Information processing system, information processing method and program

Abstract

To provide an information processing system etc. that enables a job seeker to efficiently input search conditions for searching for information on job openings that suit the job seeker. [Solution] According to one aspect of the present invention, there is provided an information processing system comprising at least one processor, the processor being configured to execute the following steps by reading a program: in an input receiving step, input information including instructions regarding search conditions for a job search is received from a job seeker in a chat format; in an output information generating step, first output information is generated based on the input information and first reference information, the first reference information including at least a relationship between the input information and the first output information, and the first output information includes search conditions according to the input information; and in a display control step, the first output information is displayed in a chat format so that it can be viewed by the job seeker.

Description

[Technical Field]

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

[0002] Patent Document 1 discloses a technique relating to an employment support system that provides employment information or job-seeking information in response to search requests from job seekers or companies offering jobs. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-331684 Summary of the Invention [Problem to be solved by the invention]

[0004] It was time-consuming to input search criteria to search for information on job openings that suited the job seeker.

[0005] In view of the above circumstances, the present invention provides an information processing system etc. that allows a job seeker to efficiently input search conditions for searching for information on job offers that suit the job seeker. [Means for solving the problem]

[0006] According to one aspect of the present invention, there is provided an information processing system comprising at least one processor, the processor being configured to execute the following steps by reading a program: an input receiving step receiving input information from a job seeker in a chat format, the input information including instructions regarding search conditions for a job search; an output information generating step generating first output information based on the input information and first reference information, the first reference information including at least a relationship between the input information and the first output information, and the first output information including search conditions according to the input information; and a display control step displaying the first output information in a chat format so that it can be viewed by the job seeker.

[0007] According to this embodiment, it is possible to assist in generating a search formula for searching for information on job openings that suit a job seeker.

[0008] Hereinafter, embodiments of the present invention will be described. Note that various features shown in the following embodiments can be combined with each other. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a configuration diagram illustrating an information processing system 1. FIG. [Figure 2] FIG. 2 is a block diagram showing the hardware configuration of the server 2. [Figure 3] 3 is a block diagram showing the hardware configuration of a job seeker terminal 3. FIG. [Figure 4] 2 is a block diagram showing functions realized by the server 2 (controller 23) and the job seeker terminal 3 (controller 33). FIG. [Figure 5] FIG. 2 is a diagram showing an outline of processing executed by the information processing system 1. [Figure 6] 2 is an activity diagram showing an example of the flow of information processing according to the embodiment, which is executed by the information processing system 1. FIG. [Figure 7] FIG. 10 is a diagram showing a screen G1 which is an example of a chat screen. [Figure 8]FIG. 10 is a diagram showing a screen G2 which is an example of a chat screen. [Figure 9] FIG. 10 is a diagram showing a screen G3, which is an example of a chat screen. DETAILED DESCRIPTION OF THE INVENTION

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the accompanying drawings. Various features shown in the following embodiments can be combined with each other.

[0011] Incidentally, the program for realizing the software appearing in one embodiment may be provided as a non-transitory computer-readable medium, or may be provided so that it can be downloaded from an external server, or may be provided so that the program is started on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0012] Furthermore, various information processing according to an embodiment may realize input and output corresponding to the input. Here, the form of information referenced in such information processing (hereinafter referred to as reference information) is not limited as long as an output is obtained as a result of the input. The reference information may be, for example, rule-based information such as a database, a lookup table, or a predetermined function (including a decision formula such as a regression formula constructed using a statistical method), a trained model that has previously learned the correlation between input and output, or a generative AI such as a large-scale language model (these models include parameters that establish the correlation between input and output) or a visual language model that can output a desired result in response to a prompt.

[0013] In one embodiment, a "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In one embodiment, various information is handled, and this information is represented, for example, by physical values ​​of signal values ​​representing voltage and current, high and low signal values ​​as a binary bit set consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculations can be performed on a circuit in the broad sense.

[0014] Furthermore, a circuit in the broad sense is a circuit realized by at least an appropriate combination of a circuit, circuitry, processor, memory, etc. The processor may be a general-purpose processor or a dedicated circuit. That is, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

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

[0016] <Information Processing System 1> FIG. 1 is a configuration diagram illustrating an information processing system 1. The information processing system 1 includes a server 2 and a job seeker terminal 3. The server 2 and the job seeker terminal 3 are configured to be able to communicate with each other via a telecommunications line (network). In an exemplary embodiment, the job seeker terminal 3 can function as a job seeker's terminal. Here, the system exemplified as the information processing system 1 is made up of one or more devices or components. Therefore, it should be noted that the information processing system 1 includes both the server 2 alone and the server 2 and the job seeker terminal 3. More specifically, the information processing system 1 may include an element selected from the group consisting of the server 2 and the job seeker terminal 3. The elements not selected may not be included in the information processing system 1, but may be external elements electrically connected to the selected elements. These components will be described below.

[0017] The information processing system 1 constitutes, for example, at least a part of a recruitment and job search system used by job seekers. The information processing system 1 provides and manages a talent matching platform and talent matching services used by job seekers. In one embodiment, the information processing system 1 is made up of one or more devices or components. For example, the information processing system 1 may include a server (e.g., server 2) having a processor (e.g., control unit 23 described below) and a terminal (e.g., job seeker terminal 3) that can access the server. These components will be described below.

[0018] <Server 2> 2 is a block diagram showing the hardware configuration of server 2. Server 2 includes a communication unit 21, a storage unit 22, and a control unit 23, and these components are electrically connected via a communication bus 20 inside server 2. Each component will be further described below.

[0019] The communication unit 21 is configured by a communication module. The communication module may be a wireless communication module conforming to standards such as IEEE802.11a / b / g / n / ac / ax, LTE, 5G, or 6G, or may be a wired communication module conforming to standards such as IEEE802.3. The communication unit 21 is configured to be able to transmit various electrical signals from the server 2 to external components. The communication unit 21 is also configured to be able to receive various electrical signals from the external components to the server 2. More preferably, the communication unit 21 has a network communication function, which allows various information to be communicated between the server 2 and external devices via a network.

[0020] The memory unit 22 stores various pieces of information defined above. This can be implemented, for example, as a storage device such as a solid state drive (SSD) that stores various programs and the like related to the server 2 executed by the control unit 23, or as a memory such as a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to the program operations. The memory unit 22 stores various programs, variables, etc. related to the server 2 executed by the control unit 23.

[0021] The control unit 23 processes and controls the overall operations related to the server 2. The control unit 23 is, for example, a central processing unit (CPU) not shown. The control unit 23 realizes various functions related to the server 2 by reading out predetermined programs stored in the storage unit 22. In other words, information processing by software stored in the storage unit 22 is specifically realized by the control unit 23, which is an example of hardware, and each step related to each function described below can be executed. This will be described in further detail in the next section. Note that the control unit 23 is not limited to being single, and multiple control units 23 may be provided for each function. A combination of these may also be used.

[0022] <Job Seeker Terminal 3> The job seeker terminal 3 is an information processing device used by the job seeker U1. "Job seekers" include, for example, employed people (those wishing to change jobs or transfer within the company), prospective new graduates (job seekers), students, etc., and also include people who are looking for a new job or job, or people who are interested in changing jobs or finding employment.

[0023] 3 is a block diagram showing the hardware configuration of the job seeker terminal 3. The job seeker terminal 3 includes a communication unit 31, a memory unit 32, a control unit 33, a display unit 34, and an input unit 35, and these components are electrically connected via a communication bus 30 inside the job seeker terminal 3. Each component will be further described below. The description of the communication unit 31, memory unit 32, and control unit 33 will be omitted as they are the same as the description of each unit in the server 2.

[0024] The display unit 34 may be included in the housing of the job seeker terminal 3 or may be attached externally. The display unit 34 displays a graphical user interface (GUI) screen that can be operated by the user. This is preferably implemented by using display devices such as a CRT display, a liquid crystal display, an organic EL display, and a plasma display, depending on the type of job seeker terminal 3.

[0025] The display unit 34 displays a screen indicated by the screen data transmitted from the server 2. The display unit 34 displays a system screen relating to the information processing system 1 indicated by the screen data transmitted from the server 2.

[0026] The input unit 35 may be included in the housing of the job seeker terminal 3, or may be attached externally. For example, the input unit 35 may be implemented as a touch panel integrated with the display unit 34. A touch panel allows the user to input operations such as tapping and swiping. Of course, switch buttons, a mouse, a QWERTY keyboard, etc. may be used instead of a touch panel. In other words, the input unit 35 accepts operation inputs made by the user. The inputs are transferred as command signals to the control unit 33 via the communication bus 30, and the control unit 33 can execute predetermined control and calculations as necessary.

[0027] 2. Functional configuration This section describes the functional configuration of this embodiment. Information processing by software stored in the storage unit 22 is specifically realized by the control unit 23, which is an example of hardware, and can be executed as each functional unit included in the control unit 23 (at least one processor included in the information processing system 1).

[0028] FIG. 4 is a block diagram showing functions realized by the server 2 (controller 23) and the job seeker terminal 3 (controller 33).

[0029] 4A is a block diagram showing functions realized by the control unit 23. The control unit 23 includes an input receiving unit 231, an output information generating unit 232, a display control unit 233, a proposal generating unit 234, a presenting unit 235, an adding unit 236, a recording unit 237, an artificial intelligence unit 238, and a determining unit 239.

[0030] As shown in FIG. 4B, the job seeker terminal 3 (control unit 33) includes a display unit 331 and an operation acquisition unit 332.

[0031] <Input Receiving Unit 231> The input acceptance unit 231 is configured to be able to accept information from the job seeker terminal 3 or other devices. The input acceptance unit 231 is also configured to be able to accept various pieces of information by reading out various pieces of information stored in a storage area that is at least a part of the memory unit 22 and writing the read out information in a work area that is at least a part of the memory unit 22. The storage area is, for example, an area of ​​the memory unit 22 that is implemented as a storage device such as an SSD. The work area is, for example, an area that is implemented as a memory such as a RAM.

[0032] The input receiving unit 231 is configured to be able to receive input based on terminal operation by the job seeker U1 on the job seeker terminal 3. Specifically, the input receiving unit 231 is configured to be able to receive input by the job seeker U1. The input receiving unit 231 may receive job seeker information of the job seeker U1 via the job seeker terminal 3.

