Diversity-Aware Weighted Majority Vote Classifier for Imbalanced Data
It addresses classification challenges in domains with imbalanced data, such as fraud detection and medical applications, but appears incremental as it builds on existing ensemble and PAC-Bayesian methods.
The paper tackles imbalanced binary classification by proposing DAMVI, a diversity-aware ensemble algorithm that reweights hard minority examples and optimizes classifier weights using the PAC-Bayesian C-Bound, showing efficiency compared to state-of-the-art models in tasks like predictive maintenance and fraud detection.
In this paper, we propose a diversity-aware ensemble learning based algorithm, referred to as DAMVI, to deal with imbalanced binary classification tasks. Specifically, after learning base classifiers, the algorithm i) increases the weights of positive examples (minority class) which are "hard" to classify with uniformly weighted base classifiers; and ii) then learns weights over base classifiers by optimizing the PAC-Bayesian C-Bound that takes into account the accuracy and diversity between the classifiers. We show efficiency of the proposed approach with respect to state-of-art models on predictive maintenance task, credit card fraud detection, webpage classification and medical applications.