Cross-lingual Parse Disambiguation based on Semantic Correspondence

Lea Frermann; Francis Bond

Cross-lingual Parse Disambiguation based on Semantic Correspondence

Lea Frermann, Francis Bond

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

Abstract

We present a system for cross-lingual parse disambiguation, exploiting the assumption that the meaning of a sentence remains unchanged during translation and the fact that different languages have different ambiguities. We simultaneously reduce ambiguity in multiple languages in a fully automatic way. Evaluation shows that the system reliably discards dispreferred parses from the raw parser output, which results in a pre-selection that can speed up manual treebanking.

Original language	English
Title of host publication	The 50th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference, July 8-14, 2012, Jeju Island, Korea - Volume 2: Short Papers
Publisher	Association for Computational Linguistics
Pages	125-129
Number of pages	5
ISBN (Print)	978-1-937284-25-1
Publication status	Published - 2012

Access to Document

P12-2025Final published version, 115 KB

http://www.aclweb.org/anthology/P12-2025

Cite this

@inproceedings{b5aa91f2ff254e16ab738043181c6917,

title = "Cross-lingual Parse Disambiguation based on Semantic Correspondence",

abstract = "We present a system for cross-lingual parse disambiguation, exploiting the assumption that the meaning of a sentence remains unchanged during translation and the fact that different languages have different ambiguities. We simultaneously reduce ambiguity in multiple languages in a fully automatic way. Evaluation shows that the system reliably discards dispreferred parses from the raw parser output, which results in a pre-selection that can speed up manual treebanking.",

author = "Lea Frermann and Francis Bond",

year = "2012",

language = "English",

isbn = "978-1-937284-25-1",

pages = "125--129",

booktitle = "The 50th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference, July 8-14, 2012, Jeju Island, Korea - Volume 2: Short Papers",

publisher = "Association for Computational Linguistics",

}

Cross-lingual Parse Disambiguation based on Semantic Correspondence. / Frermann, Lea; Bond, Francis.

The 50th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference, July 8-14, 2012, Jeju Island, Korea - Volume 2: Short Papers. Association for Computational Linguistics, 2012. p. 125-129.

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

TY - GEN

T1 - Cross-lingual Parse Disambiguation based on Semantic Correspondence

AU - Frermann, Lea

AU - Bond, Francis

PY - 2012

Y1 - 2012

N2 - We present a system for cross-lingual parse disambiguation, exploiting the assumption that the meaning of a sentence remains unchanged during translation and the fact that different languages have different ambiguities. We simultaneously reduce ambiguity in multiple languages in a fully automatic way. Evaluation shows that the system reliably discards dispreferred parses from the raw parser output, which results in a pre-selection that can speed up manual treebanking.

AB - We present a system for cross-lingual parse disambiguation, exploiting the assumption that the meaning of a sentence remains unchanged during translation and the fact that different languages have different ambiguities. We simultaneously reduce ambiguity in multiple languages in a fully automatic way. Evaluation shows that the system reliably discards dispreferred parses from the raw parser output, which results in a pre-selection that can speed up manual treebanking.

M3 - Conference contribution

SN - 978-1-937284-25-1

SP - 125

EP - 129

BT - The 50th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference, July 8-14, 2012, Jeju Island, Korea - Volume 2: Short Papers

PB - Association for Computational Linguistics

ER -

Cross-lingual Parse Disambiguation based on Semantic Correspondence

Abstract

Access to Document

Fingerprint

Cite this