fabiansinz / UniverSVM

Fast transductive and universum support vector machine

UniverSVM

Fast transductive and universum support vector machine.

Code can also be obtained from: http://mloss.org/software/view/19/

To quickly test the USVM use the accompanying dockerfile. Assuming docker is installed, use docker build --tag usvm . followed by docker run -it usvm. Inside the container use ./universvm -V 0 -o 0 -c 10 -t 0 -T data/test.dat -D output.dat data/train.dat. That shuld give about 95% accurary.

For further information see:

1. Weston, J., Collobert, R., Sinz, F., Bottou, L., Vapnik, V.: Inference with the Universum. Proceedings of the 23rd international conference on Machine learning ICML 06. pp. 1009–1016. ACM Press (2006).
2. Collobert, R., Sinz, F., Weston, J., Bottou, L.: Trading Convexity for Scalability. In: Bottou, L., Chapelle, O., Decoste, D., and Weston, J. (eds.) Proceedings of the 23rd international conference on Machine learning ICML 06. ACM Press (2006).
3. Collobert, R., Sinz, F., Weston, J., Bottou, L.: Large scale transductive SVMs. J. Mach. Learn. Res. 7, 1687–1712 (2006).
4. Sinz, F.H., Chapelle, O., Agarwal, A., Schölkopf, B.: An Analysis of Inference with the Universum. In: Platt, J.C., Koller, D., Singer, Y., and Roweis, S. (eds.) Advances in Neural Information Processing Systems 20: 21st Annual Conference on Neural Information Processing Systems 2007. pp. 1369–1376. Curran, Red Hook, NY, USA (2008).