STUDENTS PERFORMANCE PREDICTION USING DATA MINING TECHNIQUES

Abstract

Students Performance prediction using data mining techniques has become an essential approach for analyzing large-scale datasets and extracting meaningful insights for decision-making. This study presents a Student Performance Prediction using Data Mining Techniques (SPPDMT) model that integrates feature selection and classification algorithms to improve prediction accuracy. The proposed system utilizes Dynamic Feature Ensemble Evolution (DE-FS) to identify the most relevant features by combining multiple statistical methods such as correlation analysis, information gain, and chi-square techniques. These selected features are then used to train classification models such as Decision Tree (DT) and Support Vector Machine (SVM). The hybrid SPPDMT model (DE-FS) enhances prediction performance by reducing dimensionality and improving model generalization. Experimental results demonstrate that the proposed method outperforms traditional models in terms of accuracy, precision, recall, and F1-score. The system provides efficient prediction results and supports better decision-making in performance evaluation systems.