Class imbalances versus small disjuncts

TLDR

Class imbalances are commonly believed to cause major performance losses in standard classifiers. This study investigates whether class imbalance itself degrades performance or whether the effect is due to small disjuncts, proposing that targeting small disjuncts may be more effective. The authors develop and test a method that explicitly addresses the small disjunct problem. Experiments show that small disjuncts, not class imbalance per se, drive degradation, and that the small‑disjunct method outperforms standard and advanced class‑imbalance solutions.

Abstract

It is often assumed that class imbalances are responsible for significant losses of performance in standard classifiers. The purpose of this paper is to the question whether class imbalances are truly responsible for this degradation or whether it can be explained in some other way. Our experiments suggest that the problem is not directly caused by class imbalances, but rather, that class imbalances may yield small disjuncts which, in turn, will cause degradation. We argue that, in order to improve classifier performance, it may, then, be more useful to focus on the small disjuncts problem than it is to focus on the class imbalance problem. We experiment with a method that takes the small disjunct problem into consideration, and show that, indeed, it yields a performance superior to the performance obtained using standard or advanced solutions to the class imbalance problem.