Sciweavers

Share
ICPR
2008
IEEE

Multiclass spectral clustering based on discriminant analysis

10 years 4 months ago
Multiclass spectral clustering based on discriminant analysis
Many existing spectral clustering algorithms share a conventional graph partitioning criterion: normalized cuts (NC). However, one problem with NC is that it poorly captures the graph's local marginal information which is very important to graph-based clustering. In this paper, we present a discriminant analysis based graph partitioning criterion (DAC), which is designed to effectively capture the graph's local marginal information characterized by the intra-class compactness and the inter-class separability. DAC preserves the intrinsic topological structures of the similarity graph on data points by constructing a k-nearest neighboring subgraph for each data point. Consequently, the clustering results generated by the DAC-based clustering algorithm (DACA) are robust to the outlier disturbance. Theoretic analysis and experimental evaluations demonstrate the promise and effectiveness of DACA.
Xi Li, Zhongfei Zhang, Yanguo Wang, Weiming Hu
Added 05 Nov 2009
Updated 06 Nov 2009
Type Conference
Year 2008
Where ICPR
Authors Xi Li, Zhongfei Zhang, Yanguo Wang, Weiming Hu
Comments (0)
books