Evaluating the Impact of Gradient Boosting and Random Forest on Accuracy Improvement in Cardiovascular Risk Prediction

The aim of the study is to compare the performance of Random Forest and Gradient Boosting for enhancing cardiovascular risk prediction accuracy for early medical intervention. This study has two groups. Group 1 employs a Random Forest model trained on Framingham Heart Study and UCI Heart Disease data with features like age, gender, and blood pressure. Group 2 employs a Gradient Boosting model, validated against similar features, with performance indicators of accuracy, precision, recall and f1-score. Random Forest proved to be highly accurate with 84.5% accuracy at the 20th epoch, contrasted to Gradient Boosting that reached 88.0%. The models indicated improvement in prediction accuracy across 20 epochs. The Gradient Boosting model did better in accuracy (88.0%), precision (87.7%), recall 87.9% and f1-score at (87.8%) metrics than Random Forest. The Gradient Boosting mode did better across the board in cardiovascular disease prediction and is therefore a more reliable choice for detection. Gradient Boosting did better than Random Forest in accuracy (88.0%), precision (87.7%), and recall at 87.9% and F1-Score (87.8%) and is therefore more reliable for CVD prediction. Random Forest ranked lower on recall and accuracy and may be prone to missing cases of CVD. The study highlights the importance of choosing the right model in the detection early on.