Classification system and method for classifying objects.

Abstract

This classification system comprises an acquisition unit that acquires classification object information, a classification unit that classifies the classification object information in accordance with a classification criterion, and a display device. The classification unit includes: a language processing model (430); and a retrieval-augmented generation (RAG) data storage unit (420) in which a data source for causing the language processing model to execute language processing is stored on the basis of RAG. The RAG data storage unit (420) includes a classification reference data storage unit (422) in which a data source relating to the classification criterion is stored in a format conforming to the RAG. The data source relating to the classification criterion includes a plurality of pieces of classification information established according to the classification criterion. The classification unit determines classification information corresponding to the classification object information using the language processing model (430) and the data source relating to the classification criterion and assumes the configuration of a display device in a state in which the classification object information and the classification information are associated with one another.

Description

Technical Field

[0001] The present disclosure relates to a classification system and a method for classifying classification targets.

Background Art

[0002] In companies and the like, cloud sourcing of human resources is being utilized. In this type of cloud sourcing, recruiters who consider having work performed by external parties need to find human resources with the optimal capabilities for the work tasks.

[0003] Patent Document 1 describes a method of comparing whether information on human resources required for a development project matches information on the company's members and extracting, as search results, members who match the required human resources from the entire organization.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In order to match work tasks and human resources, it is conceivable to classify work tasks with some classification information. For example, for technical work tasks, classification information such as IPC classification and F-term related to patents may be used to classify the work tasks. By classifying work tasks according to classification information and associating and managing work tasks and classification information, users can appropriately search for required work tasks.

[0006] Recently, there has been a movement to utilize rapidly advancing large-scale language models for text analysis. Therefore, applying large-scale language models to the classification of business cases is conceivable. However, large-scale language models have the characteristic that it is difficult to obtain precise answers from them because they infer information during the process of creating answers. In particular, the difficulty of appropriately matching one of the vast number of business cases with one of the vast number of classification information is extremely high. For this reason, there is a problem in that it is difficult to appropriately classify business cases by simply using general large-scale language models.

[0007] This problem does not only occur when classifying job applications. Recruiters may use the resumes of applicants to make hiring decisions for job applications. The skills and other professional competency information listed in the resumes are important factors in the hiring decision. The types of professional competency information are enormous. Therefore, just as with job applications, it is conceivable to apply large-scale language models to classify this kind of professional competency information. The aforementioned problem arises when classifying professional competency information, just as it does when classifying job applications.

[0008] This disclosure was made to solve the problems described above, and its purpose is to appropriately classify the information to be classified. [Means for solving the problem]

[0009] The classification system relating to the first aspect of this disclosure is a classification system for classifying information to be classified that is used by a user, comprising: an acquisition unit for acquiring information to be classified; a classification unit for classifying the information to be classified according to classification criteria; and a display device, wherein the classification unit includes a language processing model and a RAG data storage unit which stores data sources for causing the language processing model to perform language processing based on RAG (Retrieval-Augmented Generation), the RAG data storage unit includes a classification criteria data storage unit which stores data sources relating to classification criteria in a format conforming to RAG, the data sources relating to classification criteria include a plurality of classification information defined by the classification criteria, and the classification unit uses the language processing model and the data sources relating to classification criteria to determine classification information corresponding to the information to be classified and displays the information to be classified on the display device in association with the classification information.

[0010] A method relating to the second aspect of this disclosure is a method for classifying information to be classified for use by a user, the method causing a computer to perform the steps of acquiring information to be classified and classifying the information to be classified according to classification criteria, the computer being configured to access a RAG (Retrieval-Augmented Generation) data storage unit, the RAG data storage unit including a classification criteria data storage unit in which a data source relating to classification criteria is stored in a format conforming to RAG, the data source relating to classification criteria includes a plurality of classification information defined by the classification criteria, and the classifying step includes the step of determining classification information corresponding to the information to be classified using a language processing model and a data source relating to classification criteria, and the step of displaying the information to be classified on a display device in association with the classification information. [Effects of the Invention]

[0011] According to this disclosure, the information to be classified can be appropriately classified. [Brief explanation of the drawing]

[0012] [Figure 1]It is a block diagram showing the overall configuration of the classification system related to Embodiment 1. [Figure 2] It is a block diagram showing an overview of the configurations of the server and the user device. [Figure 3] It is a block diagram for explaining the configuration related to RAG (Retrieval-Augmented Generation) in the server. [Figure 4] It is a diagram for explaining the concepts of the classification method and the matching method by the classification system. [Figure 5] It is a block diagram showing the configuration of the enterprise database. [Figure 6] It is a block diagram showing the configuration of the user database. [Figure 7] It is a block diagram showing the configuration of the case database. [Figure 8] It is a block diagram showing the configuration of the matching database. [Figure 9] It is a block diagram showing the configurations of the classification unit, the AI construction unit, the data import unit, the acquisition unit, the matching unit, and the result display unit. [Figure 10] It is a flowchart showing the processing procedure of the acquisition unit. [Figure 11] It is a flowchart showing the processing procedure of the data import unit. [Figure 12] It is a flowchart showing the processing procedure of the AI construction unit. [Figure 13] It is a flowchart showing the processing procedure of the chat unit (summarization). [Figure 14] It is a flowchart showing the processing procedure of the chat unit (classification). [Figure 15] It is a flowchart showing the processing procedure of the classification storage unit. [Figure 16] It is a flowchart showing the processing procedure of the matching unit. [Figure 17] It is a flowchart showing the processing procedure of the result display unit. [Figure 18] It is a diagram for explaining the concepts of the classification method and the matching method related to Modification 1. [Figure 19] It is a diagram for explaining the concept of the classification method and the matching method related to Modification Example 2. [Figure 20] It is a block diagram showing the outline of the classification system related to Embodiment 2. [Figure 21] It is a block diagram showing the configurations of the sharing server, the recruiter device, and the applicant device. [Figure 22] It is a block diagram showing the configuration of the generation server. [Figure 23] It is a diagram showing an example of the enterprise database. <0000​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​This is a flowchart showing the processing procedure of the dictionary information display unit. [Figure 38] This is a flowchart showing the processing procedures of the Information Organization Department. [Figure 39] This flowchart shows the processing procedure for the resume registration department. [Figure 40] This is a timing chart showing the processing procedures for the sharing server, recruiter device, applicant device, and generation server related to resumes. [Figure 41] This diagram shows the functional configuration of the generation server involved in modification. [Figure 42] This figure shows another example of interaction between the generation server and the user. [Figure 43] This figure shows another example of interaction between the generation server and the user. [Figure 44] This diagram shows the functional configuration of the generation server, including the IPC classification unit. [Figure 45] This flowchart shows the processing procedure of the IPC classification unit. [Figure 46] This figure shows an example of a resume including an IPC (Information Processing Classification). [Figure 47] This figure shows an example of a resume including an IPC (Information Processing Classification). [Figure 48] This figure shows an example of a resume including an IPC (Information Processing Classification). [Figure 49] This figure shows an example of a resume including an IPC (Information Processing Classification). [Figure 50] This figure shows an example of utilizing RAG technology in the functional configuration of a generation server. [Modes for carrying out the invention]

[0013] The embodiments of this disclosure will be described in detail below with reference to the drawings. While multiple embodiments will be described below, it has been intended from the outset that the configurations described in each embodiment may be combined as appropriate. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and their descriptions will not be repeated.

[0014] Embodiment 1. Embodiment 1 will be explained using Figures 1 to 19. In Embodiment 1, the classification system 1 will be explained using business cases and profile information as examples of classification targets. Here, profile information includes the name, age, work history, qualifications, and skills of a user who is expected to perform work related to the business case.

[0015] As described below, classification system 1 has the function of classifying business cases and profile information using classification information such as IPC. More specifically, classification system 1 identifies (collects) classification information related to each of the business cases and profile information. Classification system 1 stores classification information corresponding to business cases in the database in association with the business cases. Classification system 1 also stores classification information corresponding to profile information in the database in association with the profile information. The information stored in the database is expected to be used, for example, for business matching.

[0016] Figure 1 is a block diagram illustrating the overview of classification system 1 related to Embodiment 1. Classification system 1 can be used, for example, in inter-company crowdsourcing. Crowdsourcing is generally the process of soliciting contributions from a large number of people to obtain the services, ideas, or content that are needed.

[0017] Referring to Figure 1, the general configuration of classification system 1 will be explained. Classification system 1 comprises a server 400, recruiter devices 200A, 200B, 200C..., and applicant devices 300A, 300B, 300C....

[0018] Server 400 provides a matching service to numerous companies, facilitating the matching of business orders and acceptances between companies. Figure 1 shows Company A, Company B, Company C, etc., as examples of companies that use the matching service. Companies A, B, C, etc. are registered as corporate members of Classification System 1. Employees of Companies A, B, C, etc., who use Classification System 1 are also individually registered as members of Classification System 1.

[0019] Hereinafter, in Classification System 1, the work for which contractors are soliciting work may be referred to as "work projects" or "recruitment projects," the person providing the work project may be referred to as the "recruiter," and the person applying to receive the work project may be referred to as the "applicant."

[0020] In classification system 1, for example, an applicant from company A can also receive work from company A. Therefore, in classification system 1, it is permissible for an applicant belonging to a different department Y of company A to receive work from department X of company A.

[0021] An applicant who wins a project is considered a "contractor," and the person who commissions the project to the contractor is considered a "client." However, in the following, "applicants" may be included in the term "contractor," and "clients" may be included in the term "recruiter."

[0022] Server 400 has a database 120 built on it, which is necessary for the matching service. Database 120 contains various databases in which information necessary for providing the matching service is registered. In database 120, information on business cases is registered in a categorized state according to classification information.

[0023] In this disclosure, the act of "registering" information in a database may be referred to as "storing" information in a database.

[0024] Recruiter device 200A is operated by the administrator of company A. Recruiter device 200B is operated by the administrator of company B. Recruiter device 200C is operated by the administrator of company C. Hereinafter, recruiter devices 200A, 200B, 200C, etc. may be collectively referred to as "recruiter device 200".

[0025] Applicant device 300A is operated by an applicant from company A. Applicant device 300B is operated by an applicant from company B. Applicant device 300C is operated by an applicant from company C. Hereinafter, applicant devices 300A, 300B, 300C, etc. may be collectively referred to as "applicant device 300".

[0026] Figure 1 shows two applicants for each company, but the number of applicants is not limited to this. There may be many more applicants for each company, or some companies may have only one applicant. Server 400 may also accept applicants who are not affiliated with a company, such as freelancers.

[0027] In Embodiment 1, the managers of companies A, B, C, etc., will assume the role of recruiters. Therefore, in the following, the managers of each company may be referred to as "recruiters." Recruiters can also act as applicants for jobs advertised by other recruiters. In that case, recruiter device 200 will function as applicant device 300. In Embodiment 1, when a company manager acts as a recruiter, the device that the manager uses to utilize the matching service will be referred to as recruiter device 200.

[0028] Company A may have one administrator or multiple administrators. When assigning administrators to Company A, each administrator may be given a recruiter device 200, or a single recruiter device 200 may be shared by multiple people. The same applies to Companies B, C, etc.

[0029] Server 400 and applicant device 200 are configured to communicate via the Internet 50, which is an example of a communication network. Server 400 and applicant device 300 are configured to communicate via the Internet 50.

[0030] The recruiter device 200 accepts various operations from the recruiter. For example, the recruiter device 200 accepts operations such as inputting recruitment requests (requested tasks) and searching for members of the matching service.

[0031] The recruiting device 200 communicates with the server 400 in response to each operation performed on the recruiting device 200. The server 400 registers the recruitment request (requested work) in the database 120 in response to an operation to input the recruitment request, and provides member information to the recruiting device 200 in response to an operation to search for members.

[0032] The applicant device 300 accepts various operations from the applicant. For example, the applicant device 300 accepts operations such as searching for job postings and applying for job postings.

[0033] The applicant device 300 communicates with the server 400 in response to each operation performed on the applicant device 300. The server 400 provides the applicant device 300 with appropriate job postings in response to an operation to search for job postings, and issues a notification of acceptance or rejection to the applicant device 300 in response to an operation to apply for a job posting.

[0034] Members using Classification System 1 access Server 400 as either recruiters or applicants. Hereafter, members of Classification System 1 may be referred to as "users." Also below, recruiter equipment 200 and applicant equipment 300 may be collectively referred to as "user equipment 500."

[0035] When accepting access from user device 500, server 400 requires sign-in, which involves entering a user ID and password. Server 400 identifies the user by the user ID provided during sign-in.

[0036] Server 400 provides users with an interface for interacting with AI (Artificial Intelligence). Users interact with the AI ​​when classifying business cases using classification information. The AI ​​interacts with the business cases and determines the classification information corresponding to them. Server 400 registers the business cases in database 120, associating them with the classification information. Database 120 is an example of a business case storage unit.

[0037] Server 400 searches the database 120 for job postings that match the user's skills and provides them to the user. Figure 1 shows an example where a list of job postings that match the user's skills is displayed on the screen 550 of the user device 500. As shown in Figure 1, in addition to the list of job postings, the user's skills may also be displayed on the screen 550. In addition to job posting information, the database 120 of Server 400 contains profile information to identify the user's skills.

[0038] As explained above, classification system 1 functions as a matching system that matches users with business cases.

[0039] Figure 2 is a block diagram illustrating the configuration of the server 400 and the user device 500. The user accesses the server 400 using the user device 500. The user device 500 comprises a processor 501, memory 502, a communication interface 503, an input / output interface 504, a display 505, and an operation unit 506. The operation unit 506 consists of a mouse and keyboard, etc.

[0040] Memory 502 includes RAM (Random Access Memory), ROM (Read Only Memory), flash memory, or any other suitable memory system. Memory 502 stores programs necessary for the arithmetic processing of processor 501, as well as temporary data calculated during the arithmetic processing.

[0041] The processor 501 connects to the internet 50 via the communication interface 503, following a program stored in memory 502. The processor 501 connects to the internet 50 and communicates with the server 400.

[0042] Server 400 includes a microcomputer 410 and storage 404. The microcomputer 410 is an example of a "computer". Storage 404 consists of hard disk drives and solid-state drives, etc. Storage 404 stores a Large Language Model (LLM) 430 and a database 120.

[0043] The large-scale language model 430 is an example of a language processing model. The large-scale language model 430 is a language model (pre-trained language model) that has been pre-trained by machine learning. A huge amount of training data is used to train the large-scale language model 430. The large-scale language model 430 is formed as an autoregressive model using a transformer such as GPT (Generative Pre-trained Transformer). The large-scale language model 430 according to Embodiment 1 may be composed of GPT (GPT-2, GPT-3, GPT-4), Bard, or any other model.

[0044] Database 120 includes a company database 121, a user database 122, a case database 123, a matching database 124, and a RAG database 420. The company database 121 contains information on companies that use classification system 1. The user database 122 contains information on users that use the system. The case database 123 contains information on business cases. The matching database 124 contains information on business cases that match users. The RAG database 420 contains information to improve the quality of responses generated by the large-scale language model 430.

[0045] The microcomputer 410 accesses the storage 404 and calls the large-scale language model 430. The microcomputer 410 accesses the storage 404 and performs the processes of registering (storing) data in the database 120, reading the data registered in the database 120, and updating the data registered in the database 120.

[0046] Figure 3 is a block diagram illustrating the configuration related to RAG (Retrieval-Augmented Generation) within Server 400. Here, we will explain in detail the configuration related to RAG within Server 400, as shown in Figure 2.

[0047] The microcomputer 410 comprises a processor 401, memory 402, and a communication interface 403. Memory 402 includes RAM (Random Access Memory), ROM (Read Only Memory), flash memory, or any other suitable memory system. Memory 402 stores programs necessary for the arithmetic operations of the processor 401, as well as temporary data calculated during these operations. The processor 401 operates based on the programs stored in memory 402 and references storage 404.

[0048] RAG is a technique that allows large language models (LGAs) to generate answers based on knowledge sources containing the latest and most accurate information. Generally, LGAs can occasionally output inaccurate or misleading information, which is called hallucination. RAG can complement this imperfection in LGAs and improve the quality of the answers they generate.

[0049] The RAG database 420 is, for example, a vector database. In addition to the data source for classification information, the RAG database 420 defines instructions (behavioral information) and embedding parameters (chunk size, etc.). The RAG database 420 is an example of a RAG data storage unit. The RAG database 420 includes multiple related information databases in which various types of data related to RAG are registered. Figure 1 shows related information databases 421 to 423 as examples of the RAG database 420.

[0050] Related information databases 421-423 are examples of knowledge sources containing the latest and most accurate information. Embodiment 1 proposes a method that uses RAG technology to allow a large-scale language model 430 to appropriately classify various types of information.

[0051] In particular, in Embodiment 1, various information includes user profile information and work case information. User profile information includes the user's skills information related to the work.

