Computer program, information processing method, and information processing device

Abstract

To provide a computer program for supporting an employment interview for employing appropriate human resources, an information processing method, and an information processing device.SOLUTION: A computer program acquires information on employment reference of human resources (S103), acquires determination reference of employment corresponding to information on the acquired employment reference by using a language generation model (S104), acquires questions to employment applicants in accordance with information on the acquired employment reference by using the language generation model (S108), and makes a computer execute processing (S109) for outputting the acquired questions.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a computer program, an information processing method, and an information processing device for supporting a job interview. [Background technology]

[0002] In order to hire personnel at a business, a job interview is conducted. In recent years, job interviews are sometimes conducted via a communication network without the interviewer and the job applicant actually meeting. Patent Document 1 discloses a technology for supporting job interviews using a computer. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-31089 A Summary of the Invention [Problem to be solved by the invention]

[0004] In a job interview, the job applicant answers questions posed by the interviewer, and the job applicant is judged whether or not to hire based on the answers. In order to hire the right person, the interviewer needs to ask the right questions to be able to distinguish the people who meet the hiring criteria. However, asking the right questions according to the hiring criteria is not always easy.

[0005] The present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide a computer program, an information processing method, and an information processing device that support employment interviews to hire suitable personnel. [Means for solving the problem]

[0006] A computer program according to one embodiment of the present invention is characterized in that it causes a computer to execute a process of acquiring information regarding hiring criteria for personnel, using a language generation model to acquire hiring judgment criteria based on the acquired information regarding the hiring criteria, acquiring questions to be posed to job applicants based on the acquired information regarding the hiring criteria and the judgment criteria, and outputting the acquired questions.

[0007] A computer program according to one embodiment of the present invention is characterized in that it causes a computer to execute a process of obtaining an answer to the question, using the language generation model to obtain information regarding the hiring criteria and an evaluation of the job applicant based on the question and the answer, and outputting the obtained evaluation.

[0008] An information processing method according to one embodiment of the present invention is characterized in that it acquires information regarding hiring criteria for personnel, uses a language generation model to acquire hiring judgment criteria based on the acquired information on the hiring criteria, acquires questions for job applicants based on the acquired information on the hiring criteria and the judgment criteria using the language generation model, and outputs the acquired questions.

[0009] An information processing device according to one embodiment of the present invention includes a calculation unit that acquires information relating to hiring criteria for personnel, uses a language generation model to acquire hiring judgment criteria based on the acquired information on the hiring criteria, acquires questions for job applicants based on the acquired information on the hiring criteria and the judgment criteria using the language generation model, and outputs the acquired questions.

[0010] In one embodiment of the present invention, a language generation model is used to obtain hiring criteria corresponding to information on the hiring criteria, and a language generation model is used to obtain questions to be used in a job interview for a job applicant corresponding to the information on the hiring criteria and the hiring criteria. Appropriate questions to be used in the job interview to identify personnel who meet the hiring criteria are obtained. By using the language generation model, the appropriate questions are automatically obtained.

[0011] In one embodiment of the present invention, a language generation model is used to generate an evaluation of the job applicant based on information about the hiring criteria, questions, and answers to the questions. Based on the results of a job interview using the generated questions, an evaluation is obtained of whether the job applicant meets the hiring criteria. An objective evaluation of the job applicant is automatically obtained based on the results of the job interview. Effect of the Invention

[0012] The present invention has excellent effects such as being able to support employment interviews so that the interviews can be conducted appropriately or so that the required work can be reduced. [Brief description of the drawings]

[0013] [Figure 1] 1 is a schematic diagram showing an example of the configuration of an interview support system. [Diagram 2] 1 is a block diagram showing an example of an internal configuration of an information processing device; [Diagram 3] 11 is a flowchart illustrating an example of a procedure for generating a question executed by an information processing device. [Figure 4] 10 is a conceptual diagram showing an example of a prompt input to the language generation model in S102 and a judgment criterion output from the language generation model. FIG. [Diagram 5] 13 is a conceptual diagram showing an example of a prompt input to a language generation model in S107 and a question output from the language generation model. FIG. [Figure 6]13 is a flowchart illustrating an example of a procedure for generating a follow-up question, which is executed by the information processing device. [Figure 7] 11 is a flowchart illustrating an example of a procedure for a process of evaluating a job applicant executed by an information processing device. [Figure 8] FIG. 13 is a conceptual diagram illustrating an example of a prompt that includes instructions for generating an assessment. [Figure 9] FIG. 13 is a conceptual diagram showing an example of an evaluation output from a language generation model. [Figure 10] 11 is a flowchart showing an example of a procedure of a process executed by an information processing device according to a second embodiment. [Figure 11] 11 is a flowchart showing an example of a procedure of a process executed by an information processing device according to a second embodiment. [Figure 12] FIG. 11 is a conceptual diagram showing an overview of the processes in S402 to S405. [Figure 13] FIG. 11 is a conceptual diagram showing an overview of the processes of S408 to S411. [Figure 14] FIG. 13 is a conceptual diagram showing an overview of the processes in S412 to S415. [Figure 15] 13 is a schematic diagram showing an example of outputting a general comment, a judgment comment, and the overall level of the job applicant. FIG. [Figure 16] 13 is a flowchart showing an example of a processing procedure in which an information processing device according to a second embodiment creates a transfer item. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] The present invention will now be described in detail with reference to the drawings showing embodiments thereof. <Embodiment 1> FIG. 1 is a schematic diagram showing an example of the configuration of an interview support system 100. The interview support system 100 is a system that performs processing to support a recruitment interview for hiring personnel. The interview support system 100 includes an information processing device 1. The information processing device 1 is connected to a communication network 3 such as the Internet. The information processing device 1 is used by a user 4 such as an interviewer. A terminal device 2 is connected to the communication network 3. The terminal device 2 is used by a job applicant 5 who will take a recruitment interview.

[0015] FIG. 2 is a block diagram showing an example of the internal configuration of the information processing device 1. The information processing device 1 is configured using a computer such as a server device or a PC (personal computer). The information processing device 1 executes an information processing method. The information processing device 1 includes a calculation unit 11, a memory 12, a reading unit 13, a storage unit 14, an input unit 15, an output unit 16, and a communication unit 17. The calculation unit 11 is configured using, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a multi-core CPU. The calculation unit 11 may be configured using a quantum computer. The memory 12 stores temporary data generated in association with calculation. The memory 12 is, for example, a RAM (Random Access Memory). The reading unit 13 reads information from a recording medium 10 such as an optical disk or a portable memory. The storage unit 14 is non-volatile, for example, a hard disk or a non-volatile semiconductor memory.

[0016] The input unit 15 accepts input of information from the user 4. The input unit 15 includes an operation unit that accepts input of information such as text by accepting an operation from the user 4. The operation unit is, for example, a touch panel, a keyboard, or a pointing device. The input unit 15 may include a video input unit that accepts input of video using a photographing device, and an audio input unit such as a microphone that accepts input of audio.

[0017] The output unit 16 outputs information. The output unit 16 includes a display unit that displays an image. The display unit is, for example, a liquid crystal display or an electroluminescent display (EL display). The output unit 16 may include an audio output unit such as a speaker that outputs audio. The input unit 15 or the output unit 16 may be configured using an input / output interface. The communication unit 17 communicates with the outside of the information processing device 1 by wired or wireless communication. Specifically, the communication unit 17 communicates with the terminal device 2 via the communication network 3.

[0018] The calculation unit 11 causes the reading unit 13 to read the computer program 141 recorded on the recording medium 10, and stores the read computer program 141 in the storage unit 14. The calculation unit 11 executes processing for realizing the functions of the information processing device 1 in accordance with the computer program 141. The computer program 141 may be a program product. The computer program 141 may be stored in advance in the storage unit 14, or may be downloaded from outside the information processing device 1. In this case, the information processing device 1 does not need to include the reading unit 13.

