Toward Explainable Artificial Intelligence Through Fuzzy Systems

Explainable Artificial Intelligence is a novel paradigm conjugating the effectiveness of machine learning with the new requirements coming from the integration of intelligent systems in the human society. Explainable Artificial Intelligence can find successful application in a plethora of contexts, endowing classical intelligent systems with a crucial added value: the possibility for users to interact with machines, validate their results and ultimately trust their behavior. Fuzzy Set Theory provides a mathematical framework which is especially suitable to model concepts and perceptions of physical reality, thus injecting a kind of common-sense reasoning into machine learning algorithms and realizing a human-centric information processing which is the core of the Granular Computing paradigm. In particular, the exploitation of natural language enables Fuzzy Set Theory to act as a key-element for designing explainable models which can be able to provide explanations useful for human beings. We argue on those topics in order to show how the development of explainable fuzzy systems is a promising direction for paving the way from interpretable fuzzy systems to explainable intelligent systems.