King's College London

In this paper, we present our spambot overview on the creativity of the spammers, and the most important techniques that the spammer might use to deceive current spambot detection tools to bypass detection. These techniques include performing a series of malicious actions with variable time delays, repeating the same series of malicious actions multiple times, and interleaving legitimate and malicious actions. In response, we define our problems to detect the aforementioned techniques in addition to disguised and "don’t cares" actions. Our proposed algorithms to solve those problems are based on advanced data structures that are able to detect malicious actions efficiently in linear time, and in an innovative way.