A New Approach for Cardiovascular Disease Prediction Using Decision Tree

Abstract

Heart diseases have have been the central reason for death everywhere throughout the world in the course of the most recent couple of decades. To stay away from heart malady or coronary sickness and find signs early, people more than 55 years old must have an all-out cardiovascular exam. Scientists and experts created different shrewd procedures to improve the limits of the human services experts in acknowledgment of cardiovascular infection. In cardiovascular infection finding and treatment, single information mining procedures are giving the sensible exactness and accuracy. All things considered, the use information mining method is equipped for diminishing the quantity of tests that are required to be done. So as to diminish the figure of passing from heart diseases, there must be a snappy and proficient recognition method giving better accuracy and exactness. The point of this paper is to introduce an effective method of anticipating heart diseases utilizing machine learning draws near. In this paper, a hybridization method is proposed in which decision tree and counterfeit neural system classifiers are hybridized for better execution of forecasts of heart illness. This is finished utilizing WEKA. To approve the presentation of the proposed algorithm, a ten-times approval test is performed on the dataset of heart sickness patients, which is taken from the UCI storehouse. The accuracy, affectability, and explicitness of the individual classifier and half-and-half procedure are analyzed.