DocumentCode :
3610459
Title :
Graph clustering by congruency approximation
Author :
Weiya Ren ; Guohui Li ; Dan Tu
Author_Institution :
Nat. Univ. of Defense Technol., Changsha, China
Volume :
9
Issue :
6
fYear :
2015
Firstpage :
841
Lastpage :
849
Abstract :
The authors consider the general problem of graph clustering. Graph clustering manipulates the graph-based data structure and the entries of the solution vectors are only allowed to take non-negative discrete values. Finding the optimal solution is NP-hard, so relaxations are usually considered. Spectral clustering retains the orthogonality rigorously but ignores the non-negativity and discreteness of the solution. Sym non-negative matrix factorisation can retain the non-negativity rigorously but it is hard to reach the orthogonality. In this study, they proposed a novel method named congruent approximate graph clustering (CAC), which can retain the non-negativity rigorously and can reach the orthogonality properly by congruency approximation. Furthermore, the solution obtained by CAC is sparse, which is approximate with the ideal discrete solution. Experimental results on several real image benchmark datasets indicate that CAC achieves encouraging results compared with state-of-the-art methods.
Keywords :
computational complexity; graph theory; matrix decomposition; pattern clustering; CAC; NP-hard optimal solution; congruency approximation; congruent approximate graph clustering; graph-based data structure; ideal discrete solution; nonnegative discrete values; nonnegative matrix factorisation; real image benchmark datasets; solution vectors; spectral clustering;
fLanguage :
English
Journal_Title :
Computer Vision, IET
Publisher :
iet
ISSN :
1751-9632
Type :
jour
DOI :
10.1049/iet-cvi.2014.0131
Filename :
7328509
Link To Document :
بازگشت