An Intelligent Framework for Student Performance Analysis and Prediction Using Machine Learning and Educational Data Mining

Abstract

This paper presents an intelligent framework for student performance analysis and prediction using Machine Learning and Educational Data Mining. The proposed system analyzes student academic records, including attendance, marks, grades, and SGPA, to predict academic performance. A Random Forest Classifier is used for pass/fail prediction, while Linear Regression estimates future SGPA. Developed using Python, Streamlit, MongoDB, Pandas, and Scikit-learn, the system provides interactive dashboards and visualizations to support educators in identifying at-risk students and making data-driven academic decisions.