Title :
How the ExpLSA approach impacts the document classification tasks
Author :
Béchet, Nicolas ; Roche, Mathieu ; Chauché, Jacques
Author_Institution :
CNRS, Univ. Montpellier 2, Montpellier
Abstract :
Latent semantic analysis (LSA) is a statistical method which can be used to classify texts. This paper proposes a sentence expansion method (ExpLSA) to improve document classification tasks. We propose to study the impact of ExpLSA on the size and on the type of corpora.
Keywords :
document handling; statistical analysis; text analysis; document classification tasks; latent semantic analysis; sentence expansion method; statistical method; Classification algorithms; Heart; Humans; Matrix decomposition; Protocols; Statistical analysis; Support vector machine classification; Support vector machines; Vocabulary; Writing;
Conference_Titel :
Digital Information Management, 2008. ICDIM 2008. Third International Conference on
Conference_Location :
London
Print_ISBN :
978-1-4244-2916-5
Electronic_ISBN :
978-1-4244-2917-2
DOI :
10.1109/ICDIM.2008.4746814
