Analysis and Prediction of Heart Disease Using Decision Tree and Naive Bayes

The extraction of the hidden data from the large databases is data mining and it is also known as Knowledge Discovery Mining. It has many tasks one of them used here is predictive tasks which uses, that some variables to predict unknown or future values of other dataset. The major health problem affects large number of people is Heart disease . Unless it is treated at an early stage it causes death. Healthcare industry generates today large amount of complex data about the patients and resources of the hospitals, from the time where there has no sufficient focus on effective analysis tools to discover relationships in data especially in medical sector. The techniques of mining data are used to analyse rich collections of data from different perspectives and derive useful information, to develop diagnosis and predicting system for heart disease based on predictive mining. Number of trials are taken up to compare the performances of different techniques of data mining including Decision tree and Naïve Bayes algorithms. As proposed, the risk factors are taken into consideration ,Decision tree and Naïve Bayes are applied and performance on their diagnosis have been compared by the UCI Machine Learning Repository i,e WEKA tool. Thus, the Naïve Bayes outperforms the Decision tree .