An Empirical Assessment of Machine Learning Algorithms for Predicting Heart Disease

Abstract: Cardiovascular disorders remain a primary cause of mortality worldwide, emphasizing the urgent need for early diagnostic support systems. Machine learning (ML) techniques have emerged as promising tools for assisting clinical decision-making through data-driven prediction models. This study presents a comprehensive empirical evaluation of several supervised learning algorithms for heart disease prediction using structured clinical data. The performance of Logistic Regression, Support Vector Machine, k-Nearest Neighbours, Decision Tree, Random Forest, Gradient Boosting, and Artificial Neural Networks is systematically compared. Models are assessed using accuracy, precision, recall, F1-score, and ROC-AUC metrics. The results of the study show that the k-Nearest Neighbours algorithm achieved the best overall performance with the highest accuracy and balanced evaluation metrics. The Support Vector Machine model also showed strong and consistent performance across different validation folds. Other models such as Logistic Regression and Random Forest produced competitive results, while Decision Tree and Gradient Boosting showed comparatively lower performance. Overall, the results demonstrate that machine learning techniques can be effective tools for predicting heart disease and can help support early diagnosis. In the future, the performance of these models could be improved by using larger datasets and applying advanced feature selection or optimization techniques. These findings demonstrate that machine learning techniques can be useful for supporting early detection of heart disease and assisting healthcare professionals in decision-making.

Keywords: Cardiovascular Disease, Predictive Modelling, Healthcare Analytics, Artificial Intelligence, Machine Learning.