a video (first video) of an interview with a recruiter about the skills and abilities of the desired human resource, the conditions to be given to the human resource, and the work environment to be provided, as information regarding the job of the desired human resource. As an example, the first input unit 151 accepts the first video from the recruiter terminal 200 via the communication unit 110. The first input unit 151 transmits the accepted first video to the first generation unit 152.

[0034] The first generation unit 152 generates job information of the recruiter corresponding to the first video based on the first video transmitted from the first input unit 151. The first generation unit 152 inputs the first video to the first trained model 141 to generate the job information. In addition, if the first trained model 141 is a model that has learned the relationship between a character string extracted by speech recognition from the audio of the interview video and the job information, the first generation unit 152 performs speech recognition on the first video transmitted from the first input unit 151 to extract the character string. Then, the first generation unit 152 may input the extracted character string to the first trained model 141 to generate the job information.

[0035] The second input unit 153 accepts input of a video (second video) of an interview with a job seeker about a desired job. As an example, the second input unit 153 accepts the second video from the job seeker terminal 300 via the communication unit 110. The second input unit 153 transmits the accepted second video to the second generation unit 154.

[0036] The second generation unit 154 generates job seeker information of the job seeker based on the second video transmitted from the second input unit 153. As an example, the second generation unit 154 inputs the second video transmitted from the second generation unit into the second trained model 142 to generate job seeker information of the job seeker corresponding to the second video. In the case where the second trained model 142 is a model that has learned the relationship between a character string obtained by character recognition (text mining) of the voice of the job seeker's interview and the job seeker information, the second generation unit 154 may obtain the job seeker information by performing voice recognition on the second video, extracting the character string, and inputting the character string into the second trained model 142.

[0037] The calculation unit 155 executes a matching process between the recruiter and the job seeker based on the job information and the job seeker information. That is, the calculation unit 155 calculates the degree of correlation between the character string included in the job information and the character string included in the job seeker information. For example, the degree of correlation may be an inner product of a vector based on the character string included in the job information and a vector based on the character string included in the job seeker information. As an example of the process of converting the job information and the job seeker information into vectors, an algorithm for converting character strings into vectors such as word2vec can be used, but is not limited to this. Then, the calculation unit 155 determines whether the calculated degree of correlation is equal to or greater than a predetermined threshold. The threshold may be any value that is suitable for determining that the degree of matching is high. If the degree of correlation is equal to or greater than a predetermined threshold, it is determined that the degree of matching between the recruiter and the job seeker is high, that is, the degree to which the job seeker is the recruiter that the job seeker is looking for, or the job that the recruiter is looking for is high.

[0038] The output unit 156 outputs information on the other for each combination of recruiter and job seeker for which the calculation unit 155 determines that the combination has a high degree of matching. That is, the output unit 156 transmits job seeker information 11 suitable for the desired personnel to the recruiter terminal 200 via the communication unit 110. The output unit 156 also transmits job seeker information 12 indicating jobs matching the desired conditions to the job seeker terminal 300 via the communication unit 110. The output unit 156 may also output its own job seeker information to the recruiter terminal 200. This allows the recruiter to recognize what kind of job offers will be made in his / her company or department. The output unit 156 may also output its own job seeker information to the job seeker terminal 300. This allows the job seeker to recognize what kind of job seeker information (resume) will be made.

[0039] The above is an example of the configuration of the matching device 100. Note that the recruiter terminal 200 and the job seeker terminal 300 may be any type of information processing device capable of communication, and may be any type of computer realized by a general PC, notebook PC, tablet terminal, smartphone, or the like, and therefore a detailed description of the configuration will be omitted.

[0040] <Operation> 3 to 5 are flowcharts showing an example of the operation of the matching device 100.

[0041] FIG. 3 is a flowchart showing the job information generation process.

[0042] 3, the communication unit 110 of the matching device 100 receives from the recruiter terminal 200 the recruiter interview video 20 showing the contents of an interview about the skills and abilities of the human resource the recruiter is seeking, the atmosphere of the workplace, etc., as a first video including information about the job of the human resource desired by the recruiter (step S301). The interview video 20 may also include information about various conditions such as salary and holidays to be given to the job seeker. The communication unit 110 transmits the received recruiter interview video 20 to the control unit 150.

