Cross Language Information Retrieval for Biomedical Literature

Martijn Schuemie; Rudolf Berend Trieschnigg; Wessel Kraaij

Cross Language Information Retrieval for Biomedical Literature

Martijn Schuemie, Rudolf Berend Trieschnigg, Wessel Kraaij

Databases (Former)

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

1 Downloads (Pure)

Abstract

This workshop report discusses the collaborative work of UT, EMC and TNO on the TREC Genomics Track 2007. The biomedical information retrieval task is approached using cross language methods, in which biomedical concept detection is combined with eﬀective IR based on unigram language models. Furthermore, a co-occurrence method is used to select and ��?lter candidate answers. On its own, the cross lingual approach and the ��?ltering do not strongly improve retrieval results. However, the combination of approaches does show a strong improvement over the monolingual baseline.

Original language	Undefined
Title of host publication	The Sixteenth Text REtrieval Conference (TREC 2007) Proceedings
Editors	E.M Voorhees, L.P. Buckland
Place of Publication	Gaithersburg
Publisher	NIST
Pages	-
Number of pages	10
ISBN (Print)	not assigned
Publication status	Published - Nov 2007
Event	16th Text REtrieval Conference, TREC 2007 - Gaithersburg, United States Duration: 6 Nov 2007 → 9 Nov 2007 Conference number: 16

Publication series

Name
Publisher	NIST
Number	10

Conference

Conference	16th Text REtrieval Conference, TREC 2007
Abbreviated title	TREC
Country/Territory	United States
City	Gaithersburg
Period	6/11/07 → 9/11/07

Keywords

IR-62527
EWI-14033
METIS-252102

Access to Document

Cite this

@inproceedings{e1dac7addc0748bc8852f151d236d096,

title = "Cross Language Information Retrieval for Biomedical Literature",

abstract = "This workshop report discusses the collaborative work of UT, EMC and TNO on the TREC Genomics Track 2007. The biomedical information retrieval task is approached using cross language methods, in which biomedical concept detection is combined with eﬀective IR based on unigram language models. Furthermore, a co-occurrence method is used to select and ��?lter candidate answers. On its own, the cross lingual approach and the ��?ltering do not strongly improve retrieval results. However, the combination of approaches does show a strong improvement over the monolingual baseline.",

keywords = "IR-62527, EWI-14033, METIS-252102",

author = "Martijn Schuemie and Trieschnigg, {Rudolf Berend} and Wessel Kraaij",

year = "2007",

month = nov,

language = "Undefined",

isbn = "not assigned",

publisher = "NIST",

number = "10",

pages = "--",

editor = "E.M Voorhees and L.P. Buckland",

booktitle = "The Sixteenth Text REtrieval Conference (TREC 2007) Proceedings",

note = "16th Text REtrieval Conference, TREC 2007 ; Conference date: 06-11-2007 Through 09-11-2007",

}

Schuemie, M, Trieschnigg, RB & Kraaij, W 2007, Cross Language Information Retrieval for Biomedical Literature. in EM Voorhees & LP Buckland (eds), The Sixteenth Text REtrieval Conference (TREC 2007) Proceedings. NIST, Gaithersburg, pp. -, 16th Text REtrieval Conference, TREC 2007, Gaithersburg, Maryland, United States, 6/11/07. <http://eprints.eemcs.utwente.nl/secure2/14033/01/TREC.2007.schuemie.pdf>

TY - GEN

T1 - Cross Language Information Retrieval for Biomedical Literature

AU - Schuemie, Martijn

AU - Trieschnigg, Rudolf Berend

AU - Kraaij, Wessel

N1 - Conference code: 16

PY - 2007/11

Y1 - 2007/11

N2 - This workshop report discusses the collaborative work of UT, EMC and TNO on the TREC Genomics Track 2007. The biomedical information retrieval task is approached using cross language methods, in which biomedical concept detection is combined with eﬀective IR based on unigram language models. Furthermore, a co-occurrence method is used to select and ��?lter candidate answers. On its own, the cross lingual approach and the ��?ltering do not strongly improve retrieval results. However, the combination of approaches does show a strong improvement over the monolingual baseline.

AB - This workshop report discusses the collaborative work of UT, EMC and TNO on the TREC Genomics Track 2007. The biomedical information retrieval task is approached using cross language methods, in which biomedical concept detection is combined with eﬀective IR based on unigram language models. Furthermore, a co-occurrence method is used to select and ��?lter candidate answers. On its own, the cross lingual approach and the ��?ltering do not strongly improve retrieval results. However, the combination of approaches does show a strong improvement over the monolingual baseline.

KW - IR-62527

KW - EWI-14033

KW - METIS-252102

M3 - Conference contribution

SN - not assigned

SP - -

BT - The Sixteenth Text REtrieval Conference (TREC 2007) Proceedings

A2 - Voorhees, E.M

A2 - Buckland, L.P.

PB - NIST

CY - Gaithersburg

T2 - 16th Text REtrieval Conference, TREC 2007

Y2 - 6 November 2007 through 9 November 2007

ER -