Multilingual projection for parsing truly low-resource languages

Zeljko Agic; Anders Trærup Johannsen; Barbara Plank; Hector Martinez Alonso; Natalie Schluter; Anders Søgaard

Multilingual projection for parsing truly low-resource languages

Zeljko Agic
, Anders Trærup Johannsen
, Barbara Plank
, Hector Martinez Alonso
, Natalie Schluter
, Anders Søgaard

Research output: Journal Article or Conference Article in Journal › Journal article › Research › peer-review

Abstract

We propose a novel approach to cross-lingual part-of-speech tagging and dependency parsing for truly low-resource languages. Our annotation projection-based approach yields tagging and parsing models for over 100 languages. All that is needed are freely available parallel texts, and taggers and parsers for resource-rich languages. The empirical evaluation across 30 test languages shows that our method consistently provides top-level accuracies, close to established upper bounds, and outperforms several competitive baselines.

Original language	English
Journal	Transactions of the Association for Computational Linguistics
Volume	4
Pages (from-to)	301-312
ISSN	2307-387X
Publication status	Published - Jul 2016
Externally published	Yes

Keywords

Cross-lingual part-of-speech tagging
Dependency parsing
Low-resource languages
Annotation projection
Parallel texts

Access to Document

http://aclweb.org/anthology/Q/Q16/Q16-1022.pdf

Cite this

@article{a112ab9f9eb547eda5de2c3f344df463,

title = "Multilingual projection for parsing truly low-resource languages",

abstract = "We propose a novel approach to cross-lingual part-of-speech tagging and dependency parsing for truly low-resource languages. Our annotation projection-based approach yields tagging and parsing models for over 100 languages. All that is needed are freely available parallel texts, and taggers and parsers for resource-rich languages. The empirical evaluation across 30 test languages shows that our method consistently provides top-level accuracies, close to established upper bounds, and outperforms several competitive baselines.",

keywords = "Cross-lingual part-of-speech tagging, Dependency parsing, Low-resource languages, Annotation projection, Parallel texts, Cross-lingual part-of-speech tagging, Dependency parsing, Low-resource languages, Annotation projection, Parallel texts",

author = "Zeljko Agic and Johannsen, \{Anders Tr{\ae}rup\} and Barbara Plank and \{Martinez Alonso\}, Hector and Natalie Schluter and Anders S{\o}gaard",

year = "2016",

month = jul,

language = "English",

volume = "4",

pages = "301--312",

journal = "Transactions of the Association for Computational Linguistics",

issn = "2307-387X",

publisher = "MIT Press",

}

TY - JOUR

T1 - Multilingual projection for parsing truly low-resource languages

AU - Agic, Zeljko

AU - Johannsen, Anders Trærup

AU - Plank, Barbara

AU - Martinez Alonso, Hector

AU - Schluter, Natalie

AU - Søgaard, Anders

PY - 2016/7

Y1 - 2016/7

N2 - We propose a novel approach to cross-lingual part-of-speech tagging and dependency parsing for truly low-resource languages. Our annotation projection-based approach yields tagging and parsing models for over 100 languages. All that is needed are freely available parallel texts, and taggers and parsers for resource-rich languages. The empirical evaluation across 30 test languages shows that our method consistently provides top-level accuracies, close to established upper bounds, and outperforms several competitive baselines.

AB - We propose a novel approach to cross-lingual part-of-speech tagging and dependency parsing for truly low-resource languages. Our annotation projection-based approach yields tagging and parsing models for over 100 languages. All that is needed are freely available parallel texts, and taggers and parsers for resource-rich languages. The empirical evaluation across 30 test languages shows that our method consistently provides top-level accuracies, close to established upper bounds, and outperforms several competitive baselines.

KW - Cross-lingual part-of-speech tagging

KW - Dependency parsing

KW - Low-resource languages

KW - Annotation projection

KW - Parallel texts

KW - Cross-lingual part-of-speech tagging

KW - Dependency parsing

KW - Low-resource languages

KW - Annotation projection

KW - Parallel texts

M3 - Journal article

SN - 2307-387X

VL - 4

SP - 301

EP - 312

JO - Transactions of the Association for Computational Linguistics

JF - Transactions of the Association for Computational Linguistics

ER -

Multilingual projection for parsing truly low-resource languages

Abstract

Keywords

Access to Document

Fingerprint

Cite this