Dependency Parsing

Our dependency parser uses a transition-based, non-projective parsing algorithm showing a linear-time speed for both projective and non-projective parsing. It processes over 14K tokens per second on an Intel Xeon 2.30GHz machine, and shows the near state-of-the-art accuracy for greedy parsing (92.26% on the WSJ corpus).

• It Depends: Dependency Parser Comparison Using A Web-based Evaluation Tool, Jinho D. Choi, Amanda Stent, Joel Tetreault, Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics (ACL'15), 387–396, Beijing, China, 2015.
• Transition-based Dependency Parsing with Selectional Branching, Jinho D. Choi, Andrew McCallum, Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (ACL'13), 1052-1062, Sofia, Bulgaria, 2013.
• Getting the Most out of Transition-based Dependency Parsing, Jinho D. Choi, Martha Palmer, Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies (ACL'11), 687-692, Portland, Oregon, 2011.

English Labels