A Comparison between Different Kernels of Support Vector Machine to Predict Cardiovascular Diseases Using Phonocardiogram Signal

Early detection and treatment of cardiovascular diseases are essential in lowering the risk of mortality. The goal of this research study is to more accurately predict cardiovascular disease in a patient by integrating the outcome via the machine learning approaches, namely the Support Vector Machine. Using phonocardiogram signals as instances, the authors of this research evaluated the effectiveness of SVM with Linear, Quadratic, Cubic, Fine Gaussian, Medium Gaussian, and Coarse Gaussian kernels by comparing them in terms of Accuracy, Precision, Recall, F1-Score, and Specificity. The results indicate that SVM with the Cubic kernel performs substantially better than the other five kernels and is preferred for the prediction of cardiovascular diseases.