A Modified K-Nearest Neighbor Algoritrhm for Noisy Data Set

Abstract

Cardiovascular disease is one of the most fatal conditions in the present world. The lethalness of Cardiovascular disease can be understandable through the percentage of deaths caused due to heart attacks worldwide. Prediction of cardiovascular disease is considered as challenging task in the area of clinical data analysis. This paper presents a new Machine Learning based Decision Support System to predict the chances of a person to be affected by heart attack . It uses a dataset with attributes such as age, blood pressure, thickness of the artery etc, and the K-Nearest Neighbour (KNN) algorithm to predict risk of heart attack in a person. Experimental results prove that the proposed system can predict the heart disease at the starting stage with an accuracy above 85% by using KNN algorithm.