[0052] Users can determine whether they possess the necessary skills for a given job by reviewing the details of the job postings. However, it is time-consuming for users to review each of the numerous job postings and check whether they are suitable for them.

[0053] Therefore, Embodiment 1 proposes a classification system 1 that can classify profile information and business case information using classification information, and present the user with the most suitable business case based on the classified profile information and business case information.

[0054] Related information database 421 contains a large amount of business case information. Related information database 422 contains a large amount of classification information for classifying technical information based on classification criteria. Related information database 423 contains a large amount of profile information. Related information database 421 is an example of a business case data storage unit. Related information database 422 is an example of a classification criteria data storage unit. Related information database 423 is an example of a profile data storage unit. The data stored in related information database 422 is an example of a data source related to classification criteria. Related information databases 421 and 423 are examples of classification target data storage units.

[0055] Figure 3 shows an example in which related information databases 421 to 423 are configured on storage 404. However, the storage on which related information database 421 is configured, the storage on which related information database 422 is configured, and the storage on which related information database 423 is configured may be different. The data contained in each of related information databases 421 to 423 may be stored in table format instead of database format. Within a single database, there may be a table area for storing the data contained in related information database 421, a table area for storing the data contained in related information database 422, and a table area for storing the data contained in related information database 423.

[0056] One possible classification method is to use IPC (International Patent Classification). IPC is an internationally standardized technical classification system for classifying patented inventions. By using IPC, it is possible to classify technology in detail by dividing it into hierarchical levels such as "sections," "subsections," "classes," "subclasses," "main groups," and "subgroups."

[0057] IPC is originally an index used to classify patent documents. However, by utilizing IPC as an index for classifying project information and profile information, it is possible to classify project information and profile information (especially skill information) using a uniform standard.

[0058] For example, suppose there is a project such as "Research and development work on electric vehicle propulsion systems, electric vehicle propulsion devices, and electric vehicle battery charging methods." Such a project may be classified into "B06L (Electric Vehicle Propulsion Systems)," "B06L 50 / 00 (Electric Vehicle Propulsion Devices)," and "B06L 50 / 53 (Electric Vehicle Battery Charging Methods)."

[0059] Furthermore, instead of IPC, patent classifications such as F-terms, CPC, or FI may be used as classification information. Alternatively, the Nippon Decimal Classification (NDC) classification table for books and the International Decimal Classification may be used. Alternatively, if classification system 1 is used exclusively within a specific company, the classification information used within the company may be used. In addition, product classifications such as equipment may be applied. In other words, the classification information may be any of IPC, F-terms, CPC, FI, book classification, or company classification.

[0060] The microcomputer 410 accesses the RAG database 420 as needed and provides the large-scale language model 430 with information based on the RAG database 420 during interactions between the user and the large-scale language model 430. As a result, the large-scale language model 430 responds to the user using accurate information based on the RAG database 420.

[0061] As shown in Figure 1, the AI450 is constructed from a microcomputer 410 and storage 404. Users classify and search for business case information and profile information through chat interactions with the AI450.

[0062] In particular, in Embodiment 1, business case information and profile information are classified through dialogue between the AI450 and the user. The method for classifying business case information and profile information through dialogue will be explained below with reference to Figure 4. Furthermore, the method for matching users with business cases using the classification results will be explained below.

[0063] Figure 4 is a diagram illustrating the concepts of the classification and matching methods used by classification system 1. Here, each method is explained according to steps S100, S200, S300, and S400 shown in Figure 4.

[0064] (Step 100) Step S100 shows the process flow for constructing related information databases 421 and 422 using business case information and classification information, and then classifying the business case information instructed by the user using the constructed related information databases 421 and 422.

[0065] The user constructs related information databases 421 and 422 in advance using a large amount of classification information and a large amount of business case information (see Figure 4(i)). The user who constructs related information databases 421 and 422 is, for example, the administrator of classification system 1. The administrator manages a large amount of business case information. The administrator constructs related information database 421 using the business case information under their management.

[0066] Users who wish to classify business case information interact with the AI ​​450 using a chat screen provided by the user device 500. In particular, Embodiment 1 envisions two types of interactions for classifying business case information. One is an interaction for summarizing business case information (see Figure 4(ii)). The other is an interaction for determining classification information corresponding to business case information based on the summarized business case information (see Figure 4(iii)).

[0067] AI450 summarizes the business case information specified by the user through interaction with the user. When summarizing the business case information, AI450 refers to the related information database 421 as needed. AI450 presents the summarized business case information to the user. AI450 resummarizes the business case information according to the user's instructions. AI450 finalizes the content of the summarized business case information according to the user's instructions.

[0068] Next, AI450 classifies the summarized business case information through interaction with the user. When classifying the business case information, AI450 refers to the related information database 422 as needed. AI450 presents the user with the classification information it deems appropriate, along with the technical overview encompassed by the classification information. AI450 re-estimates the classification information according to the user's instructions. AI450 finalizes the classification information corresponding to the summary of the business case information according to the user's instructions.

[0069] When AI450 determines appropriate classification information corresponding to the summarized business case information, it registers the business case information in the case database 123, associating it with the classification information (see Figure 4(iv)). As a result, the case database 123 stores the business case information in a state classified according to the classification information.

[0070] (Step 200) Step S200 describes the process of constructing related information databases 422 and 423 using profile information and classification information, and then using the constructed RAG database 420 to classify the profile information instructed by the user. Step S200 is the same processing procedure as step S100, except that profile information is processed instead of business case information. Therefore, the same explanation as in step S100 will not be repeated here. Note that the profile information includes the user's skill information. According to step S200, the profile information (especially the skill information) is stored in the user database 122 in a state classified by the classification information.

[0071] (Step 300) Step S300 shows the process of searching for job opportunities that match the user's skills using classification information associated with job opportunity information and classification information associated with profile information (see Figure 4(v)). Step S300 performs what is known as job matching.

[0072] (Step 400) Step S400 shows the process of registering the results of the business matching in the matching database 124 (see Figure 4(vi)). Step S400 registers business cases that are considered suitable for each user in the matching database 124.

[0073] The user database 122 may also store unclassified profile information. In this case, the AI ​​450 may retrieve the unclassified profile information according to the user's specification and perform the two above-mentioned dialogues with the retrieved profile information. Alternatively, the AI ​​450 may accept profile information not stored in the user database 122 and perform the two above-mentioned dialogues. The same applies to the case database 123. That is, the case database 123 may store unclassified business case information. In this case, the AI ​​450 may retrieve the business case information according to the user's specification and perform the two above-mentioned dialogues with the retrieved business case information. Alternatively, the AI ​​450 may accept business case information not stored in the case database 123 and perform the two above-mentioned dialogues.

[0074] Here, an example has been described in which the related information databases 421-423 are constructed in the order of steps S100 and S200, but the related information databases 421-423 may also be constructed in the order of steps S200 and S100.

[0075] Figure 5 shows an example of the company database 121. As shown in Figure 5, the company database 121 registers company names, company addresses, etc., for each company ID.

[0076] Figure 6 shows an example of the user database 122. The user database 122 stores user profile information for each user ID. The profile information includes the ID of the company the user belongs to, basic information, career information, and work ability information. Basic information includes name, gender, age, and affiliation. Career information includes work history and educational background. Work ability information includes language skills, qualifications, and skills. The server 400 registers the profile information in the user database 122, associating it with classification information. The user database 122 is an example of a profile storage unit.

[0077] Figure 7 shows an example of the case database 123. The case database 123 registers business case information by case ID. The business case information includes the case title, case details, application requirements, recruitment start date, and recruitment end date. The application requirements include, for example, "Must-have case," "Want-to-applicant case," recruitment level, and age. Server 400 registers the business case information in the case database 123, associating it with classification information. The case database 123 is an example of a business case storage unit.

[0078] Figure 8 shows an example of the matching database 124. The matching database 124 registers matching information for each user ID. The matching information includes one or more case IDs. The case IDs included in the matching information indicate business cases that match the user. The server 400 uses the matching information to identify one or more business cases that match the user identified by the user ID.

[0079] Figure 9 is a block diagram showing the configuration of the classification unit 800, the AI ​​construction unit 902, the data import unit 903, the acquisition unit 904, the matching unit 905, and the result display unit 906. The classification unit 800, the AI ​​construction unit 902, the data import unit 903, the acquisition unit 904, the matching unit 905, and the result display unit 906 are realized by a microcomputer 410 and data stored in storage 404.

[0080] The classification unit 800 is an example of a classification unit that classifies "business case information" acquired by the acquisition unit according to classification criteria. The classification unit 800 is also an example of a classification unit that classifies "profile information" acquired by the acquisition unit according to classification criteria. The classification unit 800 includes a chat unit (summary) 803, a chat unit (classification) 804, and a classification storage unit 805.

[0081] The chat unit 803 includes a chat unit 803A that processes business case information and a chat unit 803B that processes profile information. The chat unit 804 includes a chat unit 804A that processes business case information and a chat unit 804B that processes profile information. The classification storage unit 805 includes a classification storage unit 805A that processes business case information and a classification storage unit 805B that processes profile information. The processing procedures for each unit will be explained below with reference to Figures 10 to 17.

[0082] Figure 10 is a flowchart showing the processing procedure of the acquisition unit 904. First, the acquisition unit 904 acquires business case information to be classified from the user and stores the acquired business case information in the case database 123 according to the user ID (step S1). At this stage, classification information is not associated with the business case information stored in the case database 123.

[0083] Next, the acquisition unit 904 acquires the profile information to be classified from the user and stores the acquired profile information in the user database 122 according to the user ID (step S2). At this stage, the profile information stored in the user database 122 is not associated with classification information.

[0084] Furthermore, the acquisition unit 904 may not store the business case information acquired from the user in the case database 123, but instead subject it to classification processing by the classification unit 800. Similarly, the acquisition unit 904 may not store the profile information acquired from the user in the user database 122, but instead subject it to classification processing by the classification unit 800. Step S1 is an example of an acquisition unit that acquires business case information. Step S2 is an example of an acquisition unit that acquires profile information. These acquisition units may be configured by a data import unit 903.

[0085] Figure 11 is a flowchart showing the processing procedure of the data import unit 903. The data import unit 903 generates the information sources necessary to build the RAG database 420 according to the following procedure.

[0086] First, the data import unit 903 receives a large amount of business case information to be analyzed from users such as administrators (step S11). Next, the data import unit 903 saves the collected business case information to a specific area of ​​the storage 404 as the information source for the business case information (step S12).

[0087] Next, the data import unit 903 receives a large amount of profile information to be analyzed from a user such as an administrator (step S13). Then, the data import unit 903 saves the collected profile information as the information source for the profile information to a specific area of ​​the storage 404 (step S14).

[0088] Next, the data import unit 903 receives a large amount of classification information to be analyzed from a user such as an administrator (step S15). Then, the data import unit 903 saves the collected classification information to a specific area of ​​the storage 404 as the information source for the classification information (step S16).

[0089] Figure 12 is a flowchart showing the processing procedure of the AI ​​construction unit 902. First, the AI ​​construction unit 902 acquires information sources (business case information) from the storage 404 (step S21). Next, the AI ​​construction unit 902 uses the acquired information sources (business case information) to construct a related information database 421 related to RAG (step S22). This generates a related information database 421 in which data sources related to business case information are stored in a format that conforms to RAG.

[0090] Next, the AI ​​construction unit 902 retrieves information sources (profile information) from the storage 404 (step S23). Then, the AI ​​construction unit 902 constructs a related information database 423 related to RAG (step S24). This generates a related information database 423 in which data sources related to profile information are stored in a format that conforms to RAG.

[0091] Next, the AI ​​construction unit 902 acquires information sources (classification information) from the storage 404 (step S25). Next, the AI ​​construction unit 902 constructs a related information database 422 related to RAG (step S26). This generates a related information database 422 in which data sources related to classification information are stored in a format that conforms to RAG.

[0092] Figure 13 is a flowchart showing the processing procedure of the chat unit (summary) 803. First, the chat unit 803 identifies the profile information or case information to be classified (step S31). Specifically, the chat unit 803 identifies the profile information to be classified by referring to the user database 122. Alternatively, the chat unit 803 identifies the business case information to be classified by referring to the case database 123.

[0093] Furthermore, the chat unit (summary) 803 may obtain profile information or business case information to be classified from the user through chat with the user.

[0094] Next, the chat unit 803 determines whether the item to be classified is profile information or business case information (step S32).

[0095] If the classification target is profile information, the chat unit 803 refers to the related information database 423 related to the profile information as needed (step S33). Next, the chat unit 803 calls the AI ​​algorithm (step S34). Next, the chat unit 803 summarizes the profile information using RAG through chat with the user (step S35). Next, the chat unit 803 passes the summary to the chat unit (classification) 804 (step S39).

[0096] Step S35 includes the steps of having the user confirm the summary of profile information created by the chat unit 803, revising the summary of profile information based on the user's instructions, and finalizing the summary of profile information after obtaining the user's approval. In other words, the chat unit 803 determines the content of the summary after interacting with the user regarding the summary of profile information.

[0097] If the classification target is business case information, the chat unit 803 refers to the related information database 421 related to the business case information as needed (step S36). Next, the chat unit 803 calls the AI ​​algorithm (step S37). Next, the chat unit 803 summarizes the business case information using RAG through chat with the user (step S38). Next, the chat unit 803 passes the summary to the chat unit (classification) 804 (step S39).

[0098] Step S38 includes the steps of having the user confirm the summary of business case information created by the chat unit 803, revising the summary of business case information based on the user's instructions, and finalizing the summary of business case information after obtaining the user's approval. In other words, the chat unit 803 determines the content of the summary after interacting with the user regarding the summary of business case information. Note that the chat unit 803 may determine the content of the summary without obtaining explicit approval from the user.

[0099] Figure 14 is a flowchart showing the processing procedure of the chat unit (classification) 804. First, the chat unit 804 determines whether the item to be classified is profile information or business case information (step S41).

[0100] If the classification target is profile information, the chat unit 804 refers to the related information database 422 related to the classification information as needed (step S42). Next, the chat unit 804 calls the AI ​​algorithm (step S43). Then, while chatting with the user, the chat unit 803 determines the classification information corresponding to the profile information using a summary of the profile information (step S44).

[0101] In this way, the chat unit (classification) 804 uses the large-scale language model 430 and the related information database 422 to determine classification information corresponding to the profile information acquired by the acquisition unit. Specifically, the chat unit (classification) 804 determines classification information corresponding to the profile information based on a summary of the profile information. Next, the chat unit 804 displays the classification information corresponding to the information to be classified (profile information) (step S48). More specifically, the chat unit 804 sends data to the user device 500 for displaying an image that associates the profile information and the classification information. Based on the received data, the user device 500 displays the profile information associated with the classification information on the display 505. In this way, the chat unit 804 displays the profile information associated with the classification information on the display device. Next, the chat unit 804 passes the determined content to the classification storage unit 805 (step S49). The determined content includes the profile information and the classification information corresponding to the profile information.

[0102] Step S44 includes the steps of having the user confirm the classification information provisionally determined by the chat unit 804, changing the classification information based on the user's instructions, and finalizing the classification information after obtaining the user's approval. In other words, after interacting with the user regarding the classification information, the chat unit 804 determines the classification information corresponding to the profile information.

[0103] If the classification target is business case information, the chat unit 804 refers to the related information database 422 related to the classification information as needed (step S45). Next, the chat unit 804 calls the AI ​​algorithm (step S46). Then, while chatting with the user, the chat unit 803 determines the classification information corresponding to the business case information using a summary of the business case information (step S47).

[0104] In this way, the chat unit (classification) 804 uses the large-scale language model 430 and the related information database 422 to determine classification information corresponding to the business case information acquired by the acquisition unit. Specifically, the chat unit (classification) 804 determines classification information corresponding to the business case information based on a summary of the business case information. Next, the chat unit 804 displays the classification information corresponding to the information to be classified (business case information) (step S48). More specifically, the chat unit 804 sends data to the user device 500 for displaying an image that associates the business case information with the classification information. Based on the received data, the user device 500 displays the business case information on the display 505 in association with the classification information. In this way, the chat unit 804 displays the business case information on the display device in association with the classification information. Next, the chat unit 804 passes the determined content to the classification storage unit 805 (step S49). The determined content includes the business case information and the classification information corresponding to the business case information.

[0105] Step S47 includes the steps of having the user confirm the classification information provisionally determined by the chat unit 804, changing the classification information based on the user's instructions, and finalizing the classification information after obtaining the user's approval. In other words, the chat unit 804 determines the classification information corresponding to the profile information after interacting with the user regarding the classification information. The classification unit 800 may determine the classification information without interacting with the user using the chat unit 804. For example, the classification unit 800 may determine the classification information corresponding to each of the many pre-registered profile information items all at once without interacting with the user.

