Analyzing High School Students' Online Learning Behavior Through Data Mining: Personalized Pathway Construction and Outcome Prediction

The objective of this research is to use data mining and machine learning methods to track the behavior of high school students through online education platforms, so that we can learn trends, create personalized pathways, and forecast academic performance. We took engagement metrics including sessions, videos watched, and assignment completion rates from 1,000 high school students. We used a k-means clustering algorithm to break students into high, moderate and low engagement groups with different learning habits and academic performance patterns. Customised learning modules were then crafted for each cohort, tailoring content to maximise participation and facilitate learning. They also employed a decision tree model to predict performance which was 82% accurate with session length and assignment completion rates being among the best predictors. The results show that targeted interventions, rooted in engagement-based data, have the potential to positively impact learning experiences on digital platforms for high school students.