Grammars and Parsers

Department of Computer Science
New York University

The Proteus Project encompasses three research efforts that explore fundamental questions about the nature of linguistic structure and how it can be exploited in NLP applications. The first is our publicly available Apple Pie Parser for English. The second is a newer syntactic parser built by discriminative training techniques. The third is a new effort on multi-dimensional grammars and parsers for parallel texts. Such grammars are a kind of structured model of translational equivalence.

Available Software

• Apple Pie parser
• newer syntactic parser

• GenPar Toolkit for Generalized Parsing
  includes integrated prototype systems for Statistical Machine Translation by Parsing

• Multitree Viewer (MTV)

Recent Publications

• Benjamin Wellington, Sonjia Waxmonsky, and I. Dan Melamed (2006). Empirical Lower Bounds on the Complexity of Translational Equivalence, Proceedings of the 44th annual conference of the Association for Computational Linguistics (ACL'06), Sydney, Australia.

• Joseph Turian and I. Dan Melamed (2005). Constituent Parsing by Classification [PS] [PDF], in Proceedings of the 9th International Workshop on Parsing Technologies (IWPT), Vancouver, BC, Canada.

• I. Dan Melamed, Giorgio Satta, and Ben Wellington (2004). Generalized Multitext Grammars [PS][PDF] Proceedings of the 42nd Annual Conference of the Association for Computational Linguistics (ACL-04), Barcelona, Spain. This version has a couple of typos corrected.

• I. Dan Melamed (May 2003), Multitext Grammars and Synchronous Parsers [PS], [PDF] Proceedings of NAACL/HLT 2003, Edmonton, Canada