Transitive probabilistic CLIR models.

W. Kraaij; Franciska M.G. de Jong

Transitive probabilistic CLIR models.

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

19 Downloads (Pure)

Abstract

Transitive translation could be a useful technique to enlarge the number of supported language pairs for a cross-language information retrieval (CLIR) system in a cost-effective manner. The paper describes several setups for transitive translation based on probabilistic translation models. The transitive CLIR models were evaluated on the CLEF test collection and yielded a retrieval effectiveness up to 83% of monolingual performance, which is significantly better than a baseline using the synonym operator.

Original language	Undefined
Title of host publication	Proceedings of the 7th International RIAO Conference (Recherche d'Information Assistée par Ordinateur)
Place of Publication	Paris, France
Publisher	Centre de Hautes Etudes Internationales d'Informatique Documentaire (CID)
Pages	69-81
Number of pages	13
ISBN (Print)	2 905450-09-6
Publication status	Published - 2004
Event	7th International Conference on Computer-Assisted Information Retrieval, RIAO 2004: (Recherche d'Information et ses Applications) - Avignon, France Duration: 1 Apr 2004 → 1 Apr 2004 Conference number: 7

Publication series

Name
Publisher	Centre de hautes etudes internationales (CID)

Conference

Conference	7th International Conference on Computer-Assisted Information Retrieval, RIAO 2004
Abbreviated title	RIAO
Country	France
City	Avignon
Period	1/04/04 → 1/04/04

Keywords

EWI-6775
IR-66328
METIS-221281

Access to Document

kra-jon-riao2004e
open
kra-jon-riao2004e
open

http://www.riao.org/sites/RIAO-2004/Proceedings-2004/

Cite this

@inproceedings{6267e531d5b2433eb6ce4a362f25eb23,

title = "Transitive probabilistic CLIR models.",

abstract = "Transitive translation could be a useful technique to enlarge the number of supported language pairs for a cross-language information retrieval (CLIR) system in a cost-effective manner. The paper describes several setups for transitive translation based on probabilistic translation models. The transitive CLIR models were evaluated on the CLEF test collection and yielded a retrieval effectiveness up to 83% of monolingual performance, which is significantly better than a baseline using the synonym operator.",

keywords = "EWI-6775, IR-66328, METIS-221281",

author = "W. Kraaij and {de Jong}, {Franciska M.G.}",

note = "Kraaij369:2004 ISBN=2 905450-09-6 ; null ; Conference date: 01-04-2004 Through 01-04-2004",

year = "2004",

language = "Undefined",

isbn = "2 905450-09-6",

publisher = "Centre de Hautes Etudes Internationales d'Informatique Documentaire (CID)",

pages = "69--81",

booktitle = "Proceedings of the 7th International RIAO Conference (Recherche d'Information Assist{\'e}e par Ordinateur)",

address = "France",

}

Kraaij, W & de Jong, FMG 2004, Transitive probabilistic CLIR models. in Proceedings of the 7th International RIAO Conference (Recherche d'Information Assistée par Ordinateur). Centre de Hautes Etudes Internationales d'Informatique Documentaire (CID), Paris, France, pp. 69-81, 7th International Conference on Computer-Assisted Information Retrieval, RIAO 2004, Avignon, France, 1/04/04. <http://www.riao.org/sites/RIAO-2004/Proceedings-2004/>

Transitive probabilistic CLIR models. / Kraaij, W.; de Jong, Franciska M.G.

Proceedings of the 7th International RIAO Conference (Recherche d'Information Assistée par Ordinateur). Paris, France : Centre de Hautes Etudes Internationales d'Informatique Documentaire (CID), 2004. p. 69-81.

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

TY - GEN

T1 - Transitive probabilistic CLIR models.

AU - Kraaij, W.

AU - de Jong, Franciska M.G.

N1 - Conference code: 7

PY - 2004

Y1 - 2004

N2 - Transitive translation could be a useful technique to enlarge the number of supported language pairs for a cross-language information retrieval (CLIR) system in a cost-effective manner. The paper describes several setups for transitive translation based on probabilistic translation models. The transitive CLIR models were evaluated on the CLEF test collection and yielded a retrieval effectiveness up to 83% of monolingual performance, which is significantly better than a baseline using the synonym operator.

AB - Transitive translation could be a useful technique to enlarge the number of supported language pairs for a cross-language information retrieval (CLIR) system in a cost-effective manner. The paper describes several setups for transitive translation based on probabilistic translation models. The transitive CLIR models were evaluated on the CLEF test collection and yielded a retrieval effectiveness up to 83% of monolingual performance, which is significantly better than a baseline using the synonym operator.

KW - EWI-6775

KW - IR-66328

KW - METIS-221281

M3 - Conference contribution

SN - 2 905450-09-6

SP - 69

EP - 81

BT - Proceedings of the 7th International RIAO Conference (Recherche d'Information Assistée par Ordinateur)

PB - Centre de Hautes Etudes Internationales d'Informatique Documentaire (CID)

CY - Paris, France

Y2 - 1 April 2004 through 1 April 2004

ER -