The Effect of Semi-supervised Learning on Parsing Long Distance Dependencies in German and Swedish

This paper shows how the best data-driven dependency parsers available today [1] can be improved by learning from unlabeled data. We focus on German and Swedish and show that labeled attachment scores

• Type: Journal
• Year: 2010
• Where: TAL
• Authors: Anders Søgaard, Christian Rishøj
• Keywords: Attachment Scores, Data-driven Dependency Parsers, Natural Language Processing, TAL 2010, Unlabeled Data