[0106] Figure 15 is a flowchart showing the processing procedure of the classification storage unit 805. First, the classification storage unit 805 determines whether the item to be stored is profile information or business case information (step S51). If the item to be stored is profile information, the classification storage unit 805 stores the profile information in the user database 122 in association with the classification information (step S52). If the item to be stored is business case information, the classification storage unit 805 stores the business case information in the case database 123 in association with the classification information (step S53).

[0107] Figure 16 is a flowchart showing the processing procedure of the matching unit 905. First, the matching unit 905 receives a matching instruction from the user (step S61). Here, the user is, for example, a user who is considering applying for a job offer. The user instructs the matching unit 905 to perform a match using the user device 500.

[0108] Next, the matching unit 905 retrieves user profile information from the user database 122 (step S62). The profile information is associated with classification information. The matching unit 905 retrieves business case information from the case database 123 (step S63). The business case information is associated with classification information.

[0109] Next, the matching unit 905 searches for business case information using the classification information associated with the profile information (step S64). More specifically, when the classification information associated with the profile information is called classification information X, the matching unit 905 finds the business case information associated with classification information X. The matching unit 905 estimates the found business case information as the matching result and stores the matching result in the matching database 124 by user ID (step S65).

[0110] The matching unit 905 may execute step S63 before step S62. The matching unit 905 may accept matching instructions on a web screen. The matching unit 905 may perform the process of deriving the matching results in batch processing.

[0111] Figure 17 is a flowchart showing the processing procedure of the results display unit 906. First, the results display unit 906 receives an instruction from the user to display the matching results (step S71). Next, the results display unit 906 obtains the matching results corresponding to the user who gave the instruction from the matching database 124 (step S72). The results display unit 906 uses the user ID to identify the matching results corresponding to the user who gave the instruction.

[0112] Next, the result display unit 906 transmits the matching result to the user device 500 (step S73). Step S73 is an example of a transmission unit. The user device 500 displays the received matching result. As a result, the screen 550 shown in Figure 1 is displayed on the user device 500. Note that the result display unit 906 may execute the processes in steps S72 and S73 even if it has not received a display instruction from the user (step S71).

[0113] [Example 1] Next, we will explain Modification 1. Figure 18 is a diagram illustrating the concepts of the classification method and matching method related to Modification 1. In Modification 1, in addition to classifying business case information and profile information using classification information, we propose a method for classifying advertising information using classification information.

[0114] As shown in Figure 18, in Modification 1, the classification system 1 classifies profile information in step S200 and advertisement information in step S500. The procedure for step S200 has been explained using Figure 4, so that explanation will not be repeated here. In addition, in Modification 1, the classification system 1 also performs steps S100, S300, and S400 shown in Figure 4, but that explanation will not be repeated here.

[0115] In the first modified example, an advertising database 125 is added to the database 120 shown in Figure 2, and a related information database 424 is added to the RAG database 420 shown in Figure 3.

[0116] (Step 500) Step S500 shows the process flow for constructing related information databases 422 and 424 using advertising information and classification information, and then classifying the advertising information instructed by the user using the constructed related information databases 422 and 424.

[0117] The user pre-constructs related information databases 422 and 424 using a large amount of classification information and a large amount of advertising information (see Figure 18(i)). Here, the user is, for example, the administrator of classification system 1. The advertising information includes, for example, various types of advertisements targeted at applicants. The advertising information may also be created targeting recruiters. Once related information databases 422 and 424 are constructed, the system is ready to classify the advertising information.

[0118] Users who wish to classify advertising information interact with AI450 using a chat screen provided by user device 500. In Modification 1, two types of interactions for classifying advertising information are envisioned. One is an interaction for summarizing the advertising information (see Figure 18(ii)). The other is an interaction for determining the classification information corresponding to the advertising information based on the summarized advertising information (see Figure 18(iii)).

[0119] AI450 summarizes the advertising information specified by the user through interaction with the user. When summarizing the advertising information, AI450 refers to the relevant information database 424 as needed. AI450 presents the summarized advertising information to the user. AI450 resummarizes the advertising information according to the user's instructions. AI450 finalizes the content of the summarized advertising information according to the user's instructions.

[0120] Next, AI450 classifies the summarized advertising information through interaction with the user. When classifying the advertising information, AI450 refers to the relevant information database 422 as needed. AI450 presents the user with the classification information it deems appropriate, along with the technical overview encompassed by the classification information. AI450 re-estimates the classification information according to the user's instructions. AI450 finalizes the classification information corresponding to the summary of the advertising information according to the user's instructions.

[0121] When AI450 determines appropriate classification information corresponding to the summarized advertising information, it registers the advertising information in the advertising database 125, associating it with the classification information (see Figure 18(iv)). As a result, the advertising database 125 stores the advertising information in a state classified by the classification information. The advertising database 125 is an example of an advertising storage unit.

[0122] (Step 300A) Step S300A shows the process of searching for advertising information that matches the user's profile information using classification information associated with advertising information and classification information associated with profile information (see Figure 18(v)).

[0123] (Step 400A) Step S400A is a process that sends advertising information that matches the user's profile information to the user's user device 500 (see Figure 18 (vi)). As a result of step S400A, advertisements that match the user's profile information are displayed on the user's user device 500. In step S400A, advertising information that matches the user's profile information may be registered in the matching database 124 for each user. Advertising information is an example of information to be classified. The chat unit 804 may display the advertising information on the user device 500 in association with the classification information, similar to step S48 in Figure 14.

[0124] [Differentiation 2] Next, we will explain Modification 2. Figure 19 is a diagram illustrating the concepts of the classification method and matching method related to Modification 2. In Modification 2, in addition to classifying business case information and profile information using classification information, we propose a method for classifying educational information using classification information. Educational information includes, for example, educational videos that improve or broaden the user's skill set.

[0125] As shown in Figure 19, in Modification 2, the classification system 1 classifies profile information in step S200 and educational information in step S600. The procedure for step S200 has been explained using Figure 4, so that explanation will not be repeated here. In addition, in Modification 2, the classification system 1 also performs steps S100, S300, and S400 shown in Figure 4, but that explanation will not be repeated here.

[0126] In the second modification, the education database 126 is added to the database 120 shown in Figure 2, and the related information database 425 is added to the RAG database 420 shown in Figure 3.

[0127] (Step 600) Step S600 shows the process flow for constructing related information databases 422 and 425 using educational information and classification information, and then classifying the educational information instructed by the user using the constructed related information databases 422 and 425.

[0128] The user pre-constructs related information databases 422 and 425 using a large amount of classification information and a large amount of educational information (see Figure 19(i)). Here, the user is, for example, the administrator of classification system 1. The educational information includes, for example, various types of educational videos targeted at applicants. Once related information databases 422 and 425 are constructed, the system is ready to classify the educational information.

[0129] A user who wishes to classify educational information interacts with AI450 using a chat screen provided by user device 500. In Modification 2, two types of interactions for classifying educational information are envisioned. One is an interaction for summarizing the educational information (see Figure 19(ii)). The other is an interaction for determining the classification information corresponding to the educational information based on the summarized educational information (see Figure 19(iii)).

[0130] AI450 summarizes educational information specified by the user through interaction with the user. When summarizing educational information, AI450 refers to the relevant information database 425 as needed. AI450 presents the summarized educational information to the user. AI450 resummarizes the educational information according to the user's instructions. AI450 finalizes the content of the summarized educational information according to the user's instructions.

[0131] Next, AI450 classifies the summarized educational information through interaction with the user. When classifying the educational information, AI450 refers to the relevant information database 422 as needed. AI450 presents the user with the classification information it deems appropriate, along with the technical overview encompassed by the classification information. AI450 re-estimates the classification information according to the user's instructions. AI450 finalizes the classification information corresponding to the summary of the educational information according to the user's instructions.

[0132] When AI450 determines appropriate classification information corresponding to the summarized educational information, it registers the educational information in the educational database 126, associating it with the classification information (see Figure 19(iv)). As a result, the educational database 126 stores the educational information in a state classified by the classification information. The educational database 126 is an example of an educational storage unit.

[0133] (Step 300B) Step S300B shows a process that searches for educational information that matches the user's profile information using classification information associated with educational information and classification information associated with profile information (see Figure 19(v)).

[0134] (Step 400B) Step S400B is a process that sends educational information matching the user's profile information to the user's user device 500 (see Figure 19 (vi)). Step S400B displays the educational information matching the user's profile information on the user's user device 500. In step S400B, the educational information matching the user's profile information may be registered in the matching database 124 for each user. Educational information is an example of information to be classified. The chat unit 804 may display the educational information on the user device 500 in association with the classification information, similar to step S48 in Figure 14.

[0135] This example illustrates how educational information matching a user's profile information is provided. However, to broaden the user's skill range, the classification system 1 may search for educational information stored in the educational database 126 that is associated with classification information different from that associated with the user's profile information. For example, if a user does not possess the skills required for a job project that interests them, the classification system 1 may provide the user with educational information to acquire those skills.

[0136] More specifically, if the classification information corresponding to a business project of interest to the user differs from the classification information associated with the user's profile information, classification system 1 may provide the user with educational videos classified according to the classification information corresponding to the business project. In addition, classification system 1 may perform the processing of modified example 2 in addition to the processing of modified example 1.

[0137] As described above, according to Embodiment 1, business case information, profile information, advertising information, and educational information can be classified uniformly and appropriately. Hereinafter, this information to be classified will be referred to as "classification target information." Furthermore, according to Embodiment 1, confidential information can also be classified by using RAG. In addition, by storing the data related to the classification process in the RAG database 420 in advance, the error rate of the large-scale language model 430 can be reduced. In this disclosure, "storage" includes the concept of continuously receiving and processing data. The transmission unit that sends data to a storage unit such as memory may send all the data to the storage unit at once. Alternatively, the transmission unit may divide the data to be sent into packets or chunks and then sequentially send the divided data to the storage unit.

[0138] [Other variations] As explained above, classification system 1 determines the summary content of the information to be classified and the classification information corresponding to the information to be classified through interaction with the user. The interaction between classification system 1 and the user is realized by the chat function of classification system 1. Such a chat function is not required.

[0139] Classification system 1 may determine the classification information corresponding to the information to be classified without going through the process of summarizing the information to be classified. When determining the classification information corresponding to the information to be classified without going through the process of summarizing the information to be classified, classification system 1 may determine the classification information through interaction with the user, or it may determine the classification information without interaction with the user. In addition to RAG, LoRA (Low-Rank Adaptation of Large Language Models) and the like may also be used.

[0140] As shown in Figure 2, database 120 includes multiple databases. One or more of these databases may be stored in storage separate from storage 404. Each of these databases functions as a storage unit.

[0141] In Embodiment 1, the classification unit 800 determines classification information corresponding to the information to be classified using the large-scale language model 430, related information databases 421, 423, 424, and 425 related to the information to be classified, and related information database 422 in which data sources related to classification criteria are stored in a format conforming to RAG. However, the classification unit 800 may determine classification information corresponding to the information to be classified using only the large-scale language model 430 and related information database 422, without using the related information databases 421, 423, 424, and 425. In this case, the classification unit 800 may determine classification information corresponding to the information to be classified using, for example, the information to be classified obtained through interaction with the user, the large-scale language model 430, and related information database 422. The information to be classified may be, for example, business case information, profile information, advertising information, and educational information.

[0142] The user device 500 is not limited to having a processor 501, memory 502, communication interface 503, and input / output interface 504, but may also be a thin client system utilizing VDI (Virtual Desktop Infrastructure). A thin client system utilizing VDI is a system that transfers and uses a desktop environment located on a server to a terminal in a remote location. The user device 500 and the server 400 do not necessarily have to be independent devices. When using such a thin client system, the functions of the user device 500 and the server 400 can be provided on the same aggregation server. Note that the classification system 1 may be configured without including the user device 500.

[0143] Server 400 is an example of a compute device. A compute device may be configured using servers (on-premise servers, cloud servers, etc.) or serverless systems. Here, an on-premise server is a server installed and managed within facilities managed by the company itself. A cloud server is a server provided by another company via a network (a leased server). A serverless system is a system that allows users to utilize compute and memory functions only when needed, without being aware of the existence of a server. A compute device includes servers and serverless systems. Servers include on-premise servers and cloud servers.

[0144] Embodiment 2. Next, Embodiment 2 will be described with reference to Figures 20 to 50. In Embodiment 2, the classification system 1000 will be explained using "job competency information" included in a resume as an example of the items to be classified. Here, job competency information includes qualifications and skills obtained from past work experience. Career history, qualifications, and skills are examples of job competency information. Job competency information may also be included in the profile information described in Embodiment 1. Job competency information is also an example of "job-related information".

[0145] As described below as a "modified version," the classification system 1000 has the function of classifying the user's work competency information using classification information such as IPC. More specifically, the classification system 1000 identifies (collects) classification information related to the user's work competency information and stores the identified classification information as part of the user's work history information. The work history information is intended to be used, for example, for job matching. Similar to classification system 1, the classification system 1000 functions as a matching system that matches users with job opportunities.

[0146] [Overall structure] Figure 20 is a block diagram illustrating the overview of the classification system 1000 related to Embodiment 2. The classification system 1000 can be used, for example, in inter-company crowdsourcing. Crowdsourcing is generally the process of soliciting contributions from a large number of people to obtain the services, ideas, or content that are needed.

[0147] Many companies are promoting side jobs to make effective use of their human resources. By using crowdsourcing between companies, they can leverage the skills of their employees.

[0148] Referring to Figure 20, the general configuration of the classification system 1000 will be described. The classification system 1000 comprises a sharing server 100, recruiter devices 200A, 200B, 200C..., applicant devices 300A, 300B, 300C..., and a generation server 4000.

[0149] The sharing server 100 provides a matching service to numerous companies, facilitating the matching of business orders and acceptances between companies. Figure 20 shows companies A, B, C, etc., as examples of companies that use the matching service. Companies A, B, C, etc. are registered as corporate members of the classification system 1000. Employees of companies A, B, C, etc., who use the classification system 1000 are also individually registered as members of the classification system 1000.

[0150] The tasks offered through the classification system 1000 are, for example, temporary tasks that are expected to be completed within a predetermined period. Therefore, those who accept tasks offered through the classification system 1000 will engage in their primary work within their specific department within the company, and the tasks offered through the classification system 1000 as a secondary job. In addition, within the classification system 1000, for example, an applicant from company A can also accept tasks from company A. Therefore, within the classification system 1000, it is permissible for an applicant belonging to a different department Y of company A to accept tasks from department X of company A.

[0151] In the following, in classification system 1000, the work for which contractors are being recruited may be referred to as "recruitment work" or "recruitment project," the person providing the recruitment project may be referred to as "recruiter," and the person applying to receive the contract for the recruitment project may be referred to as "applicant." Applying to recruitment work may be referred to as "applying to recruitment work" or "applying to recruitment project."

[0152] An applicant who wins a project is considered a "contractor," and the person who commissions the project to the contractor is considered a "client." However, in the following, "applicants" may be included in the term "contractor," and "clients" may be included in the term "recruiter."

[0153] The sharing server 100 has a database 120 built on it that is necessary for the matching service. Database 120 includes various databases in which information necessary for providing the matching service is registered. For example, database 120 contains information on members and recruitment activities. The sharing server 100 is managed and operated by a company separate from the companies that use the matching service. Any company that uses the matching service may manage and operate the sharing server 100.

[0154] Recruiter device 200A is operated by the administrator of company A. Recruiter device 200B is operated by the administrator of company B. Recruiter device 200C is operated by the administrator of company C. Hereinafter, recruiter devices 200A, 200B, 200C, etc. may be collectively referred to as "recruiter device 200".

[0155] Applicant device 300A is operated by an applicant from company A. Applicant device 300B is operated by an applicant from company B. Applicant device 300C is operated by an applicant from company C. Hereinafter, applicant devices 300A, 300B, 300C, etc. may be collectively referred to as "applicant device 300". Figure 20 depicts two applicants for each company, but the number of applicants is not limited to this. There may be many more applicants for each company, or a company may have only one applicant. The sharing server 100 may accept applicants who are not affiliated with a company, such as freelancers.

[0156] In Embodiment 2, the managers of companies A, B, C, etc., will assume the role of recruiters. Therefore, in the following, the managers of each company may be referred to as "recruiters." Recruiters can also act as applicants for jobs advertised by other recruiters. In that case, recruiter device 200 will function as applicant device 300. In Embodiment 2, when a company manager acts as a recruiter, the device that the manager uses to utilize the matching service will be referred to as recruiter device 200.

[0157] Company A may have one administrator or multiple administrators. When assigning administrators to Company A, each administrator may be given a recruiter device 200, or a single recruiter device 200 may be shared by multiple people. The same applies to Companies B, C, etc.

[0158] The sharing server 100 and the recruiter device 200 are configured to communicate via the Internet 50, which is an example of a communication network. The sharing server 100 and the recruiter device 300 are configured to communicate via the Internet 50.