[0043] The first input unit 151 of the control unit 150 accepts the input of the transmitted interview video 20 of the recruiter as the first video. The first input unit 151 transmits the received interview video 20 of the recruiter to the first generation unit 152. The first generation unit 152 inputs the transmitted interview video 20 into the first trained model 141, and generates job information including the abilities and personality desired in a job seeker, the workplace atmosphere to be conveyed to the job seeker, the conditions to be given to the job seeker (salary, vacation, etc.), etc. (step S302).

[0044] Then, first generation unit 152 stores the generated job information in storage unit 140 (step S303) and ends the process. At this time, first generation unit 152 may cause output unit 156 to transmit the generated job information to recruiter terminal 200. Then, recruiter terminal 200 may display the received job information, and the recruiter may check the contents.

[0045] Fig. 4 is a flowchart showing the process of generating job seeker information. As shown in Fig. 4, the communication unit 110 of the matching device 100 receives from the job seeker terminal 300 a job seeker interview video 30 showing the contents of an interview about the job seeker's personality, what the job seeker can do, the job position the job seeker is looking for (what kind of work the job seeker wants to do), the desired work environment, etc., as a second video including information related to the job seeker (step S401). The interview video 30 may also include information on basic conditions such as the name, age, address, sex, and desired salary of the employer, as well as information on the qualifications, experience, and skills acquired by the employer. The communication unit 110 transmits the received job seeker interview video 30 to the control unit 150.

[0046] The communication unit 110 may also receive from the job seeker terminal 300 technical documents relating to the job that the job seeker has created in the past (step S402), and when received, transmits them to the control unit 150. The technical documents may be, for example, papers, or, if permitted, documents (for example, presentation materials, reports, etc.) or data that the job seeker has created in the past in the course of his / her job.

[0047] The second input unit 153 of the control unit 150 accepts the input of the transmitted interview video 30 of the job seeker as the second video. The second input unit 153 also accepts the input of the technical documents of the job seeker. The second input unit 153 transmits the received interview video 30 of the job seeker and the technical documents to the second generation unit 154. The second generation unit 154 inputs the transmitted interview video 30 and technical documents into the second trained model 142, and generates job seeker information including the skills and personality possessed by the job seeker, the conditions required for a workplace, and the conditions desired by the job seeker (salary, vacation, etc.) (step S403).

[0048] Then, second generation unit 154 stores the generated job seeker information in storage unit 140 (step S404) and ends the process. At this time, second generation unit 154 may cause output unit 156 to transmit the generated job seeker information to job seeker terminal 300. Then, job seeker terminal 300 may display the received job seeker information so that the job seeker can check the contents.

[0049] FIG. 5 is a flowchart showing the matching process between job recruiters and job seekers.

[0050] The control unit 150 of the matching device 100 performs a process to determine whether or not a job seeker and a recruiter who are the subject of matching are matched. As shown in Fig. 5, the calculation unit 155 of the control unit 150 acquires job information of the recruiter who is the subject of the determination from the storage unit 140 (step S501).

[0051] The calculation unit 155 generates an employer vector based on the acquired job information (Step S502).

[0052] Furthermore, the calculation unit 155 acquires the job seeker information of the job seeker to be determined from the storage unit 140 (Step S502).

[0053] Then, the calculation unit 155 generates a job applicant vector based on the job applicant information (Step S503). In generating the job applicant vector, the same algorithm as that used in generating the employer vector is used.

[0054] Then, the calculation unit 155 calculates the degree of correlation between the generated recruiter vector and job seeker vector. Specifically, the calculation unit 155 calculates the distance between the generated recruiter vector and job seeker vector (Step S505). The distance may be the inner product of the recruiter vector and the job seeker vector.

[0055] The calculation unit 155 determines whether the calculated correlation degree is equal to or greater than a predetermined value. Specifically, the calculation unit 155 determines whether the calculated distance between the recruiter vector and the job seeker vector is equal to or less than a predetermined threshold value (step S506). In other words, the calculation unit 155 determines whether the recruiter vector and the job seeker vector are close (similar) to or greater than a certain value.

