Coverage-based Outlier Explanation
This addresses the under-studied issue of outlier explanation for domain experts or policy makers, offering an incremental improvement over existing methods.
The paper tackles the problem of explaining what characterizes outliers in a dataset, aiming to provide semantic explanations understandable by domain experts, and shows through experiments that their method efficiently generates better explanations compared to rule-based learners.
Outlier detection is a core task in data mining with a plethora of algorithms that have enjoyed wide scale usage. Existing algorithms are primarily focused on detection, that is the identification of outliers in a given dataset. In this paper we explore the relatively under-studied problem of the outlier explanation problem. Our goal is, given a dataset that is already divided into outliers and normal instances, explain what characterizes the outliers. We explore the novel direction of a semantic explanation that a domain expert or policy maker is able to understand. We formulate this as an optimization problem to find explanations that are both interpretable and pure. Through experiments on real-world data sets, we quantitatively show that our method can efficiently generate better explanations compared with rule-based learners.