The impact of corpus quality and type on topic based text segmentation evaluation

In this paper, we try to fathom the real impact of corpus quality on methods performances and their evaluations. The considered task is topic-based text segmentation, and two highly different unsupervised algorithms are compared: C 99, a word-based system, augmented with LSA, and Transeg, a sentence...

Full description

Saved in:
Bibliographic Details
Published in2008 International Multiconference on Computer Science and Information Technology : 20-22 October 2008 pp. 313 - 319
Main Authors Labadie, A., Prince, V.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2008
Subjects
Algorithm design and analysis
Calculus
Computer science
Concatenated codes
Frequency
Information technology
Organizing
Performance evaluation
Protocols
Robustness
Online AccessGet full text
ISBN8360810141
9788360810149
ISSN2157-5525
1896-7094
2157-5533
DOI10.1109/IMCSIT.2008.4747258

Cover

More Information
Summary:In this paper, we try to fathom the real impact of corpus quality on methods performances and their evaluations. The considered task is topic-based text segmentation, and two highly different unsupervised algorithms are compared: C 99, a word-based system, augmented with LSA, and Transeg, a sentence-based system. Two main characteristics of corpora have been investigated: Data quality (clean vs raw corpora), corpora manipulation (natural vs artificial data sets). The corpus size has also been subject to variation, and experiments related in this paper have shown that corpora characteristics highly impact recall and precision values for both algorithms.
ISBN:8360810141
9788360810149
ISSN:2157-5525
1896-7094
2157-5533
DOI:10.1109/IMCSIT.2008.4747258