Transition-based dependency DAG parsing using dynamic oracles

Alper Tokgöz; Gülşen Eryiʇit

doi:10.3115/v1/p15-3004

Transition-based dependency DAG parsing using dynamic oracles

Alper Tokgöz, Gülşen Eryiʇit

Department of Artificial Intelligence and Data Engineering

Istanbul Technical University

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

8 Citations (Scopus)

Abstract

In most of the dependency parsing studies, dependency relations within a sentence are often presented as a tree structure. Whilst the tree structure is sufficient to represent the surface relations, deep dependencies which may result to multi-headed relations require more general dependency structures, namely Directed Acyclic Graphs (DAGs). This study proposes a new dependency DAG parsing approach which uses a dynamic oracle within a shift-reduce transitionbased parsing framework. Although there is still room for improvement on performance with more feature engineering, we already obtain competitive performances compared to static oracles as a result of our initial experiments conducted on the ITU-METU-Sabanci Turkish Treebank (IMST).

Original language	English
Title of host publication	ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop
Publisher	Association for Computational Linguistics (ACL)
Pages	22-27
Number of pages	6
ISBN (Electronic)	9781941643747
DOIs	https://doi.org/10.3115/v1/p15-3004
Publication status	Published - 2015
Event	53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2015 - Beijing, China Duration: 28 Jul 2015 → …

Publication series

Name	ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop

Conference

Conference	53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2015
Country/Territory	China
City	Beijing
Period	28/07/15 → …

Bibliographical note

Publisher Copyright:
© 2015 Association for Computational Linguistics.

Access to Document

10.3115/v1/p15-3004

Cite this

Tokgöz, A., & Eryiʇit, G. (2015). Transition-based dependency DAG parsing using dynamic oracles. In ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop (pp. 22-27). (ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop). Association for Computational Linguistics (ACL). https://doi.org/10.3115/v1/p15-3004

Tokgöz, Alper ; Eryiʇit, Gülşen. / Transition-based dependency DAG parsing using dynamic oracles. ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop. Association for Computational Linguistics (ACL), 2015. pp. 22-27 (ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop).

@inproceedings{f67848244f194eb38caff0f932a9db04,

title = "Transition-based dependency DAG parsing using dynamic oracles",

abstract = "In most of the dependency parsing studies, dependency relations within a sentence are often presented as a tree structure. Whilst the tree structure is sufficient to represent the surface relations, deep dependencies which may result to multi-headed relations require more general dependency structures, namely Directed Acyclic Graphs (DAGs). This study proposes a new dependency DAG parsing approach which uses a dynamic oracle within a shift-reduce transitionbased parsing framework. Although there is still room for improvement on performance with more feature engineering, we already obtain competitive performances compared to static oracles as a result of our initial experiments conducted on the ITU-METU-Sabanci Turkish Treebank (IMST).",

author = "Alper Tokg{\"o}z and G{\"u}l{\c s}en Eryiʇit",

note = "Publisher Copyright: {\textcopyright} 2015 Association for Computational Linguistics.; 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2015 ; Conference date: 28-07-2015",

year = "2015",

doi = "10.3115/v1/p15-3004",

language = "English",

series = "ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop",

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

pages = "22--27",

booktitle = "ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop",

address = "United States",

}

Tokgöz, A & Eryiʇit, G 2015, Transition-based dependency DAG parsing using dynamic oracles. in ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop. ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop, Association for Computational Linguistics (ACL), pp. 22-27, 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2015, Beijing, China, 28/07/15. https://doi.org/10.3115/v1/p15-3004

Transition-based dependency DAG parsing using dynamic oracles. / Tokgöz, Alper; Eryiʇit, Gülşen.
ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop. Association for Computational Linguistics (ACL), 2015. p. 22-27 (ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop).

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

TY - GEN

T1 - Transition-based dependency DAG parsing using dynamic oracles

AU - Tokgöz, Alper

AU - Eryiʇit, Gülşen

PY - 2015

Y1 - 2015

N2 - In most of the dependency parsing studies, dependency relations within a sentence are often presented as a tree structure. Whilst the tree structure is sufficient to represent the surface relations, deep dependencies which may result to multi-headed relations require more general dependency structures, namely Directed Acyclic Graphs (DAGs). This study proposes a new dependency DAG parsing approach which uses a dynamic oracle within a shift-reduce transitionbased parsing framework. Although there is still room for improvement on performance with more feature engineering, we already obtain competitive performances compared to static oracles as a result of our initial experiments conducted on the ITU-METU-Sabanci Turkish Treebank (IMST).

AB - In most of the dependency parsing studies, dependency relations within a sentence are often presented as a tree structure. Whilst the tree structure is sufficient to represent the surface relations, deep dependencies which may result to multi-headed relations require more general dependency structures, namely Directed Acyclic Graphs (DAGs). This study proposes a new dependency DAG parsing approach which uses a dynamic oracle within a shift-reduce transitionbased parsing framework. Although there is still room for improvement on performance with more feature engineering, we already obtain competitive performances compared to static oracles as a result of our initial experiments conducted on the ITU-METU-Sabanci Turkish Treebank (IMST).

UR - http://www.scopus.com/inward/record.url?scp=84944346000&partnerID=8YFLogxK

U2 - 10.3115/v1/p15-3004

DO - 10.3115/v1/p15-3004

M3 - Conference contribution

AN - SCOPUS:84944346000

T3 - ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop

SP - 22

EP - 27

BT - ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop

PB - Association for Computational Linguistics (ACL)

T2 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2015

Y2 - 28 July 2015

ER -

Tokgöz A, Eryiʇit G. Transition-based dependency DAG parsing using dynamic oracles. In ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop. Association for Computational Linguistics (ACL). 2015. p. 22-27. (ACL-IJCNLP 2015 - 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Proceedings of the Student Research Workshop). doi: 10.3115/v1/p15-3004

Transition-based dependency DAG parsing using dynamic oracles

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this