TY - GEN
T1 - A non-monotonic Arc-Eager transition system for dependency parsing
AU - Honnibal, Matthew
AU - Goldberg, Yoav
AU - Johnson, Mark
N1 - Publisher Copyright:
© 2013 Association for Computational Linguistics.
PY - 2013/1/1
Y1 - 2013/1/1
N2 - Previous incremental parsers have used monotonic state transitions. However, transitions can be made to revise previous decisions quite naturally, based on further information. We show that a simple adjustment to the Arc-Eager transition system to relax its monotonicity constraints can improve accuracy, so long as the training data includes examples of mistakes for the non-monotonic transitions to repair. We evaluate the change in the context of a state-of-the-art system, and obtain a statistically significant improvement (p < 0.001) on the English evaluation and 5/10 of the CoNLL languages.
AB - Previous incremental parsers have used monotonic state transitions. However, transitions can be made to revise previous decisions quite naturally, based on further information. We show that a simple adjustment to the Arc-Eager transition system to relax its monotonicity constraints can improve accuracy, so long as the training data includes examples of mistakes for the non-monotonic transitions to repair. We evaluate the change in the context of a state-of-the-art system, and obtain a statistically significant improvement (p < 0.001) on the English evaluation and 5/10 of the CoNLL languages.
UR - http://www.scopus.com/inward/record.url?scp=84951730021&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84951730021
T3 - CoNLL 2013 - 17th Conference on Computational Natural Language Learning, Proceedings
SP - 163
EP - 172
BT - CoNLL 2013 - 17th Conference on Computational Natural Language Learning, Proceedings
PB - Association for Computational Linguistics (ACL)
T2 - 17th Conference on Computational Natural Language Learning, CoNLL 2013
Y2 - 8 August 2013 through 9 August 2013
ER -