Advertisement

Journal of Global Optimization

, Volume 62, Issue 3, pp 545–574 | Cite as

SymNMF: nonnegative low-rank approximation of a similarity matrix for graph clustering

  • Da Kuang
  • Sangwoon Yun
  • Haesun Park
Article

Abstract

Nonnegative matrix factorization (NMF) provides a lower rank approximation of a matrix by a product of two nonnegative factors. NMF has been shown to produce clustering results that are often superior to those by other methods such as K-means. In this paper, we provide further interpretation of NMF as a clustering method and study an extended formulation for graph clustering called Symmetric NMF (SymNMF). In contrast to NMF that takes a data matrix as an input, SymNMF takes a nonnegative similarity matrix as an input, and a symmetric nonnegative lower rank approximation is computed. We show that SymNMF is related to spectral clustering, justify SymNMF as a general graph clustering method, and discuss the strengths and shortcomings of SymNMF and spectral clustering. We propose two optimization algorithms for SymNMF and discuss their convergence properties and computational efficiencies. Our experiments on document clustering, image clustering, and image segmentation support SymNMF as a graph clustering method that captures latent linear and nonlinear relationships in the data.

Keywords

Symmetric nonnegative matrix factorization Low-rank approximation Graph clustering Spectral clustering 

Notes

Acknowledgments

The work of the first and third authors was supported in part by the National Science Foundation (NSF) Grants CCF-0808863 and the Defense Advanced Research Projects Agency (DARPA) XDATA program Grant FA8750-12-2-0309. The work of the second author was supported by the TJ Park Science Fellowship of POSCO TJ Park Foundation. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF, the DARPA, or the NRF.

References

  1. 1.
    Arbelaez, P., Maire, M., Fowlkes, C., Malik, J.: Contour detection and hierarchical image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 33(5), 898–916 (2011)CrossRefGoogle Scholar
  2. 2.
    Arora, R., Gupta, M.R., Kapila, A., Fazel, M.: Clustering by left-stochastic matrix factorization. In: ICML ’11: Proceedings of the 28nd International Conference on Machine Learning (2011)Google Scholar
  3. 3.
    Banerjee, A., Dhillon, I.S., Ghosh, J., Sra, S.: Clustering on the unit hypersphere using von Mises–Fisher distributions. J. Mach. Learn. Res. 6, 1345–1382 (2005)MATHMathSciNetGoogle Scholar
  4. 4.
    Berman, A., Plemmons, R.J.: Nonnegative Matrices in the Mathematical Sciences. SIAM, Philadelphia, PA (1994)MATHCrossRefGoogle Scholar
  5. 5.
    Berman, A., Shaked-Monderer, N.: Completely Positive Matrices. World Scientific, River Edge, NJ (2003)MATHCrossRefGoogle Scholar
  6. 6.
    Bertsekas, D.P.: Projected newton methods for optimization problems with simple constraints. SIAM J. Control Optim. 20(2), 221–246 (1982)MATHMathSciNetCrossRefGoogle Scholar
  7. 7.
    Bertsekas, D.P.: Nonlinear Programming, 2nd edn. Athena Scientific, Belmont, MA (1999)MATHGoogle Scholar
  8. 8.
    Cai, D., He, X., Han, J., Huang, T.S.: Graph regularized nonnegative matrix factorization for data representation. IEEE Trans. Pattern Anal. Mach. Intell. 33(8), 1548–1560 (2011)CrossRefGoogle Scholar
  9. 9.
    Catral, M., Han, L., Neumann, M., Plemmons, R.J.: On reduced rank nonnegative matrix factorization for symmetric matrices. Linear Algebra Appl. 393, 107–126 (2004)MATHMathSciNetCrossRefGoogle Scholar
  10. 10.
    Chan, P., Schlag, M., Zien, J.: Spectral k-way ratio-cut partitioning and clustering. IEEE Trans. CAD Integr. Circuits Syst. 13(9), 1088–1096 (1994)CrossRefGoogle Scholar
  11. 11.
    Choo, J., Lee, C., Reddy, C.K., Park, H.: Utopian: User-driven topic modeling based on interactive nonnegative matrix factorization. IEEE Trans. Vis. Comput. Graph. 19(12), 1992–2001 (2013)CrossRefGoogle Scholar
  12. 12.
    Cour, T., Benezit, F., Shi, J.: Spectral segmentation with multiscale graph decomposition. In: CVPR ’05: Proceedings of the 2005 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1124–1131 (2005)Google Scholar
  13. 13.
    Dhillon, I., Guan, Y., Kulis, B.: A Unified View of Kernel K-Means, Spectral Clustering and Graph Cuts. Technical Report TR-04-25, University of Texas at Austin (2005)Google Scholar
  14. 14.
    Dhillon, I., Modha, D.S.: Concept decompositions for large sparse text data using clustering. Mach. Learn. 42, 143–175 (2001)MATHCrossRefGoogle Scholar
  15. 15.
    Ding, C., He, X., Simon, H.D.: On the equivalence of nonnegative matrix factorization and spectral clustering. In: SDM ’05: Proceedings of the SIAM International Conference on Data Mining, pp. 606–610 (2005)Google Scholar
  16. 16.
    Ding, C., Li, T., Jordan, M.: Nonnegative matrix factorization for combinatorial optimization: spectral clustering, graph matching, and clique finding. In: ICDM ’08: Proceedings of the 8th IEEE International Conference on Data Mining, pp. 183–192 (2008)Google Scholar
  17. 17.
    Ding, C., Li, T., Jordan, M.I.: Convex and semi-nonnegative matrix factorization. IEEE Trans. Pattern Anal. Mach. Intell. 32(1), 45–55 (2010)CrossRefGoogle Scholar
  18. 18.
    Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification. Wiley-Interscience, New York (2000)Google Scholar
  19. 19.
    Fowlkes, C., Malik, J.: How Much Does Globalization Help Segmentation? Technical Report UCB/CSD-4-1340. University of California, Berkeley (2004)Google Scholar
  20. 20.
    Gillis, N., Kuang, D., Park, H.: Hierarchical clustering of hyperspectral images using rank-two nonnegative matrix factorization. IEEE Trans. Geosci. Remote Sens. 53(4):2066–2078, (2015)Google Scholar
  21. 21.
    Globerson, A., Chechik, G., Pereira, F., Tishby, N.: Euclidean embedding of co-occurrence data. J. Mach. Learn. Res. 8, 2265–2295 (2007)MATHMathSciNetGoogle Scholar
  22. 22.
    Gonzales, E.F., Zhang, Y.: Accelerating the Lee–Seung Algorithm for Non-Negative Matrix Factorization. Technical Report TR05-02, Rice University (2005)Google Scholar
  23. 23.
    He, Z., Xie, S., Zdunek, R., Zhou, G., Cichocki, A.: Symmetric nonnegative matrix factorization: algorithms and applications to probabilistic clustering. IEEE Trans. Neural Netw. 22(12), 2117–2131 (2011)CrossRefGoogle Scholar
  24. 24.
    Ho, N.D.: Nonnegative Matrix Factorization Algorithms and Applications. Ph.D. thesis, Universite catholique de Louvain (2008)Google Scholar
  25. 25.
    Kannan, R., Ishteva, M., Park, H.: Bounded matrix factorization for recommender system. Knowl. Inf. Syst. 39(3), 491–511 (2014)CrossRefGoogle Scholar
