基于2,1范数的超图正则化非负矩阵分解
Hypergraph Regular Nonnegative Matrix Factorization Based on 2,1 Norms
DOI: 10.12677/AAM.2023.121048, PDF,   
作者: 文学春：贵州师范大学数学科学学院，贵州 贵阳
关键词: 2；1范数；超图学习；聚类分析；非负矩阵分解；2；1 Norm； Hypergraph Learning； Cluster Analysis； Nonnegative Matrix Decomposition
摘要: 图学习的方法只考虑了样本间的关系，忽略了样本间的高维结构，基于欧氏距离的非负矩阵模型易受到噪声和异常值的影响。为解决以上问题，给出一种基于L2,1范数的超图正则化非负矩阵分解(L2,1HNMFL)方法，并给出了算法的更新规则。通过在COIL20和ORL经典数据集上与其他算法的比较验证了该算法的有效性。
Abstract: The graph learning method only considers the relationship between samples, but ignores the high-dimensional structure between samples. The non-negative matrix model based on Euclidean distance is easily affected by noise and outliers. To solve the above problems, a hypergraph regu-larization nonnegative matrix factorization (L2,1HNMFL) method based on L2,1 norm is presented, and the updating rules of the algorithm are given. The effectiveness of this algorithm is verified by comparing it with other algorithms on classical data sets COIL20 and ORL.
文章引用：文学春. 基于2,1范数的超图正则化非负矩阵分解[J]. 应用数学进展, 2023, 12(1): 451-458. https://doi.org/10.12677/AAM.2023.121048

参考文献

[1] Fukunada, K. (1990) Introduction to Statistical Pattern Recognition. In: Rheinboldt, W. and Siewiorek, D., Eds., Com-puter Science and Scientific Computing, Academic Press, Cambridge.
[2] Kalman, D. (1996) A Singularly Valuable Decomposition: The SVD of a Matrix. The College Mathematics Journal, 27, 2-23. [Google Scholar] [CrossRef] 
[3] Xu, W. and Gong, Y.H. (2004) Document Clustering by Concept Factorization. Research and Development in Information Retrieval.
[4] Lee, D.D. and Seung, H.S. (1999) Learning the Parts of Objects by Non-Negative Matrix Factorization. Nature, 401, 788-791. [Google Scholar] [CrossRef] [PubMed]
[5] Li, S.Z., Hou, X.W., Zhang, H.J., et al. (2001) Learning Spatially Localized, Parts-Based Representation. Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pat-tern Recognition, 1, I-I.
[6] Chen, Y., Zhang, J., Cai, D., et al. (2012) Nonnegative Local Coordinate Factorization for Image Representation. IEEE Transactions on Image Processing, 22, 969-979. [Google Scholar] [CrossRef] 
[7] Deng, C., et al. (2010) Graph Regularized Nonnegative Matrix Factorization for Data Representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33, 1548-1560. [Google Scholar] [CrossRef] 
[8] Zeng, K., Yu, J., Li, C., et al. (2014) Image Clustering by Hy-per-Graph Regularized Non-Negative Matrix Factorization. Neurocomputing, 138, 209-217. [Google Scholar] [CrossRef] 
[9] Kong, D.G., Ding, C. and Huang, H. (2011) Robust Nonnega-tive Matrix Factorization Using L21-Norm. Information and Knowledge Management.
[10] Zhou, D., Huang, J. and Schölkopf, B. (2006) Learning with Hypergraphs: Clustering, Classification, and Embedding. Advances in Neural In-formation Processing Systems, 19.
[11] Yu, K., Zhang, T. and Gong, Y. (2009) Nonlinear Learning Uusing Local Co-ordinate Coding. Advances in Neural Information Processing Systems, 22.
[12] Xu, W., Liu, X. and Gong, Y.H. (2003) Document Clustering Based on Non-Negative Matrix Factorization. Research and Development in Informaion Retriev-al.

为你推荐