一种多流形正则化的多视图非负矩阵分解算法

南京大学学报(自然科学版) ›› 2017, Vol. 53 ›› Issue (3): 557–.

一种多流形正则化的多视图非负矩阵分解算法

宗林林，张宪超*，赵乾利，于　红，刘馨月

出版日期:2017-05-30 发布日期:2017-05-30
作者简介: 大连理工大学软件学院，大连，116620
基金资助:
基金项目：国家自然科学基金(61272374，61300190，61428202)，国家高技术研究发展计划(863计划)(2015AA015403)
收稿日期：2016－12－18
*通讯联系人，Email：xczhang@dlut.edu.cn

A multimanifold regularized multiview nonnegative matrix factorization algorithm

Zong Linlin，Zhang Xianchao*，Zhao Qianli，Yu Hong，Liu Xinyue

Online:2017-05-30 Published:2017-05-30
About author: School of Software，Dalian University of Technology，Dalian，116620，China

摘要/Abstract

摘要： 在大数据时代，多视图数据普遍存在．多视图聚类是分析多视图数据的一种常用方法．基于多流形正则化非负矩阵分解的多视图聚类是一种极具竞争力的多视图聚类算法，但该算法没有考虑非负矩阵分解的簇排列问题，并且在实验中没有考虑每个视图的差异性．基于上述问题，提出一种优化的多流形正则化的多视图非负矩阵分解算法．该算法的关键问题包括如何利用多视图信息聚类以及如何融合多流形．对多视图数据聚类时，令所有视图的数据共享一个低维的子矩阵，并且最小化所有视图的加权目标函数，从而体现每个视图对聚类的重要性并确保所有非负矩阵分解的簇排列的一致性．在融合多流形信息时，使用基于多视图谱聚类的权重计算方法，加权寻找一致的流形，从而体现每个视图中流形的重要性．实验结果表明，提出的优化策略可以提高多视图聚类的效果．

Abstract: In the era of big data，data often comes from multiple feature extractors or consists of multiple views.Multiview clustering is one of the common approaches for the analysis of such data，which separates data into several groups using information from multiple views.Multiview clustering via multimanifold regularized nonnegative matrix factorization has become one of the most modern multiview clustering algorithms in the past decade.However，they do not consider the cluster permutation in nonnegative matrix factorization，and they equally treat each view in the experiment.Based on the above issues，in this paper，an improved multimanifold regularized multiview nonnegative matrix factorization algorithm has been proposed.The key problems of this algorithm are the way of clustering multiview data and the integration of multimanifold.In the process of clustering multiview data，the multiview data share the same low dimensional submatrix and the weighted objective function of all views is minimized，which indicates the importance of each view in clustering and ensures the consistency of cluster permutations in nonnegative matrix factorization.In the process of integrating multiple manifolds，the consensus manifold is approximated by the weighted linear combination of multiple manifolds.To show the importance of each view’s manifold，the weighting schema based on multiview spectral clustering is adopted to find the consensus manifold.To show the effectiveness of the proposed strategy，we experiment on several benchmark datasets.Experimental results show that the proposed algorithm outperforms the state of the art and the proposed optimization strategies are effective for multiview clustering.

宗林林，张宪超*，赵乾利，于　红，刘馨月. 一种多流形正则化的多视图非负矩阵分解算法[J]. 南京大学学报(自然科学版), 2017, 53(3): 557–.

Zong Linlin，Zhang Xianchao*，Zhao Qianli，Yu Hong，Liu Xinyue. A multimanifold regularized multiview nonnegative matrix factorization algorithm[J]. Journal of Nanjing University(Natural Sciences), 2017, 53(3): 557–.

参考文献

[1]　Xu C，Tao D，Xu C．A Survey on multiview learning.Computer Science，arXiv：1304.5634，2013.　
[2] Bickel S，Scheffer T．Multiview clustering.In：IEEE International Conference on Data Mining.Brighton，UK：IEEE，2004：19－26.
[3] Zhang X，Zong L，Liu X，et al.Constrained NMFbased multiview clustering on unmapped data.In：Proceedings of the 29th AAAI Conference on Artificial Intelligence.Austin，USA：AAAI，2015：3174－3180.
[4] Li Y，Nie F，Huang H，et al.Largescale multiview spectral clustering via bipartite graph.In：Proceedings of the 29th AAAI Conference on Artificial Intelligence.Austin，USA：AAAI，2015：2750－2756.
[5] Xu C，Tao D，Xu C．Multiview selfpaced learning for clustering.In：Proceedings of the 24th International Joint Conference on Artificial Intelligence.Buenos Aires，Argentina：AAAI，2015：3974－3980.
[6] Cai X，Nie F P，Huang H．Multiview kmeans clustering on big data.In：Proceedings of the 23rd International Joint Conference on Artificial Intelligence.Beijing，China：AAAI，2013：2598－2604.　
[7] Gao J，Han J W，Liu J L，et al.Multiview clustering via joint nonnegative matrix factorization.In：Proceedings of the 13th SIAM International Conference on Data Mining.Austin，USA：SIAM，2013：252－260.
[8] Singh A P，Gordon G J．Relational learning via collective matrix factorization.In：ACM SIGKDD Conference on Knowledge Discovery and Data Mining.Las Vegas，USA：ACM，2008：650－658.
[9] Cai D，He X，Han J，et al.Graph regularized nonnegative matrix factorization for data representation.IEEE Transactions on Pattern Analysis and Machine Intelligence，2011，33(8)：1548－1560.
[10] Zhang X，Zhao L，Zong L，et al.Multiview clustering via multimanifold regularized nonnegative matrix factorization.In：IEEE International Conference on Data Mining.Shenzhen，China：IEEE，2014，1103－1108.
[11] Tzortzis G，Likas A．Kernelbased weighted multiview clustering.In：IEEE International Conference on Data Mining.Brussels，Belgium：IEEE，2012：675－684.
[12] Lee D，Seung H．Learning the parts of objects by nonnegative matrix factorization.Nature，1999，401(6755)：788－791.
[13] Lee D，Seung H．Algorithms for nonnegative matrix factorization.In：Advances in Neural Information Processing Systems.San Francisco，USA：Morgan Kaufmann Publishers，2001：556－562.　
[14] Li T，Ding C．The relationships among various nonnegative matrix factorization methods for clustering.In：Proceedings of the 6th International Conference on Data Mining.Hong Kong，China：IEEE，2006：362－371.
[15] Luxburg U．A tutorial on spectral clustering.Statistics & Computing，2007，17(4)：395－416.
[16] Huang H C，Chuang Y Y，Chen C S．Affinity aggregation for spectral clustering.In：IEEE Conference on Computer Vision and Pattern Recognition.Providence，USA：IEEE，2012：773－780.　
[17] Boyd S，Vandenberghe L．Convex optimization.Cambridge：Cambridge University Press，2004，727.　
[18] Cai D，He X，Han J．Document clustering using locality preserving indexing.IEEE Transactions on Knowledge and Data Engineering，2005，17(12)：1624－1637.
[19] Zhang X，You Q．Clusterability analysis and incremental sampling for nystrom extension based spectral clustering.In：IEEE International Conference on Data Mining.Vancouver，Canada：IEEE，2011：942－951.
[20] Hubert L，Arabie P．Comparing partitions.Journal of Classification，1985，2：193－218.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed