2022 Fiscal Year Final Research Report

Research on a hybrid learning data analysis method that takes into account pen-input data and the solution process

Research Project

PDF

Project/Area Number	19K21758
Research Category	Grant-in-Aid for Challenging Research (Exploratory)
Allocation Type	Multi-year Fund
Review Section	Medium-sized Section 9:Education and related fields
Research Institution	Nagoya University
Principal Investigator	Nakamura Yasuyuki 名古屋大学, 情報学研究科, 准教授 (70273208)
Co-Investigator(Kenkyū-buntansha)	金子真隆東邦大学, 薬学部, 教授 (90311000) 高遠節夫東邦大学, 理学部, 訪問教授 (30163223)
Project Period (FY)	2019-06-28 – 2023-03-31
Keywords	数式自動採点システム / 手書きノートデータ分析
Outline of Final Research Achievements	The objective of this project was to establish a hybrid learning data analysis method that combines "answer results" obtained on an LMS and pen input data as a manifestation of thinking processes such as calculations. We took a math problem and collected logs of writing event times, erasure events, etc. from handwritten notes that describe the thinking process, which were input to a tablet, as well as the answers. The results were used to estimate the level of confidence in the answers to the questions using machine learning. In addition, we examined the relationship between the difficulty level of problems inferred from the writing data and the difficulty level of problems estimated by item response theory.
Free Research Field	教育工学
Academic Significance and Societal Importance of the Research Achievements	オンラインテストの「解答」情報だけでなく、学習者がどのような誤答、準正答を経て正答に至ったかという解答過程、および、どのような計算過程、思考様式（筆記速度、書き直しなど）に基づき解答を得たかというペン入力データは相互に密接に関連しているはずである。両データを考慮したハイブリッド型の学習データ解析手法は確立されていなかったが、本研究課題により、機械学習を利用した、ペン入力データと解答過程データを連携させて分析する手法を提案することができ、特に手書きノートデータを活用した、今後の学習データ分析の方向性に一つの可能性を示すことができた。