Predicting Student Exam Success with Machine Learning

Created using ChatSlide

This data mining project focuses on classifying student exam performance by utilising a dataset of 6,607 records with 19 predictors. Employing techniques such as PCA, SMOTE, and various models including KNN, Naive Bayes, and Decision Tree, the project aims to provide insights for educators to predict academic results effectively. With a focus on data quality, missing values were imputed and feature engineering was conducted. Random Forest emerged as the top-performing model, highlighting...