Application of Data Mining Techniques for Predicting Students' Academic Performance in Higher Education

The present chapter lists a specific literature review from 2010 till 2025 about Educational Data Mining to determine steps in developing effective predictive models for identifying low-performing students, the parameters that influence academic achievements, various data mining techniques that can efficiently predict students' academic outcomes along with the advanced methods that may improve the prediction results. Finally, it may be concluded that students' academic and background features are the most influential features. Further, among the data mining techniques that could be used for predicting students' performance classification (especially Decision Tree) was a widely used data mining technique due to its simplicity, interpretability, and capability to provide significant results consistently. The present analysis also reveals that the use of feature selection, data balancing, hyper-parameter tuning, and ensemble models may further improve the prediction results.