Applying statistical parsers developed for English to languages with freer word-order has turned out to be harder than expected. This paper investigates the adequacy of different statistical parsing models for dealing with a (relatively) free word-order language. We show that the recently proposed Relational-Realizational (RR) model consistently outperforms state-of-the-art Head-Driven (HD) models on the Hebrew Treebank. Our analysis reveals a weakness of HD models: their intrinsic focus on configurational information. We conclude that the form-function separation ingrained in RR models makes them better suited for parsing nonconfigurational phenomena.
|Number of pages||10|
|State||Published - 2009|
|Event||2009 Conference on Empirical Methods in Natural Language Processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009 - Singapore, Singapore|
Duration: 6 Aug 2009 → 7 Aug 2009
|Conference||2009 Conference on Empirical Methods in Natural Language Processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009|
|Period||6/08/09 → 7/08/09|