Document Clustering

An unsupervised technique that groups documents into clusters based on a chosen similarity measure so that items within a cluster are more similar to each other than to items in other clusters, used to reveal topical groupings, support browsing, or summarize collections.