[0019] The computer program 141 can be deployed to be executed on a single computer, or on multiple computers located at one site or distributed across multiple sites and interconnected by a communications network. That is, the information processing device 1 may be composed of multiple computers, and the computer program 141 may be executed on multiple computers connected via a communications network.

[0020] The information processing device 1 includes a language generation model 142. The language generation model 142 is a trained model trained to output a sentence corresponding to a prompt and information associated with the prompt when the prompt is input. For example, the language generation model 142 is a large-scale language model. More specifically, the language generation model 142 is BERT, GPT-4, Bard, LLaMA, or the like.

[0021] The language generation model 142 is realized by the calculation unit 11 executing information processing according to the computer program 141. The storage unit 14 stores data for realizing the language generation model 142. The language generation model 142 may be configured by hardware. The language generation model 142 may be realized using a quantum computer. Alternatively, the language generation model 142 may be provided outside the information processing device 1, and the information processing device 1 may execute processing using the external language generation model 142. For example, the language generation model 142 may be realized using a plurality of computers connected via a communication network, or may be realized using a cloud.

[0022] The terminal device 2 is a computer such as a PC, a smartphone, a tablet computer, etc. A job interview is conducted by transmitting and receiving video and audio between the information processing device 1 and the terminal device 2 via a communication network 3.

[0023] In preparation for the job interview, the information processing device 1 performs a process of generating questions to be asked to the job applicant 5 in the job interview, using the language generation model 142. In the process of generating questions, the information processing device 1 generates hiring criteria corresponding to information on the hiring criteria, using the language generation model 142. The information processing device 1 also generates questions to be used in the job interview for the job applicant 5, corresponding to the information on the hiring criteria and the criteria, using the language generation model 142.

[0024] The information on the hiring criteria includes the hiring criteria and the definition of the hiring criteria. The hiring criteria are expressed in words as the criteria for the personnel that the establishment should hire. The definition of the hiring criteria is a more specific definition of the contents of the hiring criteria. The hiring criteria and the definition of the hiring criteria are created in advance by the establishment. The memory unit 14 stores hiring criteria data that records the hiring criteria and the definition of the hiring criteria. The judgment criteria are criteria for judging whether or not an applicant is a personnel that matches the hiring criteria, and are expressed in words as the characteristics of personnel that match the hiring criteria or personnel that do not match the hiring criteria. For example, the judgment criteria include typical behaviors that personnel that match the hiring criteria should exhibit, or typical bad behaviors that personnel that match the hiring criteria should not exhibit.

[0025] 3 is a flowchart showing an example of a procedure of a process for generating a question executed by the information processing device 1. Hereinafter, step is abbreviated as S. The calculation unit 11 executes information processing according to the computer program 141, so that the information processing device 1 executes the following process. The information processing device 1 acquires the acceptance criteria and the definition of the acceptance criteria (S101). In S101, the calculation unit 11 acquires the acceptance criteria and the definition of the acceptance criteria by reading out the acceptance criteria and the definition of the acceptance criteria recorded in the acceptance criteria data stored in the storage unit 14. The user 4 may operate an operation unit included in the input unit 15 to input the acceptance criteria and the definition of the acceptance criteria, and the calculation unit 11 may acquire the input acceptance criteria and the definition of the acceptance criteria.

[0026] An example of the hiring criteria and their definitions is shown below. ·Recruitment criteria: Commitment · Employment criteria definition: Willingness to go to great lengths to achieve goals. Demonstrates self-discipline and tenacity in the process of planning, executing and evaluating results.

[0027] The information processing device 1 inputs a prompt including the acquired adoption criteria and the definition of the adoption criteria and including an instruction to generate a judgment criterion to the language generation model 142 (S102). For example, the content of the prompt instructing the generation of the judgment criteria is stored in advance in the storage unit 14 or is included in the computer program 141. The prompt instructing the generation of the judgment criteria may be input by the user 4 operating an operation unit included in the input unit 15. In S102, the calculation unit 11 generates a prompt instructing the generation of a plurality of judgment criteria by including the adoption criteria and the definition of the adoption criteria, and inputs the generated prompt to the language generation model 142.

[0028] The language generation model 142 performs calculations and outputs sentences in response to the input of a prompt. The prompt input to the language generation model 142 includes an instruction to generate an acceptance criterion, a definition of the acceptance criterion, and a plurality of judgment criteria, so that the language generation model 142 outputs a plurality of judgment criteria according to the acceptance criterion and the definition of the acceptance criterion.

[0029] Fig. 4 is a conceptual diagram showing an example of the prompt input to the language generation model 142 in S102 and the judgment criteria output from the language generation model 142. The example of the prompt shown in Fig. 4 instructs the generation of a typical behavior as the judgment criterion, and includes the adoption criteria and the definition of the adoption criterion as described above. In response to the input of the prompt, a plurality of typical behaviors are output from the language generation model 142 as a plurality of judgment criteria.

[0030] The information processing device 1 acquires the hiring criteria and hiring criteria corresponding to the definition of the hiring criteria (S103). In S103, the calculation unit 11 acquires a plurality of judgment criteria output by the language generation model 142. By acquiring a plurality of judgment criteria, it becomes possible to make hiring decisions from a plurality of viewpoints. The calculation unit 11 stores the acquired judgment criteria in the memory unit 14. The information processing device 1 outputs the acquired judgment criteria (S104). In S104, the calculation unit 11 displays an image including the judgment criteria on a display unit included in the output unit 16.

[0031] The information processing device 1 waits for receipt of an instruction to modify the judgment criteria (S105). In S105, the user 4 operates an operation unit included in the input unit 15, whereby the calculation unit 11 receives the instruction to modify the judgment criteria. If an instruction to modify the judgment criteria is received (S105: YES), the information processing device 1 modifies and outputs the judgment criteria (S106). In S106, the calculation unit 11 modifies the judgment criteria according to the received instruction, and displays an image including the modified judgment criteria on a display unit included in the output unit 16.

[0032] The processes of S105 and S106 may be repeated as necessary. The processes of S105 and S106 allow the judgment criteria to be modified to the content that the user 4 judges to be more appropriate. For example, the inappropriate judgment criteria may be deleted. The calculation unit 11 stores the modified judgment criteria in the storage unit 14. Note that the language generation model 142 may be re-trained using the modified judgment criteria as training data.

[0033] If no instruction to modify the judgment criteria is received in S105 (S105: NO), or after S106 is completed, the information processing device 1 inputs a prompt including the hiring criteria, the definition of the hiring criteria, and the judgment criteria, and including an instruction to generate a question for the job applicant 5, to the language generation model 142 (S107). For example, the content of the prompt instructing the generation of a question is stored in advance in the storage unit 14 or is included in the computer program 141. The prompt instructing the generation of a question may be input by the user 4 operating an operation unit included in the input unit 15. In S107, the calculation unit 11 generates a prompt instructing the generation of a plurality of questions, including the hiring criteria, the definition of the hiring criteria, and the judgment criteria, and inputs the generated prompt to the language generation model 142.

[0034] The language generation model 142 performs calculations and outputs sentences in response to the input of a prompt. The prompt input to the language generation model 142 includes an instruction to generate an acceptance criterion, a definition of the acceptance criterion, a judgment criterion, and a plurality of questions, so that the language generation model 142 outputs a plurality of questions according to the acceptance criterion, the definition of the acceptance criterion, and the judgment criterion.

