IEEE WCCI 2024 - Tutorial on learning from imbalanced data streams
Alberto Cano - Virginia Commonwealth University
This tutorial covers the many challenges in learning from data streams with imbalance, including data-level difficulties, concept drift, and the data and algorithm level approaches to address these issues. The tutorial will provide an overview of the state of the art, discuss benchmarks and performance metrics, and will give the participants the code of a framework for evaluating and comparing algorithms for imbalanced data streams.
Slides
Source code and experiments on Github
- A survey on learning from imbalanced data streams: taxonomy, challenges, empirical study, and reproducible experimental framework
- ROSE: Robust Online Self-Adjusting Ensemble for Continual Learning from Imbalanced Drifting Data Streams
- Enhancing Concept Drift Detection in Drifting and Imbalanced Data Streams through Meta-Learning
- Kappa Updated Ensemble for drifting data stream mining
- A comprehensive analysis of concept drift locality in data streams
- Dynamic budget allocation for sparsely labeled drifting data streams
- Aging and rejuvenating strategies for fading windows in multi-label classification on data streams