[0033] The input accepting unit 231 is configured to be able to accept input information from the job seeker U1. For example, the input accepting unit 231 may accept input information from the job seeker U1 via a chat screen displayed on the display unit 34 of the job seeker terminal 3. The input accepting unit 231 may accept free text descriptions as search conditions for the job search desired by the job seeker. The input accepting unit 231 may also accept at least one or more items of the job seeker U1's desired annual salary, job type, work location, industry, and job type as search conditions for the job search. The input information from the job seeker U1 may include the job seeker U1's desired industry and job type, as well as desired conditions for the organization to which the job seeker U1 is applying. Information related to job content includes information about organizations in which the job seeker U1 has had experience, information describing the duties and tasks the job seeker U1 has performed at the organization, etc. The job description may include, for example, the job seeker's previous organizations, department names, job titles, length of employment, positions held, projects held, work period, the job seeker's management experience, language skills, awards, etc.

[0034] The input accepting unit 231 accepts input information via the first input field and accepts corrected input information via the second input field. The first input field and the second input field are, for example, input fields that accept input in a chat format. Note that the input information and corrected input information may be accepted from the same input field. The input accepting unit 231 accepts corrected input information from the job seeker in a chat format. The corrected input information includes instructions for correcting the first output information. The corrected input information includes instructions for correcting the search conditions for the job search included in the first output information. The input accepting unit 231 accepts instructions for executing the search conditions in a chat format. The input accepting unit 231 accepts corrected input information including instructions for correcting the search results. In this way, the input accepting unit 231 may accept corrections to the search results after once displaying the search results. The input accepting unit 231 may accept corrected input information including instructions for correcting the search results in a chat format.

[0035] <Output information generation unit 232> The output information generation unit 232 can generate various types of information. The output information generation unit 232 can generate output information based on input information and reference information received from the job seeker U1. The output information generation unit 232 generates first output information based on the input information of the job seeker U1 and first reference information. The first reference information includes at least the relationship between the input information and the first output information. The first output information includes search criteria according to the input information. The output information generation unit 232 may generate the first output information based on search criteria output from a search criteria generation model included in the first reference information. The output information generation unit 232 controls input to the artificial intelligence unit 238. The output information generation unit 232 is configured, for example, to use input information input by the job seeker U1 as input and generate a prompt instructing the output of search criteria according to the input information of the job seeker U1. The output information generation unit 232 may also be configured, for example, to use input information, etc. as input, and generate a prompt instructing the output of first output information. Note that "generation" may be read as "creation."

[0036] <Display control unit 233> The display control unit 233 controls the display unit 34 to display visual information such as screens, images including still images and videos, icons, and messages (text). The control unit 23 may generate rendering information for displaying visual information on the display unit 34 and transmit the rendering information to the job seeker terminal 3, thereby causing the job seeker terminal 3 to generate the visual information and display the visual information on the display unit 34. The display control unit 233 may also generate visual information and transmit the visual information to the job seeker terminal 3, thereby causing the display unit 34 to display the visual information. The display control unit 233 executes processing for displaying a system screen related to the information processing system 1 on each terminal. For example, the display control unit 233 performs processing such as generating and transmitting an HTML (Hyper Text Markup Language) file and causing the display unit 34 to display a web page showing the system screen. The display control unit 233 may also perform processing such as generating and transmitting display data for an application for using the information processing system 1. For example, the display control unit 233 can cause the display unit 34 to display various visual information.

[0037] The display control unit 233 is configured to display various information on the job seeker terminal 3 and other devices. For example, the display control unit 233 displays a chat screen or the like on the display unit 331 of the job seeker terminal 3.

[0038] <Proposal generation unit 234> The proposal generation unit 234 can generate proposal information. The proposal information is, for example, information related to questions, options, or proposals for the job seeker U1. For example, as a first proposal generation step, the proposal generation unit 234 generates search conditions for the job seeker or options related to the search conditions based on registration information related to the job seeker registered in the database and the third reference information. As a second proposal generation step, the proposal generation unit 234 generates a proposal related to corrected input information based on registration information related to the job seeker registered in the database, input information received from the job seeker, and fourth reference information.

[0039] <Presentation part 235> The presentation unit 235 can present the information generated by the proposal generation unit 234 to the job seeker U1 in a visible manner.

[0040] <Additional Section 236> The adding unit 236 adds the specified information to the input information. The adding unit 236 can specify information that can be added to the input information and add the specified information to the input information.

[0041] <Recording Unit 237> The recording unit 237 can record and save various information and data in association with each other. The recording unit 237 controls the memory unit 22 of its own device and writes and reads data to the memory unit 22. The recording unit 237 stores, for example, job seeker information (input information) about the job seeker U1 entered on the input screen. The recording unit 237 may register the job seeker information (registration information) of the job seeker U1, which is generated or updated in response to the input by the job seeker U1, and the job seeker information of the job seeker U1 entered on the chat screen, in the job seeker database in association with the job seeker U1's identification information. The recording unit 237 can also record the input information entered by the job seeker, the corrected input information, and the first output information and second output information displayed to the job seeker as a conversation log.

[0042] <Artificial Intelligence Department 238> The artificial intelligence unit 238 is configured to receive input from each functional unit and return the instructed output. The artificial intelligence used by the server 2 for each functional unit may be a common one, or may be prepared individually for each functional unit.

[0043] The artificial intelligence unit 238 is an AI (Artificial Intelligence) equipped with learning models such as Transformers including GPT (Generative Pretrained Transformer, including GPT-1, GPT-2, GPT-3, GPT-4, and GPT-4o), BERT (Bidirectional Encoder Representations from Transformers), and BART (Bidirectional and Auto-regressive Transformer), and language models such as Recurrent Neural Networks (RNNs), and may include generative AI or AI agents.

[0044] The language model is an example of a learning model based on a machine learning algorithm. Specific examples of machine learning algorithms include nearest neighbor methods, naive Bayes methods, decision trees, support vector machines, and deep learning using neural networks. The artificial intelligence unit 238 can apply the above algorithms as appropriate.

[0045] The artificial intelligence unit 238 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). The training data consists of pairs of input data for learning and output data (correct answer data). In addition, the language model may not only be trained for a specific task, but also be a general-purpose model that can be used for a wide range of tasks.

[0046] The artificial intelligence unit 238 may use a natural language model as its artificial intelligence, or may include a general-purpose natural language processing trained model such as a large-scale language model (LLM). A large-scale language model is a learning model that has previously trained a large amount of data, such as text data (e.g., (i) web content on the Internet, or (ii) data stored in a specified database). It can perform various language processing tasks when given a task, and can perform a wide range of natural language processing tasks, such as understanding sentence patterns and contexts, answering questions, and generating sentences, according to given prompts. Such a general-purpose learning model may include a language model that can handle various tasks without fine-tuning, using one-shot learning, few-shot learning, or the like. Furthermore, a general-purpose learning model can also handle various tasks using zero-shot learning. The artificial intelligence used in each functional unit of the control unit 23 may be a separate learning model, or a common general-purpose learning model. A large-scale language model is a type of generative AI, and includes models provided by services such as OpenAI's GPT, Google's Gemini, and Microsoft's Azure AI Studio. Additionally, the artificial intelligence unit 238 can include any machine learning model, deep learning model, artificial intelligence model, etc. The artificial intelligence unit 238 may be constructed in a system external to the information processing system 1. Furthermore, the artificial intelligence unit 238 may be of an interactive type (which may be interpreted as a chat type or a conversation type) that alternately receives input for performing instructed output and generates and outputs information.

[0047] The learning model included in the artificial intelligence unit 238 can undergo additional learning using techniques such as transfer learning or fine tuning. For example, the artificial intelligence unit 238 learns whether the output content has been modified by a user or the like. That is, the artificial intelligence unit 238 may perform additional learning and fine tuning based on modifications to the content output by the learning model. Also, for example, each time new data is registered, the artificial intelligence unit 238 may perform additional learning and fine tuning using the new data as new training data. This improves the accuracy of the information output from the learning model.

[0048] The learning model included in the artificial intelligence unit 238 may be a learning model (distilled model) obtained by knowledge distillation using an original trained model. In knowledge distillation, a trained model such as a large-scale language model is used as a teacher model, and the parameters of the student model (distilled model) are adjusted to reduce the output loss (Soft Target Loss) of the student model (distilled model) relative to the output (Soft Target) of the teacher model, thereby learning the student model, which becomes the distilled model. Alternatively, the student model may be trained to reduce the output loss (Hard Target Loss) of the student model relative to the correct label (Hard Target) of the teacher data (a combination of input data and output data of the learning model). Compared to the original trained model (teacher model), the distilled model has a smaller number of parameters and a smaller processing load while maintaining performance similar to the trained model. Therefore, using a distilled model allows the information processing system 1 to reduce costs.

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

[0050] The artificial intelligence unit 238 may include an AI agent. The AI ​​agent may also be called an autonomous agent. An "AI agent" is a model that, when a target (goal, objective, etc.) such as "teach me XX" or a task such as "output XX" is input, breaks down the processing required to reach the goal or accomplish the task into subtasks, actions, etc., and performs the necessary data collection and analysis, program generation, and execution. The AI ​​agent targets information and instructions input by a user, autonomously selects and executes tasks and actions according to the target, and outputs information according to the target, without requiring user intervention (operational input). The AI ​​agent may also autonomously plan and execute the plan, evaluate the execution results, and autonomously learn to achieve the target. For example, the AI ​​agent may be autonomously updated based on the results of subtask execution (e.g., collected information, information analysis results, etc.).

[0051] The artificial intelligence unit 238 may be an external component of the server 2. In this case, the artificial intelligence unit 238, which is an external component, may be provided by, for example, an artificial intelligence service server, and configured to receive inputs from each functional unit of the server 2, receive requests to execute an artificial intelligence service, and return instructed output as a processing result to the server 2. The artificial intelligence service server may be a server that provides services using a language model as a learning model, or may be a server that executes language processing tasks using a language model. The artificial intelligence service server may be constructed using a large-scale language model. The artificial intelligence service server may receive input of prompts using text, images, voice, etc., and generate and respond to the prompts.

