Paper: A Latent Variable Model of Synchronous Syntactic-Semantic Parsing for Multiple Languages

ACL ID W09-1205
Title A Latent Variable Model of Synchronous Syntactic-Semantic Parsing for Multiple Languages
Venue International Conference on Computational Natural Language Learning
Session shared task
Year 2009
Authors

Motivated by the large number of languages (seven) and the short development time (two months) of the 2009 CoNLL shared task, we exploited latent variables to avoid the costly process of hand-crafted feature engineering, allowing the latent variables to induce features from the data. We took a pre-existing gener- ative latent variable model of joint syntactic- semantic dependency parsing, developed for English, and applied it to six new languages with minimal adjustments. The parser’s ro- bustness across languages indicates that this parser has a very general feature set. The parser’s high performance indicates that its la- tent variables succeeded in inducing effective features. This system was ranked third overall with a macro averaged F1 score of 82.14%, only 0.5% worse than the bes...