[0035] Fig. 5 is a conceptual diagram showing an example of a prompt input to the language generation model 142 in S107 and a question output from the language generation model 142. The example of the prompt shown in Fig. 5 includes the adoption criteria and the definition of the adoption criteria as described above, and typical behaviors that are acquired judgment criteria. In response to the input of the prompt, a plurality of questions are output from the language generation model 142.

[0036] The information processing device 1 acquires questions for the job applicant 5 according to the hiring criteria, the definition of the hiring criteria, and the judgment criteria (S108). In S108, the calculation unit 11 acquires a plurality of questions output by the language generation model 142. Obtaining a plurality of questions makes it possible to ask the job applicant 5 appropriate questions a plurality of times during the job interview. The information processing device 1 outputs the acquired questions (S109). In S109, the calculation unit 11 displays an image including the questions on a display unit included in the output unit 16.

[0037] The information processing device 1 waits for an instruction to revise a question for the job applicant 5 (S110). In S110, the user 4 operates an operation unit included in the input unit 15, whereby the calculation unit 11 receives an instruction to revise the question. If an instruction to revise the question is received (S110: YES), the information processing device 1 revises the question and outputs it (S111). In S111, the calculation unit 11 revise the question according to the received instruction, and displays an image including the revised question on the display unit included in the output unit 16. The processes of S110 and S111 may be repeated as necessary. The processes of S110 and S111 allow the user 4 to revise the question to a content that the user 4 judges to be more appropriate. For example, an inappropriate question can be deleted.

[0038] If an instruction to modify the question is not received in S110 (S110: NO), or after S111 is completed, the information processing device 1 stores the question (S112). In S112, the calculation unit 11 stores the acquired question or the modified question in the memory unit 14. After S112 is completed, the information processing device 1 ends the process of generating a question.

[0039] In the job interview, the questions obtained in the processes of S101 to S112 are used. For example, questions to the job applicant 5 are transmitted as text or voice from the information processing device 1 to the terminal device 2 via the communication network 3. Similarly, answers to the questions from the job applicant 5 are transmitted to the information processing device 1 from the terminal device 2.

[0040] The information processing device 1 performs a process of generating follow-up questions in response to answers to questions during a job interview. In the process of generating follow-up questions, the information processing device 1 utilizes the language generation model 142 to generate follow-up questions in response to questions for the job applicant 5, answers from the job applicant 5 to the questions, and judgment criteria.

[0041] 6 is a flowchart showing an example of the procedure of a process for generating an additional question executed by the information processing device 1. The information processing device 1 acquires an answer to the question (S21). In S21, the calculation unit 11 acquires the answer to the question by receiving the answer transmitted from the terminal device 2 via the communication unit 17. The calculation unit 11 stores the received answer in the storage unit 14. The calculation unit 11 may acquire the answer by reading out the answer stored in the storage unit 14. The calculation unit 11 may acquire the answer by having the user 4 input the content of the answer using the input unit 15.

[0042] The information processing device 1 inputs a prompt including a question, an answer to the question, and a judgment criterion, and including an instruction to generate an additional question, to the language generation model 142 (S22). For example, the content of the prompt instructing the generation of an additional question is stored in advance in the storage unit 14 or is included in the computer program 141. The prompt instructing the generation of an additional question may be input by the user 4 operating an operation unit included in the input unit 15. In S22, the calculation unit 11 generates a prompt instructing the generation of an additional question including a question, an answer to the question, and a judgment criterion, and inputs the generated prompt to the language generation model 142. In S22, the calculation unit 11 may generate a prompt instructing the generation of an additional question including a question and a judgment criterion, add an answer to the question, and input the prompt with the answer added to the language generation model 142.

[0043] An example of a prompt to generate additional questions is shown below: prompt: You are an excellent consultant. #Request You are a professional interviewer. In order to more accurately determine whether the answers of job applicants in the {#interview data} show that they have ${#self-control}, please create questions that are different from the {#questions} below. We will also list each {##typical behavior}, so please use this as a basis for your consideration. The {#interview data} will be posted after this {#request}. Please create your questions after that. #question Typical behavior For example, after the prompt shown in the above example includes a question and a typical behavior as a criterion for judgment, the prompt and interview data as answers to the question are input to the language generation model 142.

[0044] The language generation model 142 performs calculations and outputs sentences in response to the input of a prompt. The prompt input to the language generation model 142 includes a question, an answer to the question, a criterion for judgment, and an instruction for generating a follow-up question, so the language generation model 142 outputs a question for the job applicant 5 and a follow-up question in response to the answer to the question.

[0045] The information processing device 1 acquires questions for the job applicant 5, answers to the questions, and additional questions corresponding to the judgment criteria (S23). In S23, the calculation unit 11 acquires the additional questions output by the language generation model 142. The information processing device 1 outputs the acquired additional questions (S24). In S24, the calculation unit 11 displays an image including the additional questions on a display unit included in the output unit 16. The calculation unit 11 stores the acquired additional questions in the memory unit 14. After S24 is completed, the information processing device 1 terminates the process of generating additional questions.

[0046] The job interview is conducted using the additional questions obtained by the processes of S21 to S24. The additional questions are transmitted from the information processing device 1 to the terminal device 2 via the communication network 3. During the job interview, the processes of S21 to S24 may be repeatedly executed as necessary.

[0047] After the job interview, the information processing device 1 performs a process of evaluating the job applicant 5 in accordance with the answers to the questions. In the process of evaluating the job applicant 5, the information processing device 1 uses the language generation model 142 to generate an evaluation of the job applicant 5 in accordance with information on the hiring criteria, questions for the job applicant 5, and the answers from the job applicant 5 to the questions.

[0048] 7 is a flowchart showing an example of the procedure of a process executed by the information processing device 1 to evaluate a job applicant 5. The information processing device 1 acquires questions and answers (S31). The memory unit 14 stores questions used in a job interview and answers given by the job applicant 5 to the questions. In S31, the calculation unit 11 acquires the questions and answers by reading out the questions and answers to the questions stored in the memory unit 14.

[0049] The information processing device 1 inputs a prompt including hiring criteria, a definition of the hiring criteria, a question, and an answer to the question, and including an instruction to generate an evaluation for the job applicant 5, to the language generation model 142 (S32). For example, the content of the prompt instructing the generation of an evaluation for the job applicant 5 is stored in advance in the storage unit 14 or is included in the computer program 141. The prompt instructing the generation of an evaluation for the job applicant 5 may be input by the user 4 operating an operation unit included in the input unit 15. In S32, the calculation unit 11 generates a prompt instructing the generation of an evaluation for the job applicant 5 including the hiring criteria, a definition of the hiring criteria, a question, and an answer to the question, and inputs the generated prompt to the language generation model 142. The prompt input to the language generation model 142 may include a judgment criterion.

[0050] FIG. 8 is a conceptual diagram showing an example of a prompt including an instruction to generate an evaluation. The interview content included in the prompt includes a question and an answer to the question, and the prompt includes an instruction to generate an evaluation for the job applicant 5 based on the interview content. The prompt also includes "self-control" as an item of the hiring criteria, and includes a definition of the hiring criteria related to this item in association with the hiring criteria item. The example shown in FIG. 8 is an example in which the prompt includes a judgment criterion, and includes a judgment criterion and a typical good behavior example and a typical bad behavior example that are specific examples of the judgment criterion in association with the hiring criteria item.

[0051] As shown in Fig. 8, a prompt including an instruction to generate an evaluation instructs the language generation model 142 to compare the answer of the job applicant 5 with the hiring criteria and the definition of the hiring criteria, interpret the answer including the context, and evaluate whether the answer matches the hiring criteria. For example, the instruction is to determine the similarity or difference between the answer and the hiring criteria, and to evaluate according to the similarity or difference. If a judgment criterion is included in the prompt, the instruction is to make the evaluation with reference to the judgment criterion.

