Towards Knowledge-Intensive Subgroup Discovery.

Abstract

Subgroup discovery can be applied for exploration or descriptive induction in order to discover ”interesting” subgroups of the general population, given a certain property of interest. In domains with available background knowledge, the user usually wants to utilize this to improve the quality of the subgroup discovery results. We describe a knowledge-intensive approach for subgroup discovery utilizing several types of background knowledge, which can be applied incrementally. Our application area is the medical domain of sonography. The context of our work is to identify interesting diagnostic patterns using subgroup discovery techniques, to supplement a medical documentation and consultation system. We present an experimental evaluation of our approach using a case base from a real world application.