Simple Semi-Supervised Training of Part-Of-Speech Taggers

Most attempts to train part-of-speech taggers on a mixture of labeled and unlabeled data have failed. In this work stacked learning is used to reduce tagging to a classification task. This simplifies semisupervised training considerably. Our prefered semi-supervised method combines tri-training (Li and Zhou, 2005) and disagreement-based co-training. On the Wall Street Journal, we obtain an error reduction of 4.2% with SVMTool (Gimenez and Marquez, 2004).

• Type: Journal
• Year: 2010
• Where: ACL
• Authors: Anders Søgaard
• Keywords: ACL 2010, Computational Linguistics, Part-of-speech Taggers, Simplifies Semisupervised Training, Wall Street Journal