Skip to Main content Skip to Navigation

Détection de comportements et identification de rôles dans les réseaux sociaux

Abstract : Social networks (SN) are omnipresent in our lives today. Not all users have the same behavior on these networks. If some have a low activity, rarely posting messages and following few users, some others at the other extreme have a significant activity, with many followers and regularly posts. The important role of these popular SN users makes them the target of many applications for example for content monitoring or advertising. After a study of the metadata of these users, in order to detect abnormal accounts, we present an approach allowing to detect users who are becoming popular. Our approach is based on modeling the evolution of popularity in the form of frequent patterns. These patterns describe the behaviors of gaining popularity. We propose a pattern matching model which can be used with a data stream and we show its scalability and its performance by comparing it to classic models. Finally, we present a clustering approach based on PageRank. This work allow to identify groups of users sharing the same role, using the interaction graphs
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Wednesday, December 8, 2021 - 9:33:39 AM
Last modification on : Monday, February 21, 2022 - 3:38:11 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03464517, version 2



Jonathan Debure. Détection de comportements et identification de rôles dans les réseaux sociaux. Réseaux sociaux et d'information [cs.SI]. Conservatoire national des arts et metiers - CNAM, 2021. Français. ⟨NNT : 2021CNAM1290⟩. ⟨tel-03464517v2⟩



Record views


Files downloads