Cluster Kernels for Semi-Supervised Learning

Part of Advances in Neural Information Processing Systems 15 (NIPS 2002)

Bibtex Metadata Paper


Olivier Chapelle, Jason Weston, Bernhard Schölkopf


We propose a framework to incorporate unlabeled data in kernel classifier, based on the idea that two points in the same cluster are more likely to have the same label. This is achieved by modifying the eigenspectrum of the kernel matrix. Experimental results assess the validity of this approach.