Nonnegative Matrix Factorization with Zellner Penalty


Nonnegative matrix factorization (NMF) is a relatively new unsupervised learning algorithm that decomposes a nonnegative data matrix into a parts-based, lower dimensional, linear representation of the data. NMF has applications in image processing, text mining, recommendation systems and a variety of other fields. Since its inception, the NMF algorithm has been modified and explored by numerous authors. One such modification involves the addition of auxiliary constraints to the objective function of the factorization. The purpose of these auxiliary constraints is to impose task-specific penalties or restrictions on the objective function. Though many auxiliary constraints have been studied, none have made use of data-dependent penalties. In this paper, we propose Zellner nonnegative matrix factorization (ZNMF), which uses data-dependent auxiliary constraints. We assess the facial recognition performance of the ZNMF algorithm and several other well-known constrained NMF algorithms using the Cambridge ORL database.

Share and Cite:

Corsetti, M. and Fokoué, E. (2015) Nonnegative Matrix Factorization with Zellner Penalty. Open Journal of Statistics, 5, 777-786. doi: 10.4236/ojs.2015.57077.

Conflicts of Interest

The authors declare no conflicts of interest.


[1] Lee, D.D. and Seung, H.S. (1999) Learning the Parts of Objects by Nonnegative matrix Factorization. Nature, 401, 788-791.
[2] Li, S.Z., Hou, X., Zhang, H. and Cheng, Q. (2001) Learning Spatially Localized Parts-Based Representations. IEEE Conference on Computer Vision and Pattern Recognition, Kauai, 8-14 December 2001, 207-212.
[3] Wang, Y., Jia, Y., Hu, C. and Turk, M. (2004) Fisher Nonnegative Matrix Factorization for Learning Local Features. Asian Conference of Computer Vision, Jeju, January 2004, 27-30.
[4] Wang, Y., Jia, Y., Hu, C. and Turk, M. (2005) Nonnegative Matrix Factorization Framework for Face Recognition. International Journal of Pattern Recognition and Artificial Intelligence, 19, 495-511.
[5] Huang, X., Zhao, J., Ash, J. and Lai, W. (2013) Clustering Student Discussion Messages on Online Forum by Visualization and Nonnegative matrix Factorization. Journal of Software Engineering and Applications, 6, 7-12.
[6] Guillamet, D., Bressan, M. and Vitria, J. (2001) A Weighted Nonnegative Matrix Factorization for Local Representations. IEEE Conference of Computer Vision and Pattern Recognition, Kauai, 8-14 December 2001, 942-947.
[7] Cichocki, A., Zdunek, R. and Amari, S. (2006) Csiszár’s Divergence for Nonnegative Matrix Factorization: Family of New Algorithms. Lecture Notes in Computer Science, 3889, 32-39.
[8] Pauca, V.P., Piper, J. and Plemmons, R.J. (2006) Nonnegative Matrix Factorization for Spectral Data Analysis. Linear Algebra and its Applications, 416, 29-47.
[9] Hamza, A.B. and Brady, D.J. (2006) Reconstruction of Reflectance Spectra Using Robust Nonnegative Matrix Factorization. IEEE Transactions on Signal Processing, 54, 3637-3642.
[10] Lin, C.J. (2007) Projected Gradient Methods for Nonnegative Matrix Factorization. Neural Computation, 19, 2756-2779.
[11] Gonzales, E.F. and Zhang, Y. (2005) Accelerating the Lee-Seung Algorithm for Nonnegative Matrix Factorization. Technical Report, Department of Computational and Applied Mathematics, Rice University, Houston.
[12] Zdunek, R. and Cichocki, A. (2006) Nonnegative Matrix Factorization with Quasi-Newton Optimization. Lecture Notes in Computer Science, 4029, 870-879.
[13] Wild, S., Curry, J. and Dougherty, A. (2004) Improving Nonnegative Matrix Factorization through Structured Initialization. Pattern Recognition, 37, 2217-2232.
[14] Wild, S. (2002) Seeding Nonnegative Matrix Factorization with the Spherical K-Means Clustering. Master’s Thesis, University of Colorado, Colorado.
[15] Wild, S., Curry, J. and Dougherty, A. (2003) Motivating Nonnegative Matrix Factorizations. Proceedings of the 8th SIAM Conference on Applied Linear Algebra, Williamsburg, 15-19 July 2003.
[16] Piper, J., Pauca, J.P., Plemmons, R.J. and Giffin, M. (2004) Object Characterization from Spectral Data Using Nonnegative Factorization and Information Theory. Proceedings of the 2004 AMOS Technical Conference, Maui, 9-12 September 2004.
[17] Hoyer, P.O. (2002) Non-Negative Sparse Coding. Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing, Martigny, Switzerland, 4-6 September 2002, 557-565.
[18] Hoyer, P.O. (2004) Nonnegative Matrix Factorization with Sparseness Constraints. Journal of Machine Learning Research, 5, 1457-1469.
[19] Liu, W., Zheng, N. and Lu, X. (2003) Nonnegative Matrix Factorization for Visual Coding. 2003 IEEE International Conference on Acoustics, Speech and Signal Processing, 6, 293-296.
[20] Zellner, A. (1986) On Assessing Prior Distributions and Bayesian Regression Analysis with g-Prior Distributions. In: Goel, P. and Zellner, A., Eds., Bayesian Inference and Decision Techniques: Essays in Honor of Bruno de Finetti, Elsevier Science Publishers, Inc., New York, 233-243.
[21] Foster, D.P. and George, E.I. (1994) The Risk Inflation Criterion for Multiple Regression. Annals of Statistics, 22, 1947-1975.

Copyright © 2022 by authors and Scientific Research Publishing Inc.

Creative Commons License

This work and the related PDF file are licensed under a Creative Commons Attribution 4.0 International License.