[0056] If the calculated distance is equal to or smaller than a predetermined threshold, that is, if the correlation between the employer vector and the job seeker vector is higher than a predetermined value (YES in step S506), the output unit 156 of the control unit 150 transmits the job seeker information 11 indicating the information of the job seeker who is the subject of the judgment to the employer terminal 200 of the employer who is the subject of the judgment via the communication unit 110 (step S507). Here, the job seeker information 11 may be the job seeker information itself generated by the matching device 100, or may be information obtained by extracting a part of the generated job seeker information. The extracted information may be, for example, the age, skills, and minimum wage required by the job seeker of the job seeker, which are information indicating that the job seeker meets the conditions presented by the employer, or any information that the employer wishes to disclose.

[0057] Furthermore, the output unit 156 of the control unit 150 transmits the job information 12 indicating information about the employer who is the subject of the judgment to the job seeker terminal 300 of the job seeker who is the subject of the judgment via the communication unit 110 (step S508). Here, the job information 12 may be the job information itself generated by the matching device 100, or may be information in which some information is extracted from the generated job information. The extracted information may be, for example, information about the various conditions such as salary offered by the employer, but may also be any information that the job seeker wishes to disclose.

[0058] In addition, if the correlation between the employer vector and the job seeker vector is lower than a predetermined threshold, i.e., if the distance between the employer vector and the job seeker vector is greater than a predetermined threshold (NO in step S506), the matching process, i.e., the process of introducing a job seeker to an employer or introducing a employer to a job seeker, is terminated without being performed.

[0059] The operation of the matching device 100 has been described above.

[0060] <Examples of job postings and job seeker information> FIG. 6 is an example of job information created by the matching device 100. As shown in FIG. 6, the job information includes information such as the age range required for job seekers, the salary range to be paid to the job seekers, and superficial conditions regarding vacation (see information group 601 in FIG. 6), the skills required for job seekers, the personality required for job seekers, and other abilities required outside of the job (see information group 602 in FIG. 6), and the atmosphere in the workplace to be conveyed to the job seekers (see information group 603 in FIG. 6). FIG. 6 is an example of a template of job information held by the matching device 100, and the control unit 150 (first generation unit 152) of the matching device 100 may generate the job information by using the first trained model 141. Note that the example shown in FIG. 6 is a diagram showing an example of a case where job information is sent to the recruiter terminal 200, and when it is not necessary to send the job information, it may be stored in the storage unit 140 in a form in which each item is associated with each piece of information as shown in FIG. 6. The job information shown in FIG. 6 may also be sent to the job seeker terminal 300 of a job seeker who is determined to have a high possibility of matching. Moreover, the example shown in FIG. 6 is just one example of the structure of job information, and is not limited to the form shown in FIG. 6 as long as similar information is included.

[0061] FIG. 7 is an example of job seeker information created by the matching device 100. As shown in FIG. 7, the job seeker information includes basic information such as the age and sex of the job seeker, the desired minimum wage, and other information (see information group 701 in FIG. 7), the technical field and skills in which the job seeker excels, experience (see information group 702 in FIG. 7), and the personality of the job seeker (see information group 703 in FIG. 7). FIG. 7 is an example of a template of job seeker information held by the matching device 100, and the control unit 150 (second generation unit 154) of the matching device 100 may generate the job seeker information by using the second trained model 142. Note that the example shown in FIG. 7 is a diagram showing an example of a case where job seeker information is sent to the job seeker terminal 300, and when it is not necessary to send the job seeker information, it is sufficient to store the job seeker information in the storage unit 140 in a form in which each item is associated with each piece of information as shown in FIG. 7. The job seeker information shown in FIG. 7 may also be sent to the recruiter terminal 200 of a job offer determined to have a high possibility of matching. Moreover, the example shown in FIG. 7 is just one example of the configuration of job seeker information, and is not limited to the form shown in FIG. 7 as long as similar information is included.

[0062] The calculation unit converts the job information and job seeker information into vectors and calculates the degree of correlation, and the correlation between these vectors basically looks at the correlation between information group 601 shown in FIG. 6 and information group 701 shown in FIG. 7, the correlation between information group 602 and information group 702, and the correlation between information group 603 and information group 703.

