Skip to Main content Skip to Navigation
Conference papers

Une méthode non supervisée pour la vérification d'auteur à base d'un modèle gaussien multivarié

Abstract : In this paper, we present a first study on using a distance-based outlier detection method for the authorship verification task. We have considered an unsupervised method based on a multivariate Gaussian model. To evaluate the effectiveness of the proposed method, we conducted experiments on a classic French corpus. Our preliminary results show that the proposed method can achieve a high verification performance that can reach an F 1 score of 83% outperforming the baseline. MOTS-CLÉS : vérification non supervisée de l'auteur, détection des cas aberrants, modèle Gaussien multivarié.
Complete list of metadatas

Cited literature [19 references]  Display  Hide  Download

https://hal.sorbonne-universite.fr/hal-01198412
Contributor : Mohamed Amine Boukhaled <>
Submitted on : Saturday, September 12, 2015 - 2:52:45 PM
Last modification on : Thursday, March 21, 2019 - 2:44:48 PM
Long-term archiving on: : Tuesday, December 29, 2015 - 12:54:32 AM

File

authorship_verification_BOUKHA...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01198412, version 1

Citation

Mohamed Amine Boukhaled. Une méthode non supervisée pour la vérification d'auteur à base d'un modèle gaussien multivarié. 10es Rencontres Jeunes Chercheurs en Recherche d’Information (RJCRI), Mar 2015, Paris, France. pp.525-533. ⟨hal-01198412⟩

Share

Metrics

Record views

131

Files downloads

90