dTrust: a simple deep learning approach for social recommendation - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

dTrust: a simple deep learning approach for social recommendation

Résumé

Rating prediction is a key task of e-commerce recommendation mechanisms. Recent studies in social recommendation enhance the performance of rating predictors by taking advantage of user relationship. However, these prediction approaches mostly rely on user personal information which is a privacy threat. In this paper, we present a simple rating prediction approach relying on deep neural networks that does not reveal any personal information. Experiments on real-world data sets showed that the approach outperforms state-of-the-art in terms of RMSE and MAE scores for both warm-start and cold-start problems.
Fichier principal
Vignette du fichier
CIC2017.pdf (825.97 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01578316 , version 1 (29-08-2017)
hal-01578316 , version 2 (06-09-2017)
hal-01578316 , version 3 (15-09-2017)

Identifiants

  • HAL Id : hal-01578316 , version 2

Citer

Quang-Vinh Dang, Claudia-Lavinia Ignat. dTrust: a simple deep learning approach for social recommendation. The 3rd IEEE International Conference on Collaboration and Internet Computing (CIC-17), Oct 2017, San Jose, United States. ⟨hal-01578316v2⟩
595 Consultations
978 Téléchargements

Partager

Gmail Facebook X LinkedIn More