A Comparison Based Study of Supervised Machine Learning Algorithms for Prediction of Heart Disease

Recent time’s heart disease has been propagating at an intensified rate and has become a large cause of untimely deaths all over the world among non-communicable diseases. South Asian countries in particular seem to have a higher risk of heart disease suggest the studies. It is challenging to predict heart disease, the expertise required to do this is not easy to get and only highly experienced doctors have it. A large amount of data is available to us which can provide us with hidden information, therefore data analysis techniques can be used to make some effective decisions in this area. Not only would this provide us with the reliable way of predicting heart disease, but it would also reduce the pressure on the medical professionals. These algorithms can estimate the likelihood of a person developing heart disease based on factors such as age, gender, blood pressure, stress, and so on. We used a data set with 1025 samples of data including 13 attributes in our research. In this work, we used the Logistic Regression, Decision Tree, SVM, Naive Bayes, Random Forest, and KNN algorithms to predict heart disease, with the purpose of determining which method is the most reliable. In our case, Decision Tree came out at the top with 98.53% accuracy. In the end, we also compared the results of some other study with ours.