[0052] The language generation model 142 performs a calculation in response to the input of the prompt and outputs a sentence. The prompt input to the language generation model 142 includes the hiring criteria, the definition of the hiring criteria, a question, an answer to the question, and an instruction to generate an evaluation of the job applicant 5, so that the language generation model 142 outputs an evaluation of the job applicant 5 according to the hiring criteria, the definition of the hiring criteria, the question, and the answer. For example, a prompt including an instruction to evaluate the degree to which the job applicant 5 matches the hiring criteria in multiple stages is input to the language generation model 142. The language generation model 142 outputs an evaluation including a result of determining the degree to which the job applicant 5 matches the hiring criteria as one of multiple stages. The prompt may be configured so that the evaluation result is output by a method other than the method of evaluating the job applicant 5 in multiple stages. For example, an evaluation expressed in numerical values ​​may be output, or an evaluation expressed only in sentences may be output.

[0053] FIG. 9 is a conceptual diagram showing an example of an evaluation output from the language generation model 142. The evaluation shown in FIG. 9 is an example of an evaluation output from the language generation model 142 in response to the prompt shown in FIG. 8. The prompt is configured so that an evaluation including an evaluation result, which is a result of judging the degree to which the job applicant 5 matches the hiring criteria, an interpretation of the evaluation result or a reason for the evaluation result, and a grounds for the evaluation, is output. The overall evaluation included in the example shown in FIG. 9 corresponds to the interpretation of the evaluation result or the reason for the evaluation result. Although FIGS. 8 and 9 show an example in which there is one hiring criterion item, the prompt may be configured so that there are multiple hiring criterion items and an evaluation is output for each item.

[0054] The information processing device 1 acquires an evaluation of the job applicant 5 according to the hiring criteria, the definition of the hiring criteria, the questions, and the answers (S33). In S33, the calculation unit 11 acquires the evaluation of the job applicant 5 output by the language generation model 142. The information processing device 1 outputs the acquired evaluation of the job applicant 5 (S34). In S34, the calculation unit 11 displays an image including the evaluation of the job applicant 5 on a display unit included in the output unit 16. The calculation unit 11 stores the acquired evaluation of the job applicant 5 in the storage unit 14. After S34 is completed, the information processing device 1 ends the process of evaluating the job applicant 5. The selection of personnel to be hired is performed with reference to the evaluation of the job applicant 5 obtained by the processes of S31 to S34. The obtained evaluation may be appropriately corrected by the user 4. The language generation model 142 may be re-trained using the corrected evaluation as training data.

[0055] In order to conduct a job interview and hire a suitable person, it is necessary to ask suitable questions and make an evaluation so that a person who meets the hiring criteria can be identified. However, in the conventional method, it may not be easy to ask suitable questions and make suitable evaluations according to the hiring criteria. As described above, in this embodiment, the information processing device 1 uses the language generation model 142 to generate hiring criteria according to information on the hiring criteria. In addition, the information processing device 1 uses the language generation model 142 to generate questions to be used in a job interview for the job applicant 5 according to the information on the hiring criteria and the criteria. The criteria according to the hiring criteria are obtained, and the hiring criteria and questions according to the criteria are obtained, so that suitable questions to be used in a job interview to identify a person who meets the hiring criteria are obtained. By using the language generation model 142, suitable questions are automatically obtained.

[0056] This embodiment reduces the work of creating appropriate questions to identify personnel who meet the hiring criteria. The interviewer can conduct an appropriate hiring interview to identify personnel who meet the hiring criteria simply by asking the created questions to the job applicants 5. This makes it possible to conduct an appropriate hiring interview without relying on the ability of the interviewer.

[0057] The information processing device 1 uses the language generation model 142 to generate questions for the job applicant 5, answers from the job applicant 5 to the questions, and additional questions that correspond to the hiring criteria. By obtaining additional questions that correspond to the questions, answers, and criteria, appropriate additional questions can be obtained when additional questions become necessary during the job interview. By using the language generation model 142, appropriate additional questions can be obtained automatically. This eliminates the need for the interviewer to create additional questions during the job interview, reducing the burden on the interviewer. When additional questions become necessary, the job interview can be continued appropriately without relying on the interviewer's ability.

[0058] Furthermore, the information processing device 1 uses the language generation model 142 to generate an evaluation of the job applicant 5 according to information on the hiring criteria, questions to the job applicant 5, and the answers from the job applicant 5 to the questions. By obtaining an evaluation according to the hiring criteria, questions, and answers, an evaluation of whether the job applicant 5 meets the hiring criteria can be obtained based on the results of the job interview. By using the language generation model 142, an objective evaluation of the job applicant 5 based on the results of the job interview can be automatically obtained. The obtained evaluation becomes an objective material for determining whether or not to hire the job applicant 5, and enables the employer to make an appropriate decision and hire the appropriate person. As described above, the information processing device 1 can support the job interview so that the appropriate person can be hired or the necessary work can be reduced.

[0059] In the present embodiment, an example has been shown in which the information on the hiring criteria includes the hiring criteria and the definitions of the hiring criteria, but the information on the hiring criteria may include other information. For example, the information on the hiring criteria may include only each item of the hiring criteria. In the present embodiment, an example has been shown in which a hiring interview is conducted via a communication network, but the information processing device 1 may also be used in a hiring interview in which the interviewer and the job applicant actually meet. For example, the hiring interview is conducted using questions that are generated in advance by the information processing device 1. For example, answers from the job applicant are input to the information processing device 1 by voice input or text input, and the information processing device 1 generates additional questions during the hiring interview and generates an evaluation of the job applicant 5 after the hiring interview.

[0060] In the present embodiment, an example has been shown in which the process of generating questions from S101 to S112, the process of generating additional questions from S21 to S24, and the process of evaluating the employment applicant 5 from S31 to S34 are executed by a single information processing device 1. The processes of S101 to S112, S21 to S24, and S31 to S34 are not necessarily executed by a single information processing device 1, and may be executed by different information processing devices.

[0061] <Embodiment 2> In the second embodiment, the information processing device 1 does not generate questions, but evaluates the job applicant 5 according to the statements of the job applicant 5 based on the hiring criteria. The configuration of the interview support system 100 according to the second embodiment is the same as that of the first embodiment. The configuration of the information processing device 1 according to the second embodiment is also the same as that of the first embodiment. The memory unit 14 stores hiring criteria data. The hiring criteria data records the hiring criteria and an explanation of the hiring criteria. The explanation of the hiring criteria explains the content of the hiring criteria in concrete terms and is equivalent to the definition of the hiring criteria. The hiring criteria data records a plurality of hiring criteria, and an explanation of each hiring criterion. The hiring criteria and the explanation of the hiring criteria are included in the information related to the hiring criteria. The hiring criteria and the explanation of the hiring criteria are created in advance.

[0062] In a job interview, questions to the job applicant 5 are transmitted as text or voice from the information processing device 1 to the terminal device 2 via the communication network 3, and similarly, answers to the questions from the job applicant 5 are transmitted from the terminal device 2 to the information processing device 1. For example, questions are inputted from the user 4 to the information processing device 1. In addition to answers to questions, comments by the job applicant 5 may be inputted from the terminal device 2 to the information processing device 1 via the communication network 3. The information processing device 1 stores the contents of the comments by the job applicant 5, including the answers to the questions, in the memory unit 14. The contents of the comments may be data expressed as text or may be voice data.

