Detecting user's interests based on the accuracy of collaborative tagging information

Manel Mezghani (1, 2) , André Péninou (1, 3) , Corinne Amel Zayani (1, 2) , Ikram Amous (1, 2) , Florence Sèdes (1)

1 IRIT-SIG - Systèmes d’Informations Généralisées
2 MIRACL - Multimedia, InfoRmation systems and Advanced Computing Laboratory
3 UT2J - Université Toulouse - Jean Jaurès

A user profile has to reflect the XVHU¶V needs according to his characteristics (personal data, interests and preferences), his context, and his situation. ln this article, we focus on the problem of adaptation quality in social networks, which is affected by the accuracy and relevance of the user's interests. The originality of our approach is the proposal of a new technique of interest' detection by analyzing the accuracy of the tagging behaviour of the user in order to figure out the tags which actually reflect the UHVRXUFHV¶ FRQWHQW. Our approach has been tested and evaluated on the Delicious social database.

Mots clés: Recherche d'information

Domaines: Théorie de l'information [cs.IT]; Recherche d'information [cs.IR]

Manel Mezghani, André Péninou, Corinne Amel Zayani, Ikram Amous, Florence Sèdes. Detecting user's interests based on the accuracy of collaborative tagging information. 13th International Conference on Computer-Supported Cooperative Work (ECSCW 2013), Sep 2013, Paphos, Cyprus. pp.21-26. ⟨hal-01225774⟩