[0159] The sharing server 100 requires sign-in, including the input of a member ID and password, when accepting access from the recruiter device 200. Similarly, the sharing server 100 requires sign-in, including the input of a member ID and password, when accepting access from the applicant device 300. The sharing server 100 identifies each recruiter and applicant by the member ID provided during sign-in.

[0160] The recruiter device 200 accepts various operations from the recruiter. For example, the recruiter device 200 accepts operations such as inputting job postings (requested tasks), inputting evaluations of contractors who have completed the tasks, and searching for members of the matching service.

[0161] The recruiter device 200 communicates with the sharing server 100 in response to each operation performed on the recruiter device 200. The sharing server 100 registers the recruitment request (requested work) in the database 120 in response to an operation to input the recruitment request (requested work), registers the evaluation of the target applicant (contractor) in the database 120 in response to an operation to input an evaluation, and provides member information to the recruiter device 200 in response to an operation to search for members.

[0162] The applicant device 300 accepts various operations from the applicant. For example, the applicant device 300 accepts operations such as searching for job postings, applying for job postings, entering work performance, and entering evaluations of the recruiter (client).

[0163] The applicant device 300 communicates with the sharing server 100 in response to each operation performed on the applicant device 300. The sharing server 100 provides the applicant device 300 with appropriate job postings in response to an operation to search for job postings, issues a notification of acceptance or rejection to the applicant device 300 in response to an operation to apply for a job posting, registers the work performance in the database 120 in response to an operation to input work performance, and registers the evaluation of the target recruiter (client) in the database 120 in response to an operation to input an evaluation.

[0164] A recruiter belonging to a certain department of Company A can use classification system 1000 to hire applicants belonging to other departments of Company A as contractors for recruitment projects. A recruiter belonging to Company A can use classification system 1000 to hire applicants belonging to Company B as contractors for recruitment projects.

[0165] Members using the classification system 1000 access the sharing server 100 as either recruiters or applicants. Hereafter, members of the classification system 1000 may be referred to as "users." Also below, the recruiter devices 200 and applicant devices 300 operated by members may be collectively referred to as "user devices 500."

[0166] The sharing server 100 is connected to the generation server 4000 in a communicative manner. The generation server 4000 provides users with an interface to assist in creating resumes. The generation server 4000 has a database 4420 built on it, which is necessary to provide such an interface to users. The generation server 4000 may be managed by the company that manages the sharing server 100, or it may be managed by a company other than the company that manages the sharing server 100. The sharing server 100 may include the functions of the generation server 4000.

[0167] The generation server 4000 is connected to the user device 500 via the Internet 50. The user device 500 includes the recruiter device 200 and the applicant device 300. Users access the generation server 4000 using the user device 500.

[0168] The generation server 4000, like the sharing server 100, requires sign-in, which involves entering a member ID and password, when accepting access from the user device 500. The generation server 4000 identifies the user by the member ID provided during sign-in.

[0169] The generation server 4000 provides the user with an interface to assist in creating a resume. As a result, the resume creation tool is displayed on screen 1550 of the user device 500. The generation server 4000 is an example of a work history information collection device.

[0170] The generation server 4000 has the function of collecting all the necessary work history information from the user while interacting with the user by displaying a question box 1551 and an answer box 1552 on the screen 1550. Based on the collected work history information, the generation server 4000 creates a work history in a standardized format. The generation server 4000 displays the created work history on the screen 1550, giving the user the opportunity to check the work history. The generation server 4000 saves the work history that the user has checked. When a user applies for a job posting, the generation server 4000 cooperates with the sharing server 100 and sends the work history to the recruiter device 200.

[0171] The reason for introducing the generation server 4000 into classification system 1000 lies in the "complexity of creating resumes" and the "difficulty of standardizing resumes." Each of these points will be explained in detail below.

[0172] [The complexity of creating a resume / CV] To effectively utilize internal talent, it is effective to clearly define individual skills and work experience. To do this, it is necessary to collect as much personal information as possible without omission and compile it into a standardized document format.

[0173] However, for job seekers, recalling their skills and work experience and then compiling them into a standardized document format is an extremely cumbersome process.

[0174] Of course, those who wish to formally change jobs may be able to maintain a certain level of motivation for the aforementioned cumbersome procedures, as they will be preparing for rigorous document screening and interviews. In contrast, employees applying for internal projects or those seeking side jobs outside the company will likely find it difficult to maintain motivation for these cumbersome procedures. This is because, in addition to not involving a formal job change, they are not required to undergo the rigorous document screening and multifaceted interviews.

[0175] Thus, in so-called "insider" cases, applicants tend to have little motivation to engage in the cumbersome procedures described above. In such cases, there is a challenge in that job seekers are unable to comprehensively collect individual skills and work experience, and to compile the collected information into a standardized document format.

[0176] Creating a high-quality resume requires more information about an individual's skills and experience. However, in internal cases, even if a lot of information is available, many job seekers lack the motivation to write a resume and find it tedious (or cumbersome) to write down their skills and experience. In such cases, the amount of information included in the resume tends to be small. Therefore, even if a lot of information is available to the job seeker, there is a challenge in that they cannot compile that information into a standardized document format.

[0177] [The difficulty of standardizing resumes] If resumes were standardized, evaluators could more easily assess each individual's work history. Furthermore, standardized resumes would allow companies to effectively manage talent by grouping employee resumes. However, since resumes are generally written in a free format, the style varies depending on the writer. Even if writers are required to use a uniform format, differences in individual document creation abilities make it difficult to standardize resumes. There is also the risk of duplicate information creeping into resumes due to writer's carelessness. Therefore, evaluators need time to assess each individual's work history based on their resumes. Additionally, it is difficult for companies to group individual work histories based on resumes.

[0178] Based on the background described above, the generation server 4000 is installed in the classification system 1000. According to the generation server 4000, work history information is collected from the user in an interactive manner without any omissions, and a standardized work history document is created. The created work history document contains all necessary work history information without any omissions or duplication.

[0179] According to Embodiment 2, it is possible to assist users in creating a resume that includes all necessary information without any omissions or duplication. In addition, according to Embodiment 2, it is possible to assist users in creating a resume that appropriately reflects their work experience, regardless of their document creation ability. According to Embodiment 2, it is possible to collect work experience information in a way that includes all necessary information without any omissions or duplication.

[0180] Figure 21 is a block diagram showing the configuration of the sharing server 100, the recruiter device 200, and the applicant device 300.

[0181] [Configuration of Sharing Server 100] The sharing server 100 includes a processor 101, memory 102, storage 103, and a communication interface 104.

[0182] Memory 102 includes RAM (Random Access Memory), ROM (Read Only Memory), flash memory, or any other suitable memory system. Memory 102 stores programs necessary for the arithmetic processing of processor 101, as well as temporary data calculated during the arithmetic processing.

[0183] Storage 103 consists of hard disk drives and solid-state drives, etc. Storage 103 stores database 120. Database 120 contains multiple types of databases. These multiple types of databases include a corporate database (corporate DB) 121, a user database (user DB) 122, a community database (community DB) 127, and a case database (case DB) 123.

[0184] Some of these multiple types of databases may be stored in storage separate from the sharing server 100. For example, some of the multiple types of databases shown in Figure 21 may be stored on a cloud service separate from the sharing server 100. In this case, the sharing server 100 can access the necessary databases by communicating with the cloud via the internet 50.

[0185] The processor 101 connects to the Internet 50 via the communication interface 104, following a program stored in the memory 102. The processor 101 communicates with the recruiter device 200 and the applicant device 300 after connecting to the Internet 50. The processor 101 accesses the database 120 and performs processes such as extracting necessary data, registering new data in the database 120, and updating data registered in the database 120.

[0186] [Configuration of recruiter device 200] The recruiter device 200 comprises a processor 201, memory 202, communication interface 203, input / output interface 204, display 205, and operation unit 206. The operation unit 206 consists of a mouse and keyboard, etc.

[0187] Memory 202 includes RAM (Random Access Memory), ROM (Read Only Memory), flash memory, or any other suitable memory system. Memory 202 stores programs necessary for the arithmetic operations of processor 201, as well as temporary data calculated during those operations.

[0188] The processor 201 connects to the internet 50 via the communication interface 203 according to the program stored in memory 202. The processor 201 connects to the internet 50 and communicates with the sharing server 100. The processor 201 communicates with the sharing server 100 and performs processes such as sending job postings, displaying information of applicant members on the display 205, ordering work from selected contractors from among the applicants, and sending the evaluations of contractors entered by the recruiter to the sharing server 100.

[0189] Information entered through the operation of the control unit 206 is notified to the processor 201 via the input / output interface 204.

[0190] [Configuration of applicant device 300] The applicant device 300 comprises a processor 301, memory 302, communication interface 303, input / output interface 304, display 305, and operation unit 306. The operation unit 306 consists of a mouse and keyboard, etc.

[0191] Memory 302 includes RAM (Random Access Memory), ROM (Read Only Memory), flash memory, or any other suitable memory system. Memory 302 stores programs necessary for the arithmetic processing of the processor 301, as well as temporary data calculated during the arithmetic processing.

[0192] The processor 301 connects to the Internet 50 via the communication interface 303 according to the program stored in the memory 302. The processor 301 connects to the Internet 50 and communicates with the sharing server 100. The processor 301 communicates with the sharing server 100 and performs processes such as applying for job postings, displaying acceptance or rejection notifications for applied jobs on the display 305, sending the results of the received work to the sharing server 100, and sending the evaluation of the recruiter entered by the applicant to the sharing server 100.

[0193] Information entered through the operation of the control unit 306 is notified to the processor 301 via the input / output interface 304.

[0194] [Database 120] The following describes database 120. The company database 121 contains information on companies that are members of the classification system 1000. The user database 122 contains information on members who use the classification system 1000. Many of the members are employees of companies that are members of the classification system 1000.

[0195] Members registered in the user database 122 can use the classification system 1000 to act as recruiters (clients) or applicants (contractors). Members may include employees belonging to companies registered in the company database 121, as well as individuals (freelancers) who are not affiliated with any company.

[0196] The community database 127 stores information for identifying companies belonging to a community. Communities are formed by agreement between companies. Therefore, multiple communities can be formed depending on how the companies reach an agreement. The number of companies belonging to a single community can also be set in various ways. Companies that have community relationships form a relationship of trust to the extent determined by how the agreement was reached when the community was formed. The community database 127 registers information for identifying companies belonging to each community.

[0197] The project database 123 contains registered projects for which contractors are being sought. Employees of each company, while engaged in their primary duties within their respective departments, can, as members of the classification system 1000, accept projects from other departments within their own company or projects from other companies that are registered in the project database 123. In this case, members accept projects from other departments within their own company or projects from other companies as a side job.

[0198] [Configuration of Generation Server 4000] Figure 22 is a block diagram showing the configuration of the generation server 4000. The generation server 4000 comprises a processor 4001, memory 4002, storage 4003, and a communication interface 4004.

[0199] Memory 4002 includes RAM (Random Access Memory), ROM (Read Only Memory), flash memory, or any other suitable memory system. Memory 4002 stores programs necessary for the arithmetic processing of processor 4001, as well as temporary data calculated during the arithmetic processing.

[0200] Storage 4003 consists of hard disk drives and solid-state drives, etc. Storage 4003 stores database 4420 and large-scale language model 430. The large-scale language model 430 is an example of a natural language processing algorithm. Storage 4003 is an example of a memory unit that stores natural language processing algorithms.

[0201] The large-scale language model 430 is a language model (pre-trained language model) that has been pre-trained using machine learning. A huge amount of text data is used to train the large-scale language model 430. The large-scale language model 430 is formed as an autoregressive model using a transformer such as GPT (Generative Pre-trained Transformer). The large-scale language model 430 according to Embodiment 2 may include Bard and others in addition to GPT (GPT-2, GPT-3, GPT-4).

[0202] Database 4420 contains multiple types of databases. These multiple types of databases include the work history database (Work History DB) 4421, the conversational information database (Conversational Information DB) 4422, and the dictionary database (Dictionary DB) 4423.

[0203] The memory 4002 stores the program (algorithm) 4050. The processor 4001 executes the program (algorithm) 4050, utilizing the large-scale language model 430 to create a resume.

[0204] Program 4050 includes an interaction program 4051, a verification program 4052, and a creation program 4053. Processor 4001 interacts with the user by executing the interaction program 4051, utilizing the large-scale language model 430. Through this, processor 4001 obtains a large amount of work history information from the user. Processor 4001 verifies, by executing the verification program 4052, that all the work history information necessary for creating a resume has been obtained, utilizing the large-scale language model 430. Processor 4001 creates a resume by executing the creation program 4053, utilizing the large-scale language model 430 to conform to a predetermined format.

[0205] [Database 4420] The following describes database 4420. The work history database 4421 registers "work history information" and "work history documents" for each user's member ID. In this disclosure, "work history information" refers to the information used to create the "work history document." The generation server 4000 can, for example, receive work history information in file format from the user via the user device 500 before interacting with the user. When the generation server 4000 receives work history information from the user, it registers the work history information in the work history database 4421 for each user's member ID.

[0206] If the work experience information registered in the work experience database 4421 includes all the information necessary to create a work experience document, the generation server 4000 will create the work experience document without interacting with the user.

[0207] The generation server 4000 interacts with the user to obtain work history information from the user. The "work history information" obtained through the interaction is registered in the interaction information database 4422 for each member ID. The generation server 4000 uses the "work history information" registered in the interaction information database 4422 to create a "work history document". If "work history information" is registered in the work history database 4421, the generation server 4000 uses the "work history information" registered in the work history database 4421 and the "work history information" registered in the interaction information database 4422 to create a "work history document". The generation server 4000 registers the created "work history document" in the work history database 4421. The configuration of the work history database 4421 will be explained in detail later using Figure 27.

[0208] The dictionary database 4423 contains terms and their meanings extracted during the interaction between the generation server 4000 and the user. The generation server 4000 associates terms registered in the dictionary with the dictionary database 4423 from the description of the resume. The generation server 4000 (or sharing server 100) displays the resume on the user device 500. When the user clicks on a term registered in the dictionary in the description of the resume with the mouse, the generation server 4000 (or sharing server 100) displays the meaning of the term on the user device 500.

[0209] The large-scale language model 430 or a portion of the multiple types of databases may be stored in storage separate from the generation server 4000. For example, the generation server 4000 may connect to a different cloud service and store a portion of the multiple types of databases shown in Figure 22, or the large-scale language model 430, on that cloud. In this case, the generation server 4000 can access the necessary databases or large-scale language model 430 by communicating with the cloud via the Internet 50.

[0210] The processor 4001 connects to the Internet 50 via the communication interface 4004, following a program stored in the memory 4002. The processor 4001 connects to the Internet 50 and communicates with the user device 500 (see Figure 20). The processor 4001 accesses the database 4420 and performs processes such as extracting necessary data, registering new data in the database 4420, and updating data registered in the database 4420.

[0211] The processor 4001 communicates with the sharing server 100. For example, in response to a request from the sharing server 100, the processor 4001 sends the resumes of members (users) registered in the work history database 4421 to the sharing server 100.

[0212] [Company Database 121] Figure 23 shows an example of the company database 121. The company database 121 registers company IDs, company names, and company addresses for each company to identify them. In Embodiment 2, members are permitted to apply for and win contracts for work offered by various companies and departments.

[0213] [User Database 122] Figure 24 shows an example of the user database 122. The user database 122 contains various information about members. This information includes a member ID to identify the member, the ID of the company to which the member belongs, the member's name, the member's authority, and the department to which the member belongs.

[0214] Member privileges include administrator and applicant. Members with administrator privileges are authorized to use classification system 1000 as both recruiters and applicants. Members with applicant privileges are authorized to use classification system 1000 as applicants, but not as recruiters. Department heads within a company are granted administrator privileges to manage the side-job status of their subordinates. Administrators with administrator privileges are authorized to approve applications from their subordinates who are applicants. Therefore, administrators function as approvers.

[0215] [Community Database 127] Figure 25 shows an example of the community database 127. The community database 127 registers information about communities formed between companies. The community information includes a community ID to identify the community, a community name, and a list of IDs of companies belonging to the community. Each company can form various communities by agreeing with other companies. Companies belonging to a community can change the companies that belong to that community by agreement with other companies.

[0216] [Case Database 123] Figure 26 shows an example of the project database 123. The project database 123 contains information on job postings. The job posting information includes a project ID to identify the job posting, the ID of the company to which the recruiter who registered the job posting belongs, a list of non-disclosed company IDs, the disclosure level, the project title, estimated man-hours, estimated duration, and project details.

[0217] The list of non-disclosing company IDs registers the IDs of companies that are prohibited from disclosing their job postings. There are three levels of disclosure: "Our Company," "Within the Community," and "All." If the disclosure level is set to "All," applicants from outside the community will also be included in the disclosure.

