Conceptual Annotations Preserve Structure Across Translations: A French-English Case Study

Elior Sulem; Omri Abend; Ari Rappoport

doi:10.18653/v1/W15-3502

Conceptual Annotations Preserve Structure Across Translations: A French-English Case Study

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

Abstract

Divergence of syntactic structures between languages constitutes a major challenge in using linguistic structure in Machine Translation (MT) systems. Here, we examine the potential of semantic structures. While semantic annotation is appealing as a source of cross-linguistically stable structures, little has been accomplished in demonstrating this stability through a detailed corpus study. In this paper, we experiment with the UCCA conceptual-cognitive annotation scheme in an English-French case study. First, we show that UCCA can be used to annotate French, through a systematic type-level analysis of the major French grammatical phenomena. Second, we annotate a parallel English-French corpus with UCCA, and quantify the similarity of the structures on both sides. Results show a high degree of stability across translations, supporting the usage of semantic annotations over syntactic ones in structure-aware MT systems.

Original language	English
Title of host publication	Proceedings of the 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015)
Editors	Deyi Xiong, Kevin Duh, Christian Hardmeier, Roberto Navigli
Place of Publication	Beijing, China
Publisher	Association for Computational Linguistics (ACL)
Pages	11-22
Number of pages	12
ISBN (Electronic)	978-1-941643-61-7
DOIs	https://doi.org/10.18653/v1/W15-3502
State	Published - 1 Jul 2015
Event	1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015) - Beijing, China Duration: 30 Jul 2015 → 30 Jul 2015 Conference number: 1 https://aclanthology.org/volumes/W15-35/

Workshop

Workshop	1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015)
Country/Territory	China
City	Beijing
Period	30/07/15 → 30/07/15
Internet address	https://aclanthology.org/volumes/W15-35/

Keywords

linguistic structure
Machine Translation
MT
semantic structures
UCCA
semantic annotations

Access to Document

10.18653/v1/W15-3502License: CC BY-NC-SA

Cite this

Sulem, E., Abend, O., & Rappoport, A. (2015). Conceptual Annotations Preserve Structure Across Translations: A French-English Case Study. In D. Xiong, K. Duh, C. Hardmeier, & R. Navigli (Eds.), Proceedings of the 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015) (pp. 11-22). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/W15-3502

Sulem, Elior ; Abend, Omri ; Rappoport, Ari. / Conceptual Annotations Preserve Structure Across Translations : A French-English Case Study. Proceedings of the 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015). editor / Deyi Xiong ; Kevin Duh ; Christian Hardmeier ; Roberto Navigli. Beijing, China : Association for Computational Linguistics (ACL), 2015. pp. 11-22

@inproceedings{affc072bcfff493b8157f9ac252e4e34,

title = "Conceptual Annotations Preserve Structure Across Translations: A French-English Case Study",

abstract = "Divergence of syntactic structures between languages constitutes a major challenge in using linguistic structure in Machine Translation (MT) systems. Here, we examine the potential of semantic structures. While semantic annotation is appealing as a source of cross-linguistically stable structures, little has been accomplished in demonstrating this stability through a detailed corpus study. In this paper, we experiment with the UCCA conceptual-cognitive annotation scheme in an English-French case study. First, we show that UCCA can be used to annotate French, through a systematic type-level analysis of the major French grammatical phenomena. Second, we annotate a parallel English-French corpus with UCCA, and quantify the similarity of the structures on both sides. Results show a high degree of stability across translations, supporting the usage of semantic annotations over syntactic ones in structure-aware MT systems.",

keywords = "linguistic structure, Machine Translation, MT, semantic structures, UCCA, semantic annotations",

author = "Elior Sulem and Omri Abend and Ari Rappoport",

year = "2015",

month = jul,

day = "1",

doi = "10.18653/v1/W15-3502",

language = "אנגלית",

pages = "11--22",

editor = "Deyi Xiong and Kevin Duh and Christian Hardmeier and Roberto Navigli",

booktitle = "Proceedings of the 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015)",