[0063] <Summary> As described above, the matching device 100 can generate the job information of the recruiter and the job seeker information of the job seeker based on the interview video of both. In the case of the recruiter, the interview video includes an interview of the atmosphere in the workplace and the culture (working methods) respected in the workplace, so the job information includes not only general information such as the job conditions and skills possessed, but also information such as the atmosphere and culture of the workplace. Similarly, the job seeker information includes information such as the personality of the job seeker and the preferred work method. Furthermore, compared to when the recruiter creates the job information himself or when the job seeker creates the job seeker information himself, by using the video and the learning model, the matching device 100 may be able to write in the job information or job seeker information even items that the person himself or herself would not have thought of writing, and can create job information or job seeker information that is more detailed and has fewer omissions than those created by humans. As a result, the matching device 100 can provide matching with a higher compatibility rate with the workplace when the job seeker is employed than when matching is simply based on skills, and can provide matching (introduction) with high satisfaction for both the recruiter and the job seeker.

[0064] <Supplementary Information> The matching device 100 according to the above embodiment is not limited to the aspects shown in the above embodiment. Various modifications will be described below.

[0065] (1) In the above embodiment, an example was shown in which job information was created using only interview videos of the recruiter, but this is not limited to this. Information other than the interview videos may be used, and as an example, text data created by the recruiter describing the type of person they are looking for may be used.

[0066] Similarly, when creating job seeker information, video interviews with the job seeker and text data created by the job seeker himself indicating the desired conditions for the job other than technical documents created in the past by the job seeker may be used.

[0067] In addition, when information other than the interview video is used as an input, the job information may be generated by integrating information estimated from the interview video and information extracted from the text data. Specifically, written information (for example, information such as the age range of the human resources required by the employer, the salary given to the job seeker, and vacation time) is received as text data, and information other than written information is extracted using interview videos obtained by interviewing the abilities and personality of the required human resources and the atmosphere of the workplace where the person will actually work, as the interview video. In this case, the first trained model 141 may be a model that has learned the correspondence between the interview video and information such as abilities and personality required of the job seeker to be extracted from the interview video, and may be a model that inputs the interview video and estimates information on abilities and personality. The first generation unit 152 may generate the job information by filling in the job information shown in FIG. 6 using written information obtained from the text data and information such as abilities and personality obtained from the interview video.

[0068] Similarly, written information of the job seeker (job seeker's career, age, desired salary conditions, etc.) may be received as text data, and an interview video including an interview asking about the job seeker's personality and an interview asking about what the job seeker wants in a workplace may be received as the interview video. In this case, the second trained model 142 may be a model that learns the correspondence between the interview video and information such as the job seeker's personality and the atmosphere desired in a workplace to be extracted from the interview video, and may be a model that inputs the interview video and estimates information indicating the job seeker's personality and what the job seeker wants in a workplace. The second generation unit 154 may generate the job seeker information shown in FIG. 7 using the written information obtained from the text data and the information indicating the job seeker's personality and what the job seeker wants in a workplace obtained from the interview video.

[0069] (2) In the above embodiment, when job information is sent to the recruiter terminal 200, if the recruiter has an error or wishes to correct the generated job information, the information can be fed back, and the control unit 150 of the matching device 100 can correct the job information based on the fed back information. Alternatively, the recruiter terminal 200 can receive the job information corrected by the recruiter, and store it in the storage unit 140.

[0070] Similarly, when job seeker information is transmitted to the job seeker terminal 300, if the job seeker has an error or wishes to correct something in the generated job seeker information, the information may be fed back and the control unit 150 of the matching device 100 may correct the job seeker information based on the fed back information. Alternatively, the job seeker terminal 300 may receive the job seeker information corrected by the job seeker and store this in the storage unit 140.

[0071] (3) In the above embodiment, when generating job seeker information, an example has been given in which an interview video and technical documents previously created by a job seeker are acquired and input into second trained model 142 to generate job seeker information, but the method of generating job seeker information is not limited to this. The above second trained model 142 is exemplified as being capable of inputting both interview videos and technical documents, but a model for inputting interview videos and a model for inputting technical documents may be created separately and stored in storage unit 140. Then, when creating job seeker information, corresponding data may be input into both models and the information may be output based on the information.

[0072] (4) In the above embodiment, an example was shown in which the first trained model 141 and the second trained model 142 were used to generate job information and job seeker information. However, the matching device 100 may generate job information and job seeker information without using these training models. That is, voice may be extracted from an interview video, the extracted voice may be recognized and converted into text data, and information usable as job information and job seeker information may be extracted from the converted text data to generate job information and job seeker information. The usable information extracted from the text data here may be extracted by storing the type and examples of the information in advance. In this way, the matching device 100 may generate job information and job seeker information by combining voice recognition processing and text mining.