[0218] On the right side of the job database 123 in Figure 26, the IDs of companies that can view the job postings are shown. For example, for the job posting corresponding to job ID=001, the disclosure level is set to "Our Company". In this case, only members belonging to the company that registered the job posting (Company ID=00A) can view the job posting corresponding to job ID=001.

[0219] Hereafter, using the project ID, the recruitment projects corresponding to each project ID may be referred to as project 001, project 002, project 003, etc. Similarly, using the community ID, the communities corresponding to each community ID may be referred to as community 01, community 02, community 03, etc., and using the member ID, the members corresponding to each member ID may be referred to as member P1, member P2, member P3, etc. Furthermore, using a part of the company ID, the companies corresponding to each company ID may be referred to as company A, company B, company C, etc.

[0220] In case 002, the disclosure level is set to "within the community." According to the community database 127 shown in Figure 25, the companies that have a community relationship with company A, which registered case 002, are company B and company C. Therefore, as shown in Figure 26, only members belonging to company A, company B, or company C can view case 002.

[0221] Case 003 has the same registered companies and disclosure level as Case 002. However, in Case 003, "00B" is registered in the list of non-disclosing company IDs. Therefore, as shown in Figure 26, only members belonging to either Company A or Company C can view Case 003, and members belonging to Company B are not authorized to view Case 003.

[0222] For job postings where the disclosure level is set to "all," all members can view the job posting. Job posting 005, shown in Figure 26, falls into this category. If one or more company IDs are registered in the non-disclosure company ID list for job posting 005, members belonging to those companies will not be granted permission to view job posting 005.

[0223] The estimated workload and estimated duration are used by the applicant and classification system 1000 to estimate the time required to process the job postings.

[0224] Furthermore, the company database 121, user database 122, and community database 127 may each be shared between the sharing server 100 and the generation server 4000. The project database 123 may register the member IDs of recruiters corresponding to the recruitment projects.

[0225] [Work History Database] Figure 27 shows an example of the work history database 4421. As shown in Figure 27, the work history database 4421 registers the "work history information" and "work history document" of each member (user) by member ID. As mentioned above, "work history information" refers to the information used to create the "work history document". The "work history document" is created based on the "work history information". In Embodiment 2, there are cases in which the "work history information" is acquired in advance before the interaction between the generation server 4000 and the user, and cases in which it is acquired through the interaction between the generation server 4000 and the user.

[0226] The format of the resume is predetermined by the designer. This format includes the types of required information and the order in which that information should be presented.

[0227] A resume includes both "formal items" and "substantive items." Formal items include, for example, "name," "age," "gender," "work history summary," "work history (duration and content)," "qualifications and skills," and "self-introduction." As shown in Figure 27, these formal items are arranged in the resume in the order of "name," "age," "gender," "work history summary," "work history (duration and content)," "qualifications and skills," and "self-introduction." The resume created by the generation server 4000 will list the "formal items" shown in Figure 27 in the order shown in Figure 27. Note that the types and order of "formal items" shown in Figure 27 are merely examples.

[0228] A resume contains content corresponding to each of the "formal items." The content that should be included in a resume includes predetermined "substantive items." As an example of "substantive items," Embodiment 2 introduces an item called "STAR."

[0229] Generally, "STAR" is known as one of the methods used by interviewers to effectively interview job seekers. "STAR" is a coined word created by combining the first letters of "S (Situation)", "T (Task)", "A (Action)", and "R (Result)". The generation server 4000 retrieves work history information from the user so that it includes the user's work history details for each of the four items intended by "STAR", and uses the retrieved work history information to create a resume.

[0230] The work history database 4421 contains not only the work history documents (including content) created for each user, but also format information for work history documents. This format information includes the necessary information to be included in a work history document (formal and substantive information). The generation server 4000 performs processing related to assisting in the creation of work history documents while referring to the format registered in the work history database 4421.

[0231] The work experience information registered in the work experience database 4421 is the basis for creating a resume, similar to the work experience information obtained through interaction between the generation server 4000 and the user. If no work experience information is registered in the work experience database 4421, the generation server 4000 creates a resume based on the "work experience information" obtained through interaction with the user. Therefore, in this disclosure, it is not essential that work experience information is registered in the work experience database 4421. The work experience database 4421 is an example of a work experience database in which work experience information collected before interaction by the collection unit is registered.

[0232] "Work history information" includes, for example, information about the name of the company the user previously worked for, the department, job duties, work experience, and period of employment, and includes the "STAR" information mentioned above. Alternatively, "Work history information" may refer to information about the qualifications the user holds.

[0233] "Work experience information" includes fragmented information about the user's work experience. Fragmented information is, for example, information about one of several required items included in a resume. One of several required items is, for example, information about the qualifications the user holds. Alternatively, fragmented information is information that constitutes part of the required items. For example, if the user holds three qualifications, the fragmented information is information about one of those three qualifications.

[0234] The work history database 4421, together with the dialogue information database 4422 and the dictionary database 4423, constitutes database 4420. Here, the work history database 4421, part of database 4420, is described in detail using diagrams. The dialogue information database 4422 and the dictionary database 4423 have already been explained in order to understand their respective structures, so their explanations will not be repeated here.

[0235] [Example of dialogue] Figures 28 and 29 illustrate an example of interaction between the generation server 4000 and a user. The user accesses the generation server 4000 using the user device 500. The generation server 4000 interacts with the user using the large-scale language model 430 in the following manner. In particular, the generation server 4000 diverges the interaction with the user to ensure that all of the user's work history is obtained without omission from the perspective of STAR (Situation, Task, Action, Result) related to the required information. As a result, the generation server 4000 obtains all of the user's work history information necessary for creating a resume.

[0236] This section describes an example in which the generation server 4000 interacts with the user using the screen of the user device 500. Specifically, the generation server 4000 asks the user about their work history by displaying request text and question text on the screen of the user device 500. The user answers the inquiry by typing text using the keyboard or other means on the user device 500. The generation server 4000 may also interact with the user via voice.

[0237] First, the generation server 4000 displays a request on the screen that says, "Please enter your work history." (Step SA1). The user responds to the request in Step SA1 with, "I worked as a lead designer at XX company." (Step SA2).

[0238] Based on the response in Step SA2, the generation server 4000 displays the question "What was the work environment like?" on the screen (Step SA3). The user responds to the question in Step SA3 with, "The team was understaffed and had a large backlog of unprocessed cases. The account manager set unrealistic deadlines, which caused stress and lowered morale within the team" (Step SA4).

[0239] Based on the answer in Step SA4, the generation server 4000 displays the following message on the screen: "Please describe the responsibilities and roles you played in the situation and challenges at that time" (Step SA5). The user responds to the question in Step SA5 with: "My role as team leader was not only to ensure the team could complete the work on time, but also to communicate processing capabilities to other departments and maintain team motivation" (Step SA6).

[0240] In the same manner, the generation server 4000 interacts with the user in a chat format to obtain all the necessary work history information from the user to create the work history document. The generation server 4000 then creates the work history document based on the work history information. The generation server 4000 displays a request message on the screen saying, "Your work history document has been created. Please review it." (Step SA7). Furthermore, the generation server 4000 presents the created work history document to the user (Step SA8). The user checks the work history document on the screen of the user device 500.

[0241] In particular, the generation server 4000 diverges the interaction between the generation server 4000 and the user so that all of the user's work history is acquired without omission from the perspective of STAR (Situation, Task, Action, Result).

[0242] Step SA3 is an example of a query to obtain work history information corresponding to "Situation" from the user. Similarly, Step SA5 is an example of a query to obtain work history information corresponding to "Task" from the user. Although not illustrated in Figures 28 and 29, "Action" refers to the method the user used to overcome "Situation" or "Task". "Result" refers to the outcome obtained by the user's "Action".

[0243] The dialogue program 4051 includes parameters to ensure that the dialogue diverges in terms of STAR (Situation, Task, Action, Result). Such parameters are designed, for example, by the system administrator of the generation server 4000.

[0244] In addition to, or as an alternative to, the generation server 4000 may allow the user to send a CSV (Comma Separated Value) file or the like containing work history information from the user device 500 to the generation server 4000, in addition to the chat-style dialogue.

[0245] To give some examples of "diverging," see below. For example, when information obtained from a user is expanded into text, there may be situations where the number of characters in that text is insufficient to meet the character limit specified in the required information. More specifically, when asked "What kind of work environment is it?", the answer received may not reach the required character limit (for example, 100 characters or more). In such cases, the generation server 4000 will ask the user additional questions such as "What was your relationship with your immediate supervisor like?" or "What was your relationship with your colleagues like?".

[0246] Alternatively, if the user's response lacks sufficient detail, the generation server 4000 will ask the user additional questions. For example, if the reason is unclear from the user's response, the generation server 4000 will ask the user additional questions. For instance, if the user responds with "The team is understaffed...", but the response does not include any keywords or other clues indicating the reason for the understaffing, the generation server 4000 will ask the user additional questions such as "Why was the team understaffed?"

[0247] When the information regarding "reasons" in the response obtained from the user is insufficient, the generation server 4000 may use the large language model 430 to organize the information and then present reason options to the user. For example, assume that the response obtained from the user is "The team was understaffed and had a large number of unprocessed cases. Since the account manager set unreasonable deadlines, the team also felt stressed and morale was low."

[0248] In this case, the generation server 4000 may determine that, for example, (a) the reason for the understaffing and (b) the reason for the user having a large number of unprocessed cases are insufficient, and present options to the user to obtain each reason.

[0249] To obtain "(a) the reason for the understaffing", the generation server 4000 may, for example, ask a query such as "It is said that there was understaffing. Could the reason be among the following? (Multiple selections possible)" and present "A: Unable to recruit even if hiring", "B: Suddenly received an increase in orders", "C: Many people resigning", and "D: Others (free description)" as options to the user.

[0250] To obtain "(b) the reason for the user having a large number of unprocessed cases", the generation server 4000 may, for example, ask a query such as "It is said that there were a large number of unprocessed cases. Could the reason be among the following? (Multiple selections possible)" and present "A: There was insufficient information sharing between sales and development, and too many cases were received relative to the number of development staff", "B: The management ability of the account manager is insufficient", "C: The skills of the development staff are insufficient", and "D: Others (free description)" as options to the user.

[0251] The user may select three options, A, B, and C, as the answer to (a) above and two options, A and B, as the answer to (b) above.

[0252] As a result, the generation server 4000 can extract more detailed information from the user regarding the reasons for (a) and (b) above. For example, the information obtained from the user before asking additional questions was: "The team was understaffed and had a large backlog of pending cases. The account manager set unrealistic deadlines, which stressed the team and lowered morale."

[0253] In response, by asking additional questions regarding (a) and (b) above, the generating server 4000 can extract information from the user such as, "There was insufficient information sharing between sales and development, the number of projects received was excessive and too rapid compared to the number of development personnel, and the account managers lacked the management skills to handle all the projects. Due to the large number of projects and the tendency for working hours to exceed the limit, many people left the company, further exacerbating the personnel shortage. Furthermore, we were also recruiting experienced mid-career employees who could hit the ground running, but we were unable to find enough people."

[0254] [Example of dialogue (when dictionary registration occurs)] Figures 30 and 31 show an example of a dialogue between the generation server 4000 and the user (with dictionary registration). Here, we will explain the operation of the generation server 4000 when it detects a term that should be registered in the dictionary database 4423 in the dialogue shown in Figures 28 and 29, which have already been explained.

[0255] As shown in Figure 30, the generation server 4000 detects "lead designer" in the input of step SA2 as a term that is easily misunderstood. In this case, the generation server 4000 displays the inquiry message "Does 'lead designer' mean ○○○○○○?" on the screen (step SA2a).

[0256] The user responds to the inquiry in step SA2a with "No, it means xxxxxx" (step SA2b). Based on the response in step SA2b, the generation server 4000 displays the message "Understood" on the screen (step SA2c). Furthermore, in step SA2c, the generation server 4000 registers the term "lead designer" and its meaning together in the dictionary database 4423.

[0257] Subsequently, the generation server 4000 continues to interact with the user and obtains all the necessary work experience information from the user to create the resume. The generation server 4000 then creates the resume based on the work experience information. The generation server 4000 displays a request message on the screen saying, "Your work experience has been created. Please review it." (Step SA7). Furthermore, the generation server 4000 presents the created resume to the user (Step SA8).

[0258] As shown in Figure 31, the term "Lead Designer" in the resume is underlined. The user understands that the underlined term is linked to a dictionary. When the user clicks on a term displayed on the screen with a mouse, the meaning of the term registered in the dictionary database 4423 is displayed on the screen. The user checks the resume, including the dictionary link, on the user device 500 screen. The generation server 4000 registers the resume approved by the user in the resume database 4421, linked to the user's member ID.

[0259] Resumes registered in the work history database 4421 are sent to the recruiter device 200 upon user request. The recruiter device 200 displays the resume, including links to a dictionary. When a recruiter clicks on a term displayed on the screen with a mouse or other means, the meaning of that term, registered in the dictionary database 4423, is displayed on the screen. This allows recruiters to accurately understand the terms used in the resume.

[0260] [Generation Server Functions] Figure 32 is a diagram illustrating the processing performed by the generation server 4000 from the perspective of the divergence phase and the convergence phase.

[0261] The generation server 4000 includes a large-scale language model 430 and a program 4050 (dialogue program 4051, confirmation program 4052, and creation program) that executes processes to support the creation of a resume using the large-scale language model 430. The processes to support the creation of a resume are divided into a divergence phase and a convergence phase, as shown in Figure 32.

[0262] The "Conversational AI (Artificial Intelligence) 4110" is formed by combining the dialogue program 4051 and the large-scale language model 430. The Conversational AI 4110 obtains a large amount of work history information from the user that is necessary for creating a resume. Furthermore, the Conversational AI 4110 asks the user whether the created resume aligns with the user's intentions.

[0263] The combination of the verification program 4052 and the large-scale language model 430 constitutes the "Verification AI 4120". The Verification AI 4120 verifies that all work experience information corresponding to the necessary entries required for creating a resume has been acquired without any omissions.

[0264] The combination of the creation program 4053 and the large-scale language model 430 constitutes the "creation (summarization) AI 4130". The creation (summarization) AI 4130 creates a resume in accordance with a predetermined resume format.

[0265] In the divergence phase, the conversational AI 4110 engages in a chat-like conversation with the user, obtaining work experience information related to the required information for the resume from the user. At this time, the conversational AI 4110 controls the conversation so that it diverges in terms of "STAR (Situation, Task, Action, Result)" according to parameters designed by the system administrator.

[0266] In the divergence phase, the verification AI 4120 reads the resume format. The resume format contains the necessary information required to create a resume. The verification AI 4120 verifies that all work experience information corresponding to the necessary information required to create a resume has been obtained without any omissions. If work experience information is missing, the verification AI 4120 instructs the dialogue AI 4110 to continue the conversation regarding the missing work experience information.

[0267] The conversational AI 4110 stores the conversation history. Therefore, if instructed by the conversational AI 4120 to continue the conversation, the conversational AI 4110 can continue the conversation from where it left off without having to restart a conversation that has already been completed.

[0268] In the divergent phase, the conversational AI 4110 engages in divergent dialogues, and the confirmation AI 4120 repeatedly verifies the information until all necessary work history information corresponding to the required items for creating a resume is obtained without any omissions.

[0269] After the divergence phase, the generation server 4000 transitions to the convergence phase. In the convergence phase, the creation AI (abstracting AI) 4130 has the function of arranging the acquired information in a predetermined format. More specifically, the creation AI 4130 creates a resume according to the format of a pre-defined resume. The dialogue AI 4110 presents the created resume to the user. The user checks the resume. If the user determines that the resume does not conform to the user's intention, the user instructs the dialogue AI 4110 to make corrections. In this case, the creation AI 4130 corrects the resume based on the user's correction instruction. The dialogue AI 4110 presents the corrected resume to the user. The user checks the resume.

[0270] In the convergence phase, until the resume intended by the user is created, the creation (correction) of the resume by the creation AI 4130 and the dialogue between the dialogue AI 4110 and the user are repeated.

[0271] Figure 33 is a diagram showing the functional configuration of the generation server 4000. As shown in Figure 33, the generation server 4000 includes a control unit 4010. The control unit 4 is realized by the processor 4001, the memory 4002, and the communication interface 4004 shown in Figure 22.

[0272] The control unit 4010 can function as the dialogue AI 4110, the confirmation AI 4120, and the creation AI 4130 shown in Figure 32 by using the large language model 430. The control unit 4010 executes processing related to the creation support of the resume by accessing the database 4420.

[0273] The control unit 4010 includes a data import unit 4011, an information acquisition unit 4012, a dictionary information registration unit 4013, a dictionary information display unit 4014, an information sorting unit 4015, and a resume registration unit 4016.

