Search

Nowadays, cardiovascular deaths and diseases have increased at a fast rate worldwide. The early prediction of this disease is necessary to prevent the deaths. Detection of heart disease requires more experience and good knowledge about heart problems. To detect heart disease in medical science massive quantity of information is collected and accumulated as databases. All the accumulated information could not be worthwhile. Mostly, the datasets are assorted and dispersed in nature. So there is a need for extracting predictive information about heart diseases. Machine learning, an evolving technique that can be used to develop automated systems/frameworks to analyze the data in different domains. It performs well in health care also. There are many techniques available for analyzing and predicting the heart disease, this paper aims to develop predictive models to analyze the datasets relevant to heart disease based on random forest, SVM, J.48, Bayesian prediction and MLP. [ABSTRACT FROM AUTHOR]