Information processing systems, information processing methods, and programs

Abstract

To provide an information processing system that enables more efficient acquisition of responses from job candidates. [Solution] According to one aspect of the present invention, an information processing system is provided, comprising at least one processor, the processor being configured to perform the following steps by reading a program, the receiving step receiving a first condition concerning the type of purpose of a question as a condition for a first question presentation means from a recruiter engaged in recruitment activities, the presentation step presenting the first question presentation means to a candidate involved in recruitment activities, where the first question presentation means presents a question to the candidate, the content of the question in the first question presentation means is controlled based on the received first condition, and the memory management step storing the candidate's answer to the question in the first question presentation means in association with the candidate.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a technique capable of conducting a question-and-answer session with an applicant.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] On the other hand, there is still room for improvement in techniques for obtaining higher-quality answers.

[0005] In view of the above circumstances, the present invention aims to provide an information processing system and the like that can more efficiently obtain answers from candidates for adoption.

Means for Solving the Problems

[0006] According to one aspect of the present invention, an information processing system is provided, comprising at least one processor, the processor configured to perform the following steps by reading a program, wherein in the reception step, a first condition concerning the type of purpose of a question is received from a recruiter conducting recruitment activities, as a condition for a first question presentation means; in the presentation step, the first question presentation means is presented to a candidate for employment related to the recruitment activities, where the first question presentation means presents a question to the candidate for employment, the content of the question in the first question presentation means is controlled based on the received first condition; and in the memory management step, the information processing system is provided, which stores the candidate's answer to the question in the first question presentation means in association with the candidate.

[0007] This configuration provides an information processing system that can more efficiently obtain responses from job candidates. [Brief explanation of the drawing]

[0008] [Figure 1] This is a diagram showing the configuration of Information Processing System 1. [Figure 2] This is a block diagram showing the hardware configuration of server device 10. [Figure 3] This is a block diagram showing the hardware configuration of user terminal 20. [Figure 4] This is a block diagram showing the functions implemented by the server device 10 (control unit 11) and the user terminal 20 (control unit 21). [Figure 5] This is an activity diagram illustrating the information processing of this embodiment. [Figure 6] This is an example of a screen that may be displayed on a recruiter's terminal. [Figure 7] This is an example of a screen that may be displayed on a recruiter's terminal. [Figure 8] This is an example of a screen that may be displayed on a recruiter's terminal. [Figure 9] This is an example of a screen that may be displayed on a recruiter's terminal. [Figure 10] This is an example of a screen that may be displayed on a candidate's device. [Figure 11] This is an example of a screen that may be displayed on a candidate's device. [Figure 12] This is an example of a screen that may be displayed on a candidate's device. [Figure 13] This is an example of a screen that may be displayed on a candidate's device. [Figure 14] This is an example of a screen that may be displayed on a recruiter's terminal. [Modes for carrying out the invention]

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

[0010] In other words, the information processing system of this embodiment is as follows. An information processing system, Equipped with at least one processor, The aforementioned processor is configured to perform the following steps by reading a program: In the registration step, recruiters conducting recruitment activities will be asked to provide the first condition regarding the type of question concerning its purpose, as a condition for the first question presentation method. In the presentation step, the first question presentation means is presented to the candidate for employment related to the recruitment activities of the aforementioned personnel. Here, the first question presentation means presents questions to the job candidate, The content of the question in the first question presentation means is controlled based on the first condition received. In the memory management step, the information processing system stores the answers of the job candidates to the questions presented by the first question presentation means, associating them with the job candidates.

[0011] Incidentally, the program for implementing the software appearing in one embodiment may be provided as a non-transitory computer-readable medium readable by a computer, may be provided so as to be downloadable from an external server, or may be provided so that the program is started on an external computer to realize its functions on a client terminal (so-called cloud computing).

[0012] Also, in various information processes according to one embodiment, an input and an output corresponding to the input can be realized. Here, if an output is obtained as a result of the input, the mode of information (hereinafter referred to as reference information) referred to in such information processing is not limited. The reference information may be, for example, rule-based information such as a database, a lookup table, a predetermined function (including a judgment formula such as a regression formula constructed by a statistical method), a learned model in which the correlation between the input and the output has been learned in advance, or a large language model capable of outputting a desired result by inputting a prompt (these models include parameters for constructing the correlation between the input and the output), or a generative AI such as a vision language model.

[0013] Also, in one embodiment, the "unit" may include, for example, hardware resources implemented by a circuit in a broad sense and information processing of software that can be specifically realized by these hardware resources. Also, in one embodiment, various types of information are handled, and these information are represented, for example, by physical values of signal values representing voltage and current, the level of signal values as a set of binary bits composed of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculation can be executed on a circuit in a broad sense.

[0014] Furthermore, a circuit in a broad sense is a circuit realized by combining at least a suitable combination of circuits, circuits, processors, and memory. The processor may be a general-purpose processor or a dedicated circuit. In other words, it includes application-specific integrated circuits (ASICs), programmable logic devices (for example, simple programmable logic devices (SPLDs), complex programmable logic devices (CPLDs), and field programmable gate arrays (FPGAs)), etc.

[0015] 1. Hardware Configuration This section describes the hardware configuration.

[0016] <Information Processing System 1> Figure 1 is a configuration diagram representing information processing system 1. The information processing system 1 shown as an example in Figure 1 comprises a communication line 2, a server device 10, and a plurality of user terminals 20. The server device 10 and the plurality of user terminals 20 are configured to communicate with each other via the communication line 2. Although Figure 1 shows one user terminal 20 used by a recruiter U1 and one user terminal 20 used by a job candidate U2, there may be multiple recruiters U1 and multiple job candidates U2. Furthermore, the connection between the server device 10 and the plurality of user terminals 20 may be wired or wireless.

[0017] Recruiter U1 may be one or more users who conduct recruitment activities. In one embodiment, Recruiter U1 may be an employer who recruits personnel. In this embodiment, "employer" may include organizations such as for-profit corporations (e.g., companies), non-profit corporations (e.g., cooperatives, foundations, etc.), and public corporations (e.g., local governments, etc.). Furthermore, "employer" may be a person in charge (recruitment officer, etc.) of the aforementioned types of corporations, and may include a person in charge of the organization's human resources department or a person in charge of the department that recruits personnel. In addition, Recruiter U1 may include persons who support the recruitment activities of the aforementioned employer. Examples of persons who provide support here include recruitment agencies. In this embodiment, "recruitment agency" is a person who supports the recruitment activities of employers, etc., and may be an organization or a person in charge thereof that mediates the communication between job candidates U2 (described later) and employers. Such recruitment agencies are sometimes also called recruitment agencies, headhunters, agents, etc.

[0018] Candidate U2 may be one or more users equivalent to candidates in recruitment activities. In one embodiment, Candidate U2 may be a job seeker. A job seeker refers to various people who are looking for work, including, for example, currently employed people (those seeking a career change), unemployed people who wish to find work, and prospective graduates (job seekers). Candidate U2 can apply for job postings from the aforementioned employers, etc., or conduct interviews with recruiters U1, etc., in order to be hired. In addition, Candidate U2 can receive information about job postings, etc., from recruiters U1, etc., through scout messages, etc., and consider applying for relevant job postings. In this embodiment, "interview" may be appropriately rephrased as a meeting, interview, dialogue, question and answer session, etc.

[0019] Furthermore, the user terminals 20 used by recruiter U1 and candidate U2 may be referred to as "recruiter terminal," "candidate terminal," etc.

[0020] In one embodiment, the information processing system 1 may provide a platform used by recruiters U1 and job candidates U2. Such a platform may also be called a platform for talent acquisition. For example, the information processing system 1 provides and manages a talent matching platform or talent matching services used by recruiters U1 and job candidates U2. That is, from the perspective of recruiter U1, recruitment activities of employers, etc., are supported via the server device 10, so it can be said that recruitment support services are provided. Also, from the perspective of job candidate U2, their job-seeking activities are supported, so it can be said that job-seeking support services are provided.

[0021] In another embodiment, the information processing system 1 may provide a platform to support human resource management (employee management) and the formulation of personnel strategies within an organization. In such processes of human resource management and the formulation of personnel strategies, recruiters U1 and job candidates U2 can store (register) various information in the server device 10 via their own terminals (user terminals 20). In a typical embodiment, recruiter U1 may be a person belonging to a predetermined organization, and job candidate U2 may be a person belonging to the same organization (employee, etc.). In such a case, various activities can be carried out to recruit personnel within the organization. Furthermore, when such a platform is provided, various information may be sent and received between recruiter U1 and job candidate U2.

[0022] In one embodiment, the information processing system 1 consists of one or more devices or components. For example, the information processing system 1 may include a server device (e.g., server device 10) having a processor (e.g., control unit 11) and a terminal (e.g., user terminal 20) that can access the server. These components will be described below.

[0023] <Server device 10> Figure 2 is a block diagram showing the hardware configuration of the server device 10. As shown in Figure 2, the server device 10 comprises a control unit 11, a storage unit 12, a communication unit 13, and a communication bus 14. The control unit 11, the storage unit 12, and the communication unit 13 are electrically connected within the server device 10 via the communication bus 14.

[0024] <Control Unit 11> The control unit 11 performs processing and control of the overall operation related to the server device 10. The control unit 11 is, for example, a Central Processing Unit (CPU). The control unit 11 realizes various functions related to the server device 10 by reading predetermined programs stored in the memory unit 12. That is, information processing by software stored in the memory unit 12 is concretely realized by the control unit 11, which is an example of hardware, and can be executed as each functional unit included in the control unit 11. These will be described in more detail in the next section. Note that the control unit 11 is not limited to being a single unit, and the server device 10 may have multiple control units 11 for each function. The server device 10 may also be composed of a combination of these.

[0025] <Storage section 12> The storage unit 12 stores various types of information as defined above. This can be done, for example, as a storage device such as a solid-state drive (SSD) that stores various programs related to the server device 10 executed by the control unit 11, or as memory such as random access memory (RAM) that stores temporarily necessary information (arguments, arrays, etc.) related to program calculations. The storage unit 12 stores various programs, variables, etc. related to the server device 10 executed by the control unit 11.

[0026] <Communications Department 13> The communication unit 13 preferably uses wired communication methods such as USB, IEEE1394, Thunderbolt®, and wired LAN network communication, but may also include wireless LAN network communication, mobile communication such as LTE / 5G, and Bluetooth® communication as needed. In other words, it is more preferable to implement it as a collection of these multiple communication methods. That is, the server device 10 may communicate various information from the outside via the communication unit 13 and the network.

[0027] The server device 10 may be on-premises or in a cloud environment. A cloud-based server device 10 may provide the above-mentioned functions and processing in the form of, for example, SaaS (Software as a Service) or cloud computing.

[0028] <User terminal 20> Figure 3 is a block diagram showing the hardware configuration of the user terminal 20. This user terminal 20 is the terminal used by the various users described above. As shown in Figure 3, the user terminal 20 comprises a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, an output unit 25, and a communication bus 26. The control unit 21, storage unit 22, communication unit 23, input unit 24, and output unit 25 are electrically connected within the user terminal 20 via the communication bus 26. The explanation of the control unit 21, storage unit 22, and communication unit 23 is the same as the explanation of each part in the server device 10, so it will be omitted.

[0029] <Input section 24> The input unit 24 receives operation inputs made by the user. The operation inputs are transmitted as command signals to the control unit 21 via the communication bus 26. The control unit 21 can perform predetermined controls or calculations based on the transmitted command signals as needed. The input unit 24 may be included in the casing of the user terminal 20 or it may be an external component. For example, the input unit 24 may be implemented as a touch panel integrated with the output unit 25. When the input unit 24 is implemented as a touch panel, the user can input tap operations, swipe operations, etc. to the input unit 24. Instead of a touch panel, the input unit 24 can be a switch button, mouse, trackpad, QWERTY keyboard, etc.

[0030] <Output section 25> The output unit 25 displays a graphical user interface (GUI) screen that can be operated by the user. The output unit 25 may be included in the casing of the user terminal 20 or it may be an external component. Specifically, the output unit 25 can be implemented as a display device such as a CRT display, liquid crystal display, organic EL display, or plasma display. It is preferable that these display devices be used in accordance with the type of user terminal 20.

[0031] Although Figure 1 shows an example where the user terminal 20 is a laptop PC (Personal Computer), the type of terminal used by the user terminal 20 is not particularly limited in this embodiment. That is, each user terminal 20 may be a desktop PC, laptop PC, smartphone, tablet, or any other type of information processing terminal.

[0032] 2. Functional Configuration This section describes the functional configuration of this embodiment. Information processing by software stored in the memory unit 12 is specifically realized by the control unit 11, which is an example of hardware, and can be executed as each functional unit included in the control unit 11 (the processor provided by the information processing system 1).

[0033] Figure 4 is a block diagram showing the functions realized by the server device 10 (control unit 11) and the user terminal 20 (control unit 21).