[0274] The data import unit 4011 imports the user's existing work history information into the conversational AI 4110. For example, the data import unit 4011 can receive work history information from the user via the user device 500 in a predetermined file format. The control unit 4010 uses the imported work history information to create a work history document. In this way, by utilizing the user's existing work history information, the conversational cost of the conversational AI 4110 can be reduced.

[0275] The information acquisition unit 4012 uses the conversational AI 4110 to acquire work history information from the user. The information acquisition unit 4012 uses the verification AI 4120 to confirm that all work history information corresponding to the necessary entries for creating a work history document has been acquired without any omissions.

[0276] The dictionary information registration unit 4013 registers unique terms obtained during the interaction between the user and the dialogue AI 4110 in the dictionary database 4423. Here, unique terms may be information specific to an organization, such as a company.

[0277] The dictionary information display unit 4014 displays terms registered in the dictionary database 4423 as links if they appear in the resume. The information organization unit 4015 uses the creation AI 4130 to create the resume in a predetermined format.

[0278] The resume registration unit 4016 stores the created resumes in the resume database 4421. In particular, if a term registered in the dictionary database 4423 is included in the resume, the resume registration unit 4016 generates a link that associates the term in the resume with the meaning of the term registered in the dictionary database 4423, and then stores the resume information in the resume database 4421.

[0279] [Processing procedure of control unit 4010] Next, with reference to Figures 34 to 39, an example of the processing procedure of the control unit 4010 included in the generation server 4000 will be described. Here, the processing procedures of the control unit 4010 will be described for each of the following: the data import unit 4011, the information acquisition unit 4012, the dictionary information registration unit 4013, the dictionary information display unit 4014, the information organization unit 4015, and the resume registration unit 4016.

[0280] Figure 34 is a flowchart showing the processing procedure of the data import unit 4011. First, the data import unit 4011 obtains the format information of the resume and the work history information registered before the interaction between the dialogue AI 4110 and the user from the work history database 4421 (step SA11). Step SA11 is an example of a collection unit that collects work history information related to the user's work history. Furthermore, step SA11 is an example of an acquisition unit that obtains format information containing the necessary information to be included in the resume.

[0281] "Work history information registered before the interaction between the conversational AI 4110 and the user" refers to the "work history information" shown in Figure 27. The data import unit 4011 may also read the user's work history information from a predetermined local file in step SA11. The work history document format information includes the required information for the work history document, as well as the order in which these required information should be placed.

[0282] Next, the data import unit 4011 inputs the work history information obtained in step SA11 as preliminary information into the conversational AI 4110 (step SA12). Next, the data import unit 4011 determines whether the work history information obtained in step SA11 contains all the information necessary to create a work history document (step SA13). The data import unit 4011 calls the confirmation AI 4120 and makes the determination in step SA13 by referring to the work history document format information.

[0283] If the work history information obtained in step SA11 contains all the information necessary to create a resume, no interaction between the conversational AI 4110 and the user is required. Therefore, if the work history information obtained in step SA11 contains all the information necessary to create a resume, the data import unit 4011 passes the processing to the information organization unit 4015. The information organization unit 4015 then creates the resume.

[0284] If the work history information obtained in step SA11 does not contain all the information necessary to create a work history document, the data import unit 4011 passes the processing to the information acquisition unit 4012.

[0285] Figure 35 is a flowchart showing the processing procedure of the information acquisition unit 4012. First, the information acquisition unit 4012 calls the dialogue AI 4110 and the confirmation AI 4120 using the large-scale language model 430 (step SA21).

[0286] Next, the information acquisition unit 4012 interacts with the user using the conversational AI 4110 (step SA22). Through this, the information acquisition unit 4012 acquires conversational information (including work history information) from the user. Step SA22 is an example of a collection unit that collects work history information related to the user's work history. Furthermore, step SA22 is an example of a conversation unit that collects work history information by interacting with the user. Next, the information acquisition unit 4012 stores the conversational information in the conversational information database 4422 (step SA23). The conversational information is information that shows the history of conversations with the user.

[0287] Next, the information acquisition unit 4012 determines, based on the dialogue information, whether all the necessary work history information corresponding to the required entries for creating the work history document has been acquired without omission (step SA24). At this time, the information acquisition unit 4012 refers to the format information acquired in step SA11 and makes the determination in step SA24. The information acquisition unit 4012 may also acquire the format information from the work history database 4421 in a step other than step SA11. If the information acquisition unit 4012 has not acquired all the necessary work history information corresponding to the required entries without omission, it returns to step SA22.

[0288] If the information acquisition unit 4012 has successfully acquired all the necessary work history information corresponding to the required fields, it asks the user whether there is any other work history information (step SA25). If there is other work history information, the information acquisition unit 4012 returns to step SA22. If there is no other work history information, the information acquisition unit 4012 passes the process to the information organization unit 4015. The information organization unit 4015 then creates the work history document.

[0289] Figure 36 is a flowchart showing the processing procedure of the dictionary information registration unit 4013. First, the dictionary information registration unit 4013 detects words to be registered in the dictionary during the dialogue between the conversational AI 4110 and the user (step SA31). Words to be registered in the dictionary include proper nouns and polysemous word information. Words to be registered in the dictionary may also be information specific to an organization, such as a company. Next, the dictionary information registration unit 4013 confirms the meaning of the word with the user (step SA32).

[0290] Next, the dictionary information registration unit 4013 obtains the user's response (step SA33). Then, the dictionary information registration unit 4013 registers the word detected in step SA31 and its meaning based on the user's response as dictionary information in the dictionary database 4423 (step SA34), and terminates the process according to this flowchart. Step SA34 is an example of a registration unit that registers the meaning of terms included in the work history information into the dictionary database.

[0291] Figure 37 is a flowchart showing the processing procedure of the dictionary information display unit 4014. First, the dictionary information display unit 4014 retrieves the user's work history from the work history database 4421 (step SA41). Next, the dictionary information display unit 4014 retrieves terms present in the work history from the dictionary database 4423 (step SA42). Then, the dictionary information display unit 4014 displays the work history with links on the screen of the user device 500 (step SA43).

[0292] Next, the dictionary information display unit 4014 displays the dictionary information corresponding to the link on the screen of the user device 500 in response to the user's operation (step SA44), and terminates the process based on this flowchart. The dictionary information display unit 4014 is an example of a display unit that displays a resume on a display device. The user device 500 is an example of a display device. The dictionary information display unit 4014 may also display the resume on a display device equipped on the generation server 4000, either in place of the user device 500 or in addition to the user device 500.

[0293] Figure 38 is a flowchart showing the processing procedure of the information organization unit 4015. First, the information organization unit 4015 calls the created AI 4130 using the large-scale language model 430 (step SA51).

[0294] Next, the information organization unit 4015 reads the dialogue information from the dialogue information database 4422 (step SA52). Also, if pre-registered work history information exists in the work history database 4421, the information organization unit 4015 reads the work history information from the work history database 4421 in step SA52.

[0295] Next, the information organization unit 4015 creates a resume from the dialogue information (step SA53). Furthermore, if the information organization unit 4015 has read work history information from the work history database 4421, in step SA53 it creates a resume from both the work history information and the dialogue information. Step SA53 is an example of a resume creation unit.

[0296] The information organization unit 4015 refers to the format information obtained in step SA11 and creates a resume in the format that conforms to the format information. The information acquisition unit 4012 may acquire the format information from the resume database 4421 in a step separate from step SA11. Next, the information organization unit 4015 displays the resume on the screen of the user device 500 (step SA54).

[0297] Next, the information processing unit 4015 determines whether or not it has received the user's approval operation (step SA55). If the information processing unit 4015 has received the user's approval operation, it terminates the process based on this flowchart.

[0298] If the information organization unit 4015 has not received an approval operation from the user, it determines whether or not it has received a rejection operation from the user (step SA56). If the information organization unit 4015 has not received a rejection operation from the user, it returns to step SA55. If the information organization unit 4015 has received a rejection operation from the user, it passes the process to the information acquisition unit 4012. The information acquisition unit 4012 calls the dialogue AI 4110 again and interacts with the user to acquire new dialogue information from the user.

[0299] Furthermore, the information processing unit 4015 may not only acquire new dialogue information from the user, but may also engage in dialogue with the user again without changing the conditions prior to the dialogue. This is because, generally, large-scale language models may generate multiple answers (sentences) with different nuances for the same question and conditions.

[0300] Figure 39 is a flowchart showing the processing procedure of the resume registration unit 4016. First, the resume registration unit 4016 retrieves the user's resume from the resume database 4421 (step SA61). Next, the resume registration unit 4016 refers to the dictionary database 4423 to generate links between the resume and dictionary information (step SA62). Next, the resume registration unit 4016 stores the linked resume in the resume database 4421 (step SA63).

[0301] Next, the resume registration unit 4016 receives a request from the user to output the resume (step SA64). Then, the resume registration unit 4016 outputs the resume to the user device 500 (step SA65), and the process based on this flowchart ends. The user device 500 may display the resume on its screen. Alternatively, the user device 500 may store the resume data in memory until instructed by the user.

[0302] [Sequence of processes executed by the classification system 1000] Figure 40 is a timing chart showing the processing procedures for the sharing server 100, recruiter device 200, applicant device 300, and generation server 4000 regarding resumes. The processing flow of the classification system 1000 regarding resumes will be explained using the timing chart shown in Figure 40.

[0303] First, the applicant accesses the generation server using the applicant device 300 and creates a resume (step SA101). The generation server 4000 registers the created resume in the resume database 4421 (step SA102). The detailed processing procedures for steps SA101 and SA102 have already been explained using Figures 34 to 39, so that explanation will not be repeated here.

[0304] Next, the applicant accesses the sharing server 100 using the applicant device 300, searches for job postings, and then decides where to apply (step SA103). Next, the applicant operates the applicant device 300 to send a command from the applicant device 300 to the sharing server 100 requesting the applicant's resume (step SA104). Upon receiving the command, the sharing server 100 identifies the applicant (step SA105). More specifically, the sharing server 100 identifies the applicant's member ID.

[0305] Next, the sharing server 100 sends a command to the generation server 4000 requesting the submission of the resume (step SA106). This command includes the applicant's member ID. Based on the member ID included in the command, the generation server 4000 searches the resume database 4421 for the applicant's resume (step SA107).

[0306] Next, the generation server 4000 sends the resumes detected as a result of the search to the sharing server 100 (step SA108). The sharing server 100 then obtains the applicant's resume (step SA109). Next, the sharing server 100 sends the obtained resume to the applicant device 300 (step SA110).

[0307] The applicant device 300 displays the received resume on its screen (step SA111). If the resume contains terms registered in the dictionary database 4423, the resume, including a link to the dictionary, will be displayed on the screen.

[0308] Applicants confirm that there are no problems with the resume displayed on the screen. For example, applicants confirm that there is no information that needs to be updated in the resume. If there are no problems with the resume, applicants approve it using the keyboard or other means. If there are problems with the resume, applicants reject it using the keyboard or other means.

[0309] The applicant device 300 determines whether or not it has detected an approval operation (step SA112). If the applicant device 300 detects a disapproval operation instead of an approval operation, it returns to step SA101. In this case, the process of creating the resume is executed again. However, it is desirable that the generation server 4000 does not execute the process of creating the resume from the beginning, but rather interacts with the applicant regarding matters that need to be corrected and corrects the contents of the resume.

[0310] If the applicant device 300 detects an approval operation, it sends a command to the sharing server 100 instructing it to send the resume (step SA113). The sharing server 100 identifies the recruiter to whom the resume should be sent (step SA114). For example, the sharing server 100 identifies the recruiter to whom the resume should be sent based on the application destination determined in step SA103.

[0311] Next, the sharing server 100 sends the resume to the recruiter device 200 (step SA115). The recruiter device 200 displays the received resume on its screen (step SA116). If the resume contains terms registered in the dictionary database 4423, the resume, including a link to the dictionary, is displayed on the screen. The recruiter uses the resume to understand the applicant's work history. When the applicant clicks on a term registered in the dictionary within the resume using the mouse, the meaning of the term is displayed on the recruiter device 200. Therefore, if the resume contains terms that the applicant does not understand, such as internal company jargon, the recruiter can understand the meaning of those terms.

[0312] In Figure 40, the sharing server 100 and the generation server 4000 may be configured as a single server. That is, the functions of the generation server 4000 may be provided to the sharing server 100, or the functions of the sharing server 100 may be provided to the generation server 4000. Steps SA110 and SA115 are examples of output units configured to output the resumes created by the creation unit to the user device.

[0313] As described above, Embodiment 2 can assist users in creating a resume that includes all necessary information and avoids any duplicate entries. In addition, Embodiment 2 can assist users in creating a resume that appropriately reflects their work experience, regardless of their document creation abilities. Furthermore, Embodiment 2 provides the following advantages.

[0314] a. Regardless of the user's document creation ability, we can provide the user with a resume that appropriately expresses their skills and experience.

[0315] b. It can collect a wide variety of user work history information without any omissions. c. By providing users with an interactive interface, the burden on users to appropriately extract work history information can be reduced.

[0316] d. It can reduce the burden on users in creating their resumes. e. Because resumes are created in a standardized format, it is possible to improve efficiency when managing and searching for a large number of resumes.

[0317] f. For users who find it difficult to maintain motivation for the cumbersome process of creating a resume, such as employees applying for internal projects or those seeking side jobs outside the company, this provides an interface for creating high-quality resumes.

[0318] g. If a term registered in dictionary database 4423 is included in a resume, the resume will be displayed on the screen with a link to the dictionary, so that recruiters can understand the meaning of any unfamiliar terms included in the resume.

[0319] Furthermore, some components of the generation server 4000 may be configured on separate devices. For example, at least one of the multiple databases included in database 4420 and the large-scale language model 430 may be located on a cloud separate from the generation server 4000. In this case, it is sufficient that the generation server 4000 and the cloud are connected via a network to enable communication.

[0320] [Differentiation] In Embodiment 2, an example was shown in which the generation server 4000 is located within the classification system 1000. However, a generation server that creates resumes may be constructed independently of the classification system 1000. An example of a generation server that creates resumes independently of the classification system 1000 will be described.

[0321] Figure 41 shows the functional configuration of the generation server 4000A in a modified example. The generation server 4000A, like the generation server 4000, has a processor, memory, communication interface, and storage, and includes a control unit 4010A realized by these configurations.

[0322] The generation server 4000A is connected to the user device 500A, which is used by the user, via the internet or other means, enabling communication between the two devices.

[0323] The generation server 4000A includes a requirements database (requirements DB) 426, a source information storage database (source information storage DB) 427, a generation format database (generation format DB) 428, and large language models 430A and 430B. These databases and large language models are configured in the storage of the generation server 4000A.

[0324] Large-scale language models 430A and 430B, like large-scale language model 430, are natural language processing models trained using large amounts of text data. In this disclosure, Google's BERT® (LaMBDA) or OpenAI's GPT-4 may be used as the large-scale language model. In this disclosure, the large-scale language model is an example of a natural language processing algorithm. Other language models other than the large-scale language model may be used as the natural language processing algorithm. Models generated by machine learning, such as pattern matching, may be used as the natural language processing algorithm. Furthermore, instead of using the two large-scale language models 430A and 430B, one large-scale language model may be used. For example, large-scale language model 430 may be used.

[0325] As shown in Figure 41, the control unit 4010A includes an input unit 4021, a content satisfaction determination unit 4022, an input request unit 4023, a source information storage unit 4024, a generation unit 4025, a confirmation request unit 4026, and an output unit 4027.

[0326] The input unit 4021 receives information entered by the user using the user device 500. The information entered by the user includes work history information. The input unit 4021 may, for example, accept work history information in chat format. The input unit 4021 may also accept work history information in audio file format and other file formats. When using a large-scale language model such as GPT-3 and a chat-format UI such as ChatGPT, files may be attached to the chat. In this case, the generation server 4000 may be provided with a function to convert files into chat-format documents.

[0327] The output unit 4027 outputs various information to the user device 500. The output unit 4027 may output not only chat-formatted information but also other files such as audio to the user device 500. When the control unit 4010A interacts with the user in chat format, the input of information to the input unit 4021 and the output of information from the output unit 4027 are repeatedly performed.

[0328] The fulfillment requirements database 426 contains information regarding the requirements (fulfillment requirements) that the large-scale language model 430A uses to determine whether all work experience information corresponding to the necessary entries for creating a resume has been acquired without any omissions.

[0329] Information regarding multiple types of satisfaction requirements may be registered in the satisfaction requirement database 426. In this case, the control unit 4010A may transmit multiple types of satisfaction requirements to the user device 500A and allow the user to select their preferred satisfaction requirements. Alternatively, the control unit 4010A may receive the satisfaction requirements set by the user via the user device 500A.