[0063] 10 and 11 are flowcharts showing an example of a procedure of a process executed by the information processing device 1 according to the second embodiment. The information processing device 1 acquires an employment criterion and an explanation of the employment criterion (S401). In S401, the calculation unit 11 acquires a plurality of employment criteria by reading the contents of the employment criterion data, and acquires an explanation of each of the employment criteria. For example, the plurality of employment criteria are stress control, commitment, adaptability, flexibility, etc.

[0064] The information processing device 1 inputs a prompt including the acquired adoption criteria and the explanation of the adoption criteria and including an instruction to generate a judgment criterion and an explanation of the judgment criterion to the language generation model 142 (S402). The explanation of the judgment criterion is an explanation of the content of the judgment criterion in concrete words. In S402, the calculation unit 11 generates a prompt including an instruction to generate a judgment criterion and an explanation of the judgment criterion according to the adoption criteria and the explanation of the adoption criterion. The prompt includes an instruction to generate a plurality of judgment criteria and an explanation of each judgment criterion for each adoption criterion. The calculation unit 11 may generate the prompt using a template of a prompt stored in the storage unit 14 in advance or included in the computer program 141. The prompt may be input by the user 4 operating an operation unit included in the input unit 15. The calculation unit 11 inputs the generated prompt to the language generation model 142.

[0065] The language generation model 142 performs calculations in response to the input of a prompt and outputs language. The language generation model 142 outputs a plurality of judgment criteria and an explanation of each judgment criterion according to the adoption criteria and the explanation of the adoption criteria included in the input prompt.

[0066] The information processing device 1 acquires a judgment criterion and an explanation of the judgment criterion corresponding to the adoption criterion and the explanation of the adoption criterion (S403). In S403, the calculation unit 11 acquires a plurality of judgment criteria and an explanation of each of the judgment criteria output by the language generation model 142. The calculation unit 11 stores the acquired judgment criteria and the explanation of the judgment criterion in the storage unit 14. The information processing device 1 may perform a process of correcting the acquired judgment criterion and the explanation of the judgment criterion.

[0067] In S403, the calculation unit 11 acquires multiple judgment criteria and an explanation of each judgment criterion for each of the multiple adoption criteria. For example, the calculation unit 11 inputs a prompt including multiple adoption criteria and an explanation of each adoption criterion to the language generation model 142 in S402, and acquires multiple judgment criteria and an explanation of each judgment criterion for each adoption criterion in S403. Alternatively, the calculation unit 11 may repeatedly execute S402 of inputting a prompt including a single adoption criterion and an explanation of the adoption criterion to the language generation model 142 and S403 of acquiring multiple judgment criteria and an explanation of each judgment criterion for the single adoption criterion for each adoption criterion.

[0068] The information processing device 1 then inputs a prompt including the hiring criteria and an explanation of the hiring criteria, as well as the acquired judgment criteria and an explanation of the judgment criteria, and including an instruction to generate definitions of multiple levels of talent judged based on the judgment criteria to the language generation model 142 (S404). The multiple levels of talent are multiple-stage expressions of the value of talent judged based on the judgment criteria. For example, each level is expressed as high, medium, low, or poor. The multiple levels of talent may be expressed numerically, or may be expressed by multiple letters or words with a hierarchy such as A, B, and C. The level definition is a concrete expression of the characteristics of talent corresponding to the level.

[0069] In S404, the calculation unit 11 generates a prompt including an instruction to generate a definition of a plurality of levels of talent according to the hiring criteria, an explanation of the hiring criteria, and the acquired judgment criteria and an explanation of the judgment criteria. The prompt includes an instruction to generate a definition of a plurality of levels of talent to be judged based on the judgment criteria, for each judgment criterion. The calculation unit 11 may generate the prompt using a template of the prompt stored in advance in the storage unit 14 or included in the computer program 141. The prompt may be input by the user 4 operating an operation unit included in the input unit 15. The calculation unit 11 inputs the generated prompt to the language generation model 142.

[0070] The language generation model 142 performs calculations and outputs language in response to the input of a prompt. The language generation model 142 outputs a definition of multiple levels of human resources according to the hiring criteria and explanations of the hiring criteria and the judgment criteria and explanations of the judgment criteria included in the input prompt.

[0071] The information processing device 1 acquires definitions of multiple levels of human resources to be judged based on the judgment criteria, corresponding to the hiring criteria and the explanation of the hiring criteria, and the judgment criteria and the explanation of the judgment criteria (S405). In S405, the calculation unit 11 acquires definitions of multiple levels of human resources output by the language generation model 142. The calculation unit 11 stores the acquired multiple levels of human resources in the storage unit 14. For example, the calculation unit 11 stores definitions of multiple levels of human resources to be judged based on the judgment criteria, in association with the hiring criteria and the explanation of the hiring criteria, and the judgment criteria and the explanation of the judgment criteria.

[0072] In S405, the calculation unit 11 acquires, for each of the multiple judgment criteria, a definition of multiple levels of talent judged based on the judgment criteria. In addition, for each of the multiple hiring criteria, the calculation unit 11 acquires a definition of multiple levels of talent corresponding to each of the multiple judgment criteria. For example, in S404, the calculation unit 11 inputs a prompt including multiple hiring criteria and each hiring criterion, as well as multiple judgment criteria and explanations of each judgment criterion, to the language generation model 142, and acquires a definition of multiple levels of talent for each hiring criterion in S405. The calculation unit 11 may repeatedly execute, for each hiring criterion, S404 of inputting a single hiring criterion and an explanation of the hiring criterion, as well as multiple judgment criteria and explanations of the judgment criterion, to the language generation model 142, and S405 of acquiring a definition of multiple levels of talent for each of the multiple judgment criteria related to the single hiring criterion.

[0073] Alternatively, the calculation unit 11 may repeatedly execute S404 of inputting a prompt including a single hiring criterion and a description of the hiring criterion and a single judgment criterion and a description of the judgment criterion to the language generation model 142, and S405 of acquiring definitions of multiple levels of talent for each of the single judgment criterion. In this case, the calculation unit 11 repeats S404 and S405 for each hiring criterion related to the single hiring criterion, and repeats the process of S404 and S405 for each hiring criterion.

[0074] 12 is a conceptual diagram showing an overview of the processing of S402 to S405. Hiring criteria and explanations of the hiring criteria are input to the language generation model 142, and multiple judgment criteria and explanations of the judgment criteria are output from the language generation model 142. In addition, hiring criteria and explanations of the hiring criteria and rejection criteria and explanations of the judgment criteria are input to the language generation model 142, and definitions of multiple levels of human resources judged based on the judgment criteria are output from the language generation model 142. The definitions of multiple levels of human resources are acquired for each judgment criterion. The definitions of multiple levels of human resources corresponding to each of the multiple judgment criteria are acquired for each hiring criterion.

[0075] The information processing device 1 then acquires the final definitions of the multiple levels of the talent judged based on the judgment criteria (S406). The final definitions of the multiple levels of the talent are determined using at least a part of the definitions of the multiple levels acquired in S405. In S406, the calculation unit 11 deletes a part of the acquired definitions of the multiple levels, and generates the final definitions of the multiple levels of the talent judged based on the judgment criteria so as to include the part of the acquired definitions of the multiple levels that has not been deleted and the definition of the predetermined level. For example, the calculation unit 11 deletes the definition of the medium level from the acquired definitions of the high, medium, low, and poor levels, changes the acquired definition of the high level to the definition of the medium level, and adds the definition of the high level that has been predetermined, thereby generating the final definitions of the multiple levels of the talent. For example, the predetermined definition of the high level is "having abilities far exceeding the definitions of other levels." The definition of the predetermined level is stored in the storage unit 14 in advance.