[0034] As shown in Figure 4A, the server device 10 (control unit 11) may include a registration unit 110, a reception unit 111, a presentation unit 112, a memory management unit 113, a display control unit 114, a notification unit 115, a generation unit 116, and an artificial intelligence unit 117. As shown in Figure 4B, the user terminal 20 (control unit 21) may include a display control unit 210 and an operation reception unit 211.

[0035] <Registration Section 110> The registration unit 110 is configured to execute the registration step. In the registration step, the registration unit 110 performs user registration and various information registration for the platform provided by the information processing system 1. In this embodiment, the registration unit 110 registers one or more of the recruiters U1 and job candidates U2 as users of the platform. In this embodiment, the registration unit 110 may also register information indicating the career history and skills of the job candidate U2 in association with the user (job candidate U2). Such career history and skills may be registered through work history information, for example, a resume or work history document. Here, the work history document may be called a resume (resume document). Furthermore, in this embodiment, the registration unit 110 may register information regarding job postings on the platform based on terminal operations of the recruiter U1, etc. Here, information regarding a job posting may include, for example, information that could be written on a job posting, and may include information such as job title, annual salary, industry, job level, required skills, employment type, work location, working hours, holidays, corporate culture, job description, and allowances. Furthermore, information regarding a job posting may also include information such as the location, number of employees, performance, and corporate culture of the employer providing the job posting. The registration unit 110 can also register the contents set for the job posting (for example, a template combining the contents of the questions described later) and the status of the candidate U2 regarding the job posting (such as the progress of recruitment). The information registered by the registration unit 110 may be stored in a predetermined memory area based on the functions of the memory management unit 113 described later.

[0036] <Reception Desk 111> The reception unit 111 is configured to execute the reception step. In the reception step, the reception unit 111 receives various information related to the information processing system 1. In this embodiment, the reception unit 111 receives a first condition regarding the type of purpose of the question from the recruiter U1 who is conducting recruitment activities, as a condition for the first question presentation means. Details of the information received by the reception unit 111 will be explained later.

[0037] <Presentation part 112> The presentation unit 112 is configured to perform the presentation step. In the presentation step, the presentation unit 112 presents various information to the user associated with the information processing system 1. This information may be presented based on visual information or auditory information. In this embodiment, the presentation unit 112 presents the first question presentation means and the second question presentation means to the job candidate U2. Details of the processing performed by the presentation unit 112 will be described later.

[0038] <Storage management section 113> The memory management unit 113 is configured to execute the memory management step. In the memory management step, the memory management unit 113 manages the storage state of various information related to the information processing system 1. Typically, the memory management unit 113 can be configured to store information handled by the server device 10, various terminals, etc., in a memory area. This memory area is exemplified by the memory area (storage unit 12) of the server device 10 or the memory areas of various devices, but this memory area does not necessarily have to be within the system shown in Figure 1, and the memory management unit 113 can also store various information in an external memory device or the like. In the example of this embodiment, the memory management unit 113 stores the answers of the candidate U2 to the questions of the first question presentation means and the answers to the questions of the second question presentation means, associating them with the candidate U2.

[0039] <Display Control Unit 114> The display control unit 114 is configured to execute the display control step. In the display control step, the display control unit 114 controls whether or not visual information can be displayed on each user terminal 20. In the display control step, the display control unit 114 also generates various display information and controls it so that content that can be seen by the user is displayed. The display information may be the information itself that is generated in a manner that can be seen by the user, such as a screen, image, icon, or text, or it may be rendering information for displaying a screen, image, icon, text, etc. on various terminals. The content that can be displayed on the user terminal 20 will be explained later.

[0040] <Notification section 115> The notification unit 115 is configured to execute a notification step. In the notification step, the notification unit 115 provides various notifications to various users related to the information processing system 1. Such notifications may be based on visual information or auditory information. In this embodiment, the notification unit 115 provides a predetermined notification to the recruiter U1 when the answer to the first question presentation means (and / or the answer to the second question presentation means) satisfies predetermined conditions or does not satisfy predetermined conditions. Details of the content of such notifications from the notification unit 115 will be explained later.

[0041] <Generation unit 116> The generation unit 116 is configured to execute the generation step. In the generation step, the generation unit 116 generates various information based on the received information, etc. In this specification, the term "generation" may be replaced with terms such as "creation" as appropriate. Details of the content generated by the generation unit 116 will be explained later.

[0042] <Artificial Intelligence Department 117> The artificial intelligence unit 117 is configured to receive input from each functional unit and return the instructed output, thus constituting an artificial intelligence module. The artificial intelligence used by the server device 10 in each functional unit may be common to all units, or it may be individually prepared for each functional unit. Furthermore, the artificial intelligence unit 117 may function as a core module that supports the artificial intelligence processing in each of these functional units.

[0043] The artificial intelligence unit 117 may be an AI (Artificial Intelligence) equipped with trained models such as transformers including GPT (Generative Pretrained Transformer, including GPT-1 to GPT-5), BERT (Bidirectional Encoder Representations from Transformers), BART (Bidirectional and Auto-regressive Transformer), and language models such as recurrent neural networks (RNN). The artificial intelligence unit 117 may be, for example, a general-purpose learning model including various language models, large-scale language models, and generative AI, or an AI agent, and may include specific models such as OpenAI's GPT (registered trademark), Google's Gemini (registered trademark), and models provided through services and platforms such as Microsoft's Azure (registered trademark) AI Studio. Generative AI may be, for example, text generation AI, image generation AI, multimodal generation AI, etc. The trained model may be called an artificial intelligence model, machine learning model, or deep learning model. In addition, the artificial intelligence unit 117 can include any pre-trained model.

[0044] Specific machine learning algorithms used to build trained models include nearest neighbors, naive Bayes, decision trees, support vector machines, and deep learning using neural networks. The artificial intelligence unit 117 can apply these algorithms as appropriate.

[0045] The artificial intelligence unit 117 may have a trained model constructed by a learning method such as supervised learning, unsupervised learning, or self-supervised learning. In supervised learning, machine learning is performed using training data. Training data consists of pairs of input data and output data (correct answer data) for training. Furthermore, the trained model may not only be one trained for a specific task, but also a general-purpose learning model that can be used universally for a wide range of tasks.

[0046] The artificial intelligence unit 117 may include a natural language model as its artificial intelligence, or it may be a general-purpose learning model such as a Large Language Model (LLM). An LLM is a learning model that has been pre-trained on a large amount of large-scale data consisting of text data, etc. (for example, (i) web content on the internet, or (ii) data stored in a predetermined database), and can perform various language processing tasks by being given a task. According to the given prompt, it can perform a wide range of natural language processing tasks, such as understanding sentence patterns and context, responding to questions, and generating sentences. Such a general-purpose learning model may include a pre-trained model that can handle various tasks without fine-tuning by One-shot Learning or Few-shot Learning. Furthermore, the general-purpose learning model may also be configured to handle various tasks by Zero-shot Learning. The artificial intelligence used in each functional unit of the control unit 11 may be a separate pre-trained model, or it may be a common general-purpose pre-trained model. In addition, the artificial intelligence unit 117 may include a small-scale language model or a medium-scale language model that is smaller in scale than a large-scale language model as a pre-trained model. Small-scale and medium-scale language models are natural language processing models that are trained on less data (and constructed with fewer parameters) compared to large-scale language models.

[0047] The pre-trained models included in the artificial intelligence unit 117 (pre-trained models used in each functional unit) can undergo additional training using methods such as transfer learning and fine-tuning. For example, whenever new data is registered, the artificial intelligence unit 117 may perform additional training and fine-tuning using this new data as training data. This improves the accuracy of the information output from the pre-trained models.

[0048] The trained model included in the artificial intelligence unit 117 may be a trained model (distilled model) obtained by knowledge distillation using the original trained model. In knowledge distillation, a trained model such as a large-scale language model is used as the teacher model, and the student model is trained by adjusting the parameters of the student model so that the loss of the student model's output (soft target loss) relative to the teacher model's output (soft target) is small, and that student model becomes the distilled model. Alternatively, the student model may be trained so that the loss of the student model's output (hard target loss) relative to the correct labels (hard target) of the teacher data (combination of input data and output data of the trained model) is small. Compared to the original trained model (teacher model), the distilled model has performance close to that of the trained model, but with fewer parameters and a lower processing load. Therefore, by using the distilled model, the cost of the information processing system 1 can be reduced.

[0049] For example, the trained model used in each functional unit may be a distilled model trained using combinations of input and output data from a large-scale language model as training data. Alternatively, when the information processing system 1 is introduced, a large-scale language model may be used as the trained model in each functional unit, and once training data from the large-scale language model has been accumulated, the distilled model obtained by knowledge distillation using that training data may be used as the trained model in each functional unit.

[0050] An AI agent (also called an autonomous agent) is a system, program, or function that autonomously determines and executes the processes necessary to achieve a goal or accomplish a task, in response to a goal (objective, purpose, etc.) such as "teach me about XX" or a task such as "output XX" input by a user. To achieve a goal or task, the AI ​​agent may break down the task into subtasks or actions, and may perform at least one of the following: collecting and analyzing necessary data, using external tools, generating and executing programs, etc. The AI ​​agent uses information and instructions input by the user, or information obtained from external sources, etc., as goals and decision-making factors, and autonomously selects and executes tasks, actions, etc., according to the goal, and outputs information according to the goal, without requiring user intervention (operation input). However, the AI ​​agent may request confirmation or feedback from the user as needed. Furthermore, the AI ​​agent may autonomously plan and execute, evaluate the execution results itself, and autonomously perform learning (including in-context learning, etc.) aimed at achieving goals or improving the accuracy of achievement. For example, an AI agent may autonomously update its own action plan, memory, or knowledge based on the results of its subtask execution (e.g., collected information, results of information analysis, etc.). Furthermore, the AI ​​agent may be configured as a multi-agent system comprising multiple AI agents. In this case, the multiple AI agents may have different roles, capabilities, or access privileges. The multiple AI agents may cooperate to solve tasks by communicating (dialogue) with each other using natural language or a predetermined protocol. The cooperation between each AI agent is not limited to a hierarchical structure (superior-subordinate relationship); multiple AI agents may autonomously discuss and vote, and the final output may be determined by consensus.

[0051] <Display Control Unit 210> The display control unit 210 of the user terminal 20 controls the display to show the screen indicated by the screen data transmitted from the server device 10.

[0052] <Operation reception unit 211> The operation reception unit 211 of the user terminal 20 receives operations from users using the user terminal 20 (such as recruiter U1, job candidate U2, etc.).

[0053] 3. Information Processing Methods This section describes the information processing method for the server device 10, with examples. This information processing method may be executed by each part of the server device 10 as individual steps. The various features shown in this section can be combined with each other as long as they do not create technical inconsistencies.

[0054] Figure 5 is an activity diagram illustrating the information processing of this embodiment. As shown in the activity diagram in Figure 5, in the information processing method of this embodiment, a predetermined question presentation means is presented to the terminal of the job candidate U2, and the answer to this question presentation means is stored by the server device 10.

[0055] The content of the questions presented by such a question presentation means can be set, for example, by recruiter U1. In a typical embodiment, recruiter U1 requests the display of a settings screen for configuring the question presentation means, and in response, the display control unit 114 of the server device 10 displays the settings screen for configuring the question presentation means on recruiter U1's terminal (Activities A101 to A104).

[0056] In this embodiment, at least a question based on the first question presentation means is presented to the candidate U2. The first question presentation means here typically presents a question to the candidate U2, but its form can be set in various ways. In one embodiment, the first question presentation means may be a means of presenting a question to the candidate U2 in the form of an utterance. The first question presentation means may also receive a response from the candidate U2 in the form of an utterance. By communicating in this form of an utterance, the candidate U2 can more easily express their thoughts and opinions. Furthermore, the first question presentation means is not limited to the form of an utterance, and may also be a means of presenting a question in the form of a text chat that sends and receives text information in the form of a timeline, for example. In this case, the first question presentation means may receive a response from the candidate U2 in the form of text input via a keyboard or touch panel. In the case of a text chat, the candidate U2 can send a response while visually confirming the question and their own answer. More specifically, the first question presentation means may be a means of displaying an avatar or live-action video of an interviewer on the screen, and presenting questions via audio or subtitle text in accordance with the movements and facial expressions of the avatar, etc. By incorporating such visual elements, it is possible to give the candidate U2 a sense of realism similar to an actual interview. Furthermore, from this perspective, the first question presentation means may also be referred to as "interview means," "interview," "interview part," "interview section," "interview mode," "interview means," "interview," "interview part," "interview section," "dialogue section," etc., for the candidate U2.

[0057] Furthermore, from the perspective of conducting more efficient recruitment activities, the first question presentation means described here may present a question to candidate U2 and generate a new question in response to candidate U2's answer, which is received in the form of an utterance. Alternatively, the first question presentation means may display a predetermined avatar on candidate U2's terminal, and this predetermined avatar may present a question to candidate U2. Here, a means of presenting a question via a predetermined avatar (for example, an avatar modeled after an interviewer) is described, but when the first question presentation means generates a new question, the generated question may be presented by a different means. For example, the question may be presented via a live video of the person acting as the interviewer. Also, for example, in a scenario where recruiter U1 and candidate U2 conduct an online interview, the first question presentation means may present a new question to recruiter U1. In such a case, recruiter U1 may visually confirm the new question generated by the first question presentation means and then ask the question to candidate U2. In other words, the first question presentation means does not necessarily have to present questions directly to the candidate U2. For example, the generated questions may be presented to the candidate U2 through personnel associated with the first question presentation means, such as the recruitment officer U1.

