Mesurer la proximité entre corpus par de nouveaux méta-descripteurs

Abstract : Given the number of existing classification algorithms, finding the most appropriate for classifying a new corpus is a difficult task. Meta-classification appears today very useful to help to determine, by using past experiences, what should be the most suitable algorithm compared to our corpus. The underlying idea is that "if an algorithm was particularly suitable for a corpus, it should have the same behavior on a quite similar corpus.". In this paper, we propose new meta-descriptors based on the concept of similarity to improve the meta-classification step. Conducted experiments on real dataset show the relevance of our new meta-descriptors.
Complete list of metadatas

Cited literature [19 references]  Display  Hide  Download

https://hal-lirmm.ccsd.cnrs.fr/lirmm-01184560
Contributor : Mathieu Roche <>
Submitted on : Sunday, August 16, 2015 - 5:38:56 AM
Last modification on : Wednesday, September 18, 2019 - 4:04:04 PM
Long-term archiving on : Tuesday, November 17, 2015 - 10:11:15 AM

File

37.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : lirmm-01184560, version 1

Citation

Flavien Bouillot, Pascal Poncelet, Mathieu Roche. Mesurer la proximité entre corpus par de nouveaux méta-descripteurs. CORIA: Conférence en Recherche d’Information et Applications, Mar 2015, Paris, France. pp.369-383. ⟨lirmm-01184560⟩

Share

Metrics

Record views

298

Files downloads

366