[0076] In S406, the calculation unit 11 may obtain the final definitions of the multiple levels of human resources according to an instruction from the user 4. For example, the calculation unit 11 displays the definitions of the multiple levels of human resources obtained in S405 on a display unit included in the output unit 16, and the user 4 operates an operation unit included in the input unit 15 to receive an instruction from the user 4. For example, a definition of a level to be deleted from the definitions of the multiple levels obtained is specified, and a definition of a specific level to be added is input. The calculation unit 11 generates the final definitions of the multiple levels of human resources according to the received instruction. In this manner, the calculation unit 11 obtains the final definitions of the multiple levels of human resources judged based on the judgment criteria. The calculation unit 11 stores the obtained final definitions of the multiple levels of human resources in the storage unit 14.

[0077] The information processing device 1 acquires the contents of the remarks made by the job applicant 5 during the job interview (S407). In S407, the calculation unit 11 acquires the contents of the remarks made by the job applicant 5 by reading out the contents of the remarks made by the job applicant 5 stored in the memory unit 14. The contents of the remarks include answers to questions asked during the job interview. The calculation unit 11 may acquire the contents of the remarks made by the job applicant 5 by receiving the contents of the remarks made by the job applicant 5 transmitted from the terminal device 2 during the job interview via the communication unit 17. The processes of S401 to S406 may be performed at a point in time prior to the job interview.

[0078] The information processing device 1 then inputs a prompt including the content of the statement, the hiring criteria and an explanation of the hiring criteria, multiple judgment criteria and an explanation of each of the judgment criteria, and a final definition of multiple levels of human resources judged based on each of the judgment criteria, and including instructions to generate positive comments, negative comments or concerns, and judgment comments for the job applicant 5 to the language generation model 142 (S408). The positive comments are comments expressing a positive evaluation of the job applicant 5. The negative comments are comments expressing a negative evaluation of the job applicant 5, and the concerns are comments expressing matters that should be of concern to the job applicant. The judgment comments are a summary of the positive comments and the negative comments or concerns.

[0079] In S408, the calculation unit 11 generates a prompt including an instruction to generate a positive comment, a negative comment or a concern, and a judgment comment according to the input content of the statement, the hiring criteria and an explanation of the hiring criteria, a plurality of judgment criteria and an explanation of each of the judgment criteria, and a final definition of a plurality of levels of talent judged based on each of the judgment criteria. The language generation model 142 performs a calculation in response to the input of the prompt, and outputs a language. The language generation model 142 outputs a positive comment, a negative comment or a concern, and a judgment comment according to the input content of the statement, the hiring criteria and an explanation of the hiring criteria, a plurality of judgment criteria and an explanation of each of the judgment criteria, and a final definition of a plurality of levels of talent judged based on each of the judgment criteria.

[0080] The information processing device 1 acquires positive comments, negative comments or concerns, and judgment comments according to the remark content, the hiring criteria and an explanation of the hiring criteria, a plurality of judgment criteria and an explanation of each of the judgment criteria, and a final definition of a plurality of levels of human resources judged based on each of the judgment criteria (S409). In S409, the calculation unit 11 acquires the positive comments, negative comments or concerns, and judgment comments output by the language generation model 142. The calculation unit 11 stores the acquired positive comments, negative comments or concerns, and judgment comments in the storage unit 14. For example, the calculation unit 11 stores the positive comments, negative comments or concerns, and judgment comments in association with the hiring criteria and an explanation of the hiring criteria.

[0081] In S409, the calculation unit 11 acquires positive comments, negative comments or concerns, and judgment comments for each of the multiple hiring criteria. For example, in S408, the calculation unit 11 inputs a prompt including multiple hiring criteria, each hiring criteria, multiple judgment criteria for each hiring criteria, an explanation of each judgment criterion, and a final definition of multiple levels of talent judged based on each judgment criterion to the language generation model 142. At this time, the calculation unit 11 acquires positive comments, negative comments or concerns, and judgment comments for each hiring criterion in S409. Alternatively, the calculation unit 11 may perform a process of acquiring positive comments, negative comments or concerns, and judgment comments for a single hiring criterion in S408 and S409, and repeatedly execute this process for each hiring criterion.

[0082] The information processing device 1 then inputs a prompt including a plurality of positive comments and a plurality of negative comments or concerns obtained for a plurality of employment criteria and an instruction to generate a comprehensive comment for the job applicant 5 to the language generation model 142 (S410). The comprehensive comment is a comment expressing a comprehensive evaluation of the job applicant 5. In S410, the calculation unit 11 generates a prompt including a first group including a plurality of positive comments and a second group including a plurality of negative comments or concerns, and including an instruction to generate a comprehensive comment according to the first group and the second group. The language generation model 142 performs calculations in response to the input of the prompt and outputs a language. The language generation model 142 outputs a comprehensive comment according to the inputted plurality of positive comments and the plurality of negative comments or concerns.

[0083] The information processing device 1 acquires a plurality of positive comments and a plurality of negative comments or an overall comment according to concerns (S411). In S411, the calculation unit 11 acquires the overall comment output by the language generation model 142. The calculation unit 11 stores the acquired overall comment in the storage unit 14.

[0084] FIG. 13 is a conceptual diagram showing an overview of the processing of S408 to S411. The content of the remark, the hiring criteria and an explanation of the hiring criteria, a plurality of judgment criteria and an explanation of each judgment criterion, and a final definition of a plurality of levels of talent judged based on each judgment criterion are input to the language generation model 142. Positive comments, negative comments or concerns, and a judgment comment for the job applicant 5 are output from the language generation model 142. In addition, a first group collecting a plurality of positive comments obtained for the plurality of hiring criteria and a second group collecting a plurality of negative comments or concerns are input to the language generation model 142, and a comprehensive comment is output from the language generation model 142. In FIG. 13, each of the plurality of hiring criteria is represented as hiring criteria 1, hiring criteria 2, .... By inputting the first group collecting a plurality of positive comments and the second group collecting a plurality of negative comments or concerns to the language generation model 142, a more appropriate comprehensive comment can be output compared to inputting each comment individually.

[0085] The information processing device 1 then inputs a prompt including the content of the statement, the hiring criteria and an explanation of the hiring criteria, a plurality of judgment criteria and an explanation of each of the judgment criteria, and a final definition of a plurality of levels of human resources judged based on each of the judgment criteria, and an instruction to generate a level of the job applicant 5 based on each of the judgment criteria, to the language generation model 142 (S412). The level of the job applicant 5 specifies which level the job applicant 5 falls into among the plurality of levels of human resources whose definitions are set for each of the judgment criteria.

[0086] In S412, the calculation unit 11 generates a prompt including an instruction to generate a level of the job applicant 5 according to the utterance content, the hiring criteria and an explanation of the hiring criteria, a plurality of judgment criteria and an explanation of each of the judgment criteria, and a final definition of a plurality of levels of talent judged based on each of the judgment criteria. The language generation model 142 performs calculations in response to the input of the prompt, and outputs language. The language generation model 142 outputs the level of the job applicant 5 according to the input utterance content, the hiring criteria and an explanation of the hiring criteria, a plurality of judgment criteria and an explanation of each of the judgment criteria, and a final definition of a plurality of levels of talent judged based on each of the judgment criteria.

[0087] The information processing device 1 acquires the level of the job applicant 5 for each judgment criterion according to the content of the statement, the hiring criteria and an explanation of the hiring criteria, the multiple judgment criteria and an explanation of each of the judgment criteria, and the final definition of the multiple levels of personnel judged based on each of the judgment criteria (S413). In S413, the calculation unit 11 acquires the level of the job applicant 5 output by the language generation model 142. The calculation unit 11 stores the acquired level of the job applicant 5 in the storage unit 14. For example, the calculation unit 11 stores the level of the job applicant 5 in association with the hiring criteria and an explanation of the hiring criteria.

