Apprentissage de représentations probabilistes pour la prédiction de diffusions d'informations sur les réseaux sociaux

Based on the well-known Independent Cascade (IC) model, we embed users of the social network in a representation space to extract more robust diffusion probabilities than those defined by classical graphical learning approaches for social influence modeling. Better generalization abilities provided by the use of such a projection space allows our approach to present good performances on various real-world datasets, for both diffusion prediction and influence relationships inference tasks. MOTS-CLÉS : Apprentissage de Représentation, Diffusion d’Information.