Abstract
In this paper, we present an augmented Lagrangian alternating direction algorithm for symmetric nonnegative matrix factorization. The convergence of the algorithm is also proved in detail and strictly. Then we present a modified overlapping community detection method which is based on the presented symmetric nonnegative matrix factorization algorithm. We apply the modified community detection method to several real world networks. The obtained results show the capability of our method in detecting overlapping communities, hubs and outliers. We find that our experimental results have better quality than several competing methods for identifying communities.
Similar content being viewed by others
References
Lee DD, Seung HS (1999) Learning the parts of objects by non-negative matrix factorization. Nature 401(6755):788–791
Lin L (2010) Alternative gradient algorithms with applications to nonnegative matrix factorizations. Appl Math Comput 216:1763–1770
Lin L, Liu ZY (2011) An alternationg projected gradient algorithm for nonnegative matrix factorization. Appl Math Comput 217:9997–10002
Hoyer PO (2004) Non-negative matrix factorization with sparseness constraints. J Mach Learn Res 5(1):1457–1469
Berry M, Browne M, Langville A, Pauca P, Plemmons R (2007) Algorithms and applications for approximate nonnegative matrix factorization. Comput Stat Data Anal 52:155–173
Pauca P, Shahnaz F, Berry M, Plemmons R (2004) Text mining using non-negative matrix factorizations. In: Proceedings of the 2004 SIAM international conference on data mining, pp 452–456
Saul LK, Sha F, Lee DD (2003) Statistical signal processing with nonnegativity constraints. Proc EuroSpeech 2:1001–1004
Shahnaz F, Berry MW, Pauca VP, Plemmons R (2006) Document clusting using nonnegative matrix factorization. Inform Process Manag 42:373–386
Pauca VP, Piper J, Plemmons RJ (2006) Nonnegative matrix factorization for spectral data analysis. Linear Algebra Appl 416:29–47
Smaragdis P, Brown JC (2003) Non-negative matrix factorization for polyphonic music transcription. In: 2003 IEEE workshop on applications of signal processing to audio and acoustics, New Paltz, NY, 19–22 October 2003, pp 177–180
Zarei M, Izadi D, Samani KA (2009) Detecting overlapping community structure of networks based on vertex-vertex correlations. J Stat Mech Theory Exp 2009(11):209–222
Psorakis I, Roberts S, Ebden M, Sheldon B (2011) Overlapping community detection using bayesian non-negative matrix factorization. Phys Rev E 83(6):066114
Wang F, Li T, Wang X, Zhu SH, Ding C (2011) Community discovery using nonnegative matrix factorization. Data Mining Knowl Discov 22(3):493–521
Cao XC, Wang X, Jin D, Cao Y, He DX (2014) Identifying overlapping communities as well as hubs and outliers via nonnegative matrix factorization. Sci Rep 3(10):2993–2993
Hestenes MR (1969) Multiplier and gradient methods. J Optim Theory Appl 4(5):303–320
Powell MJD (1969) A method for nonlinear constraints in minimization problems. Optimization 5(6):283–298
Rockafellar RT (1974) Augmented Lagrange multiplier functions and duality in nonconvex programming. SIAM J Control Optim 12(2):268–285
Bertsekas DP (1982) Constrained optimization and Lagrangian multiplier methods. Academic Press, New York
Huang XX, Yang XQ (2001) Approximate optimal solutions and nonlinear Lagrangian functions. J Global Optim 21:51–65
Huang XX, Yang XQ (2005) Further study on augmented Lagrangian duality theory. J Global Optim 31:193–210
Polak E, Royset JO (2005) On the use of augmented Lagrangians in the solution of gengeralized semi-infinite min-max problems. Comput Optim Appl 31:173–192
Nocedal J, Wright SJ (2006) Numerical optimization. Springer series in operations research and financial engineering. Springer, New York
Pu DG, Yang P (2013) A class of new Lagrangian multiplier methods. In: 2013 sixth international conference on business intelligence and financial engineering, pp 647–651
Liu BZ, Wang CY (2007) Zero duality gap properties for a class of Lagrangian dual problem and the convergence of its optimal path. OR Trans 11(1):73–84
He B, Yang H, Wang S (2000) Alternating direction method with self-adaptive penalty parameters for monotone variational inequalities. J Optim Theory Appl 106(2):337–356