[0088] In S413, the calculation unit 11 acquires the level of the job applicant 5 judged based on each of the multiple judgment criteria. That is, for one employment criterion, the level of the job applicant 5 with respect to each of the multiple judgment criteria is obtained. In addition, the calculation unit 11 acquires the level of the job applicant 5 with respect to each of the multiple judgment criteria for each of the multiple employment criteria. For example, in S412, the calculation unit 11 inputs a prompt including the multiple employment criteria and multiple judgment criteria for each employment criterion, an explanation of each judgment criterion, and a final definition of multiple levels of human resources judged based on each judgment criterion to the language generation model 142. At this time, the calculation unit 11 acquires the level of the job applicant 5 with respect to each of the multiple judgment criteria for each employment criterion in S413. Alternatively, the calculation unit 11 may perform a process of acquiring the level of the job applicant 5 with respect to each of the multiple judgment criteria for a single employment criterion in S412 and S413, and repeatedly execute this process for each employment criterion.

[0089] The information processing device 1 then calculates the level of the job applicant 5 for each employment criterion (S414). In S413, the level of the job applicant 5 for each of the multiple judgment criteria is obtained for each employment criterion. In S414, the calculation unit 11 calculates the level of the job applicant 5 for each employment criterion based on the level of the job applicant 5 for each of the multiple judgment criteria using a predetermined calculation method. For example, the calculation unit 11 expresses the level of the job applicant 5 as a numerical value, inputs the multiple level values ​​for one employment criterion into a predetermined function, and calculates the level of the job applicant 5 for one employment criterion. The calculation unit 11 calculates the level of the job applicant 5 for each employment criterion. The calculation unit 11 stores the calculated level of the job applicant 5 for each employment criterion in the memory unit 14.

[0090] The information processing device 1 then calculates an overall level of the job applicant 5 based on the levels of the job applicant 5 with respect to the multiple hiring criteria (S415). The overall level is a value of the job applicant 5 obtained by combining the levels of the job applicant 5 obtained with respect to the multiple hiring criteria, expressed in multiple stages. In S415, the calculation unit 11 calculates the overall level of the job applicant 5 based on the level of the job applicant 5 with respect to each of the multiple hiring criteria using a predetermined calculation method. For example, the calculation unit 11 expresses the level of the job applicant 5 for each hiring criterion as a numerical value, and inputs the value of the level of the job applicant 5 with respect to the multiple hiring criteria into a predetermined function to calculate the overall level of the job applicant 5. The calculation unit 11 stores the calculated overall level of the job applicant 5 in the storage unit 14.

[0091] FIG. 14 is a conceptual diagram showing an overview of the processing of S412 to S415. The content of the statement, the hiring criteria and an explanation of the hiring criteria, a plurality of judgment criteria and an explanation of each judgment criterion, and the final definition of a plurality of levels of the personnel judged based on each judgment criterion are input to the language generation model 142. The language generation model 142 outputs the level of the job applicant 5 based on each judgment criterion. In FIG. 14, each of the judgment criteria is represented as judgment criterion 1, judgment criterion 2, ..., and the level of the job applicant 5 is represented as level ( ). The level of the job applicant 5 for each of the judgment criteria is calculated. In addition, the overall level of the job applicant 5 is calculated from the level of the job applicant 5 for each of the judgment criteria.

[0092] The information processing device 1 outputs the overall comment, the judgment comments, and the overall level of the job applicant 5 (S416). In S416, the calculation unit 11 generates an image including the overall comment, the judgment comments, and the overall level of the job applicant 5, and displays the generated image on a display unit included in the output unit 16. The output unit 16 may include a printer, and the calculation unit 11 may output the overall comment, the judgment comments, and the overall level of the job applicant 5 from the printer. After S416 is completed, the information processing device 1 ends the processing.

[0093] FIG. 15 is a schematic diagram showing an example of outputting a general comment, a judgment comment, and the overall level of the job applicant 5. The general comment, the judgment comment for each employment criterion, and the overall level of the job applicant 5 are output. The user 4 can know the evaluation of the job applicant 5 obtained using the language generation model 142 by looking at the output contents. In addition, a level graph for each employment criterion is output. The level graph for each employment criterion is a graph showing the relationship between each of a plurality of employment criteria and the level of the job applicant 5 for each employment criterion. The calculation unit 11 generates the level graph for each employment criterion and outputs the level graph for each employment criterion together with the general comment, the judgment comment, and the overall level. The user 4 can visually confirm the level of the job applicant 5 for each employment criterion.

[0094] The information processing device 1 can perform a process of creating handover items related to the job applicant 5 for recording and transmitting information about the job applicant 5. FIG. 16 is a flowchart showing an example of a process procedure for creating handover items by the information processing device 1 according to the second embodiment. The information processing device 1 acquires the contents of the utterance of the job applicant 5 (S51), and inputs a prompt including the contents of the utterance and including an instruction to generate information about the job applicant 5 to the language generation model 142 (S52). The information about the job applicant 5 to be generated is information expressing the situation of the job applicant 5 contained in the utterance contents, such as the career of the job applicant 5, the reason for changing jobs, the motivation for applying, the desired conditions of the job applicant 5, and the job-changing activity situation. In S51, the calculation unit 11 generates a prompt including an instruction to extract information related to a predetermined item such as the career of the job applicant 5 from the utterance contents and generate information about the job applicant 5. The calculation unit 11 inputs the generated prompt to the language generation model 142, and the language generation model 142 performs calculations and outputs language in response to the input of the prompt.

[0095] The information processing device 1 acquires information about the job applicant 5 according to the content of the comment (S53). In S53, the calculation unit 11 acquires the information about the job applicant 5 output by the language generation model 142. The calculation unit 11 stores the acquired information about the job applicant 5 in the storage unit 14.

[0096] The information processing device 1 acquires the positive comments, negative comments or concerns, and judgment comments for each of the multiple adoption criteria (S54). In S54, the calculation unit 11 acquires the positive comments, negative comments or concerns, and judgment comments for each of the multiple adoption criteria by reading them out from the storage unit 14 and acquired in S409.

[0097] The information processing device 1 then inputs a prompt including positive comments, negative comments or concerns, and judgment comments regarding each of the multiple hiring criteria, and an instruction to generate information regarding the treatment of the job applicant 5 to the language generation model 142 (S55). The information regarding the treatment of the job applicant 5 to be generated is information summarizing future treatments to be performed on the job applicant 5. For example, the information regarding the treatment of the job applicant 5 includes the name of the future treatment to be performed on the job applicant 5, the content of the future treatment, and the reason for the treatment. In S55, the calculation unit 11 generates a prompt including an instruction to generate information regarding the treatment of the job applicant 5 according to the positive comments, negative comments or concerns, and judgment comments regarding each of the multiple hiring criteria. The calculation unit 11 inputs the generated prompt to the language generation model 142, and the language generation model 142 performs a calculation and outputs a language in response to the input of the prompt.

[0098] The information processing device 1 acquires information regarding how to treat the job applicant 5 (S56). In S56, the calculation unit 11 acquires information regarding how to treat the job applicant 5 output by the language generation model 142. The calculation unit 11 stores the acquired information regarding how to treat the job applicant 5 in the memory unit 14. The information processing device 1 may output information regarding the job applicant 5 and information regarding how to treat the job applicant 5. After S56 is completed, the information processing device 1 terminates the process of creating the handover items.

[0099] Through the processes of S51 to S56, handover matters are created, including information about the job applicant 5 and information about how to deal with the job applicant 5. For example, the handover matters are used so that a person other than the employer 4 can learn about the job applicant 5. For example, the handover matters are used as reference material for conducting further job interviews with the job applicant 5.

