Information processing methods, programs, and information processing systems
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- 株式会社ケンシン
- Filing Date
- 2024-11-30
- Publication Date
- 2026-06-11
AI Technical Summary
【0009】 本発明者は、上述した課題を解決するため、鋭意検討を行った。その結果、受験者が受験者の端末からサーバに接続して、論述方式の設問に対する回答を添削する情報処理方法であって、サーバの制御部は、受験者の端末又は管理者の端末から設問を、受験者の端末から設問に対する受験者の回答を、それぞれ入力情報として受け付け、入力情報をサーバの記憶部に記憶し、記憶部から取得した設問及び受験者の回答、並びに回答の添削条件を含む添削プロンプトデータとして生成し、添削プロンプトデータの入力に応じて、機械学習アルゴリズムによる学習モデルが生成した添削結果を受験者の端末に提示し、添削条件は、少なくとも、設問に対する回答になっているか、かつ論理的な回答になっているかを確認し、文章でアドバイスする条件である情報処理方法によれば、記述式試験の回答の仕方に慣れていない受験生の回答を適切に添削することができることを見出し、本開示に示される技術を完成するに至った。具体的に、本開示は以下のものを提供する。
Smart Images
Figure 2026095446000001_ABST
Abstract
Claims
1. An information processing method in which an examinee connects to a server from the examinee's terminal and corrects their answers to essay-style questions, The control unit of the server is: The server receives the questions from the examinee's terminal or the administrator's terminal, and the examinee's answers to the questions from the examinee's terminal, as input information, and stores the input information in the server's storage unit. The questions and the examinee's answers obtained from the memory unit are generated as correction prompt data including the conditions for correcting the answers. In response to the input of the aforementioned correction prompt data, the correction results generated by the machine learning algorithm's learning model are presented to the examinee's terminal. The aforementioned editing conditions are that the editor will at least check whether the answer addresses the question and is logical, and will provide written advice. Information processing methods.
2. The control unit generates learning prompt data that includes learning instructions for at least a portion of the correction results, The learning model performs learning in response to the input of the learning prompt data. The information processing method according to claim 1.
3. The aforementioned examination is a qualification examination related to the construction industry. The aforementioned editing conditions further include instructions that the truthfulness of the construction examples and / or the validity of the numerical values will not be considered. The information processing method according to claim 1 or 2.
4. A program that allows an examinee to connect to a server from their terminal and execute an information processing method for correcting their answers to essay-style questions, The control unit of the server is: The server receives the questions from the examinee's terminal or the administrator's terminal, and the examinee's answers to the questions from the examinee's terminal, as input information, and stores the input information in the server's storage unit. The questions and the examinee's answers obtained from the memory unit are generated as correction prompt data including the conditions for correcting the answers. In response to the input of the aforementioned correction prompt data, the correction results generated by the machine learning algorithm's learning model are presented to the examinee's terminal. The aforementioned editing conditions are that the editor will at least check whether the answer addresses the question and is logical, and will provide written advice. program.
5. An information processing system in which an examinee connects to a server from the examinee's terminal and corrects their answers to essay-style questions, The control unit of the server is: The server receives the questions from the examinee's terminal or the administrator's terminal, and the examinee's answers to the questions from the examinee's terminal, as input information, and stores the input information in the server's storage unit. The questions and the examinee's answers obtained from the memory unit are generated as correction prompt data including the conditions for correcting the answers. In response to the input of the aforementioned correction prompt data, the correction results generated by the machine learning algorithm's learning model are presented to the examinee's terminal. The aforementioned editing conditions are that the editor will at least check whether the answer addresses the question and is logical, and will provide written advice. Information processing system.