He B, Liao L, Han D, Yang H (2002) A new inexact alternating directions method for monotone variational inequalities. Math Program 92(1, Ser. A):103–118
Wen Z, Goldfarb D, Yin W (2009) Alternating direction augmented lagrangian methods for semidefinite programming. Technical report, Dept of IEOR, Columbia University
Tseng P (1997) Alternating projection-proximal methods for convex programming and variational inequalities. SIAM J Optim 7(4):951–965
Kontogiorgis S, Meyer RR (1998) A variable-penalty alternating directions method for convex optimization. Math Program 83(1,Ser. A):29–53
Yuan X, Yang J (2009) Sparse and low rank matrix decomposition via alternating direction method. Pac J Optim 9(1):1–11
Lin Z, Chen M, Wu L, Ma Y (2010) The augmented lagrange multiplier method for exact recovery of corrupted low-rank matrices. Eprint Arxiv 9:1–20
Shen Y, Wen Z, Zhang Y (2012) Augmented Lagrangian alternating direction method for matrix separation based on low-rank factorization. Optim Method Softw 29(2):1–25
Zhang Y (2010) An alternating direction algorithm for nonnegative matrix factorization. Technical report, Rice University. http://www.caam.rice.edu/~yzhang/reports/tr1003.pdf
Wang D, Li T, Zhu S, Ding C (2008) Multi-document summarization via sentence-level semantic analysis and symmetric matrix factorization. In: Proceedings of the 31st annual international ACM SIGIR conference on Research and development in information retrieval 5(2):307–314
Zachary WW (1977) An information flow model for conflict and fission in small groups. J Anthropol Res 33:452–473
Lusseau D (2003) The emergent properties of dolphin social network. Proc Biol Sci 270(Suppl 2):186–188
Newman MEJ (2006) Modularity and community structure in networks. Proc Ntl Acad Sci 103(23):8577–8582
Newman MEJ (2006) Finding community structure in networks using the eigenvectors of matrices. Phys Rev E Stat Nonlinear Soft Matter Phys 74(3 Pt 2):92–100
Watts DJ, Strogatz SH (1998) Collective dynamics of ’small-world’ networks. Nature 393:440–442
Girvan M, Newman MEJ (2001) Community structure in social and biological networks. Proc Natl Acad Sci USA 99(12):7821–7826
Leskovec J, Lang KJ, Mahoney M (2010) Empirical comparison of algorithms for network community detection. In: Proceedings of the 19th international conference on World wide web, Raleigh, North Carolina, USA, 26–30 April 2010. ACM, New York, pp 631–640
Blondel V, Guillaume J, Lambiotte R, Lefebvre E (2008) Fast unfolding of communities in large networks. J Stat Mech Theory Exp 2008(10):155–168
Palla G, Derenyi I, Farkas I, Vicsek T (2005) Uncovering the overlapping community structure of complex networks in nature and society. Nature 435(7043):814–818
Zhang Y, Yeung D (2012) Overlapping community detection via bounded nonnegative matrix tri-factorization. In: Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining, pp 606–614
Lu S, Hong M, Wang Z (2017) A nonconvex splitting method for symmetric nonnegative matrix factorization: convergence analysis and optimality. IEEE Int Confer Acoust 99:1–1
Borhani R, Watt J, Katsaggelos A (2016) Fast and effective algorithms for symmetric nonnegative matrix factorization. Cornell University Library. arXiv:1609.05342
Acknowledgements
The authors are grateful to Prof. Z. Peng of Fuzhou University for his guidance and suggestions on optimization theory in this work. The authors thank the editors and the anonymous reviewers for their insightful and constructive comments. This work is supported by National Natural Science Foundation of China (Nos. U1805263, 11672074, 61772134), Natural Science Foundation of Fujian Province (Nos. 2018J01775, 2018J01776) and Fujian Science and Technology Department Project (No. JK2017007).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Hu, L., Guo, G. An augmented Lagrangian alternating direction method for overlapping community detection based on symmetric nonnegative matrix factorization. Int. J. Mach. Learn. & Cyber. 11, 403–415 (2020). https://doi.org/10.1007/s13042-019-00980-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13042-019-00980-z