Evolving multi-user fuzzy classifier system with advanced explainability and interpretability aspects

Edwin Lughofer, Mahardhika Pratama

Abstract: Highlights•Evolving multi-user classifier system fusioning users’ feedback behaviors.•Advanced explainability aspects to increase understanding of classifier decisions.•Advanced interpretability aspects to guide users to an improved understanding.•Integration of a sample forgetting strategy to properly handle drifts in streams.•Evaluation in an industrial application showing accuracy, decision reasons and rules.