In the present pandemic situation, health care data is generated voluminously in an unstructured format posing challenge to technology in perspective of analysis, classification and prediction. The data generated is converted to structured format. Suitability of methodology keeping in mind low computational complexity and high accuracy is a major concern which has emerged as a problem in data science. In this research work real time heart disease data set is considered to evaluate the accuracy of six supervised methods –SVM (Support Vector Machine), KNN (K-Nearest Neighbor), GNB (Gaussian Naïve Bayes), LR (Logistic Regression), DT (Decision Tree) and RF (Random Forest). Analysis through ROC curve and confusion matrix predominantly justify RF classifier and LR gives efficient results compared to other methods. This is a preprocessing stage; every researcher has to perform before deciding the methodology to be considered for further processing.
