Sulav Karki

COMPARATIVE ANALYSIS OF LOGISTIC REGRESSION, SUPPORT VECTOR MACHINE AND RANDOM FOREST ALGORITHMS FOR HEART DISEASE PREDICTION

This project presents a comparative analysis of three widely used machine learning algorithms for heart disease prediction: Logistic Regression, Support Vector Machine and Random Forest. Logistic Regression is a simple yet effective algorithm that identifies relationships between variables, while SVM excels at handling non-linear data and producing accurate classifications. Random Forest, an ensemble learning technique, combines multiple decision trees to enhance predictive accuracy and reduce overfitting. The results compare the performance of these algorithms based on metrics such as accuracy and F1-score. Logistic Regression achieved an accuracy of 87.72% and an F1-score of 0.8813, SVM recorded an accuracy of 84.21% and an F1-score of 0.8475, and Random Forest achieved an accuracy of 82.46% and an F1-score of 0.8275. This report demonstrates the potential of machine learning in heart disease prediction and offers insights into the strengths and limitations of these algorithms.