Post-processing of BRACID Rules Induced from Imbalanced Data

Issue title: Rough Sets, Rule-Based Approaches, and Knowledge Representation

Guest editors: Davide Ciucci, Dominik Ślęzak and Marcin Wolski

Article type: Research Article

Authors: Napierala, Krystyna^a | Stefanowski, Jerzy^{b; *}

Affiliations: [a] DATAX sp. z o.o., 53-609 Wroclaw, Poland. krystyna.napierala@datax.pl | [b] Institute of Computing Sciences, Poznan University of Technology, 60-965 Poznan, Poland. Jerzy.Stefanowski@cs.put.poznan.pl

Correspondence: [*] Address for correspondence: Institute of Computing Sciences, Poznan University of Technology, 60-965 Poznan, Poland

Abstract: Rule-based classifiers constructed from imbalanced data fail to correctly classify instances from the minority class. Solutions to this problem should deal with data and algorithmic difficulty factors. The new algorithm BRACID addresses these factors more comprehensively than other proposals. The experimental evaluation of classification abilities of BRACID shows that it significantly outperforms other rule approaches specialized for imbalanced data. However, it may generate too high a number of rules, which hinder the human interpretation of the discovered rules. Thus, the method for post-processing of BRACID rules is presented. It aims at selecting rules characterized by high supports, in particular for the minority class, and covering diversified subsets of examples. Experimental studies confirm its usefulness.

Keywords: Rule induction, class imbalances, interpretability of rules, filtering of rules

DOI: 10.3233/FI-2016-1422

Journal: Fundamenta Informaticae, vol. 148, no. 1-2, pp. 51-64, 2016