Local and global regularized sparse coding for data representation

NEUROCOMPUTING

卷:175

Article

Shu, Zhenqiu¹  Zhou, Jun²  Huang, Pu³  Yu, Xun²  Yang, Zhangjing⁴  Zhao, Chunxia⁵ 

[1] Jiangsu Univ Technol, Sch Comp Engn, Changzhou 213001, Peoples R China

[2] Griffith Univ, Sch Informat & Commun Technol, Nathan, Qld 4111, Australia

[3] Nanjing Univ Posts & Telecommun, Sch Comp Sci & Technol, Nanjing 210023, Jiangsu, Peoples R China

[4] Nanjing Audit Univ, Sch Technol, Nanjing 211815, Jiangsu, Peoples R China

[5] Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Nanjing 210094, Jiangsu, Peoples R China

关键词: Sparse coding;    Data representation;    Regularizer;    Regression;    Clustering;   

DOI  :  10.1016/j.neucom.2015.10.048

来源: Elsevier

【 摘 要 】

Recently, sparse coding has been widely adopted for data representation in real-world applications. In order to consider the geometric structure of data, we propose a novel method, local and global regularized sparse coding (LGSC), for data representation. LGSC not only models the global geometric structure by a global regression regularizer, but also takes into account the manifold structure using a local regression regularizer. Compared with traditional sparse coding methods, the proposed method can preserve both global and local geometric structures of the original high-dimensional data in a new representation space. Experimental results on benchmark datasets show that the proposed method can improve the performance of clustering. (C) 2015 Elsevier B.V. All rights reserved.

【 授权许可】

Free