[0330] Furthermore, the control unit 4010A may use the contents of numerous satisfaction requirements as training data to generate a trained model that functions as an AI. In that case, the generated trained model may be used instead of the large-scale language model. Alternatively, the satisfaction requirements may be used in the Prompt of the already existing large-scale language model 430A. The satisfaction requirements may include specific information, such as the number of companies that the user wants to include in their resume.

[0331] The content satisfaction determination unit 4022 uses the large-scale language model 430 to determine whether the information obtained from the input unit 4021 satisfies the satisfaction requirements. If the content satisfaction determination unit 4022 determines that the information obtained from the input unit 4021 does not satisfy the satisfaction requirements, it instructs the input request unit 4023 to obtain additional information. If the content satisfaction determination unit 4022 determines that the information obtained from the input unit 4021 satisfies the satisfaction requirements, it outputs the information obtained from the input unit 4021 to the original information storage unit 4024.

[0332] The input request unit 4023 instructs the large-scale language model 430A to continue interacting with the user so that additional information (missing information) can be obtained from the user, based on the instructions of the content satisfaction determination unit 4022. If multiple pieces of additional information are required, the large-scale language model 430A may interact with the user so that multiple pieces of additional information are obtained all at once, or it may interact with the user so that each piece of additional information is obtained sequentially. For example, if information A, B, and C are missing, the large-scale language model 430A may first query the user for information A.

[0333] Furthermore, the large-scale language model 430A may automatically generate dialogue rules to prompt the user for additional information, or it may select dialogue rules from pre-configured options.

[0334] The source information storage unit 4024 temporarily stores the work history information that will be used as the basis for generating the work history document. The source information storage unit 4024 outputs the temporarily stored work history information to the source information storage database 427. The source information storage database 427 registers the work history information. The generation unit 4025 creates the work history document using the work history information registered in the source information storage database 427. The generation unit 4025 may also have a function to create the work history document without waiting for a determination from the content satisfaction determination unit 4022.

[0335] For example, if a user already has a draft of their resume, they may only want to finalize or format the resume. In this case, the input unit 4021 may receive the draft resume data from the user via the user device 500. Furthermore, the input unit 4021 may register the received data in the source information storage database 427. The generation unit 4025 may create the resume based on the draft data registered in the source information storage database 427.

[0336] The generation format database 428 contains the formats necessary for the large-scale language model 430B to generate work history data.

[0337] Multiple formats may be registered in the generation format database 428. In this case, the control unit 4010A may send multiple formats to the user device 500A and allow the user to select their preferred format. Alternatively, the control unit 4010A may accept a format set by the user via the user device 500A.

[0338] Furthermore, the control unit 4010A may use a number of formats as training data to generate a trained model that functions as an AI. In that case, the generated trained model may be used instead of the large-scale language model. Alternatively, the format may be used in the Prompt of the already existing large-scale language model 430A.

[0339] The generation unit 4025 generates a resume using the large-scale language model 430B. The formats registered in the generation format database 428 are input into the large-scale language model 430B. The generation unit 4025 refers to the source information storage database 427 and creates a resume that reflects the work experience information in a predetermined format. Alternatively, the large-scale language model 430B may create the resume instead of the generation unit 4025. The resume generated by the generation unit 4025 is output to the confirmation request unit 4026.

[0340] The confirmation request unit 4026 requests the user to confirm the work history document. The confirmation request unit 4026 transmits the work history document to the user device 500 via the output unit 4027. The user confirms the contents of the work history document displayed on the user device 500. The confirmation request unit 4026 may also divide the contents of the work history document into multiple items and have the user confirm the contents of the work history document for each item.

[0341] For example, if the resume contains information about both Company A and Company B, the confirmation request unit 4026 may have the user confirm the information about Company A, and then have the user confirm the information about Company B. Of course, the confirmation request unit 4026 may also have the user confirm all the contents of the resume at once. The large-scale language model 430B may automatically generate dialogue rules for when the user is asked to review the resume, or it may select dialogue rules from pre-configured options.

[0342] Furthermore, some components of the generation server 4000A may be configured separately from the generation server 4000A. For example, at least one of the sufficiency requirements database 426, the source information storage database 427, the generation format database 428, the large-scale language model 430A, and the large-scale language model 430B may be configured separately from the generation server 4000A. In this case, it is sufficient that the device configured separately from the generation server 4000A and the generation server 4000A are connected to each other via a network so that they can communicate with each other.

[0343] [Variations related to the use of IPC classification] Generally, a technical professional's resume details the skills they possess. However, with the rapid development of IT-related technologies, new and previously little-known fields of technology are constantly emerging, leading to rapid diversification of technologies. On the other hand, the level of detail of skills possessed by users varies from field to field. As a result, consistency in the description of skills in resumes is being lost. This is hindering recruiters when analyzing applicants' resumes.

[0344] Such problems also exist within companies. It is important for those in a position to evaluate employees to understand and systematically organize the skills they possess. This is thought to enable the maximum utilization of the company's technical resources. However, given the circumstances described above, even if each employee is required to submit a resume, it is difficult to manage each employee's skills uniformly according to a consistent standard based on the submitted resumes.

[0345] Therefore, an index is needed that can classify various skills using a uniform standard. One possible solution is to use the IPC (International Patent Classification). The IPC is an internationally standardized technical classification system for classifying patented inventions. By using the IPC, it is possible to classify technologies in detail by dividing them into hierarchical levels such as "sections," "subsections," "classes," "subclasses," "main groups," and "subgroups."

[0346] IPC is originally an index used to classify patent documents. However, by utilizing IPC as an index to classify engineers' skills, it is possible to classify the skills of various engineers, including those working with the latest technologies, using a uniform standard. However, because IPC is a highly specialized classification index used in the patent industry, it is difficult for users unfamiliar with IPC to accurately select the IPC that corresponds to their skills from among the many IPC options.

[0347] Therefore, this document proposes a system that enables users to select appropriate IPCs (Instructional Programming Classifications) corresponding to their skills through interaction between the generation server 4000 and the user. Here, IPCs are assumed as an example of required information or format information for a resume, and user career information necessary to obtain IPCs corresponding to the user's skills is assumed as an example of career information. The generation server 4000 collects the user's career information necessary to obtain IPCs by interacting with the user using a natural language processing algorithm. Skills are an example of work ability information.

[0348] Figures 42 and 43 illustrate other examples of interactions between the generation server 4000 and the user. Figures 42 and 43 show examples in which the generation server 4000 obtains appropriate IPCs corresponding to the user's skills through interaction with the user.

[0349] First, the generation server 4000 displays a request message on the screen that reads, "Please enter your work history." (Step SA201). The user responds to the request in Step SA1 with their work history (Step SA202). In Step SA202, for example, work history related to the development of lithium-ion secondary batteries is provided.

[0350] Based on the response in step SA2, the generation server 4000 uses the large-scale language model 430 to organize the information and presents the user with IPCs that may correspond to the user's skills (step SA203). In step SA203, the user is presented with one or more "IPCs" corresponding to each of the pieces of information obtained through the dialogue, such as "Research and development of lithium-ion secondary batteries," "Development of new cathode materials," "Development of lithium cobalt nickel manganese oxide," "Development of silicon anodes," "Development of electrolytes that suppress degradation," and "Development and practical application of lithium-ion secondary batteries using lithium titanate as the anode."

[0351] Next, the generation server 4000 presents the user with several IPCs as selection candidates from among those presented in step SA203 (step SA204). At this time, the generation server 4000 may present the selection candidates along with checkboxes, as shown in Figure 43. The user considers whether there is an IPC among the options that seems appropriate. If there is an IPC among the options that seems appropriate, the user checks the checkbox corresponding to that IPC to respond. In this case, a button 601 as shown in Figure 43 may be displayed on the screen of the user device 500. After checking the checkbox, the user clicks button 601. When the generation server 4000 detects the click operation of button 601, it determines the IPC to be included in the user's resume based on the user's response.

[0352] In this way, the generation server 4000 suggests to the user relevant IPCs from among a large number of IPCs based on the work history obtained through the process. Therefore, even if the user does not have specialized knowledge of IPCs, they can select IPCs that are relevant to their skills. Here, an example of using "subgroups" is shown, but higher-level symbols such as "subclasses" or "groups" may be used depending on the level of granularity desired for classification. In addition, explanations of "subclasses," "groups," or "main groups" may be added to the description text of "subgroups," or buttons such as a question mark may be added to allow users to check the content of higher levels as needed.

[0353] While this example shows the user selecting an IPC related to their technical field, the IPC presented to the user may be a "section," "subsection," "class," "subclass," "main group," or "subgroup." For example, a subclass, such as "H01: Electrical Components," may be presented to the user. Furthermore, along with the classification code, the name of the technical field to which that classification code belongs may also be presented to the user.

[0354] If no suitable IPC exists among the options proposed in step SA205, the user can request the generation server 4000 to present other candidates or to restart the work history entry process from the beginning. In this case, buttons 602 and 603, as shown in Figure 43, may be displayed on the screen of the user device 500.

[0355] When the generation server 4000 detects a click operation of button 602, it selects an IPC from the IPCs presented in step SA203 that is different from the options presented in step SA204, and presents the selected IPC to the user. When the generation server 4000 detects a click operation of button 603, it returns to step SA204 and prompts the user to enter their work history.

[0356] Furthermore, the generation of IPC candidates may be performed automatically multiple times. The generation server 4000 may generate different answers for the same text (or the same question). Therefore, by automatically generating IPCs multiple times from step SA202 before proposing IPC candidates in step SA204, the range of IPC candidates can be broadened. This increases the likelihood that the user can select an IPC that is suitable for them.

[0357] As described above, the generation server 4000 assists the user's actions in identifying IPCs corresponding to the user's skills through interaction between the generation server 4000 and the user. Note that the processing in step SA203 is not mandatory in Embodiment 2. That is, after the generation server 4000 detects multiple IPCs that are thought to correspond to the user's skills through interaction, it may present some of the detected IPCs to the user in step SA204 without executing the processing in step SA203.

[0358] Furthermore, the generation server 4000 may autonomously present other candidates to the user, not only when requested by the user to present other candidates again. For example, the generation server 4000 may execute the process of creating candidate options multiple times and present multiple candidates to the user. In general, the answers obtained from a large-scale language model to a query may differ even if the content of the first query and the second query are the same. Therefore, by having the generation server 4000 execute the process of creating candidate options multiple times, it is possible that the generation server 4000 will derive multiple candidates from different perspectives. If the user's selection for each of these multiple candidates can be obtained, the generation server 4000 can obtain a broader range of answers regarding IPC from the user.

[0359] Figure 44 shows the functional configuration of the generation server 4000, including the IPC classification unit 4017. In this modified example, the IPC classification unit 4017 is added to the control unit 4010 of the generation server 4000 shown in Figure 33. The IPC classification unit 4017 determines the user's IPC to be included in the resume. The IPC classification unit 4017 can present the user with IPCs that are considered to correspond to the user's skills, using the user's work history information obtained through dialogue. The IPC classification unit 4017 can also present the user with IPCs that are considered to correspond to the user's skills, using the work history information imported by the data import unit 4011. After presenting the user with multiple IPCs, the IPC classification unit 4017 determines the IPC that is appropriate for the user's skills through dialogue. The determined IPC is taken into the information organization unit 4015. The information organization unit 4015 uses the creation AI 4130 to create a resume including the IPC in a predetermined format. The resume registration unit 4016 stores the created resumes in the resume database 4421.

[0360] Figure 45 is a flowchart showing the processing procedure of the IPC classification unit 4017. First, the IPC classification unit 4017 calls the conversational AI 4110 using the large-scale language model 430 (step SA71).

[0361] Next, the IPC classification unit 4017 interacts with the user using the conversational AI 4110 (step SA72). Through this, the IPC classification unit 4017 obtains conversational information (including work history information) from the user. Step SA72 is an example of a collection unit that collects work history information related to the user's work history. Furthermore, step SA72 is an example of a dialogue unit that collects work history information by interacting with the user. Note that information regarding IPCs may be stored in the dictionary database 4423. In this case, the IPC classification unit 4017 may refer to the dictionary database 4423 to determine which IPCs to present to the user.

[0362] Next, the IPC classification unit 4017 presents the user with several IPCs that are considered to correspond to the user's skills (step SA73). Then, the IPC classification unit 4017 determines whether or not the user has performed the operation of selecting an IPC from among the presented IPCs (step SA74).

[0363] The IPC classification unit 4017 determines the user's request (step SA76) if the user has not performed an operation to select an IPC. More specifically, if the user requests the IPC to be recreated, the IPC classification unit 4017 returns to step SA73, changes the IPCs it presents, and then presents several IPCs to the user again. This process is performed, for example, when a click operation of button 602 shown in Figure 43 is detected. If the user requests to restart the interaction, the IPC classification unit 4017 returns to step SA72 and interacts with the user again. This process is performed, for example, when a click operation of button 603 shown in Figure 43 is detected.

[0364] The IPC classification unit 4017 determines, for example, that the user is performing an operation to select an IPC when it detects a click operation of button 601 shown in Figure 43. In this case, the IPC classification unit 4017 determines the IPC to be included in the resume based on the user's operation (step SA75), and completes the processing according to this flowchart. The generation server 4000 registers the IPC determined in step SA75 as part of the work history information in the work history database 4421. Therefore, the work history information includes the IPC. In this flowchart, steps SA72 to SA75 are an example of a collection unit that collects patent classification information related to the user's job based on information obtained from the user through dialogue. As shown in steps SA72 to SA75, the collection unit presents the user with multiple IPCs related to the user's job based on information obtained from the user through dialogue, and then determines the IPC corresponding to the user's selection as the user's work history information. As a result, the user's qualifications and skills are classified by IPC.

[0365] Next, we will describe some examples of resumes with reference to Figures 46-49. Figures 46-49 are diagrams showing examples of resumes that include IPC classifications. The resumes shown in Figures 46-49 include basic information, job duties, and IPCs related to the job duties. In particular, Figure 46 shows a resume format in which the IPC section is listed separately from the job duties section. In contrast, Figure 47 shows a resume format in which the corresponding IPCs are incorporated into the job duties description.

[0366] A format may be adopted in which an IPC (Instructional Programming Standard) section is listed separately from the job description section, while the corresponding IPC is incorporated into the job description. Figure 48 shows an example of a resume created using such a format.

[0367] If a patent application exists in which the user is credited as an inventor, a patent section may be included in the resume, as shown in Figure 49. The patent section may include the corresponding IPC (Institute of Patent Information) along with the patent information. By including the user's patent information along with the IPC in the resume in this way, the user can more strongly showcase their expertise.

[0368] In this way, IPCs corresponding to the user's skills can be included in the resume in various formats. The generation server 4000 stores information on job content, job experience, and employment period linked with the IPC as work history information in the work history database 4421. As shown in Figure 49, when the generation server 4000 obtains information on the user's patents, it stores the patent information linked with the IPC as work history information in the work history database 4421. Furthermore, the generation server 4000 stores the resumes exemplified in Figures 46 to 49 in the work history database 4421. The generation server 4000 may also allow the user to select the format of the resume through dialogue. Dialogue is a concept that includes expressions of intent (questions and answers, etc.) exchanged between the generation server 4000 and the user.

[0369] Here, patent classification is explained as an example of format information that includes the necessary information for a resume. The patent classification is not limited to IPC; F-terms, FI-terms, and CPC (Cooperative Patent Classification) may also be used. Patent classification is an example of classification information for classifying a user's professional competence. Similar to Embodiment 1, the classification information may be internal company classification information, or product classifications such as equipment may be applied. In other words, the classification information may be IPC, F-terms, book classification, or internal company classification.

[0370] [Other variations] Figure 50 shows an example of utilizing RAG (Retrieval-Augmented Generation) technology in the functional configuration of the generation server 4000. As already explained, RAG is a technology that configures a large-scale language model to access a knowledge source containing the latest and most accurate information, and then has the large-scale language model generate answers based on that knowledge source. As is well known, large-scale language models can sometimes output inaccurate or misleading information, which is called hallucination. RAG can complement such imperfections in large-scale language models and improve the quality of the answers they generate.

[0371] To improve the accuracy of selecting IPCs, an IPC database 429 containing IPCs and patent information may be provided on the generation server 4000, as shown in Figure 50. The large-scale language model 430 retrieves IPCs from the IPC database 429 and identifies candidate IPCs that are considered to correspond to the user's skills based on the retrieved IPCs.

[0372] More specifically, the IPC database 429 is composed of the related information database 422 shown in Figure 3. The IPC classification unit 4017 uses the method already explained with reference to Figure 3, etc., to determine classification information corresponding to the user's skills using the large-scale language model 430 and the data source of classification criteria stored in the IPC database 429.

[0373] This prevents the large-scale language model 430 from outputting responses that constitute hallucination. As a result, the accuracy and reliability of the large-scale language model 430 can be improved. In addition, system administrators can easily access the information sources from which the large-scale language model 430 derived its responses. This allows system administrators to easily determine the accuracy of the responses obtained from the large-scale language model 430.

