IRIT at CheckThat! 2018 - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

IRIT at CheckThat! 2018

Résumé

The 2018 CLEF CheckThat! is composed of two tasks: (1) Check-Worthiness and (2) Factuality. We participated to task (1) only which purpose is to evaluate the check-worthiness of claims in political debates. Our method to achieve this goal is to represent each claim by a vector of five computed values that correspond to scores on five criteria. These vectors are then used with machine learning algorithms to classify claims as check-worthy or not. We submitted three runs using different machine learning algorithms. The best result we achieved using the offi-cial measure MAP ranks our primary run the 12th over the 16 submitted runs. Our first contrastive run is ranked 2nd with the Mean Precision@1 measure.
Fichier principal
Vignette du fichier
agez_22448.pdf (137.43 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02289947 , version 1 (17-09-2019)

Identifiants

  • HAL Id : hal-02289947 , version 1
  • OATAO : 22448

Citer

Romain Agez, Clément Bosc, Cédric Lespagnol, Josiane Mothe, Noémie Petitcol. IRIT at CheckThat! 2018. 9th Conference and Labs of the Evaluation Forum, Living Labs (CLEF 2018), Sep 2018, Avignon, France. ⟨hal-02289947⟩
56 Consultations
27 Téléchargements

Partager

Gmail Facebook X LinkedIn More