Prodigious explosion of social network services may trigger new business models, there trails, however, negative aspects such as personal information spill or spamming as much. Amongst conventional spam detection approaches, the studies based on vertex Degrees have been sacrificed false negative results so that normal vertices can be specified as spammer ones. In this paper, we propose a novel approach by applying the circuit structure in the social networks,, which demonstrates the advantages of our work in the experiment.