RIONIDA: A novel algorithm for imbalanced data combining instance-based learning and rule induction

Abstract

The article presents the Rule Induction with Optimal Neighbourhood for Imbalanced Data Algorithm (RIONIDA) learning algorithm based on combination of two widely-used empirical approaches: rule induction and instance-based learning for imbalanced data classification. The algorithm is a substantial extension of the well-known the Rule Induction with Optimal Neighbourhood Algorithm (RIONA) learning algorithm developed for balanced data.

RIONIDA uses rules more general than in RIONA and realises a few additional concepts in comparison to RIONA, i.e. optimisation of the explicitly given performance measure defined over the confusion matrix, optimisation of weights for two classes, the idea of scaled rules, optimisation of parameters related to two latter ideas. RIONIDA, with decisions explainable for the user, is relatively fast and significantly outperforms the state-of-the-art algorithms analysed in the article.