计算机科学 ›› 2016, Vol. 43 ›› Issue (7): 77-82.doi: 10.11896/j.issn.1002-137X.2016.07.013

• 2015年第二十四届全国多媒体学术会议 • 上一篇    下一篇

基于图正则化和稀疏约束的半监督非负矩阵分解

姜小燕,孙福明,李豪杰   

  1. 辽宁工业大学电子与信息工程学院 锦州121001,辽宁工业大学电子与信息工程学院 锦州121001,大连理工大学软件学院 大连116300
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受国家自然科学基金(61572244,61472059),辽宁省高等学校优秀人才支持计划(LR2015030)资助

Semi-supervised Nonnegative Matrix Factorization Based on Graph Regularization and Sparseness Constraints

JIANG Xiao-yan, SUN Fu-ming and LI Hao-jie   

  • Online:2018-12-01 Published:2018-12-01

摘要: 非负矩阵分解是在矩阵非负约束下的分解算法。为了提高识别率,提出了一种基于稀疏约束和图正则化的半监督非负矩阵分解方法。该方法对样本数据进行低维非负分解时,既保持数据的几何结构,又利用已知样本的标签信息进行半监督学习,而且对基矩阵施加稀疏性约束,最后将它们整合于单个目标函数中。构造了一个有效的更新算法,并且在理论上证明了该算法的收敛性。在多个人脸数据库上的仿真结果表明,相对于NMF、GNMF、CNMF等算法,GCNMFS具有更好的聚类精度和稀疏性。

关键词: 非负矩阵分解,图正则,稀疏约束,半监督

Abstract: Nonnegative matrix factorization (NMF) is a kind of matrix factorization algorithm under non-negative constraints .With the aim to enhance the recognition rate,a method called graph regularized and constrained non-negative matrix factorization with sparseness (GCNMFS) was proposed.It not only preserves the intrinsic geometry of data,but also uses the label information for semi-supervised learning and introduces sparseness constraint into base matrix.Finally,they are integrated into a single objective function.An efficient updating approach was produced and the convergence of this algorithm was also proved.Compared with NMF,GNMF and CNMF,experiments on some face databases show that the proposed method can achieve better clustering results and sparseness.

Key words: Nonnegative matrix factorization,Graph regularization,Sparseness constraints,Semi-supervised

[1] Paatero P,Tapper U.Positive Matrix Factorization:A Nonnegative Factor Model with Optimal Utilization of Error Estimates of Data Values [J].Environmetrices,1994,5(2):111-126
[2] Lee D D,Seung H S.Learning the parts of objects by non-nega-tive matrix factori-zation [J].Nature,1999,401(6755):788-791
[3] Lee D D,Seung H S.Algorithms for non-negative matrix facto-rization [J].Advances in Neural Information Processing Systems,2001,13:556-562
[4] Hoyer P O.Non-negative sparse coding[C]∥Proceedings ofIEEE Workshop on Neural Networks for Signal Processing.Martigny,Switzerland,2002:557-565
[5] Hoyer P O.Non-negative matrix factorization with sparseness constrains[J].Journal of Machine Learning Research,2004,5(9):1457-1469
[6] Wang Y,Jia Y.Fisher non-negative matrix factorization forlearning local feature[C]∥Proceedings of Asian Conf.on Comp.Vision.2004:27-30
[7] Guillamet D,Vitria J,Schiele B.Introducting a weighted non-negative matrix factorization for image classification [J].Pattern Recognition,2003,24(14):2447-2454
[8] Ding C,Li T,Jordan M.Convex and semi-nonnegative matrix factorizations[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2010,32(1):45-55
[9] Cai Deng,He Xiao-fei,Han Jia-wei,et al.Graph regularized non-negative matrix factorization for data representation[J].IEEE Trans on Pattern Anal Mach Intell,2011,33(8):1548-1560
[10] Jiang Wei,Li Hong,Yu Zhen-guo,et al.Graph regularized Non-negative Matrix Factorization with Sparseness Constraints.[J].Computer Science,2013,1(40):218-256(in Chinese) 姜伟,李宏,于震国,等.稀疏约束图正则非负矩阵分解[J].计算机科学,2013,1(40):218-256
[11] Liu Hai-feng,Wu Zhao-hui,Li Xue-long,et al.Constrained non-negative matrix factorization for image representation [J].IEEE Trans on Pattern Anal Mach,2012,34(7):1299-1311
[12] Ding C,Li T,Peng W,et al.Orthogonal nonnegative matrix t-factorizations for clustering[C]∥Proceedings of SIGKDD.2006:126-135
[13] Li S Z,Hou X,Zhang H,et al.Learning spatially localized,parts-based representation[C]∥Proc of 2001 IEEE Computer Vision and Pattern Recognition.2001:207-212
[14] Michael W,Shakhina A,Stewart G W.Computing sparse re-duced-rank approximations to sparse matrices [J].ACM Tran-sactions on mathematical software,2004,19(3):231-235
[15] Du Shi-qiang,Shi Yu-qing,Wang Wei-lan,et al.Graph regula-rized-based semi-supervised non-negative Matrix Factorization.[J].Computer Engineering and Applications,2012,48(36):194-200(in Chinese) 杜世强,石玉清,王维兰,等.基于图正则化的半监督非负矩阵分解[J].计算机工程与应用,2012,48(36):194-200
[16] Shu Zhen-qiu,Zhao Chun-xia.Graph-regularized Constrained Non-Negative Matrix Factorization algorithm and its application to image representation[J].Pattern Recognition and Artificial Intelligence,2013,26(3):300-306(in Chinese) 舒振球,赵春霞.基于图正则化的受限非负矩阵分解算法及在图像表示中的应用[J].模式识别与人工智能,2013,6(3):300-306

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!