Paper: Using Treebanking Discriminants as Parse Disambiguation Features

ACL ID W09-3836
Title Using Treebanking Discriminants as Parse Disambiguation Features
Venue International Conference on Parsing Technologies
Session Main Conference
Year 2009
Authors

This paper presents a novel approach of in- corporating fine-grained treebanking deci- sions made by human annotators as dis- criminative features for automatic parse disambiguation. To our best knowledge, this is the first work that exploits treebank- ing decisions for this task. The advan- tage of this approach is that use of human judgements is made. The paper presents comparative analyses of the performance of discriminative models built using tree- banking decisions and state-of-the-art fea- tures. We also highlight how differently these features scale when these models are tested on out-of-domain data. We show that, features extracted using treebanking decisions are more efficient, informative and robust compared to traditional fea- tures.