HEART DISEASE PREDICTION USING RANDOM FOREST ALGORITHM

Present days one of the major application areas of machine learning algorithms is medical diagnosis of diseases and treatment. Machine learning algorithms also used to find correlations and associations between different diseases. Nowadays many people are dying because of sudden heart attack .Prediction and diagnosing of heart disease becomes a challenging factor faced by doctors and hospitals both in India and abroad. In order to reduce number of deaths because of heart diseases, we have to predict whether person is at the risk of heart disease or not   in advance. Data mining techniques and machine learning algorithms play a very important role in this area. Many researchers are carrying out their research in this area to develop software that can help doctors to take decision regarding both prediction and diagnosing of heart disease. In this paper we focused on how data mining techniques can be used to predict heart disease in advance such that patient is well treated. We used different algorithms for comparative analysis but random forest algorithm has shown highest accuracy in prediction. We used Random forest machine learning algorithms supported by WEKA to predict heart disease in advance. Dataset contains 303 samples and 14 input features as well as 1output feature. The dataset is available in UCI Machine Learning Repository; we used 65% data for training and 35% data for testing. The algorithm has shown 0.763 precision and 0.935 recall in predicting negative class tupples.