The GREC challenge 2008; Overview and evaluation results

Anja Belz; Eric Kow; Jette Viethen; Albert Gatt

doi:10.1.1.140.9786

The GREC challenge 2008; Overview and evaluation results

Anja Belz, Eric Kow, Jette Viethen, Albert Gatt

University of Brighton

Research output: Chapter in Book/Conference proceeding with ISSN or ISBN › Conference contribution with ISSN or ISBN › peer-review

Abstract

The GREC Task at REG’08 required participating systems to select coreference chains to the main subject of short encyclopaedic texts collected from Wikipedia. Three teams submitted a total of 6 systems, and we additionally created four baseline systems. Systems were tested automatically using a range of existing intrinsic metrics. We also evaluated systems extrinsically by applying coreference resolution tools to the outputs and measuring the success of the tools. In addition, systems were tested in a reading/comprehension experiment involving human subjects. This report describes the GREC Task and the evaluation methods, gives brief descriptions of the participating systems, and presents the evaluation results.

Original language	English
Title of host publication	5th International Natural Language Generation Conference (INLG'08)
Place of Publication	Stroudsburg, PA, USA
Publisher	Association for Computational Linguistics
Pages	183-199
Number of pages	17
DOIs	https://doi.org/10.1.1.140.9786
Publication status	Published - 1 Jan 2008
Event	5th International Natural Language Generation Conference (INLG'08) - Ohio, USA Duration: 1 Jan 2008 → …

Conference

Conference	5th International Natural Language Generation Conference (INLG'08)
Period	1/01/08 → …

Access to Document

10.1.1.140.9786Licence: Unspecified

http://www.aclweb.org/anthology/W/W08/W08-1127Licence: Unspecified

Cite this

@inproceedings{00022589c2aa4331bad8f27f7751d143,

title = "The GREC challenge 2008; Overview and evaluation results",

abstract = "The GREC Task at REG{\textquoteright}08 required participating systems to select coreference chains to the main subject of short encyclopaedic texts collected from Wikipedia. Three teams submitted a total of 6 systems, and we additionally created four baseline systems. Systems were tested automatically using a range of existing intrinsic metrics. We also evaluated systems extrinsically by applying coreference resolution tools to the outputs and measuring the success of the tools. In addition, systems were tested in a reading/comprehension experiment involving human subjects. This report describes the GREC Task and the evaluation methods, gives brief descriptions of the participating systems, and presents the evaluation results.",

author = "Anja Belz and Eric Kow and Jette Viethen and Albert Gatt",

year = "2008",

month = jan,

day = "1",

doi = "10.1.1.140.9786",

language = "English",

pages = "183--199",

booktitle = "5th International Natural Language Generation Conference (INLG'08)",

publisher = "Association for Computational Linguistics",

note = "5th International Natural Language Generation Conference (INLG'08) ; Conference date: 01-01-2008",

}

The GREC challenge 2008; Overview and evaluation results. / Belz, Anja; Kow, Eric; Viethen, Jette et al.
5th International Natural Language Generation Conference (INLG'08). Stroudsburg, PA, USA: Association for Computational Linguistics, 2008. p. 183-199.

Research output: Chapter in Book/Conference proceeding with ISSN or ISBN › Conference contribution with ISSN or ISBN › peer-review

TY - GEN

T1 - The GREC challenge 2008; Overview and evaluation results

AU - Belz, Anja

AU - Kow, Eric

AU - Viethen, Jette

AU - Gatt, Albert

PY - 2008/1/1

Y1 - 2008/1/1

N2 - The GREC Task at REG’08 required participating systems to select coreference chains to the main subject of short encyclopaedic texts collected from Wikipedia. Three teams submitted a total of 6 systems, and we additionally created four baseline systems. Systems were tested automatically using a range of existing intrinsic metrics. We also evaluated systems extrinsically by applying coreference resolution tools to the outputs and measuring the success of the tools. In addition, systems were tested in a reading/comprehension experiment involving human subjects. This report describes the GREC Task and the evaluation methods, gives brief descriptions of the participating systems, and presents the evaluation results.

AB - The GREC Task at REG’08 required participating systems to select coreference chains to the main subject of short encyclopaedic texts collected from Wikipedia. Three teams submitted a total of 6 systems, and we additionally created four baseline systems. Systems were tested automatically using a range of existing intrinsic metrics. We also evaluated systems extrinsically by applying coreference resolution tools to the outputs and measuring the success of the tools. In addition, systems were tested in a reading/comprehension experiment involving human subjects. This report describes the GREC Task and the evaluation methods, gives brief descriptions of the participating systems, and presents the evaluation results.

U2 - 10.1.1.140.9786

DO - 10.1.1.140.9786

M3 - Conference contribution with ISSN or ISBN

SP - 183

EP - 199

BT - 5th International Natural Language Generation Conference (INLG'08)

PB - Association for Computational Linguistics

CY - Stroudsburg, PA, USA

T2 - 5th International Natural Language Generation Conference (INLG'08)

Y2 - 1 January 2008

ER -

The GREC challenge 2008; Overview and evaluation results

Abstract

Conference

Access to Document

Fingerprint

Cite this