  26. 26.
    Kelley, C.T.: Iterative Methods for Linear and Nonlinear Equations. SIAM, Philadelphia, PA (1995)MATHCrossRefGoogle Scholar
  27. 27.
    Kim, H., Park, H.: Sparse non-negative matrix factorizations via alternating non-negativity-constrained least squares for microarray data analysis. Bioinformatics 23(12), 1495–1502 (2007)CrossRefGoogle Scholar
  28. 28.
    Kim, H., Park, H.: Nonnegative matrix factorization based on alternating non-negativity-constrained least squares and the active set method. SIAM J. Matrix. Anal. Appl. 30(2), 713–730 (2008)MATHMathSciNetCrossRefGoogle Scholar
  29. 29.
    Kim, J., He, Y., Park, H.: Algorithms for nonnegative matrix and tensor factorizations: a unified view based on block coordinate descent framework. J. Glob. Optim. 58(2), 285–319 (2014)MATHMathSciNetCrossRefGoogle Scholar
  30. 30.
    Kim, J., Park, H.: Sparse Nonnegative Matrix Factorization for Clustering. Technical Report GT-CSE-08-01, Georgia Institute of Technology (2008)Google Scholar
  31. 31.
    Kim, J., Park, H.: Toward faster nonnegative matrix factorization: a new algorithm and comparisons. In: ICDM ’08: Proceedings of the 8th IEEE International Conference on Data Mining, pp. 353–362 (2008)Google Scholar
  32. 32.
    Kim, J., Park, H.: Fast nonnegative matrix factorization: An active-set-like method and comparisons. SIAM J. Sci. Comput. 33(6), 3261–3281 (2011)MATHMathSciNetCrossRefGoogle Scholar
  33. 33.
    Kleinberg, J.: An impossibility theorem for clustering. Adv. Neural Inf. Process. Syst. 15, 446–453 (2002)Google Scholar
  34. 34.
    Kuang, D., Ding, C., Park, H.: Symmetric nonnegative matrix factorization for graph clustering. In: SDM ’12: Proceedings of the SIAM International Conference on Data Mining, pp. 106–117 (2012)Google Scholar
  35. 35.
    Kuang, D., Park, H.: Fast rank-2 nonnegative matrix factorization for hierarchical document clustering. In: KDD ’13: Proceedings of the 19th ACM International Conference on Knowledge Discovery and Data Mining, pp. 739–747 (2013)Google Scholar
  36. 36.
    Kulis, B., Basu, S., Dhillon, I., Mooney, R.: Semi-supervised graph clustering: a kernel approach. In: ICML ’05: Proceedings of the 22nd Internatioal Conference on Machine Learning, pp. 457–464 (2005)Google Scholar
  37. 37.
    Lawson, C.L., Hanson, R.J.: Solving Least Squares Problems. Prentice-Hall, Englewood Cliffs, NJ (1974)MATHGoogle Scholar
  38. 38.
    Lee, D.D., Seung, H.S.: Learning the parts of objects by non-negative matrix factorization. Nature 401, 788–791 (1999)CrossRefGoogle Scholar
  39. 39.
    Lee, K.C., Ho, J., Kriegman, D.: Acquiring linear subspaces for face recognition under variable lighting. IEEE Trans. Pattern Anal. Mach. Intell. 27(5), 684–698 (2005)CrossRefGoogle Scholar
  40. 40.
    Lewis, D.D., Yang, Y., Rose, T.G., Li, F.: Rcv1: A new benchmark collection for text categorization research. J. Mach. Learn. Res. 5, 361–397 (2004)Google Scholar
  41. 41.
    Li, T., Ding, C., Jordan, M.I.: Solving consensus and semi-supervised clustering problems using nonnegative matrix factorization. In: ICDM ’07: Proceedings of the 7th IEEE International Conference on Data Mining, pp. 577–582 (2007)Google Scholar
  42. 42.
    Lin, C.J.: On the convergence of multiplicative update algorithms for nonnegative matrix factorization. Trans. Neural Netw. 18(6), 1589–1596 (2007)CrossRefGoogle Scholar
  43. 43.
    Lin, C.J.: Projected gradient methods for nonnegative matrix factorization. Neural Comput. 19(10), 2756–2779 (2007)MATHMathSciNetCrossRefGoogle Scholar
  44. 44.
    Lucena, A., Ribeiro, C., Santos, A.C.: A hybrid heuristic for the diameter constrained minimum spanning tree problem. J. Glob. Optim. 46(3), 363–381 (2010)MATHMathSciNetCrossRefGoogle Scholar
  45. 45.
    Ma, X., Gao, L., Yong, X., Fu, L.: Semi-supervised clustering algorithm for community structure detection in complex networks. Phys. A Stat. Mech. Appl. 389(1), 187–197 (2010)CrossRefGoogle Scholar
  46. 46.
    Malik, J., Belongie, S., Leung, T., Shi, J.: Contour and texture analysis for image segmentation. Int. J. Comput. Vis. 43(1), 7–27 (2001)MATHCrossRefGoogle Scholar
  47. 47.
    Manning, C.D., Raghavan, P., Schütze, H.: Introduction to Information Retrieval. Cambridge University Press, New York, NY (2008)MATHCrossRefGoogle Scholar
  48. 48.
    Martin, D., Fowlkes, C., Malik, J.: Learning to detect natural image boundaries using local brightness, color, and texture cues. IEEE Trans. Pattern Anal. Mach. Intell. 26(5), 530–549 (2004)CrossRefGoogle Scholar
  49. 49.
