TOPNMF: Topic based Document Clustering using Non-negative Matrix Factorization

Objectives: This work focuses on creating targeted content-specific topicbased clusters. They can help users to discover the topics in a set of documents information more efficiently. Methods/Statistical analysis: The Non-negative Matrix Factorization (NMF) based models learn by directly decomposing term-document matrix, which is bag-of-word matrix representation text corpus, into two low-rank factor matrices namely Word-Topic feature Matrix(WTOM) and Document-Topic Matrix(DTOM). Topic clusters Document are extracted from obtained features matrices. method does not require any statistical distribution probability. Experiments were carried out subset BBC sport Corpus. Findings: experimental results indicate that accuracy TONMF was observed as 100 percent. Novelty/Applications: NMF often fails improve given clustering result number parameters increases linearly with size corpus. computational complexity TOPNMF better than exact decomposition like Singular Value Decomposition (SVD). Keywords: cluster; factorization; K-means clustering; Word cloud