Using Symbolic Similarity to Explain Case-based Reasoning in Classification Tasks

The explanation of the results is a key point of automatic problem solvers. CBR systems solve a new problem by assessing its similarity with already solved cases and they commonly show the user the set of cases that have been assessed as the most similar to the new problem. Using the notion of symbolic similarity, our proposal is to show the user a symbolic description that makes explicit what the new problem has in common with the retrieved cases. Specifically, we use the notion of anti-unification (least general generalization) to build symbolic similarity descriptions. We also present an explanation scheme using anti-unification for CBR in classification tasks that focuses on explaining what is shared between the current problem and the retrieved cases that belong to different classes.