"Explainable" artificial intelligence (AI) and "interpretable" machine learning (ML) are taking various research communities by storm. While there are numerous research and review articles proposing new models and discussing the nature of explanation, we still do not know what the explanation is, or how to measure it. So, to some extent, the research on what constitutes an explainable AI or model is not properly defined. Also, automated machine explainability is not properly defined. Thus, in our research, we try to combine theories from the social sciences, computer science, cognitive science, and linguistics to provide a logical framework. It has the potential to become a theoretical foundation for discussing explainable AI. We also conducted experiments to test this framework. While a lot of work still needs to be done to make this framework more concrete, our experiments provide evidence in support of the framework.
Pavan Reddy Gottimukkula (Mon,) studied this question.