STUDENT PERFORMANCE PREDICTION

Abstract

Student performance prediction is a challenging task due to the complex, non-linear, and dynamic nature of educational data. This project focuses on the application of machine learning techniques, particularly Linear Regression with time-series features, to predict student academic performance. By utilizing historical data such as attendance, study hours, previous scores, and assignment performance, the system identifies patterns that influence academic outcomes. Lag features are engineered to capture temporal dependencies and improve prediction accuracy. The developed model achieves a high R² score, indicating strong predictive capability. Interactive visualizations are used to present trends and compare actual versus predicted performance, making the results easy to interpret. Additionally, the system can predict future performance based on user input, demonstrating its practical usefulness. Overall, this project highlights how simple yet effective machine learning models can be used for accurate and efficient student performance forecasting, supporting datadriven decision-making in education