DCCR: Document Clustering by Conceptual Relevance as a Factor of Unsupervised Learning
Annaluri Sreenivasa Rao, Sri Ramakrishna · 2014
Present document clustering approaches cluster the documents by term frequency, which are not considering the concept and semantic relations during unsupervised or supervised learning. In this paper, a new unsupervised learning approach that estimates similarity between any two documents given by concept similarity measure is proposed. This novel method represents the concept as set of word sequences found in given documents. In regard to demonstrate the significance of the proposal we applied on set of benchmark document datasets.