Title :
On the clustering aspect of nonnegative matrix factorization
Author :
Mirzal, Andri ; Furukawa, Masashi
Author_Institution :
Grad. Sch. of Inf. Sci. & Technol., Hokkaido Univ., Sapporo, Japan
Abstract :
This paper provides a theoretical explanation on the clustering aspect of nonnegative matrix factorization (NMF). We prove that even without imposing orthogonality nor sparsity constraint on the basis and/or coefficient matrix, NMF still has clustering capability, thus giving a theoretical support for many works, e.g., Xu et al. and Kim et al., that show the superiority of the standard NMF as a clustering method.
Keywords :
concave programming; constraint handling; data mining; learning (artificial intelligence); matrix decomposition; pattern clustering; clustering aspect; nonconvex optimization; nonnegative matrix factorization; orthogonality constraint; sparsity constraint; Clustering algorithms; Clustering methods; Data mining; Optimization; Sparse matrices; Symmetric matrices; Vectors; bound-constrained optimization; clustering method; non-convex optimization; nonnegative matrix factorization;
Conference_Titel :
Electronics and Information Engineering (ICEIE), 2010 International Conference On
Conference_Location :
Kyoto
Print_ISBN :
978-1-4244-7679-4
Electronic_ISBN :
978-1-4244-7681-7
DOI :
10.1109/ICEIE.2010.5559822
