STUDENT PERFORMANCE PREDICTION SYSTEM

Abstract

The Student Performance Prediction System is a web-based intelligent application developed using Django and Machine Learning techniques to predict academic outcomes effectively. Student performance is influenced by multiple academic and behavioral factors such as attendance, internal marks, study hours, assignment completion, and sleep patterns. The system collects these inputs through a user-friendly interface and processes them using a trained machine learning model. The Random Forest Classification algorithm is used due to its ability to handle complex datasets and improve prediction accuracy by combining multiple decision trees. Data preprocessing techniques such as normalization, encoding, and handling missing values ensure that input data is suitable for model prediction. Based on the processed data, the system classifies students into three categories: Excellent, Average, and Needs Improvement. The results are displayed instantly, allowing educators to identify weak students early and take corrective measures. The system reduces manual effort, minimizes errors, and supports datadriven decision-making in education. Additionally, the application is scalable, efficient, and easy to use, making it suitable for schools and colleges. By integrating machine learning with web technologies, the system enhances academic monitoring and improves overall student outcomes.