Heart Plaque Detection System Design using a ML Technique “Naive Bayes and LSSVM”

It is generally acknowledged that the prediction of cardiovascular disease is essential for informing practitioners and hence allowing educated health-related decisions to take place. Early identification is algorithmically seen to motivate lifestyle changes if appropriate and also ensure ideal medical therapy if needed in the population. Given that early intervention is known to be beneficial, machine learning (ML) appears as a valuable tool for predicting and gaining insight into heart disease symptoms. Our study adds to this knowledge base as it demonstrates that the Naive Bayes algorithm was a better predictor of cardiac plaque disease than the LSSVM method. The goal of the research work is to conduct heart plaque diagnosis using Naive Bayes algorithm and compare the accuracy with Least Squares SVM algorithm. The study divides the Naive Bayes method and the Least Squares SVM algorithm into two separate groups, with 20 samples being used to derive G power from the Kaggle dataset from the heart plaque disease depending on Clinical with a 95% confidence interval. LSSVM Algorithm has achieved accuracy of only 67.3 %, While Naive Bayes Algorithm gives the best accuracy of 78%. The dataset of this study found that the LSSVM algorithm had the lowest performance when compared to Naive Bayes-algorithm for detecting heart plaque.