[0058] Furthermore, in this embodiment, a second question presentation means may also be presented to the candidate U2. This second question presentation means may present a response operation screen on the candidate U2's terminal for inputting answers to the questions. From this perspective, the second question presentation means may also be referred to as a "survey means," "survey," "survey part," "survey section," "survey mode," "survey," "questionnaire," "answer form," etc., for the candidate U2. More specific examples of how the first and second question presentation means are presented will be explained in detail later with reference to Figures 10 to 12, etc.

[0059] In this embodiment, questions that are preferable for candidate U2 to answer via the operation screen may be presented to candidate U2 via the second question presentation means. The recruitment agent U1 may, through the first question presentation means, present questions to candidate U2 that they want to confirm in detail, that they want to confirm in more depth, or that they want to confirm through interaction in the form of speech, etc. (for example, matters that require further investigation through interaction with candidate U2), such as an evaluation of candidate U2's inner self and thought process. In response, the recruitment agent U1 may present questions to candidate U2 via the second question presentation means that can be answered in a businesslike or formal manner (for example, standard confirmation items), that can be answered selectively, or that they want to aggregate statistically. Specifically, the second question presentation means may present questions that can be answered in a binary "yes / no" format, matters concerning working conditions such as possible start date and desired annual salary, or standard matters that HR will confirm in a businesslike manner during a typical interview. Furthermore, for example, the questions in the second question presentation means may include questions concerning at least one of the following: the candidate U2's level of interest in recruiter U1, and the candidate U2's selection status with other recruiters. These questions often require formal answers, and it is often preferable for candidate U2 to answer them via an answer screen. By presenting essential confirmation items in the second question presentation means in this way, the interview time (time for verbal communication) in the first question presentation means can be used more effectively. For example, the questions (interview time) using the first question presentation means can be made more conducive to making a fundamental assessment of candidate U2. At the same time, by making answers mandatory in the second question presentation means, it becomes possible to prevent candidates from missing answers to confirmation items.

[0060] Here, the first question presentation means and the second question presentation means may be provided through different means (interfaces). Specifically, for example, the first question presentation means may be provided through an auditory interface (dialogue format) using voice output and input, while the second question presentation means may be provided through a visual interface (operation format) such as screen display on a display. In this way, by presenting questions using appropriate means (interfaces) depending on the content of the questions, the burden on the candidate U2 to answer can be reduced compared to conducting everything in a dialogue format. In addition, the recruiter U1 can acquire standardized information as text or numerical data, thus improving the efficiency of the selection process.

[0061] Let's return to the setting operations for the various question presentation means. The questions of the first question presentation means and the questions of the second question presentation means, as described above, can be set according to the operation of the setting screen displayed by the display control unit 114 of the server device 10.

[0062] Figure 6 is an example of a screen that may be displayed on the recruiter's terminal. Figure 7 is an example of a screen that may be displayed on the recruiter's terminal. Figure 8 is an example of a screen that may be displayed on the recruiter's terminal. As shown in these drawings, the setting screen displayed by the display control unit 114 may have a first setting area where the questions of the first question presentation means can be set, and a second setting area where the questions of the second question presentation means can be set. The questions of the first question presentation means and the questions of the second question presentation means can each be set based on operations performed by the recruiter U1 on the first setting area and the second setting area, respectively. Here, the screens where settings for the first question presentation means (interview part) can be made are shown in Figure 6 and Figure 7, and the screen where settings for the second question presentation means (questionnaire part) can be made is shown in Figure 8, and the two setting areas are shown separated by screens. On the other hand, both setting areas may be provided on the same screen.

[0063] The screen shown in Figure 6 is configured to include Form F1 for specifying the recruitment category, Form F2 for specifying a template related to the question, and Forms F3 to F6 for specifying the items of the question to be presented by the first question presentation means.

[0064] Form F1 is configured to allow the selection of the recruitment category of candidate U2, such as "new graduate / mid-career." In one embodiment, the way questions are presented by the question presentation means described later can be controlled based on the attributes of candidate U2. That is, in one embodiment, the reception unit 111 of the server device 10 can accept conditions (second condition) regarding the attributes of candidate U2 to whom questions are asked. The attributes of candidate U2 here may include at least information regarding the recruitment category of candidate U2. Note that the recruitment category here may be a category such as "new graduate / mid-career" as shown in Figure 6, or a category along hierarchy such as "management level / non-management level," a category along job type such as "generalist / administrative," "clerical / technical," "sales / development / planning," a category along employment type such as "regular employment / non-regular employment (temporary employment)," "working at the Tokyo office / working at the Osaka office," "internship / part-time / outsourcing," or a category along recruitment method such as "referral / direct application / through an agent." Furthermore, the conditions regarding the attributes of candidate U2 accepted by the reception department 111 (second condition) may be different from those for the recruitment category. For example, various conditions related to candidate U2's registration information, such as career, number of companies worked for, qualifications held, and skills held, may be accepted as conditions regarding candidate U2's attributes.

[0065] Recruiter U1 can receive conditions regarding the type of question's purpose by performing a predetermined operation on the screen shown in Figure 6. That is, the reception unit 111 of the server device 10 can receive first conditions regarding the type of question's purpose as conditions for the first question presentation means from recruiter U1 who is conducting recruitment activities (Activities A105~A106).

[0066] Specifically, in the example screen shown in Figure 6, there are areas to specify the type of question related to its purpose, such as "hearing questions," "behavioral interview questions," and "case interview questions." When a recruiter U1 sees such a screen, they can consider and set the type of questions to be presented to the candidate U2.

[0067] The questions presented to candidate U2 via the first question presentation means may relate to a single item, but in a typical embodiment, it is preferable that the first question presentation means presents candidate U2 with questions relating to multiple items. This allows recruiter U1 to evaluate candidate U2 from multiple perspectives. When the first question presentation means presents questions relating to multiple items in this manner, a first condition may be set for each of the items. That is, in the example shown in Figure 6, a configuration is shown in which questions relating to four types of items specified in forms F3 to F6 are set, and a condition is set for each of these four types of items. In other words, in this embodiment, the first condition for each of the items can be set according to the operation of the settings screen by recruiter U1. These items may be configured so that their display order can be changed on the settings screen. By presenting questions in the set order, the first question presentation means can allow recruiter U1 to delve deeper in a context that is in line with their intention. The number of questions presented to candidate U2 may be adjusted as appropriate. In the example shown in Figure 6, the number of questions in the "hearing questions" is increased based on the operation of object OBJ1. Although not shown in detail in Figure 6, objects for increasing the number of questions in "behavioral interview questions" and "case interview questions" may also be provided on the screen. Furthermore, it may be possible to decrease the number of questions in the settings screen, and objects or other means to enable such operation may be provided on the screen.

[0068] From one perspective, the "question items" here can constitute one aspect of the questions that recruiter U1 asks candidate U2. Typically, questions can be presented to candidate U2 starting from such "question items." Furthermore, by setting a first condition for each of these "question items," it becomes easier to collect information about candidate U2 according to recruiter U1's objectives and intentions. From this perspective, "question items" may also be referred to as "(question) topics," "(question) aspects," "(question) subject matter," "(question) themes," "(question) perspectives," or "(question) content."

[0069] On the other hand, the "type" associated with such question items may correspond to the purpose for which recruiter U1 asks the questions. As shown in the example in Figure 6, recruiter U1 can set the purpose (objective) of presenting the questions, such as "hearing (questions)," "behavioral interview (questions)," or "case interview (questions)." By showing recruiter U1 the purpose of the questions as a "type" in this way, recruiter U1 can efficiently set what questions to present according to the content, quality, or depth of information they want to extract from candidate U2 (for example, specific examples of past behavior and achievements, ways of thinking, or thinking ability and ability to respond to new challenges). Note that in Figure 6, the purpose itself is directly expressed as a "type," but in one embodiment, the "type" to be set does not have to directly express the purpose, and something other than the purpose may be set. In other words, as long as the predetermined control is realized as information processing, it is not necessarily required that the specific content of the purpose be visible to recruiter U1 etc. on the setting screen. For example, the system may present options that do not directly express the purpose, such as "Question Pattern A / Question Pattern B / Question Pattern C," as question types, and allow users to configure them on the settings screen. Furthermore, the question types may correspond to the questioning approach (method) and evaluation criteria used by the recruiter U1. These settings should not be limited to mere formal settings such as whether or not to delve deeper into the questions (whether or not it is a question-and-answer format), but should also define how to conduct in-depth questioning and dialogue with the candidate U2, using what framework, logic, or rules. Additionally, the names of frameworks, etc., as described later, may be directly specified as question types. For example, the model name of a framework (e.g., the GROW model), as described later, could be set as a question type.

[0070] The setting of the question items can be achieved by various means. For example, as shown in Figure 6, the system may be configured so that the question items and the first condition can be set by entering predetermined text information into a specific form, or the question items and the first condition may be set by the reception unit 111 receiving voice from the recruiter U1.

[0071] Furthermore, the artificial intelligence unit 117 may perform a process to determine a first condition corresponding to the question item. In this case, the server device 10 can determine a first condition regarding the type of question's purpose as a condition for the first question presentation means. For example, the artificial intelligence unit 117 can analyze the text of the question item entered by recruiter U1, determine a first condition that is appropriate for the content, and recommend the determined first condition to recruiter U1 or automatically set it as the first condition. For example, if recruiter U1 enters "Please tell me about your past failures" as a question item, the artificial intelligence unit 117 may determine from the meaning of this text that "a question for interviewing (interviewing about thoughts based on past actions)" is appropriate as the first condition. This reduces the burden on recruiter U1 when setting the first condition.

[0072] The first condition determination model described below can be used for this determination. If the first condition determination model is configured as a dedicated learning model, the model may be one that has learned using question items and the first condition selected for them (for example, the type of question, such as a hearing (question)) as training data. In such a first condition determination model, a correlation between the input question items and the first condition may be established by parameters calculated and tuned through learning. The dedicated learning model may also include a generative AI capable of generating answers not included in the training data. The generative AI in the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0073] If the first condition determination model is a general-purpose learning model (for example, a language model such as a large-scale language model), the server device 10 may input a prompt to the first condition determination model provided in the artificial intelligence unit 117 (artificial intelligence module) that inserts the items of the input question and an instruction to output the first condition corresponding to the items of the question, causing the first condition to be output. Alternatively, the server device 10 may generate a prompt that gives an instruction to the first condition determination model to output the first condition, and input this prompt to the first condition determination model. Furthermore, the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that, in addition to the instruction to output the first condition and the items of the question, inserts, for example, one or more sample items of the question and one or more sample items of the corresponding first condition. Note that the general-purpose learning model may include a generative AI capable of generating arbitrary output information based on the input information. The generative AI of the general-purpose learning model is a general-purpose generative AI that requires input of instructions such as the content of the output information to be generated and the content of the task to be executed. The general-purpose learning model generating AI may analyze the context of the input question items and output information to identify the optimal first condition. The first condition output in this manner may be displayed on the screen of the recruiter U1's terminal. Furthermore, automatic settings regarding the question may be made based on the output first condition.

[0074] Furthermore, Figure 6 shows a configuration in which a template for questions is selected as form F2. That is, in this embodiment, settings for multiple items may be made by free-form text, but on the other hand, the registration unit 110 may register combinations of descriptions for each of the multiple items as templates. In the example shown in Figure 6, after selecting a registered template as input content for form F2, the combination of descriptions is transferred to subsequent forms (forms F3 to F6) when button BT1 is pressed. By adopting this configuration, the operational burden on the recruiter U1 in setting questions can be reduced. After the combination of descriptions has been transferred, the content may be modified as appropriate by the recruiter U1 or others. In addition, the combination of descriptions is not limited to the content of questions presented based on the first question presentation means, but may also include the content of questions presented based on the second question presentation means. In this specification, operational objects displayed on the screen, such as "buttons," may be described, but such operational objects may be appropriately changed to various alternatives that the user can operate. From this perspective, operational objects shown in drawings, etc., may be appropriately referred to as "objects."

[0075] Such templates may be automatically generated by the artificial intelligence unit 117. For example, the server device 10 may generate a combination of interview questions suitable for the selection process of a registered job posting based on information about the job posting (job type, skill requirements, etc.) and present this as a template to the recruiter U1. The template may also be registered for use by recruiter U1 and others as appropriate after it is generated. The registered template may be configured to be shareable not only with recruiter U1 but also with other recruiters belonging to the same organization, for example. This makes it possible to standardize the quality of interviews within the organization.

[0076] The following question item generation model can be used for this generation. When the question item generation model is configured as a dedicated learning model, it is a model that has been trained using information about job postings and the questions created for them as training data. In such a question item generation model, the correlation between information about job postings and the questions is established by parameters calculated and tuned through learning.

[0077] If the question item creation model is a general-purpose learning model (for example, a language model such as a large-scale language model), the server device 10 may input a prompt to the question item creation model provided in the artificial intelligence unit 117 (artificial intelligence module) that inserts information about the job posting and instructions to output question items corresponding to the job posting, causing the server device 10 to output the question items. Alternatively, the server device 10 may generate a prompt that gives instructions to the question item creation model to output question items, and input this prompt to the question item creation model. Furthermore, the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that, in addition to the instructions to output question items and information about the job posting, inserts, for example, one or more samples of information about the job posting and one or more corresponding samples of question items. Note that the general-purpose learning model may include a generative AI capable of generating arbitrary output information based on the input information. The generative AI of the general-purpose learning model is a general-purpose generative AI that requires input of instructions such as the content of the output information to be generated and the content of the task to be executed. The generative AI of the general-purpose learning model may analyze the context of the input information about the job posting and output information to identify the optimal question items. Based on the question items generated in the manner described above, the aforementioned template can be constructed.

[0078] Furthermore, when setting up the questions, settings for evaluating the questions (evaluation criteria) may also be made at the same time. In the example shown in Figure 7, a method is shown in which the points to be evaluated for each type of question are set as "evaluation items". The screen shown in Figure 7 may be displayed, for example, after button BT2 shown in Figure 6 is pressed, but it may also be possible to set up such question evaluations prior to the screen shown in Figure 6.

[0079] In the example shown in Figure 7, form F8, etc., is configured to allow input of evaluation axes associated with various question types. This allows for efficient management of what aspects should be evaluated for the responses of candidate U2. The evaluation axes may be entered by recruiter U1, etc., via selection or as free text. The number of evaluation axes to be set may also be increased based on operations on object OBJ2, etc. Although not shown in detail in Figure 7, it may also be possible to decrease the number of evaluation axes to be set. Furthermore, Figure 7 shows an example in which a "weighting" indicating the importance in evaluation can be set for each type. For example, recruiter U1 can set the evaluation of "hearing questions" to account for 15% of the total, the evaluation of "behavioral interview questions" to account for 60%, and the evaluation of "case interview questions" to account for 25% of the total from the group of questions in the first question presentation means. Furthermore, within each category, detailed weights (percentages) may be set for each further subdivided evaluation axis (for example, "basic skills for performing work," "task execution ability," "communication skills," "logical thinking ability," etc.). In the example shown in Figure 7, it is configured so that predetermined values ​​can be entered into form F9, etc. The weighting (weight) for each category shown above may correspond to the sum of these detailed weights, and the sum of the input values ​​in form F9, etc. may be displayed on the screen in association with each category. In one embodiment, it may be possible to set the weight set for each category and the weight for each subdivided evaluation axis separately. In such cases, if the weight set for each category and the sum of the weights for the subdivided evaluation axes for that category do not match, an error may be displayed on the screen. When the server device 10 (artificial intelligence unit 117, etc.) analyzes the responses of the job candidate U2 and performs scoring, it can calculate a final evaluation (overall score) that reflects the points that the recruiter U1 considers important by multiplying them by these set weights.This makes it possible to precisely reproduce the different hiring criteria for each company and job type within the system.

[0080] Furthermore, the settings for the evaluation (evaluation axis) of such questions may be made on the screen where the type settings are made as described above. That is, the screen shown in Figure 6 may be configured to allow setting evaluation axes and "weights" indicating the importance in the evaluation for each of the multiple items that have been set. Specifically, a form may be provided in which recruiter U1 can input evaluation axes and numerical values ​​(e.g., percentages) representing the importance of the evaluation axes for the type of question (first condition) related to the purpose of the question specified in Forms F3 to F6, etc., and for multiple items (question items) associated with that type.

[0081] Furthermore, even in such cases, the final evaluation value may be calculated using the set weights. That is, by multiplying (or taking a weighted average of) the raw scores of each item calculated by the artificial intelligence unit 117 by the weights received on the setting screen, a comprehensive score or the like that reflects the points that recruiter U1 considers important can be generated.

[0082] Furthermore, the weighting set here may be used not only for calculating evaluations but also as a parameter for controlling the manner in which questions are presented by the first question presentation means. That is, the server device 10 may control the presentation of questions of a type with a high weighting, allocating more time to presenting the question, or repeatedly presenting questions (generating new questions) to allow for more detailed exploration, compared to types with a low weighting.

[0083] Specifically, if the total scheduled time for the interview is 60 minutes, the server device 10 may control the allocation of time resources by allocating approximately 9 minutes to the "hearing (questions)" type, which is weighted at 15%, approximately 36 minutes to the "behavioral interview (questions)" type, which is weighted at 60%, and approximately 15 minutes to the "case interview (questions)" type, which is weighted at 25%. Furthermore, for question types with high weights, the system may be configured to repeatedly generate new questions until the specificity of the candidate U2's answers is sufficiently high (for example, until the degree of satisfaction for each component described later reaches an extremely high level) (for example, by increasing the depth of questioning or the upper limit of the number of questions generated), while for items with low weights, the system may be controlled to move to the next topic once a certain level of answers have been obtained (when the degree of satisfaction exceeds a predetermined level). This allows recruiter U1 to not only specify how evaluations are compiled, but also to reflect their intention to focus on certain aspects in the way questions are presented using the first question presentation method.

[0084] Furthermore, Figure 7 shows an example of selecting an evaluation template as form F7. In other words, in this embodiment, the registration unit 110 may register combinations of evaluation axes to be set for a question as a template. In the example shown in Figure 7, after selecting a registered template as input content for form F7, the combination of evaluation axes is transferred to a subsequent form (form F9, etc.) when button BT3 is pressed. By adopting this method, the operational burden on the recruiter U1 in setting questions can be reduced. After the combination of evaluation axes has been transferred, the content may be modified as appropriate by the recruiter U1, etc.

[0085] Furthermore, the setting of evaluation criteria for such questions may be applied not only to questions related to the first question presentation means, but also to questions related to the second question presentation means.

[0086] Next, we will explain the settings related to the second question presentation method. As an example, recruiter U1 can configure the settings related to the second question presentation method by entering the required information into the various forms on the screen shown in Figure 7 and then pressing button BT4. In other words, the screen shown in Figure 8 is an example of a screen that is accessed when button BT4 is pressed, and on this screen shown in Figure 8, it is possible to configure the settings related to the second question presentation method.

[0087] In the example screen shown in Figure 8, a form F10 is shown containing a question about the selection status of candidate U2 in other recruitment activities. In one embodiment, the format (answer format) used to prompt candidate U2 for answers can also be set based on the answer operation screen. In the example shown in Figure 8, it is possible to set whether to use a "selection format" or a "free-text format" using radio button objects. The selectable answer formats may include, for example, a "scale format (NPS, etc.)" that asks candidate U2 to rate the candidate on a scale from 1 to 10, a "date and time input format" that allows the candidate to input their desired start date from a calendar, or a "multi-line text format". Furthermore, an embodiment is shown in which the content of the options when using a selection format is set based on the settings of forms F11 to F13. The number of options when using a selection format can be increased by manipulating the object OBJ3 on the screen. Although not shown in detail in Figure 8, it may also be possible to decrease the number of options. Furthermore, it may be possible to configure whether each question is mandatory or optional (by setting a mandatory flag). If "free-response format" is selected, the response screen accessed by the candidate U2 may be controlled to display a form (field) or similar element that allows for the specified type of response.

[0088] Furthermore, by entering the required information into the various forms in Figure 8 and then pressing button BT5, the settings for the second question presentation means can be completed. In this way, recruiter U1 can configure the settings for the first and second question presentation means.

[0089] Furthermore, settings for various question presentation methods may be reflected within predetermined ranges. These ranges may be set appropriately according to the content of the recruitment activities of recruiter U1. On the other hand, from the perspective of recruiter U1 conducting recruitment activities efficiently, questions may be set as follows, for example. That is, in one embodiment, the questions of the first question presentation method may be configured to be set for each candidate U2. This makes it easier to ask flexible questions according to the registration details and attributes of candidate U2. On the other hand, the questions of the second question presentation method may be configured to be set for each job posting related to the recruitment activity. As mentioned above, the questions of the second question presentation method may be formal, and by setting them for each job posting in this way, it is possible to improve the efficiency of recruitment activities for recruiter U1.

[0090] Furthermore, the following embodiments may be adopted in relation to such settings screens.

[0091] In other words, in one embodiment, the notification unit 115 of the server device 10 may give a predetermined notification to the recruiter U1 when the questions set in the first setting area and the questions set in the second setting area are related to each other. That is, when the questions presented by the first question presentation means and the questions presented by the second question presentation means are correlated (similar), it will result in duplicate questions being asked to the candidate U2, so the recruiter U1 may be notified that the questions are related in this way. This notification may be implemented by displaying text or marks on the recruiter U1's terminal indicating that both questions are related to each other. The server device 10 may also restrict the setting of related (duplicate) questions. For example, if a specific question is set in the first setting area, and a question highly related to that question is entered in the second setting area, the server device 10 may display an error and prevent the input from being accepted, or it may deactivate (gray out) the button for completing the setting (confirm button, etc.) to make it inoperable. The notification may be made in association with either the first setting area or the second setting area. The notifications that the notification unit 115 can perform include notifying the recruiter U1 via email or other message that the questions are related, and displaying various notifications (so-called pop-up notifications, push notifications, etc.) on the screen of the recruiter's terminal. The notifications that the notification unit 115 performs are not limited to these, and may be any means by which the recruiter U1 can understand that the questions are related. With this configuration, it is possible to prevent the candidate U2 from being asked duplicate questions, reduce the burden on the candidate U2 to answer, and prevent the deterioration of the candidate U2's interview experience. In addition, it is possible to encourage an appropriate division of roles between the first question presentation means and the second question presentation means (for example, the second question presentation means efficiently collects answers to standard questions, and the first question presentation means spends time on more essential in-depth questions), thereby improving the quality and efficiency of the entire recruitment activity.

[0092] Furthermore, the relationship between the questions set in the first setting area and the questions set in the second setting area may be evaluated as follows, for example. That is, the relationship here may be performed by referring to reference information that includes the correlation between both the questions set in the first setting area and the questions set in the second setting area, and the relationship between these two sets of questions. The reference information here may include a set of parameters for outputting the relationship between the two sets of questions from the two sets of questions. For example, the reference information may include various pre-trained models. For example, the reference information may include a first evaluation model, which is a dedicated learning model or a general-purpose learning model that takes both sets of questions as input and outputs the relationship between the two sets of questions. In this case, the server device 10 inputs both sets of questions into the first evaluation model and causes the first evaluation model to output the relationship between the two sets of questions.

[0093] The first evaluation model may be included in the artificial intelligence unit 117. The first evaluation model, which is a dedicated learning model, is a learning model that has learned using the relationships between both questions and their corresponding questions as training data. In such a first evaluation model, parameters calculated and tuned through learning construct a correlation between both questions and the relationships between them. The dedicated learning model may include a generative AI capable of generating answers not included in the training data. The generative AI of the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0094] If the first evaluation model is a general-purpose learning model (for example, a language model such as a large-scale language model), the server device 10 may input an instruction to output the relationship between the two questions based on both questions, and a prompt inserting both questions, to the artificial intelligence unit 117 (artificial intelligence module), causing it to output the relationship corresponding to both questions. Alternatively, the server device 10 may generate a prompt that gives the first evaluation model an instruction to output the relationship between the two questions, and input this prompt to the first evaluation model. In addition to the instruction to output the relationship between the two questions and both questions, the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that inserts, for example, one or more samples of both questions and one or more corresponding samples of the relationship between the two questions. Here, the parameters for constructing the first evaluation model and the prompt including the instruction to output the relationship between the two questions based on both questions construct the correlation between the two questions and the relationship between the two questions. Note that the general-purpose learning model may include a generative AI capable of generating arbitrary output information based on input information. The general-purpose learning model generation AI is a general-purpose generation AI that requires input such as instructions on the content of the output information to be generated and the content of the task to be executed. If the relevance between the two output questions exceeds a predetermined level, it can notify recruiter U1 that there is a duplication of highly relevant questions.

[0095] Furthermore, although we have described a setting screen having a first setting area and a second setting area, the setting screen displayed by the display control unit 114 of the server device 10 may be as follows. That is, the display control unit 114 of the server device 10 may display a setting screen for setting the question presentation means on the terminal of the recruiter U1 conducting recruitment activities. This setting screen may have a third setting area that displays one or more items that the recruiter U1 asks as a list. The third setting area may be configured to allow setting whether the questions corresponding to each of the one or more items are presented based on the first question presentation means or based on the second question presentation means, based on operations on the objects provided in the third setting area.

[0096] Figure 9 shows an example of a screen that may be displayed on a recruiter's terminal. In the example screen shown in Figure 9, information such as questions to be presented to candidate U2 can be entered into an input area (form F14, etc.). Here, the means of presenting the content of the entered questions (presentation method) can be selected using a selection object (radio button) to choose whether to present the questions based on a first question presentation means (e.g., interview format) or a second question presentation means (e.g., questionnaire format). Here, the selection object may be a pull-down menu or a checkbox, etc. In such a screen, the questions to be presented to candidate U2 can be listed, making it easier to prevent questions from being overlooked. Furthermore, since it is possible to set which question presentation means to use to present the content of the questions, recruiter U1 can appropriately grasp the overall picture of the questions to be asked of candidate U2. Also, recruiter U1 can list the questions and assign the question presentation means in parallel on a single screen, which helps to prevent questions from being overlooked or setting errors. Although not shown in detail in Figure 9, the configurable items for the first and second question presentation means may be the same as those described in relation to Figures 6 to 8. In the example shown in Figure 9, the content of the questions to be asked to candidate U2 can be determined by pressing button BT6. Alternatively, a question list created in a predetermined format (such as a CSV file) can be imported to automatically input question items into form F14 and select the question presentation means.

[0097] While Figure 9 shows an example of settings using a visual UI, the system is not limited to this. For example, the system may be configured to register question items via voice input and select or specify question presentation means via voice commands. Alternatively, the system may be configured to associate question content with question presentation means by dragging and dropping card-shaped objects containing question items into areas corresponding to each question presentation means (for example, an area corresponding to the first question presentation means (e.g., an interview area) and an area corresponding to the second question presentation means (e.g., a questionnaire area)). Furthermore, multiple question presentation means (e.g., the first and second question presentation means) are not limited to being selected exclusively; the system may be configured to present the same question using multiple question presentation means.

[0098] Furthermore, in the screen shown in Figure 9, when a question item is entered into form F14, the artificial intelligence unit 117 may execute a process to determine a question presentation means (for example, a first question presentation means or a second question presentation means) corresponding to the question item. The question presentation means determination model described below can be used for this determination. When the question presentation means determination model is configured as a dedicated learning model, the model is trained using question items and the question presentation means selected for them (for example, an interview format or a questionnaire format) as training data. In such a question presentation means determination model, a correlation relationship between the input question item and the question presentation means is constructed by parameters calculated and tuned through learning. The dedicated learning model may include a generative AI capable of generating answers not included in the training data. The generative AI of the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0099] If the question presentation means determination model is a general-purpose learning model (for example, a language model such as a large-scale language model), the server device 10 may input a prompt to the question presentation means determination model provided in the artificial intelligence unit 117 (artificial intelligence module) that inserts the items of the input question and an instruction to output a question presentation means (for example, a first question presentation means or a second question presentation means) corresponding to the items of the question, causing the server device 10 to output the question presentation means. Alternatively, the server device 10 may generate a prompt that gives an instruction to the question presentation means determination model to output a question presentation means, and input this prompt to the question presentation means determination model. Furthermore, the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that, in addition to the output instruction for the question presentation means and the items of the question, inserts, for example, one or more sample question items and one or more sample question presentation means corresponding to them. Note that the general-purpose learning model may include a generative AI capable of generating arbitrary output information based on the input information. The generative AI of the general-purpose learning model is a general-purpose generative AI that requires input of instructions such as the content of the output information to be generated and the content of the task to be executed. The general-purpose learning model generation AI analyzes the context of the input question items and outputs information to identify the optimal question presentation method.

[0100] In this way, based on the judgment result (question presentation means) output from the question presentation means determination model, an object on the screen in Figure 9 (for example, a radio button, etc.) may be automatically selected. Alternatively, based on the output judgment result, a recommended question presentation means may be presented to the recruiter U1. Furthermore, if the output judgment result contradicts a question presentation means that has already been selected, or if it is too biased towards a particular question presentation means, the system may be configured to notify the recruiter U1 accordingly.

[0101] In this way, after the settings for the question presentation means have been made, the presentation unit 112 of the server device 10 presents various question presentation means (first question presentation means, second question presentation means, etc.) to the job candidate U2 involved in the recruitment activities (Activities A107-A108).

[0102] Figure 10 shows an example of a screen that may be displayed on a candidate's terminal. The screen shown in Figure 10 is for candidate U2 to configure settings on their terminal before the first question presentation means asks a question. Specifically, in area Rg1, the candidate can configure settings for devices such as the camera, microphone, and speaker, and then press button BT7 to proceed to the interview.

[0103] Figure 11 shows an example of a screen that may be displayed on the candidate's terminal. The screen shown in Figure 11 is the screen that appears after the aforementioned button BT7 is pressed, and it is used to conduct an interview between the candidate U2 and the interviewer (AI interviewer). Specifically, in the screen shown in Figure 11, the avatar AVT1 is displayed in area Rg2 and is configured to present the questions set by the recruiter U1 to the candidate U2. The content spoken by the avatar AVT1 (questions, etc.) and the content spoken by the candidate U2 (answers, etc.) may be configured to be displayed in area Rg3.

[0104] In this embodiment, the content of the questions in the first question presentation means can be controlled based on the received first conditions. That is, as set in the screen shown in Figure 6, the questions in the first question presentation means can be set in association with a type related to the purpose of the question, such as "hearing questions," "behavioral interview questions," or "case interview questions." The presentation unit 112 of the server device 10 can then control the content of the questions in accordance with the content of such types and present them to the job candidate U2. For example, the storage unit 12 of the server device 10 stores the types of questions and frameworks, logic, or rules for determining the content of the questions corresponding to each type, in association with each other. The presentation unit 112 (or generation unit 116) of the server device 10 can then call up the frameworks, logic, or rules corresponding to the set question types from the storage unit 12, control the generation of additional questions sequentially in accordance with the called frameworks, and present them to the job candidate U2.

[0105] For example, in the "hearing questions," the process may be controlled to sequentially ask additional questions in accordance with a framework for uncovering the abilities of candidate U2. The framework here may be set as appropriate, but for example, a framework based on the GROW model may be used. The GROW model is a method for clarifying the elements of each of the following regarding the subject's behavior and thoughts: G (Goal), R (Reality), O (Options), and W (Will). The process may be controlled to present questions in a way that collects these elements, starting from the question items set by the recruiter U1. Separately, the type of "hearing questions" may be controlled to be in a question-and-answer format. That is, in the first question presentation method, a question may be presented to candidate U2, and if an answer is received, the process may be controlled to move on to the next question without further in-depth questioning.

[0106] In the "behavioral interview questions," for example, the process may be controlled to sequentially ask additional questions in accordance with a framework for uncovering the abilities of candidate U2. The framework here may be set as appropriate, but for example, a framework based on the STAR model may be used. The STAR model is a method for clarifying the elements of a subject's behavior: S (Situation), T (Task), A (Action), and R (Result). The questions may be controlled to be presented in a way that collects these elements, starting from the question items set by the recruiter U1. Note that the framework used in the "behavioral interview questions" is not limited to such a STAR model. For example, a model such as TAPS, which delves into the subject in accordance with each element such as T (To be), A (As is), P (Problem), and S (Solution), may be incorporated.

[0107] Furthermore, the "case interview questions" may be those that assess the candidate's analytical ability regarding a given case, presented by the first question presentation tool. Even in such questions, control may be implemented to sequentially ask additional questions in accordance with a framework designed to uncover the candidate's thoughts and analytical abilities. The framework here may be set as appropriate, but for example, a framework combining a hypothesis thinking process or an analytical process (3C analysis, SWOT analysis, etc.) may be used to allow the candidate to express their thoughts and opinions on the given case. In other words, the questions may be controlled to be presented in a way that collects elements related to the candidate's thoughts and analytical content, starting from the question items set by the recruiter U1.

[0108] The framework for asking questions is not limited to the above; any framework, logic, or rules based on other coaching, counseling, or interviewing methods may be adopted. Furthermore, the types of questions are not limited to the above; various types can be set according to the evaluation criteria, such as "behavioral fact-type questions" to confirm specific past behavioral records or "high-load questions" to confirm the candidate's stress tolerance and quick thinking ability. The generation unit 116 may generate questions for the candidate U2 by controlling the content of the questions, the depth of the questions, the time limit for answers, etc., in accordance with the framework, logic, or rules defined for each of these types.

[0109] As shown in these details, the first question presentation means of this embodiment may generate and present new questions in response to the answers of the candidate U2. The content of the new questions may be controlled based on generation conditions related to questions linked to at least the first condition.

[0110] Of course, the questions presented by the first question presentation method may be controlled by different content (models, etc.). Also, if necessary, the recruiter U1 may be able to configure, via a settings screen, the flow in which the questions are presented to the candidate U2.

[0111] Furthermore, when the first question presentation means generates a new question, the functions of the generation unit 116 may be utilized. For example, a new question may be generated based on generation conditions related to the question and predetermined reference information (first reference information). The reference information (first reference information) here may include a machine learning model that takes generation conditions related to the question as input and outputs a new question, or a first generation model which is a generation AI. If the first generation model includes a machine learning model, the presentation unit 112 may input at least the generation conditions related to the question into the first generation model and execute a process to cause the first generation model to output a new question. On the other hand, if the first generation model includes a generation AI, the presentation unit 112 may input an instruction to generate a new question based at least on the generation conditions related to the question, and the generation conditions related to the question into the generation AI and execute a process to cause the generation AI to output a new question. The generation conditions related to the question here may be associated with the answers of the job candidates and the first conditions.

[0112] To explain from another perspective, the generation of a new question may be performed by referring to reference information that includes the correlation between the generation conditions for a question and the new question. The reference information here may include a set of parameters for generating a new question from the generation conditions for a question. For example, the reference information may include various pre-trained models. For example, the reference information may include a first generation model, which is a dedicated training model or a general-purpose training model that takes the generation conditions for a question as input and outputs a new question. In this case, the generation unit 116 of the server device 10 inputs the generation conditions for a question to the first generation model and causes the first generation model to output a new question.

[0113] The first generative model may be included in the artificial intelligence unit 117. The first generative model, which is a dedicated learning model, is a learning model that has learned generation conditions related to questions and corresponding new questions as training data. In such a first generative model, parameters calculated and tuned through learning build a correlation between the generation conditions related to questions and new questions. Alternatively, the first generative model, which is a dedicated learning model, may be a model that has learned a specific framework, etc., in accordance with the type of question. That is, the model may be pre-trained (or fine-tuned) using training data that conforms to the specific framework, etc., in order to generate questions based on that framework, etc. Note that the dedicated learning model may include a generative AI capable of generating answers not included in the training data. The generative AI of the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0114] If the first generative model is a general-purpose learning model (for example, a language model such as a large-scale language model), the generation unit 116 of the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that includes an instruction to output a new question based on the generation conditions for the question, and the generation conditions for the question, causing the artificial intelligence unit 117 to output a new question that correlates with the generation conditions for the question. Alternatively, the generation unit 116 of the server device 10 may generate a prompt that gives the first generative model an instruction to output a new question, and input this prompt to the first generative model. In this case, the generation unit 116 of the server device 10 may include an instruction in the prompt to generate a new question according to a framework or the like that corresponds to the type of question. That is, the prompt may include an instruction that specifies a framework or the like. In addition to the instruction to output a new question and the generation conditions for the question, the generation unit 116 of the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that includes, for example, one or more sample generation conditions for the question and one or more sample new questions that correspond to them. Here, parameters for constructing the first generative model and prompts containing instructions to output a new question based on the generation conditions for the question establish a correlation between the generation conditions for the question and the new question. The general-purpose learning model may include a generative AI capable of generating arbitrary output information based on input information. The generative AI in the general-purpose learning model is a general-purpose generative AI that requires input instructions such as the content of the output information to be generated and the content of the task to be performed. The generated new question may then be controlled to be presented to the candidate's terminal.

[0115] The generation conditions for questions used to generate new questions can be linked to the answers of candidate U2. This makes it easier to present new questions based on candidate U2's answers, and thus makes it easier to enrich the content of candidate U2's answers. Furthermore, the generation conditions for questions can be linked to the first condition (type of question purpose) mentioned above. This makes it easier to collect answers from candidate U2 in accordance with the content set by recruiter U1, leading to more efficient recruitment activities. From one perspective, the presentation unit 112 of the server device 10 may repeatedly generate and present new questions based at least on the first condition until it reaches the level required by recruiter U1.

[0116] Furthermore, the generation conditions for the questions may include one or more components that the candidate U2's answers must satisfy. The presentation unit 112 of the server device 10 may evaluate the candidate U2's answers based on the degree of satisfaction for one or more components. If at least a portion of the degree of satisfaction for one or more components falls below a predetermined level, the presentation unit 112 of the server device 10 may generate a new question to satisfy that component. In other words, although it was explained earlier that questions may be presented in accordance with the GROW model, STAR model, etc., in this embodiment, new questions may be generated and presented according to the degree of satisfaction for the elements constituting such a model. This makes it easier to manage what kind of answers candidate U2 gave for each element constituting the model, and makes it easier to make the recruitment activities of the recruiter U1 more efficient.

