From raw text to universal dependencies – Look, no tags!

Miryam de Lhoneux; Yan Shao; Ali Basirat; Eliyahu Kiperwasser; Sara Stymne; Yoav Goldberg; Joakim Nivre

doi:10.18653/v1/k17-3022

From raw text to universal dependencies – Look, no tags!

Miryam de Lhoneux, Yan Shao, Ali Basirat, Eliyahu Kiperwasser, Sara Stymne, Yoav Goldberg, Joakim Nivre

Department of Computer Science

Uppsala University

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

31 Scopus citations

Abstract

We present the Uppsala submission to the CoNLL 2017 shared task on parsing from raw text to universal dependencies. Our system is a simple pipeline consisting of two components. The first performs joint word and sentence segmentation on raw text; the second predicts dependency trees from raw words. The parser bypasses the need for part-of-speech tagging, but uses word embeddings based on universal tag distributions. We achieved a macro-averaged LAS F1 of 65.11 in the official test run and obtained the 2nd best result for sentence segmentation with a score of 89.03. After fixing two bugs, we obtained an unofficial LAS F1 of 70.49.

Original language	English
Title of host publication	CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task
Subtitle of host publication	Multilingual Parsing from Raw Text to Universal Dependencies
Publisher	Association for Computational Linguistics (ACL)
Pages	207-217
Number of pages	11
ISBN (Electronic)	9781945626708
DOIs	https://doi.org/10.18653/v1/k17-3022
State	Published - 2017
Event	2017 SIGNLL Conference on Computational Natural Language Learning- CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2017 - Vancouver, Canada Duration: 3 Aug 2017 → 4 Aug 2017

Publication series

Name	CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies

Conference

Conference	2017 SIGNLL Conference on Computational Natural Language Learning- CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2017
Country/Territory	Canada
City	Vancouver
Period	3/08/17 → 4/08/17

Bibliographical note

Publisher Copyright:
© 2017 Association for Computational Linguistics.

Funding

We are grateful to the shared task organizers and to Dan Zeman in particular, and we acknowledge the computational resources provided by CSC in Helsinki and Sigma2 in Oslo through NeIC-NLPL (www.nlpl.eu). Our parser will be made available in the NLPL dependency parsing laboratory.

Funders	Funder number
CSC in Helsinki and Sigma2 in Oslo

Access to Document

10.18653/v1/k17-3022

Cite this

de Lhoneux, M., Shao, Y., Basirat, A., Kiperwasser, E., Stymne, S., Goldberg, Y., & Nivre, J. (2017). From raw text to universal dependencies – Look, no tags! In CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies (pp. 207-217). (CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/k17-3022

de Lhoneux, Miryam ; Shao, Yan ; Basirat, Ali et al. / From raw text to universal dependencies – Look, no tags!. CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. Association for Computational Linguistics (ACL), 2017. pp. 207-217 (CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies).

@inproceedings{bfcde80e77f3485fb138853088cd920e,

title = "From raw text to universal dependencies – Look, no tags!",

abstract = "We present the Uppsala submission to the CoNLL 2017 shared task on parsing from raw text to universal dependencies. Our system is a simple pipeline consisting of two components. The first performs joint word and sentence segmentation on raw text; the second predicts dependency trees from raw words. The parser bypasses the need for part-of-speech tagging, but uses word embeddings based on universal tag distributions. We achieved a macro-averaged LAS F1 of 65.11 in the official test run and obtained the 2nd best result for sentence segmentation with a score of 89.03. After fixing two bugs, we obtained an unofficial LAS F1 of 70.49.",

author = "{de Lhoneux}, Miryam and Yan Shao and Ali Basirat and Eliyahu Kiperwasser and Sara Stymne and Yoav Goldberg and Joakim Nivre",

note = "Publisher Copyright: {\textcopyright} 2017 Association for Computational Linguistics.; 2017 SIGNLL Conference on Computational Natural Language Learning- CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2017 ; Conference date: 03-08-2017 Through 04-08-2017",

year = "2017",

doi = "10.18653/v1/k17-3022",

language = "אנגלית",

series = "CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies",

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

pages = "207--217",

booktitle = "CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task",

address = "ארצות הברית",

}

de Lhoneux, M, Shao, Y, Basirat, A, Kiperwasser, E, Stymne, S, Goldberg, Y & Nivre, J 2017, From raw text to universal dependencies – Look, no tags! in CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, Association for Computational Linguistics (ACL), pp. 207-217, 2017 SIGNLL Conference on Computational Natural Language Learning- CoNLL Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, CoNLL 2017, Vancouver, Canada, 3/08/17. https://doi.org/10.18653/v1/k17-3022

From raw text to universal dependencies – Look, no tags! / de Lhoneux, Miryam; Shao, Yan; Basirat, Ali et al.
CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. Association for Computational Linguistics (ACL), 2017. p. 207-217 (CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 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 - From raw text to universal dependencies – Look, no tags!

AU - de Lhoneux, Miryam

AU - Shao, Yan

AU - Basirat, Ali

AU - Kiperwasser, Eliyahu

AU - Stymne, Sara

AU - Goldberg, Yoav

AU - Nivre, Joakim

PY - 2017

Y1 - 2017

N2 - We present the Uppsala submission to the CoNLL 2017 shared task on parsing from raw text to universal dependencies. Our system is a simple pipeline consisting of two components. The first performs joint word and sentence segmentation on raw text; the second predicts dependency trees from raw words. The parser bypasses the need for part-of-speech tagging, but uses word embeddings based on universal tag distributions. We achieved a macro-averaged LAS F1 of 65.11 in the official test run and obtained the 2nd best result for sentence segmentation with a score of 89.03. After fixing two bugs, we obtained an unofficial LAS F1 of 70.49.

AB - We present the Uppsala submission to the CoNLL 2017 shared task on parsing from raw text to universal dependencies. Our system is a simple pipeline consisting of two components. The first performs joint word and sentence segmentation on raw text; the second predicts dependency trees from raw words. The parser bypasses the need for part-of-speech tagging, but uses word embeddings based on universal tag distributions. We achieved a macro-averaged LAS F1 of 65.11 in the official test run and obtained the 2nd best result for sentence segmentation with a score of 89.03. After fixing two bugs, we obtained an unofficial LAS F1 of 70.49.

UR - http://www.scopus.com/inward/record.url?scp=85063163938&partnerID=8YFLogxK

U2 - 10.18653/v1/k17-3022

DO - 10.18653/v1/k17-3022

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

AN - SCOPUS:85063163938

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 - 207

EP - 217

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 -

de Lhoneux M, Shao Y, Basirat A, Kiperwasser E, Stymne S, Goldberg Y et al. From raw text to universal dependencies – Look, no tags! In CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. Association for Computational Linguistics (ACL). 2017. p. 207-217. (CoNLL 2017 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies). doi: 10.18653/v1/k17-3022

From raw text to universal dependencies – Look, no tags!

Abstract

Publication series

Conference

Bibliographical note

Funding

Access to Document

Other files and links

Fingerprint

Cite this