[0052] <Judgment section 239> The determination unit 239 is configured to determine whether various information and data satisfy predetermined standards and conditions. For example, the determination unit 239 determines whether there is registered information that can be added to the input information in order to generate search conditions.

[0053] <Display> The display unit 331 of the job seeker terminal 3 displays the screen indicated by the screen data transmitted from the server 2.

[0054] <Operation acquisition section> The operation acquisition unit 332 of the job seeker terminal 3 accepts operations by the job seeker using the job seeker terminal 3. When a job seeker U1 operates the job seeker terminal 3 as a user to use the information processing system 1, the job seeker U1 logs in by entering a user ID (Identification) and password. This associates the user ID with information generated or updated on the job seeker terminal 3, making it possible to identify which user the information pertains to.

[0055] 3. Information processing flow This section describes the flow of an information processing method executed by the information processing system 1. As shown below, the information processing method includes steps executed by the information processing system. The program of this embodiment causes a computer to execute each step of the information processing system. Note that the order of the processes can be changed as appropriate, multiple processes may be executed simultaneously, or some processes may be omitted.

[0056] 3.1 Overview

[0057] 5 is a diagram showing an overview of the processing executed by the information processing system 1. In this processing, first, the input receiving unit 231 receives input information including instructions related to search conditions for a job search from a job seeker in a chat format (step S001). Next, the output information generating unit 232 generates first output information based on the input information and first reference information (step S002). Next, the display control unit 233 displays the first output information in a chat format so that it can be viewed by the job seeker (step S003).

[0058] In summary, an information processing system according to one embodiment includes at least one processor. By reading a program, the processor includes the following units: The input accepting unit 231 accepts input information from a job seeker in a chat format, the input information including instructions regarding search conditions for a job search. The output information generating unit 232 generates first output information based on the input information and first reference information. The first reference information includes at least the relationship between the input information and the first output information. The first output information includes search conditions according to the input information. The display control unit 233 displays the first output information in a chat format so that it can be viewed by the job seeker. This aspect makes it possible to provide an information processing system, etc., that can efficiently generate search conditions for searching for information regarding job openings that suit the job seeker.

[0059] 3.2 Specific examples An example of the flow of processing executed by the information processing system 1 will be described below with reference to Figures 6 to 9. This example of the flow may fall within the scope defined in the above-mentioned outline. The information processing may include any exception processing not shown. Exception processing includes the interruption of the information processing or the omission of each process. Selection or input performed in the information processing may be based on a user operation or may be performed automatically without relying on a user operation.

[0060] First Embodiment The first embodiment shows an example of the flow of information processing when an information processing system 1 accepts input information from a job seeker U1 via chat, presents first output information including search conditions generated based on the input information, accepts modified input information including instructions to modify the presented first output information, presents second output information including the modified search conditions, and performs a job search based on the modified search conditions.

[0061] 6 is an activity diagram showing an example of the flow of information processing according to the embodiment, which is executed by the information processing system 1. Below, an explanation will be given along with each activity in this activity diagram.

[0062] First, the input receiving unit 231 receives a login from job seeker U1 to the information processing system (activity A101).

[0063] Next, as a first proposal generation step, the proposal generation unit 234 generates questions related to the job seeker's search conditions or options related to the search conditions based on the registration information related to the job seeker registered in the database and the third reference information. The third reference information includes at least the relationship between the registration information and the questions or options (activity A102).

[0064] The registration information is information that has already been registered in the job seeker information database as information about job seeker U1. The job seeker database is a database that stores information about job seekers received from multiple job seekers in association with the job seeker's identification information. The job seeker U1 performs an operation on the job seeker terminal 3 in advance to register basic information about the job seeker as registration information. When the job seeker terminal 3 receives input of the registration information via the operation acquisition unit 332, the operation acquisition unit 332 transmits the input registration information to the server 2. The server 2 stores the transmitted registration information in the job seeker information database as job seeker information. The basic information about the job seeker is registered information that includes, for example, "basic information" such as the job seeker U1's "name," "age," "address," "date of birth," and "current annual salary"; "desired conditions" such as "desired job type," "desired industry," "desired work location," and "desired annual salary"; "educational background" such as "school name" and "graduation date"; "work history" such as "company name," "department name," "job title," and "term of employment"; and other items such as "awards," "experienced job type," "experienced industry," "qualifications," and "language proficiency." The "experienced job type" and "experienced industry" fields may each allow the user to enter the number of years of experience for each job type or industry. The registered information may also include, for example, a resume, a document that the job seeker uses to inform employers of their work history, experience, skills, qualifications, etc. The registered information about the job seeker may also include information other than the document about the job seeker's work history (e.g., the job seeker's personal information, attributes, behavioral history, etc.). Furthermore, the recording unit 237 may record input information relating to search conditions for job searches received from job seekers in the registration information database.

[0065] If the third reference information is a rule set or template dictionary that associates attribute values ​​of registered information with questions or choice templates, the suggestion generation unit 234 may extract the attribute values ​​included in the registered information and generate questions or choice options by applying the corresponding rules or templates. Alternatively, the suggestion generation unit 234 may use keywords included in the registered information as keys to search for matching records from a prepared question database or choice database, and output the results as is or in a formatted form.

[0066] The third reference information includes a learning model that has been machine-learned to take the registered information as input and output questions or options, or a proposal generation model that is a generative AI. The proposal generation unit 234 inputs the registered information to the proposal generation model and causes the proposal generation model to output questions or options. When the third reference information includes a learning model that has been trained (by supervised learning, online learning, transfer learning, etc.) to take the registered information as input and output questions or options related to the registered information, the proposal generation unit 234 may input the registered information to the learning model and cause the learning model to output questions or options as model inference results. The architecture of the learning model is not limited, and the learning model may be a classification model, a sequence generation model, a re-ranking model, or the like.

[0067] If the third reference information is a proposed generative model including a generative AI, the proposal generation unit 234 may input a prompt (including few-shot examples as necessary) with the registered information inserted into the proposed generative model, and output a question or option corresponding to the registered information. After obtaining the question or option output from the proposed generative model, the proposal generation unit 234 may perform post-processing such as format verification, duplication removal, and reliability scoring. The generative AI may include a large-scale language model.

[0068] Furthermore, the suggestion generation unit 234 generates questions or options that include elements that are highly relevant to elements included in the registered information, based on the registered information and third reference information. The third reference information includes information regarding the degree of relevance. When the third reference information includes a prepared question database or option database and an extraction model (which may be rule-based, learning model, or generation AI) that evaluates the degree of correlation with the registered information, the extraction model inputs the registered information and the option DB and extracts a set of highly correlated candidates, and the generation AI summarizes and reconstructs the candidate set as necessary to generate a final question or option.

[0069] The third reference information may be a rule base, a learning model, or a generation AI, either singly or in combination. For example, the proposal generation unit 234 may perform a three-stage process in which questions or option candidates are roughly extracted using a rule base, then naturalized using a large-scale language model, and then re-ranked using a learning model. Alternatively, the proposal generation unit 234 may extract questions or option candidates from a prepared question database or option database, and then generate additional "missing items" using a large-scale language model. Note that the generation of questions or option candidates by the proposal generation unit 234 is not limited to these specific implementation examples.

[0070] Furthermore, the third reference information may be a learning model trained (by machine learning or fine tuning) to receive registered information as input and output questions or options, or a third model that is a generative AI. In other words, the third reference information may include a learning model trained by machine learning to receive registered information as input and output questions or options, or a third model that is a generative AI. In this case, the proposal generation unit 234 executes, as a first proposal generation step, a process of inputting registered information into the third model and causing the third model to output questions or options. Note that in the third model, parameters calculated, tuned, or the like by learning constitute the correlation of the third reference information.

[0071] The third model is included in the artificial intelligence unit 238. The third model trained to be able to output questions or options is, for example, a learning model trained using registered information and the corresponding questions or options as training data.

[0072] When the third model is a generative AI including a general-purpose natural language model (large-scale language model), the proposal generation unit 234 inputs instructions for creating questions or options based on registered information and the registered information to the generative AI, and executes a process of having the generative AI output the questions or options. More specifically, the proposal generation unit 234 inputs the registered information and inputs a prompt including instructions for outputting questions or options corresponding to the registered information to the third model, and causes the third model to output the questions or options. The proposal generation unit 234 may generate a prompt that instructs the third model to create questions or options and input the prompt to the third model. Furthermore, in addition to the instructions for creating and outputting questions or options and the registered information, the proposal generation unit 234 may input a prompt to the third model that inserts, for example, one or more samples of registered information and one or more samples of questions or options corresponding thereto as examples, samples, or training data of input and output pairs. According to this aspect, questions or options can be estimated with high accuracy by using a machine learning model or a generative AI.

[0073] Next, the display control unit 233 displays a screen for accepting input information (activity A103). Here, the acceptance screen is configured to be able to accept input information from the job seeker U1 in a chat format. For example, the display control unit 233 displays a first input field for accepting input of input information.

[0074] The chat format is a dialogue between the job seeker and the system, where information is exchanged two-way between the job seeker and the system. The chat screen, which displays the chat dialogue, displays inputs received from both the job seeker and the system in chronological order, allowing the user to scroll through a predetermined range of past conversation logs. The predetermined range, for example, refers to conversation logs for a predetermined period, and the "predetermined period" may be, for example, one week, one month, six months, or one year. The chat screen may include an input field for receiving input from the job seeker and be configured to accept input information from the job seeker. The input field may accept text, stamps, audio, files, etc., and may include buttons for accepting operational instructions and selectable options. The inputs received from both the job seeker and the system are displayed in association with the identification information of the job seeker and the system, respectively, so that the senders can be identified. For example, the display control unit 233 displays a first input field that accepts input of input information so that it is visible to the job seeker U1, and the presentation unit 235 presents the questions or options generated by the proposal generation unit 234 on the same screen as the first input field that accepts input of input information.

[0075] Next, the input receiving unit 231 receives input information (activity A104). The input information includes at least instructions regarding search conditions for a job search. The instructions regarding the search conditions may be instructions regarding the content of the search conditions, or may be instructions to generate search conditions or instructions to search using the search conditions.

