Explanation-driven model stacking

With advances of artificial intelligence (AI), there is a growing need for provisioning of transparency and accountability to AI systems. These properties can be achieved with eXplainable AI (XAI) methods, extensively developed over the last few years with relation for machine learning (ML) models. However, the practical usage of XAI is limited nowadays in most of the cases to the feature engineering phase of the data mining (DM) process. We argue that explainability as a property of a system should be used along with other quality metrics such as accuracy, precision, recall in order to deliver better AI models. In this paper we present a method that allows for weighted ML model stacking and demonstrates its practical use in an illustrative example.