Ulf Brefeld, Christoph Büscher, Tobias Scheffer
Discriminative learning techniques for sequential data have proven to be more effective than generative models for named entity recognition, information extraction, and other tasks of discrimination. However, semi-supervised learning mechanisms that utilize inexpensive unlabeled sequences in addition to few labeled sequences – such as the Baum-Welch algorithm – are available only for generative models. The multi-view approach is based on the principle of maximizing the consensus among multiple independent hypotheses; we develop this principle into a semisupervised hidden Markov perceptron algorithm. Experiments reveal that the resulting procedure utilizes unlabeled data effectively and discriminates more accurately than its purely supervised counterparts.
[article]