A weighted-graph based approach for solving the cold start problem in collaborative recommender systems. - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2012

A weighted-graph based approach for solving the cold start problem in collaborative recommender systems.

Nassira Chekkai
  • Fonction : Auteur
Salim Chikhi
  • Fonction : Auteur

Résumé

In recent years, social networks analysis has become a very active area of research. There has been a great focus of studying social networks in the area of graph theory. One important problem in social networks is the identification of the most influential actors which can be detected using different graph parameters. In this paper, we attempt to identify minimum sets of the most influential actors in trust social networks using weighted graphs with more focus on the critical nodes detection. We propose three methods: the first identifies the sets of the most important critical nodes using the concept of network efficiency; the second specifies conditions under which critical nodes should be controlled using connected components; and the third extracts sets of most powerful articulation points which are able to play critical nodes role. Computational results are demonstrated to confirm the effectiveness of our methods.
Fichier non déposé

Dates et versions

hal-01353179 , version 1 (10-08-2016)

Identifiants

  • HAL Id : hal-01353179 , version 1

Citer

Nassira Chekkai, Salim Chikhi, Hamamache Kheddouci. A weighted-graph based approach for solving the cold start problem in collaborative recommender systems.. International Symposium on Computers and Communications, Jul 2012, Cappadocia, Turkey. pp.01/07-04/07. ⟨hal-01353179⟩
90 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More