kossi-26/Refractory_Machine-Learning

ML models to forecast and predict based on data # Refactory Data Science and Advanced Machine Learning Program - Final Project ## Project Title **Student Dropout Prediction Using Machine Learning** ## Overview This project builds a predictive model to identify students at risk of dropping out using academic, demographic, and socioeconomic data. The goal is to help educational institutions detect at-risk students early and take proactive measures to improve retention. ## Problem Statement Student dropout is a critical issue in education systems, especially in developing regions. It leads to wasted resources, lost potential, and long-term socioeconomic consequences. Our goal is to build a predictive model that flags students likely to drop out, enabling timely support. ## Dataset Source: UCI Student Performance Dataset Size: 4,424 students, 37 original features Enhancements: Location_Type: Urban vs. Rural Internet_Access: Connectivity score Teacher_Quality: Qualification score School_Distance_km: Distance to school These features were simulated to reflect Tanzanian educational realities. ## Key Features - Data preprocessing and cleaning - Exploratory data analysis (EDA) - Model training and evaluation for binary classification - Prediction of student dropout probability based on key factors