Several instance-based large-margin classi¯ers have recently been put forward in the literature: Support Hyperplanes, Nearest Convex Hull classifier, and Soft Nearest Neighbor. We examine those techniques from a common fit-versus-complexity framework and study the links be- tween them. Finally, we compare the performance of these techniques vis-a-vis each other and other standard classification methods.

Additional Metadata
Persistent URL hdl.handle.net/1765/8218
Citation
Nalbantov, G.I, Bioch, J.C, & Groenen, P.J.F. (2007). Instance-Based penalization techniques for classification (No. EI 2007-01). Report / Econometric Institute, Erasmus University Rotterdam. Retrieved from http://hdl.handle.net/1765/8218