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

address = "ארצות הברית",

note = "1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015) ; Conference date: 30-07-2015 Through 30-07-2015",

url = "https://aclanthology.org/volumes/W15-35/",

}

Sulem, E, Abend, O & Rappoport, A 2015, Conceptual Annotations Preserve Structure Across Translations: A French-English Case Study. in D Xiong, K Duh, C Hardmeier & R Navigli (eds), Proceedings of the 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015). Association for Computational Linguistics (ACL), Beijing, China, pp. 11-22, 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015), Beijing, China, 30/07/15. https://doi.org/10.18653/v1/W15-3502

Conceptual Annotations Preserve Structure Across Translations: A French-English Case Study. / Sulem, Elior; Abend, Omri ; Rappoport, Ari.
Proceedings of the 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015). ed. / Deyi Xiong; Kevin Duh; Christian Hardmeier; Roberto Navigli. Beijing, China: Association for Computational Linguistics (ACL), 2015. p. 11-22.

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

TY - GEN

T1 - Conceptual Annotations Preserve Structure Across Translations

T2 - 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015)

AU - Sulem, Elior

AU - Abend, Omri

AU - Rappoport, Ari

N1 - Conference code: 1

PY - 2015/7/1

Y1 - 2015/7/1

N2 - Divergence of syntactic structures between languages constitutes a major challenge in using linguistic structure in Machine Translation (MT) systems. Here, we examine the potential of semantic structures. While semantic annotation is appealing as a source of cross-linguistically stable structures, little has been accomplished in demonstrating this stability through a detailed corpus study. In this paper, we experiment with the UCCA conceptual-cognitive annotation scheme in an English-French case study. First, we show that UCCA can be used to annotate French, through a systematic type-level analysis of the major French grammatical phenomena. Second, we annotate a parallel English-French corpus with UCCA, and quantify the similarity of the structures on both sides. Results show a high degree of stability across translations, supporting the usage of semantic annotations over syntactic ones in structure-aware MT systems.

AB - Divergence of syntactic structures between languages constitutes a major challenge in using linguistic structure in Machine Translation (MT) systems. Here, we examine the potential of semantic structures. While semantic annotation is appealing as a source of cross-linguistically stable structures, little has been accomplished in demonstrating this stability through a detailed corpus study. In this paper, we experiment with the UCCA conceptual-cognitive annotation scheme in an English-French case study. First, we show that UCCA can be used to annotate French, through a systematic type-level analysis of the major French grammatical phenomena. Second, we annotate a parallel English-French corpus with UCCA, and quantify the similarity of the structures on both sides. Results show a high degree of stability across translations, supporting the usage of semantic annotations over syntactic ones in structure-aware MT systems.

KW - linguistic structure

KW - Machine Translation

KW - MT

KW - semantic structures

KW - UCCA

KW - semantic annotations

U2 - 10.18653/v1/W15-3502

DO - 10.18653/v1/W15-3502

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

SP - 11

EP - 22

BT - Proceedings of the 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015)

A2 - Xiong, Deyi

A2 - Duh, Kevin

A2 - Hardmeier, Christian

A2 - Navigli, Roberto

PB - Association for Computational Linguistics (ACL)

CY - Beijing, China

Y2 - 30 July 2015 through 30 July 2015

ER -

Sulem E, Abend O , Rappoport A. Conceptual Annotations Preserve Structure Across Translations: A French-English Case Study. In Xiong D, Duh K, Hardmeier C, Navigli R, editors, Proceedings of the 1st Workshop on Semantics-Driven Statistical Machine Translation (S2MT 2015). Beijing, China: Association for Computational Linguistics (ACL). 2015. p. 11-22 doi: 10.18653/v1/W15-3502

Conceptual Annotations Preserve Structure Across Translations: A French-English Case Study

Abstract

Workshop

Keywords

Access to Document

Fingerprint

Cite this