Support Vector Clustering Toolbox

This SVC toolbox was written by Dr. Daewon Lee under supervision by Prof. Jaewook Lee. The toolbox is implemented by the Matlab and based on the statistical pattern recognition toolbox (stprtool) in parts of kernel computation and efficient QP solving. At first, you should add the stprtool in your Matlab path. Our SVC toolbox includes the complete graph (CG) based (original SVC), the proximity graph based, and also dynamical system based methods.

Download from

https://sites.google.com/site/daewonlee/research/svctoolbox

Installation

This version includes Statistical Pattern Recognition Toolbox (stprtool) Version 2.12, 12 within it downloaded from here. Instructions for GNU Octave on GNU/Linux platform are:

1. Add strptool to path:

Edit ~/.octaverc and add addpath(genpath("path/to/strptool/"));. 2. Generate mexfiles for strptool

From svc root directory run ./install.sh

Done. Now run octave Demo_SVC.m which should work.