[0076] The input receiving unit 231 can receive input information including information on job seekers related to the search criteria. The information on job seekers related to the search criteria includes at least one item from among the job seeker's annual salary, job type, work location, industry, and job type.

[0077] The control unit 23 repeats activities A102 to A104 until the input receiving unit 231 receives an instruction to generate search conditions. That is, the proposal generating unit 234 generates questions and options related to the input information of the job seeker U1 based on the input information received from the job seeker U1, the display control unit 233 displays the newly generated questions and options together with additional input fields, and the input receiving unit 231 repeatedly receives the additional input information from the job seeker U1.

[0078] The proposal generation unit 234 may use a large-scale language model to infer differences between the information required to generate search conditions and the input information, and generate questions or options for obtaining the differences. The proposal generation unit 234 may also use a large-scale language model to generate questions and options for concretizing the job seeker U1's latent needs. By displaying the questions or options thus generated in a chat format and repeatedly receiving input information from the job seeker U1, the job seeker's abstract needs can be further refined and made more specific. Next, when the input receiving unit 231 receives an instruction from the job seeker in a chat format to generate search conditions for a job search, the control unit 23 proceeds to activity A105. The input receiving unit 231 may receive text such as "Suggest search conditions" as the generation instruction.

[0079] Next, the determination unit 239 determines whether or not there is any registered information that can be added to the input information received in activity A104 (activity A105). For example, the determination unit 239 determines whether or not there is any information that is missing from the input information but can be supplemented by the registered information.

[0080] Next, when the determination unit 239 determines that there is registered information that can be added, if the information of the job seeker related to the search criteria included in the input information satisfies a predetermined condition, the adding unit 236 adds the registered information related to the job seeker registered in the database to the input information (activity A106). For example, based on an instruction related to the search criteria included in the input information, the information of the job seeker required to generate the first output information can be added to the input information from the registered information saved in the registered information database.

[0081] When the adding unit 236 adds the information to the input information, the control unit 23 returns to activity A105.

[0082] Subsequently, if the determining unit 239 determines that there is no registration information that can be added, the control unit 23 proceeds to activity A107.

[0083] Next, the output information generating unit 232 generates search conditions based on the input information (activity A107).

[0084] The output information generation unit 232 inputs the input information into the search condition generation model and causes the search condition generation model to output search conditions. The first reference information includes a search condition generation model, which is a machine-learned learning model or a generative AI that uses the input information as input and is capable of outputting search conditions. The output information generation unit 232 is configured to generate search conditions suitable for the job seeker U1 based on input information (e.g., desired job type, work location, annual salary range, possessed skills, etc.) input from the job seeker U1. The output information generation unit 232 may generate search conditions to be used for job searches based on the input information acquired by the input accepting unit 231. The output information generation unit 232 may generate search conditions by referring to a predetermined mapping rule or template dictionary and directly converting the attributes of the input information (e.g., work location = "Tokyo", remote work = "available") into query parameters. The rule-based method is lightweight and fast, and is effective when the input information is schematized. The output information generation unit 232 may generate search conditions by estimating an optimal parameter set (e.g., industry code, minimum annual salary, skill keywords, etc.) from the input information using a learning model (e.g., classification / regression / reranking model) that uses past input information and hiring records as training data. After estimation, a threshold filter or weight correction rule may be applied to improve the accuracy of the search conditions. The output information generation unit 232 may input a prompt with the input information inserted into a large-scale language model (LLM) and generate search conditions by adding an instruction such as "Generate a JSON-formatted query for job searches taking into account the following profile." The output information generation unit 232 dynamically generates prompts and templates that embed the input information to control the input to the artificial intelligence unit 238. The output information generation unit 232 may include an instruction in the prompt such as "Output the industry code, work location, minimum annual salary, and remote work availability in JSON format taking into account the input information." The output information generation unit 232 may form job seeker information by merging the input information with pre-registered information (e.g., work history, skills, qualifications, etc.), and may instruct the generation of search conditions based on that information. Note that the generation of search conditions in the output information generation unit 232 is not limited to these specific implementation examples.

[0085] Furthermore, the first reference information may be a learning model trained (by machine learning or fine tuning) to use input information as input and search conditions as output, or a first model (search condition generation model) that is a generative AI. In other words, the first reference information may include a learning model trained by machine learning to use input information as input and be able to output search conditions, or a first model that is a generative AI. In this case, the output information generation unit 232 executes, as an output information generation step, a process of inputting the input information into the first model and causing the first model to output search conditions. Note that in the first model, parameters calculated, tuned, etc. by learning constitute the correlation of the first reference information.

[0086] The first model is included in the artificial intelligence unit 238. The first model that has been trained to be able to output search conditions is, for example, a learning model that has been trained using input information and the corresponding search conditions as training data.

[0087] When the first model is a generative AI including a general-purpose natural language model (large-scale language model), the output information generation unit 232 inputs instructions for creating search conditions based on input information and the input information to the generative AI, and executes a process of causing the generative AI to output the search conditions. More specifically, the output information generation unit 232 inputs the input information and inputs a prompt including instructions for outputting search conditions corresponding to the input information to the first model, causing the first model to output the search conditions. The output information generation unit 232 may generate a prompt that instructs the first model to create search conditions and input the prompt to the first model. Furthermore, in addition to the search condition creation and output instructions and the input information, the output information generation unit 232 may input a prompt to the first model that inserts, for example, one or more samples of input information and one or more corresponding samples of search conditions as examples, samples, or training data of input and output pairs. According to this aspect, search conditions can be estimated with high accuracy by using a machine learning model or a generative AI.

[0088] The recording unit 237 records information such as input information entered by the job seeker and first output information displayed to the job seeker as a conversation log. Each search condition generation model includes a large-scale language model. The large-scale language model has access to the conversation log of the job seeker. The search condition generation model interprets the input information using the large-scale language model and generates search conditions based on the interpretation results.

[0089] Next, the output information generation unit 232 generates first output information (activity A108). The output information generation unit 232 generates the first output information based on the search conditions. The output information generation unit 232 receives the generated search conditions as input and generates the first output information to be presented to job seeker U1. The output information generation unit 232 may receive the search conditions generated by the search condition generation model or rule processing, and generate the first output information including the search conditions and information related to the search conditions.

[0090] The output information generation unit 232 may refer to a predetermined template dictionary or mapping rules, and develop each item of the search criteria received as input into a table layout to generate first output information. The output information generation unit 232 may use a first output information generation model that uses past search criteria and past first output information as training data to generate the first output information using the search criteria as input. The output information generation unit 232 may embed the search criteria in a prompt in JSON format, add instructions such as "Generate a sentence that explains the following criteria in an easy-to-understand manner for the user and presents it along with an approximate number of hits," and input the resulting information into a large-scale language model to generate first output information (e.g., natural language explanation + table format). The output information generation unit 232 dynamically generates prompts and templates with embedded search criteria to control input to the artificial intelligence unit 238.

[0091] Next, the display control unit 233 displays the first output information (activity A109). The display control unit 233 displays the first output information in a chat format so that it can be viewed by the job seeker. The first output information includes the generated search conditions and information about the generated search conditions. The first output information includes, as information about the generated search conditions, information indicating that the search conditions have been generated in response to instructions from job seeker U1 regarding the search conditions. By displaying the first output information (e.g., the generated search conditions) so that it can be viewed by the job seeker, the job seeker can determine whether the search conditions are appropriate before executing a search.

[0092] The first output information includes information about the number of hits obtained when searching using the search criteria. This allows job seeker U1 to consider whether to modify his or her search criteria based on the number of hits obtained using the current search criteria. For example, if the number of hits is greater than desired, the job seeker can change the search criteria to narrow down the number of hits further, or if the number of hits is less than desired, the job seeker can change the search criteria to increase the number of hits.

[0093] Here, the display control unit 233 may display a second input field together with the first output information. The second input field can accept input information related to correction. Furthermore, the display control unit 233 may display the second input field that accepts input of correction input information, and the presentation unit 235 may present suggestions for correction to the search criteria on the same screen as the second input field.

[0094] The proposal generation unit 234 generates a proposal for correction to the search criteria. In a second proposal generation step, the proposal generation unit 234 generates a proposal for corrected input information based on registration information about job seekers registered in the database, input information received from the job seekers, and fourth reference information. The fourth reference information includes at least the relationship between the registration information, the input information, and the proposal. For example, the fourth reference information includes a rule set or template dictionary that associates attribute values ​​of the registration information and input information with correction candidate templates. The proposal generation unit 234 compares the attribute values ​​included in the registration information with the values ​​included in the input information, detects a mismatch or deficiency, and then applies the corresponding rule to generate a proposal for correction, such as "Shall we limit the work location to 'Kanto'?". The proposal generation unit 234 also searches the correction candidate database using keywords from the registration information or input information as keys, and the obtained candidates may be used as proposals either directly or after being formatted.

[0095] The fourth reference information may also include a learning model trained to input the registered information and the input information and output correction suggestions related to them. The proposal generation unit 234 inputs both pieces of information into the learning model and obtains correction suggestions obtained as an inference result of the model. The architecture of the learning model is not limited, and a classification model, a sequence generation model, a re-ranking model, etc. can be used as appropriate. The fourth reference information may also be a proposal generation model equipped with a generative AI (including a large-scale language model). In this case, the proposal generation unit 234 generates a prompt into which the registered information and the input information are inserted. The proposal generation unit 234 may also perform a three-stage process of roughly extracting difference items using a rule base, calculating priorities using a learning model, and naturalizing and completing missing items using a large-scale language model. The proposal generation unit 234 may also reconstruct and summarize a candidate set extracted from a correction candidate database using a generative AI to generate a final correction suggestion. Note that the generation of correction suggestions by the proposal generation unit 234 is not limited to these specific implementation examples.

[0096] Furthermore, the fourth reference information may be a learning model trained (by machine learning or fine tuning) to use the registered information and the input information as input and output a correction suggestion, or a fourth model that is a generative AI. In other words, the fourth reference information may include a learning model trained by machine learning to use the registered information and the input information as input and output a correction suggestion, or a fourth model that is a generative AI. In this case, as a second proposal generation step, the proposal generation unit 234 inputs the registered information and the input information into the fourth model and executes a process of causing the fourth model to output a correction suggestion. Note that in the fourth model, parameters calculated, tuned, etc. by learning constitute the correlation of the fourth reference information.

