Comparison of SVM-based heart disease prediction with naive bayes-based prediction on accuracy.

This study's objectives are to forecast cardiac disease using the support vector machine (SVM) approach, improve the accuracy of prediction using machine learning classifiers, and evaluate the usefulness of these various methods. The Support Vector Machine (SVM) and the Naive Bayes method are the two classes being compared here. The algorithms have been tested and evaluated using a dataset that contains 1,700 records. Clincalc.com suggests that the sample size should be 540, with an estimated 80 percent power for the pretest. The statistical analysis includes twenty different samples. Following the completion of the trial, the SVM algorithm obtained a mean accuracy of 89.38 percent in predicting heart disease, but the Naive Bayes technique only reached an accuracy of 80.66 percent. The results of independent sample t-tests indicate that there is a statistically significant difference in performance between the two approaches (p 0.05). The use of machine learning strategies is going to be investigated in this project in an effort to improve heart disease prediction. During the comparison, it was determined that SVM was better than Naive Bayes in terms of the output. [ABSTRACT FROM AUTHOR]