Français Anglais
Accueil Annuaire Plan du site
Home > Research results > Dissertations & habilitations
Research results
Ph.D de

Group : Heterogeneous Modeling

Intégration du Web Social dans les systèmes de recommandation

Starts on 01/03/2014
[QUERCINI Gianluca]

Funding : Convention industrielle de formation par la recherche
Affiliation : Centrale Supélec
Laboratory :

Defended on 19/12/2017, committee :
Jérôme Azé – Université de Montpellier, LIRMM

Nicolas LABROCHE – Université François-Rabelais de Tours, LI

Nacéra SEGHOUANI BENNACER – LRI, CentraleSupélec

Gianluca QUERCINI – LRI, CentraleSupélec

Chantal REYNAUD – Université Paris-Sud, LRI

Haïfa ZARGAYOUNA – Université Paris 13, LIPN

Uriel BERDUGO – Wepingo

Research activities :

Abstract :
The social Web grows more and more and gives through the web, access to a wide variety of resources, like sharing sites such as, exchange messages as Twitter, or social networks with the professional purpose such as LinkedIn, or more generally for social purposes, such as Facebook and LiveJournal. Thus, the same individual can be registered and active on different social networks (potentially having different purposes), in which it publishes various information, which are constantly growing, such as its name, locality, communities, messages, various activities, etc. This information is important especially for applications seeking to know their users in order to better understand their needs, activities and interests. The objective of our research is to exploit essentially the textual resources extracted from the different social networks of the same individual in order to construct his characterizing profile, which can be exploited in particular by applications seeking to understand their users, such as recommendation systems. Given its international dimension, the content of the Web is inherently multilingual and intrinsically ambiguous, since individuals from different origin publish it in natural language in a free vocabulary and therefore the exploited textual resources are also multilingual and ambiguous. Nevertheless, we propose automatic, multilingual, and unsupervised approaches using Wikipedia to build an expanded profile for each user by aggregating information from its various social networks. In addition, we analyzed the correlation between user’s personality traits and his / her discovered interests, with a view to further characterizing it.