JP-7855186-B1 - Information processing system, information processing method, and information processing program

JP7855186B1JP 7855186 B1JP7855186 B1JP 7855186B1JP-7855186-B1

Abstract

[Problem] To provide an information processing system, information processing method, and information processing program that can efficiently provide diverse and high-quality self-promotion data to job seekers. [Solution] In an information processing system in which multiple evaluated self-PR data are stored in a predetermined storage unit, the PR data are evaluated based on specificity, fluency, cultural relevance, and objectivity from different evaluation perspectives, for multiple self-PR data that have been pre-generated by combining user-selectable conditions and diversification variables. The specific method involves receiving a request for conditions that combine at least job type, industry, and strengths as selectable conditions, input by the user through a selection operation, extracting the self-PR data that matches the conditions from the multiple evaluated self-PR data in the storage unit based on the received conditions, and displaying the extracted self-PR data on a predetermined terminal corresponding to the user. [Selection Diagram] Figure 5

Inventors

安藤将晃
権瓶匠
曽我部亮
堤大貴

Assignees

株式会社マイナビ
株式会社ELYZA

Dates

Publication Date: 20260508
Application Date: 20250627

Claims (6)

Equipped with a processor, Multiple evaluated self-PR data are stored in a designated storage unit. The aforementioned evaluated multiple self-PR data are generated in advance by combining user-selectable conditions and diversification variables, and the content generated as the self-PR data is evaluated based at least on specificity, fluency, cultural relevance, and objectivity from different evaluation perspectives. The aforementioned processor, As for the selectable conditions, the system accepts requests that combine at least the job title, industry, and strengths entered by the user through a selection process. Based on the received conditions, the storage unit extracts the self-PR data that matches the conditions from among the evaluated plurality of self-PR data. The extracted self-PR data is displayed on a predetermined terminal corresponding to the user. Information processing system.
The aforementioned processor, As a process for assigning evaluations to the aforementioned multiple self-PR data, The evaluation method involves a combination of evaluation using a generative model based on predefined evaluation axes, evaluation using language analysis, and evaluation of structure based on format. The information processing system according to claim 1.
The aforementioned processor, In the evaluation using the generative model, the evaluation of fluency on the evaluation axis includes evaluation of linguistic usage, structural correctness, and contextual consistency. The information processing system according to claim 2.
The aforementioned processor, As part of the extraction process, the evaluated self-PR data that matches the conditions is randomly extracted from the storage unit. If the user determines that the self-PR data displayed after the extraction is unsuitable, the system accepts a re-request from the user under the same conditions, and then extracts and displays different self-PR data under the same conditions through random extraction. The information processing system according to claim 1.
Multiple evaluated self-PR data are stored in a designated storage unit. The aforementioned evaluated multiple self-PR data are generated in advance by combining user-selectable conditions and diversification variables, and the content generated as the self-PR data is evaluated based at least on specificity, fluency, cultural relevance, and objectivity from different evaluation perspectives. Computers As for the selectable conditions, the system accepts requests that combine at least the job title, industry, and strengths entered by the user through a selection process. Based on the accepted conditions, the evaluated self-PR data is extracted from the storage unit. The extracted self-PR data is displayed on a predetermined terminal corresponding to the user. An information processing method that performs a process.
Multiple evaluated self-PR data are stored in a designated storage unit. The aforementioned evaluated multiple self-PR data are generated in advance by combining user-selectable conditions and diversification variables, and the content generated as the self-PR data is evaluated based at least on specificity, fluency, cultural relevance, and objectivity from different evaluation perspectives. On the computer, As for the selectable conditions, the system accepts requests that combine at least the job title, industry, and strengths entered by the user through a selection process. Based on the accepted conditions, the evaluated self-PR data is extracted from the storage unit. The extracted self-PR data is displayed on a predetermined terminal corresponding to the user. An information processing program used to execute a process.

Description

This disclosure relates to an information processing system, an information processing method, and an information processing program. Traditionally, there are techniques for creating job postings. For example, there is a technology that uses natural language models to obtain information useful for recruiting job seekers (see Patent Document 1). Japanese Patent Publication No. 2024-161943 Figure 1 is a block diagram showing the functional configuration of the information processing system according to this embodiment.Figure 2 is a block diagram showing the hardware configuration of the information processing system.Figure 3 is a schematic diagram showing the flow of the process for providing self-PR data.Figure 4 is a flowchart showing the flow of pre-processing by the information processing system.Figure 5 is a flowchart showing the flow of presentation processing by the information processing system.Figure 6 shows an example of a Web UI screen, specifically a job selection screen.Figure 7 shows an example of a Web UI screen, specifically the industry selection screen.Figure 8 shows an example of a Web UI screen, specifically the chat window screen and the strengths selection screen.Figure 9 shows an example of a Web UI screen, specifically the chat window screen and the screen displaying the extraction results. The following describes an example of an embodiment of the disclosed technology with reference to the drawings. In each drawing, identical or equivalent components and parts are given the same reference numerals. Furthermore, the dimensional ratios in the drawings are exaggerated for illustrative purposes and may differ from actual ratios. First, an overview of the embodiments of this disclosure will be provided. The system provided in this embodiment relates to a system that provides sample self-introduction texts (hereinafter referred to as "samples") in job placement support. The challenges in providing self-introduction texts will be explained. For example, while various job search websites sometimes provide sample self-introductions as part of their job placement support content, these samples were manually created by the content providers. Furthermore, site users (users) created their self-introductions for companies they wished to apply to, using the samples as a reference. However, the current provision of self-introduction samples faced the following major challenges: (1) Manual creation of content for job search websites was inefficient. It was impossible to create a large number of self-introduction samples manually, and the samples tended to be limited to common job types and industries. (2) It was difficult to find self-introduction samples that matched one's own job type and industry. As a result of (1), users were unable to find suitable samples and consequently had to come up with their own. (3) There were concerns about usability and the inability of people unfamiliar with AI to use it. While the development and spread of AI has made it relatively easy for users to create self-introductions using conversational AI tools such as ChatGPT, the number of users who can utilize them is limited. Furthermore, even when using existing AI tools, hallucination can occur depending on the input content. Also, using existing AI tools can result in long processing times for output, potentially impairing the user experience. Therefore, in this embodiment, the information processing system retrieves and provides pre-generated self-introduction (Self-PR) samples. This ensures a seamless job-seeking experience for the user, eliminating any perceived time lag caused by AI generation. Furthermore, by evaluating the generated Self-PR samples using an evaluation AI based on human assessment, and providing evaluated Self-PR samples, hallucination is prevented, improving the quality of the Self-PRs provided to the user. As described below, the Self-PR samples provided in this embodiment are complex data in which evaluation information is attached as metadata to the Self-PR samples. When the system references the data, it is extracted based on the evaluation information, and the Self-PR samples are provided to the user upon delivery. Therefore, to distinguish them from the existing Self-PR samples mentioned above, they are referred to as evaluated Self-PR data. (composition) Figure 1 is a block diagram showing the functional configuration of the information processing system 100 in this embodiment. The information processing system 100 is connected via a network N to a management terminal 140 that can be operated by an administrator and a user terminal 150 that can be operated by a job seeker user. The management terminal 140 accepts operations related to setting the generation model and prompts and pre-evaluation of self-PR data. The user terminal 150 accepts various operations related to requesting and acquiring self-PR data. Figure 2 is a block diagram showing the hardware configuration of the information