LIN Hongyan,DU Yuanhua,ZHOU Nan,et al.Non-negative Matrix Factorization Clustering Algorithm based on Multi-view Adaptive Graph Regularization[J].Journal of Chengdu University of Information Technology,2023,38(05):526-534.[doi:10.16836/j.cnki.jcuit.2023.05.006]
基于多视角自适应图正则的非负矩阵分解聚类
- Title:
- Non-negative Matrix Factorization Clustering Algorithm based on Multi-view Adaptive Graph Regularization
- 文章编号:
- 2096-1618(2023)05-0526-09
- Keywords:
- multi-view learning; constraint Laplacian rank; graph embedding; non-negative matrix factorization; clustering
- 分类号:
- TP181
- 文献标志码:
- A
- 摘要:
- 为充分利用各个视角数据内在几何结构关系，提出一种新的基于自适应图正则非负矩阵分解的多视角聚类。该算法在一个统一的框架内，通过各视角亲和矩阵自适应学习提取共识的亲和矩阵进行图嵌入来提取多视角数据共识局部结构信息。另外，通过非负矩阵分解来提取多视角数据全局重构信息。最终使各个视角的共识表达，既保持了数据多视角共识全局重构信息，也保持了数据多视角局部结构信息。该优化问题在考虑了所有数据的一致性和每个视角之间互补性的同时，引入了各个视角数据的局部结构信息，达到数据表达和聚类的效果。通过4组真实数据集的实验，结果表明所提出的方法与已有多视角聚类方法相比具有一定的优越性。
- Abstract:
- In order to make full use of the inherent geometric structure relationship of each view data, this paper proposes a new multi-view clustering method based on adaptive graph regular non-negative matrix factorization. In a unified framework, the algorithm adaptively learns the consensus affinity matrix to extract the consensus local structure information of multi-view data for graph embedding. The global reconstruction information of multi-view data is extracted by non-negative matrix factorization. Finally, the consensus expression of each perspective not only achieves the effect of data expression and clustering. Experiments on four real datasets show that the proposed method is superior to the multi-view clustering method.
参考文献/References:
[1] LIU J,CHI W,JING G,et al.Multi-view clustering via joint nonnegative matrix factorization[EB/OL].http://www.docin.com/p-1774193084.html,2016-11-03.
[2] Sankaran P,Asari V K.A Multi-View Approach on Modular PCA for Illumination and Pose Invariant Face Recognition[C].Applied Imagery Pattern Recognition Workshop.IEEE Computer Society,2004.
[3] Richard H,Ablin P,Hyvärinen A,et al.Adaptive Multi-View ICA:Estimation of noise levels for optimal inference[J].arXiv preprint arXiv:2102.10964,2021.
[4] ZHAN K,Shi J,Jing W,et al.Adaptive Structure Concept Factorization for Multiview Clustering[J].Neural Computation,2018,30(4):1-24.
[5] Nie F,Li J,Li X.Self-weighted Multiview Clustering with Multiple Graphs[J].Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence.2017:2564-2570.
[6] Zhan K,Nie F,Wang J,et al.Multiview consensus graph clustering[J].IEEE Transactions on Image Processing,2018,28(3):1261-1270.
[7] Wen J,Zhang Z,Zhang Z,et al.Generalized Incomplete Multiview Clustering With Flexible Locality Structure Diffusion[J].IEEE Transactions on Cybernetics,2020(99):1-14.
[8] Luo S,Cao X.Multiview Subspace Dual Clustering[J].IEEE Transactions on Neural Networks and Learning Systems,2021:1-13.
[9] Liu B Y,Huang L,Wang C D,et al.Multiview clustering via proximity learning in latent representation space[J].IEEE Transactions on Neural Networks and Learning Systems,2021:1-14.
[10] Nie F,Wang X,Jordan M,et al.The constrained laplacian rank algorithm for graph-based clustering[J].Proceedings of the AAAI conference on artificial intelligence.2016,30(1):1969-1976.
[11] Yan S,Xu D,Zhang B,et al.Graph embedding and extensions:A general framework for dimensionality reduction[J].IEEE transactions on pattern analysis and machine intelligence,2006,29(1):40-51.
[12] Xu Y,Yin W.A Block Coordinate Descent Method for Regularized Multiconvex Optimization with Applications to Nonnegative Tensor Factorization and Completion[J].SIAM Journal on Imaging ences,2015,6(3):1758-1789.
[13] Blum A,Mitchell T.Combining Labeled and Unlabeled Data with Co-Training[C].Proceedings of the 11th Annual Conference on Computational Learning Theory,1998.
[14] Xiao C,Nie F,Huang H,et al.Heterogeneous image feature integration via multi-modal spectral clustering[C].CVPR 2011.IEEE,2011.
[15] Lee D D,Seung H S.Learning the parts of objects by non-negative matrix factorization[J].Nature,1999,401(6755):788-791.
[16] Deng C,He X,Han J,et al.Graph regularized non-negative matrix factorization for data representation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(8):1548-1560.
[17] Wong J A H A.Algorithm AS 136:A K-Means Clustering Algorithm[J].Journal of the Royal Statistical Society,1979,28(1):100-108.
[18] Xu Y,Yin W.A globally convergent algorithm for nonconvex optimization based on block coordinate update[J].Journal of Scientific Computing,2017,72(2):700-734.
[19] Yangang,Xu.Alternating proximal gradient method for sparse nonnegative Tucker decomposition[J].Mathematical Programming Computation:A Publication of the Mathematical Programming Society,2015,7(1):39-70.
[20] Chen Y,Ye X.Projection onto a simplex[J].arXiv preprint arXiv,2011:1101-6081.
[21] Kumar A,Rai P,Daume H.Co-regularized Multi-view Spectral Clustering[C].Neural Information Processing Systems.Curran Associates Inc.2011.
[22] Wang Z,Kong X,Fu H,et al.Feature extraction via multi-view non-negative matrix factorization with local graph regularization[C].IEEE International Conference on Image Processing.IEEE,2015.
[23] Nie F,Li J,Li X.Parameter-free auto-weighted multiple graph learning:a framework for multiview clustering and semi-supervised classification[J]. Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence.2016:1881-1887.
[24] Nie F,Jing L,Li X.Self-weighted multiviewclustering with Multiple Graphs[C].Twenty-Sixth International Joint Conference on Artificial Intelligence.2017.
[25] 李向利,逯喜燕,范学珍.学习一致相似度矩阵的图非负矩阵分解[J].广西大学学报(自然科学版),2022,47(1):262-273.
[26] Shi S,Nie F,Wang R,et al.Self-weighting multi-view spectral clustering based on nuclear norm[J].Pattern Recognition,2022,124:108429.
备注/Memo
收稿日期:2022-08-31
基金项目:国家自然科学基金资助项目(11901063); 四川省自然科学基金资助项目(23NSFSC2919)