Heart diseases are one of the most challenging
problems faced by the Health Care sectors all over the world.
These diseases are very basic now a days. With the expanding
count of deaths because of heart illnesses, the necessity to build up
a system to foresee heart ailments precisely. The work in this
paper focuses on finding the best Machine Learning algorithm for
identification of heart diseases. Our study compares the precision
of three well known classification algorithms, Decision Tree and
Naïve Bayes, Random Forest for the prediction of heart disease by
making the use of dataset provided by Kaggle. We utilized various
characteristics which relate with this heart diseases well, to find
the better algorithm for prediction. The result of this study
indicates that the Random Forest algorithm is the most efficient
algorithm for prediction of heart disease with accuracy score of
97.17%.