pspectralclustering is a parallel C++ implementation of Parallel Spectral Clustering. We are expecting to present a highly optimized parallel implemention of all the steps of spectral clustering. We use PARPACK as underlying eigenvalue decomposition package and F2C to compile fortran code.

If you wish to publish any work based on pspectralclustering, please cite our paper as:

The bibtex format is

  author = {Wen-Yen Chen and Yangqiu Song and Hongjie Bai and Chih-Jen Lin and Edward Y. Chang},^M
  title = {Parallel Spectral Clustering in Distributed Systems},^M
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},^M
  volume = {33},^M
  number = {3},^M
  pages = {568-586},^M
  year = {2011}^M
}  ^M

You can also download our paper here.

If any problems using it, please send mail to