A Survey on Interpretable Clustering

Clustering is the process of dividing a collection of physical or abstract objects into several classes composed of similar objects. Now there are many clustering algorithms with superior performance, but the clusters generated by them are difficult for human to understand. Thus, some interpretable clustering methods are proposed, which make the clustering results have good interpretability without much impact on the clustering accuracy. This paper reviews the interpretable clustering algorithms, introduces and summarizes the previous work in this field according to the different interpretative ways, including rules, rectangular bounds and decision trees, and explores the development of interpretable clustering algorithms in the future.