A Novel Machine Learning Approaches for Heart Disease Dataset

Abstract

CVDs are concertedly contributed by hypertension, diabetes, overweight and unhealthy lifestyles. Around 17.9 million people die every year due to heart diseases accounting for 31% of all the deaths in the world. It is important for early and accurate detection of heart diseases This work focuses on the optimal solution for producing pattern by using deductive learning algorithms for heart disease dataset. The RBF Kernel has high correlation coefficient compare with other models which is 0.69. The Linear Kernel has the correlation coefficient value is 0.68. The Puk kernel produces the low correlation coefficient compare with others. The Linear kernel has very low Mean absolute error, Root Mean Squared Error, Relative absolute error and Root squared error which are 0.28, 0.36, 57.16% and 72.90% respectively. This model is comparatively good for other models.