Paper: Learning Reliability of Parses for Domain Adaptation of Dependency Parsing

ACL ID I08-2097
Title Learning Reliability of Parses for Domain Adaptation of Dependency Parsing
Venue International Joint Conference on Natural Language Processing
Session Main Conference
Year 2008
Authors

The accuracy of parsing has exceeded 90% recently, but this is not high enough to use parsing results practically in natural lan- guageprocessing(NLP)applicationssuchas paraphrase acquisition and relation extrac- tion. We present a method for detecting re- liable parses out of the outputs of a single dependency parser. This technique is also applied to domain adaptation of dependency parsing. Our goal was to improve the per- formance of a state-of-the-art dependency parser on the data set of the domain adap- tation track of the CoNLL 2007 shared task, a formidable challenge.