Problem & Context
Build a machine learning model to classify heart disease risk from structured data.
Goal
Explore supervised machine learning approaches for classification and evaluate model performance using standard classification metrics.
Approach
Performed data preprocessing and feature engineering, addressed class imbalance using SMOTE, trained Logistic Regression and Support Vector Machine models, and evaluated performance using accuracy, confusion matrix, and classification report.
Technical Implementation
Used Pandas for preprocessing and feature manipulation, applied feature engineering before training, used SMOTE for class imbalance, trained Logistic Regression and SVM classifiers, and compared performance using classification metrics. The SVM model achieved up to 88.78% accuracy in this evaluation.
Outcome
Built and evaluated classification models for heart disease risk prediction, with SVM reaching 88.78% accuracy in the project evaluation.
Important Note
This project should not be presented as a clinically validated medical prediction system.
Lessons Learned
Built practical, hands-on experience with a complete supervised learning workflow — from handling class imbalance to comparing classifier performance using proper evaluation metrics.