MEDICAL INFORMATICS IV
IMPROVING CARDIOVASCULAR DISEASE PREDICTION USING ENSEMBLE LEARNING TECHNIQUES AND DIMENSIONALITY REDUCTION
Yayıncı:
İstanbul Üniversitesi Yayınları
Cardiovascular diseases (CVDs) represent a significant global health challenge, leading to heart failure in numerous cases. Addressing this issue requires the development of effective strategies. In this study, we employ ensemble learning models, specifically “Bagging” and “Boosting”, to predict the risk of cardiovascular diseases using a Kaggle dataset comprising 11 features and 70,000 observations. Our investigation focuses on exploring the potential of ensemble models such as AdaBoost, Random Forest, Gradient Boosting, and Gaussian Naive Bayes to enhance the prediction performance for a medical dataset. Additionally, we highlight the importance of dimensionality reduction through Principal Component Analysis (PCA). The findings underscore the critical role of dimensionality reduction. Applying the Bagging and Boosting models with dimensionality reduction results in higher accuracy, precision, recall, F1-score, and Area under Curve (AUC). Leveraging dimensionality reduction significantly improves the model performance, yielding substantial enhancements in predictive capabilities.
