Adaptive learning on hierarchical data streams using window-weighted Gaussian probabilities

Eduardo Tieppo, Júlio César Nievola, Jean Paul Barddal

Abstract: Highlights•An adaptive learning method that addresses the stability–plasticity dilemma trade-off in hierarchical data stream classification using window weighted Gaussian probabilities.•A current overview of state-of-the-art techniques and definitions in hierarchical data stream classification, following established criteria in related literature.•A unified assessment protocol to compare hierarchical data stream classifiers using different evaluation metrics, enabling reproducible testbeds for future studies.•The proposed methods are compared with state-of-the-art techniques and a benchmark of the hierarchical data stream classification area is provided.