Supporting factors to improve the explanatory potential of contrast set mining: Analyzing brain ischaemia data

Abstract

The goal of exploratory pattern mining is to find patterns that exhibit yet unknown relationships in data and to provide insightful representations of detected relationships. This paper explores contrast set mining and an approach to improving its explanatory potential by using the so called supporting factors that provide additional descriptions of the detected patterns. The proposed methodology is described in a medical data analysis problem of distinguishing between similar diseases in the analysis of patients suffering from brain ischaemia.