LSC is a statistical softclustering tool which is trained with the Expectation Maximization algorithm. The statistical model is described in a paper by Mats Rooth with the title "Two-dimensional clusters in grammatical relations" which is available here.

The LSC source code can be downloaded here as a gzip-compressed tar file. It is freely available for education, research and other non-commercial purposes.

Please send comments, suggestions and bug reports to me (Helmut Schmid) at the address