iDetect: An immunity based algorithm to detect harmful content shared in Peer-to-Peer networks

A huge amount of harmful and illegal contents such as child pornography and abuse video are shared in Peer-to-Peer (P2P) network and have brought some serious social problems. Traditional detection algorithms monitor and analyze the content of the P2P traffic by deploying centralized powerful servers. The immense amount of sharing, transferring and frequently updating files content in P2P network makes these techniques quite cost-expensive and inefficient to detect the harmful elements in time. We develop the iDetect, a distributed harmful content detection algorithm inspired by the Clonal Selection mechanism of the immune system. Analogous to the B-lymphocytes secreting antibodies against antigens in human bodies, the clients in the P2P network deployed with the iDetect cooperate to detect the harmful content in a distributed and self-organized manner. Experiments show that the algorithm is efficient, effective, scalable to locate the clients sharing harmful content in the P2P network.