This tool performs spectral clustering using either sparse similarity matrix (nearest neighbors) or the Nystrom method. It is also used for comparison experiments shown in the following paper:

• PSC: Parallel Spectral Clustering
  Wen-Yen Chen, Yangqiu Song, Hongjie Bai, Chih-Jen Lin, Edward Chang, 2008
  [PDF (2.4MB)] (submitted, under review)

If you find this tool useful, please cite the above work.

Our tool can handle large data sets (200,000 RCV1 data) on a 4GB memory general machine.

All codes are implemented in MATLAB, and data are stored in .mat format. The code has been tested under 64-bit Linux environment using MATLAB 7.4.0.287 (R2007a). You will be able to regenerate experiment results in the paper. However, results may be slightly different due to the randomness, the CPU speed, and the load of your computer.

Name	Version Number	Release Date
MATLAB code and Corel data files	1.0 (.zip [16MB] ) (.tar.gz [16MB])	15 Aug 2008
RCV1 feature file	.mat [98MB]	15 Aug 2008
RCV1 label file	.mat [146KB]	15 Aug 2008

Please feel free to send your feedback to Wen-Yen Chen and Chih-Jen Lin.