SVD based initialization: A head start for nonnegative matrix factorization

  title={SVD based initialization: A head start for nonnegative matrix factorization},
  author={Christos Boutsidis and Efstratios Gallopoulos},
  journal={Pattern Recognition},
We describe Nonnegative Double Singular Value Decomposition (NNDSVD), a new method designed to enhance the initialization stage of nonnegative matrix factorization (NMF). NNDSVD can readily be combined with existing NMF algorithms. The basic algorithm contains no randomization and is based on two SVD processes, one approximating the data matrix, the other approximating positive sections of the resulting partial SVD factors utilizing an algebraic property of unit rank matrices. Simple practical… CONTINUE READING
Highly Influential
This paper has highly influenced 23 other papers. REVIEW HIGHLY INFLUENTIAL CITATIONS
Highly Cited
This paper has 383 citations. REVIEW CITATIONS
Recent Discussions
This paper has been referenced on Twitter 1 time over the past 90 days. VIEW TWEETS


Publications citing this paper.
Showing 1-10 of 192 extracted citations

383 Citations

Citations per Year
Semantic Scholar estimates that this publication has 383 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 46 references

An implicitly restarted block Lanczos method for large - scale Hermitian eigenproblems

  • D. Calvetti J. Baglama, L. Reichel
  • SIAM J . Sci . Comput .
  • 2006

Marqui , bioNMF : A versatile tool for non - negative matrix factorization in biology

  • R. Zdunek A. Cichocki, P. Carmona-Saez A. Pascual-Montano, M. Chagoyen, F. Tirado, J. Carazo, - R.Pascual
  • BMC Bioinformatics
  • 2006

Similar Papers

Loading similar papers…