Fast and Effective Subgroup Mining for Continuous Target Variables

Subgroup mining is a flexible data mining method that considers a given target variable and aims to discover interesting subgroups with re- spect to this property of interest. In this paper, we especially focus on the handling of continuous target variables: We propose novel formaliza- tions of effective pruning strategies for reducing the search space, and we present the SD-Map* algorithm that enables fast subgroup mining for continuous target variables. Furthermore, we present effective visualization techniques that are seamlessly integrated in the semi-automatic sub- group mining process. The evaluation demon- strates the efficiency and effectiveness of the pre- sented approaches.