Adaptive sample repulsion against class-specific counterfactuals for explainable imbalanced classification

Key Points

• Improved explainability in imbalanced classification enhances model insights and user trust.
• Key improvements found with adaptive sampling over traditional methods in handling counterfactuals.
• Innovative approach introduces class-specific sample repulsion strategies to enhance overall classification accuracy.
• Enhancements support better model decisions, providing clearer interpretations for user applications.