Paper: Self-training and co-training in biomedical word sense disambiguation

ACL ID W11-0223
Title Self-training and co-training in biomedical word sense disambiguation
Venue Workshop on Biomedical Natural Language Processing
Session
Year 2011
Authors

Word sense disambiguation (WSD) is an inter- mediate task within information retrieval and information extraction, attempting to select the proper sense of ambiguous words. Due to the scarcity of training data, semi-supervised learning, which profits from seed annotated examples and a large set of unlabeled data, are worth researching. We present preliminary results of two semi-supervised learning algo- rithms on biomedical word sense disambigua- tion. Both methods add relevant unlabeled ex- amples to the training set, and optimal param- eters are similar for each ambiguous word.