Trust Assessment in Online Social Networks

Assessing trust in online social networks (OSNs) is critical for many applications such as marketing and network security. It a challenging problem, however, due to the difficulties of handling complex topologies conducting accurate assessment these topologies. To address challenges, we model by proposing three-valued subjective logic (3VSL) model. 3VSL properly models uncertainties that exist trust, thus able compute arbitrary graphs. We theoretically prove capability based on Dirichlet-Categorical (DC) distribution its correctness OSN Based model, further design AssessTrust (AT) algorithm accurately between any two users connected an OSN. validate against real-world datasets: Advogato Pretty Good Privacy (PGP). Experimental results indicate can pair indirectly PGP.