Comparison of Data mining Techniques for Forecasting Diabetes Mellitus

Abstract: Data mining is the practice of examining large pre- existing databases in order to generate new information. There are different kinds of data mining techniques are available. Classification, Clustering, Association Rule and Neural Network are some of the most significant techniques in data mining. In Health care industries, Data mining plays a significant role. Most frequently the data mining is used in health care industries for the process of forecasting diseases. Diabetes is a chronic condition. This means that is lasts for a long time, often for someone’s whole life [1]. This paper studies the comparison of diabetes forecasting approaches using clustering techniques. Here we are using three different kinds of clustering techniques named as Hierarchical clustering, Density based clustering, and Simple K- Means clustering. Weka is used as a tool.

Keywords: Data mining, Diabetes, Forecast, Clustering, Hierarchical clustering, Density based clustering, K-means, Weka.