点击次数：

论文类型： 期刊论文

发表时间： 2011-10-01

发表刊物： NEUROCOMPUTING

收录刊物： Scopus、SCIE、EI

卷号： 74

期号： 17

页面范围： 3598-3608

ISSN号： 0925-2312

关键字： Semi-supervised learning; Subspace clustering; Constraint inconsistence

摘要： Selecting correct dimensions is very important to subspace clustering and is a challenging issue. This paper studies semi-supervised approach to the problem. In this setting, limited domain knowledge in the form of space level pair-wise constraints, i.e., must-links and cannot-links, are available. We propose a semi-supervised subspace clustering (S(3)C) algorithm that exploits constraint inconsistence for dimension selection. Our algorithm firstly correlates globally inconsistent constraints to dimensions in which they are consistent, then unites constraints with common correlating dimensions, and finally forms the subspaces according to the constraint unions. Experimental results show that S(3)C is superior to the typical unsupervised subspace clustering algorithm FINDIT, and the other constraint based semi-supervised subspace clustering algorithm SC-MINER. (C) 2011 Elsevier B.V. All rights reserved.