BSVM: software for large classification problems

[Chih-jen Lin]

Dear Colleagues:

We announce the release of the software BSVM, a decomposition 
method for support vector machines (SVM) for large classification 
problems.

BSVM uses a decomposition method to solve a bound-constrained 
SVM formulation. Advantages of BSVM include

1.Using a simple working set selection which leads to faster
convergences for difficult cases.

2.Using a bounded SVM formulation and a projected gradient optimization
solver which allow BSVM to quickly and stably identify support vectors.

3.It naturally uses the idea of a cache.

The current release (Version 1.0) is available from

  http://www.csie.ntu.edu.tw/~cjlin/bsvm

For additional information on BSVM, see

  Chih-Wei Hsu and Chih-Jen Lin
  A simple decomposition method for  support vector machines, January 2000
  http://www.csie.ntu.edu.tw/~cjlin/papers/decomp.ps.gz 

Any comments are very welcome.

Sincerely,
Chih-Jen Lin
Department of Computer Science and
Information Engineering
National Taiwan University 
Taipei, Taiwan
cjlin at csie.ntu.edu.tw