    Martin, D., Fowlkes, C., Tal, D., Malik, J.: A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. In: ICCV ’01: Proceedings of the 8th IEEE International Conference on Computer Vision, vol. 2, pp. 416–423 (2001)Google Scholar
  50. 50.
    Nepusz, T., Petróczi, A., Négyessy, L., Bazsó, F.: Fuzzy communities and the concept of bridgeness in complex networks. Phys. Rev. E 77(1), 016,107 (2008)Google Scholar
  51. 51.
    Ng, A.Y., Jordan, M.I., Weiss, Y.: On spectral clustering: analysis and an algorithm. Adv. Neural Inf. Process. Syst. 14, 849–856 (2001)Google Scholar
  52. 52.
    Pauca, V.P., Shahnaz, F., Berry, M.W., Plemmons, R.J.: Text mining using non-negative matrix factorizations. In: SDM ’04: Proceedings of the SIAM International Conference on Data Mining, pp. 452–456 (2004)Google Scholar
  53. 53.
    Rinnooy Kan, A., Timmer, G.: Stochastic global optimization methods, part II: multi level methods. Math. Program. 39, 57–78 (1987)MATHMathSciNetCrossRefGoogle Scholar
  54. 54.
    Rinnooy Kan, A., Timmer, G.: Global optimization. In: Kan, R., Todds, M.J. (eds.) Handbooks in Operations Research and Management Science, vol. 1, pp. 631–662. North Holland, Amsterdam (1989)Google Scholar
  55. 55.
    Shahnaz, F., Berry, M.W., Pauca, V.P., Plemmons, R.J.: Document clustering using nonnegative matrix factorization. Inf. Process. Manag. 42, 373–386 (2006)MATHCrossRefGoogle Scholar
  56. 56.
    Shi, J., Malik, J.: Normalized cuts and image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 22(8), 888–905 (2000)CrossRefGoogle Scholar
  57. 57.
    Sim, T., Baker, S., Bsat, M.: The CMU pose, illumination, and expression database. IEEE Trans. Pattern Anal. Mach. Intell. 25(12), 1615–1618 (2003)CrossRefGoogle Scholar
  58. 58.
    Stewart, G.W., Sun, J.G.: Matrix Perturbation Theory. Academic Press, New York (1990)MATHGoogle Scholar
  59. 59.
    Strogatz, S.H.: Exploring complex networks. Nature 410, 268–276 (2001)CrossRefGoogle Scholar
  60. 60.
    van der Maaten, L., Hinton, G.: Visualizing data using t-SNE. J. Mach. Learn. Res. 9, 2579–2605 (2008)MATHGoogle Scholar
  61. 61.
    von Luxburg, U.: A tutorial on spectral clustering. Stat. Comput. 17(4), 395–416 (2007)MathSciNetCrossRefGoogle Scholar
  62. 62.
    Wang, F., Li, T., Wang, X., Zhu, S., Ding, C.: Community discovery using nonnegative matrix factorization. Data Min. Knowl. Discov. 22(3), 493–521 (2011)MATHMathSciNetCrossRefGoogle Scholar
  63. 63.
    Wild, S., Curry, J., Dougherty, A.: Improving non-negative matrix factorizations through structured initialization. Pattern Recognit. 37, 2217–2232 (2004)CrossRefGoogle Scholar
  64. 64.
    Xu, W., Liu, X., Gong, Y.: Document clustering based on non-negative matrix factorization. In: SIGIR ’03: Proceedings of the 26th International ACM Conference on Research and Development in Informaion Retrieval, pp. 267–273 (2003)Google Scholar
  65. 65.
    Yang, Z., Hao, T., Dikmen, O., Chen, X., Oja, E.: Clustering by nonnegative matrix factorization using graph random walk. Adv. Neural Inf. Process. Syst. 25, 1088–1096 (2012)Google Scholar
  66. 66.
    Yang, Z., Oja, E.: Clustering by low-rank doubly stochastic matrix decomposition. In: ICML ’12: Proceedings of the 29nd International Conference on Machine Learning (2012)Google Scholar
  67. 67.
    Yu, S.X., Shi, J.: Multiclass spectral clustering. In: ICCV ’03: Proceedings of the 9th IEEE International Conference on Computer Vision, pp. 313–319 (2003)Google Scholar
  68. 68.
    Yun, S., Tseng, P., Toh, K.C.: A block coordinate gradient descent method for regularized convex separable optimization and covariance selection. Math. Program. 129, 331–355 (2011)MATHMathSciNetCrossRefGoogle Scholar
  69. 69.
    Zass, R., Shashua, A.: A unifying approach to hard and probabilistic clustering. In: ICCV ’05: Proceedings of the 10th IEEE International Conference on Computer Vision, pp. 294–301 (2005)Google Scholar
  70. 70.
    Zelnik-Manor, L., Perona, P.: Self-tuning spectral clustering. Adv. Neural Inf. Process. Syst. 17, 1601–1608 (2004)Google Scholar
  71. 71.
    Zhang, Y., Yeung, D.Y.: Overlapping community detection via bounded nonnegative matrix tri-factorization. In: KDD ’12: Proceedings of the 18th ACM International Conference on Knowledge Discovery and Data Mining, pp. 606–614 (2012)Google Scholar
  72. 72.
    Zhang, Z.Y., Wang, Y., Ahn, Y.Y.: Overlapping community detection in complex networks using symmetric binary matrix factorization. Phys. Rev. E 87(6), 062,803 (2013)Google Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  1. 1.School of Computational Science and EngineeringGeorgia Institute of TechnologyAtlantaUSA
  2. 2.Department of Mathematics EducationSungkyunkwan UniversitySeoulKorea
  3. 3.Korea Institute for Advanced StudySeoulKorea

Personalised recommendations