Abstract

Co-EM Support Vector Learning
Ulf Brefeld - Humboldt-Universitaet zu Berlin Tobias Scheffer - Humboldt-Universitaet zu Berlin
Multi-view algorithms, such as co-training and co-EM, utilize unlabeled datawhen the available attributes can be split into independent and compatiblesubsets. Co-EM outperforms co-training for many problems, but it requires theunderlying learner to estimate class probabilities, and to learn fromprobabilistically labeled data. Therefore, co-EM has so far only been studiedwith naive Bayesian learners. We cast linear classifiers into a probabilisticframework and develop a co-EM version of the Support Vector Machine. Weconduct experiments on text classification problems and compare the family ofsemi-supervised support vector algorithms under different conditions,including violations of the assumptions underlying multi-view learning. Forsome problems, such as course web page classification, we observe the mostaccurate results reported so far.

Co-EM Support Vector Learning

Ulf Brefeld - Humboldt-Universitaet zu Berlin
Tobias Scheffer - Humboldt-Universitaet zu Berlin

Multi-view algorithms, such as co-training and co-EM, utilize unlabeled datawhen the available attributes can be split into independent and compatiblesubsets. Co-EM outperforms co-training for many problems, but it requires theunderlying learner to estimate class probabilities, and to learn fromprobabilistically labeled data. Therefore, co-EM has so far only been studiedwith naive Bayesian learners. We cast linear classifiers into a probabilisticframework and develop a co-EM version of the Support Vector Machine. Weconduct experiments on text classification problems and compare the family ofsemi-supervised support vector algorithms under different conditions,including violations of the assumptions underlying multi-view learning. Forsome problems, such as course web page classification, we observe the mostaccurate results reported so far.