XStacking: Explanation-Guided Stacked Ensemble Learning
This provides a practical and scalable solution for responsible ML by making stacked models inherently explainable, addressing a key limitation for users needing interpretable predictions.
The paper tackles the lack of interpretability in ensemble machine learning, particularly stacking, by introducing XStacking, a framework that integrates dynamic feature transformation with Shapley additive explanations, achieving improvements in predictive accuracy and interpretability across 29 datasets.
Ensemble Machine Learning (EML) techniques, especially stacking, have been shown to improve predictive performance by combining multiple base models. However, they are often criticized for their lack of interpretability. In this paper, we introduce XStacking, an effective and inherently explainable framework that addresses this limitation by integrating dynamic feature transformation with model-agnostic Shapley additive explanations. This enables stacked models to retain their predictive accuracy while becoming inherently explainable. We demonstrate the effectiveness of the framework on 29 datasets, achieving improvements in both the predictive effectiveness of the learning space and the interpretability of the resulting models. XStacking offers a practical and scalable solution for responsible ML.