[0117] In one embodiment, the following configuration may be adopted in relation to the first question presentation means. That is, the presentation unit 112 of the server device 10 may be controlled so that questions related to the content answered by the candidate U2 are not presented as new questions. This can prevent the candidate U2 from being asked similar questions. From one perspective, the question generation conditions when generating questions may include a condition to exclude questions that are related to the content answered by the candidate U2.

[0118] The relationship between the answers given by candidate U2 and the new questions may be evaluated as follows, for example. That is, the relationship here may be performed by referring to reference information that includes the correlation between the answers given by candidate U2 and the new questions, and the relationship between these two. The reference information here may include a set of parameters for outputting the relationship between the answers given by candidate U2 and the new questions. For example, the reference information may include various pre-trained models. For example, the reference information may include a dedicated learning model or a general-purpose learning model, which is a second evaluation model, that takes the answers given by candidate U2 and the new questions as input and outputs the relationship between the two. In this case, the server device 10 inputs the answers given by candidate U2 and the new questions into the second evaluation model and causes the second evaluation model to output the relationship between the two.

[0119] The second evaluation model may be included in the artificial intelligence unit 117. The second evaluation model, which is a dedicated learning model, is a learning model that has learned using the content of the responses given by candidate U2 and new questions and the corresponding relationships between them as training data. In such a second evaluation model, parameters calculated and tuned through learning construct a correlation between the content of the responses given by candidate U2 and new questions and the relationships between them. The dedicated learning model may include a generative AI capable of generating answers not included in the training data. The generative AI of the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0120] If the second evaluation model is a general-purpose learning model (for example, a language model such as a large-scale language model), the server device 10 may input to the artificial intelligence unit 117 (artificial intelligence module) an instruction to output the relationship between the content answered by candidate U2 and a new question, and a prompt inserting the content answered by candidate U2 and the new question, causing the server device 10 to output the relationship corresponding to the content answered by candidate U2 and the new question. Alternatively, the server device 10 may generate a prompt that gives the second evaluation model an instruction to output the relationship between the two, and input this prompt to the second evaluation model. In addition to the instruction to output the relationship between the two and the content answered by candidate U2 and the new question, the server device 10 may input to the artificial intelligence unit 117 (artificial intelligence module) a prompt inserting, for example, one or more samples of the content answered by candidate U2 and the new question, and one or more samples of the relationship between the two that correspond to them. Here, parameters for constructing the second evaluation model and prompts including instructions to output the relationship between the content of candidate U2's answers and new questions, construct a correlation between the content of candidate U2's answers and new questions and the relationship between the two. The general-purpose learning model may include a generative AI capable of generating arbitrary output information based on input information. The generative AI of the general-purpose learning model is a general-purpose generative AI that requires input instructions such as the content of the output information to be generated and the content of the task to be performed. Furthermore, if the relationship between the two outputted items is greater than (or exceeds) a predetermined value, the presentation of such questions can be prevented, as a question with a higher relevance may be presented.

[0121] In one embodiment, the following configuration may be adopted in relation to the first question presentation means. That is, the presentation unit 112 of the server device 10 may control the range of questions based on at least a second condition. That is, as mentioned earlier, it is possible to set attributes of the job candidate U2, such as the recruitment category, but in one embodiment, the range of questions may be controlled according to such attributes. For example, if the recruitment category is set as "mid-career," questions about the work that the job candidate U2 has performed so far can be presented, while if the recruitment category is set as "new graduate," it can be controlled so that questions about such work are not presented. To explain from one perspective, the generation conditions for questions when questions are generated may be associated with such a second condition, which makes it easier to present questions that correspond to the attributes of the job candidate U2.

[0122] In one embodiment, the following configuration may be adopted in relation to the first question presentation means. That is, the reception unit 111 of the server device 10 may further receive registration information about the candidate U2. The questions in the first question presentation means may be further controlled based on the received registration information about the candidate U2. This makes it easier to ask questions that are appropriate to the candidate U2's background and skills, thereby improving the richness of the questions. To explain from one perspective, the generation conditions for questions when questions are generated may be associated with such registration information about the candidate U2, thereby making it easier to present questions that are appropriate to the candidate U2's experience and attributes. The registration information about the candidate U2 used here may be information indicating the candidate U2's background and skills, etc., which is registered by the registration unit 110. As mentioned above, such background and skills may be registered through information about work history, for example, a resume or work history document, and the presentation unit 112 may refer to such information and then present questions to the candidate U2.

[0123] Furthermore, the first question presentation means may not only present questions to candidate U2, but also accept questions from candidate U2. That is, the first question presentation means may be configured to accept questions from candidate U2 and provide predetermined answers. For example, the first question presentation means may ask candidate U2 at a predetermined point (such as the end of the interview) whether or not they have any questions. If candidate U2 has a question, the first question presentation means may be configured to provide a predetermined answer.

[0124] Regarding responses provided by such a first question presentation means, the first question presentation means may determine whether it is possible to answer the question based on the content of the question from candidate U2, and may provide an answer to candidate U2's question if it is determined that it is possible to answer. Here, the determination of whether or not an answer is possible may be made not only based on the presence or absence of knowledge, but also from the perspective of appropriateness of answering based on the company's compliance and recruitment policies. In other words, for questions that are easy to answer and for which it is acceptable to provide an answer through the first question presentation means, the first question presentation means may provide an answer on the spot. On the other hand, if it is determined that it is difficult for the first question presentation means to answer questions such as those concerning compensation or details of the work that candidate U2 is expected to perform, the first question presentation means may provide a message to candidate U2 indicating that it is difficult to answer. Such questions that are difficult to answer may be shared with the recruiter U1, and the recruiter U1 may separately provide an answer to candidate U2.

[0125] The answers to the questions (provided by the first question presentation means) may be generated, for example, as follows: The answers may be generated by referring to reference information that includes the correlation between the content of the questions from the candidate U2 and the answers. The reference information here may include a set of parameters for outputting answers from the content of the questions from the candidate U2. For example, the reference information may include various pre-trained models. For example, the reference information may include a first answer model, which is a dedicated learning model or a general-purpose learning model that is trained to take the content of the questions from the candidate U2 as input and output answers. In this case, the server device 10 inputs the content of the questions from the candidate U2 into the first answer model and causes the first answer model to output answers.

[0126] The first response model may be included in the artificial intelligence unit 117. The first response model, which is a dedicated learning model, is a learning model that has learned using the content of questions from job candidate U2 and their corresponding answers as training data. In such a first response model, parameters calculated and tuned through learning build a correlation between the content of questions from job candidate U2 and the answers. The dedicated learning model may also include a generative AI capable of generating answers not included in the training data. The generative AI of the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0127] If the first response model is a general-purpose learning model (for example, a language model such as a large-scale language model), the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that includes an instruction to output a response based on the content of the question from candidate U2, and the content of the question from candidate U2, causing the artificial intelligence unit 117 to output a response corresponding to the content of the question from candidate U2. Alternatively, the server device 10 may generate a prompt that gives an instruction to output a response to the first response model and input this prompt to the first response model. In addition to the instruction to output a response and the content of the question from candidate U2, the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that includes, for example, one or more sample questions from candidate U2 and one or more sample answers corresponding to them. Here, the parameters for constructing the first response model and the prompt including an instruction to output a response based on the content of the question from candidate U2 establish a correlation between the content of the question from candidate U2 and the answer. Note that the general-purpose learning model may include a generative AI capable of generating arbitrary output information based on input information. The general-purpose learning model generative AI is a general-purpose generative AI that requires input instructions such as the content of the output information to be generated and the content of the task to be performed. The outputted answer may be presented to the job candidate U2 (via the first question presentation means). The reliability of the outputted answer may be required as appropriate. If the reliability is deemed sufficient, the answer may be presented to the job candidate U2. On the other hand, if the reliability is deemed insufficient, the presentation of the answer to the job candidate U2 may be prevented.

[0128] In the screen shown in Figure 11, the system is configured so that the next operation proceeds when button BT8 is pressed. In area Rg3, it is noted that a questionnaire will be conducted at the end, but for example, when button BT8 is pressed, the system may be configured so that a screen for answering such a questionnaire is displayed (the second question presentation means is presented). In other words, in one embodiment, the presentation of the second question presentation means may be performed after the presentation of the first question presentation means. By placing the relatively high-load first question presentation means (e.g., the interview part) before the administrative second question presentation means (e.g., the questionnaire part), the job candidate U2 can answer important questions while maintaining a high level of concentration. Furthermore, this prevents a decline in performance caused by fatigue from answering formal questions, and makes it possible to more accurately evaluate the true abilities of the job candidate U2. In addition, the first question presentation means may display an object (button BT8) on the job candidate U2's terminal to execute the presentation of the second question presentation means. The presentation of the second question presentation means may be performed based on the job candidate U2's operation of the object. In one embodiment, button BT8 may be configured to become active when certain conditions are met. For example, button BT8 may be configured to become operational when the first question presentation means has presented all questions, or when the candidate U2 has answered all questions. If these predetermined conditions are not met, button BT8 may remain inactive (e.g., grayed out, unoperable, etc.). Such a configuration ensures that candidate U2 does not leave any answers blank. In particular, it is possible to systematically prevent transitioning to the second question presentation means (e.g., a relatively low-load question such as a multiple-choice questionnaire) before the answers to the first question presentation means have been completed. This reduces the risk for recruiter U1 of accepting answers when the answers to the first question presentation means, which are important for selection decisions, are incomplete.Furthermore, because candidate U2 can visually confirm whether the current task (the first question presentation method) is incomplete through the display status of the button, they can easily identify any missing or incomplete answers, thus improving usability.

[0129] In this example, the presentation of the second question presentation means is shown after the presentation of the first question presentation means, but other configurations can also be adopted. For example, the second question presentation means may be presented before the presentation of the first question presentation means. Also, if the presentation of questions by the first question presentation means spans multiple sections, the presentation of questions by the second question presentation means may occur between these sections.

[0130] Figure 12 shows an example of a screen that may be displayed on a candidate's terminal. The screen shown in Figure 12 may be displayed, for example, based on the pressing of button BT8. The screen shown in Figure 12 displays questions set by recruiter U1 in a questionnaire format. In other words, the screen shown in Figure 12 can correspond to the answer operation screen, which is the second question presentation means. Candidate U2, upon viewing the screen shown in Figure 12, can answer various questions using multiple-choice, free-response, etc. After answering all the questions, the candidate can finish answering by pressing button BT9.

[0131] Furthermore, the following embodiments may be adopted in relation to such a second question presentation means. That is, the presentation unit 112 of the server device 10 may control the second question presentation means so that questions related to the content answered to the first question presentation means are not asked. Alternatively, the presentation unit 112 of the server device 10 may input the content answered to the first question presentation means as a default value into the answer operation screen corresponding to the second question presentation means and present it.

[0132] In other words, it may be burdensome for candidate U2 to answer the same questions again that were answered in response to the first question presentation means. To address this, the presentation unit 112 can control the content of the second question presentation means's screen to prevent similar questions from being asked, and present a response screen to candidate U2. Similarly, to reduce the burden on candidate U2, the presentation unit 112 may also control the input content of the second question presentation means based on the content of candidate U2's answers. For example, Figure 12 shows a configuration where the option "1-2 months later" is pre-selected for item "Q2". If candidate U2 has given such an answer regarding their desired start date to the first question presentation means, the content of that answer may be reflected on the second question presentation means's screen. This can also reduce the burden on candidate U2 to answer the second question presentation means. Note that the content entered as a default value in this way may be configured to be editable (modified) by candidate U2. This ensures that even if there is an error in understanding the intent of the first question, candidate U2 can correct it and complete the answer, thereby guaranteeing the accuracy of the collected data.

[0133] As described above, after receiving a response from candidate U2, the reception unit 111 of the server device 10 receives the response (Activities A109-A110). The memory management unit 113 of the server device 10 then performs memory management processing on the received response (Activity A111). That is, the memory management unit 113 of the server device 10 stores candidate U2's response in association with candidate U2. The response here may include candidate U2's responses to the first question presentation means and / or the second question presentation means.

[0134] Furthermore, the responses that have undergone memory management processing may be made accessible to recruiter U1 as a result of the recruitment activity (Activities A112-A113). This allows recruiter U1 to efficiently carry out the recruitment activity.

[0135] Here, the responses of candidate U2 may include the content of candidate U2's speech (responses to the first question presentation means, etc.) and candidate U2's terminal operation history (responses to the second question presentation means, etc.). Furthermore, candidate U2's responses may also include elements such as candidate U2's actions when the first question presentation means is presented (gestures, body language, speaking speed, behavior, etc.) and the time taken to respond to the questions presented by the first and second question presentation means. These elements may also be stored as appropriate based on the functions of the memory management unit 113, and recruiter U1 can appropriately carry out their recruitment activities by referring to these elements.

[0136] Furthermore, the following approaches may be adopted in relation to the response of candidate U2.

