Skip navigation
Run Run Shaw Library City University of Hong KongRun Run Shaw Library

Please use this identifier to cite or link to this item: http://dspace.cityu.edu.hk/handle/2031/7110
Title: Binary- and multi-class group sparse canonical correlation analysis for discriminant feature extraction and classification
Authors: Zhang, Zhao (張召)
Department: Department of Electronic Engineering
Issue Date: 2013
Award: Zhang Zhao won the 2nd Prize (Postgraduate Section) in the IEEE HK Section Student Paper Contest 2012.
Subjects: canonical correlation analysis
group sparse representation
group sparse representation
NP-hard minimization
feature extraction
Type: Research Paper
Abstract: This paper incorporates the sparse representation into the well-known Canonical Correlation Analysis (CCA) framework and proposes a novel discriminant feature extraction technique named Group Sparse Canonical Correlation Analysis (GSCCA). GSCCA uses two sets of variables and aims to keep the group sparse (GS) characteristics of data within each set in addition to maximizing the global inter-set covariance. With GS weights computed prior to feature extraction, the locality, sparsity and discriminant information of data can be adaptively determined. The GS weights are calculated from a NP-hard group-sparsity promoting problem that considers all highly correlated data within a group. By defining one of the two variable sets as the class label matrix, GSCCA is effectively extended to multi-class case. Then GSCCA is theoretically formulated as a least squares problem. Comparative analysis between this work and other related work demonstrate that our approach is more general exhibiting attractive properties. The GSCCA projection matrix is analytically solved by using eigen-decomposition and trace ratio (TR) optimization. Extensive benchmark simulations are conducted to examine GSCCA. Results show that our approach delivers promising results, compared with other related algorithms.
Appears in Collections:Student Works With External Awards 

Files in This Item:
File SizeFormat 
award_winning_work.html161 BHTMLView/Open
Show full item record


Items in Digital CityU Collections are protected by copyright, with all rights reserved, unless otherwise indicated.

Send feedback to Library Systems
Privacy Policy | Copyright | Disclaimer