Hellinger Distance Weighted Ensemble for imbalanced data stream classification

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