Anonymisation semi-automatique de corpus d'interactions éléments pour une méthode interactive

Christophe Reffay 1 François-Marie Blondel 1 Stéphane Allaire 1 Emmanuel Giguet 2
2 Equipe Hultech - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
Abstract : In order to ease research data sharing and scientific comparison, researchers need appropriate methods and tools to anonymise their huge corpora of interaction. We first draw the legal context and present the stakes of anonymisation. We emphasise the subtleties that avoid complete automation of the anonymisation process. The target user-researcher will keep the control of the anonymisation process with the method we propose here. It is mainly based on two processes: marking and mining presented in details in this article. The whole method has been applied to two very different corpora. Preliminary evaluation of these first tests is given in the discussion and gives the floor to interesting perspectives for the method and its implementation on the Calico platform.
Document type :
Conference papers
Complete list of metadatas

Cited literature [16 references]  Display  Hide  Download

https://edutice.archives-ouvertes.fr/edutice-00720211
Contributor : Christophe Reffay <>
Submitted on : Friday, August 24, 2012 - 7:00:21 AM
Last modification on : Tuesday, November 19, 2019 - 12:47:18 PM
Long-term archiving on : Sunday, November 25, 2012 - 2:20:30 AM

File

Reffay_et_al_JOCAIR2012_comple...
Files produced by the author(s)

Identifiers

  • HAL Id : edutice-00720211, version 1

Citation

Christophe Reffay, François-Marie Blondel, Stéphane Allaire, Emmanuel Giguet. Anonymisation semi-automatique de corpus d'interactions éléments pour une méthode interactive. JOurnées Communication et Apprentissage Instrumentés en Réseau, Sep 2012, Amiens, France. ⟨edutice-00720211⟩

Share

Metrics

Record views

514

Files downloads

437