PROJECT TITLE :

Risk Assessment in Social Networks Based on User Anomalous Behaviors - 2018

ABSTRACT:

Although the dramatic increase in On-line Social Network (OSN) usage, there are still a ton of security and privacy considerations. In such a situation, it'd be terribly useful to possess a mechanism in a position to assign a risk score to each OSN user. For this reason, in this Project, we have a tendency to propose a risk assessment based mostly on the thought that the more a user behavior diverges from what it will be thought of as a 'normal behavior', the a lot of it ought to be thought-about risky. In doing this, we have taken under consideration that OSN population is really heterogeneous in observed behaviors. As such, it's not possible to outline a unique standard behavioral model that matches all OSN users' behaviors. However, we have a tendency to expect that similar folks tend to follow similar rules with the results of similar behavioral models. For this reason, we have a tendency to propose a risk assessment approach organized into 2 phases: similar users are first grouped together, then, for every identified group, we build a number of models for traditional behavior. The disbursed experiments on a real Facebook dataset show that the proposed model outperforms a simplified behavioral-based mostly risk assessment where behavioral models are engineered over the full OSN population, while not a group identification section.