Data Exploration and Machine Learning using R: SVM and Logistic Regression on Cleveland Heart Disease Dataset

Cardiovascular diseases are common these days to every age group of patient. The early stage prediction may help in adapting healthy lifestyle so that high risk of life threat can be avoided. The researchers are continuously finding links from existing data sources so that heart diseases can be predicted at early stages. There are proven data mining techniques such as decision trees, support vector machine, logistic regression useful in prognosis of heart disease. This research focuses on predicting hear diseases using support vector machine and linear regression technique. The Cleveland heart disease dataset is used as sample dataset to find accuracy of these two chosen techniques. The comparison shows that logistic regression gives accurate results than support vector machine on heart disease dataset. The research analysis is conducted in R script where Cleveland Heart Disease Dataset is analyzed and two models (SVM, logistic regression) are implemented using R. The project concentrates on applying Support Vector Machine and Logistic Regression techniques on the above mentioned dataset.