[0137] In other words, in one embodiment, the notification unit 115 of the server device 10 may give a predetermined notification to recruiter U1 when the answer to the first question presentation means and / or the answer to the second question presentation means meets or does not meet predetermined conditions. For example, if the content of the answer of candidate U2 to the various question presentation means is suitable for recruiter U1's recruitment conditions (for example, if the relationship between candidate U2's answer, as described later, and the registered information concerning recruiter U1 exceeds a predetermined threshold), or if candidate U2 is proceeding with other selection processes (for example, if the answer to the second question presentation means is "final interview" or "job offer" as the selection status of another company), or if candidate U2 is actively interested in recruiter U1's recruitment (for example, if the answer to the second question presentation means is "within one week" as the timing of the other company's decision), it may be desirable for recruiter U1 to make an appropriate approach to candidate U2. In such cases, the notification unit 115 can notify recruiter U1 of various information based on the fact that candidate U2's response meets predetermined conditions (for example, information that prompts recruiter U1 to take action, such as "immediate offer recommended," or alert information indicating that the conditions have been met). Conversely, if candidate U2's response does not meet the employment conditions, the notification unit 115 may notify recruiter U1 of various information based on the fact that the predetermined conditions have not been met. The notifications that the notification unit 115 can perform include notifying recruiter U1 of candidate U2's response results via email or other message, or displaying various notifications (so-called pop-up notifications, push notifications, etc.) on the recruiter's terminal screen. The forms of notifications that the notification unit 115 performs are not limited to these, and may be any means by which recruiter U1 can keep track of candidate U2, etc.

[0138] The specified conditions here may be based on the relationship between the responses of candidate U2 and the registration information of recruiter U1. This relationship may be determined by referring to reference information that includes the correlation between the content of candidate U2 and the registration information of recruiter U1 and the relationship between the two. This reference information may include a set of parameters for outputting the relationship between candidate U2's responses and the registration information of recruiter U1. For example, the reference information may include various pre-trained models. For example, the reference information may include a dedicated learning model or a general-purpose learning model, the third evaluation model, which is trained to take candidate U2's responses and the registration information of recruiter U1 as inputs and output the relationship between the two. In this case, the server device 10 inputs candidate U2's responses and the registration information of recruiter U1 into the third evaluation model and causes the third evaluation model to output the relationship between the two.

[0139] The third evaluation model may be included in the artificial intelligence unit 117. The third evaluation model, which is a dedicated learning model, is a learning model that has learned using the responses of candidate U2 and the registration information of recruiter U1, along with the corresponding relationships between the two, as training data. In such a third evaluation model, parameters calculated and tuned through learning construct a correlation between the responses of candidate U2 and the registration information of recruiter U1 and the relationships between the two. The dedicated learning model may include a generative AI capable of generating responses not included in the training data. The generative AI of the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0140] If the third evaluation model is a general-purpose learning model (for example, a language model such as a large-scale language model), the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that includes an instruction to output the relationship between candidate U2 and registered information about recruiter U1 based on the candidate U2's response and the recruiter U1's registration information, and the response of candidate U2 and registered information about recruiter U1, causing the server device 10 to output the relationship corresponding to candidate U2's response and registered information about recruiter U1. Alternatively, the server device 10 may generate a prompt that gives the third evaluation model an instruction to output the relationship between the two, and input this prompt to the third evaluation model. In addition to the instruction to output the relationship between the two and the registered information about candidate U2 and recruiter U1, the server device 10 may input a prompt to the artificial intelligence unit 117 (artificial intelligence module) that includes, for example, one or more samples of the relationship between candidate U2 and registered information about recruiter U1, and one or more corresponding samples of the relationship between the two. Here, parameters for constructing the third evaluation model and prompts including instructions to output the relationship between candidate U2 and registered information about recruiter U1 based on U2's responses and U1's registration information construct a correlation between U2's responses, U1's registration information and the relationship between them. The general-purpose learning model may include a generative AI capable of generating arbitrary output information based on input information. The generative AI of the general-purpose learning model is a general-purpose generative AI that requires input instructions such as the content of the output information to be generated and the content of the task to be performed. If the output relationship between the two is greater than (or exceeds) a predetermined level, or if the relationship between the two is less than (or below) a predetermined level, a predetermined notification can be sent to recruiter U1.

[0141] The registration information of recruiter U1 referenced when making a notification may correspond to the information on job postings registered by registration unit 110. That is, the notification here may be based on information that can be described in the job posting, and may be based on information such as job title, annual salary, industry, job level, required skills, employment type, work location, working hours, holidays, corporate culture, job description, and allowances. In addition, the notification may be based on information such as the location, number of employees, performance, and corporate culture of the employer providing the job posting.

[0142] In one embodiment, the generation unit 116 of the server device 10 may generate statistical information regarding the responses or proposed revisions to the registration content regarding recruitment activities based on the responses to the first question presentation means and / or the responses to the second question presentation means.

[0143] In other words, recruiter U1 may want to compile the responses of candidate U2 to various questioning methods, or modify the registration details related to recruitment activities based on candidate U2's responses. In such cases, the generation unit 116 generates predetermined information, which further streamlines the recruitment activities of recruiter U1.

[0144] In particular, if the second question presentation means is in the form of a questionnaire (e.g., radio buttons, checkboxes, or a scoring format such as a multi-stage evaluation using the Likert scale) that allows the respondent to select an answer from predetermined options, the generation unit 116 can quantitatively process the response results without relying on natural language processing of free-response answers, thereby generating more accurate and objective statistical information. Furthermore, based on the fact that this statistical result is lower than a predetermined standard, the generation unit 116 may generate specific revision suggestions (e.g., "reduce technical terms and use simpler language") for the registered information related to recruitment activities (e.g., the "Job Description" field in a job posting).

[0145] At least one of the "statistical information regarding responses" and the "proposed revisions to registration details regarding recruitment activities" will be collectively referred to as "products" in this explanation. The generation of products here may be performed by referring to reference information that includes the correlation between the responses to the first question presentation means and / or the responses to the second question presentation means and the products. The reference information here may include a set of parameters for generating products from the responses to the first question presentation means and / or the responses to the second question presentation means. For example, the reference information may include various trained models. For example, the reference information may include a second generation model, which is a dedicated training model or a general-purpose training model that takes the responses to the first question presentation means and / or the responses to the second question presentation means as input and outputs products. In this case, the generation unit 116 of the server device 10 inputs the responses to the first question presentation means and / or the responses to the second question presentation means into the second generation model and causes the second generation model to output products.

[0146] The second generative model may be included in the artificial intelligence unit 117. The second generative model, which is a dedicated learning model, is a learning model that has learned using the answers to the first question presentation means and / or the answers to the second question presentation means and their corresponding products as training data. In such a second generative model, parameters calculated and tuned through learning construct a correlation between the answers to the first question presentation means and / or the answers to the second question presentation means and the products. The dedicated learning model may include a generative AI capable of generating answers not included in the training data. The generative AI of the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0147] If the second generative model is a general-purpose learning model (for example, a language model such as a large-scale language model), the generation unit 116 of the server device 10 may input to the artificial intelligence unit 117 (artificial intelligence module) an instruction to output a product based on the answer to the first question presentation means and / or the answer to the second question presentation means, and a prompt inserting the answer to the first question presentation means and / or the answer to the second question presentation means, causing the server device 10 to output a product correlated with the answer to the first question presentation means and / or the answer to the second question presentation means. Alternatively, the generation unit 116 of the server device 10 may generate a prompt that gives the second generative model an instruction to output a product, and input this prompt to the second generative model. In addition to the instruction to output a product and the answer to the first question presentation means and / or the answer to the second question presentation means, the generation unit 116 of the server device 10 may input to the artificial intelligence unit 117 (artificial intelligence module) a prompt inserting, for example, one or more samples of the answer to the first question presentation means and / or the answer to the second question presentation means, and one or more samples of the corresponding product. Here, parameters for constructing the second generative model and prompts including instructions to output a product based on the answer to the first question presentation means and / or the answer to the second question presentation means establish a correlation between the answer to the first question presentation means and / or the product. The general-purpose learning model may include a generative AI capable of generating arbitrary output information based on input information. The generative AI of the general-purpose learning model is a general-purpose generative AI that requires input instructions such as the content of the output information to be generated and the content of the task to be performed. The output product may be controlled to be presented to the recruiter's terminal as appropriate.

[0148] Furthermore, in one embodiment, the presentation unit 112 of the server device 10 may present content introducing recruiter U1 to candidate U2 if the answer to the first question presentation means and / or the answer to the second question presentation means meets predetermined conditions. For example, if candidate U2's answers to the various question presentation means are in line with recruiter U1's recruitment conditions, or if candidate U2 is pursuing other selection processes, or if candidate U2 is actively interested in recruiter U1's recruitment project, it may be desirable for recruiter U1 to make an appropriate appeal to candidate U2. In such cases, presenting appropriate content to candidate U2 can make the matching process between recruiter U1 and candidate U2 more efficient. The predetermined conditions here may be based on the relationship between candidate U2's answers and registered information about recruiter U1. The evaluation of this relationship may be performed by referring to reference information that includes the correlation between the content of candidate U2's answers, recruiter U1's registered information, and the relationship between these two, as described above. This point is the same as what was explained in relation to the third evaluation model, so we will not repeat the explanation here.

[0149] The manner in which such content is presented will be explained using Figure 13. Figure 13 is an example of a screen that may be displayed on a candidate's terminal. The screen shown in Figure 13 is displayed on the candidate's terminal when the candidate's answer to the question presentation means meets predetermined conditions. On the screen shown in Figure 13, a video (content CT) introducing the recruiter U1 is displayed, and the candidate U2 can view the video. By allowing the candidate U2 to interact with such content CT, the candidate's interest in the recruiter U1 can be further increased. Note that viewing such a video may be skipped by pressing button BT10. Furthermore, although Figure 13 shows a manner in which video content is presented as content CT, the content presented is not limited to video content. For example, text content or image content can also be presented to the candidate U2.

[0150] On the other hand, in relation to the responses of candidate U2, the following screen may be displayed on the terminal of recruiter U1. Figure 14 is an example of a screen that may be displayed on the recruiter's terminal. As mentioned earlier, the server device 10 can analyze and evaluate (score) the responses of candidate U2. Specifically, in the example screen shown in Figure 14, there is an area showing "Overall Evaluation (Evaluation EV1)" and an area showing "Item-Specific Evaluation (EV2)". In the area showing "Item-Specific Evaluation", evaluations according to each evaluation axis set by recruiter U1 are shown, and scores and comments (short reviews) are displayed in association with them. In addition, in the area showing "Overall Evaluation", evaluation values ​​calculated using the weights set for each evaluation axis may be shown. That is, the artificial intelligence unit 117 can generate an overall score that reflects the points that recruiter U1 considers important by multiplying (or weighted averaging) the responses of candidate U2 by the weights received on the setting screen as appropriate. The generated content may also be displayed on the recruiter's terminal, etc. Furthermore, in relation to this "overall evaluation," characteristic items, items lacking information, etc., may be displayed on the recruiter's terminal, etc. Also, although Figure 14 shows the evaluation based on the answers to the first question presentation means, instead of, or in addition to, the evaluation based on the answers to the first question presentation means, the evaluation based on the answers to the second question presentation means may be shown to the recruiter U1.

[0151] The generation of an evaluation of candidate U2 may be performed by referring to reference information that includes the correlation between the answers to the first question presentation means and / or the answers to the second question presentation means and the evaluation of candidate U2. The reference information here may include a set of parameters for generating an evaluation of candidate U2 from the answers to the first question presentation means and / or the answers to the second question presentation means. For example, the reference information may include various pre-trained models. For example, the reference information may include a third generation model, which is a dedicated learning model or a general-purpose learning model, that takes the answers to the first question presentation means and / or the answers to the second question presentation means as input and outputs an evaluation of candidate U2. In this case, the generation unit 116 of the server device 10 inputs the answers to the first question presentation means and / or the answers to the second question presentation means into the third generation model and causes the third generation model to output an evaluation of candidate U2.

[0152] The third generative model may be included in the artificial intelligence unit 117. The third generative model, which is a dedicated learning model, is a learning model that has learned using the answers to the first question presentation means and / or the answers to the second question presentation means and the corresponding evaluations of the candidate U2 as training data. In such a third generative model, parameters calculated and tuned through learning construct a correlation between the answers to the first question presentation means and / or the answers to the second question presentation means and the evaluations of the candidate U2. The dedicated learning model may include a generative AI capable of generating answers not included in the training data. The generative AI of the dedicated learning model is a limited-use generative AI that does not require input of instructions such as the content of the output information to be generated or the content of the task to be executed.

