Predictive Power of a Generalized Preventive Care Segmentation Model

Abstract: Healthcare costs in America have risen sharply in recent years. One of the most effective ways to combat this trend is via preventive health care measures. We explore a dataset consisting of insurance claims provided by our sponsor, Silverlink, in order to pinpoint patterns that will enable insurance providers to identify and contact their customers who are most likely to respond to some form of preventive health care intervention. To detect these patterns, we use a variety of data mining techniques such as clustering, association analysis, decision trees, and regression.