A Comparative Study of Heart Disease Prediction Based on Principal Component Analysis and Clustering Methods

Abstract

In this article, we propose clustering approach based on Principal Component Analysis (PCA) to diagnosis of heart disease patients. At the first stage, the original dataset is reduced using PCA reduction method. Then, at the second stage, reduced dataset is applied to clustering methods which is based on fuzzy C-means and K-means algorithms. These algorithms are implemented and tested on a Cleveland heart disease dataset. We compared the clustering results with and without PCA. The results are suggesting that the combination of clustering algorithms and PCA was the most effective at heart disease diagnosis.