[0153] If the third generative model is a general-purpose learning model (for example, a language model such as a large-scale language model), the generation unit 116 of the server device 10 may input to the artificial intelligence unit 117 (artificial intelligence module) an instruction to output an evaluation of candidate U2 based on the answer to the first question presentation means and / or the answer to the second question presentation means, and a prompt inserting the answer to the first question presentation means and / or the answer to the second question presentation means, causing the artificial intelligence unit 117 to output an evaluation of candidate U2 that correlates with the answer to the first question presentation means and / or the answer to the second question presentation means. Alternatively, the generation unit 116 of the server device 10 may generate a prompt that gives the third generative model an instruction to output an evaluation of candidate U2, and input the prompt to the third generative model. The generation unit 116 of the server device 10 may input to the artificial intelligence unit 117 (artificial intelligence module) prompts that include, for example, one or more sample answers to the first and / or second question presentation means and one or more corresponding sample evaluations of candidate U2, in addition to the output instruction for the evaluation of candidate U2 and the answer to the first question presentation means and / or the answer to the second question presentation means. Here, the parameters for constructing the third generation model and the prompts that include instructions to output the evaluation of candidate U2 based on the answer to the first and / or second question presentation means establish a correlation between the answer to the first and / or second question presentation means and the evaluation of candidate U2. The general-purpose learning model may include a generation AI capable of generating arbitrary output information based on input information. The generation AI of the general-purpose learning model is a general-purpose generation AI that requires input of instructions such as the content of the output information to be generated and the content of the task to be executed. The output evaluation of candidate U2 may be controlled to be presented to the recruitment agent terminal as appropriate. Furthermore, the evaluation of such job candidate U2 may be expressed using various methods, such as numerical values ​​(scores), text, ranks, graphs, tables, etc. In other words, the evaluation that the server device 10 can generate and display may take various forms, such as the overall evaluation (rank), overall evaluation comments (text), handover notes (text), numerical values ​​(scores) as item-specific evaluations, and comments (text) as item-specific evaluations, as shown in Figure 14.

[0154] In summary, the information processing method of this embodiment allows for the storage of the responses of the candidate U2 to predetermined question presentation means, thus making the recruitment process more efficient.

[0155] 4. Others Although embodiments of the present invention have been described above, the present invention is not limited thereto and can be modified as appropriate without departing from the technical spirit of the invention.

[0156] In the above embodiment, the first question presentation means was shown to receive a question from candidate U2 and provide a predetermined answer. However, the process of receiving such a question and providing an answer may be carried out by means other than the first question presentation means. For example, a predetermined answer means may be provided on the platform provided by the information processing system 1, and this answer means may be configured to provide a predetermined answer to a question input to it. Such an answer means may have various interfaces for candidate U2. For example, the answer means may receive questions from candidate U2 in a chatbot format and provide an answer to that question. The location of the answer means may also be set as appropriate. The answer here may be provided by referring to reference information that includes the correlation between the content of the question from candidate U2 and the answer, as described above. This is the same as described in relation to the first answer model, so the explanation will not be repeated here.

[0157] In the above embodiment, a method of presenting predetermined content to the candidate U2 was shown, but such content may be presented regardless of the content of the candidate U2's answers. For example, when the candidate U2 has finished answering questions based on various question presentation means, predetermined content may be presented to the candidate U2 regardless of the quality of the answers.

[0158] In the above embodiment, the server device 10 performed various storage and control functions, but instead of the server device 10, multiple external devices may be used. That is, various information and programs may be stored in a distributed manner across multiple external devices using blockchain technology or the like. Also, the artificial intelligence unit 117 may be an external component of the server device 10. In that case, the external artificial intelligence unit 117 may be provided by, for example, an artificial intelligence service server, and is configured to receive input from each functional unit of the server device 10, receive requests to execute artificial intelligence services, and return the instructed output as a processing result to the server device 10. The artificial intelligence service server may be a server that provides services using a language model as a learning model, or a server that executes language processing tasks using a language model, and may provide an LLM, generative AI, or AI agent. The artificial intelligence service server may, for example, receive prompt input in the form of text, images, or audio, and generate and respond to the prompt. The server device 10 may also cooperate with the API (Application Programming Interface) of the service server that provides generative AI, etc., and utilize the generative AI, etc.

[0159] At least one of the devices included in the information processing system 1 may be located outside the country in which the functions of the information processing system 1 are performed.

[0160] The embodiments of this model are not limited to the information processing system 1, but may also be an information processing method or a program. The information processing method comprises each step executed by the information processing system 1. The program causes a computer to execute each step of the information processing system 1.

[0161] The product may be provided in any of the following embodiments.

[0162] (1) An information processing system comprising at least one processor, wherein the processor is configured to perform the following steps by reading a program, the system comprising: reception step, receiving a first condition concerning the type of purpose of a question as a condition for a first question presentation means from a recruiter engaged in recruitment activities; presentation step, presenting the first question presentation means to a candidate for employment related to the recruitment activities, wherein the first question presentation means presents a question to the candidate for employment, the content of the question in the first question presentation means is controlled based on the received first condition; and storage management step, storing the candidate's answer to the question in the first question presentation means in association with the candidate for employment.

[0163] (2) An information processing system as described in (1) above, wherein in the presentation step, the first question presentation means generates and presents a new question in response to the answer of the candidate for employment, and the content of the new question is controlled based on generation conditions relating to at least the first condition.

[0164] (3) An information processing system as described in (2) above, wherein the generation conditions for the question include one or more components that the candidate's answer must satisfy, the presentation step evaluates the candidate's answer based on the degree of satisfaction of each of the one or more components, and the new question is generated to satisfy the degree of satisfaction of at least a portion of the degree of satisfaction of the one or more components if such a portion falls below a predetermined level.

[0165] (4) An information processing system as described in (2) or (3) above, wherein the new question is generated based on generation conditions relating to the question and first reference information, the first reference information includes a first generative model which is a machine learning model or generative AI that is capable of taking the generation conditions relating to the question as input and outputting the new question, the presentation step, if the first generative model includes the machine learning model, at least the generation conditions relating to the question are input to the first generative model and a process is executed to cause the first generative model to output the new question, if the first generative model includes the generative AI, at least an instruction to generate the new question based on the generation conditions relating to the question and the generation conditions relating to the question are input to the generative AI and a process is executed to cause the generative AI to output the new question, the generation conditions relating to the question are associated with the answers of the job candidate and the first conditions.

[0166] (5) An information processing system according to any one of (2) to (4) above, wherein in the presentation step, the generation and presentation of new questions are repeated at least based on the first condition until the level required by the recruiter is reached.

[0167] (6) An information processing system described in any one of (2) to (5) above, wherein in the presentation step, the system is controlled so that questions related to the content answered by the candidate are not presented as new questions.

[0168] (7) An information processing system according to any one of (1) to (6) above, wherein in the reception step, a second condition relating to the attributes of the candidate for employment to whom the questions are asked is received, and in the presentation step, the scope of the questions is controlled based at least on the second condition.

[0169] (8) An information processing system as described in (7) above, wherein the attributes of the candidate for employment include at least information relating to the employment category of the candidate for employment.

[0170] (9) An information processing system according to any one of (1) to (8) above, wherein the first question presentation means presents the candidate with questions relating to multiple items, and the first condition is set for each of the multiple items.

[0171] (10) An information processing system as described in (9) above, wherein in the display control step, a setting screen for setting the question presentation means is displayed on the terminal of the recruiter, and the first condition for each of the multiple items is set according to the operation of the setting screen.

[0172] (11) An information processing system as described in (10) above, wherein the settings for the multiple items are made by free description, and in the registration step, a combination of the descriptions for each of the multiple items is registered as a template.

[0173] (12) An information processing system according to any one of (1) to (11) above, wherein the reception step further receives registration information relating to the candidate for employment, and the content of the questions in the first question presentation means is further controlled based on the received registration information relating to the candidate for employment.

[0174] (13) An information processing system described in any one of (1) to (12) above, wherein in the notification step, a predetermined notification is given to the recruiter if the answer to the first question presentation means satisfies predetermined conditions or does not satisfy predetermined conditions.

[0175] (14) An information processing system according to any one of (1) to (13) above, wherein the first question presentation means displays a predetermined avatar on the terminal of the job candidate, and the predetermined avatar presents a question to the job candidate.

[0176] (15) An information processing system according to any one of (1) to (14) above, comprising a server device having the processor and a terminal that can access the server device.

[0177] (16) An information processing method comprising each step performed by the information processing system described in any one of (1) to (15) above.

[0178] (17) A program that causes a computer to perform each step of the information processing system described in any one of (1) through (15) above. Of course, this is not always the case.

[0179] Finally, while various embodiments relating to this disclosure have been described, these are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be implemented in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0180] 1: Information Processing System 2: Communication lines 10: Server device 11: Control Unit 12: Storage section 13: Communications Department 14: Communications bus 20: User terminal 21: Control Unit 22: Storage section 23: Communications Department 24: Input section 25: Output section 26: Communications bus 110: Registration Department 111: Reception Department 112: Presentation part 113: Memory management department 114: Display Control Unit 115: Notification Department 116 :Generation part 117: Artificial Intelligence Department 210: Display Control Unit 211: Operation Reception Section AVT1: Avatar BT1~BT10: Buttons CT: Content EV1, EV2: Evaluation F1~F14: Form OBJ1~OBJ3: Objects Rg1~Rg3: area U1: Recruitment agent U2: Candidate

Claims

1. An information processing system, Equipped with at least one processor, The aforementioned processor is configured to perform the following steps by reading a program: In the display control step, a settings screen for configuring the question presentation means is displayed on the terminal of the recruiter conducting the recruitment activities. Here, the settings screen allows you to configure the questions that will be presented to job candidates in connection with the recruitment activities for the aforementioned personnel. The aforementioned settings screen is configured to allow input of topics or content for each of the multiple questions as multiple topics or content, The settings screen is configured to allow each of the multiple topics or content to be set in association with a first condition regarding the type of the purpose of the question. In the reception step, based on the operation of the setting screen by the recruiter, the conditions for the first question presentation means are received as the multiple topics or content and the first conditions associated with each of the multiple topics or content. In the presentation step, the candidate for employment is presented with the first question presentation method. Here, the first question presentation means presents the candidate with a question relating to each of the multiple topics or content that it has received. Each of the questions presented by the first question presentation means is controlled in content based on the first condition associated with the topic or content corresponding to the question, In the memory management step, an information processing system stores the answers of the job candidates to the questions presented by the first question presentation means, associating them with the job candidates.

2. In the information processing system described in claim 1, In the aforementioned presentation step, the first question presentation means generates and presents a new question in response to the candidate's answer. An information processing system in which the content of the new question is controlled based on generation conditions relating to at least the first condition.

3. In the information processing system described in claim 2, The generation conditions for the aforementioned question include one or more components that the candidate's answer must satisfy. An information processing system in which, in the presentation step, the responses of the job candidate are evaluated based on the degree of satisfaction of each of the one or more components, and the new questions are generated in such a way that the degree of satisfaction of at least some of the components falls below a predetermined level.

4. In the information processing system described in claim 2, The aforementioned new question is generated based on the generation conditions for the aforementioned question and the first reference information. The first reference information includes a first generative model, which is a machine learning model or generative AI, that is capable of taking the generation conditions for the question as input and outputting the new question. In the aforementioned presentation step, If the first generation model includes the learning model, at least the generation conditions relating to the question are input to the first generation model, and the process is executed to cause the first generation model to output the new question. If the first generation model includes the generation AI, the generation AI is given an instruction to generate a new question based on generation conditions relating to the question, and generation conditions relating to the question, and the generation AI is given an instruction to generate a new question based on generation conditions relating to the question, and the generation AI is given an instruction to generate a new model is given an instruction to generate a new question based on generation conditions relating to the question, and the generation model is given an instruction to generate a new question based on generation An information processing system in which the generation conditions for the aforementioned question are associated with the answers of the job applicants and the first condition.

5. In the information processing system described in claim 2, An information processing system that, in the presentation step, repeatedly generates and presents new questions to the level required by the recruiter, based at least on the first condition.

6. In the information processing system described in claim 2, An information processing system that, in the presentation step, controls the system so that questions related to the answers given by the candidate are not presented as new questions.

7. In the information processing system described in claim 1, In the aforementioned application step, the second condition regarding the attributes of the candidate for employment to whom the aforementioned questions are asked is received, An information processing system that, in the presentation step, controls the scope of the question based on at least the second condition.

8. In the information processing system described in claim 7, An information processing system that includes at least information regarding the employment category of the candidate for employment.

9. In the information processing system described in claim 1, The settings for the aforementioned multiple topics or content are made through free-form text. The registration step involves registering combinations of descriptive content for each of the aforementioned multiple topics or contents as templates in an information processing system.

10. In the information processing system described in claim 1, In the aforementioned application step, registration information regarding the candidate for employment is further received. An information processing system in which the content of the questions in the first question presentation means is further controlled based on the registered information of the received job candidate.

11. In the information processing system described in claim 1, In the notification step, the information processing system provides a predetermined notification to the recruiter if the answer to the first question presentation means meets or does not meet predetermined conditions.

12. In the information processing system described in claim 1, The first question presentation means is an information processing system that displays a predetermined avatar on the terminal of the job candidate, and the predetermined avatar presents a question to the job candidate.

13. In the information processing system described in claim 1, A server device having the aforementioned processor, An information processing system comprising a terminal capable of accessing the aforementioned server device.

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

15. It is a program, A program for causing a computer to perform each step of the information processing system described in any one of claims 1 to 13.

Images