“Early Malicious Activity Discovery in Microblogs by Social Bridges Detection”

“Early Malicious Activity Discovery in Microblogs by Social Bridges Detection” paper

The paper “Early Malicious Activity Discovery in Microblogs by Social Bridges Detection” by Gogoglou Antonia, Theodosiou Zenonas, Kounoudes Tasos, Vakali Athena and Manolopoulos Yannis has been accepted for publication at the 16th International Symposium on Signal Processing and Information Technology (ISSPIT2016) and will be soon available via IEEExplore. This work was completed for the purposes of H2020 project ENCASE, that aims to protect youngsters in Online Social Networks and the Web in general. The current publication proposes a framework to detect potentially malicious users in Twitter and defines the concept of “social bridges”, meaning the influential users that are utilized by the malicious ones to help penetrate the large component of an OSN. A topology based classification model was introduced to allow for early detection of any dangerous new connection of underage Twitter users.