BSVM 2.06

ChihWei Hsu, ChihJen Lin — Wed, 30 Jan 2008 10:27:13 -0000

BSVM solves support vector machines (SVM) for the solution of large classification and regression problems. It includes three methods

* One vs. One multi-class classification using a bound-constrained formulation

* Multi-class classification by solving a single optimization problem (again, a bounded formulation).

* Multi-class classification using Crammer and Singer's formulation.

* Regression using a bound-constrained formulation

It also has an efficient implementation for linear SVMs.

The current implementation borrows the structure of libsvm. Similar options are also adopted. For the bound-constrained formulation for classification and regression, BSVM uses a decomposition method. BSVM uses a simple working set selection which leads to faster convergences for difficult cases. The use of a special implementation of the opmization solver TRON allows BSVM to stably identify bounded variables.

mloss.org BSVM

BSVM 2.06