Unsupervised SPITters detection scheme for unbalanced callers

As the IP-based telephony service is getting popular, new attackers called SPITters (Spam over Internet Telephony callers) who advertise products and conduct a survey is being emerged and it is urgent demand to detect them. Recently, a novel unsupervised SPITters detection scheme, which leverages a clustering algorithm, has been proposed. However, this scheme does not work well when the SPITters account for a small fraction of the entire caller. In this paper, we propose a new unsupervised SPITters detection scheme by adding artificial data to solve such unbalanced situation. Our scheme will avoid some of the legitimate callers from being clustered into the SPITters' cluster and the classification performance will be improved. We show the efficiency of the proposed scheme by means of computer simulation with real and artificial call log datasets.