Deception Detection and Social Engineering Attacks - Livestream

Humans are naturally concerned about being deceived, so having assistance in identifying malicious verbal or nonverbal behaviors that may indicate deception would be extremely beneficial. With its emergence in courtrooms, detecting lies and deception has attracted the attention of psychologists, law enforcement professionals, and cognitive scientists. Despite extensive research in deception detection, it remains an open research question, and it is especially difficult to find a coherent framework for automatic deception detection. One application of such a framework could be in cybersecurity and the detection of automated social engineering attacks.

Social engineering attacks include a wide range of manipulative activities aimed at obtaining sensitive information from humans. This type of attack has become one of the preferred methods of attackers because it is easier to manipulate the weakest link in cybersecurity ( i.e., humans). Characteristics such as taking advantage of humans’ tendency to trust rather than find technical methods to exploit, publicly available online data about victims provided by social media, as well as the low cost and high effectiveness of such attacks, are making them more common by the day. Despite the growing number and significance of social engineering attacks, this field of study is still in its early stages.

This talk discusses current research on deception detection. Furthermore, interactions in social engineering attacks as one type of deceptive communication are explained, and a generic social engineering attack is modeled using the Markov Decision Process (MDP). Finally, future work and research directions are presented.

Speaker: Faranak Abri, San Jose State University