Clustering as Data Mining Technique in Risk Factors Analysis of Diabetes, Hypertension and Obesity.

Abstract

This investigation explores data mining using open source software WEKA in health care application. The cluster analysis technique is utilized to study the effects of diabetes, obesity and hypertension from the database obtained from Virginia school of Medicine. The simple k-means cluster techniques are adopted to form ten clusters which are clearly discernible to distinguish the differences among the risk factors such as diabetes, obesity and hypertension. Cluster formation was tried by trial and error method and also kept the SSE as low as possible. The SSE is low when numbers of clusters are more. Less than ten clusters formation unable to yield distinguishable information. In this work each cluster is revealing quit important information about the diabetes, obesity, hypertension and their interrelation. Cluster 0: Diabetes ? Obesity ? Hypertension = Healthy patient, Cluster 1: Diabetes ? Obesity ? Hypertension = Healthy patient, Cluster2: Diabetes ? Obesity ? Hypertension = Obesity, Cluster3: Diabetes ? Obesity ? Hypertension = Patients with Obesity and Hypertension, Cluster4: Boarder line Diabetes ? Obesity ? Hypertension = Sever obesity, Cluster5: Obesity ? Hyper tension ? Diabetes = Hypertension, Cluster6: Border line obese ? Border line hypertension ? Diabetes = No serious complications, Cluster 7: Obesity ? Hypertension ? Diabetes= Healthy patients, Cluster 8: Obesity ? Hypertension ? Diabetes= Healthy patients, and Cluster 9: Diabetes ? Hyper tension ? Obesity = High risk unhealthy patients.