[0100] As described above, in the second embodiment, the information processing device 1 uses the language generation model 142 to generate an evaluation of the job applicant 5 according to the remarks of the job applicant 5 in the job interview and the hiring criteria. By obtaining an evaluation of the job applicant 5 using the language generation model 142, an objective evaluation of the job applicant 5 based on the hiring criteria and the remarks is automatically obtained. The obtained evaluation can be a reference for determining whether to hire a suitable person. In the second embodiment, the information processing device 1 generates a comment on the job applicant 5 or a level expressing the value of the job applicant 5 in multiple stages as the evaluation of the job applicant 5. This makes it easier for the user 4 to understand the evaluation of the job applicant 5 obtained using the language generation model 142. Therefore, in the second embodiment as well, the information processing device 1 can support a job interview so that a suitable person can be hired or so that the necessary work can be reduced.

[0101] The present invention is not limited to the contents of the above-described embodiment, and various modifications are possible within the scope of the claims. In other words, embodiments obtained by combining technical means appropriately modified within the scope of the claims are also included in the technical scope of the present invention.

[0102] The matters described in each embodiment can be combined with each other. In addition, the independent claims and dependent claims described in the claims can be combined with each other in any and all combinations regardless of the citation format. Furthermore, the claims use a format in which a claim cites two or more other claims (multiple claim format), but this is not limited to this. A multiple claim (multi-multi claim) that cites at least one multiple claim may also be used. [Explanation of symbols]

[0103] 100 Interview Support System 1. Information processing device 10 Recording media 11 Arithmetic section 14 Storage section 141 Computer Programs 142 Language Generation Model 2 Terminal Device 3. Communication Network 4 User 5 Applicants for employment

Claims

1. Obtain information about recruitment criteria, Using a language generation model, obtain hiring criteria according to the obtained information on the hiring criteria; Using the language generation model, acquire information on the hiring criteria and questions to ask the job applicant in accordance with the judgment criteria; Output the obtained question A computer program that causes a computer to execute a process.

2. Accept instructions for modifying the criteria or the questions; Modifying the judgment criteria or the question in accordance with the received instruction and outputting the modified judgment criteria or the question.

2. The computer program product according to claim 1, which causes a computer to execute a process.

3. As information regarding the hiring criteria, hiring criteria for personnel at the establishment and definitions of the hiring criteria are obtained.

2. The computer program product according to claim 1, which causes a computer to execute a process.

4. A prompt including the acquired adoption criteria, the definition, and an instruction for generating a plurality of the judgment criteria is input to the language generation model, and the plurality of the judgment criteria output by the language generation model are acquired.

4. The computer program according to claim 3, which causes a computer to execute a process.

5. inputting the hiring criteria and the definition, the acquired plurality of judgment criteria, and a prompt including an instruction to generate a plurality of questions for the job applicant to the language generation model, and acquiring a plurality of questions output by the language generation model; Output the acquired questions 5. The computer program according to claim 4, which causes a computer to execute a process.

6. Obtaining answers to said questions; using the language generation model to obtain additional questions corresponding to the question, the answer, and the judgment criteria; Output the obtained additional question 2. The computer program product according to claim 1, which causes a computer to execute a process.

7. The question, the answer, the criterion, and a prompt including an instruction for generating the follow-up question are input to the language generation model, and the follow-up question is output by the language generation model.

7. A computer program according to claim 6, which causes a computer to execute a process.

8. Obtaining answers to said questions; Using the language generation model, an evaluation of the job applicant is obtained according to information on the hiring criteria, the questions, and the answers; Output the obtained evaluation 2. The computer program product according to claim 1, which causes a computer to execute a process.

9. Obtain information on recruitment criteria for personnel, Using a language generation model, obtain hiring criteria according to the obtained information on the hiring criteria; Using the language generation model, acquire information on the hiring criteria and questions to ask the job applicant in accordance with the judgment criteria; Output the obtained question 23. An information processing method comprising:

10. A calculation unit is provided, The calculation unit is Obtain information about recruitment criteria, Using a language generation model, obtain hiring criteria according to the obtained information on the hiring criteria; Using the language generation model, acquire information on the hiring criteria and questions to ask the job applicant in accordance with the judgment criteria; Output the obtained question 23. An information processing apparatus comprising:

11. Obtain information about recruitment criteria, Using a language generation model, obtain information regarding hiring criteria according to the obtained information regarding the hiring criteria; Using the language generation model, obtain definitions of a plurality of levels of human resources to be judged based on the judgment criteria according to information on the hiring criteria and information on the judgment criteria; Obtain the statements of job applicants, Using the language generation model, an evaluation of the job applicant is obtained according to the acquired utterance content, information on the hiring criteria, information on the judgment criteria, and a final definition of the plurality of levels determined using at least a part of the acquired definitions of the plurality of levels. A computer program that causes a computer to execute a process.

12. for each of the plurality of hiring criteria, using the language generation model, in accordance with the content of the remark, information on the hiring criteria, information on the judgment criteria, and a final definition of the plurality of levels, the final definition being determined using at least a portion of the definitions of the plurality of levels, to obtain positive comments expressing a positive evaluation of the job applicant, and negative comments expressing a negative evaluation of the job applicant or concerns about the job applicant; Using the language generation model, a comprehensive comment expressing a comprehensive evaluation of the job applicant is obtained in accordance with the positive comments regarding the plurality of hiring criteria and the negative comments or concerns.

12. A computer program product according to claim 11, which causes a computer to execute a process.

13. obtaining a definition of said plurality of levels for each of said plurality of criteria; obtaining a level of the job applicant based on the judgment criteria for each of the judgment criteria using the language generation model in accordance with the content of the remark, information on the hiring criteria, information on the judgment criteria, and a final definition of the plurality of levels that is determined using at least a part of the definitions of the plurality of levels; Calculating a level of the job candidate based on the hiring criteria based on the level of the job candidate with respect to the plurality of criteria.

12. A computer program product according to claim 11, which causes a computer to execute a process.

14. For each of a plurality of the hiring criteria, calculating a level of the job applicant based on the hiring criteria; Calculating an overall level of the job candidate based on the job candidate's levels with respect to a plurality of the hiring criteria 14. A computer program product according to claim 13, which causes a computer to execute a process.

15. The final definition of the plurality of levels includes a part of the definition of the plurality of levels and a definition of a predetermined level.

15. A computer program according to any one of claims 11 to 14.

16. Obtain information about recruitment criteria, Using a language generation model, obtain information regarding hiring criteria according to the obtained information regarding the hiring criteria; Using the language generation model, obtain definitions of a plurality of levels of human resources to be judged based on the judgment criteria according to information on the hiring criteria and information on the judgment criteria; Obtain the contents of statements made by job applicants, Using the language generation model, an evaluation of the job applicant is obtained according to the acquired utterance content, information on the hiring criteria, information on the judgment criteria, and a final definition of the plurality of levels determined using at least a part of the acquired definitions of the plurality of levels.

23. An information processing method comprising:

17. A calculation unit is provided, The calculation unit is Obtain information about recruitment criteria, Using a language generation model, obtain information regarding hiring criteria according to the obtained information regarding the hiring criteria; Using the language generation model, obtain definitions of a plurality of levels of human resources to be judged based on the judgment criteria according to information on the hiring criteria and information on the judgment criteria; Obtain the contents of statements made by job applicants, Using the language generation model, an evaluation of the job applicant is obtained according to the acquired utterance content, information on the hiring criteria, information on the judgment criteria, and a final definition of the plurality of levels determined using at least a part of the acquired definitions of the plurality of levels.

23. An information processing apparatus comprising:

Images