Study on Relation between Student Learning Situation and Lecture Evaluation Using Student Comments Collected After Every Lessons

2016 Fiscal Year Final Research Report

• Project/Area Number: 26540183
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Learning support system
• Research Institution: Kyushu University
• Principal Investigator: MINE Tsunenori 九州大学, システム情報科学研究院, 准教授 (30243851)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: テキストマイニング / ラーニングアナリティクス / 機械学習 / コメントマイニング / 授業後報告文 / 成績推定 / 学習状況推定

Outline of Final Research Achievements

We analyzed student free-style comments written by themselves just after every lesson and extract words related to student learning characteristics, learning performance and/or class assessment. We set five learning attributes:
effort, attitude, understanding, collaboration, findings and proposed novel methods to estimate student learning performance using the learning attributes. The methods use several machine learning methods such as SVM, Artificial Neural Network, Random Forest, LDA, pLSA, LSA, Word2Vec, Multi-instance Learning. We found there were high correlations between combination of the attributes and student grades. We also showed the effect of teacher interventions for improving the quality of student comments, understanding student learning situations and estimating student learning performance.

Free Research Field

情報知能工学