Troika - An improved stacking schema for classification tasks Academic Article uri icon

abstract

  • Stacking is a general ensemble method in which a number of base classifiers are combined using one meta-classifier which learns the ir outputs. Such an approach provides certain advantages: simplicity; performanc e that is similar to the best classifier; and the capability of combining classif iers induced by different inducers. The disadvantage of stacking is that on multiclass problems, stacking seems to perform worse than other meta-learning approaches. In this paper we present Troika , a new stacking method for improving ensemble classifiers. The new scheme is built from three layers of combining classifiers. The new method was tested on various datasets and the results indicate the superiority o f the proposed method to other legacy ensemble schemes, Stacking and StackingC, especially when the classification task consists of more than two classes

publication date

  • December 15, 2009