[0097] The fourth model is included in the artificial intelligence unit 238. The fourth model trained to be able to output correction suggestions is, for example, a learning model trained using the registered information, input information, and corresponding correction suggestions as training data.

[0098] When the fourth model is a generative AI including a general-purpose natural language model (large-scale language model), the proposal generation unit 234 inputs the registration information and input information, along with an instruction to create a correction proposal based on the registration information and input information, into the generative AI, and executes a process of causing the generative AI to output the correction proposal. More specifically, the proposal generation unit 234 inputs the registration information and input information, inputs a prompt including an instruction to output a correction proposal corresponding to the registration information and input information into the fourth model, and causes the fourth model to output the correction proposal. The proposal generation unit 234 may generate a prompt that instructs the fourth model to create a correction proposal and input the prompt to the fourth model. Furthermore, the proposal generation unit 234 may input, in addition to the instruction to create and output the correction proposal and the registration information and input information, a prompt that inserts, for example, one or more samples of the registration information and the input information and one or more corresponding correction proposal samples as examples, samples, or training data of input and output pairs into the fourth model. According to this aspect, by using a machine learning model or a generative AI, it is possible to estimate correction suggestions with high accuracy.

[0099] The generated suggested revisions are presented to the job seeker U1, who then accepts or modifies the proposal, inputting it as revised input information. The input revised input information is added to the recording unit 237 as a conversation log, and is also provided to the search criteria generation model, where it is used to regenerate search criteria and update output information. This enables the information processing system 1 to gradually optimize search criteria through an interactive feedback loop.

[0100] Incidentally, when optimizing the search conditions step by step, the job seeker U1 may repeatedly input the modified input information while gradually changing the content of the modified input information. Therefore, when multiple pieces of modified input information are received, the output information generation unit 232 generates the second output information using the modified input information received later.

[0101] Next, job seeker U1 determines whether the displayed search conditions require correction, and if it is determined that correction is necessary, he / she inputs correction input information, and operation acquisition unit 332 accepts the correction input information from the job seeker. Input acceptance unit 231 accepts correction input information from control unit 33 via the network. Input acceptance unit 231 accepts correction input information via a second input field (activity A110). The second input field is displayed on the chat screen, and input acceptance unit 231 accepts correction input information including instructions for correcting the search conditions from the job seeker in a chat format.

[0102] Next, the output information generating unit 232 generates the corrected search conditions (activity A111).

[0103] The output information generation unit 232 inputs the search criteria and the modified input information into the modified search criteria generation model and causes the modified search criteria generation model to output the modified search criteria. The second reference information includes the modified search criteria generation model, which is a learning model or generation AI that has been machine-learned to be able to output modified search criteria using the search criteria and the modified input information as input.

[0104] The output information generation unit 232 receives existing search criteria and modified input information received from job seeker U1 as input and supplies them to a modified search criteria generation model to generate modified search criteria. The modified search criteria generation model is defined as second reference information and may be implemented using one or a combination of the following: (1) rule-based processing; (2) machine learning model; or (3) generative AI (including large-scale language models), configured to receive search criteria and modified input information and output modified search criteria. For example, when rule-based processing is used, the output information generation unit 232 references a preset modified mapping rule or template dictionary to extract differences between the existing search criteria and the modified input information. The output information generation unit 232 then generates modified search criteria by inserting new values ​​into templates corresponding to the differences. When a learning model is used, an estimation model (such as a classification model, sequence generation model, or reranking model) is trained using past search conditions, modified input information, hiring records, etc. as training data, and the output information generation unit 232 inputs the search conditions and modified input information into the model to generate modified search conditions. When a generation AI is used, the output information generation unit 232 embeds the search conditions and modified input information in a prompt in JSON format, adds a command such as "Generate a new job search query in JSON while reflecting the following modification instructions," and inputs the resulting command into the large-scale language model. The large-scale language model interprets the prompt and generates modified search conditions.

[0105] Furthermore, when the correction input information is input in natural language, the output information generation unit 232 may first execute a step of interpreting the correction instruction using a large-scale language model and extracting the attribute to be changed and the old and new values. Based on the interpretation result, the output information generation unit 232 can flexibly respond to correction instructions written in free text by generating corrected search conditions using the rule-based processing, machine learning model, or generation AI. Note that the generation of corrected search conditions by the output information generation unit 232 is not limited to these specific implementation examples.

[0106] Furthermore, the second reference information may be a learning model trained (by machine learning or fine tuning) to receive search criteria and modified input information as input and output modified search criteria, or a second model (modified search criteria generation model) that is a generative AI. In other words, the second reference information may include a learning model trained by machine learning to receive search criteria and modified input information as input and output modified search criteria, or a second model that is a generative AI. In this case, the output information generation unit 232 executes, as an output information generation step, a process of inputting the search criteria and modified input information into the second model and outputting the search criteria modified by the second model. Note that in the second model, parameters calculated, tuned, etc. by learning constitute the correlation of the second reference information.

[0107] The second model is included in the artificial intelligence unit 238. The second model trained to be able to output modified search conditions is, for example, a learning model trained using search conditions, modified input information, and the corresponding modified search conditions as training data.

[0108] When the second model is a generation AI including a general-purpose natural language model (large-scale language model), the output information generation unit 232 executes a process of inputting an instruction to create modified search conditions based on the search conditions and the modified input information, as well as the search conditions and the modified input information, to the generation AI, and causing the generation AI to output the modified search conditions. More specifically, the output information generation unit 232 inputs the search conditions and the modified input information, and inputs a prompt to the second model including an instruction to output modified search conditions corresponding to the search conditions and the modified input information, and causes the second model to output the modified search conditions. The output information generation unit 232 may generate a prompt that instructs the second model to create modified search conditions, and input the prompt to the second model. Furthermore, the output information generation unit 232 may input to the second model, in addition to the instruction to create and output the modified search conditions, the search conditions, and the modified input information, a prompt that includes, for example, one or more samples of the search conditions and the modified input information and one or more corresponding samples of the modified search conditions as examples, samples, or training data of input and output pairs. According to this aspect, by using a machine learning model or a generation AI, it is possible to estimate modified search conditions with high accuracy.

[0109] The generated revised search criteria are formatted as second output information together with an estimate of the number of hits and a search screen transition object (e.g., link information for executing a search), and presented to job seeker U1. If job seeker U1 makes further revisions after presentation, the revised input information is added to the conversation log, and the above series of processes is repeated. This allows the system to gradually optimize the search criteria through an interactive feedback loop, enabling it to accurately reflect job seeker U1's potential needs.

[0110] The revised search criteria generation model is a machine-learned learning model or generation AI capable of outputting revised search criteria using search criteria and a conversation log containing revised input information as input. Both revised search criteria generation models include a large-scale language model. The large-scale language model has access to the job seeker's conversation log. The revised search criteria generation model interprets the search criteria and the conversation log containing the revised input information using the large-scale language model and generates revised search criteria based on the interpretation results. The output information generation unit 232 may implement a feedback loop in which the list of missing items obtained as the interpretation result of the large-scale language model is presented as an additional question, and the additional input information is added to the conversation log again and the search criteria are regenerated. This allows the search criteria to be gradually optimized as the dialogue progresses. This allows job seeker U1 to arrive at optimal search criteria without having to change the search criteria and repeat the search.

[0111] Next, the output information generation unit 232 generates second output information (activity A112). The output information generation unit 232 generates the second output information including the modified search criteria based on the modified input information, the search criteria, and the second reference information. The second reference information includes at least the relationship between the modified input information, the first output information, and the second output information. The output information generation unit 232 may receive search criteria modified by a modified search criteria generation model or rule processing and generate second output information for visualizing and explaining the content of the modified search criteria. The second reference information includes the correspondence between the modified input information and the first and second output information. The output information generation unit 232 may refer to a predetermined template dictionary or mapping rule, and develop the modified search criteria received as input into a table layout to generate the second output information. The output information generation unit 232 may use a modified search criteria generation model that uses past modified input information and search criteria as training data to estimate optimal descriptions, highlight order, supplemental tags, etc. using the modified search criteria as input, and generate the second output information. The output information generation unit 232 may highlight the corrected parts. The output information generation unit 232 may embed the corrected search conditions in a prompt in JSON format, add instructions such as "Explain the following corrected search conditions in an easy-to-understand manner for job seekers, and present them together with the expected number of hits," and input them into a large-scale language model to generate second output information.

[0112] Next, the display control unit 233 displays the second output information (activity A113). The display control unit 233 displays the second output information in a chat format so that it can be viewed by the job seeker. Here, the second output information may highlight the portions of the search conditions that have been corrected based on the instruction for correction. This makes it easier for the job seeker U1 to recognize the corrections. The second output information may also include information regarding the number of hits when a search is performed using the corrected search conditions. The display control unit 233 may also display the number of hits before and after the correction so that they can be compared.

[0113] Next, the job seeker determines whether or not the displayed search conditions need to be modified, and if it is determined that no modifications are necessary and the search conditions are OK as they are, the operation acquisition unit 332 receives an instruction to execute the search conditions from the job seeker (activity A114). The input acceptance unit 231 accepts the instruction to execute the search conditions in a chat format.

[0114] The control unit 23 searches for job information using the search criteria and extracts job information that matches the search criteria. Here, job information is information in a job posting by a job recruiter. Employers include organizations such as for-profit corporations (e.g., companies), non-profit corporations (e.g., cooperatives, foundations), and public corporations (e.g., local governments). Employers also include recruitment agencies that act as agents of organizations to mediate between job seekers and organizations. Recruitment agencies are also called headhunters or agents. A job posting contains the details of a job offer, and the job information is listed in multiple categories, such as the name of the position being offered, the job content and working conditions (annual salary, job type, industry, work location, working style, work environment, etc.), application qualifications (skills), the desired personality, and appealing points. Job postings may also include items such as the job title, headline, and information about the recruiter (company size (sales, number of employees, etc.), industry, etc.). The job search includes extracting job information that matches the search criteria from job listings registered in the job database. The search results are the results of extracting the job information.

