electronic international standard serial number (EISSN)
Abstract. Stacking is a widely used technique for combining classiﬁers and improving prediction accuracy. Early research in Stacking showed that selecting the right classiﬁers, their parameters and the meta-classiﬁers was a critical issue. Most of the research on this topic hand picks the right combination of classiﬁers and their parameters. Instead of starting from these initial strong assumptions, our approach uses genetic algorithms to search for good Stacking conﬁgurations. Since this can lead to overﬁtting, one of the goals of this paper is to empirically evaluate the overall efﬁciency of the approach. A second goal is to compare our approach with the current best Stacking building techniques. The results show that our approach ﬁnds Stacking conﬁgurations that, in the worst case, perform as well as the best techniques, with the advantage of not having to manually set up the structure of the Stacking system.