Dynamic ensemble has significantly greater potential space to improve the classification of imbalanced data compared to static ensemble. However, dynamic ensemble schemes are far less successful than static ensemble methods in the imbalanced learning field. Through an in-depth analysis on the behavior characteristics of dynamic ensemble, we find that there are some important problems that need to be addressed to release the full potential of dynamic ensemble, including but not limited to, correcting the component classifiers’ bias towards the majority classes, increasing the proportions of the positive classifiers (i.e., the component classifiers making correct prediction) for difficult samples, and providing the accurate competence estimations on the hard-to-classify samples w.r.t the classifier pool. Inspired by these, we propose a Dynamic Ensemble Framework for imbalanced data classification (imDEF). imDEF first uses the data generation method OREM$\mathrm{_{G}}$to generate multiple artificial synthetic datasets, which have diverse class distributions by rebalancing the original imbalanced data. Based on each of such synthetic datasets, imDEF then utilizes a Classification Error-aware Self-Paced Sampling Ensemble (SPSE$\mathrm{_{CE}}$) method to gradually focus more on difficult samples, to create a low-biased classifier pool and increase the proportions of the positive classifiers for the difficult samples. Finally, imDEF constructs a referee system to achieve the competence estimations by leveraging an Ensemble Margin-aware Self-Paced Sampling Ensemble (SPSE$\mathrm{_{EM}}$) method. SPSE$\mathrm{_{EM}}$incrementally strengthens the learning of the hard-to-classify samples, so that the competent levels of component classifiers could be estimated accurately. Extensive experiments demonstrate the effectiveness of imDEF. The source codes have been made publicly available on GitHub.

This paper proposes a simple interpolation Oversampling method with the purpose of Reliably Expanding the Minority class regions (OREM). OREM first finds the candidate minority region around each original minority sample, then exploits this region to further identify those clean subregions without distributing any majority sample. The synthetic samples are only allowed to generate in the clean subregions, so that the regions of the minority class can be broadened reliably. Given that the learning from multiclass imbalanced data is more challenging as compared to two-class scenarios, we also extend OREM to handle multiclass imbalance problems by leveraging an iteration procedure of generating synthetic samples, consequently leading to a multiclass oversampling algorithm OREM-M. The key peculiarity of OREM-M is to reduce the class overlapping not only between the synthetic minority and original samples, but also from the synthetic samples of different minority classes. In this way, OREM-M ensures that the data of each class after oversampling can be modeled well. In addition, we embed OREM into boosting framework to develop a new ensemble method OREMBoost addressing class imbalance problems. Extensive experiments demonstrate the effectiveness of the proposed OREM, OREM-M, and OREMBoost.