[0115] Next, the display control unit 233 displays the search results (activity A115). For example, the display control unit 233 displays a list of job information extracted as meeting the search criteria as the search results. Here, the display control unit 233 displays the search results based on the search criteria on a screen different from the screen displaying the chat format in a manner that is visible to the job seeker.

[0116] Subsequently, in activity A115, the input accepting unit 231 may display the search results and then accept correction input information including instructions for correcting the search results. More specifically, when the control unit 23 accepts correction input information from the operation acquiring unit 332 on the chat-style screen on which the instruction to execute the search conditions was accepted, the process returns to activity A110, and the input accepting unit 231 accepts the entered correction input information. In this manner, even after the search results have been displayed, correction input information can be accepted in chat style to correct the search conditions. In this manner, search conditions for searching for information on job openings that suit the job seeker can be efficiently generated.

[0117] 7 is a diagram showing a screen G1 which is an example of a chat screen. The screen G1 includes an area G11, an area G12, an area G13, an area G14, an area G15, and an object G16, and the area G14 includes areas G141 to G145.

[0118] Area G11 is an example of a first input field that accepts input information from job seeker U1. Area G12 is an area that displays identification information for job seeker U1. The identification information may be, for example, the name, account name, photo, or avatar icon of job seeker U1. Area G12 is displayed in association with area G11, and it is clear that area G11 is input information entered by job seeker U1. Area G13 is displayed in association with area G14, and is an area that indicates that area G14 is a response from the information processing system 1. Area G13 displays identification information that indicates that the response is from the information processing system 1. The identification information may be, for example, an avatar icon or character that indicates that it is the information processing system 1. Area G14 is an area that displays first output information that includes search conditions corresponding to the input information. Area G12 is an example of an area that displays all of the response messages generated by the information processing system 1. Area G14 has multiple areas G141 to G145 arranged therein.

[0119] Area G11 is an example of input information received by the input receiving unit 231 in activity A102. The input receiving unit 231 can receive input information from a job seeker in a chat format, including instructions regarding search conditions for a job search. Area G11 displays the input information "Suggest search conditions" entered by the job seeker U1. The input information including instructions regarding search conditions for a job search may be entered directly by the job seeker U1 in a chat format, or may be an instruction to suggest search conditions without including information about the job seeker, as in area G11. When the input receiving unit 231 receives an instruction to suggest search conditions, the process proceeds to activity A105. In activity A105, if the determination unit 239 determines that there is registration information that can be added (e.g., desired job type, desired industry, desired work location, desired annual salary), the addition unit 236 adds the registration information to the input information, and in activity A107, the output information generation unit 232 generates search conditions based on the registration information of the job seeker U1 and the first reference information. Proposal pattern 1 in area G142 is an example of a search condition. Furthermore, in activity A108, the output information generation unit 232 generates first output information including the search condition (proposal pattern 1). Area G14 is an example of this first output information.

[0120] Area G141 is an example of an area displaying a message from the system to job seeker U1 regarding the first output information. Area G141 displays, "We've created search criteria based on the information you provided. How about the following criteria?" Area G141 may also include information indicating that search criteria have been generated. An explanation of the proposed search criteria can be displayed. Area G142 is an example of an area displaying generated search criteria. Area G142 contains a table for "Proposal Pattern 1," with headings for industry, job type, annual salary, work location, remote work availability, other criteria, and keywords, and the contents of each item displayed in the corresponding cell. Area G143 is an example of an area displaying the expected number of job search hits (the number of jobs matching the search criteria) when the search criteria for "Proposal Pattern 1" in area G142 are executed. Area G143 displays, "120 hits for these criteria." This allows job seeker U1 to grasp the number of results before executing a search. The number of hits may be an estimated value calculated by the output information generation unit 232 based on the search criteria of proposed pattern 1 and the job information database. The number of hits for the job search in area G143 may or may not be displayed. Area G144 is an example of an area that accepts an instruction to execute the search criteria presented in area G142. Area G144 displays "Search with these criteria." Area G145 is an example of an area that displays comments to the job seeker U1 regarding the search criteria presented in area G142. Area G145 displays "Are these criteria OK? If you have any further corrections or additions, please fill out the form below and submit." An instruction to execute the search criteria presented in area G142 may be accepted from the job seeker U1 through area G144, and a search may be executed. This allows for efficient generation and execution of search criteria for searching for information related to job openings that suit the job seeker.

[0121] Area G15 is an area that accepts chat-style input from job seeker U1. Area G15 is an area that accepts chat-style input to the system and can accept input information or revised input information. The input accepting unit 231 can accept free text input from area G15. The free text input may describe desired conditions, or may accept numerical values ​​or character strings such as "negotiable." The input accepting unit 231 can also accept additional conditions not included in the above items. The text entered in area G15 is sent to the control unit 23 when object G16 is pressed. When input information or revised input information is entered in area G15 and object G16 is pressed, the acquisition unit 231 accepts the input information or revised input information entered in area G15.

[0122] FIG. 8 shows screen G2, an example of a chat screen. Screen G2 includes areas G143, G144, G145, G21, G22, G23, G24, G25, G26, G27, and object G28. Screen G2 is a chat screen following screen G1, and areas G143, G144, and G145 were also included in screen G1. Areas G21 to G26 are input fields that accept correction input information from job seeker U1. Areas G21 to G26 are configured to be able to accept correction input information. Areas G21 to G26 correspond to each item presented in proposal pattern 1, and may allow partial correction on an item-by-item basis or input that corrects multiple items collectively. Display control unit 233 displays the first output information in a manner that allows for the acceptance of correction input information for search conditions for each search item. In this manner, the job seeker U1 can give instructions to correct each item of the search conditions.

[0123] Area G21 is an input field for accepting input or editing of information regarding the desired industry from job seeker U1. Area G22 is an input field for accepting input or editing of information regarding the desired job type from job seeker U1. Area G23 is an input field for accepting input or editing of information regarding the desired annual salary from job seeker U1. Area G23 is an input field for entering the lower limit or range of the desired annual salary. Area G24 is an input field for accepting input or editing of information regarding the desired work location from job seeker U1. Area G24 is a field for entering the desired work location. Area G25 is an input field for accepting input or editing of information regarding remote work preferences from job seeker U1. Area G26 is an input field for accepting input or editing of information regarding other conditions from job seeker U1. Area G27 is an input field for accepting free input from job seeker U1.

[0124] The presentation unit 235 may present the content of proposal pattern 1 (e.g., consulting) as an initial value for each item in areas G21 to G26. The presentation unit 235 may also present proposals generated by the proposal generation unit 234 in areas G21 to G26. For example, the proposal may be "Tokyo" or "full remote" for the work location item, or may exemplify the availability and frequency of remote work, such as "yes, no, up to two days a week available" for the remote work item. The presentation unit 235 may also present options for the proposal. Areas G21 to G26 may be configured to accept selection of options using buttons, pull-down menus, or the like. This configuration can assist job seeker U1 in making corrections. Areas G21 to G26 may accept free text input, or free text input may be accepted from area G27. The free text input may describe desired conditions, or may allow numerical values ​​or character strings such as "negotiable." The input receiving unit 231 can also receive additional conditions not included in the above items.

[0125] On screen G2, "18 million yen or more" has been entered as correction input information in area G23. The display form of "18 million yen or more" in area G23 is different from the proposals presented in the other areas G21, G22, and areas G24 to G26, and it can be seen that a correction has been entered in area G23.

[0126] Object G28 is a button for transmitting the corrected input information that the job seeker U1 has input in areas G21 to G27 to the information processing system 1. When the operation acquisition unit 332 receives a selection for object G28 from the job seeker U1, it transmits the corrected input information to the control unit 23. When the input reception unit 231 receives the corrected input information, it records the received information in the conversation log and provides the corrected input information to the corrected search condition generation model. The output information generation unit 232 regenerates the corrected search conditions, and the proposal generation unit 234 generates additional proposals in accordance with the corrected search conditions. The output information generation unit 232 generates second output information including the corrected search conditions and the newly generated additional proposals, and the display control unit 233 can display new search results and additional questions.

[0127] 9 is a diagram showing screen G3, which is an example of a chat screen. Screen G3 includes areas G30 to G40 and an object G41. Screen G3 is a chat screen following Fig. 7 (screen G1) and Fig. 8 (screen G2), and is a screen that displays an example of second output information including proposed pattern 2 that reflects corrected input information from job seeker U1.

[0128] Area G30 displays an example of the second output information. Area G31 displays identification information for the information processing system 1 and is associated with area G30, indicating that area G30 is output information from the system. Area G32 displays an example of a message from the system to job seeker U1 in the second output information displayed in area G30: "We have created search conditions based on the information you have provided. How about the following conditions?" Area G32 displays an explanation of the purpose of the second output information. Area G33 displays a table for "Proposal Pattern 2," with row headings for industry, job type, annual salary, work location, remote work availability, other conditions, and keywords. Items modified from Proposal Pattern 1 are highlighted with shading, etc. This is an example of an area displaying the expected number of job search hits when the search conditions of "Proposal Pattern 2" in area G34 are executed. Area G34 displays "50 hits for these conditions (for reference, 120 hits for the conditions before modification)." The display control unit 233 may display the number of hits before and after modification in this way so that the number of hits can be compared. Area G35 is an example of an area that accepts an instruction to execute the search conditions presented in area G33. Area G35 displays "Search with these conditions." Area G36 is an example of an area that displays comments to the job seeker U1 regarding the search conditions presented in area G33. Area G36 displays a comment to the job seeker U1, such as "How about these conditions? If you have any further requests for corrections or additions, please fill out the form below and submit it." Area G37 is an input field for the job seeker U1 to input or edit the industry desired. Area G38 is an input field for the job seeker U1 to input or edit the job type desired. Area G39 is an input field for job seeker U1 to input or edit the lower limit or range of the desired annual income. Area G40 is an input field for receiving free input from job seeker U1.

[0129] After the user inputs additional corrections in areas G36 to G40, when the user operates object G41, the adding unit 236 receives the input content, adds the input content to the conversation log, and provides it to the corrected search condition generation model. Subsequently, a similar feedback loop is used to generate and display proposed patterns 3 and 4, search conditions that are optimized to the user's needs in stages. As described above, screen G3 realizes an interactive chat UI that allows the second stage of search condition proposals that reflect user feedback and further fine-tuning input on the same screen.

