Hellinger Distance Weighted Ensemble for Imbalanced Data Stream Classification

TL;DR

This paper introduces the Hellinger Distance Weighted Ensemble (HDWE), a novel method for classifying imbalanced, non-stationary data streams by effectively pruning ensembles to improve minority and majority class accuracy.

Contribution

The paper proposes a new ensemble method using Hellinger Distance for pruning, specifically designed for imbalanced, non-stationary data streams, and demonstrates its effectiveness through extensive experiments.

Findings

HDWE outperforms state-of-the-art methods in imbalanced data stream classification.

The choice of base classifier impacts the ensemble's performance.

HDWE is effective across various imbalanced data stream scenarios.

Abstract

The imbalanced data classification remains a vital problem. The key is to find such methods that classify both the minority and majority class correctly. The paper presents the classifier ensemble for classifying binary, non-stationary and imbalanced data streams where the Hellinger Distance is used to prune the ensemble. The paper includes an experimental evaluation of the method based on the conducted experiments. The first one checks the impact of the base classifier type on the quality of the classification. In the second experiment, the Hellinger Distance Weighted Ensemble (HDWE) method is compared to selected state-of-the-art methods using a statistical test with two base classifiers. The method was profoundly tested based on many imbalanced data streams and obtained results proved the HDWE method's usefulness.