Privacy-Preserving Distributed Collaborative Filtering

Abstract : We propose a mechanism to preserve privacy while leveraging user profiles in distributed recommender systems. Our approach relies on (i ) an original obfuscation mechanism hiding the exact profiles of users without significantly decreasing their utility, as well as (ii ) a randomized dissemination algorithm ensuring differential privacy during the dissemination process. We evaluate our system against an alternative providing differential privacy both during profile construction and dissemination. Results show that our solution preserves accuracy without the need for users to reveal their preferences. Our approach is also flexible and more robust to censorship.
Type de document :
Rapport
[Research Report] RR-8253, INRIA. 2013
Liste complète des métadonnées

Littérature citée [24 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00799209
Contributeur : Antoine Boutet <>
Soumis le : lundi 11 mars 2013 - 18:13:20
Dernière modification le : mercredi 11 avril 2018 - 02:01:25
Document(s) archivé(s) le : lundi 17 juin 2013 - 11:43:04

Fichier

RR-8253.pdf
Accord explicite pour ce dépôt

Identifiants

  • HAL Id : hal-00799209, version 1

Citation

Antoine Boutet, Anne-Marie Kermarrec, Davide Frey, Rachid Guerraoui, Arnaud Jégou. Privacy-Preserving Distributed Collaborative Filtering. [Research Report] RR-8253, INRIA. 2013. 〈hal-00799209〉

Partager

Métriques

Consultations de la notice

1642

Téléchargements de fichiers

1073