Clustering transactions using large items

Abstract

In traditional data clustering, similarity of a cluster of objects is measured by pairwise similarity of objects in that cluster. We argue that such measures are not appropriate for transactions that are sets of items. We propose the notion of large items, i.e., items contained in some minimum fraction of transactions in a cluster, to measure the similarity of a cluster of transactions. The intuition of our clustering criterion is that there should be many large items within a cluster and little overlapping of such items across clusters. We discuss the rationale behind our approach and its implication on providing a better solution to the clustering problem. We present a clustering algorithm based on the new clustering criterion and evaluate its effectiveness.