The goal of Knowledge Discovery is to extract knowledge from a set of data. Most common techniques used in knowledge discovery are clustering methods, whose goal is to analyze a set of objects and obtain clusters based on the similarity among these objects. A desirable characteristic of clustering results is that these should be easily understandable by domain experts. In fact, these are characteristics that exhibit the results of eager learning methods (such as ID3) and lazy learning methods when used for building lazy domain theories. In this paper we propose LazyCL, a procedure using a lazy learning method to produce explanations on clusters of unlabeled cases. The analysis of the relations among these explanations converges to a correct clustering of the data set.