Development of a Data Clustering Algorithm for Predicting Heart

Abstract

This research paper proposes the findings of the accuracy of the result by using the K-Means clustering technique in prediction of heart disease diagnosis with real and artificial datasets. K-Means Clustering is a method of cluster analysis which aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean. Each cluster is assigned a random target number of clusters-k and started from a random initialization. The proposed technique classifies the group of the objects based on attributes into K number of groups. The grouping is done by minimizing the sum of squares of distances between data using Euclidean distance formula and the corresponding cluster centroid. The research result shows that the integration of clustering gives promising results with highest accuracy rate and robustness.