Cross-lingual transfer parsing for low-resourced languages: an Irish case study

Teresa Lynn; Jennifer Foster; Mark Dras; Lamia Tounsi

Cross-lingual transfer parsing for low-resourced languages: an Irish case study

Teresa Lynn, Jennifer Foster, Mark Dras, Lamia Tounsi

School of Computing

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

79 Downloads (Pure)

Abstract

We present a study of cross-lingual direct transfer parsing for the Irish language. Firstly we discuss mapping of the annotation scheme of the Irish Dependency Treebank to a universal dependency scheme. We explain our dependency label mapping choices and the structural changes required in the Irish Dependency Treebank. We then experiment with the universally annotated treebanks of ten languages from four language family groups to assess which languages are the most useful for cross-lingual parsing of Irish by using these treebanks to train delexicalised parsing models which are then applied to sentences from the Irish Dependency Treebank. The best results are achieved when using Indonesian, a language from the Austronesian language family.

Original language	English
Title of host publication	Proceedings of the Celtic Language Technology Workshop (CLTW)
Editors	John Judge, Teresa Lynn, Monica Ward, Brian Ó Raghallaigh
Place of Publication	Dublin, Ireland
Publisher	Association for Computational Linguistics
Pages	41-49
Number of pages	9
ISBN (Print)	9781873769324
Publication status	Published - 2014
Event	Celtic Language Technology Workshop (1st : 2014) - Dublin, Ireland Duration: 23 Aug 2014 → 23 Aug 2014

Workshop

Workshop	Celtic Language Technology Workshop (1st : 2014)
City	Dublin, Ireland
Period	23/08/14 → 23/08/14

Bibliographical note

Copyright the Author(s) 2014. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Access to Document

Publisher version (open access)Final published version, 181 KBLicence: Unspecified

http://www.aclweb.org/anthology/W/W14/#4600

Cite this

@inproceedings{e9d1c65a8a724c2b96dac092318f9eb0,

title = "Cross-lingual transfer parsing for low-resourced languages: an Irish case study",

abstract = "We present a study of cross-lingual direct transfer parsing for the Irish language. Firstly we discuss mapping of the annotation scheme of the Irish Dependency Treebank to a universal dependency scheme. We explain our dependency label mapping choices and the structural changes required in the Irish Dependency Treebank. We then experiment with the universally annotated treebanks of ten languages from four language family groups to assess which languages are the most useful for cross-lingual parsing of Irish by using these treebanks to train delexicalised parsing models which are then applied to sentences from the Irish Dependency Treebank. The best results are achieved when using Indonesian, a language from the Austronesian language family.",

author = "Teresa Lynn and Jennifer Foster and Mark Dras and Lamia Tounsi",

note = "Copyright the Author(s) 2014. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.; Celtic Language Technology Workshop (1st : 2014) ; Conference date: 23-08-2014 Through 23-08-2014",

year = "2014",

language = "English",

isbn = "9781873769324",

pages = "41--49",

editor = "John Judge and Teresa Lynn and Monica Ward and Raghallaigh, {Brian {\'O}}",

booktitle = "Proceedings of the Celtic Language Technology Workshop (CLTW)",

publisher = "Association for Computational Linguistics",

}

Lynn, T, Foster, J, Dras, M & Tounsi, L 2014, Cross-lingual transfer parsing for low-resourced languages: an Irish case study. in J Judge, T Lynn, M Ward & BÓ Raghallaigh (eds), Proceedings of the Celtic Language Technology Workshop (CLTW). Association for Computational Linguistics, Dublin, Ireland, pp. 41-49, Celtic Language Technology Workshop (1st : 2014), Dublin, Ireland, 23/08/14. <http://www.aclweb.org/anthology/W/W14/#4600>

Cross-lingual transfer parsing for low-resourced languages: an Irish case study. / Lynn, Teresa; Foster, Jennifer; Dras, Mark et al.
Proceedings of the Celtic Language Technology Workshop (CLTW). ed. / John Judge; Teresa Lynn; Monica Ward; Brian Ó Raghallaigh. Dublin, Ireland: Association for Computational Linguistics, 2014. p. 41-49.

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

TY - GEN

T1 - Cross-lingual transfer parsing for low-resourced languages

T2 - Celtic Language Technology Workshop (1st : 2014)

AU - Lynn, Teresa

AU - Foster, Jennifer

AU - Dras, Mark

AU - Tounsi, Lamia

N1 - Copyright the Author(s) 2014. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2014

Y1 - 2014

N2 - We present a study of cross-lingual direct transfer parsing for the Irish language. Firstly we discuss mapping of the annotation scheme of the Irish Dependency Treebank to a universal dependency scheme. We explain our dependency label mapping choices and the structural changes required in the Irish Dependency Treebank. We then experiment with the universally annotated treebanks of ten languages from four language family groups to assess which languages are the most useful for cross-lingual parsing of Irish by using these treebanks to train delexicalised parsing models which are then applied to sentences from the Irish Dependency Treebank. The best results are achieved when using Indonesian, a language from the Austronesian language family.

AB - We present a study of cross-lingual direct transfer parsing for the Irish language. Firstly we discuss mapping of the annotation scheme of the Irish Dependency Treebank to a universal dependency scheme. We explain our dependency label mapping choices and the structural changes required in the Irish Dependency Treebank. We then experiment with the universally annotated treebanks of ten languages from four language family groups to assess which languages are the most useful for cross-lingual parsing of Irish by using these treebanks to train delexicalised parsing models which are then applied to sentences from the Irish Dependency Treebank. The best results are achieved when using Indonesian, a language from the Austronesian language family.

M3 - Conference proceeding contribution

SN - 9781873769324

SP - 41

EP - 49

BT - Proceedings of the Celtic Language Technology Workshop (CLTW)

A2 - Judge, John

A2 - Lynn, Teresa

A2 - Ward, Monica

A2 - Raghallaigh, Brian Ó

PB - Association for Computational Linguistics

CY - Dublin, Ireland

Y2 - 23 August 2014 through 23 August 2014

ER -

Cross-lingual transfer parsing for low-resourced languages: an Irish case study

Abstract

Workshop

Bibliographical note

Access to Document

Fingerprint

Cite this