Text Clustering Using Cosine Similarity and Matrix Factorization

Clustering is a useful technique that organizes a large quantity of unordered text documents into a small number of meaningful and coherent clusters, thereby providing a basis for intuitive and informative navigation and browsing mechanisms. Text-clustering is to divide a collection of textdocuments into different categories so that documents in the same category describe the same topic such as classical music. Text Clustering efficiently groups documents with similar content into same cluster: Similarity between objects is measured within the use of similarity function. The hierarchical clustering schemes can be effectively used for processing large datasets. In this paper, it is proposed to use the hierarchical clustering technique entitled “sub leader algorithm” along with cosine similarity is to cluster the documents. Key words-Similarity measures, text clustering, data