[0073] (5) The matching device 100 may also provide (or include in the job seeker information 11) information on the percentage to what extent a job seeker corresponding to the job seeker information 11 provided to the recruiter terminal 200 matches the content of a job offer (degree of correlation). Similarly, the matching device 100 may also provide (or include in the job seeker information 12) information on the percentage to what extent a job seeker corresponding to the job seeker information 12 provided to the job seeker terminal 300 matches the job that the job seeker desires.

[0074] (6) The matching device 100 may receive an incentive from the recruiter of the recruiter terminal 200 or the job seeker of the job seeker terminal 300 for introducing a human resource or a job. The incentive is a reward for introducing a human resource or a job, and may basically be money or points equivalent thereto, but is not limited to these and may also be information. As an example, the incentive information may be introduction information of other job seekers, other recruitment information, etc.

[0075] (7) In the above embodiment, the matching device 100 receives an interview video of a recruiter from the recruiter terminal 200 and acquires an interview video of a job seeker from the job seeker terminal 300, but this is not limiting. These interview videos may be, for example, captured by a camera connected to the matching device 100.

[0076] (8) In the above embodiment, the functions realized by the matching device 100 may be realized by an application executed by a smartphone or the like. In this case, when the application is executed on the recruiter terminal 200, it may be configured to sequentially acquire job seeker information from the matching device 100 and perform the matching process each time the job seeker information is acquired. Also, when the application is executed on the job seeker terminal 300, it may be configured to sequentially acquire job seeker information from the matching device 100 and perform the matching process each time the recruiter information is acquired. Also, the application may display information obtained as a result of the matching process in the matching device 100 on each terminal, in which case the actual matching process may be executed in the matching device 100 as described above.

[0077] (9) In the above embodiment, the matching device 100 receives a video including information on the job duties of the recruiter's desired personnel from the recruiter terminal 200 as the first video, and a video including information on the job seeker from the job seeker terminal 300 as the second video, and performs matching. Here, it is considered that some people may find it difficult to create the first video or the second video, or may not know how to create them. In this case, the matching device 100 may be provided with a support function for creating such a first video or the second video. For example, it is considered that an online conference is held between the matching device 100 and the recruiter terminal 200 or the job seeker terminal 300, and an operator of the matching device 100 interviews the recruiter or the job seeker to create a video, which is used as the first video or the second video.

[0078] The matching device 100 itself may substitute for the operator. That is, the matching device 100 may be provided with a function of asking questions to the recruiter or the job seeker and automatically responding to the questions, so that the first video or the second video contains information sufficient to create job information or job seeker information. That is, the control unit 150 of the matching device 100 may use a so-called generative AI to generate an interview video in a form in which a question-and-answer session regarding the job is conducted between the matching device 100 and the recruiter or the job seeker, or if the question-and-answer session is text-based, the text may be substituted for the first video or the second video. The generative AI is an AI that automatically generates content such as text, programs, images, and music that are created by humans. In this embodiment, the interview interaction may be realized using a learning model that is fine-tuned using data of interactions in various types of job-related interviews (which may be text or video) and data of interactions in interviews with job seekers (which may be text or video). As a specific example, a user of the employer terminal 200 or the job seeker terminal 300 may request the matching device 100 to "conduct an interview to create employment information (or job seeker information)," and the matching device 100 may use the generation AI to ask questions and provide topics for the interview.

[0079] Also, although an example using a generation AI has been shown here, the matching device 100 may store questions likely to be used in interviews in advance, extract questions appropriate to the attributes of the user from among the stored questions (may be randomly extracted), and pose the questions to the recruiter or job seeker to conduct an interview. In this way, the matching device 100 may be provided with a support function for creating the first video or the second video, and a matching device 100 that is easier to use for users who are not good at self-promotion or explaining can be provided.

[0080] (10) In the above embodiment, it has been shown that the threshold value used by the calculation unit to determine whether the correlation between the job information and the job seeker information is high may be any value set by the operator of the matching device. This threshold value may be set to a different value for each job information or job seeker information. In addition, this threshold value may be set when the recruiter or the job seeker uses the matching device 100. That is, when the recruiter requests matching with a job seeker and transmits the first video to the matching device 100, the degree of matching with the job seeker may be set to X% (e.g., 80%) or more. Similarly, when the job seeker requests matching with a job seeker to the matching device 100 and transmits the second video to the matching device 100, the degree of matching with the job seeker may be set to Y% (e.g., 70%) or more. In this case, it is possible that one matches but the other does not match due to the difference in the threshold value. In this case, information may be sent only to the matched one, or information may not be sent to both.

