| Popis: |
Academic procrastination is common among college students, which has seriously affected their academic and physical and mental health. The analysis of online learning behavior log data can help to diagnose students' procrastination tendency. Aiming at the phenomenon of academic procrastination of college students in the process of online learning, this paper firstly extracts five characteristic variables that represent the degree of task procrastination, and divides them into two categories: the degree of task procrastination in class and the degree of task procrastination after class according to different scenarios, students' behavior patterns may be different. Then, multi-view clustering method is used to divide students into three types of procrastinators. Finally, clustering tags are extracted to analyze the characteristics of procrastinators by combining five feature variables and learning performance. The experimental results show that the clustering performance of multi-view clustering algorithm is better than that of K-means algorithm; students' learning behavior patterns are different when facing in-class tasks and after-class tasks. Compared with after-class tasks, the procrastination degree of students in the face of in-class tasks is smaller; the procrastination degree of students’ after-class tasks is more consistent. |
