TY - GEN
T1 - The HIT-SCIR system for end-to-end parsing of universal dependencies
AU - Che, Wanxiang
AU - Guo, Jiang
AU - Wang, Yuxuan
AU - Zheng, Bo
AU - Zhao, Huaipeng
AU - Liu, Yang
AU - Teng, Dechuan
AU - Liu, Ting
N1 - Publisher Copyright:
© 2017 Association for Computational Linguistics.
PY - 2017
Y1 - 2017
N2 - This paper describes our system (HIT-SCIR) for the CoNLL 2017 shared task: Multilingual Parsing from Raw Text to Universal Dependencies. Our system includes three pipelined components: tokenization, Part-of-Speech (POS) tagging and dependency parsing. We use character-based bidirectional long short-term memory (LSTM) networks for both tokenization and POS tagging. After-wards, we employ a list-based transition-based algorithm for general non-projective parsing and present an improved Stack-LSTM-based architecture for representing each transition state and making predictions. Furthermore, to parse low/zero-resource languages and cross-domain data, we use a model transfer approach to make effective use of existing resources. We demonstrate substantial gains against the UDPipe baseline, with an average improvement of 3.76% in LAS of all languages. And finally, we rank the 4th place on the official test sets.
AB - This paper describes our system (HIT-SCIR) for the CoNLL 2017 shared task: Multilingual Parsing from Raw Text to Universal Dependencies. Our system includes three pipelined components: tokenization, Part-of-Speech (POS) tagging and dependency parsing. We use character-based bidirectional long short-term memory (LSTM) networks for both tokenization and POS tagging. After-wards, we employ a list-based transition-based algorithm for general non-projective parsing and present an improved Stack-LSTM-based architecture for representing each transition state and making predictions. Furthermore, to parse low/zero-resource languages and cross-domain data, we use a model transfer approach to make effective use of existing resources. We demonstrate substantial gains against the UDPipe baseline, with an average improvement of 3.76% in LAS of all languages. And finally, we rank the 4th place on the official test sets.
UR - https://www.scopus.com/pages/publications/85060458993
U2 - 10.18653/v1/k17-3005
DO - 10.18653/v1/k17-3005
M3 - 会议稿件
AN - SCOPUS:85060458993
T3 - CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies
SP - 52
EP - 62
BT - CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task
PB - Association for Computational Linguistics (ACL)
T2 - 2017 SIGNLL Conference on Computational Natural Language Learning- CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2017
Y2 - 3 August 2017 through 4 August 2017
ER -