Efficiently Discovering Locally Exceptional Yet Globally Representative Subgroups

Abstract

Abstract-Subgroup discovery is a local pattern mining technique to find interpretable descriptions of sub-populations that stand out on a given target variable. That is, these subpopulations are exceptional with regard to the global distribution. In this paper we argue that in many applications, such as scientific discovery, subgroups are only useful if they are additionally representative of the global distribution with regard to a control variable. That is, when the distribution of this control variable is the same, or almost the same, as over the whole data.We formalise this objective function and give an efficient algorithm to compute its tight optimistic estimator for the case of a numeric target and a binary control variable. This enables us to use the branch-and-bound framework to efficiently discover the top-k subgroups that are both exceptional as well as representative. Experimental evaluation on a wide range of datasets shows that with this algorithm we discover meaningful representative patterns and are up to orders of magnitude faster in terms of node evaluations as well as time.

Information efficient learning of complexly structured preferences: Elicitation procedures and their application to decision making under uncertainty

“…Fortunately, for the case of subgroup discovery, for example an explicit and computationally feasible analysis in terms of Vapnik-Chervonenkis theory (see, e.g., Vapnik (2006)) is possible. Because of the close connection between formal concept analysis and subgroup discovery (c.f., e.g., Boley and Grosskreutz (2009); Boley et al (2010)), this does also allow for the application of regularization strategies within this discrete setting, where, for instance, the ideas developed in (Schollmeyer et al, 2017, p. 26 ff) could proof fruitful (c.f., also Kalofolias et al (2017); Mandros et al (2018)).…”