Analysis of Blended Learning Behaviour Based on K-Means Clustering Algorithm

• Published in: 2024 6th International Conference on Computer Science and Technologies in Education (CSTE) pp. 315 - 318
• Main Authors: Sun, Yue, Xiao, Wen
• Format: Conference Proceeding
• Language: English
• Published: IEEE 19.04.2024
• Subjects: Analytical models; blended learning; Clustering algorithms; clustering analysis; Correlation; correlation analysis; Education; Federated learning; learning behavior analysis; Measurement; Solids
• DOI: 10.1109/CSTE62025.2024.00065

As blended learning models gain popularity, analyzing the online learning behaviors of college students has emerged as a crucial approach to enhancing teaching effectiveness. The objective of this study is to conduct a comprehensive analysis of the learning behavior data from 44 students specializing in computer-related subjects on the online platform of the "Introduction to Computational Thinking" course, utilizing clustering algorithms. Subsequently, we utilized the Pearson correlation coefficient method to delve into the intricate relationship between diverse behavioral traits and academic performance. Our research has revealed that cluster analysis is adept at discerning behavioral disparities among students. Specifically, we identified four types of learning groups with different learning characteristics. Furthermore, a notable correlation was observed between homework scores and key evaluation metrics, including academic performance. Based on these findings, this study proposes targeted learning strategies and teaching suggestions, aiming to help educators better guide students in learning and improve the effectiveness of blended learning.