[0081] (11) When each functional unit of the matching device 100 is realized by software, the matching device 100 includes a CPU that executes instructions of a program that is software that realizes each function, a ROM (Read Only Memory) or a storage device (these are referred to as "recording media") in which the program and various information are recorded so as to be readable by a computer (or CPU), and a RAM (Random Access Memory) in which the program is expanded. The object of the present invention is achieved by the computer (or CPU) reading and executing the program from the recording medium. That is, the matching device 100 according to the present invention functions as each of the above-mentioned components by the CPU executing the program loaded on the RAM. As the recording medium, a "non-transient tangible medium", for example, a semiconductor memory, a programmable logic circuit, etc. can be used. In addition, the program may be supplied to the computer via any transmission medium (such as a communication network or a broadcast wave) capable of transmitting the program. The present invention may also be realized in the form of an information signal embedded in a carrier wave in which the program is embodied by electronic transmission.

[0082] The above program may be implemented using, for example, a scripting language such as ActionScript, JavaScript (registered trademark), Python, or Ruby, an object-oriented programming language such as C language, C++, C#, Objective-C, Swift, or Java (registered trademark), or a markup language such as HTML5. Furthermore, the term "section, module, or unit" in the claims may be read as "means" or "circuit." For example, a communication section may be read as a communication means or a communication circuit.

[0083] Furthermore, the program of the present disclosure may be provided to the matching device 100 via any transmission medium (such as a communication network or broadcast waves) capable of transmitting the program.

[0084] (12) The order and timing of the processes described in the above embodiment may be changed as appropriate as long as the same effects as those described in the above embodiment can be achieved. For example, in FIG. 4, the processes of step S401 and step S402 may be performed in sequence. [Explanation of symbols]

[0085] 100 Matching Device 110 Communications Department 120 Input section 130 Output section 140 Storage section 150 Control section 151 First input section 152 1st generation part 153 Second input section 154 Second generation part 155 Calculation Section 156 Output section 200 Recruiter Terminals 300 Job Seeker Terminal 400 Network

Claims

1. A first input unit that receives an input of a first video including information related to the job of the human resource desired by the recruiter; A first generation unit that analyzes the first video and generates job information including an atmosphere of the workplace of the job seeker; a second input unit that accepts an input of a second video including information related to the job seeker; a second generation unit that analyzes the second video and generates job seeker information including a temperament of the job seeker; A calculation unit that calculates a matching degree between the job information and the job seeker information; and an output unit that outputs the matching degree. Matching device.

2. The matching device described in claim 1, characterized in that the first generation unit generates the job information by inputting the first video accepted by the first input unit into a first trained model that has learned the relationship between a video including information related to the job duties of the personnel desired by the employer and the job information to be generated based on the video.

3. The matching device described in claim 1, characterized in that the second generation unit generates the job seeker information by inputting the second video accepted by the second input unit into a second trained model that has learned the relationship between a video including information related to a job seeker and job seeker information to be generated based on the video.

4. The second trained model is a model that has further learned the relationship between technical documents and job seeker information, The second input unit further receives input of documents previously filled out by the job seeker.

4. The matching device according to claim 3.

5. The computer A first input step of receiving an input of a first video including information related to the job of a human resource desired by a recruiter; A first generation step of analyzing the first video and generating job information including an atmosphere of the workplace of the job seeker; a second input step of receiving an input of a second image including information related to the job seeker; a second generation step of analyzing the second video to generate job seeker information including a temperament of the job seeker; a calculation step of calculating a degree of matching between the job information and the job seeker information; and outputting the degree of matching. Matching method.

6. On the computer, A first input function for receiving an input of a first video including information related to the job of the human resource desired by the recruiter; A first generation function that analyzes the first video and generates job information including an atmosphere of the workplace of the job seeker; a second input function for receiving an input of a second image including information related to the job seeker; a second generation function for analyzing the second video and generating job seeker information including a temperament of the job seeker; A calculation function for calculating a degree of matching between the job information and the job seeker information; and an output function for outputting the degree of matching. Matching program.

Images