Universal dependency parsing with a general transition-based DAG parser

Daniel Hershcovich; Omri Abend; Ari Rappoport

doi:10.18653/v1/K18-2010

Universal dependency parsing with a general transition-based DAG parser

Daniel Hershcovich, Omri Abend, Ari Rappoport

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

6 Scopus citations

Abstract

This paper presents our experiments with applying TUPA to the CoNLL 2018 UD shared task. TUPA is a general neural transition-based DAG parser, which we use to present the first experiments on recovering enhanced dependencies as part of the general parsing task. TUPA was designed for parsing UCCA, a cross-linguistic semantic annotation scheme, exhibiting reentrancy, discontinuity and non-terminal nodes. By converting UD trees and graphs to a UCCA-like DAG format, we train TUPA almost without modification on the UD parsing task. The generic nature of our approach lends itself naturally to multitask learning. Our code is available at https://github.com/CoNLL-UD-2018/HUJI.

Original language	English
Title of host publication	CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task
Subtitle of host publication	Multilingual Parsing from Raw Text to Universal Dependencies
Publisher	Association for Computational Linguistics (ACL)
Pages	103-112
Number of pages	10
ISBN (Electronic)	9781948087827
DOIs	https://doi.org/10.18653/v1/K18-2010
State	Published - 2018
Event	2018 SIGNLL Conference on Computational Natural Language Learning, CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2018 - Brussels, Belgium Duration: 31 Oct 2018 → 1 Nov 2018

Publication series

Name	CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies

Conference

Conference	2018 SIGNLL Conference on Computational Natural Language Learning, CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2018
Country/Territory	Belgium
City	Brussels
Period	31/10/18 → 1/11/18

Bibliographical note

Publisher Copyright:
© 2018 Association for Computational Linguistics

Access to Document

10.18653/v1/K18-2010

Cite this

Hershcovich, D., Abend, O., & Rappoport, A. (2018). Universal dependency parsing with a general transition-based DAG parser. In CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies (pp. 103-112). (CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/K18-2010

Hershcovich, Daniel ; Abend, Omri ; Rappoport, Ari. / Universal dependency parsing with a general transition-based DAG parser. CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. Association for Computational Linguistics (ACL), 2018. pp. 103-112 (CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies).

@inproceedings{9d5f102bc16f4c89a7dd2f54f3aedec9,

title = "Universal dependency parsing with a general transition-based DAG parser",

abstract = "This paper presents our experiments with applying TUPA to the CoNLL 2018 UD shared task. TUPA is a general neural transition-based DAG parser, which we use to present the first experiments on recovering enhanced dependencies as part of the general parsing task. TUPA was designed for parsing UCCA, a cross-linguistic semantic annotation scheme, exhibiting reentrancy, discontinuity and non-terminal nodes. By converting UD trees and graphs to a UCCA-like DAG format, we train TUPA almost without modification on the UD parsing task. The generic nature of our approach lends itself naturally to multitask learning. Our code is available at https://github.com/CoNLL-UD-2018/HUJI.",

author = "Daniel Hershcovich and Omri Abend and Ari Rappoport",

note = "Publisher Copyright: {\textcopyright} 2018 Association for Computational Linguistics; 2018 SIGNLL Conference on Computational Natural Language Learning, CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2018 ; Conference date: 31-10-2018 Through 01-11-2018",

year = "2018",

doi = "10.18653/v1/K18-2010",

language = "אנגלית",

series = "CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies",

publisher = "Association for Computational Linguistics (ACL)",

pages = "103--112",

booktitle = "CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task",

address = "ארצות הברית",

}

Hershcovich, D, Abend, O & Rappoport, A 2018, Universal dependency parsing with a general transition-based DAG parser. in CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, Association for Computational Linguistics (ACL), pp. 103-112, 2018 SIGNLL Conference on Computational Natural Language Learning, CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2018, Brussels, Belgium, 31/10/18. https://doi.org/10.18653/v1/K18-2010

Universal dependency parsing with a general transition-based DAG parser. / Hershcovich, Daniel; Abend, Omri ; Rappoport, Ari.
CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. Association for Computational Linguistics (ACL), 2018. p. 103-112 (CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Universal dependency parsing with a general transition-based DAG parser

AU - Hershcovich, Daniel

AU - Abend, Omri

AU - Rappoport, Ari

PY - 2018

Y1 - 2018

N2 - This paper presents our experiments with applying TUPA to the CoNLL 2018 UD shared task. TUPA is a general neural transition-based DAG parser, which we use to present the first experiments on recovering enhanced dependencies as part of the general parsing task. TUPA was designed for parsing UCCA, a cross-linguistic semantic annotation scheme, exhibiting reentrancy, discontinuity and non-terminal nodes. By converting UD trees and graphs to a UCCA-like DAG format, we train TUPA almost without modification on the UD parsing task. The generic nature of our approach lends itself naturally to multitask learning. Our code is available at https://github.com/CoNLL-UD-2018/HUJI.

AB - This paper presents our experiments with applying TUPA to the CoNLL 2018 UD shared task. TUPA is a general neural transition-based DAG parser, which we use to present the first experiments on recovering enhanced dependencies as part of the general parsing task. TUPA was designed for parsing UCCA, a cross-linguistic semantic annotation scheme, exhibiting reentrancy, discontinuity and non-terminal nodes. By converting UD trees and graphs to a UCCA-like DAG format, we train TUPA almost without modification on the UD parsing task. The generic nature of our approach lends itself naturally to multitask learning. Our code is available at https://github.com/CoNLL-UD-2018/HUJI.

UR - http://www.scopus.com/inward/record.url?scp=85072896810&partnerID=8YFLogxK

U2 - 10.18653/v1/K18-2010

DO - 10.18653/v1/K18-2010

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85072896810

T3 - CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies

SP - 103

EP - 112

BT - CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task

PB - Association for Computational Linguistics (ACL)

T2 - 2018 SIGNLL Conference on Computational Natural Language Learning, CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2018

Y2 - 31 October 2018 through 1 November 2018

ER -

Hershcovich D, Abend O , Rappoport A. Universal dependency parsing with a general transition-based DAG parser. In CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. Association for Computational Linguistics (ACL). 2018. p. 103-112. (CoNLL 2018 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies). doi: 10.18653/v1/K18-2010

Universal dependency parsing with a general transition-based DAG parser

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this