An Expert System For Heart Disease Prediction Applying On Different Clustering Approach

Abstract

Heart attacks are believed to be the most common of all hazardous disorders. Medical professionals perform several surveys on Heart Disease(HD) and gather data on patients, their illness progression, and symptoms.In many levels of disease progression, it is difficult to connect a heart patient's symptoms to the heart disease. In this study, Data Mining (DM) is applied to a database in order to find a hidden trend in a clinical dataset. Clustering is a habitual DM technique that organizing data elements based on their commensurable criterion. This study revealed how to capture clusters and establish the new centroid utilizing K-Means (KM) , Canopy Clustering (CC) algorithms and Farthest First Algorithms. The KM technique  is a broadly used clustering technique that is employed in a extensive range of scientific and industrial applications. Canopy Clustering is a straight forward and rapid approach for accurately grouping items into clusters. Farthest First Techniques (FFT) is fitting for the massive dataset and its fabricates for the non-uniform cluster. In the end, the execution of  three algorithms were investigated in this work. The American Heart Association (AHA) provided a dataset of 50 participants to examine the heart disease dataset.