Abstract
This article proposes two innovative approaches, the one-channel learning image recognition and the three-channel learning image recognition, to convert student's course involvements into images for early warning predictive analysis. Multiple experiments with 5235 students and 576 absolute/1728 relative input variables were conducted to verify their effectiveness. The results indicate that both methods can significantly capture more at-risk students (the highest average recall rate is equal to 77.26%) than the following machine learning algorithms-support vector machine, random forest, and deep neural network-in the middle of the semester. In addition, the innovative approaches allow minor subtypes of at-risk student identification and provide visual insights for personalized interventions. Implications and future directions are also discussed in this article.
Original language | English |
---|---|
Article number | 9072352 |
Pages (from-to) | 617-630 |
Number of pages | 14 |
Journal | IEEE Transactions on Learning Technologies |
Volume | 13 |
Issue number | 3 |
DOIs | |
State | Published - 1 Jul 2020 |
Keywords
- At-risk
- convolutional neural networks (CNNs)
- distance learning
- early warning
- image recognition
- machine learning
- prediction