[0130] In the above embodiment, the input receiving unit 231 receives text information as input information, but the input information does not have to be limited to text information. The input information received by the input receiving unit 231 is not limited to text information, and may include a network address (URL) of public information, a data file uploaded from the job seeker terminal 3, voice data, etc. The same applies to corrected input information.

[0131] In the above embodiment, the input receiving unit 231 receives an instruction to generate search conditions as an instruction related to the search conditions, but the input receiving unit 231 may also receive an instruction to search for job information as an instruction related to the search conditions. The output information generating unit 232 can generate search conditions in response to the search instruction.

[0132] In the above embodiment, the first output information includes the search criteria and an object that accepts an instruction to execute the search criteria, but the first output information may also include search results. The display control unit 233 may display the search results together with the search criteria in the first output information, and the input accepting unit 231 may accept an instruction to modify the search criteria from the job seeker U1. In this manner, the job seeker U1 can confirm the search results and then issue an instruction to modify the search criteria.

[0133] In the above embodiment, the output information generation unit 232 generates search criteria and then generates first output information including the search criteria. However, the output information generation unit 232 may generate the first output information including the search criteria based on the input information and the first reference information. Similarly, in the above embodiment, the output information generation unit 232 generates modified search criteria and then generates second output information including the modified search criteria. However, the output information generation unit 232 may generate the second output information including the modified search criteria based on the modified input information, the search criteria, and the second reference information. Furthermore, the suggestion generation unit 234 may generate questions or options including elements highly related to elements included in the registered information based on the registered information and the third reference information. Furthermore, the suggestion generation unit 234 may generate suggestions regarding the modified input information based on the registered information, the input information, and the fourth reference information.

[0134] In the above embodiment, the output information generation unit 232 generates first output information including search criteria based on the input information and first reference information, and generates second output information including modified search criteria based on the modified input information, search criteria, and second reference information. However, the first reference information and the second reference information may include the same model. In this case, the output information generation unit 232 may share the same model, but use "initial weights" when generating the first output information and dynamically replace "weights for modified input information" when generating the second output information. The output information generation unit 232 may share the same large-scale language model, but use different prompts when generating the first output information and the second output information. Similarly, the first reference information, second reference information, third reference information, and fourth reference information may include the same model. In this case, the proposal generation unit 234 may be configured to share the same model as the output information generation unit 232, but use "weights for questions or option options including elements highly related to elements included in the registered information" when generating questions or option options including elements highly related to elements included in the registered information, and dynamically replace "weights for proposals related to the corrected input information" when generating proposals related to the corrected input information. The proposal generation unit 234 may be configured to share the same large-scale language model as the output information generation unit 232, but use different prompts when generating questions or option options including elements highly related to elements included in the registered information and when generating proposals related to the corrected input information.

[0135] 4. Effect This embodiment can assist job seeker U1 in generating a search query to search for information related to job openings that suit him / her. Job seeker U1 inputs input information into the system in chat format. During the chat-style dialogue, the presentation unit 235 may present questions and options based on the job seeker's registration information already received or information received during the chat-style dialogue. Job seeker U1 can then enter answers based on the questions and options presented by the presentation unit 235. This embodiment can concretely express the abstract desires of job seeker U1. Furthermore, the display control unit 233 displays search conditions that meet the job seeker U1's needs in chat format to the job seeker U1. The job seeker U1 can then respond in chat format to requests for revisions to the displayed search conditions. The output information generation unit 232 generates revised search conditions in response to the job seeker U1's request for revisions. This embodiment allows search conditions that are more suited to the job seeker to be generated through chat-style interactions. Because search conditions are refined through exchanges within the same chat, the output information generation unit 232 can generate search conditions suited to job seeker U1 while referring to past exchanges within the chat.

[0136] Although the embodiment of the present invention has been described above, the present invention is not limited to this and can be modified as appropriate within the scope of the technical idea of ​​the invention.

[0137] 5.Other In the above embodiment, the server 2 performs various storage and control functions. However, multiple external devices may be used instead of the server 2. That is, various information and programs may be distributed and stored in multiple external devices using blockchain technology or the like. In particular, the artificial intelligence unit 238 may be configured external to the server 2. In this case, the artificial intelligence unit 238, which is an external component, is provided, for example, by an artificial intelligence service server and configured to receive inputs from each functional unit of the server 2, receive requests to execute artificial intelligence services, and return the instructed output as a processing result to the server 2. The artificial intelligence service server may be a server that provides services using a language model as a learning model, or a server that executes language processing tasks using a language model. The artificial intelligence service server may be constructed using a large-scale language model. The artificial intelligence service server receives inputs of prompts such as text, images, and voice, and generates and responds to the prompts.

[0138] The aspect of this embodiment is not limited to the information processing system 1, and may be an information processing method or a program. The information processing method includes steps executed by the information processing system 1. The program causes a computer to execute the steps of the information processing system 1.

[0139] It may be provided in the following manner.

[0140] (1) An information processing system comprising at least one processor, the processor being configured to execute each of the following steps by reading a program: an input receiving step, in which input information including instructions regarding search conditions for a job search is received from a job seeker in a chat format; an output information generating step, in which first output information is generated based on the input information and first reference information, wherein the first reference information includes at least a relationship between the input information and the first output information, and the first output information includes the search conditions according to the input information; and a display control step, in which the first output information is displayed in a manner visible to the job seeker in the chat format.

[0141] (2) In the information processing system described in (1) above, the input receiving step receives, in the chat format, corrected input information including instructions for modifying the search conditions from the job seeker, and the output information generating step generates second output information including the corrected search conditions based on the corrected input information, the first output information, and second reference information, wherein the second reference information includes at least the relationship between the corrected input information, the first output information, and the second output information, and the display control step displays the second output information in a manner that is visible to the job seeker in the chat format.

[0142] (3) In the information processing system described in (2) above, further, in the first proposal generation step, questions to the job seeker regarding the job seeker's search conditions or options regarding the search conditions are generated based on registration information about the job seeker registered in a database and third reference information, wherein the third reference information includes at least the relationship between the registration information and the questions or the options, and further, in the presentation step, the questions or the options are presented on the same screen as the first input field that accepts input of the input information in chat format.

[0143] (4) In the information processing system described in (3) above, the third reference information includes a proposal generation model, which is a learning model or a generative AI that has been machine-learned to use the registered information as input and be able to output the question or the options, and in the first proposal generation step, the registered information is input into the proposal generation model and the proposal generation model is caused to output the question or the options.

[0144] (5) In the information processing system described in (3) or (4) above, in the first proposal generation step, the question or the option is generated based on the registered information and the third reference information, including elements that are highly related to elements included in the registered information, and wherein the third reference information includes information regarding the degree of relevance.

[0145] (6) In the information processing system described in any one of (1) to (5) above, in the input receiving step, the input information including information on the job seeker related to the search conditions is received, wherein the information on the job seeker related to the search conditions includes at least one item of the job seeker's annual salary, job type, work location, industry, and job type, and in the output information generating step, the first output information is generated based on the input information including the information on the job seeker related to the search conditions and the first reference information.

[0146] (7) In the information processing system described in (6) above, in the addition step, if the information of the job seeker related to the search criteria included in the input information satisfies a predetermined condition, registered information related to the job seeker registered in a database is added to the input information, and in the output information generation step, the first output information is generated based on the added input information and the first reference information.

[0147] (8) In the information processing system according to any one of (1) to (7) above, the first output information includes information on the number of hits when searching using the search criteria.

[0148] (9) In the information processing system described in any one of (2) to (8) above, in the display control step, the first output information is displayed in a manner that allows the correction input information of the search conditions to be accepted for each search item.

[0149] (10) In the information processing system described in (9) above, further, in the second proposal generation step, a proposal regarding the corrected input information is generated based on registration information regarding the job seeker registered in a database, the input information received from the job seeker, and fourth reference information, wherein the fourth reference information includes at least the relationship between the registration information, the input information, and the proposal, in the display control step, a second input field for accepting input of the corrected input information is displayed, and the proposal is presented on the same screen as the second input field, and in the input acceptance step, the corrected input information is accepted via the second input field.

[0150] (11) In the information processing system described in any one of (2) to (10) above, in the output information generation step, when multiple pieces of corrected input information are received, the second output information is generated using the corrected input information received later.

[0151] (12) In the information processing system described in any one of (1) to (11) above, in the input receiving step, an instruction to execute the search conditions is received in the chat format, and in the display control step, search results based on the search conditions are displayed in a manner that is visible to the job seeker on a screen different from the screen that displays the chat format.

[0152] (13) In the information processing system described in (12) above, the input receiving step receives correction input information including an instruction for correcting the search results.

[0153] (14) In the information processing system described in any one of (2) to (13) above, in the display control step, the second output information highlights the parts of the search conditions that have been corrected based on the instructions for correction.

[0154] (15) In the information processing system described in any one of (2) to (14) above, the first reference information includes a search condition generation model that is a learning model or a generative AI that has been machine-learned to take the input information as input and be able to output the search conditions, and the second reference information includes a modified search condition generation model that is a learning model or a generative AI that has been machine-learned to take the search conditions and the modified input information as input and be able to output the modified search conditions, and in the output information generation step, the input information is input to the search condition generation model and the search condition generation model is caused to output the search conditions, and the search conditions and the modified input information are input to the modified search condition generation model and the modified search condition generation model is caused to output the modified search conditions.

[0155] (16) In the information processing system described in (15) above, further, in the recording step, the input information input by the job seeker, the corrected input information, and the first output information and the second output information displayed to the job seeker are recorded as a conversation log, the search condition generation model is a learning model or generation AI that has been machine-learned to take the input information as input and be able to output the search conditions, and the corrected search condition generation model is a learning model or generation AI that has been machine-learned to take the search conditions and the conversation log including the corrected input information as input and be able to output the corrected search conditions.

[0156] (17) In the information processing system described in (16) above, the search condition generation model and the modified search condition generation model both include a large-scale language model, the large-scale language model has access to the conversation log of the job seeker, the search condition generation model interprets the input information using the large-scale language model and generates the search conditions based on the interpretation results, and the modified search condition generation model interprets the conversation log including the search conditions and the modified input information using the large-scale language model and generates the modified search conditions based on the interpretation results.

