Social networks have become an integral part of the digital life of nowadays society. Indeed, the preponderance of these networks has increased significantly in the last years, allowing users to share different types of multimedia contents with the community [FB, FK, YT, TW, LN, Ala08]. In the last years, we have also assisted to a dramatic increase on the number of attacks on these networks in order to obtain and access confidential data from several users. In this manner, it is possible to create and execute powerful targeted social network attacks which, by exploiting the information shared on such networks, can become more credible and difficult to detect. Indeed, according to a recent report, the number of phishing attacks over social networks has increased more than 1200% during 2010 [Microsoft11].
The project Social Network User Behavior Modeling aims at analyzing and modeling the behaviors of social networks users in order to create behavior models. With such models, the project aims at increasing the knowledge the users have of their connections and of the social network in general, creating a trust relationship between the user and the remaining community. These models will also allow the detection of profile changes, or behavior models, in order to detect compromised accounts or accounts created by hackers with the purpose of accessing informations other users may share over the mentioned networks. The creation of such models enables the social network to map each user to group, allowing a more efficient, direct and secure management of existing groups of users as well as the creation of novel groups. Novel recommendation systems can also be proposed. All the mentioned aspects illustrate the importance of this project.
|Start Date: 01-07-2013|
|End Date: 01-06-2015|
|Team: Eduardo Oliveira Estanqueiro Rocha, Paulo Jorge Salvador Serra Ferreira, Antonio Manuel Duarte Nogueira, Diogo Nuno Pereira Gomes, Ivo Petiz|
|Groups: Administrative Aveiro|
|Local Coordinator: Eduardo Oliveira Estanqueiro Rocha|
|Links: Internal Page|