Publication Details

Abstract

Student performance prediction has become an important task in educational data mining. This research proposes a machine learning-based system to predict student academic performance using factors such as attendance rate, study hours, parental education level, and extracurricular participation. A synthetic dataset of 5000 student records was generated and analyzed using visualization techniques. A Linear Regression model was trained and evaluated using MAE, RMSE, and R² metrics. The system also includes an AI-based dashboard that provides predictions, analytics, and personalized study recommendations. The experimental results demonstrate that the model effectively predicts student outcomes and can assist educators in identifying at-risk students.