Clustering is an important technique for discovering relatively dense sub-regions or sub-spaces of a multi-dimension data distribution. describes a clustering method for a general metric space which is a common model of data relevant to information retrieval.