[0157] (18) The information processing system according to any one of (1) to (17) above, further comprising: a server having the processor; and a terminal that can access the server.

[0158] (19) An information processing method, comprising steps executed by the information processing system according to any one of (1) to (17) above.

[0159] (20) A program for causing a computer to execute each step of the information processing system described in any one of (1) to (17) above. Of course, this is not the case.

[0160] Finally, while various embodiments of the present disclosure have been described, they are presented as examples and are not intended to limit the scope of the invention. The novel embodiments may be embodied in various other forms, and various omissions, substitutions, and modifications may be made without departing from the spirit of the invention. Such embodiments and modifications are intended to be included within the scope and spirit of the invention, as well as within the scope of the inventions and their equivalents as defined in the claims. [Explanation of symbols]

[0161] 1: Information processing system 2: Server 20: Communication bus 21: Communications Department 22: Storage section 23: Control section 231: Input reception section 232: Output information generation unit 233: Display control unit 234: Proposal generation section 235: Presentation section 236: Additional section 237: Recording Department 238: Artificial Intelligence Department 239: Judgment section 3: Job seeker terminal 30: Communication bus 31: Communications Department 32: Storage section 33: Control section 331: Display section 332: Operation acquisition section 34:Display section 35: Input section G1: Screen G11 :Area G12 :Area G13 :Area G14 :Area G141 :Area G142 :Area G143 :Area G144 :Area G145 :Area G15 :Area G16: Objects G2: Screen G21 :Area G22 :Area G23 :Area G24 :Area G25 :Area G26 :Area G27 :Area G28 :Object G3: Screen G30 :Area G31 :Area G32 :Area G33 :Area G34 :Area G35 :Area G36 :Area G37 :Area G38 :Area G39 :Area G40 :Area G41 :Object U1: Job seeker

Claims

1. An information processing system, at least one processor; The processor is configured to execute the following steps by reading the program: In the input reception step, input information including instructions on job search conditions is received from the job seeker via chat, In the output information generating step, first output information is generated based on the input information and first reference information, wherein: the first reference information includes at least a relationship between the input information and the first output information; the first output information includes the search conditions according to the input information and information on the number of hits when searching using the search conditions; In the display control step, the first output information is displayed in the chat format so as to be visible to the job seeker.

2. 2. The information processing system according to claim 1, In the input receiving step, correction input information including instructions for correcting the search conditions is received from the job seeker in the chat format; the output information generating step generates second output information including the modified search criteria based on the modified input information, the first output information, and second reference information, wherein the second reference information includes at least a relationship between the modified input information, the first output information, and the second output information; In the display control step, the second output information is displayed in the chat format so as to be visible to the job seeker.

3. 3. The information processing system according to claim 2, Furthermore, in the first proposal generation step, questions to the job seeker regarding the search conditions of the job seeker or options regarding the search conditions are generated based on registration information about the job seeker registered in a database and third reference information, wherein the third reference information includes at least a relationship between the registration information and the questions or the options; Furthermore, in the presenting step, the question or the options are presented on the same screen as the first input field that accepts input of the input information in the chat format.

4. 4. The information processing system according to claim 3, the third reference information includes a learning model that has been machine-learned to use the registered information as input and that is capable of outputting the question or the option, or a proposed generative model that is a generative AI; In the first proposal generation step, the registered information is input to the proposal generation model, and the proposal generation model is caused to output the question or the options.

5. 4. The information processing system according to claim 3, An information processing system in which, in the first proposal generation step, the question or the option is generated based on the registered information and the third reference information, the question or the option including elements that are highly relevant to elements included in the registered information, and wherein the third reference information includes information regarding the relevance.

6. 2. The information processing system according to claim 1, In the input receiving step, the input information including information of the job seeker related to the search conditions is received, wherein the information of the job seeker related to the search conditions includes at least one item of the job seeker's desired annual salary, job type, work location, industry, and job type; an information processing system, wherein in the output information generating step, the first output information is generated based on the input information including information of the job seeker related to the search criteria and the first reference information.

7. 7. The information processing system according to claim 6, Furthermore, in the adding step, if information about the job seeker related to the search criteria included in the input information satisfies a predetermined condition, registration information about the job seeker registered in a database is added to the input information; In the output information generating step, the first output information is generated based on the added input information and the first reference information.

8. An information processing system, at least one processor; The processor is configured to execute the following steps by reading the program: In the input reception step, input information including instructions on job search conditions is received from the job seeker via chat, In the output information generating step, first output information is generated based on the input information and first reference information, wherein: the first reference information includes at least a relationship between the input information and the first output information; the first output information includes the search condition according to the input information; In the second proposal generation step, a proposal regarding corrected input information for the search criteria is generated based on registration information regarding the job seeker registered in a database, the input information received from the job seeker, and fourth reference information, wherein the fourth reference information includes at least a relationship between the registration information, the input information, and the proposal; In the display control step, the first output information, a second input field for receiving input of the correction input information for each search item, and the proposal are displayed in a manner that is visible to the job seeker in the chat format; the modification input information includes an instruction for modifying the search criteria; the suggestion is presented on the same screen as the second input field; In the input receiving step, the corrected input information is received from the job seeker via the second input field in the chat format; the output information generating step generates second output information including the modified search criteria based on the modified input information, the first output information, and second reference information, wherein the second reference information includes at least a relationship between the modified input information, the first output information, and the second output information; In the display control step, the second output information is displayed in the chat format so as to be visible to the job seeker.

9. An information processing system, at least one processor; The processor is configured to execute the following steps by reading the program: In the input reception step, input information including instructions on job search conditions is received from the job seeker via chat, In the output information generating step, first output information is generated based on the input information and first reference information, wherein: the first reference information includes at least a relationship between the input information and the first output information; the first output information includes the search condition according to the input information; In the display control step, the first output information is displayed in the chat format so as to be visible to the job seeker; In the input receiving step, correction input information including instructions for correcting the search conditions is received from the job seeker in the chat format; In the output information generating step, second output information including the modified search criteria is generated based on the modified input information, the first output information, and second reference information, wherein: the second reference information includes at least a relationship between the modified input information, the first output information, and the second output information; When a plurality of pieces of corrected input information are received, the second output information is generated using the corrected input information received later; In the display control step, the second output information is displayed in the chat format so as to be visible to the job seeker.

10. An information processing system, at least one processor; The processor is configured to execute the following steps by reading the program: In the input receiving step, input information including instructions regarding search conditions for job search and correction input information including instructions regarding corrections to the search conditions are received from the job seeker in a chat format, The search conditions include at least one search item, the correction input information includes at least an instruction regarding correction to the search item received from the job seeker; In the output information generating step, first output information is generated based on the input information and first reference information, and second output information including the modified search criteria is generated based on the modified input information, the first output information, and second reference information, wherein: the first reference information includes at least a relationship between the input information and the first output information; the first output information includes the search condition according to the input information; the second reference information includes at least a relationship between the modified input information, the first output information, and the second output information; In the display control step, the information processing system causes the first output information and the second output information to be displayed in the chat format so as to be visible to the job seeker.

11. An information processing system, at least one processor; The processor is configured to execute the following steps by reading the program: In the input reception step, input information including instructions on job search conditions is received from the job seeker via chat, In a third proposal generation step, questions or options for clarifying the search conditions are generated based on the input information received from the job seeker and fifth reference information, and the questions or options are presented in the chat format, wherein the fifth reference information includes at least a correlation between the input information and the questions or options; In the input receiving step, answers to the questions or the options are received from the job seeker as the additional input information; repeating the input receiving step and the third proposal generating step a plurality of times; In the output information generating step, first output information is generated based on the input information and first reference information, wherein: the first reference information includes at least a relationship between the input information and the first output information; the first output information includes the search condition according to the input information; In the display control step, the first output information is displayed in the chat format so as to be visible to the job seeker.

12. 2. The information processing system according to claim 1, In the input receiving step, an instruction to execute the search condition is received in the chat format; In the display control step, the information processing system displays search results based on the search criteria in a manner that is visible to the job seeker on a screen different from the screen that displays the chat format.

13. 13. The information processing system according to claim 12, In the input receiving step, correction input information including an instruction for correcting the search results is received.

14. 3. The information processing system according to claim 2, In the display control step, the second output information highlights a portion of the search condition that has been corrected based on the instruction for correction.

15. 3. The information processing system according to claim 2, the first reference information includes a search condition generation model that is a learning model or a generation AI that is machine-learned to use the input information as input and be able to output the search conditions, the second reference information includes a modified search criterion generation model that is a machine-learned learning model or a generation AI that is capable of inputting the search criterion and the modified input information and outputting the modified search criterion, In the output information generating step, inputting the input information into the search condition generation model and causing the search condition generation model to output the search conditions; an information processing system that inputs the search criteria and the modified input information into the modified search criteria generation model and causes the modified search criteria generation model to output the modified search criteria;

16. 16. The information processing system according to claim 15, Furthermore, in the recording step, the input information input by the job seeker, the corrected input information, and the first output information and the second output information displayed to the job seeker are recorded as a conversation log; the search condition generation model is a learning model or a generation AI that has been machine-learned to use the input information as input and be able to output the search conditions, The modified search condition generation model is a machine-learned learning model or a generative AI that is capable of outputting the modified search conditions using the search conditions and the conversation log including the modified input information as input.

17. 17. The information processing system according to claim 16, the search criteria generation model and the revised search criteria generation model each include a large-scale language model; the large-scale language model has access to the conversation logs of the job seeker; the search condition generation model interprets the input information using the large-scale language model and generates the search conditions based on the interpretation result; The modified search criterion generation model interprets the conversation log including the search criterion and the modified input information using the large-scale language model, and generates the modified search criterion based on the interpretation result.

18. 18. The information processing system according to claim 1, a server having the processor; and a terminal that can access the server.

19. An information processing method, comprising: An information processing method comprising the steps executed by the information processing system according to any one of claims 1 to 17.

20. A program, A program for causing a computer to execute each step of the information processing system according to any one of claims 1 to 17.

Images