[0374] In general, by adopting RAG technology, it is possible to create conversational AI that specializes in the desired function and can provide more precise answers. For example, by including company confidential information, it is possible to create conversational AI intended for use exclusively within the company. It is also possible to input a large amount of information tailored to the purpose into the large-scale language model 430. Furthermore, there is the advantage of being able to easily switch between different types of large-scale language models 430.

[0375] Generally, to obtain appropriate answers from generative AI, it is considered important to improve the quality of the prompts that users input when interacting with the AI. Prompt engineering is a technique known for improving the quality of such prompts. Prompt engineering is a technique for developing and optimizing prompts given to large-scale language models in order to use them efficiently. Prompt engineering is a method for optimizing the output for a specific task, based on the premise of leveraging existing models.

[0376] In contrast, there is the concept of fine-tuning. Fine-tuning is a technique that improves the performance of an existing model for a specific task by further training it. In fine-tuning, at least a portion of a pre-trained model, which was generated based on one dataset, is further trained on another dataset. This fine-tunes the parameters of the machine learning model for a specific task. Fine-tuning is sometimes interpreted as a type of transfer learning in a broad sense. However, the two differ in that fine-tuning is a technique that fine-tunes the weights of all layers of a pre-trained model, while transfer learning is a technique that fixes the weights of the pre-trained model and trains only using the added layers.

[0377] Fine-tuning has the problem of requiring enormous computational resources because it necessitates further training of large language models with numerous parameters. Prompt tuning solves this problem through a different approach than fine-tuning. In prompt tuning, the prompts themselves are the target of training. In other words, in prompt tuning, the parameters corresponding to the prompts are the target of optimization.

[0378] RLHF (Reinforcement Learning from Human Feedback) is a model learning method that combines "supervised learning," "reinforcement learning," and "inverse reinforcement learning." According to RLHF, it is possible to train AI on complex tasks such as natural language processing while minimizing elements that require human involvement, such as supervised learning. Therefore, it is possible that the large-scale language model 430 may be trained using such RLHF.

[0379] In Embodiment 2, a large-scale language model was given as an example of a natural language processing algorithm. However, algorithms that can be adopted as natural language processing algorithms are not limited to large-scale language models. For example, instead of a large-scale language model, an algorithm generated by a rule-based method such as pattern matching may be adopted.

[0380] The user device 500 (recruiter device 200 and applicant device 300) does not necessarily have to be equipped with all of the processor, memory, communication interface, and input / output interface shown in Figure 21, but may also be a thin client system using VDI (Virtual Desktop Infrastructure). A thin client system using VDI is a system that transfers and uses a desktop environment located on a server to a terminal in a remote location. The user device 500 (recruiter device 200 and applicant device 300), sharing server 100, and generation server 4000 do not necessarily have to be independent devices. When using such a thin client system, the functions of the user device 500, sharing server 100, and generation server 4000 can be provided on the same aggregation server.

[0381] Databases 120 and 4420 are not limited to relational databases; object-oriented databases, NoSQL databases, and other types of databases may also be used.

[0382] Sharing server 100 and generation server 4000 are examples of compute devices. Compute devices may also be configured using servers (on-premise servers, cloud servers, etc.) or serverless systems. Here, an on-premise server is a server installed and managed within facilities managed by the company itself. A cloud server is a server provided by another company via a network (a leased server). A serverless system is a system that allows the use of compute and memory functions only when needed, without being aware of the existence of a server. A compute device includes servers and serverless systems. Servers include on-premise servers and cloud servers.

[0383] The following are examples of components included in this disclosure. (a) The work history information collection device (generation server 4000, 4000A) comprises an acquisition unit (step SA11) that acquires format information containing the required information for a work history, a collection unit (step SA11, step SA22) that collects work history information related to the user's work history, and a storage unit (storage 4003) that stores a natural language processing algorithm. The collection unit collects work history information corresponding to the required information by interacting with the user using the natural language processing algorithm stored in the storage unit (step SA22).

[0384] (b) In addition to the work history information collection device described in (a) above, the creation unit (step SA53) further comprises a creation unit that creates a work history document using the work history information collected by the collection unit, and when the collection unit has completed collecting the work history information relating to the required information, the creation unit uses a natural language processing algorithm stored in the storage unit to create a work history document according to the format information (step SA53).

[0385] (c) In addition to the work history information collection device described in (b) above, the work history information collection device further includes a work history database (work history database 4421) in which work history information collected before interaction by the collection unit is registered, and the creation unit creates a work history document based on the work history information registered in the work history database if the work history information registered in the work history database includes information corresponding to the required entries (step SA13).

[0386] (d) In addition to the work history information collection device relating to (b) or (c) above, the work history information collection device further comprises an output unit configured to output the work history created by the creation unit to a user device (step SA110, step SA115: the generation server 4000 may also have the function of a sharing server 100), and the user device is a recruiter device (200) operated by a recruiter, or an applicant device (300) operated by an applicant, in a matching system that matches recruiters who are looking for contractors for work with applicants.

[0387] (e) In addition to the work history information collection device relating to any of (a) to (d) above, the device further comprises a dictionary database (dictionary database 4423), a registration unit (step SA34) that registers the meanings of terms included in the work history information in the dictionary database, and a display unit (step SA43) that displays the work history document on a display device (user device 500), wherein the display unit displays the meaning of the term on the display device (step SA44) if the term registered in the dictionary database is included in the work history document.

[0388] (f) In addition to the work history information collection device described in (e) above, the collection unit inquires with the user about the meaning of terms while interacting with the user (step SA32).

[0389] (g) In addition to the employment history information collection device relating to any of (a) to (f) above, the natural language processing algorithm includes a large-scale language model (large-scale language models 430, 430A, 430B).

[0390] (h) In addition to the employment history information collection device relating to any of (a) to (g) above, the collection unit collects patent classification information related to the user's job based on information obtained from the user through dialogue (step SA75), and the employment history information includes the said patent classification information.

[0391] (i) In addition to the work history information collection device described in (h) above, the collection unit presents the user with multiple patent classification information related to the user's job based on the information obtained from the user through dialogue, and then confirms the patent classification information according to the user's selection as work history information (steps SA72 to SA75).

[0392] (j) A method for creating a resume information collection involves having a computer perform the steps of obtaining format information containing the required information for a resume, and collecting resume information about the user's work history, wherein the collection step includes collecting resume information corresponding to the required information by interacting with the user using a natural language processing algorithm.

[0393] [Aspect] The following are the aspects of this disclosure.

[0394] (Article 1) The classification system described in Article 1 is a classification system for classifying information to be classified that is used by a user, comprising: an acquisition unit for acquiring information to be classified; a classification unit for classifying the information to be classified according to classification criteria; and a display device, wherein the classification unit includes a language processing model and a RAG data storage unit which stores data sources for causing the language processing model to perform language processing based on RAG (Retrieval-Augmented Generation), the RAG data storage unit includes a classification criteria data storage unit which stores data sources relating to classification criteria in a format that conforms to RAG, the data sources relating to classification criteria include a plurality of classification information defined by the classification criteria, and the classification unit uses the language processing model and the data sources relating to classification criteria to determine classification information corresponding to the information to be classified, and displays the information to be classified on the display device in association with the classification information.

[0395] (Section 2) The classification system described in Section 2, in addition to the classification system described in Section 1, includes a RAG data storage unit in which data sources relating to the information to be classified are stored in a format conforming to RAG, and the classification unit determines classification information corresponding to the information to be classified using a language processing model, data sources relating to classification criteria, and data sources relating to the information to be classified.

[0396] (Clause 3) In addition to the classification system described in paragraph 1 or 2, the classification system functions as a matching system for matching users with business opportunities, and the information to be classified includes one of the following: business opportunity information, user profile information, educational information for educating users, or advertising information provided to users.

[0397] (Article 4) The classification system described in Article 4, in addition to the classification system described in Article 3, includes user work capability information in the profile information.

[0398] (Clause 5) The classification system described in paragraph 5, in addition to the classification system described in paragraph 3 or 4, further comprises a business case storage unit for storing business case information, a profile storage unit for storing profile information, a user device operated by a user, a search unit for searching business case information stored in the business case storage unit using classification information associated with the profile information, and a transmission unit for transmitting the business case information found by the search unit to the user device.

[0399] (Clause 6) The classification system described in paragraph 6, in addition to the classification system described in paragraph 3 or 4, further comprises an advertising storage unit for storing advertising information, a profile storage unit for storing profile information, a user device operated by a user, a search unit for searching for advertising information stored in the advertising storage unit using classification information associated with the profile information, and a transmission unit for transmitting the advertising information found by the search unit to the user device.

[0400] (Clause 7) The classification system described in paragraph 7, in addition to the classification system described in claim 3 or claim 4, further comprises: an education storage unit for storing education information; a profile storage unit for storing profile information; a user device operated by a user; a search unit for searching for education information stored in the education storage unit using classification information associated with the profile information; and a transmission unit for transmitting the education information found by the search unit to the user device.

[0401] (Clause 8) In addition to the classification system described in Clause 7, the classification system described in Clause 8 allows the search unit to search for educational information stored in the educational storage unit that is associated with classification information different from the classification information associated with the user's profile information.

[0402] (Clause 9) In addition to the classification system described in any one of Clauses 1 to 8, the classification unit generates summary information of the information to be classified, determines the classification information corresponding to the information to be classified based on the generated summary information, and generates the summary information while interacting with the user.

[0403] (Clause 10) In addition to the classification systems described in any one of Clauses 1 to 9, the classification unit determines classification information corresponding to the information to be classified while interacting with the user.

[0404] (Clause 11) The classification system described in paragraph 11 is, in addition to the classification system described in any one of paragraphs 1 to 10, the classification information is one of the following: IPC, F-term, CPC, FI, book classification, or internal classification.

[0405] (Section 12) The method described in Section 12 is a method for classifying information to be classified for use by a user, the method comprising: causing a computer to perform the steps of acquiring information to be classified and classifying the information to be classified according to classification criteria, wherein the computer is configured to access a RAG (Retrieval-Augmented Generation) data storage unit, the RAG data storage unit includes a classification criteria data storage unit in which a data source relating to classification criteria is stored in a format conforming to RAG, the data source relating to classification criteria includes a plurality of classification information defined by the classification criteria, and the classifying step comprises the steps of determining classification information corresponding to the information to be classified using a language processing model and a data source relating to classification criteria, and displaying the information to be classified on a display device in association with the classification information.

[0406] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the present invention is indicated by the claims rather than by the description of the embodiments above, and all modifications within the meaning and scope of the claims are intended to be included. [Explanation of symbols]

[0407] 110000 Classification system, 50 Internet, 100 Sharing server, 101 Processor, 102 Memory, 103 Storage, 104 Communication interface, 120 Database, 121 Corporate database, 122 User database, 123 Project database, 124 Matching database, 125 Advertising database, 126 Education database, 127 Community database, 200 Recruiter device, 200A~200C Recruiter device, 201 Processor, 202 Memory, 203 Communication interface, 204 Input / Output interface, 205 Display, 206 Control unit, 300 Applicant device, 300A~300C Applicant device, 301 Processor, 302 Memory, 303 Communication interface, 304 Input / Output interface, 305 Display, 306 Control unit, 400 Server, 401 Processor, 402 Memory, 403 Communication interface, 404 Storage, 410 Microcomputer, 420 RAG database, 421 Related information database, 422 Related information database, 423 Related information database, 424 Related information database, 425 Related information database, 426 Satisfaction requirements database, 427 Original information storage database, 428 Generated format database, 429 IPC database, 430, 430A, 430B Large-scale language model (LLM), 450 AI, 500, 500A User device, 501 Processor, 502 Memory, 503 Communication interface, 504 Input / Output interface, 505 Display, 506 Operation unit, 550 Screen, 601-603 Buttons, 800A Profile classification unit, 800B Case classification unit, 803A, 803B Chat unit (summary), 804A, 804B Chat unit (classification), 805A, 805B Classification storage unit, 902 AI construction unit, 903 Data import unit, 904 Acquisition unit, 905 Matching unit, 906 Result display unit, 1550 Screen, 1551 Question frame, 1552 Answer frame, 4000, 4000A Generation server, 4001 Processor, 4002 Memory, 4003 Storage, 4004 Communication interface, 4010,4010A Control Unit, 4011 Data Import Unit, 4012 Information Acquisition Unit, 4013 Dictionary Information Registration Unit, 4014 Dictionary Information Display Unit, 4015 Information Organization Unit, 4016 Resume Registration Unit, 4017 IPC Classification Unit, 4021 Input Unit, 4022 Content Sufficiency Determination Unit, 4023 Input Request Unit, 4024 Original Information Storage Unit, 4025 Generation Unit, 4026 Confirmation Request Unit, 4027 Output Unit, 4050 Program, 4051 Dialogue Program, 4052 Confirmation Program, 4053 Creation Program, 4110 Dialogue AI, 4120 Confirmation AI, 4130 Creation AI, 4420 Database, 4421 Resume Database, 4422 Dialogue Information Database, 4423 Dictionary Database.

Claims

1. A classification system that classifies information to be classified by users, An acquisition unit that acquires the classification target information, A classification unit that classifies the aforementioned information to be classified according to classification criteria, Equipped with a display device, The aforementioned classification unit is Language processing models and, It includes a RAG data storage unit which stores data sources for causing the language processing model to perform language processing based on RAG (Retrieval-Augmented Generation), The RAG data storage unit includes a classification criteria data storage unit in which data sources relating to the classification criteria are stored in a format conforming to the RAG. The data source relating to the classification criteria includes multiple classification pieces of information defined by the classification criteria, The aforementioned classification unit is Using the language processing model and the data source relating to the classification criteria, the classification information corresponding to the information to be classified is determined. A classification system that displays the classification target information on the display device in association with the classification information.

2. The RAG data storage unit includes a classification target data storage unit in which data sources relating to the classification target information are stored in a format conforming to the RAG, The classification system according to claim 1, wherein the classification unit determines classification information corresponding to the information to be classified using the language processing model, the data source relating to the classification criteria, and the data source relating to the information to be classified.

3. The aforementioned classification system functions as a matching system that matches users with business cases. The classification system according to claim 1 or claim 2, wherein the classification target information includes one of the following: business case information, user profile information, educational information for educating the user, and advertising information provided to the user.

4. The classification system according to claim 3, wherein the profile information includes the user's work capability information.

5. The aforementioned classification information includes the aforementioned business case information and the aforementioned profile information, The aforementioned classification system is A business case storage unit for storing the aforementioned business case information, A profile storage unit for storing the aforementioned profile information, A user device operated by the aforementioned user, A search unit that searches for business case information stored in the business case storage unit using classification information associated with the profile information, The classification system according to claim 4, further comprising a transmission unit that transmits business case information found by the search unit to the user device.

6. The information subject to classification includes the advertising information and the profile information, The aforementioned classification system is An advertising storage unit for storing the aforementioned advertising information, A profile storage unit for storing the aforementioned profile information, A user device operated by the aforementioned user, A search unit that searches for advertising information stored in the advertising storage unit using classification information associated with the profile information, The classification system according to claim 4, further comprising a transmission unit that transmits advertising information found by the search unit to the user device.

7. The classification information includes the educational information and the profile information, The aforementioned classification system is The educational storage unit for storing the aforementioned educational information, A profile storage unit for storing the aforementioned profile information, A user device operated by the aforementioned user, A search unit that searches for educational information stored in the educational storage unit using classification information associated with the profile information, The classification system according to claim 4, further comprising a transmission unit that transmits educational information found by the search unit to the user device.

8. The classification system according to claim 7, wherein the search unit searches for educational information stored in the educational storage unit that is associated with classification information different from the classification information associated with the user's profile information.

9. The classification unit generates summary information of the information to be classified, and determines the classification information corresponding to the information to be classified based on the generated summary information. The classification system according to claim 1 or 2, wherein the classification unit generates the summary information while interacting with the user.

10. The classification system according to claim 1 or 2, wherein the classification unit determines classification information corresponding to the information to be classified while interacting with the user.

11. The classification system according to claim 1 or claim 2, wherein the classification information is one of IPC, F-term, CPC, FI, book classification, and in-house classification.

12. A method for classifying information used by users, The above method involves a computer, The steps include obtaining the classification target information, The process involves executing the step of classifying the information to be classified according to the classification criteria. The computer is configured to access the RAG (Retrieval-Augmented Generation) data storage unit. The RAG data storage unit includes a classification criteria data storage unit in which data sources relating to the classification criteria are stored in a format conforming to the RAG. The data source relating to the classification criteria includes multiple classification pieces of information defined by the classification criteria, The aforementioned classification step is, A step of determining classification information corresponding to the information to be classified using a language processing model and a data source relating to the classification criteria, A method comprising the steps of displaying the classification target information on a display device in association with the classification information.

Images