Implementation of partitional clustering on ILPD dataset to predict liver disorders

Abstract

Cluster Analysis methods are very important, popular data summarization techniques applied in diverse environments. These techniques retrieve the hidden patterns in large datasets in the form of characterized patterns which can be interpreted further in different contexts. Widespread use of medical information systems and explosive growth of medical databases require traditional manual data analysis coupled with efficient computer assisted analysis. Medical Diagnosis is a difficult process which needs proficiency as well as experience to cope with a disease. Data segmentation is an application in medical domain used to analyze patient records, disease trends and health care resource utilization, which in turn assist a physician in Medical Diagnosis. In the present paper a technique based on classification techniques is proposed to predict liver disorders accurately. The main objective is to examine whether the proposed method can obtain better prediction accuracy to traditional classification algorithms. The classification results using the proposed method are found to be very promising and accurate.