Rahul Zaru

I’m a computer engineering student focused on building practical machine learning pipelines in Python, especially for prediction and classification problems.

I’ve developed a Credit Card Fraud Detection System on highly imbalanced financial data, using preprocessing and class-imbalance handling (like SMOTE/undersampling). I trained and compared Logistic Regression, Random Forest, and XGBoost, evaluating with Precision, Recall, F1-score, and ROC-AUC to prioritize fraud detection accuracy.

I also built a Heart Disease Prediction System using structured patient data, handling missing values, encoding categorical features, and applying feature scaling. Using scikit-learn, I compared Logistic Regression, Random Forest, SVM, and KNN with accuracy, precision, recall, F1-score, and confusion matrices.